WorldWideScience

Sample records for hyperspectral image classification

  1. Hyperspectral image classification using functional data analysis.

    Science.gov (United States)

    Li, Hong; Xiao, Guangrun; Xia, Tian; Tang, Y Y; Li, Luoqing

    2014-09-01

    The large number of spectral bands acquired by hyperspectral imaging sensors allows us to better distinguish many subtle objects and materials. Unlike other classical hyperspectral image classification methods in the multivariate analysis framework, in this paper, a novel method using functional data analysis (FDA) for accurate classification of hyperspectral images has been proposed. The central idea of FDA is to treat multivariate data as continuous functions. From this perspective, the spectral curve of each pixel in the hyperspectral images is naturally viewed as a function. This can be beneficial for making full use of the abundant spectral information. The relevance between adjacent pixel elements in the hyperspectral images can also be utilized reasonably. Functional principal component analysis is applied to solve the classification problem of these functions. Experimental results on three hyperspectral images show that the proposed method can achieve higher classification accuracies in comparison to some state-of-the-art hyperspectral image classification methods.

  2. Synergetics Framework for Hyperspectral Image Classification

    Science.gov (United States)

    Müller, R.; Cerra, D.; Reinartz, P.

    2013-05-01

    In this paper a new classification technique for hyperspectral data based on synergetics theory is presented. Synergetics - originally introduced by the physicist H. Haken - is an interdisciplinary theory to find general rules for pattern formation through selforganization and has been successfully applied in fields ranging from biology to ecology, chemistry, cosmology, and thermodynamics up to sociology. Although this theory describes general rules for pattern formation it was linked also to pattern recognition. Pattern recognition algorithms based on synergetics theory have been applied to images in the spatial domain with limited success in the past, given their dependence on the rotation, shifting, and scaling of the images. These drawbacks can be discarded if such methods are applied to data acquired by a hyperspectral sensor in the spectral domain, as each single spectrum, related to an image element in the hyperspectral scene, can be analysed independently. The classification scheme based on synergetics introduces also methods for spatial regularization to get rid of "salt and pepper" classification results and for iterative parameter tuning to optimize class weights. The paper reports an experiment on a benchmark data set frequently used for method comparisons. This data set consists of a hyperspectral scene acquired by the Airborne Visible Infrared Imaging Spectrometer AVIRIS sensor of the Jet Propulsion Laboratory acquired over the Salinas Valley in CA, USA, with 15 vegetation classes. The results are compared to state-of-the-art methodologies like Support Vector Machines (SVM), Spectral Information Divergence (SID), Neural Networks, Logistic Regression, Factor Graphs or Spectral Angle Mapper (SAM). The outcomes are promising and often outperform state-of-the-art classification methodologies.

  3. Classifications of objects on hyperspectral images

    DEFF Research Database (Denmark)

    Kucheryavskiy, Sergey

    tablets have the same or similar excipient and different active ingredients, some of the pixels chemically will be identical. But these similar pixels will be associated with different classes when a classification model is being calibrated. This can give unstable model and poor classification results...... information about spatial relations of the pixels. This works well in general, especially for exploratory analysis or multivariate curve resolution, but for some specific tasks it is not beneficial at all. One of such tasks is classification or clustering of objects on hyperspectral images. An object here....... In the present work a classification method that combines classic image classification approach and MIA is proposed. The basic idea is to group all pixels and calculate spectral properties of the pixel group to be used further as a vector of predictors for calibration and class prediction. The grouping can...

  4. Spectral-Spatial Hyperspectral Image Classification Based on KNN

    Science.gov (United States)

    Huang, Kunshan; Li, Shutao; Kang, Xudong; Fang, Leyuan

    2016-12-01

    Fusion of spectral and spatial information is an effective way in improving the accuracy of hyperspectral image classification. In this paper, a novel spectral-spatial hyperspectral image classification method based on K nearest neighbor (KNN) is proposed, which consists of the following steps. First, the support vector machine is adopted to obtain the initial classification probability maps which reflect the probability that each hyperspectral pixel belongs to different classes. Then, the obtained pixel-wise probability maps are refined with the proposed KNN filtering algorithm that is based on matching and averaging nonlocal neighborhoods. The proposed method does not need sophisticated segmentation and optimization strategies while still being able to make full use of the nonlocal principle of real images by using KNN, and thus, providing competitive classification with fast computation. Experiments performed on two real hyperspectral data sets show that the classification results obtained by the proposed method are comparable to several recently proposed hyperspectral image classification methods.

  5. Dictionary-Based, Clustered Sparse Representation for Hyperspectral Image Classification

    Directory of Open Access Journals (Sweden)

    Zhen-tao Qin

    2015-01-01

    Full Text Available This paper presents a new, dictionary-based method for hyperspectral image classification, which incorporates both spectral and contextual characteristics of a sample clustered to obtain a dictionary of each pixel. The resulting pixels display a common sparsity pattern in identical clustered groups. We calculated the image’s sparse coefficients using the dictionary approach, which generated the sparse representation features of the remote sensing images. The sparse coefficients are then used to classify the hyperspectral images via a linear SVM. Experiments show that our proposed method of dictionary-based, clustered sparse coefficients can create better representations of hyperspectral images, with a greater overall accuracy and a Kappa coefficient.

  6. a Diversified Deep Belief Network for Hyperspectral Image Classification

    Science.gov (United States)

    Zhong, P.; Gong, Z. Q.; Schönlieb, C.

    2016-06-01

    In recent years, researches in remote sensing demonstrated that deep architectures with multiple layers can potentially extract abstract and invariant features for better hyperspectral image classification. Since the usual real-world hyperspectral image classification task cannot provide enough training samples for a supervised deep model, such as convolutional neural networks (CNNs), this work turns to investigate the deep belief networks (DBNs), which allow unsupervised training. The DBN trained over limited training samples usually has many "dead" (never responding) or "potential over-tolerant" (always responding) latent factors (neurons), which decrease the DBN's description ability and thus finally decrease the hyperspectral image classification performance. This work proposes a new diversified DBN through introducing a diversity promoting prior over the latent factors during the DBN pre-training and fine-tuning procedures. The diversity promoting prior in the training procedures will encourage the latent factors to be uncorrelated, such that each latent factor focuses on modelling unique information, and all factors will be summed up to capture a large proportion of information and thus increase description ability and classification performance of the diversified DBNs. The proposed method was evaluated over the well-known real-world hyperspectral image dataset. The experiments demonstrate that the diversified DBNs can obtain much better results than original DBNs and comparable or even better performances compared with other recent hyperspectral image classification methods.

  7. Geographical classification of apple based on hyperspectral imaging

    Science.gov (United States)

    Guo, Zhiming; Huang, Wenqian; Chen, Liping; Zhao, Chunjiang; Peng, Yankun

    2013-05-01

    Attribute of apple according to geographical origin is often recognized and appreciated by the consumers. It is usually an important factor to determine the price of a commercial product. Hyperspectral imaging technology and supervised pattern recognition was attempted to discriminate apple according to geographical origins in this work. Hyperspectral images of 207 Fuji apple samples were collected by hyperspectral camera (400-1000nm). Principal component analysis (PCA) was performed on hyperspectral imaging data to determine main efficient wavelength images, and then characteristic variables were extracted by texture analysis based on gray level co-occurrence matrix (GLCM) from dominant waveband image. All characteristic variables were obtained by fusing the data of images in efficient spectra. Support vector machine (SVM) was used to construct the classification model, and showed excellent performance in classification results. The total classification rate had the high classify accuracy of 92.75% in the training set and 89.86% in the prediction sets, respectively. The overall results demonstrated that the hyperspectral imaging technique coupled with SVM classifier can be efficiently utilized to discriminate Fuji apple according to geographical origins.

  8. Spatial-spectral method for classification of hyperspectral images.

    Science.gov (United States)

    Bian, Xiaoyong; Zhang, Tianxu; Yan, Luxin; Zhang, Xiaolong; Fang, Houzhang; Liu, Hai

    2013-03-15

    Spatial-spectral approach with spatially adaptive classification of hyperspectral images is proposed. The rotation-invariant spatial texture information for each object is exploited and incorporated into the classifier by using the modified local Gabor binary pattern to distinguish different types of classes of interest. The proposed method can effectively suppress anisotropic texture in spatially separate classes as well as improve the discrimination among classes. Moreover, it becomes more robust with the within-class variation. Experimental results on the classification of three real hyperspectral remote sensing images demonstrate the effectiveness of the proposed approach.

  9. Hyperspectral image classification based on NMF Features Selection Method

    Science.gov (United States)

    Abe, Bolanle T.; Jordaan, J. A.

    2013-12-01

    Hyperspectral instruments are capable of collecting hundreds of images corresponding to wavelength channels for the same area on the earth surface. Due to the huge number of features (bands) in hyperspectral imagery, land cover classification procedures are computationally expensive and pose a problem known as the curse of dimensionality. In addition, higher correlation among contiguous bands increases the redundancy within the bands. Hence, dimension reduction of hyperspectral data is very crucial so as to obtain good classification accuracy results. This paper presents a new feature selection technique. Non-negative Matrix Factorization (NMF) algorithm is proposed to obtain reduced relevant features in the input domain of each class label. This aimed to reduce classification error and dimensionality of classification challenges. Indiana pines of the Northwest Indiana dataset is used to evaluate the performance of the proposed method through experiments of features selection and classification. The Waikato Environment for Knowledge Analysis (WEKA) data mining framework is selected as a tool to implement the classification using Support Vector Machines and Neural Network. The selected features subsets are subjected to land cover classification to investigate the performance of the classifiers and how the features size affects classification accuracy. Results obtained shows that performances of the classifiers are significant. The study makes a positive contribution to the problems of hyperspectral imagery by exploring NMF, SVMs and NN to improve classification accuracy. The performances of the classifiers are valuable for decision maker to consider tradeoffs in method accuracy versus method complexity.

  10. Hyperspectral image classification based on volumetric texture and dimensionality reduction

    Science.gov (United States)

    Su, Hongjun; Sheng, Yehua; Du, Peijun; Chen, Chen; Liu, Kui

    2015-06-01

    A novel approach using volumetric texture and reduced-spectral features is presented for hyperspectral image classification. Using this approach, the volumetric textural features were extracted by volumetric gray-level co-occurrence matrices (VGLCM). The spectral features were extracted by minimum estimated abundance covariance (MEAC) and linear prediction (LP)-based band selection, and a semi-supervised k-means (SKM) clustering method with deleting the worst cluster (SKMd) bandclustering algorithms. Moreover, four feature combination schemes were designed for hyperspectral image classification by using spectral and textural features. It has been proven that the proposed method using VGLCM outperforms the gray-level co-occurrence matrices (GLCM) method, and the experimental results indicate that the combination of spectral information with volumetric textural features leads to an improved classification performance in hyperspectral imagery.

  11. Advances in Spectral-Spatial Classification of Hyperspectral Images

    Science.gov (United States)

    Fauvel, Mathieu; Tarabalka, Yuliya; Benediktsson, Jon Atli; Chanussot, Jocelyn; Tilton, James C.

    2012-01-01

    Recent advances in spectral-spatial classification of hyperspectral images are presented in this paper. Several techniques are investigated for combining both spatial and spectral information. Spatial information is extracted at the object (set of pixels) level rather than at the conventional pixel level. Mathematical morphology is first used to derive the morphological profile of the image, which includes characteristics about the size, orientation and contrast of the spatial structures present in the image. Then the morphological neighborhood is defined and used to derive additional features for classification. Classification is performed with support vector machines using the available spectral information and the extracted spatial information. Spatial post-processing is next investigated to build more homogeneous and spatially consistent thematic maps. To that end, three presegmentation techniques are applied to define regions that are used to regularize the preliminary pixel-wise thematic map. Finally, a multiple classifier system is defined to produce relevant markers that are exploited to segment the hyperspectral image with the minimum spanning forest algorithm. Experimental results conducted on three real hyperspectral images with different spatial and spectral resolutions and corresponding to various contexts are presented. They highlight the importance of spectral-spatial strategies for the accurate classification of hyperspectral images and validate the proposed methods.

  12. Advances in Spectral-Spatial Classification of Hyperspectral Images

    Science.gov (United States)

    Fauvel, Mathieu; Tarabalka, Yuliya; Benediktsson, Jon Atli; Chanussot, Jocelyn; Tilton, James C.

    2012-01-01

    Recent advances in spectral-spatial classification of hyperspectral images are presented in this paper. Several techniques are investigated for combining both spatial and spectral information. Spatial information is extracted at the object (set of pixels) level rather than at the conventional pixel level. Mathematical morphology is first used to derive the morphological profile of the image, which includes characteristics about the size, orientation, and contrast of the spatial structures present in the image. Then, the morphological neighborhood is defined and used to derive additional features for classification. Classification is performed with support vector machines (SVMs) using the available spectral information and the extracted spatial information. Spatial postprocessing is next investigated to build more homogeneous and spatially consistent thematic maps. To that end, three presegmentation techniques are applied to define regions that are used to regularize the preliminary pixel-wise thematic map. Finally, a multiple-classifier (MC) system is defined to produce relevant markers that are exploited to segment the hyperspectral image with the minimum spanning forest algorithm. Experimental results conducted on three real hyperspectral images with different spatial and spectral resolutions and corresponding to various contexts are presented. They highlight the importance of spectral–spatial strategies for the accurate classification of hyperspectral images and validate the proposed methods.

  13. Hyperspectral remote sensing image classification based on decision level fusion

    Institute of Scientific and Technical Information of China (English)

    Peijun Du; Wei Zhang; Junshi Xia

    2011-01-01

    @@ To apply decision level fusion to hyperspectral remote sensing (HRS) image classification, three decision level fusion strategies are experimented on and compared, namely, linear consensus algorithm, improved evidence theory, and the proposed support vector machine (SVM) combiner.To evaluate the effects of the input features on classification performance, four schemes are used to organize input features for member classifiers.In the experiment, by using the operational modular imaging spectrometer (OMIS) II HRS image, the decision level fusion is shown as an effective way for improving the classification accuracy of the HRS image, and the proposed SVM combiner is especially suitable for decision level fusion.The results also indicate that the optimization of input features can improve the classification performance.%To apply decision level fusion to hyperspectral remote sensing (HRS) image classification, three decision level fusion strategies are experimented on and compared, namely, linear consensus algorithm, improved evidence theory, and the proposed support vector machine (SVM) combiner. To evaluate the effects of the input features on classification performance, four schemes are used to organize input features for member classifiers. In the experiment, by using the operational modular imaging spectrometer (OMIS) Ⅱ HRS image, the decision level fusion is shown as an effective way for improving the classification accuracy of the HRS image, and the proposed SVM combiner is especially suitable for decision level fusion. The results also indicate that the optimization of input features can improve the classification performance.

  14. Classification of maize kernels using NIR hyperspectral imaging

    DEFF Research Database (Denmark)

    Williams, Paul; Kucheryavskiy, Sergey V.

    2016-01-01

    NIR hyperspectral imaging was evaluated to classify maize kernels of three hardness categories: hard, medium and soft. Two approaches, pixel-wise and object-wise, were investigated to group kernels according to hardness. The pixel-wise classification assigned a class to every pixel from individual...... and specificity of 0.95 and 0.93). Both feature extraction methods can be recommended for classification of maize kernels on production scale....

  15. Super pixel-level dictionary learning for hyperspectral image classification

    Science.gov (United States)

    Zhao, Wei; Zhu, Wen; Liao, Bo; Fu, Xiangzheng

    2017-08-01

    This paper presents a superpixel-level dictionary learning model for hyperspectral data. The idea is to divide the hyperspectral image into a number of super-pixels by means of the super-pixel segmentation method. Each super-pixel is a spatial neighborhood called contextual group. That is, each pixel is represented using a linear combination of a few dictionary items learned from the train data, but since pixels inside a super-pixel are often consisting of the same materials, their linear combinations are constrained to use common items from the dictionary. To this end, the sparse coefficients of the context group have a common sparse pattern by using the joint sparse regularizer for dictionary learning. The sparse coefficients are then used for classification using linear support vector machines. The validity of the proposed method is experimentally verified on a real hyperspectral images.

  16. Going Deeper With Contextual CNN for Hyperspectral Image Classification.

    Science.gov (United States)

    Lee, Hyungtae; Kwon, Heesung

    2017-10-01

    In this paper, we describe a novel deep convolutional neural network (CNN) that is deeper and wider than other existing deep networks for hyperspectral image classification. Unlike current state-of-the-art approaches in CNN-based hyperspectral image classification, the proposed network, called contextual deep CNN, can optimally explore local contextual interactions by jointly exploiting local spatio-spectral relationships of neighboring individual pixel vectors. The joint exploitation of the spatio-spectral information is achieved by a multi-scale convolutional filter bank used as an initial component of the proposed CNN pipeline. The initial spatial and spectral feature maps obtained from the multi-scale filter bank are then combined together to form a joint spatio-spectral feature map. The joint feature map representing rich spectral and spatial properties of the hyperspectral image is then fed through a fully convolutional network that eventually predicts the corresponding label of each pixel vector. The proposed approach is tested on three benchmark data sets: the Indian Pines data set, the Salinas data set, and the University of Pavia data set. Performance comparison shows enhanced classification performance of the proposed approach over the current state-of-the-art on the three data sets.

  17. Classification of visible and infrared hyperspectral images based on image segmentation and edge-preserving filtering

    Science.gov (United States)

    Cui, Binge; Ma, Xiudan; Xie, Xiaoyun; Ren, Guangbo; Ma, Yi

    2017-03-01

    The classification of hyperspectral images with a few labeled samples is a major challenge which is difficult to meet unless some spatial characteristics can be exploited. In this study, we proposed a novel spectral-spatial hyperspectral image classification method that exploited spatial autocorrelation of hyperspectral images. First, image segmentation is performed on the hyperspectral image to assign each pixel to a homogeneous region. Second, the visible and infrared bands of hyperspectral image are partitioned into multiple subsets of adjacent bands, and each subset is merged into one band. Recursive edge-preserving filtering is performed on each merged band which utilizes the spectral information of neighborhood pixels. Third, the resulting spectral and spatial feature band set is classified using the SVM classifier. Finally, bilateral filtering is performed to remove "salt-and-pepper" noise in the classification result. To preserve the spatial structure of hyperspectral image, edge-preserving filtering is applied independently before and after the classification process. Experimental results on different hyperspectral images prove that the proposed spectral-spatial classification approach is robust and offers more classification accuracy than state-of-the-art methods when the number of labeled samples is small.

  18. Refinement of Hyperspectral Image Classification with Segment-Tree Filtering

    Directory of Open Access Journals (Sweden)

    Lu Li

    2017-01-01

    Full Text Available This paper proposes a novel method of segment-tree filtering to improve the classification accuracy of hyperspectral image (HSI. Segment-tree filtering is a versatile method that incorporates spatial information and has been widely applied in image preprocessing. However, to use this powerful framework in hyperspectral image classification, we must reduce the original feature dimensionality to avoid the Hughes problem; otherwise, the computational costs are high and the classification accuracy by original bands in the HSI is unsatisfactory. Therefore, feature extraction is adopted to produce new salient features. In this paper, the Semi-supervised Local Fisher (SELF method of discriminant analysis is used to reduce HSI dimensionality. Then, a tree-structure filter that adaptively incorporates contextual information is constructed. Additionally, an initial classification map is generated using multi-class support vector machines (SVMs, and segment-tree filtering is conducted using this map. Finally, a simple Winner-Take-All (WTA rule is applied to determine the class of each pixel in an HSI based on the maximum probability. The experimental results demonstrate that the proposed method can improve HSI classification accuracy significantly. Furthermore, a comparison between the proposed method and the current state-of-the-art methods, such as Extended Morphological Profiles (EMPs, Guided Filtering (GF, and Markov Random Fields (MRFs, suggests that our method is both competitive and robust.

  19. A Robust Sparse Representation Model for Hyperspectral Image Classification.

    Science.gov (United States)

    Huang, Shaoguang; Zhang, Hongyan; Pižurica, Aleksandra

    2017-09-12

    Sparse representation has been extensively investigated for hyperspectral image (HSI) classification and led to substantial improvements in the performance over the traditional methods, such as support vector machine (SVM). However, the existing sparsity-based classification methods typically assume Gaussian noise, neglecting the fact that HSIs are often corrupted by different types of noise in practice. In this paper, we develop a robust classification model that admits realistic mixed noise, which includes Gaussian noise and sparse noise. We combine a model for mixed noise with a prior on the representation coefficients of input data within a unified framework, which produces three kinds of robust classification methods based on sparse representation classification (SRC), joint SRC and joint SRC on a super-pixels level. Experimental results on simulated and real data demonstrate the effectiveness of the proposed method and clear benefits from the introduced mixed-noise model.

  20. Classification of objects on hyperspectral images — further method development

    DEFF Research Database (Denmark)

    Kucheryavskiy, Sergey V.; Williams, Paul James

    Classification of objects (such as tablets, cereals, fruits, etc.) is one of the very important applications of hyperspectral imaging and image analysis. Quite often, a hyperspectral image is represented and analyzed just as a bunch of spectra without taking into account spatial information about...... and competitors will be shown based on several real cases. References: [1] S. Kucheryavskiy. A new approach for discrimination of objects on hyperspectral images. Chemometrics and Intelligent Laboratory Systems 120, 126 (2013)....

  1. Hyperspectral image classification based on spatial and spectral features and sparse representation

    Institute of Scientific and Technical Information of China (English)

    Yang Jing-Hui; Wang Li-Guo; Qian Jin-Xi

    2014-01-01

    To minimize the low classification accuracy and low utilization of spatial information in traditional hyperspectral image classification methods, we propose a new hyperspectral image classification method, which is based on the Gabor spatial texture features and nonparametric weighted spectral features, and the sparse representation classification method (Gabor–NWSF and SRC), abbreviated GNWSF–SRC. The proposed (GNWSF–SRC) method first combines the Gabor spatial features and nonparametric weighted spectral features to describe the hyperspectral image, and then applies the sparse representation method. Finally, the classification is obtained by analyzing the reconstruction error. We use the proposed method to process two typical hyperspectral data sets with different percentages of training samples. Theoretical analysis and simulation demonstrate that the proposed method improves the classification accuracy and Kappa coefficient compared with traditional classification methods and achieves better classification performance.

  2. Classification of maize kernels using NIR hyperspectral imaging.

    Science.gov (United States)

    Williams, Paul J; Kucheryavskiy, Sergey

    2016-10-15

    NIR hyperspectral imaging was evaluated to classify maize kernels of three hardness categories: hard, medium and soft. Two approaches, pixel-wise and object-wise, were investigated to group kernels according to hardness. The pixel-wise classification assigned a class to every pixel from individual kernels and did not give acceptable results because of high misclassification. However by using a predefined threshold and classifying entire kernels based on the number of correctly predicted pixels, improved results were achieved (sensitivity and specificity of 0.75 and 0.97). Object-wise classification was performed using two methods for feature extraction - score histograms and mean spectra. The model based on score histograms performed better for hard kernel classification (sensitivity and specificity of 0.93 and 0.97), while that of mean spectra gave better results for medium kernels (sensitivity and specificity of 0.95 and 0.93). Both feature extraction methods can be recommended for classification of maize kernels on production scale.

  3. Sparse Representation Based Binary Hypothesis Model for Hyperspectral Image Classification

    Directory of Open Access Journals (Sweden)

    Yidong Tang

    2016-01-01

    Full Text Available The sparse representation based classifier (SRC and its kernel version (KSRC have been employed for hyperspectral image (HSI classification. However, the state-of-the-art SRC often aims at extended surface objects with linear mixture in smooth scene and assumes that the number of classes is given. Considering the small target with complex background, a sparse representation based binary hypothesis (SRBBH model is established in this paper. In this model, a query pixel is represented in two ways, which are, respectively, by background dictionary and by union dictionary. The background dictionary is composed of samples selected from the local dual concentric window centered at the query pixel. Thus, for each pixel the classification issue becomes an adaptive multiclass classification problem, where only the number of desired classes is required. Furthermore, the kernel method is employed to improve the interclass separability. In kernel space, the coding vector is obtained by using kernel-based orthogonal matching pursuit (KOMP algorithm. Then the query pixel can be labeled by the characteristics of the coding vectors. Instead of directly using the reconstruction residuals, the different impacts the background dictionary and union dictionary have on reconstruction are used for validation and classification. It enhances the discrimination and hence improves the performance.

  4. Mitigating noise in global manifold coordinates for hyperspectral image classification

    Science.gov (United States)

    Jin, Can; Bachmann, Charles M.

    2016-09-01

    Over the past decade, manifold and graph representations of hyperspectral imagery (HSI) have been explored widely in HSI applications. Among many data-driven approaches to deriving manifold coordinate representations including Isometric Mapping (ISOMAP), Local Linear Embedding (LLE), Laplacian Eigenmaps (LE), and Diffusion Kernels (DK), ISOMAP is the only global method that well represents the large scale nonlinear geometric structure of the data. In recent years, methods such as ENH-ISOMAP as well as its parallel computing accelerations makes ISOMAP practical for hyperspectral image dimensionality reduction. However, the noise problem in these methods has not been well addressed, which is critical to classification accuracy based on the manifold coordinates derived from these methods. While standard linear techniques to reduce the effects of noise can be applied as a preliminary step, these are based on global statistics and are applied globally across the entire data set, resulting in the risk of losing subtle nonlinear features before classification. To solve this problem, in this paper, we explore several approaches to modeling and mitigating noise in HSI in a local sense to improve the performance of the ENH-ISOMAP algorithm, aiming to reduce the noise effect on the manifold representations of the HSI. A new method to split data into local spectral subsets is introduced. Based on the local spectral subsets obtained with this method, a local noise model guided landmark selection scheme is proposed. In addition, a new robust adaptive neighborhood method using intrinsic dimensionality information to construct the k-Nearest Neighbor graph is introduced to increase the fidelity of the graph, based on the same framework of local spectral subsetting. The improved algorithm produces manifold coordinates with less noise, and shows a better classification accuracy using k-Nearest Neighbor classifier.

  5. [Hyperspectral image classification based on 3-D gabor filter and support vector machines].

    Science.gov (United States)

    Feng, Xiao; Xiao, Peng-feng; Li, Qi; Liu, Xiao-xi; Wu, Xiao-cui

    2014-08-01

    A three-dimensional Gabor filter was developed for classification of hyperspectral remote sensing image. This method is based on the characteristics of hyperspectral image and the principle of texture extraction with 2-D Gabor filters. Three-dimensional Gabor filter is able to filter all the bands of hyperspectral image simultaneously, capturing the specific responses in different scales, orientations, and spectral-dependent properties from enormous image information, which greatly reduces the time consumption in hyperspectral image texture extraction, and solve the overlay difficulties of filtered spectrums. Using the designed three-dimensional Gabor filters in different scales and orientations, Hyperion image which covers the typical area of Qi Lian Mountain was processed with full bands to get 26 Gabor texture features and the spatial differences of Gabor feature textures corresponding to each land types were analyzed. On the basis of automatic subspace separation, the dimensions of the hyperspectral image were reduced by band index (BI) method which provides different band combinations for classification in order to search for the optimal magnitude of dimension reduction. Adding three-dimensional Gabor texture features successively according to its discrimination to the given land types, supervised classification was carried out with the classifier support vector machines (SVM). It is shown that the method using three-dimensional Gabor texture features and BI band selection based on automatic subspace separation for hyperspectral image classification can not only reduce dimensions; but also improve the classification accuracy and efficiency of hyperspectral image.

  6. Superpixel-based spectral classification for the detection of head and neck cancer with hyperspectral imaging

    Science.gov (United States)

    Chung, Hyunkoo; Lu, Guolan; Tian, Zhiqiang; Wang, Dongsheng; Chen, Zhuo Georgia; Fei, Baowei

    2016-03-01

    Hyperspectral imaging (HSI) is an emerging imaging modality for medical applications. HSI acquires two dimensional images at various wavelengths. The combination of both spectral and spatial information provides quantitative information for cancer detection and diagnosis. This paper proposes using superpixels, principal component analysis (PCA), and support vector machine (SVM) to distinguish regions of tumor from healthy tissue. The classification method uses 2 principal components decomposed from hyperspectral images and obtains an average sensitivity of 93% and an average specificity of 85% for 11 mice. The hyperspectral imaging technology and classification method can have various applications in cancer research and management.

  7. Spatial context driven manifold learning for hyperspectral image classification

    CSIR Research Space (South Africa)

    Zhang, Y

    2014-06-01

    Full Text Available Manifold learning techniques have demonstrated various levels of success in their ability to represent spectral signature characteristics in hyperspectral imagery. Such images consists of spectral features with very subtle differences and at times...

  8. Classification of organic beef freshness using VNIR hyperspectral imaging.

    Science.gov (United States)

    Crichton, Stuart O J; Kirchner, Sascha M; Porley, Victoria; Retz, Stefanie; von Gersdorff, Gardis; Hensel, Oliver; Weygandt, Martin; Sturm, Barbara

    2017-07-01

    Consumer trust in the food industry is heavily reliant upon accurate labelling of meat products. As such, methods, which can verify whether meat is correctly labelled are of great value to producers, retailers, and consumers. This paper illustrates two approaches to classify between, fresh and frozen thawed, and in a novel manner matured and matured frozen-thawed, as well as fresh and matured beef using the 500-1010nm waveband, captured using hyperspectral imaging, and CIELAB measurements. The results show successful classification based upon CIELAB between 1) fresh and frozen-thawed (CCR=0.93), and 2) fresh and matured (CCR=0.92). With successful classification between matured and matured frozen-thawed beef using the entire spectral range (CCR=1.00). The performance of reduced spectral models is also investigated. Overall it was found that CIELAB co-ordinates can be used for successful classification for all comparisons except between matured and matured frozen-thawed. Biochemical and physical changes of the meat are thoroughly discussed for each condition. Copyright © 2017 Elsevier Ltd. All rights reserved.

  9. A New Methodology for Spectral-Spatial Classification of Hyperspectral Images

    Directory of Open Access Journals (Sweden)

    Zelang Miao

    2016-01-01

    Full Text Available Recent developments in hyperspectral images have heightened the need for advanced classification methods. To reach this goal, this paper proposed an improved spectral-spatial method for hyperspectral image classification. The proposed method mainly consists of three steps. First, four band selection strategies are proposed to utilize the statistical region merging (SRM method to segment the hyperspectral image. The segmentation map is subsequently integrated with the pixel-wise classification method to classify the hyperspectral image. Finally, the final classification result is obtained using the decision fusion rule. Validation tests are performed to evaluate the performance of the proposed approach, and the results indicate that the new proposed approach outperforms the state-of-the-art methods.

  10. USE SATELLITE IMAGES AND IMPROVE THE ACCURACY OF HYPERSPECTRAL IMAGE WITH THE CLASSIFICATION

    Directory of Open Access Journals (Sweden)

    P. Javadi

    2015-12-01

    Full Text Available The best technique to extract information from remotely sensed image is classification. The problem of traditional classification methods is that each pixel is assigned to a single class by presuming all pixels within the image. Mixed pixel classification or spectral unmixing, is a process that extracts the proportions of the pure components of each mixed pixel. This approach is called spectral unmixing. Hyper spectral images have higher spectral resolution than multispectral images. In this paper, pixel-based classification methods such as the spectral angle mapper, maximum likelihood classification and subpixel classification method (linear spectral unmixing were implemented on the AVIRIS hyper spectral images. Then, pixel-based and subpixel based classification algorithms were compared. Also, the capabilities and advantages of spectral linear unmixing method were investigated. The spectral unmixing method that implemented here is an effective technique for classifying a hyperspectral image giving the classification accuracy about 89%. The results of classification when applying on the original images are not good because some of the hyperspectral image bands are subject to absorption and they contain only little signal. So it is necessary to prepare the data at the beginning of the process. The bands can be stored according to their variance. In bands with a high variance, we can distinguish the features from each other in a better mode in order to increase the accuracy of classification. Also, applying the MNF transformation on the hyperspectral images increase the individual classes accuracy of pixel based classification methods as well as unmixing method about 20 percent and 9 percent respectively.

  11. An improved hyperspectral image classification approach based on ISODATA and SKR method

    Science.gov (United States)

    Hong, Pu; Ye, Xiao-feng; Yu, Hui; Zhang, Zhi-jie; Cai, Yu-fei; Tang, Xin; Tang, Wei; Wang, Chensheng

    2016-11-01

    Hyper-spectral images can not only provide spatial information but also a wealth of spectral information. A short list of applications includes environmental mapping, global change research, geological research, wetlands mapping, assessment of trafficability, plant and mineral identification and abundance estimation, crop analysis, and bathymetry. A crucial aspect of hyperspectral image analysis is the identification of materials present in an object or scene being imaged. Classification of a hyperspectral image sequence amounts to identifying which pixels contain various spectrally distinct materials that have been specified by the user. Several techniques for classification of multi-hyperspectral pixels have been used from minimum distance and maximum likelihood classifiers to correlation matched filter-based approaches such as spectral signature matching and the spectral angle mapper. In this paper, an improved hyperspectral images classification algorithm is proposed. In the proposed method, an improved similarity measurement method is applied, in which both the spectrum similarity and space similarity are considered. We use two different weighted matrix to estimate the spectrum similarity and space similarity between two pixels, respectively. And then whether these two pixels represent the same material can be determined. In order to reduce the computational cost the wavelet transform is also applied prior to extract the spectral and space features. The proposed method is tested using hyperspectral imagery collected by the National Aeronautics and Space Administration Jet Propulsion Laboratory. Experimental results the efficiency of this new method on hyperspectral images associated with space object material identification.

  12. Hyperspectral Image Classification Based on the Combination of Spatial-spectral Feature and Sparse Representation

    Directory of Open Access Journals (Sweden)

    YANG Zhaoxia

    2015-07-01

    Full Text Available In order to avoid the problem of being over-dependent on high-dimensional spectral feature in the traditional hyperspectral image classification, a novel approach based on the combination of spatial-spectral feature and sparse representation is proposed in this paper. Firstly, we extract the spatial-spectral feature by reorganizing the local image patch with the first d principal components(PCs into a vector representation, followed by a sorting scheme to make the vector invariant to local image rotation. Secondly, we learn the dictionary through a supervised method, and use it to code the features from test samples afterwards. Finally, we embed the resulting sparse feature coding into the support vector machine(SVM for hyperspectral image classification. Experiments using three hyperspectral data show that the proposed method can effectively improve the classification accuracy comparing with traditional classification methods.

  13. Spectral-spatial hyperspectral image classification using super-pixel-based spatial pyramid representation

    Science.gov (United States)

    Fan, Jiayuan; Tan, Hui Li; Toomik, Maria; Lu, Shijian

    2016-10-01

    Spatial pyramid matching has demonstrated its power for image recognition task by pooling features from spatially increasingly fine sub-regions. Motivated by the concept of feature pooling at multiple pyramid levels, we propose a novel spectral-spatial hyperspectral image classification approach using superpixel-based spatial pyramid representation. This technique first generates multiple superpixel maps by decreasing the superpixel number gradually along with the increased spatial regions for labelled samples. By using every superpixel map, sparse representation of pixels within every spatial region is then computed through local max pooling. Finally, features learned from training samples are aggregated and trained by a support vector machine (SVM) classifier. The proposed spectral-spatial hyperspectral image classification technique has been evaluated on two public hyperspectral datasets, including the Indian Pines image containing 16 different agricultural scene categories with a 20m resolution acquired by AVIRIS and the University of Pavia image containing 9 land-use categories with a 1.3m spatial resolution acquired by the ROSIS-03 sensor. Experimental results show significantly improved performance compared with the state-of-the-art works. The major contributions of this proposed technique include (1) a new spectral-spatial classification approach to generate feature representation for hyperspectral image, (2) a complementary yet effective feature pooling approach, i.e. the superpixel-based spatial pyramid representation that is used for the spatial correlation study, (3) evaluation on two public hyperspectral image datasets with superior image classification performance.

  14. Evaluating the Initialization Methods of Wavelet Networks for Hyperspectral Image Classification

    Science.gov (United States)

    Hsu, Pai-Hui

    2016-06-01

    The idea of using artificial neural network has been proven useful for hyperspectral image classification. However, the high dimensionality of hyperspectral images usually leads to the failure of constructing an effective neural network classifier. To improve the performance of neural network classifier, wavelet-based feature extraction algorithms can be applied to extract useful features for hyperspectral image classification. However, the extracted features with fixed position and dilation parameters of the wavelets provide insufficient characteristics of spectrum. In this study, wavelet networks which integrates the advantages of wavelet-based feature extraction and neural networks classification is proposed for hyperspectral image classification. Wavelet networks is a kind of feed-forward neural networks using wavelets as activation function. Both the position and the dilation parameters of the wavelets are optimized as well as the weights of the network during the training phase. The value of wavelet networks lies in their capabilities of optimizing network weights and extracting essential features simultaneously for hyperspectral images classification. In this study, the influence of the learning rate and momentum term during the network training phase is presented, and several initialization modes of wavelet networks were used to test the performance of wavelet networks.

  15. A Spectral-Texture Kernel-Based Classification Method for Hyperspectral Images

    Directory of Open Access Journals (Sweden)

    Yi Wang

    2016-11-01

    Full Text Available Classification of hyperspectral images always suffers from high dimensionality and very limited labeled samples. Recently, the spectral-spatial classification has attracted considerable attention and can achieve higher classification accuracy and smoother classification maps. In this paper, a novel spectral-spatial classification method for hyperspectral images by using kernel methods is investigated. For a given hyperspectral image, the principle component analysis (PCA transform is first performed. Then, the first principle component of the input image is segmented into non-overlapping homogeneous regions by using the entropy rate superpixel (ERS algorithm. Next, the local spectral histogram model is applied to each homogeneous region to obtain the corresponding texture features. Because this step is performed within each homogenous region, instead of within a fixed-size image window, the obtained local texture features in the image are more accurate, which can effectively benefit the improvement of classification accuracy. In the following step, a contextual spectral-texture kernel is constructed by combining spectral information in the image and the extracted texture information using the linearity property of the kernel methods. Finally, the classification map is achieved by the support vector machines (SVM classifier using the proposed spectral-texture kernel. Experiments on two benchmark airborne hyperspectral datasets demonstrate that our method can effectively improve classification accuracies, even though only a very limited training sample is available. Specifically, our method can achieve from 8.26% to 15.1% higher in terms of overall accuracy than the traditional SVM classifier. The performance of our method was further compared to several state-of-the-art classification methods of hyperspectral images using objective quantitative measures and a visual qualitative evaluation.

  16. UNLABELED SELECTED SAMPLES IN FEATURE EXTRACTION FOR CLASSIFICATION OF HYPERSPECTRAL IMAGES WITH LIMITED TRAINING SAMPLES

    Directory of Open Access Journals (Sweden)

    A. Kianisarkaleh

    2015-12-01

    Full Text Available Feature extraction plays a key role in hyperspectral images classification. Using unlabeled samples, often unlimitedly available, unsupervised and semisupervised feature extraction methods show better performance when limited number of training samples exists. This paper illustrates the importance of selecting appropriate unlabeled samples that used in feature extraction methods. Also proposes a new method for unlabeled samples selection using spectral and spatial information. The proposed method has four parts including: PCA, prior classification, posterior classification and sample selection. As hyperspectral image passes these parts, selected unlabeled samples can be used in arbitrary feature extraction methods. The effectiveness of the proposed unlabeled selected samples in unsupervised and semisupervised feature extraction is demonstrated using two real hyperspectral datasets. Results show that through selecting appropriate unlabeled samples, the proposed method can improve the performance of feature extraction methods and increase classification accuracy.

  17. Experimental comparison of support vector machines with random forests for hyperspectral image land cover classification

    Indian Academy of Sciences (India)

    B T Abe; O O Olugbara; T Marwala

    2014-06-01

    The performances of regular support vector machines and random forests are experimentally compared for hyperspectral imaging land cover classification. Special characteristics of hyperspectral imaging dataset present diverse processing problems to be resolved under robust mathematical formalisms such as image classification. As a result, pixel purity index algorithm is used to obtain endmember spectral responses from Indiana pine hyperspectral image dataset. The generalized reduced gradient optimization algorithm is thereafter executed on the research data to estimate fractional abundances in the hyperspectral image and thereby obtain the numeric values for land cover classification. The Waikato environment for knowledge analysis (WEKA) data mining framework is selected as a tool to carry out the classification process by using support vector machines and random forests classifiers. Results show that performance of support vector machines is comparable to that of random forests. This study makes a positive contribution to the problem of land cover classification by exploring generalized reduced gradient method, support vector machines, and random forests to improve producer accuracy and overall classification accuracy. The performance comparison of these classifiers is valuable for a decision maker to consider tradeoffs in method accuracy versus method complexity.

  18. Weighted Chebyshev distance classification method for hyperspectral imaging

    Science.gov (United States)

    Demirci, S.; Erer, I.; Ersoy, O.

    2015-06-01

    The main objective of classification is to partition the surface materials into non-overlapping regions by using some decision rules. For supervised classification, the hyperspectral imagery (HSI) is compared with the reflectance spectra of the material containing similar spectral characteristic. As being a spectral similarity based classification method, prediction of different level of upper and lower spectral boundaries of all classes spectral signatures across spectral bands constitutes the basic principles of the Multi-Scale Vector Tunnel Algorithm (MS-VTA) classification algorithm. The vector tunnel (VT) scaling parameters obtained from means and standard deviations of the class references are used. In this study, MS-VT method is improved and a spectral similarity based technique referred to as Weighted Chebyshev Distance (WCD) method for the supervised classification of HSI is introduced. This is also shown to be equivalent to the use of the WCD in which the weights are chosen as an inverse power of the standard deviation per spectral band. The use of WCD measures in terms of the inverse power of standard deviations and optimization of power parameter constitute the most important side of the study. The algorithms are trained with the same kinds of training sets, and their performances are calculated for the power of the standard deviation. During these studies, various levels of the power parameters are evaluated based on the efficiency of the algorithms for choosing the best values of the weights.

  19. Hyperspectral Image Classification Based on the Weighted Probabilistic Fusion of Multiple Spectral-spatial Features

    Directory of Open Access Journals (Sweden)

    ZHANG Chunsen

    2015-08-01

    Full Text Available A hyperspectral images classification method based on the weighted probabilistic fusion of multiple spectral-spatial features was proposed in this paper. First, the minimum noise fraction (MNF approach was employed to reduce the dimension of hyperspectral image and extract the spectral feature from the image, then combined the spectral feature with the texture feature extracted based on gray level co-occurrence matrix (GLCM, the multi-scale morphological feature extracted based on OFC operator and the end member feature extracted based on sequential maximum angle convex cone (SMACC method to form three spectral-spatial features. Afterwards, support vector machine (SVM classifier was used for the classification of each spectral-spatial feature separately. Finally, we established the weighted probabilistic fusion model and applied the model to fuse the SVM outputs for the final classification result. In order to verify the proposed method, the ROSIS and AVIRIS image were used in our experiment and the overall accuracy reached 97.65% and 96.62% separately. The results indicate that the proposed method can not only overcome the limitations of traditional single-feature based hyperspectral image classification, but also be superior to conventional VS-SVM method and probabilistic fusion method. The classification accuracy of hyperspectral images was improved effectively.

  20. Denoising Of Hyperspectral Image

    Directory of Open Access Journals (Sweden)

    Ashumati Dhuppe

    2014-05-01

    Full Text Available The amount of noise included in a Hyperspectral images limits its application and has a negative impact on Hyperspectral image classification, unmixing, target detection, so on. Hyperspectral imaging (HSI systems can acquire both spectral and spatial information of ground surface simultaneously and have been used in a variety of applications such as object detection, material identification, land cover classification etc. In Hyperspectral images, because the noise intensity in different bands is different, to better suppress the noise in the high noise intensity bands & preserve the detailed information in the low noise intensity bands, the denoising strength should be adaptively adjusted with noise intensity in different bands. We propose a Hyperspectral image denoising algorithms employing a spectral spatial adaptive total variation (TV model, in which the spectral noise difference & spatial information differences are both considered in the process of noise reduction.

  1. Hyperspectral remote sensing image classification based on combined SVM and LDA

    Science.gov (United States)

    Zhang, Chunsen; Zheng, Yiwei

    2014-11-01

    This paper presents a novel method for hyperspectral image classification based on the minimum noise fraction (MNF) and an approach combining support vector machine (SVM) and linear discriminant analysis (LDA). A new SVM/LDA algorithm is used for the classification. First, we use MNF method to reduce the dimension and extract features of the image, and then use the SVM/LDA algorithm to transform the extracted features. Next, we train the result of transformation, optimize the parameters through cross-validation and grid search method, then get a optimal hyperspectral image classifier. Finally, we use this classifier to complete classification. In order to verify the proposed method, the AVIRIS Indian Pines image was used. The experimental results show that the proposed method can solve the contradiction between the small amount of samples and high dimension, improve classification accuracy compared to the classical SVM method.

  2. Classification of objects on hyperspectral images — further developments

    DEFF Research Database (Denmark)

    Kucheryavskiy, Sergey V.; Williams, Paul

    Classification of objects (such as tablets, cereals, fruits, etc.) is one of the very important applications of hyperspectral imaging and image analysis. Quite often, a hyperspectral image is represented and analyzed just as a bunch of spectra without taking into account spatial information about...... component space. The method modifications include both the way the principal component space is built as well as the use of new descriptors for the patterns. The comparison of the modified method with its previous version and competitors will be shown on several real cases....

  3. Maize kernel hardness classification by near infrared (NIR) hyperspectral imaging and multivariate data analysis.

    Science.gov (United States)

    Williams, Paul; Geladi, Paul; Fox, Glen; Manley, Marena

    2009-10-27

    The use of near infrared (NIR) hyperspectral imaging and hyperspectral image analysis for distinguishing between hard, intermediate and soft maize kernels from inbred lines was evaluated. NIR hyperspectral images of two sets (12 and 24 kernels) of whole maize kernels were acquired using a Spectral Dimensions MatrixNIR camera with a spectral range of 960-1662 nm and a sisuChema SWIR (short wave infrared) hyperspectral pushbroom imaging system with a spectral range of 1000-2498 nm. Exploratory principal component analysis (PCA) was used on absorbance images to remove background, bad pixels and shading. On the cleaned images, PCA could be used effectively to find histological classes including glassy (hard) and floury (soft) endosperm. PCA illustrated a distinct difference between glassy and floury endosperm along principal component (PC) three on the MatrixNIR and PC two on the sisuChema with two distinguishable clusters. Subsequently partial least squares discriminant analysis (PLS-DA) was applied to build a classification model. The PLS-DA model from the MatrixNIR image (12 kernels) resulted in root mean square error of prediction (RMSEP) value of 0.18. This was repeated on the MatrixNIR image of the 24 kernels which resulted in RMSEP of 0.18. The sisuChema image yielded RMSEP value of 0.29. The reproducible results obtained with the different data sets indicate that the method proposed in this paper has a real potential for future classification uses.

  4. Fusion of Hyperspectral and Vhr Multispectral Image Classifications in Urban Areas

    Science.gov (United States)

    Hervieu, Alexandre; Le Bris, Arnaud; Mallet, Clément

    2016-06-01

    An energetical approach is proposed for classification decision fusion in urban areas using multispectral and hyperspectral imagery at distinct spatial resolutions. Hyperspectral data provides a great ability to discriminate land-cover classes while multispectral data, usually at higher spatial resolution, makes possible a more accurate spatial delineation of the classes. Hence, the aim here is to achieve the most accurate classification maps by taking advantage of both data sources at the decision level: spectral properties of the hyperspectral data and the geometrical resolution of multispectral images. More specifically, the proposed method takes into account probability class membership maps in order to improve the classification fusion process. Such probability maps are available using standard classification techniques such as Random Forests or Support Vector Machines. Classification probability maps are integrated into an energy framework where minimization of a given energy leads to better classification maps. The energy is minimized using a graph-cut method called quadratic pseudo-boolean optimization (QPBO) with ?-expansion. A first model is proposed that gives satisfactory results in terms of classification results and visual interpretation. This model is compared to a standard Potts models adapted to the considered problem. Finally, the model is enhanced by integrating the spatial contrast observed in the data source of higher spatial resolution (i.e., the multispectral image). Obtained results using the proposed energetical decision fusion process are shown on two urban multispectral/hyperspectral datasets. 2-3% improvement is noticed with respect to a Potts formulation and 3-8% compared to a single hyperspectral-based classification.

  5. Spatial and Spectral Nonparametric Linear Feature Extraction Method for Hyperspectral Image Classification

    Directory of Open Access Journals (Sweden)

    Jinn-Min Yang

    2016-11-01

    Full Text Available Feature extraction (FE or dimensionality reduction (DR plays quite an important role in the field of pattern recognition. Feature extraction aims to reduce the dimensionality of the high-dimensional dataset to enhance the classification accuracy and foster the classification speed, particularly when the training sample size is small, namely the small sample size (SSS problem. Remotely sensed hyperspectral images (HSIs are often with hundreds of measured features (bands which potentially provides more accurate and detailed information for classification, but it generally needs more samples to estimate parameters to achieve a satisfactory result. The cost of collecting ground-truth of remotely sensed hyperspectral scene can be considerably difficult and expensive. Therefore, FE techniques have been an important part for hyperspectral image classification. Unlike lots of feature extraction methods are based only on the spectral (band information of the training samples, some feature extraction methods integrating both spatial and spectral information of training samples show more effective results in recent years. Spatial contexture information has been proven to be useful to improve the HSI data representation and to increase classification accuracy. In this paper, we propose a spatial and spectral nonparametric linear feature extraction method for hyperspectral image classification. The spatial and spectral information is extracted for each training sample and used to design the within-class and between-class scatter matrices for constructing the feature extraction model. The experimental results on one benchmark hyperspectral image demonstrate that the proposed method obtains stable and satisfactory results than some existing spectral-based feature extraction.

  6. A multiple classifier system based on Ant-Colony Optimization for Hyperspectral image classification

    Science.gov (United States)

    Tang, Ke; Xie, Li; Li, Guangyao

    2017-01-01

    Hyperspectral images which hold a large quantity of land information enables image classification. Traditional classification methods usually works on multispectral images. However, the high dimensionality in feature space influences the accuracy while using these classification algorithms, such as statistical classifiers or decision trees. This paper proposes a multiple classifier system (MCS) based on ant colony optimization (ACO) algorithm to improve the classification ability. ACO method has been implemented on multispectral images in researches, but seldom to hyperspectral images. In order to overcome the limitation of ACO method on dealing with high dimensionality, MCS is introduced to combine the outputs of each single ACO classifier based on the credibility of rules. Mutual information is applied to discretizing features from the data set and provides the criterion of band selection and band grouping algorithms. The performance of the proposed method is validated with ROSIS Pavia data set, and compared to k-nearest neighbour (KNN) algorithm. Experimental results prove that the proposed method is feasible to classify hyperspectral images.

  7. Improving SVDD classification performance on hyperspectral images via correlation based ensemble technique

    Science.gov (United States)

    Uslu, Faruk Sukru; Binol, Hamidullah; Ilarslan, Mustafa; Bal, Abdullah

    2017-02-01

    Support Vector Data Description (SVDD) is a nonparametric and powerful method for target detection and classification. The SVDD constructs a minimum hypersphere enclosing the target objects as much as possible. It has advantages of sparsity, good generalization and using kernel machines. In many studies, different methods have been offered in order to improve the performance of the SVDD. In this paper, we have presented ensemble methods to improve classification performance of the SVDD in remotely sensed hyperspectral imagery (HSI) data. Among various ensemble approaches we have selected bagging technique for training data set with different combinations. As a novel technique for weighting we have proposed a correlation based weight coefficients assignment. In this technique, correlation between each bagged classifier is calculated to give coefficients to weighted combinators. To verify the improvement performance, two hyperspectral images are processed for classification purpose. The obtained results show that the ensemble SVDD has been found to be significantly better than conventional SVDD in terms of classification accuracy.

  8. Incremental Classification Algorithm of Hyperspectral Remote Sensing Images Based on Spectral-spatial Information

    Directory of Open Access Journals (Sweden)

    WANG Junshu

    2015-09-01

    Full Text Available An incremental classification algorithm INC_SPEC_MPext was proposed for hyperspectral remote sensing images based on spectral and spatial information. The spatial information was extracted by building morphological profiles based on several principle components of hyperspectral image. The morphological profiles were combined together in extended morphological profiles (MPext. Combine spectral and MPext to enrich knowledge and utilize the useful information of unlabeled data at the most extent to optimize the classifier. Pick out high confidence data and add to training set, then retrain the classifier with augmented training set to predict the rest samples. The process was performed iteratively. The proposed algorithm was tested on AVIRIS Indian Pines and Hyperion EO-1 Botswana data, which take on different covers, and experimental results show low classification cost and significant improvements in terms of accuracies and Kappa coefficient under limited training samples compared with the classification results based on spectral, MPext and the combination of sepctral and MPext.

  9. Semisupervised classification for hyperspectral image based on multi-decision labeling and deep feature learning

    Science.gov (United States)

    Ma, Xiaorui; Wang, Hongyu; Wang, Jie

    2016-10-01

    Semisupervised learning is widely used in hyperspectral image classification to deal with the limited training samples, however, some more information of hyperspectral image should be further explored. In this paper, a novel semisupervised classification based on multi-decision labeling and deep feature learning is presented to exploit and utilize as much information as possible to realize the classification task. First, the proposed method takes two decisions to pre-label each unlabeled sample: local decision based on weighted neighborhood information is made by the surrounding samples, and global decision based on deep learning is performed by the most similar training samples. Then, some unlabeled ones with high confidence are selected to extent the training set. Finally, self decision, which depends on the self features exploited by deep learning, is employed on the updated training set to extract spectral-spatial features and produce classification map. Experimental results with real data indicate that it is an effective and promising semisupervised classification method for hyperspectral image.

  10. a Two-Step Decision Fusion Strategy: Application to Hyperspectral and Multispectral Images for Urban Classification

    Science.gov (United States)

    Ouerghemmi, W.; Le Bris, A.; Chehata, N.; Mallet, C.

    2017-05-01

    Very high spatial resolution multispectral images and lower spatial resolution hyperspectral images are complementary sources for urban object classification. The first enables a fine delineation of objects, while the second can better discriminate classes and consider richer land cover semantics. This paper presents a decision fusion scheme taking advantage of both sources classification maps, to produce a better classification map. The proposed method aims at dealing with both semantic and spatial uncertainties and consists in two steps. First, class membership maps are merged at pixel level. Several fusion rules are considered and compared in this study. Secondly, classification is obtained from a global regularization of a graphical model, involving a fit-to-data term related to class membership measures and an image based contrast sensitive regularization term. Results are presented on three datasets. The classification accuracy is improved up to 5 %, with comparison to the best single source classification accuracy.

  11. Parallel multilayer perceptron neural network used for hyperspectral image classification

    Science.gov (United States)

    Garcia-Salgado, Beatriz P.; Ponomaryov, Volodymyr I.; Robles-Gonzalez, Marco A.

    2016-04-01

    This study is focused on time optimization for the classification problem presenting a comparison of five Artificial Neural Network Multilayer Perceptron (ANN-MLP) architectures. We use the Artificial Neural Network (ANN) because it allows to recognize patterns in data in a lower time rate. Time and classification accuracy are taken into account together for the comparison. According to time comparison, two paradigms in the computational field for each ANN-MLP architecture are analysed with three schemes. Firstly, sequential programming is applied by using a single CPU core. Secondly, parallel programming is employed over a multi-core CPU architecture. Finally, a programming model running on GPU architecture is implemented. Furthermore, the classification accuracy is compared between the proposed five ANN-MLP architectures and a state-of.the-art Support Vector Machine (SVM) with three classification frames: 50%,60% and 70% of the data set's observations are randomly selected to train the classifiers. Also, a visual comparison of the classified results is presented. The Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) criteria are also calculated to characterise visual perception. The images employed were acquired by the Airborne Visible/Infrared Imaging Spectrometer (AVIRIS), the Reflective Optics System Imaging Spectrometer (ROSIS) and the Hyperion sensor.

  12. Subspace Learning via Local Probability Distribution for Hyperspectral Image Classification

    Directory of Open Access Journals (Sweden)

    Huiwu Luo

    2015-01-01

    Full Text Available The computational procedure of hyperspectral image (HSI is extremely complex, not only due to the high dimensional information, but also due to the highly correlated data structure. The need of effective processing and analyzing of HSI has met many difficulties. It has been evidenced that dimensionality reduction has been found to be a powerful tool for high dimensional data analysis. Local Fisher’s liner discriminant analysis (LFDA is an effective method to treat HSI processing. In this paper, a novel approach, called PD-LFDA, is proposed to overcome the weakness of LFDA. PD-LFDA emphasizes the probability distribution (PD in LFDA, where the maximum distance is replaced with local variance for the construction of weight matrix and the class prior probability is applied to compute the affinity matrix. The proposed approach increases the discriminant ability of the transformed features in low dimensional space. Experimental results on Indian Pines 1992 data indicate that the proposed approach significantly outperforms the traditional alternatives.

  13. A multiple criteria-based spectral partitioning method for remotely sensed hyperspectral image classification

    Science.gov (United States)

    Liu, Yi; Li, Jun; Plaza, Antonio; Sun, Yanli

    2016-10-01

    Hyperspectral remote sensing offers a powerful tool in many different application contexts. The imbalance between the high dimensionality of the data and the limited availability of training samples calls for the need to perform dimensionality reduction in practice. Among traditional dimensionality reduction techniques, feature extraction is one of the most widely used approaches due to its flexibility to transform the original spectral information into a subspace. In turn, band selection is important when the application requires preserving the original spectral information (especially the physically meaningful information) for the interpretation of the hyperspectral scene. In the case of hyperspectral image classification, both techniques need to discard most of the original features/bands in order to perform the classification using a feature set with much lower dimensionality. However, the discriminative information that allows a classifier to provide good performance is usually classdependent and the relevant information may live in weak features/bands that are usually discarded or lost through subspace transformation or band selection. As a result, in practice, it is challenging to use either feature extraction or band selection for classification purposes. Relevant lines of attack to address this problem have focused on multiple feature selection aiming at a suitable fusion of diverse features in order to provide relevant information to the classifier. In this paper, we present a new dimensionality reduction technique, called multiple criteria-based spectral partitioning, which is embedded in an ensemble learning framework to perform advanced hyperspectral image classification. Driven by the use of a multiple band priority criteria that is derived from classic band selection techniques, we obtain multiple spectral partitions from the original hyperspectral data that correspond to several band subgroups with much lower spectral dimensionality as compared with

  14. [Classification technique for hyperspectral image based on subspace of bands feature extraction and LS-SVM].

    Science.gov (United States)

    Gao, Heng-zhen; Wan, Jian-wei; Zhu, Zhen-zhen; Wang, Li-bao; Nian, Yong-jian

    2011-05-01

    The present paper proposes a novel hyperspectral image classification algorithm based on LS-SVM (least squares support vector machine). The LS-SVM uses the features extracted from subspace of bands (SOB). The maximum noise fraction (MNF) method is adopted as the feature extraction method. The spectral correlations of the hyperspectral image are used in order to divide the feature space into several SOBs. Then the MNF is used to extract characteristic features of the SOBs. The extracted features are combined into the feature vector for classification. So the strong bands correlation is avoided and the spectral redundancies are reduced. The LS-SVM classifier is adopted, which replaces inequality constraints in SVM by equality constraints. So the computation consumption is reduced and the learning performance is improved. The proposed method optimizes spectral information by feature extraction and reduces the spectral noise. The classifier performance is improved. Experimental results show the superiorities of the proposed algorithm.

  15. [Hyperspectral remote sensing image classification based on radical basis function neural network].

    Science.gov (United States)

    Tan, Kun; Du, Pei-jun

    2008-09-01

    Based on the radial basis function neural network (RBFNN) theory and the specialty of hyperspectral remote sensing data, the effective feature extraction model was designed, and those extracted features were connected to the input layer of RBFNN, finally the classifier based on radial basis function neural network was constructed. The hyperspectral image with 64 bands of OMIS II made by Chinese was experimented, and the case study area was zhongguancun in Beijing. Minimum noise fraction (MNF) was conducted, and the former 20 components were extracted for further processing. The original data (20 dimension) of extraction by MNF, the texture transformation data (20 dimension) extracted from the former 20 components after MNF, and the principal component analysis data (20 dimension) of extraction were combined to 60 dimension. For classification by RBFNN, the sizes of training samples were less than 6.13% of the whole image. That classifier has a simple structure and fast convergence capacity, and can be easily trained. The classification precision of radial basis function neural network classifier is up to 69.27% in contrast with the 51.20% of back propagation neural network (BPNN) and 40. 88% of traditional minimum distance classification (MDC), so RBFNN classifier performs better than the other three classifiers. It proves that RBFNN is of validity in hyperspectral remote sensing classification.

  16. [LLE-SVM classification of apple mealiness based on hyperspectral scattering image].

    Science.gov (United States)

    Zhao, Gui-lin; Zhu, Qi-bing; Huang, Min

    2010-10-01

    Apple mealiness degree is an important factor for its internal quality. hyperspectral scattering, as a promising technique, was investigated for noninvasive measurement of apple mealiness. In the present paper, a locally linear embedding (LLE) coupled with support vector machine (SVM) was proposed to achieve classification because of large number of image data. LLE is a nonlinear lowering dimension method, which reveals the structure of the global nonlinearity by the local linear joint. This method can effectively calculate high-dimensional input data embedded in a low-dimensional space manifold. The dimension reduction of hyperspectral data was classified by SVM. Comparing the LLE-SVM classification method with the traditional SVM classification, the results indicated that the training accuracy obtained with the LLE-SVM was higher than that just with SVM; and the testing accuracy of the classifier changed a little before and after dimensionality reduction, and the range of fluctuation was less than 5%. It is expected that LLE-SVM method would provide an effective classification method for apple mealiness nondestructive detection using hyperspectral scattering image technique.

  17. Spectral-Spatial Classification of Hyperspectral Image Based on Kernel Extreme Learning Machine

    OpenAIRE

    Chen Chen; Wei Li; Hongjun Su; Kui Liu

    2014-01-01

    Extreme learning machine (ELM) is a single-layer feedforward neural network based classifier that has attracted significant attention in computer vision and pattern recognition due to its fast learning speed and strong generalization. In this paper, we propose to integrate spectral-spatial information for hyperspectral image classification and exploit the benefits of using spatial features for the kernel based ELM (KELM) classifier. Specifically, Gabor filtering and multihypothesis (MH) pred...

  18. Classification of hyperspectral remote sensing images based on simulated annealing genetic algorithm and multiple instance learning

    Institute of Scientific and Technical Information of China (English)

    高红民; 周惠; 徐立中; 石爱业

    2014-01-01

    A hybrid feature selection and classification strategy was proposed based on the simulated annealing genetic algorithm and multiple instance learning (MIL). The band selection method was proposed from subspace decomposition, which combines the simulated annealing algorithm with the genetic algorithm in choosing different cross-over and mutation probabilities, as well as mutation individuals. Then MIL was combined with image segmentation, clustering and support vector machine algorithms to classify hyperspectral image. The experimental results show that this proposed method can get high classification accuracy of 93.13%at small training samples and the weaknesses of the conventional methods are overcome.

  19. Individual Tree Detection and Classification with UAV-Based Photogrammetric Point Clouds and Hyperspectral Imaging

    Directory of Open Access Journals (Sweden)

    Olli Nevalainen

    2017-02-01

    Full Text Available Small unmanned aerial vehicle (UAV based remote sensing is a rapidly evolving technology. Novel sensors and methods are entering the market, offering completely new possibilities to carry out remote sensing tasks. Three-dimensional (3D hyperspectral remote sensing is a novel and powerful technology that has recently become available to small UAVs. This study investigated the performance of UAV-based photogrammetry and hyperspectral imaging in individual tree detection and tree species classification in boreal forests. Eleven test sites with 4151 reference trees representing various tree species and developmental stages were collected in June 2014 using a UAV remote sensing system equipped with a frame format hyperspectral camera and an RGB camera in highly variable weather conditions. Dense point clouds were measured photogrammetrically by automatic image matching using high resolution RGB images with a 5 cm point interval. Spectral features were obtained from the hyperspectral image blocks, the large radiometric variation of which was compensated for by using a novel approach based on radiometric block adjustment with the support of in-flight irradiance observations. Spectral and 3D point cloud features were used in the classification experiment with various classifiers. The best results were obtained with Random Forest and Multilayer Perceptron (MLP which both gave 95% overall accuracies and an F-score of 0.93. Accuracy of individual tree identification from the photogrammetric point clouds varied between 40% and 95%, depending on the characteristics of the area. Challenges in reference measurements might also have reduced these numbers. Results were promising, indicating that hyperspectral 3D remote sensing was operational from a UAV platform even in very difficult conditions. These novel methods are expected to provide a powerful tool for automating various environmental close-range remote sensing tasks in the very near future.

  20. Shortwave infrared hyperspectral Imaging for cotton foreign matter classification

    Science.gov (United States)

    Various types of cotton foreign matter seriously reduce the commercial value of cotton lint and further degrade the quality of textile products for consumers. This research was aimed to investigate the potential of a non-contact technique, i.e., liquid crystal tunable filter (LCTF) hyperspectral ima...

  1. Combined Kernel-Based BDT-SMO Classification of Hyperspectral Fused Images

    Directory of Open Access Journals (Sweden)

    Fenghua Huang

    2014-01-01

    Full Text Available To solve the poor generalization and flexibility problems that single kernel SVM classifiers have while classifying combined spectral and spatial features, this paper proposed a solution to improve the classification accuracy and efficiency of hyperspectral fused images: (1 different radial basis kernel functions (RBFs are employed for spectral and textural features, and a new combined radial basis kernel function (CRBF is proposed by combining them in a weighted manner; (2 the binary decision tree-based multiclass SMO (BDT-SMO is used in the classification of hyperspectral fused images; (3 experiments are carried out, where the single radial basis function- (SRBF- based BDT-SMO classifier and the CRBF-based BDT-SMO classifier are used, respectively, to classify the land usages of hyperspectral fused images, and genetic algorithms (GA are used to optimize the kernel parameters of the classifiers. The results show that, compared with SRBF, CRBF-based BDT-SMO classifiers display greater classification accuracy and efficiency.

  2. About Classification Methods Based on Tensor Modelling for Hyperspectral Images

    Directory of Open Access Journals (Sweden)

    Salah Bourennane

    2010-03-01

    Full Text Available Denoising and Dimensionality Reduction (DR are key issue to improve the classifiers efficiency for Hyper spectral images (HSI. The multi-way Wiener filtering recently developed is used, Principal and independent component analysis (PCA; ICA and projection pursuit(PP approaches to DR have been investigated. These matrix algebra methods are applied on vectorized images. Thereof, the spatial rearrangement is lost. To jointly take advantage of the spatial and spectral information, HSI has been recently represented as tensor. Offering multiple ways to decompose data orthogonally, we introduced filtering and DR methods based on multilinear algebra tools. The DR is performed on spectral way using PCA, or PP joint to an orthogonal projection onto a lower subspace dimension of the spatial ways. Weshow the classification improvement using the introduced methods in function to existing methods. This experiment is exemplified using real-world HYDICE data. Multi-way filtering, Dimensionality reduction, matrix and multilinear algebra tools, tensor processing.

  3. Active Learning Algorithms for the Classification of Hyperspectral Sea Ice Images

    Directory of Open Access Journals (Sweden)

    Yanling Han

    2015-01-01

    Full Text Available Sea ice is one of the most critical marine disasters, especially in the polar and high latitude regions. Hyperspectral image is suitable for monitoring the sea ice, which contains continuous spectrum information and has better ability of target recognition. The principal bottleneck for the classification of hyperspectral image is a large number of labeled training samples required. However, the collection of labeled samples is time consuming and costly. In order to solve this problem, we apply the active learning (AL algorithm to hyperspectral sea ice detection which can select the most informative samples. Moreover, we propose a novel investigated AL algorithm based on the evaluation of two criteria: uncertainty and diversity. The uncertainty criterion is based on the difference between the probabilities of the two classes having the highest estimated probabilities, while the diversity criterion is based on a kernel k-means clustering technology. In the experiments of Baffin Bay in northwest Greenland on April 12, 2014, our proposed AL algorithm achieves the highest classification accuracy of 89.327% compared with other AL algorithms and random sampling, while achieving the same classification accuracy, the proposed AL algorithm needs less labeling cost.

  4. Accelerating Relevance-Vector-Machine-Based Classification of Hyperspectral Image with Parallel Computing

    Directory of Open Access Journals (Sweden)

    Chao Dong

    2012-01-01

    Full Text Available Benefiting from the kernel skill and the sparse property, the relevance vector machine (RVM could acquire a sparse solution, with an equivalent generalization ability compared with the support vector machine. The sparse property requires much less time in the prediction, making RVM potential in classifying the large-scale hyperspectral image. However, RVM is not widespread influenced by its slow training procedure. To solve the problem, the classification of the hyperspectral image using RVM is accelerated by the parallel computing technique in this paper. The parallelization is revealed from the aspects of the multiclass strategy, the ensemble of multiple weak classifiers, and the matrix operations. The parallel RVMs are implemented using the C language plus the parallel functions of the linear algebra packages and the message passing interface library. The proposed methods are evaluated by the AVIRIS Indian Pines data set on the Beowulf cluster and the multicore platforms. It shows that the parallel RVMs accelerate the training procedure obviously.

  5. Combined multi-kernel support vector machine and wavelet analysis for hyperspectral remote sensing image classification

    Institute of Scientific and Technical Information of China (English)

    Kun Tan; Peijun Du

    2011-01-01

    @@ Many remote sensing image classifiers are limited in their ability to combine spectral features with spatial features. Multi-kernel classifiers, however, are capable of integrating spectral features with spatial or structural features using multiple kernels and summing them for final outputs. Using a support vector machine (SVM) as classifier, different multi-kernel classifiers are constructed and tested using 64-band Operational Modular Imaging Spectrometer Ⅱ hyperspectral image of Changping Area, Beijing City. Results show that by integrating spectral and wavelet texture information, multi-kernel SVM classifiers can obtain more accurate classification results than sole-kernel SVM classifiers and cross-information SVM kernel classifiers. Moreover, when the multi-kernel SVM classifier is used, the combination of the first four principal components from principal component analysis and wavelet texture provides the highest accuracy (97.06%). Multi-kernel SVM is therefore an effective approach to improve the accuracy of hyperspectral image classification and to expand possibilities for remote sensing image interpretation and application.%Many remote sensing image classifiers are limited in their ability to combine spectral features with spatial features. Multi-kernel classifiers, however, are capable of integrating spectral features with spatial or structural features using multiple kernels and summing them for final outputs. Using a support vector machine (SVM) as classifier, different multi-kernel classifiers are constructed and tested using 64-band Operational Modular Imaging Spectrometer Ⅱ hyperspectral image of Changping Area, Beijing City. Results show that by integrating spectral and wavelet texture information, multi-kernel SVM classifiers can obtain more accurate classification results than sole-kernel SVM classifiers and cross-information SVM kernel classifiers. Moreover, when the multi-kernel SVM classifier is used, the combination of the first four

  6. Deep learning based classification for head and neck cancer detection with hyperspectral imaging in an animal model

    Science.gov (United States)

    Ma, Ling; Lu, Guolan; Wang, Dongsheng; Wang, Xu; Chen, Zhuo Georgia; Muller, Susan; Chen, Amy; Fei, Baowei

    2017-03-01

    Hyperspectral imaging (HSI) is an emerging imaging modality that can provide a noninvasive tool for cancer detection and image-guided surgery. HSI acquires high-resolution images at hundreds of spectral bands, providing big data to differentiating different types of tissue. We proposed a deep learning based method for the detection of head and neck cancer with hyperspectral images. Since the deep learning algorithm can learn the feature hierarchically, the learned features are more discriminative and concise than the handcrafted features. In this study, we adopt convolutional neural networks (CNN) to learn the deep feature of pixels for classifying each pixel into tumor or normal tissue. We evaluated our proposed classification method on the dataset containing hyperspectral images from 12 tumor-bearing mice. Experimental results show that our method achieved an average accuracy of 91.36%. The preliminary study demonstrated that our deep learning method can be applied to hyperspectral images for detecting head and neck tumors in animal models.

  7. Nonparametric feature extraction for classification of hyperspectral images with limited training samples

    Science.gov (United States)

    Kianisarkaleh, Azadeh; Ghassemian, Hassan

    2016-09-01

    Feature extraction plays a crucial role in improvement of hyperspectral images classification. Nonparametric feature extraction methods show better performance compared to parametric ones when distribution of classes is non normal-like. Moreover, they can extract more features than parametric methods do. In this paper, a new nonparametric linear feature extraction method is introduced for classification of hyperspectral images. The proposed method has no free parameter and its novelty can be discussed in two parts. First, neighbor samples are specified by using Parzen window idea for determining local mean. Second, two new weighting functions are used. Samples close to class boundaries will have more weight in the between-class scatter matrix formation and samples close to class mean will have more weight in the within-class scatter matrix formation. The experimental results on three real hyperspectral data sets, Indian Pines, Salinas and Pavia University, demonstrate that the proposed method has better performance in comparison with some other nonparametric and parametric feature extraction methods.

  8. Hyperspectral Image Land Cover Classification Algorithm Based on Spatial-spectral Coordination Embedding

    Directory of Open Access Journals (Sweden)

    HUANG Hong

    2016-08-01

    Full Text Available Aiming at the problem that in hyperspectral image land cover classification, the traditional classification methods just apply the spectral information while they ignore the relationship between the spatial neighbors, a new dimensionality algorithm called spatial-spectral coordination embedding (SSCE and a new classifier called spatial-spectral coordination nearest neighbor (SSCNN were proposed in this paper. Firstly, the proposed method defines a spatial-spectral coordination distance and the distance is applied to the neighbor selection and low-dimensional embedding. Then, it constructs a spatial-spectral neighborhood graph to maintain the manifold structure of the data set, and enhances the aggregation of data through raising weight of the spatial neighbor points to extract the discriminant features. Finally, it uses the SSCNN to classify the reduced dimensional data. Experimental results using PaviaU and Salinas data set show that the proposed method can effectively improve ground objects classification accuracy comparing with traditional spectral classification methods.

  9. Quantitative wavelength analysis and image classification for intraoperative cancer diagnosis with hyperspectral imaging

    Science.gov (United States)

    Lu, Guolan; Qin, Xulei; Wang, Dongsheng; Chen, Zhuo Georgia; Fei, Baowei

    2015-03-01

    Complete surgical removal of tumor tissue is essential for postoperative prognosis after surgery. Intraoperative tumor imaging and visualization are an important step in aiding surgeons to evaluate and resect tumor tissue in real time, thus enabling more complete resection of diseased tissue and better conservation of healthy tissue. As an emerging modality, hyperspectral imaging (HSI) holds great potential for comprehensive and objective intraoperative cancer assessment. In this paper, we explored the possibility of intraoperative tumor detection and visualization during surgery using HSI in the wavelength range of 450 nm - 900 nm in an animal experiment. We proposed a new algorithm for glare removal and cancer detection on surgical hyperspectral images, and detected the tumor margins in five mice with an average sensitivity and specificity of 94.4% and 98.3%, respectively. The hyperspectral imaging and quantification method have the potential to provide an innovative tool for image-guided surgery.

  10. [Hyperspectral remote sensing image classification based on SVM optimized by clonal selection].

    Science.gov (United States)

    Liu, Qing-Jie; Jing, Lin-Hai; Wang, Meng-Fei; Lin, Qi-Zhong

    2013-03-01

    Model selection for support vector machine (SVM) involving kernel and the margin parameter values selection is usually time-consuming, impacts training efficiency of SVM model and final classification accuracies of SVM hyperspectral remote sensing image classifier greatly. Firstly, based on combinatorial optimization theory and cross-validation method, artificial immune clonal selection algorithm is introduced to the optimal selection of SVM (CSSVM) kernel parameter a and margin parameter C to improve the training efficiency of SVM model. Then an experiment of classifying AVIRIS in India Pine site of USA was performed for testing the novel CSSVM, as well as a traditional SVM classifier with general Grid Searching cross-validation method (GSSVM) for comparison. And then, evaluation indexes including SVM model training time, classification overall accuracy (OA) and Kappa index of both CSSVM and GSSVM were all analyzed quantitatively. It is demonstrated that OA of CSSVM on test samples and whole image are 85.1% and 81.58, the differences from that of GSSVM are both within 0.08% respectively; And Kappa indexes reach 0.8213 and 0.7728, the differences from that of GSSVM are both within 0.001; While the ratio of model training time of CSSVM and GSSVM is between 1/6 and 1/10. Therefore, CSSVM is fast and accurate algorithm for hyperspectral image classification and is superior to GSSVM.

  11. Hyperspectral image processing

    CERN Document Server

    Wang, Liguo

    2016-01-01

    Based on the authors’ research, this book introduces the main processing techniques in hyperspectral imaging. In this context, SVM-based classification, distance comparison-based endmember extraction, SVM-based spectral unmixing, spatial attraction model-based sub-pixel mapping, and MAP/POCS-based super-resolution reconstruction are discussed in depth. Readers will gain a comprehensive understanding of these cutting-edge hyperspectral imaging techniques. Researchers and graduate students in fields such as remote sensing, surveying and mapping, geosciences and information systems will benefit from this valuable resource.

  12. Hyperspectral band selection and classification of Hyperion image of Bhitarkanika mangrove ecosystem, eastern India

    Science.gov (United States)

    Ashokkumar, L.; Shanmugam, S.

    2014-10-01

    Tropical mangrove forests along the coast evolve dynamically due to constant changes in the natural ecosystem and ecological cycle. Remote sensing has paved the way for periodic monitoring and conservation of such floristic resources, compared to labour intensive in-situ observations. With the laboratory quality image spectra obtained from hyperspectral image data, species level discrimination in habitats and ecosystems is attainable. One of the essential steps before classification of hyperspectral image data is band selection. It is important to eliminate the redundant bands to mitigate the problems of Hughes effect that are likely to affect further image analysis and classification accuracy. This paper presents a methodology for the selection of appropriate hyperspectral bands from the EO-1 Hyperion image for the identification and mapping of mangrove species and coastal landcover types in the Bhitarkanika coastal forest region, eastern India. Band selection procedure follows class based elimination procedure and the separability of the classes are tested in the band selection process. Individual bands are de-correlated and redundant bands are removed from the bandwise correlation matrix. The percent contribution of class variance in each band is analysed from the factors of PCA component ranking. Spectral bands are selected from the wavelength groups and statistically tested. Further, the band selection procedure is compared with similar techniques (Band Index and Mutual information) for validation. The number of bands in the Hyperion image was reduced from 196 to 88 by the Factor-based ranking approach. Classification was performed by Support Vector Machine approach. It is observed that the proposed Factor-based ranking approach performed well in discriminating the mangrove species and other landcover units compared to the other statistical approaches. The predominant mangrove species Heritiera fomes, Excoecaria agallocha and Cynometra ramiflora are spectral

  13. Recent development of feature extraction and classification multispectral/hyperspectral images: a systematic literature review

    Science.gov (United States)

    Setiyoko, A.; Dharma, I. G. W. S.; Haryanto, T.

    2017-01-01

    Multispectral data and hyperspectral data acquired from satellite sensor have the ability in detecting various objects on the earth ranging from low scale to high scale modeling. These data are increasingly being used to produce geospatial information for rapid analysis by running feature extraction or classification process. Applying the most suited model for this data mining is still challenging because there are issues regarding accuracy and computational cost. This research aim is to develop a better understanding regarding object feature extraction and classification applied for satellite image by systematically reviewing related recent research projects. A method used in this research is based on PRISMA statement. After deriving important points from trusted sources, pixel based and texture-based feature extraction techniques are promising technique to be analyzed more in recent development of feature extraction and classification.

  14. Identification of staphylococcus species with hyperspectral microscope imaging and classification algrorithms

    Science.gov (United States)

    Hyperspectral microscope imaging is presented as a rapid and efficient tool to classify foodborne bacteria species. The spectral data were obtained from five different species of Staphylococcus spp. with a hyperspectral microscope imaging system that provided a maximum of 89 contiguous spectral imag...

  15. Taking advantage of hyperspectral imaging classification of urinary stones against conventional infrared spectroscopy.

    Science.gov (United States)

    Blanco, Francisco; Lumbreras, Felipe; Serrat, Joan; Siener, Roswitha; Serranti, Silvia; Bonifazi, Giuseppe; López-Mesas, Montserrat; Valiente, Manuel

    2014-12-01

    The analysis of urinary stones is mandatory for the best management of the disease after the stone passage in order to prevent further stone episodes. Thus the use of an appropriate methodology for an individualized stone analysis becomes a key factor for giving the patient the most suitable treatment. A recently developed hyperspectral imaging methodology, based on pixel-to-pixel analysis of near-infrared spectral images, is compared to the reference technique in stone analysis, infrared (IR) spectroscopy. The developed classification model yields >90% correct classification rate when compared to IR and is able to precisely locate stone components within the structure of the stone with a 15 µm resolution. Due to the little sample pretreatment, low analysis time, good performance of the model, and the automation of the measurements, they become analyst independent; this methodology can be considered to become a routine analysis for clinical laboratories.

  16. Band selection for hyperspectral image classification using extreme learning machine

    Science.gov (United States)

    Li, Jiaojiao; Kingsdorf, Benjamin; Du, Qian

    2017-05-01

    Extreme learning machine (ELM) is a feedforward neural network with one hidden layer, which is similar to a multilayer perceptron (MLP). To reduce the complexity in the training process of MLP using the traditional backpropagation algorithm, the weights in ELM between input and hidden layers are random variables. The output layer in the ELM is linear, as in a radial basis function neural network (RBFNN), so the output weights can be easily estimated with a least squares solution. It has been demonstrated in our previous work that the computational cost of ELM is much lower than the standard support vector machine (SVM), and a kernel version of ELM can offer comparable performance as SVM. In our previous work, we also investigate the impact of the number of hidden neurons to the performance of ELM. Basically, more hidden neurons are needed if the number of training samples and data dimensionality are large, which results in a very large matrix inversion problem. To avoid handling such a large matrix, we propose to conduct band selection to reduce data dimensionality (i.e., the number of input neurons), thereby reducing network complexity. Experimental results show that ELM using selected bands can yield similar or even better classification accuracy than using all the original bands.

  17. Contextual Detection of Anomalies within Hyperspectral Images

    Science.gov (United States)

    2011-03-01

    Hyperspectral Imagery (HSI), Unsupervised Target Detection, Target Identification, Contextual Anomaly Detection 16. SECURITY CLASSIFICATION OF: 17. LIMITATION...processing. Hyperspectral imaging has a wide range of applications within remote sensing, not limited to terrain classification , environmental monitoring...Johnson, R. J. (2008). Improved feature extraction, feature selection, and identification techniques that create a fast unsupervised hyperspectral

  18. Use of high dimensional model representation in dimensionality reduction: application to hyperspectral image classification

    Science.gov (United States)

    Taşkin, Gülşen

    2016-05-01

    Recently, information extraction from hyperspectral images (HI) has become an attractive research area for many practical applications in earth observation due to the fact that HI provides valuable information with a huge number of spectral bands. In order to process such a huge amount of data in an effective way, traditional methods may not fully provide a satisfactory performance because they do not mostly consider high dimensionality of the data which causes curse of dimensionality also known as Hughes phenomena. In case of supervised classification, a poor generalization performance is achieved as a consequence resulting in availability of limited training samples. Therefore, advance methods accounting for the high dimensionality need to be developed in order to get a good generalization capability. In this work, a method of High Dimensional Model Representation (HDMR) was utilized for dimensionality reduction, and a novel feature selection method was introduced based on global sensitivity analysis. Several implementations were conducted with hyperspectral images in comparison to state-of-art feature selection algorithms in terms of classification accuracy, and the results showed that the proposed method outperforms the other feature selection methods even with all considered classifiers, that are support vector machines, Bayes, and decision tree j48.

  19. Hyperspectral imaging-based classification and wavebands selection for internal defect detection of pickling cucumbers

    Science.gov (United States)

    Hyperspectral imaging is useful for detecting internal defect of pickling cucumbers. The technique, however, is not yet suitable for high-speed online implementation due to the challenges for analyzing large-scale hyperspectral images. This research was aimed to select the optimal wavebands from the...

  20. Hyperspectral imaging with multivariate analysis for technological parameters prediction and classification of muscle foods: A review.

    Science.gov (United States)

    Cheng, Jun-Hu; Nicolai, Bart; Sun, Da-Wen

    2017-01-01

    Muscle foods are very important for a well-balanced daily diet. Due to their perishability and vulnerability, there is a need for quality and safety evaluation of such foods. Hyperspectral imaging (HSI) coupled with multivariate analysis is becoming increasingly popular for the non-destructive, non-invasive, and rapid determination of important quality attributes and the classification of muscle foods. This paper reviews recent advances of application of HSI for predicting some significant muscle foods parameters, including color, tenderness, firmness, springiness, water-holding capacity, drip loss and pH. In addition, algorithms for the rapid classification of muscle foods are also reported and discussed. It will be shown that this technology has great potential to replace traditional analytical methods for predicting various quality parameters and classifying muscle foods. Copyright © 2016 Elsevier Ltd. All rights reserved.

  1. Hyperspectral image preprocessing with bilateral filter for improving the classification accuracy of support vector machines

    Science.gov (United States)

    Sahadevan, Anand S.; Routray, Aurobinda; Das, Bhabani S.; Ahmad, Saquib

    2016-04-01

    Bilateral filter (BF) theory is applied to integrate spatial contextual information into the spectral domain for improving the accuracy of the support vector machine (SVM) classifier. The proposed classification framework is a two-stage process. First, an edge-preserved smoothing is carried out on a hyperspectral image (HSI). Then, the SVM multiclass classifier is applied on the smoothed HSI. One of the advantages of the BF-based implementation is that it considers the spatial as well as spectral closeness for smoothing the HSI. Therefore, the proposed method provides better smoothing in the homogeneous region and preserves the image details, which in turn improves the separability between the classes. The performance of the proposed method is tested using benchmark HSIs obtained from the airborne-visible-infrared-imaging-spectrometer (AVIRIS) and the reflective-optics-system-imaging-spectrometer (ROSIS) sensors. Experimental results demonstrate the effectiveness of the edge-preserved filtering in the classification of the HSI. Average accuracies (with 10% training samples) of the proposed classification framework are 99.04%, 98.11%, and 96.42% for AVIRIS-Salinas, ROSIS-Pavia University, and AVIRIS-Indian Pines images, respectively. Since the proposed method follows a combination of BF and the SVM formulations, it will be quite simple and practical to implement in real applications.

  2. Hyperspectral image analysis. A tutorial

    DEFF Research Database (Denmark)

    Amigo Rubio, Jose Manuel; Babamoradi, Hamid; Elcoroaristizabal Martin, Saioa

    2015-01-01

    This tutorial aims at providing guidelines and practical tools to assist with the analysis of hyperspectral images. Topics like hyperspectral image acquisition, image pre-processing, multivariate exploratory analysis, hyperspectral image resolution, classification and final digital image processi...... to differentiate between several types of plastics by using Near infrared hyperspectral imaging and Partial Least Squares - Discriminant Analysis. Thus, the reader is guided through every single step and oriented in order to adapt those strategies to the user's case....... will be exposed, and some guidelines given and discussed. Due to the broad character of current applications and the vast number of multivariate methods available, this paper has focused on an industrial chemical framework to explain, in a step-wise manner, how to develop a classification methodology...

  3. Spectral and Image Integrated Analysis of Hyperspectral Data for Waxy Corn Seed Variety Classification

    Directory of Open Access Journals (Sweden)

    Xiaoling Yang

    2015-07-01

    Full Text Available The purity of waxy corn seed is a very important index of seed quality. A novel procedure for the classification of corn seed varieties was developed based on the combined spectral, morphological, and texture features extracted from visible and near-infrared (VIS/NIR hyperspectral images. For the purpose of exploration and comparison, images of both sides of corn kernels (150 kernels of each variety were captured and analyzed. The raw spectra were preprocessed with Savitzky-Golay (SG smoothing and derivation. To reduce the dimension of spectral data, the spectral feature vectors were constructed using the successive projections algorithm (SPA. Five morphological features (area, circularity, aspect ratio, roundness, and solidity and eight texture features (energy, contrast, correlation, entropy, and their standard deviations were extracted as appearance character from every corn kernel. Support vector machines (SVM and a partial least squares–discriminant analysis (PLS-DA model were employed to build the classification models for seed varieties classification based on different groups of features. The results demonstrate that combining spectral and appearance characteristic could obtain better classification results. The recognition accuracy achieved in the SVM model (98.2% and 96.3% for germ side and endosperm side, respectively was more satisfactory than in the PLS-DA model. This procedure has the potential for use as a new method for seed purity testing.

  4. Hyperspectral image processing methods

    Science.gov (United States)

    Hyperspectral image processing refers to the use of computer algorithms to extract, store and manipulate both spatial and spectral information contained in hyperspectral images across the visible and near-infrared portion of the electromagnetic spectrum. A typical hyperspectral image processing work...

  5. Spectral-spatial classification of hyperspectral images using trilateral filter and stacked sparse autoencoder

    Science.gov (United States)

    Zhao, Chunhui; Wan, Xiaoqing; Zhao, Genping; Yan, Yiming

    2017-01-01

    A spectral-spatial classification method using a trilateral filter (TF) and stacked sparse autoencoder (SSA) for improving the classification accuracy of hyperspectral image (HSI) is proposed. The operation is carried out in two main stages: edge-preserved smoothing and high-level feature learning. First, a reference image obtained from dual tree complex wavelet transform is adopted in a TF for smoothing the HSI. As expected, the filter not only can effectively attenuate the mixed noise (e.g., Gaussian noise and impulse noise) where the bilateral filter shows poor performance but also can produce useful spectral-spatial features from HSI by considering geometric closeness and photometric similarity between pixels simultaneously. Second, an artificial fish swarm algorithm (AFSA) is first introduced into a SSA, and the proposed deep learning architecture is used to adaptively exploit more abstract and differentiable high-level feature representations from the smoothed HSI, based on the factor that AFSA provides better trade-off among concurrency, search efficiency, and convergence rate compared with gradient descent and back-propagation algorithms in a traditional SSA. Finally, a random forest classifier is utilized to perform supervised fine-tuning and classification. Experimental results on two real HSI data sets demonstrate that the proposed method generates competitive performance compared with those of conventional methods.

  6. Classification of fresh and frozen-thawed pork muscles using visible and near infrared hyperspectral imaging and textural analysis.

    Science.gov (United States)

    Pu, Hongbin; Sun, Da-Wen; Ma, Ji; Cheng, Jun-Hu

    2015-01-01

    The potential of visible and near infrared hyperspectral imaging was investigated as a rapid and nondestructive technique for classifying fresh and frozen-thawed meats by integrating critical spectral and image features extracted from hyperspectral images in the region of 400-1000 nm. Six feature wavelengths (400, 446, 477, 516, 592 and 686 nm) were identified using uninformative variable elimination and successive projections algorithm. Image textural features of the principal component images from hyperspectral images were obtained using histogram statistics (HS), gray level co-occurrence matrix (GLCM) and gray level-gradient co-occurrence matrix (GLGCM). By these spectral and textural features, probabilistic neural network (PNN) models for classification of fresh and frozen-thawed pork meats were established. Compared with the models using the optimum wavelengths only, optimum wavelengths with HS image features, and optimum wavelengths with GLCM image features, the model integrating optimum wavelengths with GLGCM gave the highest classification rate of 93.14% and 90.91% for calibration and validation sets, respectively. Results indicated that the classification accuracy can be improved by combining spectral features with textural features and the fusion of critical spectral and textural features had better potential than single spectral extraction in classifying fresh and frozen-thawed pork meat.

  7. Classification of Salmonella enterica serotypes with selective bands using visible/NIR hyperspectral microscope images.

    Science.gov (United States)

    Eady, M; Park, B

    2016-07-01

    Optical detection of foodborne bacteria such as Salmonella classifies bacteria by analysing spectral data, and has potential for rapid detection. In this experiment hyperspectral microscopy is explored as a means for classifying five Salmonella serotypes. Initially, the microscope collects 89 spectral measurements between 450 and 800 nm. Here, the objective was to develop correct classification of five serotypes with optimal spectral bands selected through multivariate data analysis (MVDA), thus reducing the data processing and storage requirement necessary for practical application in the food industry. An upright digital microscope is equipped with an acousto-optical tuneable filter, electron multiplying charge-coupled device, and metal halide lighting source. Images for each of the five serotypes were collected, and informative bands were identified through a principal component analysis, for four abbreviated spectral ranges containing 3, 7, 12 and 20 spectral bands. The experiment was repeated with an independent repetition and images were collected at each of the reduced band sets, identified by the first repetition. A support vector machine (SVM) was used to classify serotypes. Results showed that with the first repetition, classification accuracy decreased from 99.5% (89 bands) to 84.5% (3 bands), whereas the second repetition showed classification accuracies of 100%, possibly due to a reduction in spectral noise. The support vector machine regression (SVMR) was applied with cross-validation, and had R(2) calibration and validation values >0.922. Although classification accuracies through SVM classification showed that as little as 3 bands were able to classify 100% of the samples, the SVMR shows that the smallest root-mean squared-error values were 0.001 and 0.002 for 20 and 12 bands, respectively, suggesting that the 12 band range collected between 586 and 630 nm is optimal for classifying bacterial serotypes, with only the informative HMI bands selected.

  8. Hyperspectral image analysis. A tutorial

    Energy Technology Data Exchange (ETDEWEB)

    Amigo, José Manuel, E-mail: jmar@food.ku.dk [Spectroscopy and Chemometrics Group, Department of Food Sciences, Faculty of Science, University of Copenhagen, Rolighedsvej 30, Frederiksberg C DK–1958 (Denmark); Babamoradi, Hamid [Spectroscopy and Chemometrics Group, Department of Food Sciences, Faculty of Science, University of Copenhagen, Rolighedsvej 30, Frederiksberg C DK–1958 (Denmark); Elcoroaristizabal, Saioa [Spectroscopy and Chemometrics Group, Department of Food Sciences, Faculty of Science, University of Copenhagen, Rolighedsvej 30, Frederiksberg C DK–1958 (Denmark); Chemical and Environmental Engineering Department, School of Engineering, University of the Basque Country, Alameda de Urquijo s/n, E-48013 Bilbao (Spain)

    2015-10-08

    This tutorial aims at providing guidelines and practical tools to assist with the analysis of hyperspectral images. Topics like hyperspectral image acquisition, image pre-processing, multivariate exploratory analysis, hyperspectral image resolution, classification and final digital image processing will be exposed, and some guidelines given and discussed. Due to the broad character of current applications and the vast number of multivariate methods available, this paper has focused on an industrial chemical framework to explain, in a step-wise manner, how to develop a classification methodology to differentiate between several types of plastics by using Near infrared hyperspectral imaging and Partial Least Squares – Discriminant Analysis. Thus, the reader is guided through every single step and oriented in order to adapt those strategies to the user's case. - Highlights: • Comprehensive tutorial of Hyperspectral Image analysis. • Hierarchical discrimination of six classes of plastics containing flame retardant. • Step by step guidelines to perform class-modeling on hyperspectral images. • Fusion of multivariate data analysis and digital image processing methods. • Promising methodology for real-time detection of plastics containing flame retardant.

  9. Generalization of spectral fidelity with flexible measures for the sparse representation classification of hyperspectral images

    Science.gov (United States)

    Wu, Bo; Zhu, Yong; Huang, Xin; Li, Jiayi

    2016-10-01

    Sparse representation classification (SRC) is becoming a promising tool for hyperspectral image (HSI) classification, where the Euclidean spectral distance (ESD) is widely used to reflect the fidelity between the original and reconstructed signals. In this paper, a generalized model is proposed to extend SRC by characterizing the spectral fidelity with flexible similarity measures. To validate the flexibility, several typical similarity measures-the spectral angle similarity (SAS), spectral information divergence (SID), the structural similarity index measure (SSIM), and the ESD-are included in the generalized model. Furthermore, a general solution based on a gradient descent technique is used to solve the nonlinear optimization problem formulated by the flexible similarity measures. To test the generalized model, two actual HSIs were used, and the experimental results confirm the ability of the proposed model to accommodate the various spectral similarity measures. Performance comparisons with the ESD, SAS, SID, and SSIM criteria were also conducted, and the results consistently show the advantages of the generalized model for HSI classification in terms of overall accuracy and kappa coefficient.

  10. Qualitative properties of roasting defect beans and development of its classification methods by hyperspectral imaging technology.

    Science.gov (United States)

    Cho, Jeong-Seok; Bae, Hyung-Jin; Cho, Byoung-Kwan; Moon, Kwang-Deog

    2017-04-01

    Qualitative properties of roasting defect coffee beans and their classification methods were studied using hyperspectral imaging (HSI). The roasting defect beans were divided into 5 groups: medium roasting (Cont), under developed (RD-1), over roasting (RD-2), interior under developed (RD-3), and interior scorching (RD-4). The following qualitative properties were assayed: browning index (BI), moisture content (MC), chlorogenic acid (CA), trigonelline (TG), and caffeine (CF) content. Their HSI spectra (1000-1700nm) were also analysed to develop the classification methods of roasting defect beans. RD-2 showed the highest BI and the lowest MC, CA, and TG content. The accuracy of classification model of partial least-squares discriminant was 86.2%. The most powerful wavelength to classify the defective beans was approximately 1420nm (related to OH bond). The HSI reflectance values at 1420nm showed similar tendency with MC, enabling the use of this technology to classify the roasting defect beans. Copyright © 2016. Published by Elsevier Ltd.

  11. Region-based geometric active contour for classification using hyperspectral remote sensing images

    Science.gov (United States)

    Yan, Lin

    2011-12-01

    The high spectral resolution of hyperspectral imaging (HSI) systems greatly enhances the capabilities of discrimination, identification and quantification of objects of different materials from remote sensing images, but they also bring challenges to the processing and analysis of HSI data. One issue is the high computation cost and the curse of dimensionality associated with the high dimensions of HSI data. A second issue is how to effectively utilize the information including spectral and spatial information embedded in HSI data. Geometric Active Contour (GAC) is a widely used image segmentation method that utilizes the geometric information of objects within images. One category of GAC models, the region-based GAC models (RGAC), have good potential for remote sensing image processing because they use both spectral and geometry information in images are robust to initial contour placement. These models have been introduced to target extractions and classifications on remote sensing images. However, there are some restrictions on the applications of the RGAC models on remote sensing. First, the heavy involvement of iterative contour evolutions makes GAC applications time-consuming and inconvenient to use. Second, the current RGAC models must be based on a certain distance metric and the performance of RGAC classifiers are restricted by the performance of the employed distance metrics. According to the key features of the RGAC models analyzed in this dissertation, a classification framework is developed for remote sensing image classifications using the RGAC models. This framework allows the RGAC models to be combined with conventional pixel-based classifiers to promote them to spectral-spatial classifiers and also greatly reduces the iterations of contour evolutions. An extended Chan-Vese (ECV) model is proposed that is able to incorporate the widely used distance metrics in remote sensing image processing. A new type of RGAC model, the edge-oriented RGAC model

  12. Sparse Representations of Hyperspectral Images

    KAUST Repository

    Swanson, Robin J.

    2015-11-23

    Hyperspectral image data has long been an important tool for many areas of sci- ence. The addition of spectral data yields significant improvements in areas such as object and image classification, chemical and mineral composition detection, and astronomy. Traditional capture methods for hyperspectral data often require each wavelength to be captured individually, or by sacrificing spatial resolution. Recently there have been significant improvements in snapshot hyperspectral captures using, in particular, compressed sensing methods. As we move to a compressed sensing image formation model the need for strong image priors to shape our reconstruction, as well as sparse basis become more important. Here we compare several several methods for representing hyperspectral images including learned three dimensional dictionaries, sparse convolutional coding, and decomposable nonlocal tensor dictionaries. Addi- tionally, we further explore their parameter space to identify which parameters provide the most faithful and sparse representations.

  13. Automated classification and visualization of healthy and pathological dental tissues based on near-infrared hyper-spectral imaging

    Science.gov (United States)

    Usenik, Peter; Bürmen, Miran; Vrtovec, Tomaž; Fidler, Aleš; Pernuš, Franjo; Likar, Boštjan

    2011-03-01

    Despite major improvements in dental healthcare and technology, dental caries remains one of the most prevalent chronic diseases of modern society. The initial stages of dental caries are characterized by demineralization of enamel crystals, commonly known as white spots which are difficult to diagnose. If detected early enough, such demineralization can be arrested and reversed by non-surgical means through well established dental treatments (fluoride therapy, anti-bacterial therapy, low intensity laser irradiation). Near-infrared (NIR) hyper-spectral imaging is a new promising technique for early detection of demineralization based on distinct spectral features of healthy and pathological dental tissues. In this study, we apply NIR hyper-spectral imaging to classify and visualize healthy and pathological dental tissues including enamel, dentin, calculus, dentin caries, enamel caries and demineralized areas. For this purpose, a standardized teeth database was constructed consisting of 12 extracted human teeth with different degrees of natural dental lesions imaged by NIR hyper-spectral system, X-ray and digital color camera. The color and X-ray images of teeth were presented to a clinical expert for localization and classification of the dental tissues, thereby obtaining the gold standard. Principal component analysis was used for multivariate local modeling of healthy and pathological dental tissues. Finally, the dental tissues were classified by employing multiple discriminant analysis. High agreement was observed between the resulting classification and the gold standard with the classification sensitivity and specificity exceeding 85 % and 97 %, respectively. This study demonstrates that NIR hyper-spectral imaging has considerable diagnostic potential for imaging hard dental tissues.

  14. A Grid Service for Automatic Land Cover Classification Using Hyperspectral Images

    Science.gov (United States)

    Jasso, H.; Shin, P.; Fountain, T.; Pennington, D.; Ding, L.; Cotofana, N.

    2004-12-01

    Hyperspectral images are collected using Airborne Visible/Infrared Imaging Spectrometer (Aviris) optical sensors [1]. 224 contiguous channels are measured across the spectral range, from 400 to 2500 nanometers. We present a system for the automatic classification of land cover using hyperspectral images, and propose an architecture for deploying the system in a grid environment that harnesses distributed file storage and CPU resources for the task. Originally, we ran the following data mining algorithms on a 300x300 image of a section of the Sevilleta National Wildlife Refuge in New Mexico [2]: Maximum Likelihood, Naive Bayes Classifier, Minimum Distance, and Support Vector Machine (SVM). For this, ground truth for 673 pixels was manually collected according to eight possible land covers: river, riparian, agriculture, arid upland, semi-arid upland, barren, pavement, or clouds. The classification accuracies for these algorithms were of 96.4%, 90.9%, 88.4%, and 77.6%, respectively [3]. In this study, we noticed that the slope between adjacent frequencies produces specific patterns across the whole spectrum, giving a good indication of the pixel's land cover type. Wavelet analysis makes these global patterns explicit, by breaking down the signal into variable-sized windows, where long time windows capture low-frequency information and short time windows capture high-frequency information. High frequency information translates to information among close neighbors while low frequency information displays the overall trend of the features. We pre-processed the data using different families of wavelets, resulting in an increase in the performance of the Naive Bayesian Classifier and SVM to 94.2% and 90.1%, respectively. Classification accuracy with SVM was further increased to 97.1 % by modifying the mechanism by which multi-class is achieved using basic two-class SVMs. The original winner-take-all SVM scheme was replaced with a one-against-one scheme, in which k(k-1

  15. Classification of gram-positive and gram-negative foodborne pathogenic bacteria with hyperspectral microscope imaging

    Science.gov (United States)

    Optical method with hyperspectral microscope imaging (HMI) has potential for identification of foodborne pathogenic bacteria from microcolonies rapidly with a cell level. A HMI system that provides both spatial and spectral information could be an effective tool for analyzing spectral characteristic...

  16. Hyperspectral imaging-based wound analysis using mixture-tuned matched filtering classification method

    Science.gov (United States)

    Calin, Mihaela Antonina; Coman, Toma; Parasca, Sorin Viorel; Bercaru, Nicolae; Savastru, Roxana; Manea, Dragos

    2015-04-01

    Hyperspectral imaging is a technology that is beginning to occupy an important place in medical research with good prospects in future clinical applications. We evaluated the role of hyperspectral imaging in association with a mixture-tuned matched filtering method in the characterization of open wounds. The methodology and the processing steps of the hyperspectral image that have been performed in order to obtain the most useful information about the wound are described in detail. Correlations between the hyperspectral image and clinical examination are described, leading to a pattern that permits relative evaluation of the square area of the wound and its different components in comparison with the surrounding normal skin. Our results showed that the described method can identify different types of tissues that are present in the wounded area and can objectively measure their respective abundance, which proves its value in wound characterization. In conclusion, the method that was described in this preliminary case presentation shows promising results, but needs further evaluation in order to become a reliable and useful tool.

  17. Performance portability study of an automatic target detection and classification algorithm for hyperspectral image analysis using OpenCL

    Science.gov (United States)

    Bernabe, Sergio; Igual, Francisco D.; Botella, Guillermo; Garcia, Carlos; Prieto-Matias, Manuel; Plaza, Antonio

    2015-10-01

    Recent advances in heterogeneous high performance computing (HPC) have opened new avenues for demanding remote sensing applications. Perhaps one of the most popular algorithm in target detection and identification is the automatic target detection and classification algorithm (ATDCA) widely used in the hyperspectral image analysis community. Previous research has already investigated the mapping of ATDCA on graphics processing units (GPUs) and field programmable gate arrays (FPGAs), showing impressive speedup factors that allow its exploitation in time-critical scenarios. Based on these studies, our work explores the performance portability of a tuned OpenCL implementation across a range of processing devices including multicore processors, GPUs and other accelerators. This approach differs from previous papers, which focused on achieving the optimal performance on each platform. Here, we are more interested in the following issues: (1) evaluating if a single code written in OpenCL allows us to achieve acceptable performance across all of them, and (2) assessing the gap between our portable OpenCL code and those hand-tuned versions previously investigated. Our study includes the analysis of different tuning techniques that expose data parallelism as well as enable an efficient exploitation of the complex memory hierarchies found in these new heterogeneous devices. Experiments have been conducted using hyperspectral data sets collected by NASA's Airborne Visible Infra- red Imaging Spectrometer (AVIRIS) and the Hyperspectral Digital Imagery Collection Experiment (HYDICE) sensors. To the best of our knowledge, this kind of analysis has not been previously conducted in the hyperspectral imaging processing literature, and in our opinion it is very important in order to really calibrate the possibility of using heterogeneous platforms for efficient hyperspectral imaging processing in real remote sensing missions.

  18. Improved classification and visualization of healthy and pathological hard dental tissues by modeling specular reflections in NIR hyperspectral images

    Science.gov (United States)

    Usenik, Peter; Bürmen, Miran; Fidler, Aleš; Pernuš, Franjo; Likar, Boštjan

    2012-03-01

    Despite major improvements in dental healthcare and technology, dental caries remains one of the most prevalent chronic diseases of modern society. The initial stages of dental caries are characterized by demineralization of enamel crystals, commonly known as white spots, which are difficult to diagnose. Near-infrared (NIR) hyperspectral imaging is a new promising technique for early detection of demineralization which can classify healthy and pathological dental tissues. However, due to non-ideal illumination of the tooth surface the hyperspectral images can exhibit specular reflections, in particular around the edges and the ridges of the teeth. These reflections significantly affect the performance of automated classification and visualization methods. Cross polarized imaging setup can effectively remove the specular reflections, however is due to the complexity and other imaging setup limitations not always possible. In this paper, we propose an alternative approach based on modeling the specular reflections of hard dental tissues, which significantly improves the classification accuracy in the presence of specular reflections. The method was evaluated on five extracted human teeth with corresponding gold standard for 6 different healthy and pathological hard dental tissues including enamel, dentin, calculus, dentin caries, enamel caries and demineralized regions. Principal component analysis (PCA) was used for multivariate local modeling of healthy and pathological dental tissues. The classification was performed by employing multiple discriminant analysis. Based on the obtained results we believe the proposed method can be considered as an effective alternative to the complex cross polarized imaging setups.

  19. Quantitative Hyperspectral Reflectance Imaging

    Directory of Open Access Journals (Sweden)

    Ted A.G. Steemers

    2008-09-01

    Full Text Available Hyperspectral imaging is a non-destructive optical analysis technique that can for instance be used to obtain information from cultural heritage objects unavailable with conventional colour or multi-spectral photography. This technique can be used to distinguish and recognize materials, to enhance the visibility of faint or obscured features, to detect signs of degradation and study the effect of environmental conditions on the object. We describe the basic concept, working principles, construction and performance of a laboratory instrument specifically developed for the analysis of historical documents. The instrument measures calibrated spectral reflectance images at 70 wavelengths ranging from 365 to 1100 nm (near-ultraviolet, visible and near-infrared. By using a wavelength tunable narrow-bandwidth light-source, the light energy used to illuminate the measured object is minimal, so that any light-induced degradation can be excluded. Basic analysis of the hyperspectral data includes a qualitative comparison of the spectral images and the extraction of quantitative data such as mean spectral reflectance curves and statistical information from user-defined regions-of-interest. More sophisticated mathematical feature extraction and classification techniques can be used to map areas on the document, where different types of ink had been applied or where one ink shows various degrees of degradation. The developed quantitative hyperspectral imager is currently in use by the Nationaal Archief (National Archives of The Netherlands to study degradation effects of artificial samples and original documents, exposed in their permanent exhibition area or stored in their deposit rooms.

  20. Feature extraction for target identification and image classification of OMIS hyperspectral image

    Institute of Scientific and Technical Information of China (English)

    DU Pei-jun; TAN Kun; SU Hong-jun

    2009-01-01

    In order to combine feature extraction operations with specific hyperspectrai remote sensing information processing objectives, two aspects of feature extraction were explored. Based on clustering and decision tree algorithm, spectral absorption index (SAI), continuum-removal and derivative spectral analysis were employed to discover characterized spectral features of dif-ferent targets, and decision trees for identifying a specific class and discriminating different classes were generated. By combining support vector machine (SVM) classifier with different feature extraction strategies including principal component analysis (PCA), minimum noise fraction (MNF), grouping PCA, and derivate spectral analysis, the performance of feature extraction approaches in classification was evaluated. The results show that feature extraction by PCA and derivate spectral analysis are effective to OMIS (operational modular imaging spectrometer) image classification using SVM, and SVM outperforms traditional SAM and MLC classifiers for OMIS data.

  1. Comparative analysis of different implementations of a parallel algorithm for automatic target detection and classification of hyperspectral images

    Science.gov (United States)

    Paz, Abel; Plaza, Antonio; Plaza, Javier

    2009-08-01

    Automatic target detection in hyperspectral images is a task that has attracted a lot of attention recently. In the last few years, several algoritms have been developed for this purpose, including the well-known RX algorithm for anomaly detection, or the automatic target detection and classification algorithm (ATDCA), which uses an orthogonal subspace projection (OSP) approach to extract a set of spectrally distinct targets automatically from the input hyperspectral data. Depending on the complexity and dimensionality of the analyzed image scene, the target/anomaly detection process may be computationally very expensive, a fact that limits the possibility of utilizing this process in time-critical applications. In this paper, we develop computationally efficient parallel versions of both the RX and ATDCA algorithms for near real-time exploitation of these algorithms. In the case of ATGP, we use several distance metrics in addition to the OSP approach. The parallel versions are quantitatively compared in terms of target detection accuracy, using hyperspectral data collected by NASA's Airborne Visible Infra-Red Imaging Spectrometer (AVIRIS) over the World Trade Center in New York, five days after the terrorist attack of September 11th, 2001, and also in terms of parallel performance, using a massively Beowulf cluster available at NASA's Goddard Space Flight Center in Maryland.

  2. Classification of corn kernels contaminated with aflatoxins using fluorescence and reflectance hyperspectral images analysis

    Science.gov (United States)

    Zhu, Fengle; Yao, Haibo; Hruska, Zuzana; Kincaid, Russell; Brown, Robert; Bhatnagar, Deepak; Cleveland, Thomas

    2015-05-01

    Aflatoxins are secondary metabolites produced by certain fungal species of the Aspergillus genus. Aflatoxin contamination remains a problem in agricultural products due to its toxic and carcinogenic properties. Conventional chemical methods for aflatoxin detection are time-consuming and destructive. This study employed fluorescence and reflectance visible near-infrared (VNIR) hyperspectral images to classify aflatoxin contaminated corn kernels rapidly and non-destructively. Corn ears were artificially inoculated in the field with toxigenic A. flavus spores at the early dough stage of kernel development. After harvest, a total of 300 kernels were collected from the inoculated ears. Fluorescence hyperspectral imagery with UV excitation and reflectance hyperspectral imagery with halogen illumination were acquired on both endosperm and germ sides of kernels. All kernels were then subjected to chemical analysis individually to determine aflatoxin concentrations. A region of interest (ROI) was created for each kernel to extract averaged spectra. Compared with healthy kernels, fluorescence spectral peaks for contaminated kernels shifted to longer wavelengths with lower intensity, and reflectance values for contaminated kernels were lower with a different spectral shape in 700-800 nm region. Principal component analysis was applied for data compression before classifying kernels into contaminated and healthy based on a 20 ppb threshold utilizing the K-nearest neighbors algorithm. The best overall accuracy achieved was 92.67% for germ side in the fluorescence data analysis. The germ side generally performed better than endosperm side. Fluorescence and reflectance image data achieved similar accuracy.

  3. Comparing techniques for vegetation classification using multi- and hyperspectral images and ancillary environmental data

    NARCIS (Netherlands)

    Sluiter, R; Pebesma, E.J.

    2010-01-01

    This paper evaluates the predictive power of innovative and more conventional statistical classification techniques. We use Landsat 7 Enhanced Thematic Mapper Plus (ETMþ), Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) and airborne imaging spectrometer (HyMap) images

  4. Automated Classification and Correlation of Drill Cores using High-Resolution Hyperspectral Images and Supervised Pattern Classification Algorithms. Applications to Paleoseismology

    Science.gov (United States)

    Ragona, D. E.; Minster, B.; Rockwell, T.; Jasso, H.

    2006-12-01

    The standard methodology to describe, classify and correlate geologic materials in the field or lab rely on physical inspection of samples, sometimes with the assistance of conventional analytical techniques (e. g. XRD, microscopy, particle size analysis). This is commonly both time-consuming and inherently subjective. Many geological materials share identical visible properties (e.g. fine grained materials, alteration minerals) and therefore cannot be mapped using the human eye alone. Recent investigations have shown that ground- based hyperspectral imaging provides an effective method to study and digitally store stratigraphic and structural data from cores or field exposures. Neural networks and Naive Bayesian classifiers supply a variety of well-established techniques towards pattern recognition, especially for data examples with high- dimensionality input-outputs. In this poster, we present a new methodology for automatic mapping of sedimentary stratigraphy in the lab (drill cores, samples) or the field (outcrops, exposures) using short wave infrared (SWIR) hyperspectral images and these two supervised classification algorithms. High-spatial/spectral resolution data from large sediment samples (drill cores) from a paleoseismic excavation site were collected using a portable hyperspectral scanner with 245 continuous channels measured across the 960 to 2404 nm spectral range. The data were corrected for geometric and radiometric distortions and pre-processed to obtain reflectance at each pixel of the images. We built an example set using hundreds of reflectance spectra collected from the sediment core images. The examples were grouped into eight classes corresponding to materials found in the samples. We constructed two additional example sets by computing the 2-norm normalization, the derivative of the smoothed original reflectance examples. Each example set was divided into four subsets: training, training test, verification and validation. A multi

  5. Multiple Spectral-Spatial Classification Approach for Hyperspectral Data

    Science.gov (United States)

    Tarabalka, Yuliya; Benediktsson, Jon Atli; Chanussot, Jocelyn; Tilton, James C.

    2010-01-01

    A .new multiple classifier approach for spectral-spatial classification of hyperspectral images is proposed. Several classifiers are used independently to classify an image. For every pixel, if all the classifiers have assigned this pixel to the same class, the pixel is kept as a marker, i.e., a seed of the spatial region, with the corresponding class label. We propose to use spectral-spatial classifiers at the preliminary step of the marker selection procedure, each of them combining the results of a pixel-wise classification and a segmentation map. Different segmentation methods based on dissimilar principles lead to different classification results. Furthermore, a minimum spanning forest is built, where each tree is rooted on a classification -driven marker and forms a region in the spectral -spatial classification: map. Experimental results are presented for two hyperspectral airborne images. The proposed method significantly improves classification accuracies, when compared to previously proposed classification techniques.

  6. Multipurpose Hyperspectral Imaging System

    Science.gov (United States)

    Mao, Chengye; Smith, David; Lanoue, Mark A.; Poole, Gavin H.; Heitschmidt, Jerry; Martinez, Luis; Windham, William A.; Lawrence, Kurt C.; Park, Bosoon

    2005-01-01

    A hyperspectral imaging system of high spectral and spatial resolution that incorporates several innovative features has been developed to incorporate a focal plane scanner (U.S. Patent 6,166,373). This feature enables the system to be used for both airborne/spaceborne and laboratory hyperspectral imaging with or without relative movement of the imaging system, and it can be used to scan a target of any size as long as the target can be imaged at the focal plane; for example, automated inspection of food items and identification of single-celled organisms. The spectral resolution of this system is greater than that of prior terrestrial multispectral imaging systems. Moreover, unlike prior high-spectral resolution airborne and spaceborne hyperspectral imaging systems, this system does not rely on relative movement of the target and the imaging system to sweep an imaging line across a scene. This compact system (see figure) consists of a front objective mounted at a translation stage with a motorized actuator, and a line-slit imaging spectrograph mounted within a rotary assembly with a rear adaptor to a charged-coupled-device (CCD) camera. Push-broom scanning is carried out by the motorized actuator which can be controlled either manually by an operator or automatically by a computer to drive the line-slit across an image at a focal plane of the front objective. To reduce the cost, the system has been designed to integrate as many as possible off-the-shelf components including the CCD camera and spectrograph. The system has achieved high spectral and spatial resolutions by using a high-quality CCD camera, spectrograph, and front objective lens. Fixtures for attachment of the system to a microscope (U.S. Patent 6,495,818 B1) make it possible to acquire multispectral images of single cells and other microscopic objects.

  7. Improved hyperspectral imaging technologies Project

    Data.gov (United States)

    National Aeronautics and Space Administration — Improved hyperspectral imaging technologies could enable lower-cost analysis for planetary science including atmospheric studies, mineralogical investigations, and...

  8. Classification of Salmonella Enterica serotypes with selective bands using visible/NIR hyperspectral imaging

    Science.gov (United States)

    Optical detection of foodborne bacteria such as Salmonella classifies bacteria by analyzing spectral data, and has potential for rapid detection. In this experiment hyperspectral microscopy is explored as a means for classifying five Salmonella serotypes. Initially, the microscope collects 89 spect...

  9. Airborne Hyperspectral Imaging System

    Science.gov (United States)

    Behar, Alberto E.; Cooper, Moogega; Adler, John; Jacobson, Tobias

    2012-01-01

    A document discusses a hyperspectral imaging instrument package designed to be carried aboard a helicopter. It was developed to map the depths of Greenland's supraglacial lakes. The instrument is capable of telescoping to twice its original length, allowing it to be retracted with the door closed during takeoff and landing, and manually extended in mid-flight. While extended, the instrument platform provides the attached hyperspectral imager a nadir-centered and unobstructed view of the ground. Before flight, the instrument mount is retracted and securely strapped down to existing anchor points on the floor of the helicopter. When the helicopter reaches the destination lake, the door is opened and the instrument mount is manually extended. Power to the instrument package is turned on, and the data acquisition computer is commanded via a serial cable from an onboard user-operated laptop to begin data collection. After data collection is complete, the instrument package is powered down and the mount retracted, allowing the door to be closed in preparation for landing. The present design for the instrument mount consists of a three-segment telescoping cantilever to allow for a sufficient extended length to see around the landing struts and provide a nadir-centered and unobstructed field of view for the hyperspectral imager. This instrument works on the premise that water preferentially absorbs light with longer wavelengths on the red side of the visible spectrum. This property can be exploited in order to remotely determine the depths of bodies of pure freshwater. An imager flying over such a lake receives light scattered from the surface, the bulk of the water column, and from the lake bottom. The strength of absorption of longer-wavelength light depends on the depth of the water column. Through calibration with in situ measurements of the water depths, a depth-determining algorithm may be developed to determine lake depth from these spectral properties of the

  10. Discrimination of grassland species and their classification in botanical families by laboratory scale hyperspectral imaging NIR: preliminary results

    Science.gov (United States)

    The objective of this study was to discriminate by on-line hyperspectral imaging, taxonomic plant families comprised of different grassland species. Plants were collected from semi-natural meadows of the National Apuseni Park, Apuseni Mountains, Gârda area (Romania) according to botanical families. ...

  11. Hyperspectral Imager-Tracker

    Science.gov (United States)

    Agurok, Llya

    2013-01-01

    The Hyperspectral Imager-Tracker (HIT) is a technique for visualization and tracking of low-contrast, fast-moving objects. The HIT architecture is based on an innovative and only recently developed concept in imaging optics. This innovative architecture will give the Light Prescriptions Innovators (LPI) HIT the possibility of simultaneously collecting the spectral band images (hyperspectral cube), IR images, and to operate with high-light-gathering power and high magnification for multiple fast- moving objects. Adaptive Spectral Filtering algorithms will efficiently increase the contrast of low-contrast scenes. The most hazardous parts of a space mission are the first stage of a launch and the last 10 kilometers of the landing trajectory. In general, a close watch on spacecraft operation is required at distances up to 70 km. Tracking at such distances is usually associated with the use of radar, but its milliradian angular resolution translates to 100- m spatial resolution at 70-km distance. With sufficient power, radar can track a spacecraft as a whole object, but will not provide detail in the case of an accident, particularly for small debris in the onemeter range, which can only be achieved optically. It will be important to track the debris, which could disintegrate further into more debris, all the way to the ground. Such fragmentation could cause ballistic predictions, based on observations using high-resolution but narrow-field optics for only the first few seconds of the event, to be inaccurate. No optical imager architecture exists to satisfy NASA requirements. The HIT was developed for space vehicle tracking, in-flight inspection, and in the case of an accident, a detailed recording of the event. The system is a combination of five subsystems: (1) a roving fovea telescope with a wide 30 field of regard; (2) narrow, high-resolution fovea field optics; (3) a Coude optics system for telescope output beam stabilization; (4) a hyperspectral

  12. Hyperspectral Systems Increase Imaging Capabilities

    Science.gov (United States)

    2010-01-01

    In 1983, NASA started developing hyperspectral systems to image in the ultraviolet and infrared wavelengths. In 2001, the first on-orbit hyperspectral imager, Hyperion, was launched aboard the Earth Observing-1 spacecraft. Based on the hyperspectral imaging sensors used in Earth observation satellites, Stennis Space Center engineers and Institute for Technology Development researchers collaborated on a new design that was smaller and used an improved scanner. Featured in Spinoff 2007, the technology is now exclusively licensed by Themis Vision Systems LLC, of Richmond, Virginia, and is widely used in medical and life sciences, defense and security, forensics, and microscopy.

  13. Transferring results from NIR-hyperspectral to NIR-multispectral imaging systems: A filter-based simulation applied to the classification of Arabica and Robusta green coffee.

    Science.gov (United States)

    Calvini, Rosalba; Amigo, Jose Manuel; Ulrici, Alessandro

    2017-05-15

    Due to the differences in terms of both price and quality, the availability of effective instrumentation to discriminate between Arabica and Robusta coffee is extremely important. To this aim, the use of multispectral imaging systems could provide reliable and accurate real-time monitoring at relatively low costs. However, in practice the implementation of multispectral imaging systems is not straightforward: the present work investigates this issue, starting from the outcome of variable selection performed using a hyperspectral system. Multispectral data were simulated considering four commercially available filters matching the selected spectral regions, and used to calculate multivariate classification models with Partial Least Squares-Discriminant Analysis (PLS-DA) and sparse PLS-DA. Proper strategies for the definition of the training set and the selection of the most effective combinations of spectral channels led to satisfactory classification performances (100% classification efficiency in prediction of the test set). Copyright © 2017 Elsevier B.V. All rights reserved.

  14. Hyperspectral data classification improved by minimum spanning forests

    Science.gov (United States)

    Dutra da Silva, Ricardo; Pedrini, Helio

    2016-04-01

    Remote sensing technology has applications in various knowledge domains, such as agriculture, meteorology, land use, environmental monitoring, military surveillance, and mineral exploration. The increasing advances in image acquisition techniques have allowed the generation of large volumes of data at high spectral resolution with several spectral bands representing images collected simultaneously. We propose and evaluate a supervised classification method composed of three stages. Initially, hyperspectral values and entropy information are employed by support vector machines to produce an initial classification. Then, the K-nearest neighbor technique searches for pixels with high probability of being correctly classified. Finally, minimum spanning forests are applied to these pixels to reclassify the image taking spatial restrictions into consideration. Experiments on several hyperspectral images are conducted to show the effectiveness of the proposed method.

  15. Spectral Contextual Classification of Hyperspectral Imagery With Probabilistic Relaxation Labeling.

    Science.gov (United States)

    Kumar, Brajesh; Dikshit, Onkar

    2016-09-29

    In this paper, a spectral-spatial classification framework based on probabilistic relaxation labeling using compatibility coefficients is proposed for hyperspectral images. It is a two-stage classifier that uses maximum a posteriori (MAP) estimation to maximize posterior probabilities of classification map obtained in first stage to incorporate spatial information for better classification accuracy. Two different forms of compatibility coefficients based on correlation and mutual information are used for MAP estimation. The initial probability estimates are obtained from probabilistic support vector machine (SVM) classifier. The combination of SVM with MAP estimation is investigated and compared with benchmark Markov random field and extended morphological profile-based approaches and some other recent methods. The experimental results are presented for three airborne hyperspectral images. The results reveal that incorporation of contextual information with both forms of compatibility coefficients statistically significantly improved SVM results. The compatibility coefficients based on correlation produced the best results among the relaxation methods outperforming many existing methods.

  16. Improving the performance of extreme learning machine for hyperspectral image classification

    Science.gov (United States)

    Li, Jiaojiao; Du, Qian; Li, Wei; Li, Yunsong

    2015-05-01

    Extreme learning machine (ELM) and kernel ELM (KELM) can offer comparable performance as the standard powerful classifier―support vector machine (SVM), but with much lower computational cost due to extremely simple training step. However, their performance may be sensitive to several parameters, such as the number of hidden neurons. An empirical linear relationship between the number of training samples and the number of hidden neurons is proposed. Such a relationship can be easily estimated with two small training sets and extended to large training sets so as to greatly reduce computational cost. Other parameters, such as the steepness parameter in the sigmodal activation function and regularization parameter in the KELM, are also investigated. The experimental results show that classification performance is sensitive to these parameters; fortunately, simple selections will result in suboptimal performance.

  17. Medical hyperspectral imaging: a review

    Science.gov (United States)

    Lu, Guolan; Fei, Baowei

    2014-01-01

    Abstract. Hyperspectral imaging (HSI) is an emerging imaging modality for medical applications, especially in disease diagnosis and image-guided surgery. HSI acquires a three-dimensional dataset called hypercube, with two spatial dimensions and one spectral dimension. Spatially resolved spectral imaging obtained by HSI provides diagnostic information about the tissue physiology, morphology, and composition. This review paper presents an overview of the literature on medical hyperspectral imaging technology and its applications. The aim of the survey is threefold: an introduction for those new to the field, an overview for those working in the field, and a reference for those searching for literature on a specific application. PMID:24441941

  18. Medical hyperspectral imaging: a review.

    Science.gov (United States)

    Lu, Guolan; Fei, Baowei

    2014-01-01

    Hyperspectral imaging (HSI) is an emerging imaging modality for medical applications, especially in disease diagnosis and image-guided surgery. HSI acquires a three-dimensional dataset called hypercube, with two spatial dimensions and one spectral dimension. Spatially resolved spectral imaging obtained by HSI provides diagnostic information about the tissue physiology, morphology, and composition. This review paper presents an overview of the literature on medical hyperspectral imaging technology and its applications. The aim of the survey is threefold: an introduction for those new to the field, an overview for those working in the field, and a reference for those searching for literature on a specific application.

  19. Development of classification models to detect Salmonella Enteritidis and Salmonella Typhimurium found in poultry carcass rinses by visible-near infrared hyperspectral imaging

    Science.gov (United States)

    Seo, Young Wook; Yoon, Seung Chul; Park, Bosoon; Hinton, Arthur; Windham, William R.; Lawrence, Kurt C.

    2013-05-01

    Salmonella is a major cause of foodborne disease outbreaks resulting from the consumption of contaminated food products in the United States. This paper reports the development of a hyperspectral imaging technique for detecting and differentiating two of the most common Salmonella serotypes, Salmonella Enteritidis (SE) and Salmonella Typhimurium (ST), from background microflora that are often found in poultry carcass rinse. Presumptive positive screening of colonies with a traditional direct plating method is a labor intensive and time consuming task. Thus, this paper is concerned with the detection of differences in spectral characteristics among the pure SE, ST, and background microflora grown on brilliant green sulfa (BGS) and xylose lysine tergitol 4 (XLT4) agar media with a spread plating technique. Visible near-infrared hyperspectral imaging, providing the spectral and spatial information unique to each microorganism, was utilized to differentiate SE and ST from the background microflora. A total of 10 classification models, including five machine learning algorithms, each without and with principal component analysis (PCA), were validated and compared to find the best model in classification accuracy. The five machine learning (classification) algorithms used in this study were Mahalanobis distance (MD), k-nearest neighbor (kNN), linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), and support vector machine (SVM). The average classification accuracy of all 10 models on a calibration (or training) set of the pure cultures on BGS agar plates was 98% (Kappa coefficient = 0.95) in determining the presence of SE and/or ST although it was difficult to differentiate between SE and ST. The average classification accuracy of all 10 models on a training set for ST detection on XLT4 agar was over 99% (Kappa coefficient = 0.99) although SE colonies on XLT4 agar were difficult to differentiate from background microflora. The average classification

  20. New generation handheld hyperspectral imager

    Science.gov (United States)

    Wu, Huawen (Owen); Li, Hui; Tang, Shengjun

    2016-10-01

    A miniaturized hyper-spectral imager is enabled with image sensor integrated with dispersing elements in a very compact form factor, removing the need for expensive, moving, bulky and complex optics that have been used in conventional hyper-spectral imagers for decades. The result is a handheld spectral imager that can be installed on miniature UAV drones or conveyor belts in production lines. Eventually, small handhelds can be adapted for use in outpatient medical clinics for point-of-care diagnostics and other in-field applications.

  1. Research of Classification Methods Based on Hyperspectral Image%基于高光谱图像的分类方法研究

    Institute of Scientific and Technical Information of China (English)

    杨仁欣; 杨燕; 原晶晶

    2015-01-01

    随着高光谱图像技术的发展,高光谱图像在众多领域得到了广泛应用。高光谱图像分类是其应用领域的一个重要分支,其中高精度的分类算法则是实现准确分类的前提。高光谱图像分辨率高、波段数多、数据量大等特点给传统分类技术带来了巨大挑战。该文综述了基于高光谱图像的支持向量机分类法、人工神经元网络分类法、决策树分类法、最大似然分类法等监督分类方法以及 K -均值聚类法和迭代自组织方法等非监督分类方法,并结合实际高光谱图像数据给出应用实例。基于不同应用需求,以上两类分类方法均能最大程度地挖掘高光谱图像的图谱信息,从而实现更加准确和精细的模式识别。%With the development of hyperspectral image technology ,hyperspectral image analysis has been widely applied in many fields .The classification of hyperspectral images is an important branch in the whole process .The higher precision is the aim we are pursuing after while the classifica‐tion algorithm is the vital premise .For hyperspectral image ,versus the traditional classification tech‐niques ,the characteristics such as high resolution in spectrum and cube data structure have brought great challenges .Based on extracted feature mentioned in former paper ,the support vector machine , artificial neural network ,decision tree ,maximum likelihood method are discussed ,which belong to supervised state .Meanwhile , the K‐mean clustering method and iterative self‐organizing method w hich belong to unsupervised state are also involved .Finally ,the application examples are also listed to illuminate the theory .Based on different application requirements ,the above of two category classi‐fication methods (supervised and unsupervised) can dig the information to the largest extent and at‐tain more accurate result .

  2. Detection of Honey Adulteration using Hyperspectral Imaging

    NARCIS (Netherlands)

    Shafiee, Sahameh; Polder, Gerrit; Minaei, Saeid; Moghadam-charkari, Nasrolah; Ruth, Van Saskia; Kuś, Piotr M.

    2016-01-01

    This study investigates the application of hyperspectral imaging system and data mining based classifiers for honey adulteration detection. Hyperspectral images from pure and adulterated samples were captured in using a VIS-NIR hyperspectral camera (400 – 1000 nm). After preprocessing the images, fi

  3. Detection of Honey Adulteration using Hyperspectral Imaging

    NARCIS (Netherlands)

    Shafiee, Sahameh; Polder, Gerrit; Minaei, Saeid; Moghadam-charkari, Nasrolah; Ruth, Van Saskia; Kuś, Piotr M.

    2016-01-01

    This study investigates the application of hyperspectral imaging system and data mining based classifiers for honey adulteration detection. Hyperspectral images from pure and adulterated samples were captured in using a VIS-NIR hyperspectral camera (400 – 1000 nm). After preprocessing the images,

  4. Piecewise flat embeddings for hyperspectral image analysis

    Science.gov (United States)

    Hayes, Tyler L.; Meinhold, Renee T.; Hamilton, John F.; Cahill, Nathan D.

    2017-05-01

    Graph-based dimensionality reduction techniques such as Laplacian Eigenmaps (LE), Local Linear Embedding (LLE), Isometric Feature Mapping (ISOMAP), and Kernel Principal Components Analysis (KPCA) have been used in a variety of hyperspectral image analysis applications for generating smooth data embeddings. Recently, Piecewise Flat Embeddings (PFE) were introduced in the computer vision community as a technique for generating piecewise constant embeddings that make data clustering / image segmentation a straightforward process. In this paper, we show how PFE arises by modifying LE, yielding a constrained ℓ1-minimization problem that can be solved iteratively. Using publicly available data, we carry out experiments to illustrate the implications of applying PFE to pixel-based hyperspectral image clustering and classification.

  5. A spectral-spatial kernel-based method for hyperspectral imagery classification

    Science.gov (United States)

    Li, Li; Ge, Hongwei; Gao, Jianqiang

    2017-02-01

    Spectral-based classification methods have gained increasing attention in hyperspectral imagery classification. Nevertheless, the spectral cannot fully represent the inherent spatial distribution of the imagery. In this paper, a spectral-spatial kernel-based method for hyperspectral imagery classification is proposed. Firstly, the spatial feature was extracted by using area median filtering (AMF). Secondly, the result of the AMF was used to construct spatial feature patch according to different window sizes. Finally, using the kernel technique, the spectral feature and the spatial feature were jointly used for the classification through a support vector machine (SVM) formulation. Therefore, for hyperspectral imagery classification, the proposed method was called spectral-spatial kernel-based support vector machine (SSF-SVM). To evaluate the proposed method, experiments are performed on three hyperspectral images. The experimental results show that an improvement is possible with the proposed technique in most of the real world classification problems.

  6. Monitoring, analysis and classification of vegetation and soil data collected by a small and lightweight hyperspectral imaging system

    Science.gov (United States)

    Mönnig, Carsten

    2014-05-01

    The increasing precision of modern farming systems requires a near-real-time monitoring of agricultural crops in order to estimate soil condition, plant health and potential crop yield. For large sized agricultural plots, satellite imagery or aerial surveys can be used at considerable costs and possible time delays of days or even weeks. However, for small to medium sized plots, these monitoring approaches are cost-prohibitive and difficult to assess. Therefore, we propose within the INTERREG IV A-Project SMART INSPECTORS (Smart Aerial Test Rigs with Infrared Spectrometers and Radar), a cost effective, comparably simple approach to support farmers with a small and lightweight hyperspectral imaging system to collect remotely sensed data in spectral bands in between 400 to 1700nm. SMART INSPECTORS includes the whole remote sensing processing chain of small scale remote sensing from sensor construction, data processing and ground truthing for analysis of the results. The sensors are mounted on a remotely controlled (RC) Octocopter, a fixed wing RC airplane as well as on a two-seated Autogyro for larger plots. The high resolution images up to 5cm on the ground include spectra of visible light, near and thermal infrared as well as hyperspectral imagery. The data will be analyzed using remote sensing software and a Geographic Information System (GIS). The soil condition analysis includes soil humidity, temperature and roughness. Furthermore, a radar sensor is envisaged for the detection of geomorphologic, drainage and soil-plant roughness investigation. Plant health control includes drought stress, vegetation health, pest control, growth condition and canopy temperature. Different vegetation and soil indices will help to determine and understand soil conditions and plant traits. Additional investigation might include crop yield estimation of certain crops like apples, strawberries, pasture land, etc. The quality of remotely sensed vegetation data will be tested with

  7. Massively parallel processing of remotely sensed hyperspectral images

    Science.gov (United States)

    Plaza, Javier; Plaza, Antonio; Valencia, David; Paz, Abel

    2009-08-01

    In this paper, we develop several parallel techniques for hyperspectral image processing that have been specifically designed to be run on massively parallel systems. The techniques developed cover the three relevant areas of hyperspectral image processing: 1) spectral mixture analysis, a popular approach to characterize mixed pixels in hyperspectral data addressed in this work via efficient implementation of a morphological algorithm for automatic identification of pure spectral signatures or endmembers from the input data; 2) supervised classification of hyperspectral data using multi-layer perceptron neural networks with back-propagation learning; and 3) automatic target detection in the hyperspectral data using orthogonal subspace projection concepts. The scalability of the proposed parallel techniques is investigated using Barcelona Supercomputing Center's MareNostrum facility, one of the most powerful supercomputers in Europe.

  8. Research on hyperspectral dynamic scene and image sequence simulation

    Science.gov (United States)

    Sun, Dandan; Liu, Fang; Gao, Jiaobo; Sun, Kefeng; Hu, Yu; Li, Yu; Xie, Junhu; Zhang, Lei

    2016-10-01

    This paper presents a simulation method of hyperspectral dynamic scene and image sequence for hyperspectral equipment evaluation and target detection algorithm. Because of high spectral resolution, strong band continuity, anti-interference and other advantages, in recent years, hyperspectral imaging technology has been rapidly developed and is widely used in many areas such as optoelectronic target detection, military defense and remote sensing systems. Digital imaging simulation, as a crucial part of hardware in loop simulation, can be applied to testing and evaluation hyperspectral imaging equipment with lower development cost and shorter development period. Meanwhile, visual simulation can produce a lot of original image data under various conditions for hyperspectral image feature extraction and classification algorithm. Based on radiation physic model and material characteristic parameters this paper proposes a generation method of digital scene. By building multiple sensor models under different bands and different bandwidths, hyperspectral scenes in visible, MWIR, LWIR band, with spectral resolution 0.01μm, 0.05μm and 0.1μm have been simulated in this paper. The final dynamic scenes have high real-time and realistic, with frequency up to 100 HZ. By means of saving all the scene gray data in the same viewpoint image sequence is obtained. The analysis results show whether in the infrared band or the visible band, the grayscale variations of simulated hyperspectral images are consistent with the theoretical analysis results.

  9. Road Classification and Condition Determination Using Hyperspectral Imagery

    Science.gov (United States)

    Mohammadi, M.

    2012-07-01

    Hyperspectral data has remarkable capabilities for automatic identification and mapping of urban surface materials because of its high spectral resolution. It includes a wealth of information which facilitates an understanding of the ground material properties. For identification of road surface materials, information about their relation to hyperspectral sensor measurements is needed. In this study an approach for classification of road surface materials using hyperspectral data is developed. The condition of the road surface materials, in particular asphalt is also investigated. Hyperspectral data with 4m spatial resolution of the city of Ludwigsburg, Germany consisting of 125 bands (wavelength range of 0.4542μm to 2.4846 μm) is used. Different supervised classification methods such as spectral angle mapper are applied based on a spectral library established from field measurements and in-situ inspection. It is observed that using the spectral angle mapper approach with regions of interest is helpful for road surface material identification. Additionally, spectral features are tested using their spectral functions in order to achieve better classification results. Spectral functions such as mean and standard deviation are suitable for discriminating asphalt, concrete and gravel. Different asphalt conditions (good, intermediate and bad) are distinguished using the spectral functions such as mean and image ratio. The mean function gives reliable results. Automatisierte Liegenschaftskarte (ALK) vector data for roads is integrated in order to confine the analysis to roads. Reliable reference spectra are useful in evaluation of classification results for spectrally similar road surface materials. The classification results are assessed using orthophotos and field visits information.

  10. Segmentation in dermatological hyperspectral images: dedicated methods

    OpenAIRE

    Koprowski, Robert; Olczyk, Paweł

    2016-01-01

    Background Segmentation of hyperspectral medical images is one of many image segmentation methods which require profiling. This profiling involves either the adjustment of existing, known image segmentation methods or a proposal of new dedicated methods of hyperspectral image segmentation. Taking into consideration the size of analysed data, the time of analysis is of major importance. Therefore, the authors proposed three new dedicated methods of hyperspectral image segmentation with special...

  11. Unsupervised hyperspectral image analysis using independent component analysis (ICA)

    Energy Technology Data Exchange (ETDEWEB)

    S. S. Chiang; I. W. Ginsberg

    2000-06-30

    In this paper, an ICA-based approach is proposed for hyperspectral image analysis. It can be viewed as a random version of the commonly used linear spectral mixture analysis, in which the abundance fractions in a linear mixture model are considered to be unknown independent signal sources. It does not require the full rank of the separating matrix or orthogonality as most ICA methods do. More importantly, the learning algorithm is designed based on the independency of the material abundance vector rather than the independency of the separating matrix generally used to constrain the standard ICA. As a result, the designed learning algorithm is able to converge to non-orthogonal independent components. This is particularly useful in hyperspectral image analysis since many materials extracted from a hyperspectral image may have similar spectral signatures and may not be orthogonal. The AVIRIS experiments have demonstrated that the proposed ICA provides an effective unsupervised technique for hyperspectral image classification.

  12. Hyperspectral all-sky imaging of auroras.

    Science.gov (United States)

    Sigernes, Fred; Ivanov, Yuriy; Chernouss, Sergey; Trondsen, Trond; Roldugin, Alexey; Fedorenko, Yury; Kozelov, Boris; Kirillov, Andrey; Kornilov, Ilia; Safargaleev, Vladimir; Holmen, Silje; Dyrland, Margit; Lorentzen, Dag; Baddeley, Lisa

    2012-12-03

    A prototype auroral hyperspectral all-sky camera has been constructed and tested. It uses electro-optical tunable filters to image the night sky as a function of wavelength throughout the visible spectrum with no moving mechanical parts. The core optical system includes a new high power all-sky lens with F-number equal to f/1.1. The camera has been tested at the Kjell Henriksen Observatory (KHO) during the auroral season of 2011/2012. It detects all sub classes of aurora above ~½ of the sub visual 1kR green intensity threshold at an exposure time of only one second. Supervised classification of the hyperspectral data shows promise as a new method to process and identify auroral forms.

  13. Classification performance of a block-compressive sensing algorithm for hyperspectral data processing

    Science.gov (United States)

    Arias, Fernando X.; Sierra, Heidy; Arzuaga, Emmanuel

    2016-05-01

    Compressive Sensing is an area of great recent interest for efficient signal acquisition, manipulation and reconstruction tasks in areas where sensor utilization is a scarce and valuable resource. The current work shows that approaches based on this technology can improve the efficiency of manipulation, analysis and storage processes already established for hyperspectral imagery, with little discernible loss in data performance upon reconstruction. We present the results of a comparative analysis of classification performance between a hyperspectral data cube acquired by traditional means, and one obtained through reconstruction from compressively sampled data points. To obtain a broad measure of the classification performance of compressively sensed cubes, we classify a commonly used scene in hyperspectral image processing algorithm evaluation using a set of five classifiers commonly used in hyperspectral image classification. Global accuracy statistics are presented and discussed, as well as class-specific statistical properties of the evaluated data set.

  14. Restricted Bipartite Graphs Based Target Detection for Hyperspectral Image Classification with GFA-LFDA Multi Feature Selection

    Directory of Open Access Journals (Sweden)

    T. Karthikeyan

    2015-06-01

    Full Text Available Hyper spectral imaging has recently become one of the most active research areas in remote sensing. Hyper spectral imagery possesses more spectral information than multispectral imagery because the number of spectral bands in hyper spectral imagery is in the hundreds rather than in the tens. However, the high dimensions of hyper spectral images cause redundancy in spatial-spectral feature domain and consider only spectral and spatial features only and ability of the classifier to excel even as training HSI images are limited. However, unless develop suitable algorithms for target detection or classification of the hyper spectral images data becomes difficult. Therefore, it is becomes essential to consider different features and find exact target detection rate to improve classification rate. In order to overcome this problem in this study presents a novel classification framework for hyper spectral data. Proposed system uses a graph based representation, Restricted Bipartite Graphs (RBG for exact detection of the class values. Before that the feature of the HSI images are selected using the Gaussian Firefly Algorithm (GFA for multiple feature selection and Local-Fisher’s Discriminant Analysis (LFDA based feature projection are performed in a raw spectral-spatial feature space for effective dimensionality reduction. Then RBG is proposed to represent the reduced feature results into graphical manner to solve exact target class matching problem, in hyper spectral imaginary. Classification is performed using the Hybrid Genetic Fuzzy Neural Network (HGFNN, Genetic algorithm is used to optimize the weights of the fuzzifier and the defuzzifier for labeled and unlabeled data samples. Experimentation results show that the proposed GFA-LFDA-RBG-HGFNN method outperforms in terms of the classification accuracy and less misclassification results than traditional methods.

  15. Hyperspectral image reconstruction using RGB color for foodborne pathogen detection on agar plates

    Science.gov (United States)

    Yoon, Seung-Chul; Shin, Tae-Sung; Park, Bosoon; Lawrence, Kurt C.; Heitschmidt, Gerald W.

    2014-03-01

    This paper reports the latest development of a color vision technique for detecting colonies of foodborne pathogens grown on agar plates with a hyperspectral image classification model that was developed using full hyperspectral data. The hyperspectral classification model depended on reflectance spectra measured in the visible and near-infrared spectral range from 400 and 1,000 nm (473 narrow spectral bands). Multivariate regression methods were used to estimate and predict hyperspectral data from RGB color values. The six representative non-O157 Shiga-toxin producing Eschetichia coli (STEC) serogroups (O26, O45, O103, O111, O121, and O145) were grown on Rainbow agar plates. A line-scan pushbroom hyperspectral image sensor was used to scan 36 agar plates grown with pure STEC colonies at each plate. The 36 hyperspectral images of the agar plates were divided in half to create training and test sets. The mean Rsquared value for hyperspectral image estimation was about 0.98 in the spectral range between 400 and 700 nm for linear, quadratic and cubic polynomial regression models and the detection accuracy of the hyperspectral image classification model with the principal component analysis and k-nearest neighbors for the test set was up to 92% (99% with the original hyperspectral images). Thus, the results of the study suggested that color-based detection may be viable as a multispectral imaging solution without much loss of prediction accuracy compared to hyperspectral imaging.

  16. Two Linear Unmixing Algorithms to Recognize Targets Using Supervised Classification and Orthogonal Rotation in Airborne Hyperspectral Images

    Directory of Open Access Journals (Sweden)

    Michael Zheludev

    2012-02-01

    Full Text Available The goal of the paper is to detect pixels that contain targets of known spectra. The target can be present in a sub- or above pixel. Pixels without targets are classified as background pixels. Each pixel is treated via the content of its neighborhood. A pixel whose spectrum is different from its neighborhood is classified as a “suspicious point”. In each suspicious point there is a mix of target(s and background. The main objective in a supervised detection (also called “target detection” is to search for a specific given spectral material (target in hyperspectral imaging (HSI where the spectral signature of the target is known a priori from laboratory measurements. In addition, the fractional abundance of the target is computed. To achieve this we present two linear unmixing algorithms that recognize targets with known (given spectral signatures. The CLUN is based on automatic feature extraction from the target’s spectrum. These features separate the target from the background. The ROTU algorithm is based on embedding the spectra space into a special space by random orthogonal transformation and on the statistical properties of the embedded result. Experimental results demonstrate that the targets’ locations were extracted correctly and these algorithms are robust and efficient.

  17. Multiband and Lossless Compression of Hyperspectral Images

    Directory of Open Access Journals (Sweden)

    Raffaele Pizzolante

    2016-02-01

    Full Text Available Hyperspectral images are widely used in several real-life applications. In this paper, we investigate on the compression of hyperspectral images by considering different aspects, including the optimization of the computational complexity in order to allow implementations on limited hardware (i.e., hyperspectral sensors, etc.. We present an approach that relies on a three-dimensional predictive structure. Our predictive structure, 3D-MBLP, uses one or more previous bands as references to exploit the redundancies among the third dimension. The achieved results are comparable, and often better, with respect to the other state-of-art lossless compression techniques for hyperspectral images.

  18. A new semi-supervised classification strategy combining active learning and spectral unmixing of hyperspectral data

    Science.gov (United States)

    Sun, Yanli; Zhang, Xia; Plaza, Antonio; Li, Jun; Dópido, Inmaculada; Liu, Yi

    2016-10-01

    Hyperspectral remote sensing allows for the detailed analysis of the surface of the Earth by providing high-dimensional images with hundreds of spectral bands. Hyperspectral image classification plays a significant role in hyperspectral image analysis and has been a very active research area in the last few years. In the context of hyperspectral image classification, supervised techniques (which have achieved wide acceptance) must address a difficult task due to the unbalance between the high dimensionality of the data and the limited availability of labeled training samples in real analysis scenarios. While the collection of labeled samples is generally difficult, expensive, and time-consuming, unlabeled samples can be generated in a much easier way. Semi-supervised learning offers an effective solution that can take advantage of both unlabeled and a small amount of labeled samples. Spectral unmixing is another widely used technique in hyperspectral image analysis, developed to retrieve pure spectral components and determine their abundance fractions in mixed pixels. In this work, we propose a method to perform semi-supervised hyperspectral image classification by combining the information retrieved with spectral unmixing and classification. Two kinds of samples that are highly mixed in nature are automatically selected, aiming at finding the most informative unlabeled samples. One kind is given by the samples minimizing the distance between the first two most probable classes by calculating the difference between the two highest abundances. Another kind is given by the samples minimizing the distance between the most probable class and the least probable class, obtained by calculating the difference between the highest and lowest abundances. The effectiveness of the proposed method is evaluated using a real hyperspectral data set collected by the airborne visible infrared imaging spectrometer (AVIRIS) over the Indian Pines region in Northwestern Indiana. In the

  19. Classification of Grassland Successional Stages Using Airborne Hyperspectral Imagery

    Directory of Open Access Journals (Sweden)

    Thomas Möckel

    2014-08-01

    Full Text Available Plant communities differ in their species composition, and, thus, also in their functional trait composition, at different stages in the succession from arable fields to grazed grassland. We examine whether aerial hyperspectral (414–2501 nm remote sensing can be used to discriminate between grazed vegetation belonging to different grassland successional stages. Vascular plant species were recorded in 104.1 m2 plots on the island of Öland (Sweden and the functional properties of the plant species recorded in the plots were characterized in terms of the ground-cover of grasses, specific leaf area and Ellenberg indicator values. Plots were assigned to three different grassland age-classes, representing 5–15, 16–50 and >50 years of grazing management. Partial least squares discriminant analysis models were used to compare classifications based on aerial hyperspectral data with the age-class classification. The remote sensing data successfully classified the plots into age-classes: the overall classification accuracy was higher for a model based on a pre-selected set of wavebands (85%, Kappa statistic value = 0.77 than one using the full set of wavebands (77%, Kappa statistic value = 0.65. Our results show that nutrient availability and grass cover differences between grassland age-classes are detectable by spectral imaging. These techniques may potentially be used for mapping the spatial distribution of grassland habitats at different successional stages.

  20. Infrared upconversion hyperspectral imaging

    DEFF Research Database (Denmark)

    Kehlet, Louis Martinus; Tidemand-Lichtenberg, Peter; Dam, Jeppe Seidelin

    2015-01-01

    conversion process. From this, a sequence of monochromatic images in the 3.2-3.4 mu m range is generated. The imaged object consists of a standard United States Air Force resolution target combined with a polystyrene film, resulting in the presence of both spatial and spectral information in the infrared...

  1. Hyperspectral imaging applied to forensic medicine

    Science.gov (United States)

    Malkoff, Donald B.; Oliver, William R.

    2000-03-01

    Remote sensing techniques now include the use of hyperspectral infrared imaging sensors covering the mid-and- long wave regions of the spectrum. They have found use in military surveillance applications due to their capability for detection and classification of a large variety of both naturally occurring and man-made substances. The images they produce reveal the spatial distributions of spectral patterns that reflect differences in material temperature, texture, and composition. A program is proposed for demonstrating proof-of-concept in using a portable sensor of this type for crime scene investigations. It is anticipated to be useful in discovering and documenting the affects of trauma and/or naturally occurring illnesses, as well as detecting blood spills, tire patterns, toxic chemicals, skin injection sites, blunt traumas to the body, fluid accumulations, congenital biochemical defects, and a host of other conditions and diseases. This approach can significantly enhance capabilities for determining the circumstances of death. Potential users include law enforcement organizations (police, FBI, CIA), medical examiners, hospitals/emergency rooms, and medical laboratories. Many of the image analysis algorithms already in place for hyperspectral remote sensing and crime scene investigations can be applied to the interpretation of data obtained in this program.

  2. Common hyperspectral image database design

    Science.gov (United States)

    Tian, Lixun; Liao, Ningfang; Chai, Ali

    2009-11-01

    This paper is to introduce Common hyperspectral image database with a demand-oriented Database design method (CHIDB), which comprehensively set ground-based spectra, standardized hyperspectral cube, spectral analysis together to meet some applications. The paper presents an integrated approach to retrieving spectral and spatial patterns from remotely sensed imagery using state-of-the-art data mining and advanced database technologies, some data mining ideas and functions were associated into CHIDB to make it more suitable to serve in agriculture, geological and environmental areas. A broad range of data from multiple regions of the electromagnetic spectrum is supported, including ultraviolet, visible, near-infrared, thermal infrared, and fluorescence. CHIDB is based on dotnet framework and designed by MVC architecture including five main functional modules: Data importer/exporter, Image/spectrum Viewer, Data Processor, Parameter Extractor, and On-line Analyzer. The original data were all stored in SQL server2008 for efficient search, query and update, and some advance Spectral image data Processing technology are used such as Parallel processing in C#; Finally an application case is presented in agricultural disease detecting area.

  3. Kernel based subspace projection of hyperspectral images

    DEFF Research Database (Denmark)

    Larsen, Rasmus; Nielsen, Allan Aasbjerg; Arngren, Morten

    In hyperspectral image analysis an exploratory approach to analyse the image data is to conduct subspace projections. As linear projections often fail to capture the underlying structure of the data, we present kernel based subspace projections of PCA and Maximum Autocorrelation Factors (MAF). Th......). The MAF projection exploits the fact that interesting phenomena in images typically exhibit spatial autocorrelation. The analysis is based on nearinfrared hyperspectral images of maize grains demonstrating the superiority of the kernelbased MAF method....

  4. Classification of Herbaceous Vegetation Using Airborne Hyperspectral Imagery

    Directory of Open Access Journals (Sweden)

    Péter Burai

    2015-02-01

    Full Text Available Alkali landscapes hold an extremely fine-scale mosaic of several vegetation types, thus it seems challenging to separate these classes by remote sensing. Our aim was to test the applicability of different image classification methods of hyperspectral data in this complex situation. To reach the highest classification accuracy, we tested traditional image classifiers (maximum likelihood classifier—MLC, machine learning algorithms (support vector machine—SVM, random forest—RF and feature extraction (minimum noise fraction (MNF-transformation on training datasets of different sizes. Digital images were acquired from an AISA EAGLE II hyperspectral sensor of 128 contiguous bands (400–1000 nm, a spectral sampling of 5 nm bandwidth and a ground pixel size of 1 m. For the classification, we established twenty vegetation classes based on the dominant species, canopy height, and total vegetation cover. Image classification was applied to the original and MNF (minimum noise fraction transformed dataset with various training sample sizes between 10 and 30 pixels. In order to select the optimal number of the transformed features, we applied SVM, RF and MLC classification to 2–15 MNF transformed bands. In the case of the original bands, SVM and RF classifiers provided high accuracy irrespective of the number of the training pixels. We found that SVM and RF produced the best accuracy when using the first nine MNF transformed bands; involving further features did not increase classification accuracy. SVM and RF provided high accuracies with the transformed bands, especially in the case of the aggregated groups. Even MLC provided high accuracy with 30 training pixels (80.78%, but the use of a smaller training dataset (10 training pixels significantly reduced the accuracy of classification (52.56%. Our results suggest that in alkali landscapes, the application of SVM is a feasible solution, as it provided the highest accuracies compared to RF and MLC

  5. Dried fruits quality assessment by hyperspectral imaging

    Science.gov (United States)

    Serranti, Silvia; Gargiulo, Aldo; Bonifazi, Giuseppe

    2012-05-01

    Dried fruits products present different market values according to their quality. Such a quality is usually quantified in terms of freshness of the products, as well as presence of contaminants (pieces of shell, husk, and small stones), defects, mould and decays. The combination of these parameters, in terms of relative presence, represent a fundamental set of attributes conditioning dried fruits humans-senses-detectable-attributes (visual appearance, organolectic properties, etc.) and their overall quality in terms of marketable products. Sorting-selection strategies exist but sometimes they fail when a higher degree of detection is required especially if addressed to discriminate between dried fruits of relatively small dimensions and when aiming to perform an "early detection" of pathogen agents responsible of future moulds and decays development. Surface characteristics of dried fruits can be investigated by hyperspectral imaging (HSI). In this paper, specific and "ad hoc" applications addressed to propose quality detection logics, adopting a hyperspectral imaging (HSI) based approach, are described, compared and critically evaluated. Reflectance spectra of selected dried fruits (hazelnuts) of different quality and characterized by the presence of different contaminants and defects have been acquired by a laboratory device equipped with two HSI systems working in two different spectral ranges: visible-near infrared field (400-1000 nm) and near infrared field (1000-1700 nm). The spectra have been processed and results evaluated adopting both a simple and fast wavelength band ratio approach and a more sophisticated classification logic based on principal component (PCA) analysis.

  6. Lesion detection in magnetic resonance brain images by hyperspectral imaging algorithms

    Science.gov (United States)

    Xue, Bai; Wang, Lin; Li, Hsiao-Chi; Chen, Hsian Min; Chang, Chein-I.

    2016-05-01

    Magnetic Resonance (MR) images can be considered as multispectral images so that MR imaging can be processed by multispectral imaging techniques such as maximum likelihood classification. Unfortunately, most multispectral imaging techniques are not particularly designed for target detection. On the other hand, hyperspectral imaging is primarily developed to address subpixel detection, mixed pixel classification for which multispectral imaging is generally not effective. This paper takes advantages of hyperspectral imaging techniques to develop target detection algorithms to find lesions in MR brain images. Since MR images are collected by only three image sequences, T1, T2 and PD, if a hyperspectral imaging technique is used to process MR images it suffers from the issue of insufficient dimensionality. To address this issue, two approaches to nonlinear dimensionality expansion are proposed, nonlinear correlation expansion and nonlinear band ratio expansion. Once dimensionality is expanded hyperspectral imaging algorithms are readily applied. The hyperspectral detection algorithm to be investigated for lesion detection in MR brain is the well-known subpixel target detection algorithm, called Constrained Energy Minimization (CEM). In order to demonstrate the effectiveness of proposed CEM in lesion detection, synthetic images provided by BrainWeb are used for experiments.

  7. Locality-preserving sparse representation-based classification in hyperspectral imagery

    Science.gov (United States)

    Gao, Lianru; Yu, Haoyang; Zhang, Bing; Li, Qingting

    2016-10-01

    This paper proposes to combine locality-preserving projections (LPP) and sparse representation (SR) for hyperspectral image classification. The LPP is first used to reduce the dimensionality of all the training and testing data by finding the optimal linear approximations to the eigenfunctions of the Laplace Beltrami operator on the manifold, where the high-dimensional data lies. Then, SR codes the projected testing pixels as sparse linear combinations of all the training samples to classify the testing pixels by evaluating which class leads to the minimum approximation error. The integration of LPP and SR represents an innovative contribution to the literature. The proposed approach, called locality-preserving SR-based classification, addresses the imbalance between high dimensionality of hyperspectral data and the limited number of training samples. Experimental results on three real hyperspectral data sets demonstrate that the proposed approach outperforms the original counterpart, i.e., SR-based classification.

  8. Onboard Image Processing System for Hyperspectral Sensor

    National Research Council Canada - National Science Library

    Hihara, Hiroki; Moritani, Kotaro; Inoue, Masao; Hoshi, Yoshihiro; Iwasaki, Akira; Takada, Jun; Inada, Hitomi; Suzuki, Makoto; Seki, Taeko; Ichikawa, Satoshi; Tanii, Jun

    2015-01-01

    .... Since more than 100 channels are required for hyperspectral sensors on Earth observation satellites, fast and small-footprint lossless image compression capability is essential for reducing the size...

  9. Hybrid Prediction and Fractal Hyperspectral Image Compression

    Directory of Open Access Journals (Sweden)

    Shiping Zhu

    2015-01-01

    Full Text Available The data size of hyperspectral image is too large for storage and transmission, and it has become a bottleneck restricting its applications. So it is necessary to study a high efficiency compression method for hyperspectral image. Prediction encoding is easy to realize and has been studied widely in the hyperspectral image compression field. Fractal coding has the advantages of high compression ratio, resolution independence, and a fast decoding speed, but its application in the hyperspectral image compression field is not popular. In this paper, we propose a novel algorithm for hyperspectral image compression based on hybrid prediction and fractal. Intraband prediction is implemented to the first band and all the remaining bands are encoded by modified fractal coding algorithm. The proposed algorithm can effectively exploit the spectral correlation in hyperspectral image, since each range block is approximated by the domain block in the adjacent band, which is of the same size as the range block. Experimental results indicate that the proposed algorithm provides very promising performance at low bitrate. Compared to other algorithms, the encoding complexity is lower, the decoding quality has a great enhancement, and the PSNR can be increased by about 5 dB to 10 dB.

  10. [Maize seed identification using hyperspectral imaging and SVDD algorithm].

    Science.gov (United States)

    Zhu, Qi-Bing; Feng, Zhao-Li; Huang, Min; Zhu, Xiao

    2013-02-01

    The sufficiency of feature extraction and the rationality of classifier design are two key issues affecting the accuracy of maize seed recognition. In the present study, the hyperspectral images of maize seeds were acquired using hyperspectral image system, and the image entropy of maize seeds for each wavelength was extracted as classification features. Then, support vector data description (SVDD) algorithm was used to develop the classifier model for each variety of maize seeds. The SVDD models yielded 94.14% average test accuracy for known variety samples and 92.28% average test accuracy for new variety samples, respectively. The simulation results showed that the proposed method implemented accurate identification of maize seeds and solved the problem of misclassification by the traditional classification algorithm for new variety maize seeds.

  11. EXhype: A tool for mineral classification using hyperspectral data

    Science.gov (United States)

    Adep, Ramesh Nityanand; shetty, Amba; Ramesh, H.

    2017-02-01

    Various supervised classification algorithms have been developed to classify earth surface features using hyperspectral data. Each algorithm is modelled based on different human expertises. However, the performance of conventional algorithms is not satisfactory to map especially the minerals in view of their typical spectral responses. This study introduces a new expert system named 'EXhype (Expert system for hyperspectral data classification)' to map minerals. The system incorporates human expertise at several stages of it's implementation: (i) to deal with intra-class variation; (ii) to identify absorption features; (iii) to discriminate spectra by considering absorption features, non-absorption features and by full spectra comparison; and (iv) finally takes a decision based on learning and by emphasizing most important features. It is developed using a knowledge base consisting of an Optimal Spectral Library, Segmented Upper Hull method, Spectral Angle Mapper (SAM) and Artificial Neural Network. The performance of the EXhype is compared with a traditional, most commonly used SAM algorithm using Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) data acquired over Cuprite, Nevada, USA. A virtual verification method is used to collect samples information for accuracy assessment. Further, a modified accuracy assessment method is used to get a real users accuracies in cases where only limited or desired classes are considered for classification. With the modified accuracy assessment method, SAM and EXhype yields an overall accuracy of 60.35% and 90.75% and the kappa coefficient of 0.51 and 0.89 respectively. It was also found that the virtual verification method allows to use most desired stratified random sampling method and eliminates all the difficulties associated with it. The experimental results show that EXhype is not only producing better accuracy compared to traditional SAM but, can also rightly classify the minerals. It is proficient in avoiding

  12. Nonparametric, Coupled ,Bayesian ,Dictionary ,and Classifier Learning for Hyperspectral Classification.

    Science.gov (United States)

    Akhtar, Naveed; Mian, Ajmal

    2017-10-03

    We present a principled approach to learn a discriminative dictionary along a linear classifier for hyperspectral classification. Our approach places Gaussian Process priors over the dictionary to account for the relative smoothness of the natural spectra, whereas the classifier parameters are sampled from multivariate Gaussians. We employ two Beta-Bernoulli processes to jointly infer the dictionary and the classifier. These processes are coupled under the same sets of Bernoulli distributions. In our approach, these distributions signify the frequency of the dictionary atom usage in representing class-specific training spectra, which also makes the dictionary discriminative. Due to the coupling between the dictionary and the classifier, the popularity of the atoms for representing different classes gets encoded into the classifier. This helps in predicting the class labels of test spectra that are first represented over the dictionary by solving a simultaneous sparse optimization problem. The labels of the spectra are predicted by feeding the resulting representations to the classifier. Our approach exploits the nonparametric Bayesian framework to automatically infer the dictionary size--the key parameter in discriminative dictionary learning. Moreover, it also has the desirable property of adaptively learning the association between the dictionary atoms and the class labels by itself. We use Gibbs sampling to infer the posterior probability distributions over the dictionary and the classifier under the proposed model, for which, we derive analytical expressions. To establish the effectiveness of our approach, we test it on benchmark hyperspectral images. The classification performance is compared with the state-of-the-art dictionary learning-based classification methods.

  13. Data Field Modeling and Spectral-Spatial Feature Fusion for Hyperspectral Data Classification

    Science.gov (United States)

    Liu, Da; Li, Jianxun

    2016-01-01

    Classification is a significant subject in hyperspectral remote sensing image processing. This study proposes a spectral-spatial feature fusion algorithm for the classification of hyperspectral images (HSI). Unlike existing spectral-spatial classification methods, the influences and interactions of the surroundings on each measured pixel were taken into consideration in this paper. Data field theory was employed as the mathematical realization of the field theory concept in physics, and both the spectral and spatial domains of HSI were considered as data fields. Therefore, the inherent dependency of interacting pixels was modeled. Using data field modeling, spatial and spectral features were transformed into a unified radiation form and further fused into a new feature by using a linear model. In contrast to the current spectral-spatial classification methods, which usually simply stack spectral and spatial features together, the proposed method builds the inner connection between the spectral and spatial features, and explores the hidden information that contributed to classification. Therefore, new information is included for classification. The final classification result was obtained using a random forest (RF) classifier. The proposed method was tested with the University of Pavia and Indian Pines, two well-known standard hyperspectral datasets. The experimental results demonstrate that the proposed method has higher classification accuracies than those obtained by the traditional approaches. PMID:27999259

  14. Data Field Modeling and Spectral-Spatial Feature Fusion for Hyperspectral Data Classification.

    Science.gov (United States)

    Liu, Da; Li, Jianxun

    2016-12-16

    Classification is a significant subject in hyperspectral remote sensing image processing. This study proposes a spectral-spatial feature fusion algorithm for the classification of hyperspectral images (HSI). Unlike existing spectral-spatial classification methods, the influences and interactions of the surroundings on each measured pixel were taken into consideration in this paper. Data field theory was employed as the mathematical realization of the field theory concept in physics, and both the spectral and spatial domains of HSI were considered as data fields. Therefore, the inherent dependency of interacting pixels was modeled. Using data field modeling, spatial and spectral features were transformed into a unified radiation form and further fused into a new feature by using a linear model. In contrast to the current spectral-spatial classification methods, which usually simply stack spectral and spatial features together, the proposed method builds the inner connection between the spectral and spatial features, and explores the hidden information that contributed to classification. Therefore, new information is included for classification. The final classification result was obtained using a random forest (RF) classifier. The proposed method was tested with the University of Pavia and Indian Pines, two well-known standard hyperspectral datasets. The experimental results demonstrate that the proposed method has higher classification accuracies than those obtained by the traditional approaches.

  15. Parallel computation for blood cell classification in medical hyperspectral imagery

    Science.gov (United States)

    Li, Wei; Wu, Lucheng; Qiu, Xianbo; Ran, Qiong; Xie, Xiaoming

    2016-09-01

    With the advantage of fine spectral resolution, hyperspectral imagery provides great potential for cell classification. This paper provides a promising classification system including the following three stages: (1) band selection for a subset of spectral bands with distinctive and informative features, (2) spectral-spatial feature extraction, such as local binary patterns (LBP), and (3) followed by an effective classifier. Moreover, these three steps are further implemented on graphics processing units (GPU) respectively, which makes the system real-time and more practical. The GPU parallel implementation is compared with the serial implementation on central processing units (CPU). Experimental results based on real medical hyperspectral data demonstrate that the proposed system is able to offer high accuracy and fast speed, which are appealing for cell classification in medical hyperspectral imagery.

  16. Hyperspectral imaging and its applications

    Science.gov (United States)

    Serranti, S.; Bonifazi, G.

    2016-04-01

    Hyperspectral imaging (HSI) is an emerging technique that combines the imaging properties of a digital camera with the spectroscopic properties of a spectrometer able to detect the spectral attributes of each pixel in an image. For these characteristics, HSI allows to qualitatively and quantitatively evaluate the effects of the interactions of light with organic and/or inorganic materials. The results of this interaction are usually displayed as a spectral signature characterized by a sequence of energy values, in a pre-defined wavelength interval, for each of the investigated/collected wavelength. Following this approach, it is thus possible to collect, in a fast and reliable way, spectral information that are strictly linked to chemical-physical characteristics of the investigated materials and/or products. Considering that in an hyperspectral image the spectrum of each pixel can be analyzed, HSI can be considered as one of the best nondestructive technology allowing to perform the most accurate and detailed information extraction. HSI can be applied in different wavelength fields, the most common are the visible (VIS: 400-700 nm), the near infrared (NIR: 1000-1700 nm) and the short wave infrared (SWIR: 1000-2500 nm). It can be applied for inspections from micro- to macro-scale, up to remote sensing. HSI produces a large amount of information due to the great number of continuous collected spectral bands. Such an approach, when successful, is quite challenging being usually reliable, robust and characterized by lower costs, if compared with those usually associated to commonly applied analytical off-line and/or on-line analytical approaches. More and more applications have been thus developed and tested, in these last years, especially in food inspection, with a large range of investigated products, such as fruits and vegetables, meat, fish, eggs and cereals, but also in medicine and pharmaceutical sector, in cultural heritage, in material characterization and in

  17. Novel hyperspectral imager for lightweight UAVs

    Science.gov (United States)

    Saari, Heikki; Aallos, Ville-Veikko; Holmlund, Christer; Mäkynen, Jussi; Delauré, Bavo; Nackaerts, Kris; Michiels, Bart

    2010-04-01

    VTT Technical Research Centre of Finland has developed a new miniaturized staring hyperspectral imager with a weight of 350 g making the system compatible with lightweight UAS platforms. The instrument is able to record 2D spatial images at the selected wavelength bands simultaneously. The concept of the hyperspectral imager has been published in the SPIE Proc. 74741. The operational wavelength range of the imager can be tuned in the range 400 - 1100 nm and spectral resolution is in the range 5 - 10 nm @ FWHM. Presently the spatial resolution is 480 × 750 pixels but it can be increased simply by changing the image sensor. The field of view of the system is 20 × 30 degrees and ground pixel size at 100 m flying altitude is around 7.5 cm. The system contains batteries, image acquisition control system and memory for the image data. It can operate autonomously recording hyperspectral data cubes continuously or controlled by the autopilot system of the UAS. The new hyperspectral imager prototype was first tried in co-operation with the Flemish Institute for Technological Research (VITO) on their UAS helicopter. The instrument was configured for the spectral range 500 - 900 nm selected for the vegetation and natural water monitoring applications. The design of the UAS hyperspectral imager and its characterization results together with the analysis of the spectral data from first test flights will be presented.

  18. Endmember extraction algorithms from hyperspectral images

    Directory of Open Access Journals (Sweden)

    M. C. Cantero

    2006-06-01

    Full Text Available During the last years, several high-resolution sensors have been developed for hyperspectral remote sensing applications. Some of these sensors are already available on space-borne devices. Space-borne sensors are currently acquiring a continual stream of hyperspectral data, and new efficient unsupervised algorithms are required to analyze the great amount of data produced by these instruments. The identification of image endmembers is a crucial task in hyperspectral data exploitation. Once the individual endmembers have been identified, several methods can be used to map their spatial distribution, associations and abundances. This paper reviews the Pixel Purity Index (PPI, N-FINDR and Automatic Morphological Endmember Extraction (AMEE algorithms developed to accomplish the task of finding appropriate image endmembers by applying them to real hyperspectral data. In order to compare the performance of these methods a metric based on the Root Mean Square Error (RMSE between the estimated and reference abundance maps is used.

  19. Detecting red blotch disease in grape leaves using hyperspectral imaging

    Science.gov (United States)

    Mehrubeoglu, Mehrube; Orlebeck, Keith; Zemlan, Michael J.; Autran, Wesley

    2016-05-01

    Red blotch disease is a viral disease that affects grapevines. Symptoms appear as irregular blotches on grape leaves with pink and red veins on the underside of the leaves. Red blotch disease causes a reduction in the accumulation of sugar in grapevines affecting the quality of grapes and resulting in delayed harvest. Detecting and monitoring this disease early is important for grapevine management. This work focuses on the use of hyperspectral imaging for detection and mapping red blotch disease in grape leaves. Grape leaves with known red blotch disease have been imaged with a portable hyperspectral imaging system both on and off the vine to investigate the spectral signature of red blotch disease as well as to identify the diseased areas on the leaves. Modified reflectance calculated at spectral bands corresponding to 566 nm (green) and 628 nm (red), and modified reflectance ratios computed at two sets of bands (566 nm / 628 nm, 680 nm / 738 nm) were selected as effective features to differentiate red blotch from healthy-looking and dry leaf. These two modified reflectance and two ratios of modified reflectance values were then used to train the support vector machine classifier in a supervised learning scheme. Once the SVM classifier was defined, two-class classification was achieved for grape leaf hyperspectral images. Identification of the red blotch disease on grape leaves as well as mapping different stages of the disease using hyperspectral imaging are presented in this paper.

  20. Hyperspectral hybrid method classification for detecting altered mucosa of the human larynx

    Directory of Open Access Journals (Sweden)

    Martin Ron

    2012-06-01

    Full Text Available Abstract Background In the field of earth observation, hyperspectral detector systems allow precise target detections of surface components from remote sensing platforms. This enables specific land covers to be identified without the need to physically travel to the areas examined. In the medical field, efforts are underway to develop optical technologies that detect altering tissue surfaces without the necessity to perform an excisional biopsy. With the establishment of expedient classification procedures, hyperspectral imaging may provide a non-invasive diagnostic method that allows determination of pathological tissue with high reliability. In this study, we examined the performance of a hyperspectral hybrid method classification for the automatic detection of altered mucosa of the human larynx. Materials and methods Hyperspectral Imaging was performed in vivo and 30 bands from 390 to 680 nm for 5 cases of laryngeal disorders (2x hemorrhagic polyp, 3x leukoplakia were obtained. Image stacks were processed with unsupervised clustering (linear spectral unmixing, spectral signatures were extracted from unlabeled cluster maps and subsequently applied as end-members for supervised classification (spectral angle mapper of further medical cases with identical diagnosis. Results Linear spectral unmixing clearly highlighted altered mucosa as single spectral clusters in all cases. Matching classes were identified, and extracted spectral signatures could readily be applied for supervised classifications. Automatic target detection performed well, as the considered classes showed notable correspondence with pathological tissue locations. Conclusions Using hyperspectral classification procedures derived from remote sensing applications for diagnostic purposes can create concrete benefits for the medical field. The approach shows that it would be rewarding to collect spectral signatures from histologically different lesions of laryngeal disorders in

  1. Classification of Hyperspectral Images with Spectral-Spatial Sparse Representation%光谱与空间维双重稀疏表达的高光谱影像分类

    Institute of Scientific and Technical Information of China (English)

    朱勇; 吴波

    2016-01-01

    A novel sparse representation classification model with spectral-spatial sparsity properties is presented to improve the classification accuracy of hyperspectral images. Firstly, this method uses the wavelet dictionary as the core dictionary to extract spectral domain sparse information, and then the spectral dimension sparse representation classification is transformed into the wavelet domain (WSRC) by inverse wavelet transformation. After that, we actually extract the sparse spectral features of the hyper-spectral images and increase the recognition of the original dictionary. Secondly, considering the unity and diversity of the spatial adjacent object, we realize the sparse coding of the neighborhood pixels, and then accumulate the sparse codes. At the same time, we classify the hyperspectral images using a linear classifier that is based on the accumulated sparse codes. This method ensures that we extract the main sparse signal of the neighborhood pixels on the basis of the personality features of sparse encoding, and it performs better than the joint sparse representation model (JSRC) which is directly based on the neighborhood pixels. Finally, two commonly used hyperspectral images are utilized to validate the proposed model. The experimental results demonstrate that the pro-posed algorithm outperforms other models in terms of overall accuracy and kappa coefficient measurements.%高光谱遥感影像的稀疏分类是当前遥感信息处理的研究热点。本文提出一种光谱与空间双重稀疏表达的高光谱遥感影像分类方法(WSSRC)。首先利用小波字典对光谱维进行稀疏表示,将光谱维稀疏分类转化到小波域稀疏分类;其次,考虑空间邻域地物光谱的统一性和差异性,对邻域内像元分别进行稀疏编码,并对编码进行累加聚合;然后,利用聚合后的稀疏编码构造线性分类器对高光谱影像进行分类;最后,通过2幅标准的高光谱影像数据验证

  2. MEMS FPI-based smartphone hyperspectral imager

    Science.gov (United States)

    Rissanen, Anna; Saari, Heikki; Rainio, Kari; Stuns, Ingmar; Viherkanto, Kai; Holmlund, Christer; Näkki, Ismo; Ojanen, Harri

    2016-05-01

    This paper demonstrates a mobile phone- compatible hyperspectral imager based on a tunable MEMS Fabry-Perot interferometer. The realized iPhone 5s hyperspectral imager (HSI) demonstrator utilizes MEMS FPI tunable filter for visible-range, which consist of atomic layer deposited (ALD) Al2O3/TiO2-thin film Bragg reflectors. Characterization results for the mobile phone hyperspectral imager utilizing MEMS FPI chip optimized for 500 nm is presented; the operation range is λ = 450 - 550 nm with FWHM between 8 - 15 nm. Also a configuration of two cascaded FPIs (λ = 500 nm and λ = 650 nm) combined with an RGB colour camera is presented. With this tandem configuration, the overall wavelength tuning range of MEMS hyperspectral imagers can be extended to cover a larger range than with a single FPI chip. The potential applications of mobile hyperspectral imagers in the vis-NIR range include authentication, counterfeit detection and potential health/wellness and food sensing applications.

  3. Hyperspectral range imaging for transportation systems evaluation

    Science.gov (United States)

    Bridgelall, Raj; Rafert, J. B.; Atwood, Don; Tolliver, Denver D.

    2016-04-01

    Transportation agencies expend significant resources to inspect critical infrastructure such as roadways, railways, and pipelines. Regular inspections identify important defects and generate data to forecast maintenance needs. However, cost and practical limitations prevent the scaling of current inspection methods beyond relatively small portions of the network. Consequently, existing approaches fail to discover many high-risk defect formations. Remote sensing techniques offer the potential for more rapid and extensive non-destructive evaluations of the multimodal transportation infrastructure. However, optical occlusions and limitations in the spatial resolution of typical airborne and space-borne platforms limit their applicability. This research proposes hyperspectral image classification to isolate transportation infrastructure targets for high-resolution photogrammetric analysis. A plenoptic swarm of unmanned aircraft systems will capture images with centimeter-scale spatial resolution, large swaths, and polarization diversity. The light field solution will incorporate structure-from-motion techniques to reconstruct three-dimensional details of the isolated targets from sequences of two-dimensional images. A comparative analysis of existing low-power wireless communications standards suggests an application dependent tradeoff in selecting the best-suited link to coordinate swarming operations. This study further produced a taxonomy of specific roadway and railway defects, distress symptoms, and other anomalies that the proposed plenoptic swarm sensing system would identify and characterize to estimate risk levels.

  4. Advanced hyperspectral imaging system with edge enhancement

    Science.gov (United States)

    Yushkov, K. B.; Molchanov, V. Y.

    2017-03-01

    We developed an acousto-optic hyperspectral imaging system with edge enhancement capability. The system is an add-on to a standard light microscope. Edge enhancement operation mode is aimed for analysis of low-contrast microscopic samples, e.g. unstained cytological smears and histological samples, live cells. Edge-enhancement imaging mode is based on a feature of acousto-optic tunable filters to perform band-pass spatial filtering when unturned from noncritical phase matching geometry is diffraction. Switching between standard hyperspectral imaging and edge-enhancement modes is performed by means of a telecentric amplitude mask.

  5. Hyperspectral imaging of neoplastic progression in a mouse model of oral carcinogenesis

    Science.gov (United States)

    Lu, Guolan; Qin, Xulei; Wang, Dongsheng; Muller, Susan; Zhang, Hongzheng; Chen, Amy; Chen, Zhuo Georgia; Fei, Baowei

    2016-03-01

    Hyperspectral imaging (HSI) is an emerging modality for medical applications and holds great potential for noninvasive early detection of cancer. It has been reported that early cancer detection can improve the survival and quality of life of head and neck cancer patients. In this paper, we explored the possibility of differentiating between premalignant lesions and healthy tongue tissue using hyperspectral imaging in a chemical induced oral cancer animal model. We proposed a novel classification algorithm for cancer detection using hyperspectral images. The method detected the dysplastic tissue with an average area under the curve (AUC) of 0.89. The hyperspectral imaging and classification technique may provide a new tool for oral cancer detection.

  6. A multiple classifier approach for spectral-spatial classification of hyperspectral data

    OpenAIRE

    Tarabalka, Yuliya; Benediktsson, Jon Atli; Tilton, James; Chanussot, Jocelyn

    2010-01-01

    International audience; A new multiple classifier method for spectral-spatial classification of hyperspectral images is proposed. Several classifiers are used independently to classify an image. For every pixel, if all the classifiers have assigned this pixel to the same class, the pixel is kept as a marker, i.e., a seed of the spatial region, with the corresponding class label. We propose to use spectral-spatial classifiers at the preliminary step of the marker selection procedure, each of t...

  7. Kernel based subspace projection of hyperspectral images

    DEFF Research Database (Denmark)

    Larsen, Rasmus; Nielsen, Allan Aasbjerg; Arngren, Morten

    In hyperspectral image analysis an exploratory approach to analyse the image data is to conduct subspace projections. As linear projections often fail to capture the underlying structure of the data, we present kernel based subspace projections of PCA and Maximum Autocorrelation Factors (MAF...

  8. Super-resolution reconstruction of hyperspectral images

    Science.gov (United States)

    Elbakary, Mohamed; Alam, Mohammad S.

    2007-04-01

    Hyperspectral imagery is used for a wide variety of applications, including target detection, tacking, agricultural monitoring and natural resources exploration. The main reason for using hyperspectral imagery is that these images reveal spectral information about the scene that are not available in a single band. Unfortunately, many factors such as sensor noise and atmospheric scattering degrade the spatial quality of these images. Recently, many algorithms are introduced in the literature to improve the resolution of hyperspectral images [7]. In this paper, we propose a new method to produce high resolution bands from low resolution bands that are strongly correlated to the corresponding high resolution panchromatic image. The proposed method is based on using the local correlation instead of using the global correlation to improve the estimated interpolation in order to construct the high resolution image. The utilization of local correlation significantly improved the resolution of high resolution images when compared to the corresponding results obtained using the traditional algorithms. The local correlation is implemented by using predefined small windows across the low resolution image. In addition, numerous experiments are conducted to investigate the effect of the chosen window size in the image quality. Experiments results obtained using real life hyperspectral imagery is presented to verify the effectiveness of the proposed algorithm.

  9. Spectral Band Selection for Urban Material Classification Using Hyperspectral Libraries

    Science.gov (United States)

    Le Bris, A.; Chehata, N.; Briottet, X.; Paparoditis, N.

    2016-06-01

    In urban areas, information concerning very high resolution land cover and especially material maps are necessary for several city modelling or monitoring applications. That is to say, knowledge concerning the roofing materials or the different kinds of ground areas is required. Airborne remote sensing techniques appear to be convenient for providing such information at a large scale. However, results obtained using most traditional processing methods based on usual red-green-blue-near infrared multispectral images remain limited for such applications. A possible way to improve classification results is to enhance the imagery spectral resolution using superspectral or hyperspectral sensors. In this study, it is intended to design a superspectral sensor dedicated to urban materials classification and this work particularly focused on the selection of the optimal spectral band subsets for such sensor. First, reflectance spectral signatures of urban materials were collected from 7 spectral libraires. Then, spectral optimization was performed using this data set. The band selection workflow included two steps, optimising first the number of spectral bands using an incremental method and then examining several possible optimised band subsets using a stochastic algorithm. The same wrapper relevance criterion relying on a confidence measure of Random Forests classifier was used at both steps. To cope with the limited number of available spectra for several classes, additional synthetic spectra were generated from the collection of reference spectra: intra-class variability was simulated by multiplying reference spectra by a random coefficient. At the end, selected band subsets were evaluated considering the classification quality reached using a rbf svm classifier. It was confirmed that a limited band subset was sufficient to classify common urban materials. The important contribution of bands from the Short Wave Infra-Red (SWIR) spectral domain (1000-2400 nm) to material

  10. Radiation calibration for LWIR Hyperspectral Imager Spectrometer

    Science.gov (United States)

    Yang, Zhixiong; Yu, Chunchao; Zheng, Wei-jian; Lei, Zhenggang; Yan, Min; Yuan, Xiaochun; Zhang, Peizhong

    2014-11-01

    The radiometric calibration of LWIR Hyperspectral imager Spectrometer is presented. The lab has been developed to LWIR Interferometric Hyperspectral imager Spectrometer Prototype(CHIPED-I) to study Lab Radiation Calibration, Two-point linear calibration is carried out for the spectrometer by using blackbody respectively. Firstly, calibration measured relative intensity is converted to the absolute radiation lightness of the object. Then, radiation lightness of the object is is converted the brightness temperature spectrum by the method of brightness temperature. The result indicated †that this method of Radiation Calibration calibration was very good.

  11. Biologically-inspired data decorrelation for hyper-spectral imaging

    Directory of Open Access Journals (Sweden)

    Ghita Ovidiu

    2011-01-01

    Full Text Available Abstract Hyper-spectral data allows the construction of more robust statistical models to sample the material properties than the standard tri-chromatic color representation. However, because of the large dimensionality and complexity of the hyper-spectral data, the extraction of robust features (image descriptors is not a trivial issue. Thus, to facilitate efficient feature extraction, decorrelation techniques are commonly applied to reduce the dimensionality of the hyper-spectral data with the aim of generating compact and highly discriminative image descriptors. Current methodologies for data decorrelation such as principal component analysis (PCA, linear discriminant analysis (LDA, wavelet decomposition (WD, or band selection methods require complex and subjective training procedures and in addition the compressed spectral information is not directly related to the physical (spectral characteristics associated with the analyzed materials. The major objective of this article is to introduce and evaluate a new data decorrelation methodology using an approach that closely emulates the human vision. The proposed data decorrelation scheme has been employed to optimally minimize the amount of redundant information contained in the highly correlated hyper-spectral bands and has been comprehensively evaluated in the context of non-ferrous material classification

  12. Hyperspectral Image Recovery via Hybrid Regularization

    Science.gov (United States)

    Arablouei, Reza; de Hoog, Frank

    2016-12-01

    Natural images tend to mostly consist of smooth regions with individual pixels having highly correlated spectra. This information can be exploited to recover hyperspectral images of natural scenes from their incomplete and noisy measurements. To perform the recovery while taking full advantage of the prior knowledge, we formulate a composite cost function containing a square-error data-fitting term and two distinct regularization terms pertaining to spatial and spectral domains. The regularization for the spatial domain is the sum of total-variation of the image frames corresponding to all spectral bands. The regularization for the spectral domain is the l1-norm of the coefficient matrix obtained by applying a suitable sparsifying transform to the spectra of the pixels. We use an accelerated proximal-subgradient method to minimize the formulated cost function. We analyze the performance of the proposed algorithm and prove its convergence. Numerical simulations using real hyperspectral images exhibit that the proposed algorithm offers an excellent recovery performance with a number of measurements that is only a small fraction of the hyperspectral image data size. Simulation results also show that the proposed algorithm significantly outperforms an accelerated proximal-gradient algorithm that solves the classical basis-pursuit denoising problem to recover the hyperspectral image.

  13. Hyperspectral laser-induced autofluorescence imaging of dental caries

    Science.gov (United States)

    Bürmen, Miran; Fidler, Aleš; Pernuš, Franjo; Likar, Boštjan

    2012-01-01

    Dental caries is a disease characterized by demineralization of enamel crystals leading to the penetration of bacteria into the dentine and pulp. Early detection of enamel demineralization resulting in increased enamel porosity, commonly known as white spots, is a difficult diagnostic task. Laser induced autofluorescence was shown to be a useful method for early detection of demineralization. The existing studies involved either a single point spectroscopic measurements or imaging at a single spectral band. In the case of spectroscopic measurements, very little or no spatial information is acquired and the measured autofluorescence signal strongly depends on the position and orientation of the probe. On the other hand, single-band spectral imaging can be substantially affected by local spectral artefacts. Such effects can significantly interfere with automated methods for detection of early caries lesions. In contrast, hyperspectral imaging effectively combines the spatial information of imaging methods with the spectral information of spectroscopic methods providing excellent basis for development of robust and reliable algorithms for automated classification and analysis of hard dental tissues. In this paper, we employ 405 nm laser excitation of natural caries lesions. The fluorescence signal is acquired by a state-of-the-art hyperspectral imaging system consisting of a high-resolution acousto-optic tunable filter (AOTF) and a highly sensitive Scientific CMOS camera in the spectral range from 550 nm to 800 nm. The results are compared to the contrast obtained by near-infrared hyperspectral imaging technique employed in the existing studies on early detection of dental caries.

  14. Single aflatoxin contaminated corn kernel analysis with fluorescence hyperspectral image

    Science.gov (United States)

    Yao, Haibo; Hruska, Zuzana; Kincaid, Russell; Ononye, Ambrose; Brown, Robert L.; Cleveland, Thomas E.

    2010-04-01

    Aflatoxins are toxic secondary metabolites of the fungi Aspergillus flavus and Aspergillus parasiticus, among others. Aflatoxin contaminated corn is toxic to domestic animals when ingested in feed and is a known carcinogen associated with liver and lung cancer in humans. Consequently, aflatoxin levels in food and feed are regulated by the Food and Drug Administration (FDA) in the US, allowing 20 ppb (parts per billion) limits in food and 100 ppb in feed for interstate commerce. Currently, aflatoxin detection and quantification methods are based on analytical tests including thin-layer chromatography (TCL) and high performance liquid chromatography (HPLC). These analytical tests require the destruction of samples, and are costly and time consuming. Thus, the ability to detect aflatoxin in a rapid, nondestructive way is crucial to the grain industry, particularly to corn industry. Hyperspectral imaging technology offers a non-invasive approach toward screening for food safety inspection and quality control based on its spectral signature. The focus of this paper is to classify aflatoxin contaminated single corn kernels using fluorescence hyperspectral imagery. Field inoculated corn kernels were used in the study. Contaminated and control kernels under long wavelength ultraviolet excitation were imaged using a visible near-infrared (VNIR) hyperspectral camera. The imaged kernels were chemically analyzed to provide reference information for image analysis. This paper describes a procedure to process corn kernels located in different images for statistical training and classification. Two classification algorithms, Maximum Likelihood and Binary Encoding, were used to classify each corn kernel into "control" or "contaminated" through pixel classification. The Binary Encoding approach had a slightly better performance with accuracy equals to 87% or 88% when 20 ppb or 100 ppb was used as classification threshold, respectively.

  15. Hyperspectral optical imaging of two different species of lepidoptera

    Science.gov (United States)

    Medina, José Manuel; Nascimento, Sérgio Miguel Cardoso; Vukusic, Pete

    2011-05-01

    In this article, we report a hyperspectral optical imaging application for measurement of the reflectance spectra of photonic structures that produce structural colors with high spatial resolution. The measurement of the spectral reflectance function is exemplified in the butterfly wings of two different species of Lepidoptera: the blue iridescence reflected by the nymphalid Morpho didius and the green iridescence of the papilionid Papilio palinurus. Color coordinates from reflectance spectra were calculated taking into account human spectral sensitivity. For each butterfly wing, the observed color is described by a characteristic color map in the chromaticity diagram and spreads over a limited volume in the color space. The results suggest that variability in the reflectance spectra is correlated with different random arrangements in the spatial distribution of the scales that cover the wing membranes. Hyperspectral optical imaging opens new ways for the non-invasive study and classification of different forms of irregularity in structural colors.

  16. Hyperspectral optical imaging of two different species of lepidoptera

    Directory of Open Access Journals (Sweden)

    Vukusic Pete

    2011-01-01

    Full Text Available Abstract In this article, we report a hyperspectral optical imaging application for measurement of the reflectance spectra of photonic structures that produce structural colors with high spatial resolution. The measurement of the spectral reflectance function is exemplified in the butterfly wings of two different species of Lepidoptera: the blue iridescence reflected by the nymphalid Morpho didius and the green iridescence of the papilionid Papilio palinurus. Color coordinates from reflectance spectra were calculated taking into account human spectral sensitivity. For each butterfly wing, the observed color is described by a characteristic color map in the chromaticity diagram and spreads over a limited volume in the color space. The results suggest that variability in the reflectance spectra is correlated with different random arrangements in the spatial distribution of the scales that cover the wing membranes. Hyperspectral optical imaging opens new ways for the non-invasive study and classification of different forms of irregularity in structural colors.

  17. The hyperspectral imaging trade-off

    DEFF Research Database (Denmark)

    Carstensen, Jens Michael

    Although it has no clear-cut definition, hyperspectral imaging in the UV-Visible-NIR wavelength region seems to mean spectral image sampling in bands from 10 nm width or narrower that enables spectral reconstruction over some wavelength interval. For non-imaging spectral applications, this will b......Although it has no clear-cut definition, hyperspectral imaging in the UV-Visible-NIR wavelength region seems to mean spectral image sampling in bands from 10 nm width or narrower that enables spectral reconstruction over some wavelength interval. For non-imaging spectral applications...... acquisition 8. System price Spectral resolution has an obvious interest, but it is the experience of the author, that the vast majority of real spectral imaging applications only need to unmix 2-5 different components. This should be possible with a sensible choice of 10-20 wavelengths. This presentation...

  18. Tree Classification with Fused Mobile Laser Scanning and Hyperspectral Data

    Directory of Open Access Journals (Sweden)

    Juha Hyyppä

    2011-05-01

    Full Text Available Mobile Laser Scanning data were collected simultaneously with hyperspectral data using the Finnish Geodetic Institute Sensei system. The data were tested for tree species classification. The test area was an urban garden in the City of Espoo, Finland. Point clouds representing 168 individual tree specimens of 23 tree species were determined manually. The classification of the trees was done using first only the spatial data from point clouds, then with only the spectral data obtained with a spectrometer, and finally with the combined spatial and hyperspectral data from both sensors. Two classification tests were performed: the separation of coniferous and deciduous trees, and the identification of individual tree species. All determined tree specimens were used in distinguishing coniferous and deciduous trees. A subset of 133 trees and 10 tree species was used in the tree species classification. The best classification results for the fused data were 95.8% for the separation of the coniferous and deciduous classes. The best overall tree species classification succeeded with 83.5% accuracy for the best tested fused data feature combination. The respective results for paired structural features derived from the laser point cloud were 90.5% for the separation of the coniferous and deciduous classes and 65.4% for the species classification. Classification accuracies with paired hyperspectral reflectance value data were 90.5% for the separation of coniferous and deciduous classes and 62.4% for different species. The results are among the first of their kind and they show that mobile collected fused data outperformed single-sensor data in both classification tests and by a significant margin.

  19. Food inspection using hyperspectral imaging and SVDD

    Science.gov (United States)

    Uslu, Faruk Sukru; Binol, Hamidullah; Bal, Abdullah

    2016-05-01

    Nowadays food inspection and evaluation is becoming significant public issue, therefore robust, fast, and environmentally safe methods are studied instead of human visual assessment. Optical sensing is one of the potential methods with the properties of being non-destructive and accurate. As a remote sensing technology, hyperspectral imaging (HSI) is being successfully applied by researchers because of having both spatial and detailed spectral information about studied material. HSI can be used to inspect food quality and safety estimation such as meat quality assessment, quality evaluation of fish, detection of skin tumors on chicken carcasses, and classification of wheat kernels in the food industry. In this paper, we have implied an experiment to detect fat ratio in ground meat via Support Vector Data Description which is an efficient and robust one-class classifier for HSI. The experiments have been implemented on two different ground meat HSI data sets with different fat percentage. Addition to these implementations, we have also applied bagging technique which is mostly used as an ensemble method to improve the prediction ratio. The results show that the proposed methods produce high detection performance for fat ratio in ground meat.

  20. LIFTERS-hyperspectral imaging at LLNL

    Energy Technology Data Exchange (ETDEWEB)

    Fields, D. [Lawrence Livermore National Lab., CA (United States); Bennett, C.; Carter, M.

    1994-11-15

    LIFTIRS, the Livermore Imaging Fourier Transform InfraRed Spectrometer, recently developed at LLNL, is an instrument which enables extremely efficient collection and analysis of hyperspectral imaging data. LIFTIRS produces a spatial format of 128x128 pixels, with spectral resolution arbitrarily variable up to a maximum of 0.25 inverse centimeters. Time resolution and spectral resolution can be traded off for each other with great flexibility. We will discuss recent measurements made with this instrument, and present typical images and spectra.

  1. A Gimbal-Stabilized Compact Hyperspectral Imaging System Project

    Data.gov (United States)

    National Aeronautics and Space Administration — The Gimbal-stabilized Compact Hyperspectral Imaging System (GCHIS) fully integrates multi-sensor spectral imaging, stereovision, GPS and inertial measurement,...

  2. A Gimbal-Stabilized Compact Hyperspectral Imaging System Project

    Data.gov (United States)

    National Aeronautics and Space Administration — The Gimbal-stabilized Compact Hyperspectral Imaging System (GCHIS) fully integrates multi-sensor spectral imaging, stereovision, GPS and inertial measurement,...

  3. Object-Based Tree Species Classification in Urban Ecosystems Using LiDAR and Hyperspectral Data

    Directory of Open Access Journals (Sweden)

    Zhongya Zhang

    2016-06-01

    Full Text Available In precision forestry, tree species identification is key to evaluating the role of forest ecosystems in the provision of ecosystem services, such as carbon sequestration and assessing their effects on climate regulation and climate change. In this study, we investigated the effectiveness of tree species classification of urban forests using aerial-based HyMap hyperspectral imagery and light detection and ranging (LiDAR data. First, we conducted an object-based image analysis (OBIA to segment individual tree crowns present in LiDAR-derived Canopy Height Models (CHMs. Then, hyperspectral values for individual trees were extracted from HyMap data for band reduction through Minimum Noise Fraction (MNF transformation which allowed us to reduce the data to 20 significant bands out of 118 bands acquired. Finally, we compared several different classifications using Random Forest (RF and Multi Class Classifier (MCC methods. Seven tree species were classified using all 118 bands which resulted in 46.3% overall classification accuracy for RF versus 79.6% for MCC. Using only the 20 optimal bands extracted through MNF, both RF and MCC achieved an increase in overall accuracy to 87.0% and 88.9%, respectively. Thus, the MNF band selection process is a preferable approach for tree species classification when using hyperspectral data. Further, our work also suggests that RF is heavily disadvantaged by the high-dimensionality and noise present in hyperspectral data, while MCC is more robust when handling high-dimensional datasets with small sample sizes. Our overall results indicated that individual tree species identification in urban forests can be accomplished with the fusion of object-based LiDAR segmentation of crowns and hyperspectral characterization.

  4. Wide area Hyperspectral Motion Imaging

    Science.gov (United States)

    2017-02-03

    gigapixel resolution from an airborne platform . This huge amount of visual data requires algorithms for automated detection and tracking of targets of...constraints, it is not surprising that no military or commercial WAMI platform has a hyperspectral sensing capability. Therefore, today’s systems must...digital pulse stream , and a counter counts the number of pulses within a given integration period [2]. The magnitude of the count is proportional to

  5. Parallel Implementation of Morphological Profile Based Spectral-Spatial Classification Scheme for Hyperspectral Imagery

    Science.gov (United States)

    Kumar, B.; Dikshit, O.

    2016-06-01

    Extended morphological profile (EMP) is a good technique for extracting spectral-spatial information from the images but large size of hyperspectral images is an important concern for creating EMPs. However, with the availability of modern multi-core processors and commodity parallel processing systems like graphics processing units (GPUs) at desktop level, parallel computing provides a viable option to significantly accelerate execution of such computations. In this paper, parallel implementation of an EMP based spectralspatial classification method for hyperspectral imagery is presented. The parallel implementation is done both on multi-core CPU and GPU. The impact of parallelization on speed up and classification accuracy is analyzed. For GPU, the implementation is done in compute unified device architecture (CUDA) C. The experiments are carried out on two well-known hyperspectral images. It is observed from the experimental results that GPU implementation provides a speed up of about 7 times, while parallel implementation on multi-core CPU resulted in speed up of about 3 times. It is also observed that parallel implementation has no adverse impact on the classification accuracy.

  6. Mapping Soil Organic Matter with Hyperspectral Imaging

    Science.gov (United States)

    Moni, Christophe; Burud, Ingunn; Flø, Andreas; Rasse, Daniel

    2014-05-01

    Soil organic matter (SOM) plays a central role for both food security and the global environment. Soil organic matter is the 'glue' that binds soil particles together, leading to positive effects on soil water and nutrient availability for plant growth and helping to counteract the effects of erosion, runoff, compaction and crusting. Hyperspectral measurements of samples of soil profiles have been conducted with the aim of mapping soil organic matter on a macroscopic scale (millimeters and centimeters). Two soil profiles have been selected from the same experimental site, one from a plot amended with biochar and another one from a control plot, with the specific objective to quantify and map the distribution of biochar in the amended profile. The soil profiles were of size (30 x 10 x 10) cm3 and were scanned with two pushbroomtype hyperspectral cameras, one which is sensitive in the visible wavelength region (400 - 1000 nm) and one in the near infrared region (1000 - 2500 nm). The images from the two detectors were merged together into one full dataset covering the whole wavelength region. Layers of 15 mm were removed from the 10 cm high sample such that a total of 7 hyperspectral images were obtained from the samples. Each layer was analyzed with multivariate statistical techniques in order to map the different components in the soil profile. Moreover, a 3-dimensional visalization of the components through the depth of the sample was also obtained by combining the hyperspectral images from all the layers. Mid-infrared spectroscopy of selected samples of the measured soil profiles was conducted in order to correlate the chemical constituents with the hyperspectral results. The results show that hyperspectral imaging is a fast, non-destructive technique, well suited to characterize soil profiles on a macroscopic scale and hence to map elements and different organic matter quality present in a complete pedon. As such, we were able to map and quantify biochar in our

  7. Illumination compensation in ground based hyperspectral imaging

    Science.gov (United States)

    Wendel, Alexander; Underwood, James

    2017-07-01

    Hyperspectral imaging has emerged as an important tool for analysing vegetation data in agricultural applications. Recently, low altitude and ground based hyperspectral imaging solutions have come to the fore, providing very high resolution data for mapping and studying large areas of crops in detail. However, these platforms introduce a unique set of challenges that need to be overcome to ensure consistent, accurate and timely acquisition of data. One particular problem is dealing with changes in environmental illumination while operating with natural light under cloud cover, which can have considerable effects on spectral shape. In the past this has been commonly achieved by imaging known reference targets at the time of data acquisition, direct measurement of irradiance, or atmospheric modelling. While capturing a reference panel continuously or very frequently allows accurate compensation for illumination changes, this is often not practical with ground based platforms, and impossible in aerial applications. This paper examines the use of an autonomous unmanned ground vehicle (UGV) to gather high resolution hyperspectral imaging data of crops under natural illumination. A process of illumination compensation is performed to extract the inherent reflectance properties of the crops, despite variable illumination. This work adapts a previously developed subspace model approach to reflectance and illumination recovery. Though tested on a ground vehicle in this paper, it is applicable to low altitude unmanned aerial hyperspectral imagery also. The method uses occasional observations of reference panel training data from within the same or other datasets, which enables a practical field protocol that minimises in-field manual labour. This paper tests the new approach, comparing it against traditional methods. Several illumination compensation protocols for high volume ground based data collection are presented based on the results. The findings in this paper are

  8. Development of algorithms for detection of mechanical injury on white mushrooms (Agaricus bisporus) using hyperspectral imaging

    Science.gov (United States)

    Gowen, A. A.; O'Donnell, C. P.

    2009-05-01

    White mushrooms were subjected to mechanical injury by controlled shaking in a plastic box at 400 rpm for different times (0, 60, 120, 300 and 600 s). Immediately after shaking, hyperspectral images were obtained using two pushbroom line-scanning hyperspectral imaging instruments, one operating in the wavelength range of 400 - 1000 nm with spectroscopic resolution of 5 nm, the other operating in the wavelength range of 950 - 1700 nm with spectroscopic resolution of 7 nm. Different spectral and spatial pretreatments were investigated to reduce the effect of sample curvature on hyperspectral data. Algorithms based on Chemometric techniques (Principal Component Analysis and Partial Least Squares Discriminant Analysis) and image processing methods (masking, thresholding, morphological operations) were developed for pixel classification in hyperspectral images. In addition, correlation analysis, spectral angle mapping and scaled difference of sample spectra were investigated and compared with the chemometric approaches.

  9. Hyperspectral Image Analysis of Food Quality

    DEFF Research Database (Denmark)

    Arngren, Morten

    Assessing the quality of food is a vital step in any food processing line to ensurethe best food quality and maximum profit for the farmer and food manufacturer.Traditional quality evaluation methods are often destructive and labourintensive procedures relying on wet chemistry or subjective human...... inspection.Near-infrared spectroscopy can address these issues by offering a fast and objectiveanalysis of the food quality. A natural extension to these single spectrumNIR systems is to include image information such that each pixel holds a NIRspectrum. This augmented image information offers several...... extensions to the analysis offood quality. This dissertation is concerned with hyperspectral image analysisused to assess the quality of single grain kernels. The focus is to highlight thebenefits and challenges of using hyperspectral imaging for food quality presentedin two research directions. Initially...

  10. Online hyperspectral imaging system for evaluating quality of agricultural products

    Science.gov (United States)

    Mo, Changyeun; Kim, Giyoung; Lim, Jongguk

    2017-06-01

    The consumption of fresh-cut agricultural produce in Korea has been growing. The browning of fresh-cut vegetables that occurs during storage and foreign substances such as worms and slugs are some of the main causes of consumers' concerns with respect to safety and hygiene. The purpose of this study is to develop an on-line system for evaluating quality of agricultural products using hyperspectral imaging technology. The online evaluation system with single visible-near infrared hyperspectral camera in the range of 400 nm to 1000 nm that can assess quality of both surfaces of agricultural products such as fresh-cut lettuce was designed. Algorithms to detect browning surface were developed for this system. The optimal wavebands for discriminating between browning and sound lettuce as well as between browning lettuce and the conveyor belt were investigated using the correlation analysis and the one-way analysis of variance method. The imaging algorithms to discriminate the browning lettuces were developed using the optimal wavebands. The ratio image (RI) algorithm of the 533 nm and 697 nm images (RI533/697) for abaxial surface lettuce and the ratio image algorithm (RI533/697) and subtraction image (SI) algorithm (SI538-697) for adaxial surface lettuce had the highest classification accuracies. The classification accuracy of browning and sound lettuce was 100.0% and above 96.0%, respectively, for the both surfaces. The overall results show that the online hyperspectral imaging system could potentially be used to assess quality of agricultural products.

  11. Classification of Salmonella serotypes with hyperspectral microscope imagery

    Science.gov (United States)

    Previous research has demonstrated an optical method with acousto-optic tunable filter (AOTF) based hyperspectral microscope imaging (HMI) had potential for classifying gram-negative from gram-positive foodborne pathogenic bacteria rapidly and nondestructively with a minimum sample preparation. In t...

  12. The hyperspectral imaging trade-off

    DEFF Research Database (Denmark)

    Carstensen, Jens Michael

    structures in these projects. However, hyperspectral imaging is a sampling choice within spectral imaging that typically will impose some trade-offs, and these trade-offs will not be optimal for many applications. The purpose of this presentation is to point out and increase the awareness of these trade...... will show how spectral resolution in these situations can be traded off with improvements in the parameters above....

  13. A Fast Feature Extraction Method Based on Integer Wavelet Transform for Hyperspectral Images

    Institute of Scientific and Technical Information of China (English)

    GUYanfeng; ZHANGYe; YUShanshan

    2004-01-01

    Hyperspectral remote sensing provides high-resolution spectral data and the potential for remote discrimination between subtle differences in ground covers. However, the high-dimensional data space generated by the hyperspectral sensors creates a new challenge for conventional spectral data analysis techniques. A challenging problem in using hyperspectral data is to eliminate redundancy and preserve useful spectral information for applications. In this paper, a Fast feature extraction (FFE) method based on integer wavelet transform is proposed to extract useful features and reduce dimensionality of hyperspectral images. The FFE method can be directly used to extract useful features from spectral vector of each pixel resident in the hyperspectral images. The FFE method has two main merits: high computational efficiency and good ability to extract spectral features. In order to better testify the effectiveness and the performance of the proposed method, classification experiments of hyperspectral images are performed on two groups of AVIRIS (Airborne visible/infrared imaging spectrometer) data respectively. In addition, three existing methods for feature extraction of hyperspectral images, i.e. PCA, SPCT and Wavelet Transform, are performed on the same data for comparison with the proposed method. The experimental investigation shows that the efficiency of the FFE method for feature extraction outclasses those of the other three methods mentioned above.

  14. Classification of hyperspectral imagery with neural networks: comparison to conventional tools

    Science.gov (United States)

    Merényi, Erzsébet; Farrand, William H.; Taranik, James V.; Minor, Timothy B.

    2014-12-01

    Efficient exploitation of hyperspectral imagery is of great importance in remote sensing. Artificial intelligence approaches have been receiving favorable reviews for classification of hyperspectral data because the complexity of such data challenges the limitations of many conventional methods. Artificial neural networks (ANNs) were shown to outperform traditional classifiers in many situations. However, studies that use the full spectral dimensionality of hyperspectral images to classify a large number of surface covers are scarce if non-existent. We advocate the need for methods that can handle the full dimensionality and a large number of classes to retain the discovery potential and the ability to discriminate classes with subtle spectral differences. We demonstrate that such a method exists in the family of ANNs. We compare the maximum likelihood, Mahalonobis distance, minimum distance, spectral angle mapper, and a hybrid ANN classifier for real hyperspectral AVIRIS data, using the full spectral resolution to map 23 cover types and using a small training set. Rigorous evaluation of the classification accuracies shows that the ANN outperforms the other methods and achieves ≈90% accuracy on test data.

  15. Spectral Reconstruction for Obtaining Virtual Hyperspectral Images

    Science.gov (United States)

    Perez, G. J. P.; Castro, E. C.

    2016-12-01

    Hyperspectral sensors demonstrated its capabalities in identifying materials and detecting processes in a satellite scene. However, availability of hyperspectral images are limited due to the high development cost of these sensors. Currently, most of the readily available data are from multi-spectral instruments. Spectral reconstruction is an alternative method to address the need for hyperspectral information. The spectral reconstruction technique has been shown to provide a quick and accurate detection of defects in an integrated circuit, recovers damaged parts of frescoes, and it also aids in converting a microscope into an imaging spectrometer. By using several spectral bands together with a spectral library, a spectrum acquired by a sensor can be expressed as a linear superposition of elementary signals. In this study, spectral reconstruction is used to estimate the spectra of different surfaces imaged by Landsat 8. Four atmospherically corrected surface reflectance from three visible bands (499 nm, 585 nm, 670 nm) and one near-infrared band (872 nm) of Landsat 8, and a spectral library of ground elements acquired from the United States Geological Survey (USGS) are used. The spectral library is limited to 420-1020 nm spectral range, and is interpolated at one nanometer resolution. Singular Value Decomposition (SVD) is used to calculate the basis spectra, which are then applied to reconstruct the spectrum. The spectral reconstruction is applied for test cases within the library consisting of vegetation communities. This technique was successful in reconstructing a hyperspectral signal with error of less than 12% for most of the test cases. Hence, this study demonstrated the potential of simulating information at any desired wavelength, creating a virtual hyperspectral sensor without the need for additional satellite bands.

  16. A collection of hyperspectral images for imaging systems research

    Science.gov (United States)

    Skauli, Torbjørn; Farrell, Joyce

    2013-01-01

    A set of hyperspectral image data are made available, intended for use in modeling of imaging systems. The set contains images of faces, landscapes and buildings. The data cover wavelengths from 0.4 to 2.5 micrometers, spanning the visible, NIR and SWIR electromagnetic spectral ranges. The images have been recorded with two HySpex line-scan imaging spectrometers covering the spectral ranges 0.4 to 1 micrometers and 1 to 2.5 micrometers. The hyperspectral data set includes measured illuminants and software for converting the radiance data to estimated reflectance. The images are being made available for download at http://scien.stanford.edu

  17. Hyperspectral imaging using a color camera and its application for pathogen detection

    Science.gov (United States)

    Yoon, Seung-Chul; Shin, Tae-Sung; Heitschmidt, Gerald W.; Lawrence, Kurt C.; Park, Bosoon; Gamble, Gary

    2015-02-01

    This paper reports the results of a feasibility study for the development of a hyperspectral image recovery (reconstruction) technique using a RGB color camera and regression analysis in order to detect and classify colonies of foodborne pathogens. The target bacterial pathogens were the six representative non-O157 Shiga-toxin producing Escherichia coli (STEC) serogroups (O26, O45, O103, O111, O121, and O145) grown in Petri dishes of Rainbow agar. The purpose of the feasibility study was to evaluate whether a DSLR camera (Nikon D700) could be used to predict hyperspectral images in the wavelength range from 400 to 1,000 nm and even to predict the types of pathogens using a hyperspectral STEC classification algorithm that was previously developed. Unlike many other studies using color charts with known and noise-free spectra for training reconstruction models, this work used hyperspectral and color images, separately measured by a hyperspectral imaging spectrometer and the DSLR color camera. The color images were calibrated (i.e. normalized) to relative reflectance, subsampled and spatially registered to match with counterpart pixels in hyperspectral images that were also calibrated to relative reflectance. Polynomial multivariate least-squares regression (PMLR) was previously developed with simulated color images. In this study, partial least squares regression (PLSR) was also evaluated as a spectral recovery technique to minimize multicollinearity and overfitting. The two spectral recovery models (PMLR and PLSR) and their parameters were evaluated by cross-validation. The QR decomposition was used to find a numerically more stable solution of the regression equation. The preliminary results showed that PLSR was more effective especially with higher order polynomial regressions than PMLR. The best classification accuracy measured with an independent test set was about 90%. The results suggest the potential of cost-effective color imaging using hyperspectral image

  18. Supervised Classification Performance of Multispectral Images

    CERN Document Server

    Perumal, K

    2010-01-01

    Nowadays government and private agencies use remote sensing imagery for a wide range of applications from military applications to farm development. The images may be a panchromatic, multispectral, hyperspectral or even ultraspectral of terra bytes. Remote sensing image classification is one amongst the most significant application worlds for remote sensing. A few number of image classification algorithms have proved good precision in classifying remote sensing data. But, of late, due to the increasing spatiotemporal dimensions of the remote sensing data, traditional classification algorithms have exposed weaknesses necessitating further research in the field of remote sensing image classification. So an efficient classifier is needed to classify the remote sensing images to extract information. We are experimenting with both supervised and unsupervised classification. Here we compare the different classification methods and their performances. It is found that Mahalanobis classifier performed the best in our...

  19. A mutual information-Dempster-Shafer based decision ensemble system for land cover classification of hyperspectral data

    Science.gov (United States)

    Pahlavani, Parham; Bigdeli, Behnaz

    2016-12-01

    Hyperspectral images contain extremely rich spectral information that offer great potential to discriminate between various land cover classes. However, these images are usually composed of tens or hundreds of spectrally close bands, which result in high redundancy and great amount of computation time in hyperspectral classification. Furthermore, in the presence of mixed coverage pixels, crisp classifiers produced errors, omission and commission. This paper presents a mutual information-Dempster-Shafer system through an ensemble classification approach for classification of hyperspectral data. First, mutual information is applied to split data into a few independent partitions to overcome high dimensionality. Then, a fuzzy maximum likelihood classifies each band subset. Finally, Dempster-Shafer is applied to fuse the results of the fuzzy classifiers. In order to assess the proposed method, a crisp ensemble system based on a support vector machine as the crisp classifier and weighted majority voting as the crisp fusion method are applied on hyperspectral data. Furthermore, a dimension reduction system is utilized to assess the effectiveness of mutual information band splitting of the proposed method. The proposed methodology provides interesting conclusions on the effectiveness and potentiality of mutual information-Dempster-Shafer based classification of hyperspectral data.

  20. Hyperspectral image data compression based on DSP

    Science.gov (United States)

    Fan, Jiming; Zhou, Jiankang; Chen, Xinhua; Shen, Weimin

    2010-11-01

    The huge data volume of hyperspectral image challenges its transportation and store. It is necessary to find an effective method to compress the hyperspectral image. Through analysis and comparison of current various algorithms, a mixed compression algorithm based on prediction, integer wavelet transform and embedded zero-tree wavelet (EZW) is proposed in this paper. We adopt a high-powered Digital Signal Processor (DSP) of TMS320DM642 to realize the proposed algorithm. Through modifying the mixed algorithm and optimizing its algorithmic language, the processing efficiency of the program was significantly improved, compared the non-optimized one. Our experiment show that the mixed algorithm based on DSP runs much faster than the algorithm on personal computer. The proposed method can achieve the nearly real-time compression with excellent image quality and compression performance.

  1. Lossless compression of hyperspectral images based on the prediction error block

    Science.gov (United States)

    Li, Yongjun; Li, Yunsong; Song, Juan; Liu, Weijia; Li, Jiaojiao

    2016-05-01

    A lossless compression algorithm of hyperspectral image based on distributed source coding is proposed, which is used to compress the spaceborne hyperspectral data effectively. In order to make full use of the intra-frame correlation and inter-frame correlation, the prediction error block scheme are introduced. Compared with the scalar coset based distributed compression method (s-DSC) proposed by E.Magli et al., that is , the bitrate of the whole block is determined by its maximum prediction error, and the s-DSC-classify scheme proposed by Song Juan that is based on classification and coset coding, the prediction error block scheme could reduce the bitrate efficiently. Experimental results on hyperspectral images show that the proposed scheme can offer both high compression performance and low encoder complexity and decoder complexity, which is available for on-board compression of hyperspectral images.

  2. Online unmixing of multitemporal hyperspectral images accounting for spectral variability

    CERN Document Server

    Thouvenin, Pierre-Antoine; Tourneret, Jean-Yves

    2015-01-01

    Hyperspectral unmixing is aimed at identifying the reference spectral signatures composing an hyperspectral image -- referred to as endmembers -- and their relative abundance fractions in each pixel. In practice, the identified signatures may vary spectrally from an image to another due to varying acquisition conditions inducing possibly significant estimation errors. Against this background, hyperspectral unmixing of several images acquired over the same area is of considerable interest. Indeed, such an analysis enables the endmembers of the scene to be tracked and the corresponding endmember variability to be characterized. Sequential endmember estimation from a set of hyperspectral images is expected to provide improved performance when compared to methods analyzing the images independently. However, the significant size of hyperspectral data precludes the use of batch procedures to jointly estimate the mixture parameters of a sequence of hyperspectral images. Provided that each elementary component is pre...

  3. Unified Lagrangian neural network method for subpixel classification in hyperspectral imagery

    Science.gov (United States)

    Szu, Harold H.; Ren, Hsuan

    2001-03-01

    Because of the foot print of a pixel is relative large, one pixel usually contains more than one material. The spectrum of each pixel vector can be considered as spectral mixture of all materials present in that pixel. How to unmix the spectra while dealing with constraint at the same time is a challenging problem in spectral mixture analysis. One approach is supervised constrained least squares approach, which has full knowledge about the spectrum signatures of endmembers resident in the image scene. The other is unsupervised unmixing method using Lagrangian Artificial Neural Network (LANN) with no knowledge about the image scene. The concept of unit-sum constraint is identical in both cases, the difference is whether one uses the known spectral characteristics for the supervised training or not. In practice, we need both methodologies for efficiency reasons, since there is 'curse of dimensionality' (too large degree of freedom in hyperspectral). The supervised one can reduce the search and ID size and the unsupervised one can discover the unknown interferences and thus in constrained least squares algorithms help ID. The constrained least squares method has been discussed in digital signal processing and applied to hyperspectral imagery while LANN to multispectral imagery. In this paper, we expend LANN to hyperspectral image classification and also discuss the relationship between the constrained least squares method and LANN. These two methods alleviate this problem by adopting the Lagrange multiplier in neural network to relax the sum-to-one constraint. To evaluate this designed algorithm a series of experiments using Airborne Visible/Infrared Imaging Spectrometer (AVIRIS images are conducted to show its potential usefulness in hyperspectral image classification.

  4. Spectral-spatial Classification of Hyperspectral Images Based on Neighborhood Collaboration%基于近邻协同的高光谱图像谱-空联合分类

    Institute of Scientific and Technical Information of China (English)

    倪鼎; 马洪兵

    2015-01-01

    遥感高光谱成像能够获得丰富的地物光谱信息,为高精度的地物分析提供了可能。针对高光谱图像分类中通常面临的数据维数高、标记样本少、计算量大等问题,提出了一种简单有效的谱–空联合分类方法。利用高光谱图像丰富的光谱信息和地物分布的空间平滑特性,该算法首先对光谱数据进行特征提取和空间滤波,然后利用本文提出的基于近邻协同的支持向量机(Neighborhood collaborative support vector machine, NC-SVM)进行分类。近邻协同进一步利用地物分布的空间平滑特性,通过联合空间近邻的判决信息进行中心像素的类别判定,有效减小了只有少量训练样本下的错分概率。实验表明,相比已有的相关方法,该算法在不明显增加计算量的情况下可获得更高的分类正确率,能够实现少量训练样本下高光谱图像的快速高精度分类。%Hyperspectral imaging can provide rich spectral information about land covers, allowing detailed analysis of the materials on the earth. To address the high dimensionality, lack of sufficient labeled samples, and computationally intensive problems involved in hyperspectral image classification, this article presents a simple and efficient method to realize high accuracy classification with a limited training set. Taking advantage of the rich spectral features and the spatially homogeneous property of land covers0 distribution, the proposed method firstly employs feature extraction and spatial filtering to reduce the dimension and the noise of the original hyperspectral data, and then uses the proposed neighborhood collaborative support vector machine (NC-SVM) to classify each pixel. The NC-SVM further exploits the spatially homogeneous property of land covers0 distribution by combining the discriminant information of neighboring pix-els to make reliable class judgement for the central one. Neighborhood collaboration can efficiently reduce

  5. Hyperspectral Imaging for Cancer Surgical Margin Delineation: Registration of Hyperspectral and Histological Images.

    Science.gov (United States)

    Lu, Guolan; Halig, Luma; Wang, Dongsheng; Chen, Zhuo Georgia; Fei, Baowei

    2014-03-12

    The determination of tumor margins during surgical resection remains a challenging task. A complete removal of malignant tissue and conservation of healthy tissue is important for the preservation of organ function, patient satisfaction, and quality of life. Visual inspection and palpation is not sufficient for discriminating between malignant and normal tissue types. Hyperspectral imaging (HSI) technology has the potential to noninvasively delineate surgical tumor margin and can be used as an intra-operative visual aid tool. Since histological images provide the ground truth of cancer margins, it is necessary to warp the cancer regions in ex vivo histological images back to in vivo hyperspectral images in order to validate the tumor margins detected by HSI and to optimize the imaging parameters. In this paper, principal component analysis (PCA) is utilized to extract the principle component bands of the HSI images, which is then used to register HSI images with the corresponding histological image. Affine registration is chosen to model the global transformation. A B-spline free form deformation (FFD) method is used to model the local non-rigid deformation. Registration experiment was performed on animal hyperspectral and histological images. Experimental results from animals demonstrated the feasibility of the hyperspectral imaging method for cancer margin detection.

  6. AOTF hyperspectral microscopic imaging for foodborne pathogenic bacteria detection

    Science.gov (United States)

    Park, Bosoon; Lee, Sangdae; Yoon, Seung-Chul; Sundaram, Jaya; Windham, William R.; Hinton, Arthur, Jr.; Lawrence, Kurt C.

    2011-06-01

    Hyperspectral microscope imaging (HMI) method which provides both spatial and spectral information can be effective for foodborne pathogen detection. The AOTF-based hyperspectral microscope imaging method can be used to characterize spectral properties of biofilm formed by Salmonella enteritidis as well as Escherichia coli. The intensity of spectral imagery and the pattern of spectral distribution varied with system parameters (integration time and gain) of HMI system. The preliminary results demonstrated determination of optimum parameter values of HMI system and the integration time must be no more than 250 ms for quality image acquisition from biofilm formed by S. enteritidis. Among the contiguous spectral imagery between 450 and 800 nm, the intensity of spectral images at 498, 522, 550 and 594 nm were distinctive for biofilm; whereas, the intensity of spectral images at 546 nm was distinctive for E. coli. For more accurate comparison of intensity from spectral images, a calibration protocol, using neutral density filters and multiple exposures, need to be developed to standardize image acquisition. For the identification or classification of unknown food pathogen samples, ground truth regions-of-interest pixels need to be selected for "spectrally pure fingerprints" for the Salmonella and E. coli species.

  7. Compressive Hyperspectral Imaging and Anomaly Detection

    Science.gov (United States)

    2010-02-01

    Examples include the discrete cosine basis and various wavelets based bases. They have been thoroughly studied and widely considered in applications...the desired jointly sparse a"s, one shall adjust a and b. 4.4 Hyperspectral Image Reconstruction and Denoising We apply the model x* = Da’ + e! to...iteration for compressive sensing and sparse denoising ,’" Communications in Mathematical Sciences , 2008. W. Yin, "Analysis and generalizations of

  8. A DIMENSION REDUCTION-BASED METHOD FOR CLASSIFICATION OF HYPERSPECTRAL AND LIDAR DATA

    Directory of Open Access Journals (Sweden)

    B. Abbasi

    2015-12-01

    Full Text Available The existence of various natural objects such as grass, trees, and rivers along with artificial manmade features such as buildings and roads, make it difficult to classify ground objects. Consequently using single data or simple classification approach cannot improve classification results in object identification. Also, using of a variety of data from different sensors; increase the accuracy of spatial and spectral information. In this paper, we proposed a classification algorithm on joint use of hyperspectral and Lidar (Light Detection and Ranging data based on dimension reduction. First, some feature extraction techniques are applied to achieve more information from Lidar and hyperspectral data. Also Principal component analysis (PCA and Minimum Noise Fraction (MNF have been utilized to reduce the dimension of spectral features. The number of 30 features containing the most information of the hyperspectral images is considered for both PCA and MNF. In addition, Normalized Difference Vegetation Index (NDVI has been measured to highlight the vegetation. Furthermore, the extracted features from Lidar data calculated based on relation between every pixel of data and surrounding pixels in local neighbourhood windows. The extracted features are based on the Grey Level Co-occurrence Matrix (GLCM matrix. In second step, classification is operated in all features which obtained by MNF, PCA, NDVI and GLCM and trained by class samples. After this step, two classification maps are obtained by SVM classifier with MNF+NDVI+GLCM features and PCA+NDVI+GLCM features, respectively. Finally, the classified images are fused together to create final classification map by decision fusion based majority voting strategy.

  9. Morphology-based fusion method of hyperspectral image

    Science.gov (United States)

    Yue, Song; Zhang, Zhijie; Ren, Tingting; Wang, Chensheng; Yu, Hui

    2014-11-01

    Hyperspectral image analysis method is widely used in all kinds of application including agriculture identification and forest investigation and atmospheric pollution monitoring. In order to accurately and steadily analyze hyperspectral image, considering the spectrum and spatial information which is provided by hyperspectral data together is necessary. The hyperspectral image has the characteristics of large amount of wave bands and information. Corresponding to the characteristics of hyperspectral image, a fast image fusion method that can fuse the hyperspectral image with high fidelity is studied and proposed in this paper. First of all, hyperspectral image is preprocessed before the morphological close operation. The close operation is used to extract wave band characteristic to reduce dimensionality of hyperspectral image. The spectral data is smoothed at the same time to avoid the discontinuity of the data by combination of spatial information and spectral information. On this basis, Mean-shift method is adopted to register key frames. Finally, the selected key frames by fused into one fusing image by the pyramid fusion method. The experiment results show that this method can fuse hyper spectral image in high quality. The fused image's attributes is better than the original spectral images comparing to the spectral images and reach the objective of fusion.

  10. Hyperspectral imaging of skin and lung cancers

    Science.gov (United States)

    Zherdeva, Larisa A.; Bratchenko, Ivan A.; Alonova, Marina V.; Myakinin, Oleg O.; Artemyev, Dmitry N.; Moryatov, Alexander A.; Kozlov, Sergey V.; Zakharov, Valery P.

    2016-04-01

    The problem of cancer control requires design of new approaches for instrumental diagnostics, as the accuracy of cancer detection on the first step of diagnostics in clinics is slightly more than 50%. In this study, we present a method of visualization and diagnostics of skin and lung tumours based on registration and processing of tissues hyperspectral images. In a series of experiments registration of hyperspectral images of skin and lung tissue samples is carried out. Melanoma, basal cell carcinoma, nevi and benign tumours are studied in skin ex vivo and in vivo experiments; adenocarcinomas and squamous cell carcinomas are studied in ex vivo lung experiments. In a series of experiments the typical features of diffuse reflection spectra for pathological and normal tissues were found. Changes in tissues morphology during the tumour growth lead to the changes of blood and pigments concentration, such as melanin in skin. That is why tumours and normal tissues maybe differentiated with information about spectral response in 500-600 nm and 600 - 670 nm areas. Thus, hyperspectral imaging in the visible region may be a useful tool for cancer detection as it helps to estimate spectral properties of tissues and determine malignant regions for precise resection of tumours.

  11. Subpixel Target Enhancement in Hyperspectral Images

    Directory of Open Access Journals (Sweden)

    Manoj K. Arora

    2013-01-01

    Full Text Available Hyperspectral images due to their higher spectral resolution are increasingly being used for various remote sensing applications including information extraction at subpixel level. Typically whenever an object gets spectrally resolved but not spatially, mixed pixels in the images result. Numerous man made and/or natural disparate targets may thus occur inside such mixed pixels giving rise to subpixel target detection problem. Various spectral unmixing models such as linear mixture modeling (LMM are in vogue to recover components of a mixed pixel. Spectral unmixing outputs both the endmember spectrum and their corresponding abundance fractions inside the pixel. It, however, does not provide spatial distribution of these abundance fractions within a pixel. This limits the applicability of hyperspectral data for subpixel target detection. In this paper, a new inverse Euclidean distance based super-resolution mapping method has been presented. In this method, the subpixel target detection is performed by adjusting spatial distribution of abundance fraction within a pixel of an hyperspectral image. Results obtained at different resolutions indicate that super-resolution mapping may effectively be utilized in enhancing the target detection at sub-pixel level.

  12. Subpixel Target Enhancement in Hyperspectral Images

    Directory of Open Access Journals (Sweden)

    Manoj K. Arora

    2013-01-01

    Full Text Available Hyperspectral images due to their higher spectral resolution are increasingly being used for various remote sensing applications including information extraction at subpixel level. Typically whenever an object gets spectrally resolved but not spatially, mixed pixels in the images result. Numerous man made and/or natural disparatetar gets may thus occur inside such mixed pixels giving rise to subpixel target detection problem. Various spectral unmixing models such as linear mixture modeling (LMM are in vogue to recover components of a mixed pixel. Spectral unmixing outputs both the endmember spectrum and their corresponding a bundance fractions inside the pixel. It, however, does not provide spatial distribution of these abundance fractions within a pixel. This limits the applicability of hyperspectral data for subpixel target detection. In this paper, a new inverse Euclidean distance based super-resolution mapping method has been presented. In this method, the subpixel target detection is performed by adjusting spatial distribution of abundance fraction within a pixel of an hyperspectral image. Results obtainedat different resolutions indicate that super-resolution mapping may effectively be utilized in enhancing the target detection at sub-pixel level.Defence Science Journal, 2013, 63(1, pp.63-68, DOI:http://dx.doi.org/10.14429/dsj.63.3765

  13. Hyperspectral Transformation from EO-1 ALI Imagery Using Pseudo-Hyperspectral Image Synthesis Algorithm

    Science.gov (United States)

    Tien Hoang, Nguyen; Koike, Katsuaki

    2016-06-01

    Hyperspectral remote sensing is more effective than multispectral remote sensing in many application fields because of having hundreds of observation bands with high spectral resolution. However, hyperspectral remote sensing resources are limited both in temporal and spatial coverage. Therefore, simulation of hyperspectral imagery from multispectral imagery with a small number of bands must be one of innovative topics. Based on this background, we have recently developed a method, Pseudo-Hyperspectral Image Synthesis Algorithm (PHISA), to transform Landsat imagery into hyperspectral imagery using the correlation of reflectance at the corresponding bands between Landsat and EO-1 Hyperion data. This study extends PHISA to simulate pseudo-hyperspectral imagery from EO-1 ALI imagery. The pseudo-hyperspectral imagery has the same number of bands as that of high-quality Hyperion bands and the same swath width as ALI scene. The hyperspectral reflectance data simulated from the ALI data show stronger correlation with the original Hyperion data than the one simulated from Landsat data. This high correlation originates from the concurrent observation by the ALI and Hyperion sensors that are on-board the same satellite. The accuracy of simulation results are verified by a statistical analysis and a surface mineral mapping. With a combination of the advantages of both ALI and Hyperion image types, the pseudo-hyperspectral imagery is proved to be useful for detailed identification of minerals for the areas outside the Hyperion coverage.

  14. 基于高光谱技术的培养基上细菌菌落分类方法研究%Classification of Bacterial Colonies on Agar Plates Using Hyperspectral Imaging Technology

    Institute of Scientific and Technical Information of China (English)

    余伟; 彭宽宽; 陈伟; 穆渴心; 谭臣; 王湘如; 冯耀泽

    2016-01-01

    利用高光谱技术对培养基上细菌(大肠杆菌、李斯特菌和金黄色葡萄球菌)菌落进行快速识别和分类。采集琼脂培养基上细菌菌落的高光谱反射图像(390~1040 nm),在对波段差图像进行大津阈值分割的基础上自动提取细菌菌落光谱,并建立细菌分类检测的全波长和简化偏最小二乘判别( PLS-DA)模型。全波长模型对预测集样本的分类准确率和置信预测分类准确率分别为100%和95.9%。此外,利用竞争性自适应重加权算法( CARS)、遗传算法( GA)和最小角回归算法( LARS-Lasso)进行波长优选并建立对应简化模型。其中,CARS简化模型在精度、稳定性及分类准确率方面均优于GA和LARS-Lasso简化模型,其对预测集样本的分类准确率和置信预测分类准确率分别达到了100%和98.0%。研究表明,高光谱是一种细菌菌落高精度、快速、无损识别检测的有效方法。简化模型中优选的波长可以为开发低成本检测仪器提供理论依据。%Rapid detection and classification of bacteria colonies ( Escherichia coli, Listeria monocytogens and Staphylococcus aureus) were investigated by using hyperspectral imaging. The hyperspectral reflectance images (390-1040 nm ) of bacterial colonies on agar plates were collected. Bacterial spectra were extracted automatically based on the masks produced by segmenting a band difference image using the OTSU method. Full wavelength and simplified PLS-DA models were established for classification of bacterial colonies. For the full wavelength model, the overall correct classification rate ( OCCR) and confident OCCR for the prediction set were 100% and 95. 9%, respectively. Besides, competitive adaptive reweighted sampling ( CARS), genetic algorithm ( GA ) and least angle regression-least absolute shrinkage and selection operator ( LARS-Lasso) were used to select feature wavelengths for the development of simplified models. Among them, the CARS

  15. Hyperspectral image classification by combining empirical mode decomposition with Gabor filtering%结合经验模态分解和Gabor滤波的高光谱图像分类

    Institute of Scientific and Technical Information of China (English)

    王立国; 宛宇美; 路婷婷; 杨月霜

    2016-01-01

    Considering the shortcomings of traditional texture extraction methods implemented in original data space, in this paper we use the empirical mode decomposition theory to extract the intrinsic mode components of a distinct spatial structure from a hyperspectral image, and perform Gabor filtering on these extracted components. We trans⁃ferred the traditional texture extraction method to the transform domain. In this way, we propose using a high⁃preci⁃sion texture extraction algorithm for decomposing and integrating spatial information based on two⁃dimensional em⁃pirical mode decomposition. We carried out simulation tests on two datasets, and the results show that the improved algorithm effectively improves the classification accuracy of hyperspectral images and has good noise suppression performance. The proposed algorithm is thus clearly superior to the traditional Gabor⁃PCA algorithm, and can mine hyperspectral image spatial information to a greater extent.%针对传统实施于原始数据空间的纹理提取方法的不足,采用经验模态分解理论提取高光谱图像中空间结构明显的固有模态分量,并在提取出的分量上进行Gabor滤波操作,将传统纹理提取方式转移到变换域上进行,提出了一种基于二维经验模态分解融合空间信息的高精度纹理提取算法。对两个数据集进行仿真实验,实验结果表明改进算法有效地提高了高光谱图像分类精度且抗噪性能良好,提出算法性能明显优于传统Gabor⁃PCA算法,能够更大程度挖掘高光谱图像空间信息。

  16. 基于最小关联窗口的高光谱图像非监督分类%Unsupervised Classification of Hyperspectral Images Based on Min-related-window

    Institute of Scientific and Technical Information of China (English)

    岳江; 柏连发; 张毅; 徐杭威

    2012-01-01

    为了提高分类精度和减少麻点现象,该文在最小关联窗口模型基础上,提出了一种高光谱图像非监督分类方法.首先该方法推导了一个目标辨识量——泛光谱曲线;基于此辨识量和图像空间一致性,提出了像元最小关联窗口继承式分类.类间距离使用巴氏距离( Bhattacharyya distance)度量,将相似类合并,完成最终分类.最后利用美国AVIRIS数据进行性能评价,并分别与K均值(K-MEANS)和迭代自组织数据分析技术(ISODATA)进行了性能对比.实验结果表明,该方法分类精度和麻点现象均优于K-MEANS和ISODATA.%In order to improve classification accuracy and reduce discrete points, an unsupervised classification of hyperspectral images based on a min-related-window model is proposed. A variable called ex-spectral curve is introduced for target distinguishing;with this variable and spatial coherence property, the algorithm of inherited classification based on a pixel's min-related-window is proposed. Bhattacharyya distance is used to measure an interlayer distance,and similar classes are combined to achieve the classification. AVIRIS data are utilized to evaluate the performance of the proposed algorithm, which is compared with K-MEANS and ISODATA. Experimental results show that the proposed algorithm outperforms K-MEANS and ISODATA in classification accuracy and has less discrete points.

  17. Onboard Image Processing System for Hyperspectral Sensor

    Directory of Open Access Journals (Sweden)

    Hiroki Hihara

    2015-09-01

    Full Text Available Onboard image processing systems for a hyperspectral sensor have been developed in order to maximize image data transmission efficiency for large volume and high speed data downlink capacity. Since more than 100 channels are required for hyperspectral sensors on Earth observation satellites, fast and small-footprint lossless image compression capability is essential for reducing the size and weight of a sensor system. A fast lossless image compression algorithm has been developed, and is implemented in the onboard correction circuitry of sensitivity and linearity of Complementary Metal Oxide Semiconductor (CMOS sensors in order to maximize the compression ratio. The employed image compression method is based on Fast, Efficient, Lossless Image compression System (FELICS, which is a hierarchical predictive coding method with resolution scaling. To improve FELICS’s performance of image decorrelation and entropy coding, we apply a two-dimensional interpolation prediction and adaptive Golomb-Rice coding. It supports progressive decompression using resolution scaling while still maintaining superior performance measured as speed and complexity. Coding efficiency and compression speed enlarge the effective capacity of signal transmission channels, which lead to reducing onboard hardware by multiplexing sensor signals into a reduced number of compression circuits. The circuitry is embedded into the data formatter of the sensor system without adding size, weight, power consumption, and fabrication cost.

  18. Onboard Image Processing System for Hyperspectral Sensor.

    Science.gov (United States)

    Hihara, Hiroki; Moritani, Kotaro; Inoue, Masao; Hoshi, Yoshihiro; Iwasaki, Akira; Takada, Jun; Inada, Hitomi; Suzuki, Makoto; Seki, Taeko; Ichikawa, Satoshi; Tanii, Jun

    2015-09-25

    Onboard image processing systems for a hyperspectral sensor have been developed in order to maximize image data transmission efficiency for large volume and high speed data downlink capacity. Since more than 100 channels are required for hyperspectral sensors on Earth observation satellites, fast and small-footprint lossless image compression capability is essential for reducing the size and weight of a sensor system. A fast lossless image compression algorithm has been developed, and is implemented in the onboard correction circuitry of sensitivity and linearity of Complementary Metal Oxide Semiconductor (CMOS) sensors in order to maximize the compression ratio. The employed image compression method is based on Fast, Efficient, Lossless Image compression System (FELICS), which is a hierarchical predictive coding method with resolution scaling. To improve FELICS's performance of image decorrelation and entropy coding, we apply a two-dimensional interpolation prediction and adaptive Golomb-Rice coding. It supports progressive decompression using resolution scaling while still maintaining superior performance measured as speed and complexity. Coding efficiency and compression speed enlarge the effective capacity of signal transmission channels, which lead to reducing onboard hardware by multiplexing sensor signals into a reduced number of compression circuits. The circuitry is embedded into the data formatter of the sensor system without adding size, weight, power consumption, and fabrication cost.

  19. Development of a new spectral library classifier for airborne hyperspectral images on heterogeneous environments

    OpenAIRE

    Mende, Andre; Heiden, Uta; Bachmann, Martin; Hoja, Danielle; Buchroithner, Manfred

    2011-01-01

    The classification of hyperspectral images on heterogeneous environments without prior knowledge about the study area is a challenging task. Finding potential pure spectral signatures or endmembers (EM) of the various surface materials within an image is essential for obtaining accurate classification results. Automated endmember selection techniques, in many cases, return an unlabelled result without a relationship to a known material. This study demonstrates the potential of an automated sp...

  20. A maximum noise fraction transform with improved noise estimation for hyperspectral images

    Institute of Scientific and Technical Information of China (English)

    LIU Xiang; ZHANG Bing; GAO LianRu; CHEN DongMei

    2009-01-01

    Feature extraction is often performed to reduce spectral dimension of hyperspectral images before image classification.The maximum noise fraction (MNF) transform is one of the most commonly used spectral feature extraction methods.The spectral features in several bands of hyperspectral images are submerged by the noise.The MNF transform is advantageous over the principle component (PC) transform because it takes the noise information in the spatial domain into consideration.However,the experiments described in this paper demonstrate that classification accuracy is greatly influenced by the MNF transform when the ground objects are mixed together.The underlying mechanism of it is revealed and analyzed by mathematical theory.In order to improve the performance of classification after feature extraction when ground objects are mixed in hyperspectral images,a new MNF transform,with an Improved method of estimating hyperspectral Image noise covariance matrix (NCM),is presented.This improved MNF transform is applied to both the simulated data and real data.The results show that compared with the classical MNF transform,this new method enhanced the ability of feature extraction and increased classification accuracy.

  1. Investigation of Latent Traces Using Infrared Reflectance Hyperspectral Imaging

    Science.gov (United States)

    Schubert, Till; Wenzel, Susanne; Roscher, Ribana; Stachniss, Cyrill

    2016-06-01

    The detection of traces is a main task of forensics. Hyperspectral imaging is a potential method from which we expect to capture more fluorescence effects than with common forensic light sources. This paper shows that the use of hyperspectral imaging is suited for the analysis of latent traces and extends the classical concept to the conservation of the crime scene for retrospective laboratory analysis. We examine specimen of blood, semen and saliva traces in several dilution steps, prepared on cardboard substrate. As our key result we successfully make latent traces visible up to dilution factor of 1:8000. We can attribute most of the detectability to interference of electromagnetic light with the water content of the traces in the shortwave infrared region of the spectrum. In a classification task we use several dimensionality reduction methods (PCA and LDA) in combination with a Maximum Likelihood classifier, assuming normally distributed data. Further, we use Random Forest as a competitive approach. The classifiers retrieve the exact positions of labelled trace preparation up to highest dilution and determine posterior probabilities. By modelling the classification task with a Markov Random Field we are able to integrate prior information about the spatial relation of neighboured pixel labels.

  2. A comparative study on manifold learning of hyperspectral data for land cover classification

    Science.gov (United States)

    Ozturk, Ceyda Nur; Bilgin, Gokhan

    2015-03-01

    This paper focuses on the land cover classification problem by employing a number of manifold learning algorithms in the feature extraction phase, then by running single and ensemble of classifiers in the modeling phase. Manifolds are learned on training samples selected randomly within available data, while the transformation of the remaining test samples is realized for linear and nonlinear methods via the learnt mappings and a radial-basis function neural network based interpolation method, respectively. The classification accuracies of the original data and the embedded manifolds are investigated with several classifiers. Experimental results on a 200-band hyperspectral image indicated that support vector machine was the best classifier for most of the methods, being nearly as accurate as the best classification rate of the original data. Furthermore, our modified version of random subspace classifier could even outperform the classification accuracy of the original data for local Fisher's discriminant analysis method despite of a considerable decrease in the extrinsic dimension.

  3. 条带消除对于高光谱图像分类效果的影响--以CHRIS/PROBA图像为例%Effect of destriping on hyperspectral image classification, take CHRIS/PROBA as an example

    Institute of Scientific and Technical Information of China (English)

    梁建; 张杰; 马毅

    2015-01-01

    This paper takes CHRIS/PROBA hyperspectral images as an example. Image classifications are carried out on these original and destriped images by both unsupervised algorithm (ISODATA)and supervised algorithm (Maximum Likelihood, ML and Support Vector Machine, SVM ), and the results are analyzed. The results show that, destriping can well improve the visual effect of the classification by eliminating the burr between patches of different classes. This is very important for some researches that focus on the shape, geometry and distribution, such as landscape ecology. However, destriping cannot significantly improve the precision of classification;the maximum value of accuracy improvment is less than 2%.%以CHRIS/PROBA高光谱图像数据为例,使用非监督分类的ISODATA和监督分类的最大似然法、支持向量机等3种经典的图像分类算法,对消条带前后的图像分别开展分类实验,并对分类结果做了分析。实验结果表明,消条带处理可以较好地改善高光谱图像分类结果的目视效果,能够消除不同类别斑块边缘因条带而产生的“毛刺”现象,这对地物斑块的形状及几何分布敏感的研究(如景观生态学)至关重要;但消条带处理对于提高分类精度的效果并不显著,精度提高最大值不到2%。

  4. Tree Species Classification Using Hyperspectral Imagery: A Comparison of Two Classifiers

    Directory of Open Access Journals (Sweden)

    Laurel Ballanti

    2016-05-01

    Full Text Available The identification of tree species can provide a useful and efficient tool for forest managers for planning and monitoring purposes. Hyperspectral data provide sufficient spectral information to classify individual tree species. Two non-parametric classifiers, support vector machines (SVM and random forest (RF, have resulted in high accuracies in previous classification studies. This research takes a comparative classification approach to examine the SVM and RF classifiers in the complex and heterogeneous forests of Muir Woods National Monument and Kent Creek Canyon in Marin County, California. The influence of object- or pixel-based training samples and segmentation size on the object-oriented classification is also explored. To reduce the data dimensionality, a minimum noise fraction transform was applied to the mosaicked hyperspectral image, resulting in the selection of 27 bands for the final classification. Each classifier was also assessed individually to identify any advantage related to an increase in training sample size or an increase in object segmentation size. All classifications resulted in overall accuracies above 90%. No difference was found between classifiers when using object-based training samples. SVM outperformed RF when additional training samples were used. An increase in training samples was also found to improve the individual performance of the SVM classifier.

  5. Hyperspectral imaging for detection of black tip damage in wheat kernels

    Science.gov (United States)

    Delwiche, Stephen R.; Yang, I.-Chang; Kim, Moon S.

    2009-05-01

    A feasibility study was conducted on the use of hyperspectral imaging to differentiate sound wheat kernels from those with the fungal condition called black point or black tip. Individual kernels of hard red spring wheat were loaded in indented slots on a blackened machined aluminum plate. Damage conditions, determined by official (USDA) inspection, were either sound (no damage) or damaged by the black tip condition alone. Hyperspectral imaging was separately performed under modes of reflectance from white light illumination and fluorescence from UV light (~380 nm) illumination. By cursory inspection of wavelength images, one fluorescence wavelength (531 nm) was selected for image processing and classification analysis. Results indicated that with this one wavelength alone, classification accuracy can be as high as 95% when kernels are oriented with their dorsal side toward the camera. It is suggested that improvement in classification can be made through the inclusion of multiple wavelength images.

  6. Hyperspectral Imaging for Defect Detection of Pickling Cucumber

    Science.gov (United States)

    This book chapter reviews the recent progress on hyperspectral imaging technology for defect inspection of pickling cucumbers. The chapter first describes near-infrared hyperspectral reflectance imaging technique for the detection of bruises on pickling cucumbers. The technique showed good detection...

  7. Accounting for Variance in Hyperspectral Data Coming from Limitations of the Imaging System

    Science.gov (United States)

    Shurygin, B.; Shestakova, M.; Nikolenko, A.; Badasen, E.; Strakhov, P.

    2016-06-01

    Over the course of the past few years, a number of methods was developed to incorporate hyperspectral imaging specifics into generic data mining techniques, traditionally used for hyperspectral data processing. Projection pursuit methods embody the largest class of methods empoyed for hyperspectral image data reduction, however, they all have certain drawbacks making them either hard to use or inefficient. It has been shown that hyperspectral image (HSI) statistics tend to display "heavy tails" (Manolakis2003)(Theiler2005), rendering most of the projection pursuit methods hard to use. Taking into consideration the magnitude of described deviations of observed data PDFs from normal distribution, it is apparent that a priori knowledge of variance in data caused by the imaging system is to be employed in order to efficiently classify objects on HSIs (Kerr, 2015), especially in cases of wildly varying SNR. A number of attempts to describe this variance and compensating techniques has been made (Aiazzi2006), however, new data quality standards are not yet set and accounting for the detector response is made under large set of assumptions. Current paper addresses the issue of hyperspectral image classification in the context of different variance sources based on the knowledge of calibration curves (both spectral and radiometric) obtained for each pixel of imaging camera. A camera produced by ZAO NPO Lepton (Russia) was calibrated and used to obtain a test image. A priori known values of SNR and spectral channel cross-correlation were incorporated into calculating test statistics used in dimensionality reduction and feature extraction. Expectation-Maximization classification algorithm modification for non-Gaussian model as described by (Veracini2010) was further employed. The impact of calibration data coarsening by ignoring non-uniformities on false alarm rate was studied. Case study shows both regions of scene-dominated variance and sensor-dominated variance, leading

  8. Online Unmixing of Multitemporal Hyperspectral Images Accounting for Spectral Variability.

    Science.gov (United States)

    Thouvenin, Pierre-Antoine; Dobigeon, Nicolas; Tourneret, Jean-Yves

    2016-09-01

    Hyperspectral unmixing is aimed at identifying the reference spectral signatures composing a hyperspectral image and their relative abundance fractions in each pixel. In practice, the identified signatures may vary spectrally from an image to another due to varying acquisition conditions, thus inducing possibly significant estimation errors. Against this background, the hyperspectral unmixing of several images acquired over the same area is of considerable interest. Indeed, such an analysis enables the endmembers of the scene to be tracked and the corresponding endmember variability to be characterized. Sequential endmember estimation from a set of hyperspectral images is expected to provide improved performance when compared with methods analyzing the images independently. However, the significant size of the hyperspectral data precludes the use of batch procedures to jointly estimate the mixture parameters of a sequence of hyperspectral images. Provided that each elementary component is present in at least one image of the sequence, we propose to perform an online hyperspectral unmixing accounting for temporal endmember variability. The online hyperspectral unmixing is formulated as a two-stage stochastic program, which can be solved using a stochastic approximation. The performance of the proposed method is evaluated on synthetic and real data. Finally, a comparison with independent unmixing algorithms illustrates the interest of the proposed strategy.

  9. In vivo hyperspectral imaging and differentiation of skin cancer

    Science.gov (United States)

    Zherdeva, Larisa A.; Bratchenko, Ivan A.; Myakinin, Oleg O.; Moryatov, Alexander A.; Kozlov, Sergey V.; Zakharov, Valery P.

    2016-10-01

    Results of hyperspectral imaging analysis for in vivo visualization of skin neoplasms are presented. 16 melanomas, 19 basal cell carcinomas and 10 benign tumors with different stages of neoplasm growth were tested. The HSI system provide skin tissue images with 5 nm spectral resolution in the range of 450-750 nm with automatic stabilization of each frame compensating displacement of the scanning area due to spontaneous macro-movements of the patient. The integrated optical densities in 530-600 and 600-670 nm ranges are used for real-time hemoglobin and melanin distribution imaging in skin tissue. It was shown that the total accuracy of skin cancer identification exceeds 90% and 70% for differentiation of melanomas from BCC and begihn tumors. It was demonstrated the possibility for HSI classification of melanomas of different stages.

  10. High speed measurement of corn seed viability using hyperspectral imaging

    Science.gov (United States)

    Ambrose, Ashabahebwa; Kandpal, Lalit Mohan; Kim, Moon S.; Lee, Wang-Hee; Cho, Byoung-Kwan

    2016-03-01

    Corn is one of the most cultivated crops all over world as food for humans as well as animals. Optimized agronomic practices and improved technological interventions during planting, harvesting and post-harvest handling are critical to improving the quantity and quality of corn production. Seed germination and vigor are the primary determinants of high yield notwithstanding any other factors that may play during the growth period. Seed viability may be lost during storage due to unfavorable conditions e.g. moisture content and temperatures, or physical damage during mechanical processing e.g. shelling, or over heating during drying. It is therefore vital for seed companies and farmers to test and ascertain seed viability to avoid losses of any kind. This study aimed at investigating the possibility of using hyperspectral imaging (HSI) technique to discriminate viable and nonviable corn seeds. A group of corn samples were heat treated by using microwave process while a group of seeds were kept as control group (untreated). The hyperspectral images of corn seeds of both groups were captured between 400 and 2500 nm wave range. Partial least squares discriminant analysis (PLS-DA) was built for the classification of aged (heat treated) and normal (untreated) corn seeds. The model showed highest classification accuracy of 97.6% (calibration) and 95.6% (prediction) in the SWIR region of the HSI. Furthermore, the PLS-DA and binary images were capable to provide the visual information of treated and untreated corn seeds. The overall results suggest that HSI technique is accurate for classification of viable and non-viable seeds with non-destructive manner.

  11. Comparing methods for analysis of biomedical hyperspectral image data

    Science.gov (United States)

    Leavesley, Silas J.; Sweat, Brenner; Abbott, Caitlyn; Favreau, Peter F.; Annamdevula, Naga S.; Rich, Thomas C.

    2017-02-01

    Over the past 2 decades, hyperspectral imaging technologies have been adapted to address the need for molecule-specific identification in the biomedical imaging field. Applications have ranged from single-cell microscopy to whole-animal in vivo imaging and from basic research to clinical systems. Enabling this growth has been the availability of faster, more effective hyperspectral filtering technologies and more sensitive detectors. Hence, the potential for growth of biomedical hyperspectral imaging is high, and many hyperspectral imaging options are already commercially available. However, despite the growth in hyperspectral technologies for biomedical imaging, little work has been done to aid users of hyperspectral imaging instruments in selecting appropriate analysis algorithms. Here, we present an approach for comparing the effectiveness of spectral analysis algorithms by combining experimental image data with a theoretical "what if" scenario. This approach allows us to quantify several key outcomes that characterize a hyperspectral imaging study: linearity of sensitivity, positive detection cut-off slope, dynamic range, and false positive events. We present results of using this approach for comparing the effectiveness of several common spectral analysis algorithms for detecting weak fluorescent protein emission in the midst of strong tissue autofluorescence. Results indicate that this approach should be applicable to a very wide range of applications, allowing a quantitative assessment of the effectiveness of the combined biology, hardware, and computational analysis for detecting a specific molecular signature.

  12. Wide-Field, Deep UV Raman Hyperspectral Imager Project

    Data.gov (United States)

    National Aeronautics and Space Administration — ChemImage Sensor Systems (CISS), teaming with the University of South Carolina, proposes a revolutionary wide-field Raman hyperspectral imaging system capable of...

  13. Dental caries imaging using hyperspectral stimulated Raman scattering microscopy

    Science.gov (United States)

    Wang, Zi; Zheng, Wei; Jian, Lin; Huang, Zhiwei

    2016-03-01

    We report the development of a polarization-resolved hyperspectral stimulated Raman scattering (SRS) imaging technique based on a picosecond (ps) laser-pumped optical parametric oscillator system for label-free imaging of dental caries. In our imaging system, hyperspectral SRS images (512×512 pixels) in both fingerprint region (800-1800 cm-1) and high-wavenumber region (2800-3600 cm-1) are acquired in minutes by scanning the wavelength of OPO output, which is a thousand times faster than conventional confocal micro Raman imaging. SRS spectra variations from normal enamel to caries obtained from the hyperspectral SRS images show the loss of phosphate and carbonate in the carious region. While polarization-resolved SRS images at 959 cm-1 demonstrate that the caries has higher depolarization ratio. Our results demonstrate that the polarization resolved-hyperspectral SRS imaging technique developed allows for rapid identification of the biochemical and structural changes of dental caries.

  14. Using hyperspectral imaging technology to identify diseased tomato leaves

    Science.gov (United States)

    Li, Cuiling; Wang, Xiu; Zhao, Xueguan; Meng, Zhijun; Zou, Wei

    2016-11-01

    In the process of tomato plants growth, due to the effect of plants genetic factors, poor environment factors, or disoperation of parasites, there will generate a series of unusual symptoms on tomato plants from physiology, organization structure and external form, as a result, they cannot grow normally, and further to influence the tomato yield and economic benefits. Hyperspectral image usually has high spectral resolution, not only contains spectral information, but also contains the image information, so this study adopted hyperspectral imaging technology to identify diseased tomato leaves, and developed a simple hyperspectral imaging system, including a halogen lamp light source unit, a hyperspectral image acquisition unit and a data processing unit. Spectrometer detection wavelength ranged from 400nm to 1000nm. After hyperspectral images of tomato leaves being captured, it was needed to calibrate hyperspectral images. This research used spectrum angle matching method and spectral red edge parameters discriminant method respectively to identify diseased tomato leaves. Using spectral red edge parameters discriminant method produced higher recognition accuracy, the accuracy was higher than 90%. Research results have shown that using hyperspectral imaging technology to identify diseased tomato leaves is feasible, and provides the discriminant basis for subsequent disease control of tomato plants.

  15. Study on hyperspectral image technology based on manifold fuzzy clustering for pork quality classification%基于流形模糊聚类算法的高光谱图像猪肉品质分类研究

    Institute of Scientific and Technical Information of China (English)

    曾山; 王海滨; 张凤兵; 白俊

    2015-01-01

    ABSTRACT:Objective The classification of fresh chilled meat, frozen-thawed meat and spoiled meat based on the hyperspectral image technology as research object, and aiming at the problem of high feature dimension, a classification model based on manifold fuzzy clustering algorithm was established.Methods The 2D Gabor wavelet transform was used to extract the 8 texture features of image respectively under 14 characteristic wavelengths reflecting 3 kinds of quality index: drip loss, pH and color and thus forms a 112-dimensional feature variables as the feature of pork quality. Then the pork quality classification model was established through the fuzzy C-means clustering algorithm (FCM) based on Isometric Mapping (ISOMAP) dimensional reduction.ResultsThe experiment of pork quality classification showed that 2D Gabor wavelet transform could effectively extract texture feature of pork. Compared with the traditional FCM clustering algorithm, FCM clustering algorithm based on ISOMAP dimensional reduction could better resolve the clustering problem of high dimensional samples and accurately distinguish the fresh chilled meat, frozen-thawed meat and spoiled meat.ConclusionThe hyperspectral image technology could be used for pork quality classification.%目的:以基于高光谱图像技术的冷鲜肉、解冻肉和变质肉的分类为研究对象,针对特征维数过高的问题,构建了一种基于流形模糊聚类算法的分类模型。方法首先采用二维Gabor小波变换分别提取反应猪肉滴水损失、pH、颜色三种品质指标的14个特征波长下图像的8个纹理特征,组成一个112维的特征变量作为猪肉品质的特征;然后采用基于等距映射降维的模糊 C 均值聚类算法来构建猪肉品质分类模型。结果通过猪肉品质分类实验得出,二维 Gabor 小波变换能较好地提取猪肉的纹理特征;与传统模糊 C 均值聚类算法相比,基于等距映射降维的模糊C均值聚类算

  16. ADVANCES IN HYPERSPECTRAL AND MULTISPECTRAL IMAGE FUSION AND SPECTRAL UNMIXING

    OpenAIRE

    C. Lanaras; E. Baltsavias; K. Schindler

    2015-01-01

    In this work, we jointly process high spectral and high geometric resolution images and exploit their synergies to (a) generate a fused image of high spectral and geometric resolution; and (b) improve (linear) spectral unmixing of hyperspectral endmembers at subpixel level w.r.t. the pixel size of the hyperspectral image. We assume that the two images are radiometrically corrected and geometrically co-registered. The scientific contributions of this work are (a) a simultaneous approa...

  17. Recent applications of hyperspectral imaging in microbiology.

    Science.gov (United States)

    Gowen, Aoife A; Feng, Yaoze; Gaston, Edurne; Valdramidis, Vasilis

    2015-05-01

    Hyperspectral chemical imaging (HSI) is a broad term encompassing spatially resolved spectral data obtained through a variety of modalities (e.g. Raman scattering, Fourier transform infrared microscopy, fluorescence and near-infrared chemical imaging). It goes beyond the capabilities of conventional imaging and spectroscopy by obtaining spatially resolved spectra from objects at spatial resolutions varying from the level of single cells up to macroscopic objects (e.g. foods). In tandem with recent developments in instrumentation and sampling protocols, applications of HSI in microbiology have increased rapidly. This article gives a brief overview of the fundamentals of HSI and a comprehensive review of applications of HSI in microbiology over the past 10 years. Technical challenges and future perspectives for these techniques are also discussed. Copyright © 2015 Elsevier B.V. All rights reserved.

  18. Hyperspectral imaging using ultraviolet light

    OpenAIRE

    Porter, Michael A.

    2005-01-01

    The LINEATE IMAGING NEAR ULTRAVIOLET SPECTROMETER (LINUS) instrument has been used to remotely detect and measure sulfur dioxide (SO2). The sensor was calibrated in the lab, with curves of growth created for the 0.29 0.31 æ - spectral range of the LINUS sensor. Field observations were made of a coal burning plant in St. Johnâ s, Arizona at a range of 537 m. The Salt River Coronado plant stacks were emitting on average about 100 ppm and 200 ppm from the left and right stacks respectively. An...

  19. Hyperspectral Fluorescence and Reflectance Imaging Instrument

    Science.gov (United States)

    Ryan, Robert E.; O'Neal, S. Duane; Lanoue, Mark; Russell, Jeffrey

    2008-01-01

    The system is a single hyperspectral imaging instrument that has the unique capability to acquire both fluorescence and reflectance high-spatial-resolution data that is inherently spatially and spectrally registered. Potential uses of this instrument include plant stress monitoring, counterfeit document detection, biomedical imaging, forensic imaging, and general materials identification. Until now, reflectance and fluorescence spectral imaging have been performed by separate instruments. Neither a reflectance spectral image nor a fluorescence spectral image alone yields as much information about a target surface as does a combination of the two modalities. Before this system was developed, to benefit from this combination, analysts needed to perform time-consuming post-processing efforts to co-register the reflective and fluorescence information. With this instrument, the inherent spatial and spectral registration of the reflectance and fluorescence images minimizes the need for this post-processing step. The main challenge for this technology is to detect the fluorescence signal in the presence of a much stronger reflectance signal. To meet this challenge, the instrument modulates artificial light sources from ultraviolet through the visible to the near-infrared part of the spectrum; in this way, both the reflective and fluorescence signals can be measured through differencing processes to optimize fluorescence and reflectance spectra as needed. The main functional components of the instrument are a hyperspectral imager, an illumination system, and an image-plane scanner. The hyperspectral imager is a one-dimensional (line) imaging spectrometer that includes a spectrally dispersive element and a two-dimensional focal plane detector array. The spectral range of the current imaging spectrometer is between 400 to 1,000 nm, and the wavelength resolution is approximately 3 nm. The illumination system consists of narrowband blue, ultraviolet, and other discrete

  20. Semisupervised Classification of Hyperspectral Image Based on Clustering Kernel and LS-SVM%基于聚类核函数的最小二乘支持向量机高光谱图像半监督分类

    Institute of Scientific and Technical Information of China (English)

    高恒振; 万建伟; 许可; 钱林杰

    2011-01-01

    When classifying large scale hyperspectral image data, there are a lot of unlabeled samples. In order to enhance the classifier's performance by using the information contained in the unlabeled data, this paper presents a semisupervised classification method. The proposed algorithm modifies the kernel function based on the clustering assumption. It assumes that the samples belonged to the same class will be assigned to the same cluster in the clustering at high probability. The algorithm clusters the unlabeled samples using K-means clustering algorithm. The K-means method uses spectral angle to measure the differences between the samples. The bagged kemel is constructed based on the multi clustering results of data set. Then the bagged kernel is combined with the RBF kernel using sum or product operation. So the information in the unlabeled samples is merged into the classification procedure. The proposed algorithm adopts the least squares SVM ( LS-SVM). Instead of solving the quadratic problem of SVM, the LS-SVM changes it to linear equations. The proposed method is validated by the hyperspectral data. In the experiments the proposed method shows some superiority.%针对大规模的高光谱数据分类,为了利用未标签样本所含信息,来提升分类器性能,提出了一种半监督分类算法.该算法根据聚类假设,即属于同一类地物的样本点在聚类中被分为同一类的可能性较大的原则来改进核函数,采用基于光谱角度最的K均值聚类算法对样本集进行聚类,根据多次聚类的结果,构造包袋核函数,然后利用加法和乘法运算将包袋核函数和RBF核函数组合成新的核函数,从而把未标签样本信息融入分类器.而且采用最小二乘支持向量机,将标准支持向量机的二次规划问题转换为求解线性方程组的问题.高光谱实测数据实验表明了本文方法的优越性.

  1. Performance of Spectral Angle Mapper and Parallelepiped Classifiers in Agriculture Hyperspectral Image

    Directory of Open Access Journals (Sweden)

    Sahar A. El_Rahman

    2016-05-01

    Full Text Available Hyperspectral Imaging (HSI is used to provide a wealth of information which can be used to address a variety of problems in different applications. The main requirement in all applications is the classification of HSI data. In this paper, supervised HSI classification algorithms are used to extract agriculture areas that specialize in wheat growing and get a classified image. In particular, Parallelepiped and Spectral Angel Mapper (SAM algorithms are used. They are implemented by a software tool used to analyse and process geospatial images that is an Environment of Visualizing Images (ENVI. They are applied on Al-Kharj, Saudi Arabia as the study area. The overall accuracy after applying the algorithms on the image of the study area for SAM classification was 66.67%, and 33.33% for Parallelepiped classification. Therefore, SAM algorithm has provided a better a study area image classification.

  2. Metric Learning for Hyperspectral Image Segmentation

    Science.gov (United States)

    Bue, Brian D.; Thompson, David R.; Gilmore, Martha S.; Castano, Rebecca

    2011-01-01

    We present a metric learning approach to improve the performance of unsupervised hyperspectral image segmentation. Unsupervised spatial segmentation can assist both user visualization and automatic recognition of surface features. Analysts can use spatially-continuous segments to decrease noise levels and/or localize feature boundaries. However, existing segmentation methods use tasks-agnostic measures of similarity. Here we learn task-specific similarity measures from training data, improving segment fidelity to classes of interest. Multiclass Linear Discriminate Analysis produces a linear transform that optimally separates a labeled set of training classes. The defines a distance metric that generalized to a new scenes, enabling graph-based segmentation that emphasizes key spectral features. We describe tests based on data from the Compact Reconnaissance Imaging Spectrometer (CRISM) in which learned metrics improve segment homogeneity with respect to mineralogical classes.

  3. Processing of hyperspectral medical images applications in dermatology using Matlab

    CERN Document Server

    Koprowski, Robert

    2017-01-01

    This book presents new methods of analyzing and processing hyperspectral medical images, which can be used in diagnostics, for example for dermatological images. The algorithms proposed are fully automatic and the results obtained are fully reproducible. Their operation was tested on a set of several thousands of hyperspectral images and they were implemented in Matlab. The presented source code can be used without licensing restrictions. This is a valuable resource for computer scientists, bioengineers, doctoral students, and dermatologists interested in contemporary analysis methods.

  4. A survey of landmine detection using hyperspectral imaging

    Science.gov (United States)

    Makki, Ihab; Younes, Rafic; Francis, Clovis; Bianchi, Tiziano; Zucchetti, Massimo

    2017-02-01

    Hyperspectral imaging is a trending technique in remote sensing that finds its application in many different areas, such as agriculture, mapping, target detection, food quality monitoring, etc. This technique gives the ability to remotely identify the composition of each pixel of the image. Therefore, it is a natural candidate for the purpose of landmine detection, thanks to its inherent safety and fast response time. In this paper, we will present the results of several studies that employed hyperspectral imaging for the purpose of landmine detection, discussing the different signal processing techniques used in this framework for hyperspectral image processing and target detection. Our purpose is to highlight the progresses attained in the detection of landmines using hyperspectral imaging and to identify possible perspectives for future work, in order to achieve a better detection in real-time operation mode.

  5. Framework for hyperspectral image processing and quantification for cancer detection during animal tumor surgery.

    Science.gov (United States)

    Lu, Guolan; Wang, Dongsheng; Qin, Xulei; Halig, Luma; Muller, Susan; Zhang, Hongzheng; Chen, Amy; Pogue, Brian W; Chen, Zhuo Georgia; Fei, Baowei

    2015-01-01

    Hyperspectral imaging (HSI) is an imaging modality that holds strong potential for rapid cancer detection during image-guided surgery. But the data from HSI often needs to be processed appropriately in order to extract the maximum useful information that differentiates cancer from normal tissue. We proposed a framework for hyperspectral image processing and quantification, which includes a set of steps including image preprocessing, glare removal, feature extraction, and ultimately image classification. The framework has been tested on images from mice with head and neck cancer, using spectra from 450- to 900-nm wavelength. The image analysis computed Fourier coefficients, normalized reflectance, mean, and spectral derivatives for improved accuracy. The experimental results demonstrated the feasibility of the hyperspectral image processing and quantification framework for cancer detection during animal tumor surgery, in a challenging setting where sensitivity can be low due to a modest number of features present, but potential for fast image classification can be high. This HSI approach may have potential application in tumor margin assessment during image-guided surgery, where speed of assessment may be the dominant factor.

  6. Framework for hyperspectral image processing and quantification for cancer detection during animal tumor surgery

    Science.gov (United States)

    Lu, Guolan; Wang, Dongsheng; Qin, Xulei; Halig, Luma; Muller, Susan; Zhang, Hongzheng; Chen, Amy; Pogue, Brian W.; Chen, Zhuo Georgia; Fei, Baowei

    2015-12-01

    Hyperspectral imaging (HSI) is an imaging modality that holds strong potential for rapid cancer detection during image-guided surgery. But the data from HSI often needs to be processed appropriately in order to extract the maximum useful information that differentiates cancer from normal tissue. We proposed a framework for hyperspectral image processing and quantification, which includes a set of steps including image preprocessing, glare removal, feature extraction, and ultimately image classification. The framework has been tested on images from mice with head and neck cancer, using spectra from 450- to 900-nm wavelength. The image analysis computed Fourier coefficients, normalized reflectance, mean, and spectral derivatives for improved accuracy. The experimental results demonstrated the feasibility of the hyperspectral image processing and quantification framework for cancer detection during animal tumor surgery, in a challenging setting where sensitivity can be low due to a modest number of features present, but potential for fast image classification can be high. This HSI approach may have potential application in tumor margin assessment during image-guided surgery, where speed of assessment may be the dominant factor.

  7. Food quality assessment by NIR hyperspectral imaging

    Science.gov (United States)

    Whitworth, Martin B.; Millar, Samuel J.; Chau, Astor

    2010-04-01

    Near infrared reflectance (NIR) spectroscopy is well established in the food industry for rapid compositional analysis of bulk samples. NIR hyperspectral imaging provides new opportunities to measure the spatial distribution of components such as moisture and fat, and to identify and measure specific regions of composite samples. An NIR hyperspectral imaging system has been constructed for food research applications, incorporating a SWIR camera with a cooled 14 bit HgCdTe detector and N25E spectrograph (Specim Ltd, Finland). Samples are scanned in a pushbroom mode using a motorised stage. The system has a spectral resolution of 256 pixels covering a range of 970-2500 nm and a spatial resolution of 320 pixels covering a swathe adjustable from 8 to 300 mm. Images are acquired at a rate of up to 100 lines s-1, enabling samples to be scanned within a few seconds. Data are captured using SpectralCube software (Specim) and analysed using ENVI and IDL (ITT Visual Information Solutions). Several food applications are presented. The strength of individual absorbance bands enables the distribution of particular components to be assessed. Examples are shown for detection of added gluten in wheat flour and to study the effect of processing conditions on fat distribution in chips/French fries. More detailed quantitative calibrations have been developed to study evolution of the moisture distribution in baguettes during storage at different humidities, to assess freshness of fish using measurements of whole cod and fillets, and for prediction of beef quality by identification and separate measurement of lean and fat regions.

  8. Algorithms for Hyperspectral Signature Classification in Non-resolved Object Characterization Using Tabular Nearest Neighbor Encoding

    Science.gov (United States)

    Schmalz, M.; Key, G.

    significantly strengthen analysis of classifier performance in support of classifier optimization. We show that AM-based classification can be modified to include dynamic tracking of input statistical changes, to achieve accurate signature classification in the presence of noise, closely spaced or interleaved signatures, and simulated optical distortions. In particular, we examine two critical cases: (1) classification of multiple closely spaced signatures that are difficult to separate using distance measures, and (2) classification of materials in simulated hyperspectral images of spaceborne satellites. In each case, test data are derived from a NASA database of space material signatures. Additional analysis pertains to computational complexity and noise sensitivity, which are superior to Bayesian techniques based on classical neural networks.

  9. Black Beauty's Rainbow: Hyperspectral Imaging of Northwest Africa 7034

    Science.gov (United States)

    Cannon, K. M.; Mustard, J. F.; Agee, C. B.; Wilson, J. H.; Greenberger, R. N.

    2014-07-01

    Hyperspectral imaging is used to characterize the first basaltic breccia from Mars, Northwest Africa 7034. Initial results show the spectral character of NWA 7034 is unlike other SNC meteorites and may be more representative of average martian crust.

  10. Onboard Classification of Hyperspectral Data on the Earth Observing One Mission

    Science.gov (United States)

    Chien, Steve; Tran, Daniel; Schaffer, Steve; Rabideau, Gregg; Davies, Ashley Gerard; Doggett, Thomas; Greeley, Ronald; Ip, Felipe; Baker, Victor; Doubleday, Joshua; Castano, Rebecca; Mandl, Daniel; Frye, Stuart; Ong, Lawrence; Rogez, Francois; Oaida, Bogdan

    2009-01-01

    Remote-sensed hyperspectral data represents significant challenges in downlink due to its large data volumes. This paper describes a research program designed to process hyperspectral data products onboard spacecraft to (a) reduce data downlink volumes and (b) decrease latency to provide key data products (often by enabling use of lower data rate communications systems). We describe efforts to develop onboard processing to study volcanoes, floods, and cryosphere, using the Hyperion hyperspectral imager and onboard processing for the Earth Observing One (EO-1) mission as well as preliminary work targeting the Hyperspectral Infrared Imager (HyspIRI) mission.

  11. Commodity cluster and hardware-based massively parallel implementations of hyperspectral imaging algorithms

    Science.gov (United States)

    Plaza, Antonio; Chang, Chein-I.; Plaza, Javier; Valencia, David

    2006-05-01

    The incorporation of hyperspectral sensors aboard airborne/satellite platforms is currently producing a nearly continual stream of multidimensional image data, and this high data volume has soon introduced new processing challenges. The price paid for the wealth spatial and spectral information available from hyperspectral sensors is the enormous amounts of data that they generate. Several applications exist, however, where having the desired information calculated quickly enough for practical use is highly desirable. High computing performance of algorithm analysis is particularly important in homeland defense and security applications, in which swift decisions often involve detection of (sub-pixel) military targets (including hostile weaponry, camouflage, concealment, and decoys) or chemical/biological agents. In order to speed-up computational performance of hyperspectral imaging algorithms, this paper develops several fast parallel data processing techniques. Techniques include four classes of algorithms: (1) unsupervised classification, (2) spectral unmixing, and (3) automatic target recognition, and (4) onboard data compression. A massively parallel Beowulf cluster (Thunderhead) at NASA's Goddard Space Flight Center in Maryland is used to measure parallel performance of the proposed algorithms. In order to explore the viability of developing onboard, real-time hyperspectral data compression algorithms, a Xilinx Virtex-II field programmable gate array (FPGA) is also used in experiments. Our quantitative and comparative assessment of parallel techniques and strategies may help image analysts in selection of parallel hyperspectral algorithms for specific applications.

  12. Hyperspectral imaging fluorescence excitation scanning for colon cancer detection

    Science.gov (United States)

    Leavesley, Silas J.; Walters, Mikayla; Lopez, Carmen; Baker, Thomas; Favreau, Peter F.; Rich, Thomas C.; Rider, Paul F.; Boudreaux, Carole W.

    2016-10-01

    Optical spectroscopy and hyperspectral imaging have shown the potential to discriminate between cancerous and noncancerous tissue with high sensitivity and specificity. However, to date, these techniques have not been effectively translated to real-time endoscope platforms. Hyperspectral imaging of the fluorescence excitation spectrum represents new technology that may be well suited for endoscopic implementation. However, the feasibility of detecting differences between normal and cancerous mucosa using fluorescence excitation-scanning hyperspectral imaging has not been evaluated. The goal of this study was to evaluate the initial feasibility of using fluorescence excitation-scanning hyperspectral imaging for measuring changes in fluorescence excitation spectrum concurrent with colonic adenocarcinoma using a small pre-pilot-scale sample size. Ex vivo analysis was performed using resected pairs of colorectal adenocarcinoma and normal mucosa. Adenocarcinoma was confirmed by histologic evaluation of hematoxylin and eosin (H&E) permanent sections. Specimens were imaged using a custom hyperspectral imaging fluorescence excitation-scanning microscope system. Results demonstrated consistent spectral differences between normal and cancerous tissues over the fluorescence excitation range of 390 to 450 nm that could be the basis for wavelength-dependent detection of colorectal cancers. Hence, excitation-scanning hyperspectral imaging may offer an alternative approach for discriminating adenocarcinoma from surrounding normal colonic mucosa, but further studies will be required to evaluate the accuracy of this approach using a larger patient cohort.

  13. Liquid-crystal-based hyperspectral image projector

    Science.gov (United States)

    Linnenberger, Anna; Masterson, Hugh; Rice, Joseph P.; Stockley, Jay

    2010-04-01

    A hyperspectral image projector (HIP) is introduced that is built with liquid crystal based spatial light modulators (SLM) as opposed to micromirror arrays. The use of an SLM as a broadband intensity modulator presents several benefits to this application. With slight modifications to the SLM design, SLMs can be built for a wide range of spectral regimes, ranging from the ultraviolet (UV) to the long-wavelength infrared (LWIR). SLMs can have a large pixel pitch, significantly reducing diffraction in the mid-wavelength infrared (MWIR) and LWIR. Liquid crystal based devices offer direct analog intensity modulation, thus eliminating flicker from time sequential drive schemes. SLMs allow for an on-axis configuration, enabling a simple and compact optical layout. The design of the HIP system is broken into two parts consisting of a spectral and spatial engine. In the spectral engine a diffraction grating is used to disperse a broadband source into spectral components, where an SLM modulates the relative intensity of the components to dynamically generate complex spectra. The recombined output is fed to the spatial engine which is used to construct two-dimensional scenes. The system is used to simulate a broad range of real world environments, and will be delivered to the National Institute of Standards and Technology as an enabling tool for the development of calibration standards and performance testing techniques for multispectral and hyperspectral imagers. The focus of this paper is on a visible-band HIP system; however, related work is presented with regard to SLM use in the MWIR and LWIR.

  14. Upconversion applied for mid-IR hyperspectral image acquisition

    DEFF Research Database (Denmark)

    Tidemand-Lichtenberg, Peter; Kehlet, Louis Martinus; Sanders, Nicolai Højer;

    2015-01-01

    Different schemes for upconversion mid-IR hyperspectral imaging is implemented and compared in terms of spectral coverage, spectral resolution, speed and noise. Phasematch scanning and scanning of the object within the field of view is considered.......Different schemes for upconversion mid-IR hyperspectral imaging is implemented and compared in terms of spectral coverage, spectral resolution, speed and noise. Phasematch scanning and scanning of the object within the field of view is considered....

  15. Hyperspectral image classification based on joint structured sparse representation%基于联合结构化稀疏表示的高光谱图像分类

    Institute of Scientific and Technical Information of China (English)

    薄纯娟; 张汝波; 杨大伟; 龚涛

    2015-01-01

    为了实现精确的高光谱图像分类,结合稀疏表示及光谱信息提出基于联合结构化稀疏表示的高光谱分类算法,该算法能够同时利用像元的空间语义信息和字典集内部样本结构性。建立了联合结构化稀疏表示模型,并利用交替方向迭代算法设计求解方案。在此基础上,设计了基于联合结构化稀疏表示的高光谱分类框架,其中利用类特定残留来判定被测试像元的类别。实验结果表明所提算法相比其他经典或前沿算法具有更高的分类精度。%In order to achieve accurate hyperspectral image (HSI)classification,with combining sparse representation and the spectral information,a novel HSI classification algorithm was proposed based on joint structured sparse representation.This algorithm is able to exploit the spatial contextual information of the testing pixels and the structure information among dictionaries in the meanwhile.A joint sparse structured sparse representation model was built,and an effective solution method was developed by using the alternating direction method of multipliers (ADMM)method.Based on the proposed model,a HSI classification framework was designed based on joint structured sparse repre-sentation,in which the class-specific residue manner was adopted to determine the class of testing pix-els.The experimental results demonstrate that the proposed method can achieve better accurate classi-fication performance than other classical or state-of-the-arts algorithms.

  16. Geometric and Reflectance Signature Characterization of Complex Canopies Using Hyperspectral Stereoscopic Images from Uav and Terrestrial Platforms

    Science.gov (United States)

    Honkavaara, E.; Hakala, T.; Nevalainen, O.; Viljanen, N.; Rosnell, T.; Khoramshahi, E.; Näsi, R.; Oliveira, R.; Tommaselli, A.

    2016-06-01

    Light-weight hyperspectral frame cameras represent novel developments in remote sensing technology. With frame camera technology, when capturing images with stereoscopic overlaps, it is possible to derive 3D hyperspectral reflectance information and 3D geometric data of targets of interest, which enables detailed geometric and radiometric characterization of the object. These technologies are expected to provide efficient tools in various environmental remote sensing applications, such as canopy classification, canopy stress analysis, precision agriculture, and urban material classification. Furthermore, these data sets enable advanced quantitative, physical based retrieval of biophysical and biochemical parameters by model inversion technologies. Objective of this investigation was to study the aspects of capturing hyperspectral reflectance data from unmanned airborne vehicle (UAV) and terrestrial platform with novel hyperspectral frame cameras in complex, forested environment.

  17. GEOMETRIC AND REFLECTANCE SIGNATURE CHARACTERIZATION OF COMPLEX CANOPIES USING HYPERSPECTRAL STEREOSCOPIC IMAGES FROM UAV AND TERRESTRIAL PLATFORMS

    Directory of Open Access Journals (Sweden)

    E. Honkavaara

    2016-06-01

    Full Text Available Light-weight hyperspectral frame cameras represent novel developments in remote sensing technology. With frame camera technology, when capturing images with stereoscopic overlaps, it is possible to derive 3D hyperspectral reflectance information and 3D geometric data of targets of interest, which enables detailed geometric and radiometric characterization of the object. These technologies are expected to provide efficient tools in various environmental remote sensing applications, such as canopy classification, canopy stress analysis, precision agriculture, and urban material classification. Furthermore, these data sets enable advanced quantitative, physical based retrieval of biophysical and biochemical parameters by model inversion technologies. Objective of this investigation was to study the aspects of capturing hyperspectral reflectance data from unmanned airborne vehicle (UAV and terrestrial platform with novel hyperspectral frame cameras in complex, forested environment.

  18. Hyperspectral Imaging Using Flexible Endoscopy for Laryngeal Cancer Detection

    Science.gov (United States)

    Regeling, Bianca; Thies, Boris; Gerstner, Andreas O. H.; Westermann, Stephan; Müller, Nina A.; Bendix, Jörg; Laffers, Wiebke

    2016-01-01

    Hyperspectral imaging (HSI) is increasingly gaining acceptance in the medical field. Up until now, HSI has been used in conjunction with rigid endoscopy to detect cancer in vivo. The logical next step is to pair HSI with flexible endoscopy, since it improves access to hard-to-reach areas. While the flexible endoscope’s fiber optic cables provide the advantage of flexibility, they also introduce an interfering honeycomb-like pattern onto images. Due to the substantial impact this pattern has on locating cancerous tissue, it must be removed before the HS data can be further processed. Thereby, the loss of information is to minimize avoiding the suppression of small-area variations of pixel values. We have developed a system that uses flexible endoscopy to record HS cubes of the larynx and designed a special filtering technique to remove the honeycomb-like pattern with minimal loss of information. We have confirmed its feasibility by comparing it to conventional filtering techniques using an objective metric and by applying unsupervised and supervised classifications to raw and pre-processed HS cubes. Compared to conventional techniques, our method successfully removes the honeycomb-like pattern and considerably improves classification performance, while preserving image details. PMID:27529255

  19. Hyperspectral Imaging Using Flexible Endoscopy for Laryngeal Cancer Detection

    Directory of Open Access Journals (Sweden)

    Bianca Regeling

    2016-08-01

    Full Text Available Hyperspectral imaging (HSI is increasingly gaining acceptance in the medical field. Up until now, HSI has been used in conjunction with rigid endoscopy to detect cancer in vivo. The logical next step is to pair HSI with flexible endoscopy, since it improves access to hard-to-reach areas. While the flexible endoscope’s fiber optic cables provide the advantage of flexibility, they also introduce an interfering honeycomb-like pattern onto images. Due to the substantial impact this pattern has on locating cancerous tissue, it must be removed before the HS data can be further processed. Thereby, the loss of information is to minimize avoiding the suppression of small-area variations of pixel values. We have developed a system that uses flexible endoscopy to record HS cubes of the larynx and designed a special filtering technique to remove the honeycomb-like pattern with minimal loss of information. We have confirmed its feasibility by comparing it to conventional filtering techniques using an objective metric and by applying unsupervised and supervised classifications to raw and pre-processed HS cubes. Compared to conventional techniques, our method successfully removes the honeycomb-like pattern and considerably improves classification performance, while preserving image details.

  20. Development of practical thermal infrared hyperspectral imaging system

    Science.gov (United States)

    Wang, Jianyu; Li, Chunlai; Lv, Gang; Yuan, Liyin; Liu, Enguang; Jin, Jian; Ji, Hongzhen

    2014-11-01

    As an optical remote sensing equipment, the thermal infrared hyperspectral imager operates in the thermal infrared spectral band and acquires about 180 wavebands in range of 8.0~12.5μm. The field of view of this imager is 13° and the spatial resolution is better than 1mrad. Its noise equivalent temperature difference (NETD) is less than 0.2K@300K(average). 1 The influence of background radiation of the thermal infrared hyperspectral imager,and a simulation model of simplified background radiation is builded. 2 The design and implementationof the Cryogenic Optics. 3 Thermal infrared focal plane array (FPA) and special dewar component for the thermal infrared hyperspectral imager. 4 Parts of test results of the thermal infrared hyperspectral imager.The hyperspectral imaging system is China's first success in developing this type of instrument, whose flight validation experiments have already been embarked on. The thermal infrared hyperspectral data acquired will play an important role in fields such as geological exploration and air pollutant identification.

  1. A hyperspectral imaging prototype for online quality evaluation of pickling cucumbers

    Science.gov (United States)

    A hyperspectral imaging prototype was developed for online evaluation of external and internal quality of pickling cucumbers. The prototype had several new, unique features including simultaneous reflectance and transmittance imaging and inline, real time calibration of hyperspectral images of each ...

  2. Hyperspectral landcover classification for the Yakima Training Center, Yakima, Washington

    Energy Technology Data Exchange (ETDEWEB)

    Steinmaus, K.L.; Perry, E.M.; Petrie, G.M.; Irwin, D.E.; Foote, H.P.; Wurstner, S.K.; Stephen, A.J.

    1998-04-01

    The US Department of Energy`s (DOE`s) Pacific Northwest National Laboratory (PNNL) was tasked in FY97-98 to conduct a multisensor feature extraction project for the Terrain Modeling Project Office (TMPO) of the National Imagery and Mapping Agency (NIMA). The goal of this research is the development of near-autonomous methods to remotely classify and characterize regions of military interest, in support of the TMPO of NIMA. These methods exploit remotely sensed datasets including hyperspectral (HYDICE) imagery, near-infrared and thermal infrared (Daedalus 3600), radar, and terrain datasets. The study site for this project is the US Army`s Yakima Training Center (YTC), a 326,741-acre training area located near Yakima, Washington. Two study areas at the YTC were selected to conduct and demonstrate multisensor feature extraction, the 2-km x 2-km Cantonment Area and the 3-km x 3-km Choke Point area. Classification of the Cantonment area afforded a comparison of classification results at different scales.

  3. A Multichannel Gray Level Co-Occurrence Matrix for Multi/Hyperspectral Image Texture Representation

    Directory of Open Access Journals (Sweden)

    Xin Huang

    2014-09-01

    Full Text Available This study proposes a novel method for multichannel image gray level co-occurrence matrix (GLCM texture representation. It is well known that the standard procedure for the automatic extraction of GLCM textures is based on a mono-spectral image. In real applications, however, the GLCM texture feature extraction always refers to multi/hyperspectral images. The widely used strategy to deal with this issue is to calculate the GLCM from the first principal component or the panchromatic band, which do not include all the useful information. Accordingly, in this study, we propose to represent the multichannel textures for multi/hyperspectral imagery by the use of: (1 clustering algorithms; and (2 sparse representation, respectively. In this way, the multi/hyperspectral images can be described using a series of quantized codes or dictionaries, which are more suitable for multichannel texture representation than the traditional methods. Specifically, K-means and fuzzy c-means methods are adopted to generate the codes of an image from the clustering point of view, while a sparse dictionary learning method based on two coding rules is proposed to produce the texture primitives. The proposed multichannel GLCM textural extraction methods were evaluated with four multi/hyperspectral datasets: GeoEye-1 and QuickBird multispectral images of the city of Wuhan, the well-known AVIRIS hyperspectral dataset from the Indian Pines test site, and the HYDICE airborne hyperspectral dataset from the Washington DC Mall. The results show that both the clustering-based and sparsity-based GLCM textures outperform the traditional method (extraction based on the first principal component in terms of classification accuracies in all the experiments.

  4. Application of hyperspectral imaging and chemometric calibrations for variety discrimination of maize seeds.

    Science.gov (United States)

    Zhang, Xiaolei; Liu, Fei; He, Yong; Li, Xiaoli

    2012-12-12

    Hyperspectral imaging in the visible and near infrared (VIS-NIR) region was used to develop a novel method for discriminating different varieties of commodity maize seeds. Firstly, hyperspectral images of 330 samples of six varieties of maize seeds were acquired using a hyperspectral imaging system in the 380-1,030 nm wavelength range. Secondly, principal component analysis (PCA) and kernel principal component analysis (KPCA) were used to explore the internal structure of the spectral data. Thirdly, three optimal wavelengths (523, 579 and 863 nm) were selected by implementing PCA directly on each image. Then four textural variables including contrast, homogeneity, energy and correlation were extracted from gray level co-occurrence matrix (GLCM) of each monochromatic image based on the optimal wavelengths. Finally, several models for maize seeds identification were established by least squares-support vector machine (LS-SVM) and back propagation neural network (BPNN) using four different combinations of principal components (PCs), kernel principal components (KPCs) and textural features as input variables, respectively. The recognition accuracy achieved in the PCA-GLCM-LS-SVM model (98.89%) was the most satisfactory one. We conclude that hyperspectral imaging combined with texture analysis can be implemented for fast classification of different varieties of maize seeds.

  5. Application of Hyperspectral Imaging and Chemometric Calibrations for Variety Discrimination of Maize Seeds

    Directory of Open Access Journals (Sweden)

    Xiaolei Zhang

    2012-12-01

    Full Text Available Hyperspectral imaging in the visible and near infrared (VIS-NIR region was used to develop a novel method for discriminating different varieties of commodity maize seeds. Firstly, hyperspectral images of 330 samples of six varieties of maize seeds were acquired using a hyperspectral imaging system in the 380–1,030 nm wavelength range. Secondly, principal component analysis (PCA and kernel principal component analysis (KPCA were used to explore the internal structure of the spectral data. Thirdly, three optimal wavelengths (523, 579 and 863 nm were selected by implementing PCA directly on each image. Then four textural variables including contrast, homogeneity, energy and correlation were extracted from gray level co-occurrence matrix (GLCM of each monochromatic image based on the optimal wavelengths. Finally, several models for maize seeds identification were established by least squares-support vector machine (LS-SVM and back propagation neural network (BPNN using four different combinations of principal components (PCs, kernel principal components (KPCs and textural features as input variables, respectively. The recognition accuracy achieved in the PCA-GLCM-LS-SVM model (98.89% was the most satisfactory one. We conclude that hyperspectral imaging combined with texture analysis can be implemented for fast classification of different varieties of maize seeds.

  6. Improved Scanners for Microscopic Hyperspectral Imaging

    Science.gov (United States)

    Mao, Chengye

    2009-01-01

    Improved scanners to be incorporated into hyperspectral microscope-based imaging systems have been invented. Heretofore, in microscopic imaging, including spectral imaging, it has been customary to either move the specimen relative to the optical assembly that includes the microscope or else move the entire assembly relative to the specimen. It becomes extremely difficult to control such scanning when submicron translation increments are required, because the high magnification of the microscope enlarges all movements in the specimen image on the focal plane. To overcome this difficulty, in a system based on this invention, no attempt would be made to move either the specimen or the optical assembly. Instead, an objective lens would be moved within the assembly so as to cause translation of the image at the focal plane: the effect would be equivalent to scanning in the focal plane. The upper part of the figure depicts a generic proposed microscope-based hyperspectral imaging system incorporating the invention. The optical assembly of this system would include an objective lens (normally, a microscope objective lens) and a charge-coupled-device (CCD) camera. The objective lens would be mounted on a servomotor-driven translation stage, which would be capable of moving the lens in precisely controlled increments, relative to the camera, parallel to the focal-plane scan axis. The output of the CCD camera would be digitized and fed to a frame grabber in a computer. The computer would store the frame-grabber output for subsequent viewing and/or processing of images. The computer would contain a position-control interface board, through which it would control the servomotor. There are several versions of the invention. An essential feature common to all versions is that the stationary optical subassembly containing the camera would also contain a spatial window, at the focal plane of the objective lens, that would pass only a selected portion of the image. In one version

  7. A Kernel—based Nonlinear Subspace Projection Method for Dimensionality Reduction of Hyperspectral Image Data

    Institute of Scientific and Technical Information of China (English)

    GUYanfeng; ZHANGYe; QUANTaifan

    2003-01-01

    A challenging problem in using hyper-spectral data is to eliminate redundancy and preserve useful spectral information for applications. In this pa-per, a kernel-based nonlinear subspace projection (KNSP)method is proposed for feature extraction and dimension-ality reduction in hyperspectral images. The proposed method includes three key steps: subspace partition of hyperspectral data, feature extraction using kernel-based principal component analysis (KPCA) and feature selec-tion based on class separability in the subspaces. Accord-ing to the strong correlation between neighboring bands,the whole data space is partitioned to requested subspaces.In each subspace, the KPCA method is used to effectively extract spectral feature and eliminate redundancies. A criterion function based on class discrimination and sepa-rability is used for the transformed feature selection. For the purpose of testifying its effectiveness, the proposed new method is compared with the classical principal component analysis (PCA) and segmented principal component trans-formation (SPCT). A hyperspectral image classification is performed on AVIRIS data. which have 224 svectral bands.Experimental results show that KNSP is very effective for feature extraction and dimensionality reduction of hyper-spectral data and provides significant improvement over classical PCA and current SPCT technique.

  8. Algorithms for Hyperspectral Endmember Extraction and Signature Classification with Morphological Dendritic Networks

    Science.gov (United States)

    Schmalz, M.; Ritter, G.

    Accurate multispectral or hyperspectral signature classification is key to the nonimaging detection and recognition of space objects. Additionally, signature classification accuracy depends on accurate spectral endmember determination [1]. Previous approaches to endmember computation and signature classification were based on linear operators or neural networks (NNs) expressed in terms of the algebra (R, +, x) [1,2]. Unfortunately, class separation in these methods tends to be suboptimal, and the number of signatures that can be accurately classified often depends linearly on the number of NN inputs. This can lead to poor endmember distinction, as well as potentially significant classification errors in the presence of noise or densely interleaved signatures. In contrast to traditional CNNs, autoassociative morphological memories (AMM) are a construct similar to Hopfield autoassociatived memories defined on the (R, +, ?,?) lattice algebra [3]. Unlimited storage and perfect recall of noiseless real valued patterns has been proven for AMMs [4]. However, AMMs suffer from sensitivity to specific noise models, that can be characterized as erosive and dilative noise. On the other hand, the prior definition of a set of endmembers corresponds to material spectra lying on vertices of the minimum convex region covering the image data. These vertices can be characterized as morphologically independent patterns. It has further been shown that AMMs can be based on dendritic computation [3,6]. These techniques yield improved accuracy and class segmentation/separation ability in the presence of highly interleaved signature data. In this paper, we present a procedure for endmember determination based on AMM noise sensitivity, which employs morphological dendritic computation. We show that detected endmembers can be exploited by AMM based classification techniques, to achieve accurate signature classification in the presence of noise, closely spaced or interleaved signatures, and

  9. Meat quality evaluation by hyperspectral imaging technique: an overview.

    Science.gov (United States)

    Elmasry, Gamal; Barbin, Douglas F; Sun, Da-Wen; Allen, Paul

    2012-01-01

    During the last two decades, a number of methods have been developed to objectively measure meat quality attributes. Hyperspectral imaging technique as one of these methods has been regarded as a smart and promising analytical tool for analyses conducted in research and industries. Recently there has been a renewed interest in using hyperspectral imaging in quality evaluation of different food products. The main inducement for developing the hyperspectral imaging system is to integrate both spectroscopy and imaging techniques in one system to make direct identification of different components and their spatial distribution in the tested product. By combining spatial and spectral details together, hyperspectral imaging has proved to be a promising technology for objective meat quality evaluation. The literature presented in this paper clearly reveals that hyperspectral imaging approaches have a huge potential for gaining rapid information about the chemical structure and related physical properties of all types of meat. In addition to its ability for effectively quantifying and characterizing quality attributes of some important visual features of meat such as color, quality grade, marbling, maturity, and texture, it is able to measure multiple chemical constituents simultaneously without monotonous sample preparation. Although this technology has not yet been sufficiently exploited in meat process and quality assessment, its potential is promising. Developing a quality evaluation system based on hyperspectral imaging technology to assess the meat quality parameters and to ensure its authentication would bring economical benefits to the meat industry by increasing consumer confidence in the quality of the meat products. This paper provides a detailed overview of the recently developed approaches and latest research efforts exerted in hyperspectral imaging technology developed for evaluating the quality of different meat products and the possibility of its widespread

  10. Detection of fruit fly infestation in pickling cucumbers using hyperspectral imaging

    Science.gov (United States)

    Lu, Renfu; Ariana, Diwan P.

    2011-06-01

    Fruit fly infestation can be a serious problem in pickling cucumber production. In the United States and many other countries, there is zero tolerance for fruit flies in pickled products. Currently, processors rely on manual inspection to detect and remove fruit fly-infested cucumbers, which is labor intensive and also prone to error due to human fatigue and the difficulty of visually detecting infestation that is hidden inside the fruit. In this research, a laboratory hyperspectral imaging system was used to detect fruit fly-infested pickling cucumbers. Hyperspectral reflectance (450-740 nm) and transmittance (740-1,000 nm) images were acquired simultaneously for 329 normal (infestation free) and fruit flyinfested pickling cucumbers of three size classes with the mean diameters of 16.8, 22.1, and 27.6 mm, respectively. Mean spectra were extracted from the hyperspectral image of each cucumber, and they were then corrected for the fruit size effect using a diameter correction equation. Partial least squares discriminant analyses for the reflectance, transmittance and their combined data were performed for differentiating normal and infested pickling cucumbers. With reflectance mode, the overall classification accuracies for the three size classes and mixed class were between 82% and 88%, whereas transmittance achieved better classification results with the overall accuracies of 88%-93%. Integration of reflectance and transmittance did not result in noticeable improvements, compared to transmittance mode. Overall, the hyperspectral imaging system performed better than manual inspection, which had an overall accuracy of 75% and decreased significantly for smaller size cucumbers. This research demonstrated that hyperspectral imaging is potentially useful for detecting fruit fly-infested pickling cucumbers.

  11. SVM-based feature extraction and classification of aflatoxin contaminated corn using fluorescence hyperspectral data

    Science.gov (United States)

    Support Vector Machine (SVM) was used in the Genetic Algorithms (GA) process to select and classify a subset of hyperspectral image bands. The method was applied to fluorescence hyperspectral data for the detection of aflatoxin contamination in Aspergillus flavus infected single corn kernels. In the...

  12. Pixel-Based Land Cover Classification by Fusing Hyperspectral and LIDAR Data

    Science.gov (United States)

    Jahan, F.; Awrangjeb, M.

    2017-09-01

    Land cover classification has many applications like forest management, urban planning, land use change identification and environment change analysis. The passive sensing of hyperspectral systems can be effective in describing the phenomenology of the observed area over hundreds of (narrow) spectral bands. On the other hand, the active sensing of LiDAR (Light Detection and Ranging) systems can be exploited for characterising topographical information of the area. As a result, the joint use of hyperspectral and LiDAR data provides a source of complementary information, which can greatly assist in the classification of complex classes. In this study, we fuse hyperspectral and LiDAR data for land cover classification. We do a pixel-wise classification on a disjoint set of training and testing samples for five different classes. We propose a new feature combination by fusing features from both hyperspectral and LiDAR, which achieves competent classification accuracy with low feature dimension, while the existing method requires high dimensional feature vector to achieve similar classification result. Also, for the reduction of the dimension of the feature vector, Principal Component Analysis (PCA) is used as it captures the variance of the samples with a limited number of Principal Components (PCs). We tested our classification method using PCA applied on hyperspectral bands only and combined hyperspectral and LiDAR features. Classification with support vector machine (SVM) and decision tree shows that our feature combination achieves better classification accuracy compared to the existing feature combination, while keeping the similar number of PCs. The experimental results also show that decision tree performs better than SVM and requires less execution time.

  13. Standoff midwave infrared hyperspectral imaging of ship plumes

    Science.gov (United States)

    Gagnon, Marc-André; Gagnon, Jean-Philippe; Tremblay, Pierre; Savary, Simon; Farley, Vincent; Guyot, Éric; Lagueux, Philippe; Chamberland, Martin; Marcotte, Frédérick

    2016-05-01

    Characterization of ship plumes is very challenging due to the great variety of ships, fuel, and fuel grades, as well as the extent of a gas plume. In this work, imaging of ship plumes from an operating ferry boat was carried out using standoff midwave (3-5 μm) infrared hyperspectral imaging. Quantitative chemical imaging of combustion gases was achieved by fitting a radiative transfer model. Combustion efficiency maps and mass flow rates are presented for carbon monoxide (CO) and carbon dioxide (CO2). The results illustrate how valuable information about the combustion process of a ship engine can be successfully obtained using passive hyperspectral remote sensing imaging.

  14. Toward prediction of hyperspectral target detection performance after lossy image compression

    Science.gov (United States)

    Kaufman, Jason R.; Vongsy, Karmon M.; Dill, Jeffrey C.

    2016-05-01

    Hyperspectral imagery (HSI) offers numerous advantages over traditional sensing modalities with its high spectral content that allows for classification, anomaly detection, target discrimination, and change detection. However, this imaging modality produces a huge amount of data, which requires transmission, processing, and storage resources; hyperspectral compression is a viable solution to these challenges. It is well known that lossy compression of hyperspectral imagery can impact hyperspectral target detection. Here we examine lossy compressed hyperspectral imagery from data-centric and target-centric perspectives. The compression ratio (CR), root mean square error (RMSE), the signal to noise ratio (SNR), and the correlation coefficient are computed directly from the imagery and provide insight to how the imagery has been affected by the lossy compression process. With targets present in the imagery, we perform target detection with the spectral angle mapper (SAM) and adaptive coherence estimator (ACE) and evaluate the change in target detection performance by examining receiver operating characteristic (ROC) curves and the target signal-to-clutter ratio (SCR). Finally, we observe relationships between the data- and target-centric metrics for selected visible/near-infrared to shortwave infrared (VNIR/SWIR) HSI data, targets, and backgrounds that motivate potential prediction of change in target detection performance as a function of compression ratio.

  15. Undersampled Hyperspectral Image Reconstruction Based on Surfacelet Transform

    Directory of Open Access Journals (Sweden)

    Lei Liu

    2015-01-01

    Full Text Available Hyperspectral imaging is a crucial technique for military and environmental monitoring. However, limited equipment hardware resources severely affect the transmission and storage of a huge amount of data for hyperspectral images. This limitation has the potentials to be solved by compressive sensing (CS, which allows reconstructing images from undersampled measurements with low error. Sparsity and incoherence are two essential requirements for CS. In this paper, we introduce surfacelet, a directional multiresolution transform for 3D data, to sparsify the hyperspectral images. Besides, a Gram-Schmidt orthogonalization is used in CS random encoding matrix, two-dimensional and three-dimensional orthogonal CS random encoding matrixes and a patch-based CS encoding scheme are designed. The proposed surfacelet-based hyperspectral images reconstruction problem is solved by a fast iterative shrinkage-thresholding algorithm. Experiments demonstrate that reconstruction of spectral lines and spatial images is significantly improved using the proposed method than using conventional three-dimensional wavelets, and growing randomness of encoding matrix can further improve the quality of hyperspectral data. Patch-based CS encoding strategy can be used to deal with large data because data in different patches can be independently sampled.

  16. Hyperspectral laser-induced flourescence imaging for assessing internal quality of kiwi fruit

    Science.gov (United States)

    Liu, Muhua; Liao, Yifeng; Zhou, Xiaomei

    2008-03-01

    This paper describes an experimental study on non-destructive methods for predicting quality of kiwifruits using fluorescence imaging. The method is based on hyperspectral laser-induced fluorescence imaging in the region between 700 and 1110 nm, and estimates the kiwifruits quality in terms of internal sugar content and firmness. A station for acquiring hyperspectral laser-induced fluorescence imaging has been designed and carefully choosing each component. The fluorescence imaging acquired by the station has been pre-processed by selecting regions of interest (ROIs) of 50 100 × pixels. A line regressing prediction method estimates the quality of kiwifruit samples. The results obtained in classification show that the station and prediction model enables the correct discrimination of kiwifruits internal sugar content and firmness with a percentage of r= 98.5%, SEP=0.4 and r=99.9%, SEP=0.62.

  17. Cellular image classification

    CERN Document Server

    Xu, Xiang; Lin, Feng

    2017-01-01

    This book introduces new techniques for cellular image feature extraction, pattern recognition and classification. The authors use the antinuclear antibodies (ANAs) in patient serum as the subjects and the Indirect Immunofluorescence (IIF) technique as the imaging protocol to illustrate the applications of the described methods. Throughout the book, the authors provide evaluations for the proposed methods on two publicly available human epithelial (HEp-2) cell datasets: ICPR2012 dataset from the ICPR'12 HEp-2 cell classification contest and ICIP2013 training dataset from the ICIP'13 Competition on cells classification by fluorescent image analysis. First, the reading of imaging results is significantly influenced by one’s qualification and reading systems, causing high intra- and inter-laboratory variance. The authors present a low-order LP21 fiber mode for optical single cell manipulation and imaging staining patterns of HEp-2 cells. A focused four-lobed mode distribution is stable and effective in optical...

  18. Detection of ochratoxin A contamination in stored wheat using near-infrared hyperspectral imaging

    Science.gov (United States)

    Senthilkumar, T.; Jayas, D. S.; White, N. D. G.; Fields, P. G.; Gräfenhan, T.

    2017-03-01

    Near-infrared (NIR) hyperspectral imaging system was used to detect five concentration levels of ochratoxin A (OTA) in contaminated wheat kernels. The wheat kernels artificially inoculated with two different OTA producing Penicillium verrucosum strains, two different non-toxigenic P. verrucosum strains, and sterile control wheat kernels were subjected to NIR hyperspectral imaging. The acquired three-dimensional data were reshaped into readable two-dimensional data. Principal Component Analysis (PCA) was applied to the two dimensional data to identify the key wavelengths which had greater significance in detecting OTA contamination in wheat. Statistical and histogram features extracted at the key wavelengths were used in the linear, quadratic and Mahalanobis statistical discriminant models to differentiate between sterile control, five concentration levels of OTA contamination in wheat kernels, and five infection levels of non-OTA producing P. verrucosum inoculated wheat kernels. The classification models differentiated sterile control samples from OTA contaminated wheat kernels and non-OTA producing P. verrucosum inoculated wheat kernels with a 100% accuracy. The classification models also differentiated between five concentration levels of OTA contaminated wheat kernels and between five infection levels of non-OTA producing P. verrucosum inoculated wheat kernels with a correct classification of more than 98%. The non-OTA producing P. verrucosum inoculated wheat kernels and OTA contaminated wheat kernels subjected to hyperspectral imaging provided different spectral patterns.

  19. M-estimation for robust sparse unmixing of hyperspectral images

    Science.gov (United States)

    Toomik, Maria; Lu, Shijian; Nelson, James D. B.

    2016-10-01

    Hyperspectral unmixing methods often use a conventional least squares based lasso which assumes that the data follows the Gaussian distribution. The normality assumption is an approximation which is generally invalid for real imagery data. We consider a robust (non-Gaussian) approach to sparse spectral unmixing of remotely sensed imagery which reduces the sensitivity of the estimator to outliers and relaxes the linearity assumption. The method consists of several appropriate penalties. We propose to use an lp norm with 0 Gaussian likelihood with a fixed function ρ(e) that restrains outliers. The M-estimate function reduces the effect of errors with large amplitudes or even assigns the outliers zero weights. Our experimental results on real hyperspectral data show that noise with large amplitudes (outliers) often exists in the data. This ability to mitigate the influence of such outliers can therefore offer greater robustness. Qualitative hyperspectral unmixing results on real hyperspectral image data corroborate the efficacy of the proposed method.

  20. Resolving Mixed Algal Species in Hyperspectral Images

    Directory of Open Access Journals (Sweden)

    Mehrube Mehrubeoglu

    2013-12-01

    Full Text Available We investigated a lab-based hyperspectral imaging system’s response from pure (single and mixed (two algal cultures containing known algae types and volumetric combinations to characterize the system’s performance. The spectral response to volumetric changes in single and combinations of algal mixtures with known ratios were tested. Constrained linear spectral unmixing was applied to extract the algal content of the mixtures based on abundances that produced the lowest root mean square error. Percent prediction error was computed as the difference between actual percent volumetric content and abundances at minimum RMS error. Best prediction errors were computed as 0.4%, 0.4% and 6.3% for the mixed spectra from three independent experiments. The worst prediction errors were found as 5.6%, 5.4% and 13.4% for the same order of experiments. Additionally, Beer-Lambert’s law was utilized to relate transmittance to different volumes of pure algal suspensions demonstrating linear logarithmic trends for optical property measurements.

  1. In vivo and in vitro hyperspectral imaging of cervical neoplasia

    Science.gov (United States)

    Wang, Chaojian; Zheng, Wenli; Bu, Yanggao; Chang, Shufang; Tong, Qingping; Zhang, Shiwu; Xu, Ronald X.

    2014-02-01

    Cervical cancer is a prevalent disease in many developing countries. Colposcopy is the most common approach for screening cervical intraepithelial neoplasia (CIN). However, its clinical efficacy heavily relies on the examiner's experience. Spectroscopy is a potentially effective method for noninvasive diagnosis of cervical neoplasia. In this paper, we introduce a hyperspectral imaging technique for noninvasive detection and quantitative analysis of cervical neoplasia. A hyperspectral camera is used to collect the reflectance images of the entire cervix under xenon lamp illumination, followed by standard colposcopy examination and cervical tissue biopsy at both normal and abnormal sites in different quadrants. The collected reflectance data are calibrated and the hyperspectral signals are extracted. Further spectral analysis and image processing works are carried out to classify tissue into different types based on the spectral characteristics at different stages of cervical intraepithelial neoplasia. The hyperspectral camera is also coupled with a lab microscope to acquire the hyperspectral transmittance images of the pathological slides. The in vivo and the in vitro imaging results are compared with clinical findings to assess the accuracy and efficacy of the method.

  2. Classification of iconic images

    OpenAIRE

    Zrianina, Mariia; Kopf, Stephan

    2016-01-01

    Iconic images represent an abstract topic and use a presentation that is intuitively understood within a certain cultural context. For example, the abstract topic “global warming” may be represented by a polar bear standing alone on an ice floe. Such images are widely used in media and their automatic classification can help to identify high-level semantic concepts. This paper presents a system for the classification of iconic images. It uses a variation of the Bag of Visual Words approach wi...

  3. Plugin procedure in segmentation and application to hyperspectral image segmentation

    CERN Document Server

    Girard, R

    2010-01-01

    In this article we give our contribution to the problem of segmentation with plug-in procedures. We give general sufficient conditions under which plug in procedure are efficient. We also give an algorithm that satisfy these conditions. We give an application of the used algorithm to hyperspectral images segmentation. Hyperspectral images are images that have both spatial and spectral coherence with thousands of spectral bands on each pixel. In the proposed procedure we combine a reduction dimension technique and a spatial regularisation technique. This regularisation is based on the mixlet modelisation of Kolaczyck and Al.

  4. Hyperspectral imaging using the single-pixel Fourier transform technique

    Science.gov (United States)

    Jin, Senlin; Hui, Wangwei; Wang, Yunlong; Huang, Kaicheng; Shi, Qiushuai; Ying, Cuifeng; Liu, Dongqi; Ye, Qing; Zhou, Wenyuan; Tian, Jianguo

    2017-03-01

    Hyperspectral imaging technology is playing an increasingly important role in the fields of food analysis, medicine and biotechnology. To improve the speed of operation and increase the light throughput in a compact equipment structure, a Fourier transform hyperspectral imaging system based on a single-pixel technique is proposed in this study. Compared with current imaging spectrometry approaches, the proposed system has a wider spectral range (400-1100 nm), a better spectral resolution (1 nm) and requires fewer measurement data (a sample rate of 6.25%). The performance of this system was verified by its application to the non-destructive testing of potatoes.

  5. Investigating coral hyperspectral properties across coral species and coral state using hyperspectral imaging

    Science.gov (United States)

    Mehrubeoglu, Mehrube; Smith, Dustin K.; Smith, Shane W.; Strychar, Kevin B.; McLauchlan, Lifford

    2013-09-01

    Coral reefs are one of the most diverse and threatened ecosystems in the world. Corals worldwide are at risk, and in many instances, dying due to factors that affect their environment resulting in deteriorating environmental conditions. Because corals respond quickly to the quality of the environment that surrounds them, corals have been identified as bioindicators of water quality and marine environmental health. The hyperspectral imaging system is proposed as a noninvasive tool to monitor different species of corals as well as coral state over time. This in turn can be used as a quick and non-invasive method to monitor environmental health that can later be extended to climate conditions. In this project, a laboratory-based hyperspectral imaging system is used to collect spectral and spatial information of corals. In the work presented here, MATLAB and ENVI software tools are used to view and process spatial information and coral spectral signatures to identify differences among the coral data. The results support the hypothesis that hyperspectral properties of corals vary among different coral species, and coral state over time, and hyperspectral imaging can be a used as a tool to document changes in coral species and state.

  6. Robust multitask learning with three-dimensional empirical mode decomposition-based features for hyperspectral classification

    Science.gov (United States)

    He, Zhi; Liu, Lin

    2016-11-01

    Empirical mode decomposition (EMD) and its variants have recently been applied for hyperspectral image (HSI) classification due to their ability to extract useful features from the original HSI. However, it remains a challenging task to effectively exploit the spectral-spatial information by the traditional vector or image-based methods. In this paper, a three-dimensional (3D) extension of EMD (3D-EMD) is proposed to naturally treat the HSI as a cube and decompose the HSI into varying oscillations (i.e. 3D intrinsic mode functions (3D-IMFs)). To achieve fast 3D-EMD implementation, 3D Delaunay triangulation (3D-DT) is utilized to determine the distances of extrema, while separable filters are adopted to generate the envelopes. Taking the extracted 3D-IMFs as features of different tasks, robust multitask learning (RMTL) is further proposed for HSI classification. In RMTL, pairs of low-rank and sparse structures are formulated by trace-norm and l1,2 -norm to capture task relatedness and specificity, respectively. Moreover, the optimization problems of RMTL can be efficiently solved by the inexact augmented Lagrangian method (IALM). Compared with several state-of-the-art feature extraction and classification methods, the experimental results conducted on three benchmark data sets demonstrate the superiority of the proposed methods.

  7. Hybrid tenso-vectorial compressive sensing for hyperspectral imaging

    Science.gov (United States)

    Li, Qun; Bernal, Edgar A.

    2016-05-01

    Hyperspectral imaging has a wide range of applications relying on remote material identification, including astronomy, mineralogy, and agriculture; however, due to the large volume of data involved, the complexity and cost of hyperspectral imagers can be prohibitive. The exploitation of redundancies along the spatial and spectral dimensions of a hyperspectral image of a scene has created new paradigms that overcome the limitations of traditional imaging systems. While compressive sensing (CS) approaches have been proposed and simulated with success on already acquired hyperspectral imagery, most of the existing work relies on the capability to simultaneously measure the spatial and spectral dimensions of the hyperspectral cube. Most real-life devices, however, are limited to sampling one or two dimensions at a time, which renders a significant portion of the existing work unfeasible. We propose a new variant of the recently proposed serial hybrid vectorial and tensorial compressive sensing (HCS-S) algorithm that, like its predecessor, is compatible with real-life devices both in terms of the acquisition and reconstruction requirements. The newly introduced approach is parallelizable, and we abbreviate it as HCS-P. Together, HCS-S and HCS-P comprise a generalized framework for hybrid tenso-vectorial compressive sensing, or HCS for short. We perform a detailed analysis that demonstrates the uniqueness of the signal reconstructed by both the original HCS-S and the proposed HCS-P algorithms. Last, we analyze the behavior of the HCS reconstruction algorithms in the presence of measurement noise, both theoretically and experimentally.

  8. Instrumentation challenges of a pushbroom hyperspectral imaging system for currency counterfeit applications

    Science.gov (United States)

    Lim, Hoong-Ta; Vadakke Matham, Murukeshan

    2015-07-01

    Hyperspectral imaging allows the intensity of narrow and adjacent spectral bands over a large spectral range to be recorded, giving rich spectral information for each pixel in the imaged region. The spectral characteristics of each point in the imaged region can thus be detected, which is useful for quantification and classification. Hyperspectral imaging has been used in many applications such as remote sensing, quality assessment of agro-food products, biomedical imaging and document counterfeit application. This paper presents a pushbroom spatial-scanning imager, which gives a higher spectral resolution over a broad spectral range. Although a spatial-scanning imager may be slower due to the need to perform mechanical scanning, such a high spectral resolution is especially important in applications where the capability to perform classification is much more important than speed. The application of this system is demonstrated for currency counterfeit detection applications. The high spectral resolution of a pushbroom imager is able to capture fine spectral details of the samples used in this research, providing important information required for classification. Using this technique, the reflectance is acquired from specific regions of a genuine and counterfeit note. The spectra of the same region from both notes are then compared to distinguish and delineate the differences between them. The spectrum acquired from a genuine note can then be used as a reference from which future comparison can be based upon for identifying currency counterfeit and related relevant applications.

  9. Sparse dimensionality reduction of hyperspectral image based on semi-supervised local Fisher discriminant analysis

    Science.gov (United States)

    Shao, Zhenfeng; Zhang, Lei

    2014-09-01

    This paper presents a novel sparse dimensionality reduction method of hyperspectral image based on semi-supervised local Fisher discriminant analysis (SELF). The proposed method is designed to be especially effective for dealing with the out-of-sample extrapolation to realize advantageous complementarities between SELF and sparsity preserving projections (SPP). Compared to SELF and SPP, the method proposed herein offers highly discriminative ability and produces an explicit nonlinear feature mapping for the out-of-sample extrapolation. This is due to the fact that the proposed method can get an explicit feature mapping for dimensionality reduction and improve the classification performance of classifiers by performing dimensionality reduction. Experimental analysis on the sparsity and efficacy of low dimensional outputs shows that, sparse dimensionality reduction based on SELF can yield good classification results and interpretability in the field of hyperspectral remote sensing.

  10. An airborne pushbroom hyperspectral imager with wide field of view

    Institute of Scientific and Technical Information of China (English)

    Peixin Hu; Qirnin Lu; Rong Shu; Jianyu Wang

    2005-01-01

    @@ An airborne pushbroom hyperspectral imager (APHI) with wide field (42° field of view) is presented.It is composed of two 22°field of view (FOV) imagers and can provide 1304 pixels in spatial dimension,124 bands in spectral dimension in one frame. APHI has a bandwidth ranging from 400 to 900 nm.

  11. [Hyperspectral image compression technology research based on EZW].

    Science.gov (United States)

    Wei, Jun-Xia; Xiangli, Bin; Duan, Xiao-Feng; Xu, Zhao-Hui; Xue, Li-Jun

    2011-08-01

    Along with the development of hyperspectral remote sensing technology, hyperspectral imaging technology has been applied in the aspect of aviation and spaceflight, which is different from multispectral imaging, and with the band width of nanoscale spectral imaging the target continuously, the image resolution is very high. However, with the increasing number of band, spectral data quantity will be more and more, and these data storage and transmission is the problem that the authors must face. Along with the development of wavelet compression technology, in field of image compression, many people adopted and improved EZW, the present paper used the method in hyperspectral spatial dimension compression, but does not involved the spectrum dimension compression. From hyperspectral image compression reconstruction results, whether from the peak signal-to-noise ratio (PSNR) and spectral curve or from the subjective comparison of source and reconstruction image, the effect is well. If the first compression of image from spectrum dimension is made, then compression on space dimension, the authors believe the effect will be better.

  12. Fast algorithm for exploring and compressing of large hyperspectral images

    DEFF Research Database (Denmark)

    Kucheryavskiy, Sergey

    2011-01-01

    A new method for calculation of latent variable space for exploratory analysis and dimension reduction of large hyperspectral images is proposed. The method is based on significant downsampling of image pixels with preservation of pixels’ structure in feature (variable) space. To achieve this, in...

  13. Identification of seedling cabbages and weeds using hyperspectral imaging

    Science.gov (United States)

    Target detectionis one of research focues for precision chemical application. This study developed a method to identify seedling cabbages and weeds using hyperspectral spectral imaging. In processing the image data, with ENVI software, after dimension reduction, noise reduction, de-correlation for h...

  14. Noise-Tolerant Hyperspectral Signature Classification in Unresolved Object Detection with Adaptive Tabular Nearest Neighbor Encoding

    Science.gov (United States)

    Schmalz, M.; Key, G.

    Accurate spectral signature classification is a crucial step in the nonimaging detection and recognition of spaceborne objects. In classical hyperspectral recognition applications, especially where linear mixing models are employed, signature classification accuracy depends on accurate spectral endmember discrimination. In selected target recognition (ATR) applications, previous non-adaptive techniques for signature classification have yielded class separation and classifier refinement results that tend to be suboptimal. In practice, the number of signatures accurately classified often depends linearly on the number of inputs. This can lead to potentially severe classification errors in the presence of noise or densely interleaved signatures. In this paper, we present an enhancement of an emerging technology for nonimaging spectral signature classification based on a highly accurate, efficient search engine called Tabular Nearest Neighbor Encoding (TNE). Adaptive TNE can optimize its classifier performance to track input nonergodicities and yield measures of confidence or caution for evaluation of classification results. Unlike neural networks, TNE does not have a hidden intermediate data structure (e.g., a neural net weight matrix). Instead, TNE generates and exploits a user-accessible data structure called the agreement map (AM), which can be manipulated by Boolean logic operations to effect accurate classifier refinement through programmable algorithms. The open architecture and programmability of TNE's pattern-space (AM) processing allows a TNE developer to determine the qualitative and quantitative reasons for classification accuracy, as well as characterize in detail the signatures for which TNE does not obtain classification matches, and why such mis-matches occur. In this study AM-based classification has been modified to partially compensate for input statistical changes, in response to performance metrics such as probability of correct classification (Pd

  15. Crop stress detection and classification using hyperspectral remote sensing

    Science.gov (United States)

    Irby, Jon Trenton

    Agricultural production has observed many changes in technology over the last 20 years. Producers are able to utilize technologies such as site-specific applicators and remotely sensed data to assist with decision making for best management practices which can improve crop production and provide protection to the environment. It is known that plant stress can interfere with photosynthetic reactions within the plant and/or the physical structure of the plant. Common types of stress associated with agricultural crops include herbicide induced stress, nutrient stress, and drought stress from lack of water. Herbicide induced crop stress is not a new problem. However, with increased acreage being planting in varieties/hybrids that contain herbicide resistant traits, herbicide injury to non-target crops will continue to be problematic for producers. With rapid adoption of herbicide-tolerant cropping systems, it is likely that herbicide induced stress will continue to be a major concern. To date, commercially available herbicide-tolerant varieties/hybrids contain traits which allow herbicides like glyphosate and glufosinate-ammonium to be applied as a broadcast application during the growing season. Both glyphosate and glufosinate-ammonium are broad spectrum herbicides which have activity on a large number of plant species, including major crops like non-transgenic soybean, corn, and cotton. Therefore, it is possible for crop stress from herbicide applications to occur in neighboring fields that contain susceptible crop varieties/hybrids. Nutrient and moisture stress as well as stress caused by herbicide applications can interact to influence yields in agricultural fields. If remotely sensed data can be used to accurately identify specific levels of crop stress, it is possible that producers can use this information to better assist them in crop management to maximize yields and protect their investments. This research was conducted to evaluate classification of specific

  16. Parallel hyperspectral image reconstruction using random projections

    Science.gov (United States)

    Sevilla, Jorge; Martín, Gabriel; Nascimento, José M. P.

    2016-10-01

    Spaceborne sensors systems are characterized by scarce onboard computing and storage resources and by communication links with reduced bandwidth. Random projections techniques have been demonstrated as an effective and very light way to reduce the number of measurements in hyperspectral data, thus, the data to be transmitted to the Earth station is reduced. However, the reconstruction of the original data from the random projections may be computationally expensive. SpeCA is a blind hyperspectral reconstruction technique that exploits the fact that hyperspectral vectors often belong to a low dimensional subspace. SpeCA has shown promising results in the task of recovering hyperspectral data from a reduced number of random measurements. In this manuscript we focus on the implementation of the SpeCA algorithm for graphics processing units (GPU) using the compute unified device architecture (CUDA). Experimental results conducted using synthetic and real hyperspectral datasets on the GPU architecture by NVIDIA: GeForce GTX 980, reveal that the use of GPUs can provide real-time reconstruction. The achieved speedup is up to 22 times when compared with the processing time of SpeCA running on one core of the Intel i7-4790K CPU (3.4GHz), with 32 Gbyte memory.

  17. Hyperspectral imaging applied to complex particulate solids systems

    Science.gov (United States)

    Bonifazi, Giuseppe; Serranti, Silvia

    2008-04-01

    HyperSpectral Imaging (HSI) is based on the utilization of an integrated hardware and software (HW&SW) platform embedding conventional imaging and spectroscopy to attain both spatial and spectral information from an object. Although HSI was originally developed for remote sensing, it has recently emerged as a powerful process analytical tool, for non-destructive analysis, in many research and industrial sectors. The possibility to apply on-line HSI based techniques in order to identify and quantify specific particulate solid systems characteristics is presented and critically evaluated. The originally developed HSI based logics can be profitably applied in order to develop fast, reliable and lowcost strategies for: i) quality control of particulate products that must comply with specific chemical, physical and biological constraints, ii) performance evaluation of manufacturing strategies related to processing chains and/or realtime tuning of operative variables and iii) classification-sorting actions addressed to recognize and separate different particulate solid products. Case studies, related to recent advances in the application of HSI to different industrial sectors, as agriculture, food, pharmaceuticals, solid waste handling and recycling, etc. and addressed to specific goals as contaminant detection, defect identification, constituent analysis and quality evaluation are described, according to authors' originally developed application.

  18. Fusion of Airborne Discrete-Return LiDAR and Hyperspectral Data for Land Cover Classification

    Directory of Open Access Journals (Sweden)

    Shezhou Luo

    2015-12-01

    Full Text Available Accurate land cover classification information is a critical variable for many applications. This study presents a method to classify land cover using the fusion data of airborne discrete return LiDAR (Light Detection and Ranging and CASI (Compact Airborne Spectrographic Imager hyperspectral data. Four LiDAR-derived images (DTM, DSM, nDSM, and intensity and CASI data (48 bands with 1 m spatial resolution were spatially resampled to 2, 4, 8, 10, 20 and 30 m resolutions using the nearest neighbor resampling method. These data were thereafter fused using the layer stacking and principal components analysis (PCA methods. Land cover was classified by commonly used supervised classifications in remote sensing images, i.e., the support vector machine (SVM and maximum likelihood (MLC classifiers. Each classifier was applied to four types of datasets (at seven different spatial resolutions: (1 the layer stacking fusion data; (2 the PCA fusion data; (3 the LiDAR data alone; and (4 the CASI data alone. In this study, the land cover category was classified into seven classes, i.e., buildings, road, water bodies, forests, grassland, cropland and barren land. A total of 56 classification results were produced, and the classification accuracies were assessed and compared. The results show that the classification accuracies produced from two fused datasets were higher than that of the single LiDAR and CASI data at all seven spatial resolutions. Moreover, we find that the layer stacking method produced higher overall classification accuracies than the PCA fusion method using both the SVM and MLC classifiers. The highest classification accuracy obtained (OA = 97.8%, kappa = 0.964 using the SVM classifier on the layer stacking fusion data at 1 m spatial resolution. Compared with the best classification results of the CASI and LiDAR data alone, the overall classification accuracies improved by 9.1% and 19.6%, respectively. Our findings also demonstrated that the

  19. Hyperspectral Image Super-Resolution via Non-Negative Structured Sparse Representation.

    Science.gov (United States)

    Dong, Weisheng; Fu, Fazuo; Shi, Guangming; Cao, Xun; Wu, Jinjian; Li, Guangyu; Li, Guangyu

    2016-05-01

    Hyperspectral imaging has many applications from agriculture and astronomy to surveillance and mineralogy. However, it is often challenging to obtain high-resolution (HR) hyperspectral images using existing hyperspectral imaging techniques due to various hardware limitations. In this paper, we propose a new hyperspectral image super-resolution method from a low-resolution (LR) image and a HR reference image of the same scene. The estimation of the HR hyperspectral image is formulated as a joint estimation of the hyperspectral dictionary and the sparse codes based on the prior knowledge of the spatial-spectral sparsity of the hyperspectral image. The hyperspectral dictionary representing prototype reflectance spectra vectors of the scene is first learned from the input LR image. Specifically, an efficient non-negative dictionary learning algorithm using the block-coordinate descent optimization technique is proposed. Then, the sparse codes of the desired HR hyperspectral image with respect to learned hyperspectral basis are estimated from the pair of LR and HR reference images. To improve the accuracy of non-negative sparse coding, a clustering-based structured sparse coding method is proposed to exploit the spatial correlation among the learned sparse codes. The experimental results on both public datasets and real LR hypspectral images suggest that the proposed method substantially outperforms several existing HR hyperspectral image recovery techniques in the literature in terms of both objective quality metrics and computational efficiency.

  20. Fast, electrically tunable filters for hyperspectral imaging

    Science.gov (United States)

    Liberman, V.; Parameswaran, L.; Gear, C.; Cabral, A.; Rothschild, M.

    2014-06-01

    Tunable, narrow-wavelength spectral filters with a ms response in the mid-wave/long-wave infrared (MW/LWIR) are an enabling technology for hyperspectral imaging systems. Few commercial off-the-shelf (COTS) components for this application exist, including filter wheels, movable gratings, and Fabry-Perot (FP) etalon-based devices. These devices can be bulky, fragile and often do not have the required response speed. Here, we present a fundamentally different approach for tunable reflective IR filters, based on coupling subwavelength plasmonic antenna arrays with liquid crystals (LCs). Our device operates in reflective mode and derives its narrow bandwidth from diffractive coupling of individual antenna elements. The wavelength tunability of the device arises from electrically-induced re-orientation of the LC material in intimate contact with antenna array. This re-orientation, in turn, induces a change in the local dielectric environment of the antenna array, leading to a wavelength shift. We will first present results of full-field optimization of micron-size antenna geometries to account for complex 3D LC anisotropy. We have fabricated these antenna arrays on IR-transparent CaF2 substrates utilizing electron beam lithography, and have demonstrated tunability using 5CB, a commercially available LC. However, the design can be extended to high-birefringence liquid crystals for an increased tuning range. Our initial results demonstrate <60% peak reflectance in the 4- 6 μm wavelength range with a tunability of 0.2 μm with re-orientation of the surface alignment layers. Preliminary electrical switching has been demonstrated and is being optimized.

  1. Reflectance Hyperspectral Imaging for Investigation of Works of Art: Old Master Paintings and Illuminated Manuscripts.

    Science.gov (United States)

    Cucci, Costanza; Delaney, John K; Picollo, Marcello

    2016-10-18

    Diffuse reflectance hyperspectral imaging, or reflectance imaging spectroscopy, is a sophisticated technique that enables the capture of hundreds of images in contiguous narrow spectral bands (bandwidth pixel in the image has an associated reflectance spectrum. This "big data" image-cube allows for the mining of artists' materials and mapping their distribution across the surface of a work of art. Reflectance hyperspectral imaging, introduced in the 1980s by Goetz and co-workers, led to a revolution in the field of remote sensing of the earth and near planets ( Goetz, F. H.; Vane, G.; Solomon, B. N.; Rock, N. Imaging Spectrometry for Earth Remote Sensing . Science , 1985 , 228 , 1147 - 1152 ). In the subsequent decades, thanks to rapid advances in solid-state sensor technology, reflectance hyperspectral imaging, once only available to large government laboratories, was extended to new fields of application, such as monitoring agri-foods, pharmaceutical products, the environment, and cultural heritage. In the 2000s, the potential of this noninvasive technology for the study of artworks became evident and, consequently, the methodology is becoming more widely used in the art conservation science field. Typically hyperspectral reflectance image-cubes contain millions of spectra. Many of these spectra are similar, making the reduction of the data set size an important first step. Thus, image-processing tools based on multivariate techniques, such as principal component analysis (PCA), automated classification methods, or expert knowledge systems, that search for known spectral features are often applied. These algorithms seek to reduce the large number of high-quality spectra to a common subset, which allow identifying and mapping artists' materials and alteration products. Hence, reflectance hyperspectral imaging is finding its place as the starting point to find sites on polychrome surfaces for spot analytical techniques, such as X-ray fluorescence, Raman

  2. Spatial-scanning hyperspectral imaging probe for bio-imaging applications.

    Science.gov (United States)

    Lim, Hoong-Ta; Murukeshan, Vadakke Matham

    2016-03-01

    The three common methods to perform hyperspectral imaging are the spatial-scanning, spectral-scanning, and snapshot methods. However, only the spectral-scanning and snapshot methods have been configured to a hyperspectral imaging probe as of today. This paper presents a spatial-scanning (pushbroom) hyperspectral imaging probe, which is realized by integrating a pushbroom hyperspectral imager with an imaging probe. The proposed hyperspectral imaging probe can also function as an endoscopic probe by integrating a custom fabricated image fiber bundle unit. The imaging probe is configured by incorporating a gradient-index lens at the end face of an image fiber bundle that consists of about 50,000 individual fiberlets. The necessary simulations, methodology, and detailed instrumentation aspects that are carried out are explained followed by assessing the developed probe's performance. Resolution test targets such as United States Air Force chart as well as bio-samples such as chicken breast tissue with blood clot are used as test samples for resolution analysis and for performance validation. This system is built on a pushbroom hyperspectral imaging system with a video camera and has the advantage of acquiring information from a large number of spectral bands with selectable region of interest. The advantages of this spatial-scanning hyperspectral imaging probe can be extended to test samples or tissues residing in regions that are difficult to access with potential diagnostic bio-imaging applications.

  3. Denoising of hyperspectral images by best multilinear rank approximation of a tensor

    Science.gov (United States)

    Marin-McGee, Maider; Velez-Reyes, Miguel

    2010-04-01

    The hyperspectral image cube can be modeled as a three dimensional array. Tensors and the tools of multilinear algebra provide a natural framework to deal with this type of mathematical object. Singular value decomposition (SVD) and its variants have been used by the HSI community for denoising of hyperspectral imagery. Denoising of HSI using SVD is achieved by finding a low rank approximation of a matrix representation of the hyperspectral image cube. This paper investigates similar concepts in hyperspectral denoising by using a low multilinear rank approximation the given HSI tensor representation. The Best Multilinear Rank Approximation (BMRA) of a given tensor A is to find a lower multilinear rank tensor B that is as close as possible to A in the Frobenius norm. Different numerical methods to compute the BMRA using Alternating Least Square (ALS) method and Newton's Methods over product of Grassmann manifolds are presented. The effect of the multilinear rank, the numerical method used to compute the BMRA, and different parameter choices in those methods are studied. Results show that comparable results are achievable with both ALS and Newton type methods. Also, classification results using the filtered tensor are better than those obtained either with denoising using SVD or MNF.

  4. Extended SWIR imaging sensors for hyperspectral imaging applications

    Science.gov (United States)

    Weber, A.; Benecke, M.; Wendler, J.; Sieck, A.; Hübner, D.; Figgemeier, H.; Breiter, R.

    2016-05-01

    AIM has developed SWIR modules including FPAs based on liquid phase epitaxy (LPE) grown MCT usable in a wide range of hyperspectral imaging applications. Silicon read-out integrated circuits (ROIC) provide various integration and readout modes including specific functions for spectral imaging applications. An important advantage of MCT based detectors is the tunable band gap. The spectral sensitivity of MCT detectors can be engineered to cover the extended SWIR spectral region up to 2.5μm without compromising in performance. AIM developed the technology to extend the spectral sensitivity of its SWIR modules also into the VIS. This has been successfully demonstrated for 384x288 and 1024x256 FPAs with 24μm pitch. Results are presented in this paper. The FPAs are integrated into compact dewar cooler configurations using different types of coolers, like rotary coolers, AIM's long life split linear cooler MCC030 or extreme long life SF100 Pulse Tube cooler. The SWIR modules include command and control electronics (CCE) which allow easy interfacing using a digital standard interface. The development status and performance results of AIM's latest MCT SWIR modules suitable for hyperspectral systems and applications will be presented.

  5. Hyperspectral and Multispectral Imaging Technique for Food Quality and Safety Evaluation

    Science.gov (United States)

    In this chapter, recently developed ARS line-scan hyperspectral-based sensing technologies to address agro-food safety concerns are presented including a case study using the laboratory-based hyperspectral imaging platforms. An online line-scan imaging system capable of both hyperspectral and multi...

  6. Vegetation Classification in Hyperspectral Image with CART Decision Tree%高光谱图像植被类型的CART决策树分类

    Institute of Scientific and Technical Information of China (English)

    董连英; 邢立新; 潘军; 王静; 李丽丽; 焦健楠

    2013-01-01

    为提高植被分类的精度,在利用高光谱图像提取植被信息时需要考虑训练样本和地形等其他因素的影响.以长白山为研究背景,基于CART(Classification And Regression Tree)算法构建决策树模型,对高光谱图像进行植被分类.由于混合像元的影响,以采用PPI(Pixel Purity Index)提取的纯净像元作为训练样本,提取植被指数、纹理和地形等分类特征变量.基于这些变量构建CART决策树对植被分类,并将结果与最大似然法分类结果进行比较.结果表明,CART决策树分类法可实现光谱、纹理和地形特征的有效组合,有较好的分类效果.

  7. Hyperspectral Imaging and Related Field Methods: Building the Science

    Science.gov (United States)

    Goetz, Alexander F. H.; Steffen, Konrad; Wessman, Carol

    1999-01-01

    The proposal requested funds for the computing power to bring hyperspectral image processing into undergraduate and graduate remote sensing courses. This upgrade made it possible to handle more students in these oversubscribed courses and to enhance CSES' summer short course entitled "Hyperspectral Imaging and Data Analysis" provided for government, industry, university and military. Funds were also requested to build field measurement capabilities through the purchase of spectroradiometers, canopy radiation sensors and a differential GPS system. These instruments provided systematic and complete sets of field data for the analysis of hyperspectral data with the appropriate radiometric and wavelength calibration as well as atmospheric data needed for application of radiative transfer models. The proposed field equipment made it possible to team-teach a new field methods course, unique in the country, that took advantage of the expertise of the investigators rostered in three different departments, Geology, Geography and Biology.

  8. a Review of Hyperspectral Imaging in Close Range Applications

    Science.gov (United States)

    Kurz, T. H.; Buckley, S. J.

    2016-06-01

    Hyperspectral imaging is an established method for material mapping, which has been conventionally applied from airborne and spaceborne platforms for a range of applications, including mineral and vegetation mapping, change detection and environmental studies. The main advantage of lightweight hyperspectral imagers lies in the flexibility to deploy them from various platforms (terrestrial imaging and from unmanned aerial vehicles; UAVs), as well as the high spectral resolution to cover an expanding wavelength range. In addition, spatial resolution allows object sampling distances from micrometres to tens of centimetres - complementary to conventional nadir-looking systems. When this new type of imaging device was initially released, few instruments were available and the applicability and potential of the method was restricted. Today, a wider range of instruments, with a range of specifications, is available, with significant improvements over the first generation of technology. In this contribution, the state-of-the-art of hyperspectral imaging will be reviewed from a close range measurement perspective, highlighting how the method supplements geometric modelling techniques. An overview of the processing workflow, adjusted to the more complex close range imaging scenario will be given. This includes the integration with 3D laser scanning and photogrammetric models to provide a geometric framework and real world coordinate system for the hyperspectral imagery.

  9. A REVIEW OF HYPERSPECTRAL IMAGING IN CLOSE RANGE APPLICATIONS

    Directory of Open Access Journals (Sweden)

    T. H. Kurz

    2016-06-01

    Full Text Available Hyperspectral imaging is an established method for material mapping, which has been conventionally applied from airborne and spaceborne platforms for a range of applications, including mineral and vegetation mapping, change detection and environmental studies. The main advantage of lightweight hyperspectral imagers lies in the flexibility to deploy them from various platforms (terrestrial imaging and from unmanned aerial vehicles; UAVs, as well as the high spectral resolution to cover an expanding wavelength range. In addition, spatial resolution allows object sampling distances from micrometres to tens of centimetres – complementary to conventional nadir-looking systems. When this new type of imaging device was initially released, few instruments were available and the applicability and potential of the method was restricted. Today, a wider range of instruments, with a range of specifications, is available, with significant improvements over the first generation of technology. In this contribution, the state-of-the-art of hyperspectral imaging will be reviewed from a close range measurement perspective, highlighting how the method supplements geometric modelling techniques. An overview of the processing workflow, adjusted to the more complex close range imaging scenario will be given. This includes the integration with 3D laser scanning and photogrammetric models to provide a geometric framework and real world coordinate system for the hyperspectral imagery.

  10. Evaluation of cross-polarized near infrared hyperspectral imaging for early detection of dental caries

    Science.gov (United States)

    Usenik, Peter; Bürmen, Miran; Fidler, Aleš; Pernuš, Franjo; Likar, Boštjan

    2012-01-01

    Despite major improvements in dental healthcare and oral hygiene, dental caries remains one of the most prevalent oral diseases and represents the primary cause of oral pain and tooth loss. The initial stages of dental caries are characterized by demineralization of enamel crystals and are difficult to diagnose. Near infrared (NIR) hyperspectral imaging is a new promising technique for detection of early changes in the surfaces of carious teeth. This noninvasive imaging technique can characterize and differentiate between the sound tooth surface and initial or advanced tooth caries. The absorbing and scattering properties of dental tissues reflect in distinct spectral features, which can be measured, quantified and used to accurately classify and map different dental tissues. Specular reflections from the tooth surface, which appear as bright spots, mostly located around the edges and the crests of the teeth, act as a noise factor which can significantly interfere with the spectral measurements and analysis of the acquired images, degrading the accuracy of the classification and diagnosis. Employing cross-polarized imaging setup can solve this problem, however has yet to be systematically evaluated, especially in broadband hyperspectral imaging setups. In this paper, we employ cross-polarized illumination setup utilizing state-of-the-art high-contrast broadband wire-grid polarizers in the spectral range from 900 nm to 1700 nm for hyperspectral imaging of natural and artificial carious lesions of various degrees.

  11. Classification in Medical Imaging

    DEFF Research Database (Denmark)

    Chen, Chen

    Classification is extensively used in the context of medical image analysis for the purpose of diagnosis or prognosis. In order to classify image content correctly, one needs to extract efficient features with discriminative properties and build classifiers based on these features. In addition......, a good metric is required to measure distance or similarity between feature points so that the classification becomes feasible. Furthermore, in order to build a successful classifier, one needs to deeply understand how classifiers work. This thesis focuses on these three aspects of classification...... to segment breast tissue and pectoral muscle area from the background in mammogram. The second focus is the choices of metric and its influence to the feasibility of a classifier, especially on k-nearest neighbors (k-NN) algorithm, with medical applications on breast cancer prediction and calcification...

  12. Real-time progressive hyperspectral image processing endmember finding and anomaly detection

    CERN Document Server

    Chang, Chein-I

    2016-01-01

    The book covers the most crucial parts of real-time hyperspectral image processing: causality and real-time capability. Recently, two new concepts of real time hyperspectral image processing, Progressive Hyperspectral Imaging (PHSI) and Recursive Hyperspectral Imaging (RHSI). Both of these can be used to design algorithms and also form an integral part of real time hyperpsectral image processing. This book focuses on progressive nature in algorithms on their real-time and causal processing implementation in two major applications, endmember finding and anomaly detection, both of which are fundamental tasks in hyperspectral imaging but generally not encountered in multispectral imaging. This book is written to particularly address PHSI in real time processing, while a book, Recursive Hyperspectral Sample and Band Processing: Algorithm Architecture and Implementation (Springer 2016) can be considered as its companion book. Includes preliminary background which is essential to those who work in hyperspectral ima...

  13. Classification of M1/M2-polarized human macrophages by label-free hyperspectral reflectance confocal microscopy and multivariate analysis.

    Science.gov (United States)

    Bertani, Francesca R; Mozetic, Pamela; Fioramonti, Marco; Iuliani, Michele; Ribelli, Giulia; Pantano, Francesco; Santini, Daniele; Tonini, Giuseppe; Trombetta, Marcella; Businaro, Luca; Selci, Stefano; Rainer, Alberto

    2017-08-21

    The possibility of detecting and classifying living cells in a label-free and non-invasive manner holds significant theranostic potential. In this work, Hyperspectral Imaging (HSI) has been successfully applied to the analysis of macrophagic polarization, given its central role in several pathological settings, including the regulation of tumour microenvironment. Human monocyte derived macrophages have been investigated using hyperspectral reflectance confocal microscopy, and hyperspectral datasets have been analysed in terms of M1 vs. M2 polarization by Principal Components Analysis (PCA). Following PCA, Linear Discriminant Analysis has been implemented for semi-automatic classification of macrophagic polarization from HSI data. Our results confirm the possibility to perform single-cell-level in vitro classification of M1 vs. M2 macrophages in a non-invasive and label-free manner with a high accuracy (above 98% for cells deriving from the same donor), supporting the idea of applying the technique to the study of complex interacting cellular systems, such in the case of tumour-immunity in vitro models.

  14. Hyperspectral imaging system for whole corn ear surface inspection

    Science.gov (United States)

    Yao, Haibo; Kincaid, Russell; Hruska, Zuzana; Brown, Robert L.; Bhatnagar, Deepak; Cleveland, Thomas E.

    2013-05-01

    Aflatoxin is a mycotoxin produced mainly by Aspergillus flavus (A.flavus) and Aspergillus parasitiucus fungi that grow naturally in corn. Very serious health problems such as liver damage and lung cancer can result from exposure to high toxin levels in grain. Consequently, many countries have established strict guidelines for permissible levels in consumables. Conventional chemical-based analytical methods used to screen for aflatoxin such as thin-layer chromatography (TLC) and high performance liquid chromatography (HPLC) are time consuming, expensive, and require the destruction of samples as well as proper training for data interpretation. Thus, it has been a continuing effort within the research community to find a way to rapidly and non-destructively detect and possibly quantify aflatoxin contamination in corn. One of the more recent developments in this area is the use of spectral technology. Specifically, fluorescence hyperspectral imaging offers a potential rapid, and non-invasive method for contamination detection in corn infected with toxigenic A.flavus spores. The current hyperspectral image system is designed for scanning flat surfaces, which is suitable for imaging single or a group of corn kernels. In the case of a whole corn cob, it is preferred to be able to scan the circumference of the corn ear, appropriate for whole ear inspection. This paper discusses the development of a hyperspectral imaging system for whole corn ear imaging. The new instrument is based on a hyperspectral line scanner using a rotational stage to turn the corn ear.

  15. Penetration Depth Measurement of Near-Infrared Hyperspectral Imaging Light for Milk Powder

    Directory of Open Access Journals (Sweden)

    Min Huang

    2016-03-01

    Full Text Available The increasingly common application of the near-infrared (NIR hyperspectral imaging technique to the analysis of food powders has led to the need for optical characterization of samples. This study was aimed at exploring the feasibility of quantifying penetration depth of NIR hyperspectral imaging light for milk powder. Hyperspectral NIR reflectance images were collected for eight different milk powder products that included five brands of non-fat milk powder and three brands of whole milk powder. For each milk powder, five different powder depths ranging from 1 mm–5 mm were prepared on the top of a base layer of melamine, to test spectral-based detection of the melamine through the milk. A relationship was established between the NIR reflectance spectra (937.5–1653.7 nm and the penetration depth was investigated by means of the partial least squares-discriminant analysis (PLS-DA technique to classify pixels as being milk-only or a mixture of milk and melamine. With increasing milk depth, classification model accuracy was gradually decreased. The results from the 1-mm, 2-mm and 3-mm models showed that the average classification accuracy of the validation set for milk-melamine samples was reduced from 99.86% down to 94.93% as the milk depth increased from 1 mm–3 mm. As the milk depth increased to 4 mm and 5 mm, model performance deteriorated further to accuracies as low as 81.83% and 58.26%, respectively. The results suggest that a 2-mm sample depth is recommended for the screening/evaluation of milk powders using an online NIR hyperspectral imaging system similar to that used in this study.

  16. Penetration Depth Measurement of Near-Infrared Hyperspectral Imaging Light for Milk Powder.

    Science.gov (United States)

    Huang, Min; Kim, Moon S; Chao, Kuanglin; Qin, Jianwei; Mo, Changyeun; Esquerre, Carlos; Delwiche, Stephen; Zhu, Qibing

    2016-03-25

    The increasingly common application of the near-infrared (NIR) hyperspectral imaging technique to the analysis of food powders has led to the need for optical characterization of samples. This study was aimed at exploring the feasibility of quantifying penetration depth of NIR hyperspectral imaging light for milk powder. Hyperspectral NIR reflectance images were collected for eight different milk powder products that included five brands of non-fat milk powder and three brands of whole milk powder. For each milk powder, five different powder depths ranging from 1 mm-5 mm were prepared on the top of a base layer of melamine, to test spectral-based detection of the melamine through the milk. A relationship was established between the NIR reflectance spectra (937.5-1653.7 nm) and the penetration depth was investigated by means of the partial least squares-discriminant analysis (PLS-DA) technique to classify pixels as being milk-only or a mixture of milk and melamine. With increasing milk depth, classification model accuracy was gradually decreased. The results from the 1-mm, 2-mm and 3-mm models showed that the average classification accuracy of the validation set for milk-melamine samples was reduced from 99.86% down to 94.93% as the milk depth increased from 1 mm-3 mm. As the milk depth increased to 4 mm and 5 mm, model performance deteriorated further to accuracies as low as 81.83% and 58.26%, respectively. The results suggest that a 2-mm sample depth is recommended for the screening/evaluation of milk powders using an online NIR hyperspectral imaging system similar to that used in this study.

  17. Hyperspectral imaging of endogenous fluorescent metabolic molecules to identify pain states in central nervous system tissue

    Science.gov (United States)

    Staikopoulos, Vasiliki; Gosnell, Martin E.; Anwer, Ayad G.; Mustafa, Sanam; Hutchinson, Mark R.; Goldys, Ewa M.

    2016-12-01

    Fluorescence-based bio-imaging methods have been extensively used to identify molecular changes occurring in biological samples in various pathological adaptations. Auto-fluorescence generated by endogenous fluorescent molecules within these samples can interfere with signal to background noise making positive antibody based fluorescent staining difficult to resolve. Hyperspectral imaging uses spectral and spatial imaging information for target detection and classification, and can be used to resolve changes in endogenous fluorescent molecules such as flavins, bound and free NADH and retinoids that are involved in cell metabolism. Hyperspectral auto-fluorescence imaging of spinal cord slices was used in this study to detect metabolic differences within pain processing regions of non-pain versus sciatic chronic constriction injury (CCI) animals, an established animal model of peripheral neuropathy. By using an endogenous source of contrast, subtle metabolic variations were detected between tissue samples, making it possible to distinguish between animals from non-injured and injured groups. Tissue maps of native fluorophores, flavins, bound and free NADH and retinoids unveiled subtle metabolic signatures and helped uncover significant tissue regions with compromised mitochondrial function. Taken together, our results demonstrate that hyperspectral imaging provides a new non-invasive method to investigate central changes of peripheral neuropathic injury and other neurodegenerative disease models, and paves the way for novel cellular characterisation in health, disease and during treatment, with proper account of intrinsic cellular heterogeneity.

  18. [Study on the Color Determination of Tomato Leaves Stressed by the High Temperature Based on Hyperspectral Imaging].

    Science.gov (United States)

    Xie, Chuan-qi; Saho, Yong-ni; Gao, Jun-feng; He, Yong

    2015-12-01

    Determination of color values on tomato leaves stressed by the high temperature using hyperspectral imaging technique was studied in this paper. Hyperspectral images of sixty healthy and sixty unhealthy tomato leaves in the wavelengths of 380-1023 nm were acquired by the hyperspectral imaging system. Simultaneously, three color parameters (L*, a* and b*) were measured by a colorimeter. Reflectance of all pixels in the region of interest (ROI) was extracted from the corrected hyperspectral image. Partial Least Squares (PLS) models were established based on different preprocessing methods. Successive Projections Algorithm (SPA) was identified to select effective wavelengths. Finally, Partial Least Squares-Discriminant Analysis (PLS-DA) models were built to classify different types of samples. The results showed that the determination coefficient (R²) were 0. 818, 0. 109 and 0. 896 in the prediction sets of PLS modes; 0.591, 0.244 and 0.673 in the prediction sets of SPA-PLS models. The overall classification accuracy in the prediction sets of PLS-DA models were over 77.50%. It demonstrated that it is feasible to measure color values on tomato leaves and identify different types of samples using hyperspectral imaging technique.

  19. Dimension Reduction and Classification of Hyperspectral Remote Sensing Images Based on Sensitivity Analysis of Artificial Neural Network%神经网络敏感性分析的高光谱遥感影像降维与分类方法

    Institute of Scientific and Technical Information of China (English)

    高红民; 李臣明; 周惠; 张振; 陈玲慧; 何振宇

    2016-01-01

    The high dimensions of hyperspectral remote sensing images will cause the redundancy of information and complexity of data processing, which also brings tremendous computing workload and damages application accuracy. Therefore, before the analysis of hyperspectral image processing, it is necessary to reduce the high dimensions of hyperspectral data. The Sensitivity Analysis (SA) of artificial neural network can be used in dimension reduction of the model. Now the Sensitivity Analysis of artificial neural network is applied to dimension reduction for hyperspectral remote sensing images in the paper. First of all, all bands are divided into several groups as long as a lower correlation exists between adjacent bands. Furthermore, Differential Evolution (DE) algorithm is used for optimizing neural network structure. Moreover, the bands which make small contribution will be given up based on Ruck sensitivity analysis method. Finally, experiments are conducted with AVIRIS images. The results show that the proposed method can get high classification accuracy of 85.83%at small training samples, 0.31%higher than the best one among other similar methods of dimension reduction and classification.%高光谱遥感影像由于其巨大的波段数直接导致信息的高冗余和数据处理的复杂,这不仅带来庞大的计算量,而且会损害分类精度.因此,在对高光谱影像进行处理、分析之前进行降维变得非常必要.神经网络敏感性分析可以用于对模型的简化降维,该文将该方法运用于高光谱遥感影像降维中,通过子空间划分弱化波段之间的相关性,利用差分进化算法(DE)优化神经网络结构,采用Ruck敏感性分析方法剔除掉对分类贡献较小的波段,从而实现降维.最后,采用AVIRIS影像进行实验,所提算法相比其他相近的降维与分类方法能获得更高的分类精度,达到85.83%,比其他相近方法中最优方法高出0.31%.

  20. Simultaneous Spectral-Spatial Feature Selection and Extraction for Hyperspectral Images.

    Science.gov (United States)

    Zhang, Lefei; Zhang, Qian; Du, Bo; Huang, Xin; Tang, Yuan Yan; Tao, Dacheng

    2016-09-12

    In hyperspectral remote sensing data mining, it is important to take into account of both spectral and spatial information, such as the spectral signature, texture feature, and morphological property, to improve the performances, e.g., the image classification accuracy. In a feature representation point of view, a nature approach to handle this situation is to concatenate the spectral and spatial features into a single but high dimensional vector and then apply a certain dimension reduction technique directly on that concatenated vector before feed it into the subsequent classifier. However, multiple features from various domains definitely have different physical meanings and statistical properties, and thus such concatenation has not efficiently explore the complementary properties among different features, which should benefit for boost the feature discriminability. Furthermore, it is also difficult to interpret the transformed results of the concatenated vector. Consequently, finding a physically meaningful consensus low dimensional feature representation of original multiple features is still a challenging task. In order to address these issues, we propose a novel feature learning framework, i.e., the simultaneous spectral-spatial feature selection and extraction algorithm, for hyperspectral images spectral-spatial feature representation and classification. Specifically, the proposed method learns a latent low dimensional subspace by projecting the spectral-spatial feature into a common feature space, where the complementary information has been effectively exploited, and simultaneously, only the most significant original features have been transformed. Encouraging experimental results on three public available hyperspectral remote sensing datasets confirm that our proposed method is effective and efficient.

  1. Bread Water Content Measurement Based on Hyperspectral Imaging

    DEFF Research Database (Denmark)

    Liu, Zhi; Møller, Flemming

    2011-01-01

    for bread quality based on near-infrared hyperspectral imaging against the conventional manual loss-in-weight method. For this purpose, the hyperspectral components unmixing technology is used for measuring the water content quantitatively. And the definition on bread water content index is presented......Water content is one of the most important properties of the bread for tasting assesment or store monitoring. Traditional bread water content measurement methods mostly are processed manually, which is destructive and time consuming. This paper proposes an automated water content measurement...... for this measurement. The proposed measurement scheme is relatively inexpensive to implement, easy to set up. The experimental results demonstrate the effectiveness....

  2. A linear mixture analysis-based compression for hyperspectral image analysis

    Energy Technology Data Exchange (ETDEWEB)

    C. I. Chang; I. W. Ginsberg

    2000-06-30

    In this paper, the authors present a fully constrained least squares linear spectral mixture analysis-based compression technique for hyperspectral image analysis, particularly, target detection and classification. Unlike most compression techniques that directly deal with image gray levels, the proposed compression approach generates the abundance fractional images of potential targets present in an image scene and then encodes these fractional images so as to achieve data compression. Since the vital information used for image analysis is generally preserved and retained in the abundance fractional images, the loss of information may have very little impact on image analysis. In some occasions, it even improves analysis performance. Airborne visible infrared imaging spectrometer (AVIRIS) data experiments demonstrate that it can effectively detect and classify targets while achieving very high compression ratios.

  3. Recent Advances in Techniques for Hyperspectral Image Processing

    Science.gov (United States)

    Plaza, Antonio; Benediktsson, Jon Atli; Boardman, Joseph W.; Brazile, Jason; Bruzzone, Lorenzo; Camps-Valls, Gustavo; Chanussot, Jocelyn; Fauvel, Mathieu; Gamba, Paolo; Gualtieri, Anthony; Marconcini, Mattia; Tilton, James C.; Trianni, Giovanna

    2009-01-01

    Imaging spectroscopy, also known as hyperspectral imaging, has been transformed in less than 30 years from being a sparse research tool into a commodity product available to a broad user community. Currently, there is a need for standardized data processing techniques able to take into account the special properties of hyperspectral data. In this paper, we provide a seminal view on recent advances in techniques for hyperspectral image processing. Our main focus is on the design of techniques able to deal with the highdimensional nature of the data, and to integrate the spatial and spectral information. Performance of the discussed techniques is evaluated in different analysis scenarios. To satisfy time-critical constraints in specific applications, we also develop efficient parallel implementations of some of the discussed algorithms. Combined, these parts provide an excellent snapshot of the state-of-the-art in those areas, and offer a thoughtful perspective on future potentials and emerging challenges in the design of robust hyperspectral imaging algorithms

  4. Infrared hyperspectral upconversion imaging using spatial object translation

    DEFF Research Database (Denmark)

    Kehlet, Louis Martinus; Sanders, Nicolai Højer; Tidemand-Lichtenberg, Peter;

    2015-01-01

    In this paper hyperspectral imaging in the mid-infrared wavelength region is realised using nonlinear frequency upconversion. The infrared light is converted to the near-infrared region for detection with a Si-based CCD camera. The object is translated in a predefined grid by motorized actuators...

  5. Using independent component analysis for material estimation in hyperspectral images.

    Science.gov (United States)

    Kuan, Chia-Yun; Healey, Glenn

    2004-06-01

    We develop a method for automated material estimation in hyperspectral images. The method models a hyperspectral pixel as a linear mixture of unknown materials. The method is particularly useful for applications in which material regions in a scene are smaller than one pixel. In contrast to many material estimation methods, the new method uses the statistics of large numbers of pixels rather than attempting to identify a small number of the purest pixels. The method is based on maximizing the independence of material abundances at each pixel. We show how independent component analysis algorithms can be adapted for use with this problem. We demonstrate properties of the method by application to airborne hyperspectral data.

  6. Passive IR polarimetric hyperspectral imaging contributions to multisensor humanitarian demining

    Science.gov (United States)

    Iannarilli, Frank J., Jr.; Scott, Herman E.; Jones, Stephen H.

    2001-10-01

    Supported by the Army Humanitarian Demining MURI, we most recently have focused on determining the unique strengths of passive IR sensing as a function of attribute diversity. Our initial findings identify polarimetric hyperspectral imaging.as a robust means to rapidly survey and detect partially exposed, non-metallic anti-personnel (AP) mines. We are investigating the discrimination gains expected from the combined polarimetric hyperspectral attributes under laboratory and field conditions. A principal components analysis of our earliest data indicates that this combination of attributes is about three times more effective in discriminating AP mines or mine-like materials than conventional hyperspectral sensing. In addition, we have uncovered a distinguishing spectral behavior of the Fresnel reflectance across resonance features that can be measured only by spectrally-resolved polarimetry.

  7. Hyperspectral imaging in medicine: image pre-processing problems and solutions in Matlab.

    Science.gov (United States)

    Koprowski, Robert

    2015-11-01

    The paper presents problems and solutions related to hyperspectral image pre-processing. New methods of preliminary image analysis are proposed. The paper shows problems occurring in Matlab when trying to analyse this type of images. Moreover, new methods are discussed which provide the source code in Matlab that can be used in practice without any licensing restrictions. The proposed application and sample result of hyperspectral image analysis.

  8. Automatic Denoising and Unmixing in Hyperspectral Image Processing

    Science.gov (United States)

    Peng, Honghong

    This thesis addresses two important aspects in hyperspectral image processing: automatic hyperspectral image denoising and unmixing. The first part of this thesis is devoted to a novel automatic optimized vector bilateral filter denoising algorithm, while the remainder concerns nonnegative matrix factorization with deterministic annealing for unsupervised unmixing in remote sensing hyperspectral images. The need for automatic hyperspectral image processing has been promoted by the development of potent hyperspectral systems, with hundreds of narrow contiguous bands, spanning the visible to the long wave infrared range of the electromagnetic spectrum. Due to the large volume of raw data generated by such sensors, automatic processing in the hyperspectral images processing chain is preferred to minimize human workload and achieve optimal result. Two of the mostly researched processing for such automatic effort are: hyperspectral image denoising, which is an important preprocessing step for almost all remote sensing tasks, and unsupervised unmixing, which decomposes the pixel spectra into a collection of endmember spectral signatures and their corresponding abundance fractions. Two new methodologies are introduced in this thesis to tackle the automatic processing problems described above. Vector bilateral filtering has been shown to provide good tradeoff between noise removal and edge degradation when applied to multispectral/hyperspectral image denoising. It has also been demonstrated to provide dynamic range enhancement of bands that have impaired signal to noise ratios. Typical vector bilateral filtering usage does not employ parameters that have been determined to satisfy optimality criteria. This thesis also introduces an approach for selection of the parameters of a vector bilateral filter through an optimization procedure rather than by ad hoc means. The approach is based on posing the filtering problem as one of nonlinear estimation and minimizing the Stein

  9. Real-time implementation of optimized maximum noise fraction transform for feature extraction of hyperspectral images

    Science.gov (United States)

    Wu, Yuanfeng; Gao, Lianru; Zhang, Bing; Zhao, Haina; Li, Jun

    2014-01-01

    We present a parallel implementation of the optimized maximum noise fraction (G-OMNF) transform algorithm for feature extraction of hyperspectral images on commodity graphics processing units (GPUs). The proposed approach explored the algorithm data-level concurrency and optimized the computing flow. We first defined a three-dimensional grid, in which each thread calculates a sub-block data to easily facilitate the spatial and spectral neighborhood data searches in noise estimation, which is one of the most important steps involved in OMNF. Then, we optimized the processing flow and computed the noise covariance matrix before computing the image covariance matrix to reduce the original hyperspectral image data transmission. These optimization strategies can greatly improve the computing efficiency and can be applied to other feature extraction algorithms. The proposed parallel feature extraction algorithm was implemented on an Nvidia Tesla GPU using the compute unified device architecture and basic linear algebra subroutines library. Through the experiments on several real hyperspectral images, our GPU parallel implementation provides a significant speedup of the algorithm compared with the CPU implementation, especially for highly data parallelizable and arithmetically intensive algorithm parts, such as noise estimation. In order to further evaluate the effectiveness of G-OMNF, we used two different applications: spectral unmixing and classification for evaluation. Considering the sensor scanning rate and the data acquisition time, the proposed parallel implementation met the on-board real-time feature extraction.

  10. Determining Spectral Reflectance Coefficients from Hyperspectral Images Obtained from Low Altitudes

    Science.gov (United States)

    Walczykowski, P.; Jenerowicz, A.; Orych, A.; Siok, K.

    2016-06-01

    Remote Sensing plays very important role in many different study fields, like hydrology, crop management, environmental and ecosystem studies. For all mentioned areas of interest different remote sensing and image processing techniques, such as: image classification (object and pixel- based), object identification, change detection, etc. can be applied. Most of this techniques use spectral reflectance coefficients as the basis for the identification and distinction of different objects and materials, e.g. monitoring of vegetation stress, identification of water pollutants, yield identification, etc. Spectral characteristics are usually acquired using discrete methods such as spectrometric measurements in both laboratory and field conditions. Such measurements however can be very time consuming, which has led many international researchers to investigate the reliability and accuracy of using image-based methods. According to published and ongoing studies, in order to acquire these spectral characteristics from images, it is necessary to have hyperspectral data. The presented article describes a series of experiments conducted using the push-broom Headwall MicroHyperspec A-series VNIR. This hyperspectral scanner allows for registration of images with more than 300 spectral channels with a 1.9 nm spectral bandwidth in the 380- 1000 nm range. The aim of these experiments was to establish a methodology for acquiring spectral reflectance characteristics of different forms of land cover using such sensor. All research work was conducted in controlled conditions from low altitudes. Hyperspectral images obtained with this specific type of sensor requires a unique approach in terms of post-processing, especially radiometric correction. Large amounts of acquired imagery data allowed the authors to establish a new post- processing approach. The developed methodology allowed the authors to obtain spectral reflectance coefficients from a hyperspectral sensor mounted on an

  11. DETERMINING SPECTRAL REFLECTANCE COEFFICIENTS FROM HYPERSPECTRAL IMAGES OBTAINED FROM LOW ALTITUDES

    Directory of Open Access Journals (Sweden)

    P. Walczykowski

    2016-06-01

    Full Text Available Remote Sensing plays very important role in many different study fields, like hydrology, crop management, environmental and ecosystem studies. For all mentioned areas of interest different remote sensing and image processing techniques, such as: image classification (object and pixel- based, object identification, change detection, etc. can be applied. Most of this techniques use spectral reflectance coefficients as the basis for the identification and distinction of different objects and materials, e.g. monitoring of vegetation stress, identification of water pollutants, yield identification, etc. Spectral characteristics are usually acquired using discrete methods such as spectrometric measurements in both laboratory and field conditions. Such measurements however can be very time consuming, which has led many international researchers to investigate the reliability and accuracy of using image-based methods. According to published and ongoing studies, in order to acquire these spectral characteristics from images, it is necessary to have hyperspectral data. The presented article describes a series of experiments conducted using the push-broom Headwall MicroHyperspec A-series VNIR. This hyperspectral scanner allows for registration of images with more than 300 spectral channels with a 1.9 nm spectral bandwidth in the 380- 1000 nm range. The aim of these experiments was to establish a methodology for acquiring spectral reflectance characteristics of different forms of land cover using such sensor. All research work was conducted in controlled conditions from low altitudes. Hyperspectral images obtained with this specific type of sensor requires a unique approach in terms of post-processing, especially radiometric correction. Large amounts of acquired imagery data allowed the authors to establish a new post- processing approach. The developed methodology allowed the authors to obtain spectral reflectance coefficients from a hyperspectral sensor

  12. Enhancement of Hyperspectral Real World Images Using Hybrid Domain Approach

    Directory of Open Access Journals (Sweden)

    Shyam Lal

    2013-04-01

    Full Text Available This paper presents enhancement of hyperspectral real world images using hybrid domain approach. The proposed method consists of three phases: In first phase the discrete wavelet transform is applied and approximation coefficient is selected. In second phase approximation coefficient of discrete wavelet transform of image is process by automatic contrast adjustment technique and in third phase it takes logarithmic of output of second phase and after that adaptive filtering is applied for image enhancement in frequency domain. To judge the superiority of proposed method the image quality parameters such as measure of enhancement (EME and measure of enhancement factor (EMF is evaluated. Therefore, a better value of EME and EMF implies that the visual quality of the enhanced image is good. Simulation results indicates that proposed method provides better results as compared to other state-of-art contrast enhancement algorithms for hyperspectral real world images. The proposed method is efficient and very effective method for contrast enhancement of hyperspectral real world images. This method can also be used in different applications where images are suffering from different contrast problems.

  13. Hyperspectral Image Classification via Kernel Sparse Representation

    Science.gov (United States)

    2013-01-01

    Vapnik, The Nature of Statistical Learning Theory. New York: Springer, 1995. [3] J. A. Gualtieri and R. F. Cromp, “Support vector machines for...Bruzzone, G. Camps-Valls, J. Chanussot, M. Fauvel, P. Gamba, A. Gualtieri , M. Marconcini, J. C. Tilton, and G. Trianni, “Recent advances in tech

  14. Hyperspectral image superresolution: An edge-preserving convex formulation

    CERN Document Server

    Simões, Miguel; Almeida, Luis B; Chanussot, Jocelyn

    2014-01-01

    Hyperspectral remote sensing images (HSIs) are characterized by having a low spatial resolution and a high spectral resolution, whereas multispectral images (MSIs) are characterized by low spectral and high spatial resolutions. These complementary characteristics have stimulated active research in the inference of images with high spatial and spectral resolutions from HSI-MSI pairs. In this paper, we formulate this data fusion problem as the minimization of a convex objective function containing two data-fitting terms and an edge-preserving regularizer. The data-fitting terms are quadratic and account for blur, different spatial resolutions, and additive noise; the regularizer, a form of vector Total Variation, promotes aligned discontinuities across the reconstructed hyperspectral bands. The optimization described above is rather hard, owing to its non-diagonalizable linear operators, to the non-quadratic and non-smooth nature of the regularizer, and to the very large size of the image to be inferred. We tac...

  15. Exploiting Structural Complexity for Robust and Rapid Hyperspectral Imaging

    CERN Document Server

    Ely, Gregory; Miller, Eric L

    2013-01-01

    This paper presents several strategies for spectral de-noising of hyperspectral images and hypercube reconstruction from a limited number of tomographic measurements. In particular we show that the non-noisy spectral data, when stacked across the spectral dimension, exhibits low-rank. On the other hand, under the same representation, the spectral noise exhibits a banded structure. Motivated by this we show that the de-noised spectral data and the unknown spectral noise and the respective bands can be simultaneously estimated through the use of a low-rank and simultaneous sparse minimization operation without prior knowledge of the noisy bands. This result is novel for for hyperspectral imaging applications. In addition, we show that imaging for the Computed Tomography Imaging Systems (CTIS) can be improved under limited angle tomography by using low-rank penalization. For both of these cases we exploit the recent results in the theory of low-rank matrix completion using nuclear norm minimization.

  16. Hyperspectral vibrational photoacoustic imaging of lipids and collagen

    Science.gov (United States)

    Wang, Pu; Wang, Ping; Wang, Han-Wei; Cheng, Ji-Xin

    2012-02-01

    The recently developed vibrational photoacoustic (VPA) microscopy allows bond-selective imaging of deep tissues by taking advantage of intrinsic contrast from harmonic vibration of C-H bonds. Due to the spectral similarity of molecules in the overtone vibration region, the compositional information is not available from VPA images acquired by single wavelength excitation. Here we demonstrate that lipids and collagen, two critical markers in many kinds of diseases, can be distinguished by hyperspectral VPA imaging. A phantom consisted of rat tail tendon (collagen) and fat tissue (lipids) was constructed. Wavelengths between 1650 and 1850 nm were scanned to excite the first overtone/combination vibration of C-H bond. B-scan hyperspectral VPA images, in which each pixel contains a spectrum, was analyzed by a Multivariate Curve Resolution - Alternating Least Squares (MCR-ALS) algorism to recover the spatial distribution of two chemical components in the phantom.

  17. Classification in Medical Imaging

    DEFF Research Database (Denmark)

    Chen, Chen

    Classification is extensively used in the context of medical image analysis for the purpose of diagnosis or prognosis. In order to classify image content correctly, one needs to extract efficient features with discriminative properties and build classifiers based on these features. In addition...... to segment breast tissue and pectoral muscle area from the background in mammogram. The second focus is the choices of metric and its influence to the feasibility of a classifier, especially on k-nearest neighbors (k-NN) algorithm, with medical applications on breast cancer prediction and calcification...

  18. Fast algorithm for exploring and compressing of large hyperspectral images

    DEFF Research Database (Denmark)

    Kucheryavskiy, Sergey

    2011-01-01

    A new method for calculation of latent variable space for exploratory analysis and dimension reduction of large hyperspectral images is proposed. The method is based on significant downsampling of image pixels with preservation of pixels’ structure in feature (variable) space. To achieve this, in...... can be used first of all for fast compression of large data arrays with principal component analysis or similar projection techniques....

  19. Hyperspectral image visualization using t-distributed stochastic neighbor embedding

    Science.gov (United States)

    Zhang, Biyin; Yu, Xin

    2015-12-01

    Hyperspectral image visualization reduces high-dimensional spectral bands to three color channels, which are sought in order to explain well the nonlinear data characteristics that are hidden in the high-dimensional spectral bands. Despite the surge in the linear visualization techniques, the development of nonlinear visualization has been limited. The paper presents a new technique for visualization of hyperspectral image using t-distributed stochastic neighbor embedding, called VHI-tSNE, which learns a nonlinear mapping between the high-dimensional spectral space and the three-dimensional color space. VHI-tSNE transforms hyperspectral data into bilateral probability similarities, and employs a heavy-tailed distribution in three-dimensional color space to alleviate the crowding problem and optimization problem in SNE technique. We evaluate the performance of VHI-tSNE in experiments on several hyperspectral imageries, in which we compare it to the performance of other state-of-art techniques. The results of experiments demonstrated the strength of the proposed technique.

  20. Detection of Fungus Infection on Petals of Rapeseed (Brassica napus L.) Using NIR Hyperspectral Imaging

    Science.gov (United States)

    Zhao, Yan-Ru; Yu, Ke-Qiang; Li, Xiaoli; He, Yong

    2016-12-01

    Infected petals are often regarded as the source for the spread of fungi Sclerotinia sclerotiorum in all growing process of rapeseed (Brassica napus L.) plants. This research aimed to detect fungal infection of rapeseed petals by applying hyperspectral imaging in the spectral region of 874–1734 nm coupled with chemometrics. Reflectance was extracted from regions of interest (ROIs) in the hyperspectral image of each sample. Firstly, principal component analysis (PCA) was applied to conduct a cluster analysis with the first several principal components (PCs). Then, two methods including X-loadings of PCA and random frog (RF) algorithm were used and compared for optimizing wavebands selection. Least squares-support vector machine (LS-SVM) methodology was employed to establish discriminative models based on the optimal and full wavebands. Finally, area under the receiver operating characteristics curve (AUC) was utilized to evaluate classification performance of these LS-SVM models. It was found that LS-SVM based on the combination of all optimal wavebands had the best performance with AUC of 0.929. These results were promising and demonstrated the potential of applying hyperspectral imaging in fungus infection detection on rapeseed petals.

  1. Background Radiance Estimation for Gas Plume Quantification for Airborne Hyperspectral Thermal Imaging

    Directory of Open Access Journals (Sweden)

    Ramzi Idoughi

    2016-01-01

    Full Text Available Hyperspectral imaging in the long-wave infrared (LWIR is a mean that is proving its worth in the characterization of gaseous effluent. Indeed the spectral and spatial resolution of acquisition instruments is steadily decreasing, making the gases characterization increasingly easy in the LWIR domain. The majority of literature algorithms exploit the plume contribution to the radiance corresponding to the difference of radiance between the plume-present and plume-absent pixels. Nevertheless, the off-plume radiance is unobservable using a single image. In this paper, we propose a new method to retrieve trace gas concentration from airborne infrared hyperspectral data. More particularly the outlined method improves the existing background radiance estimation approach to deal with heterogeneous scenes corresponding to industrial scenes. It consists in performing a classification of the scene and then applying a principal components analysis based method to estimate the background radiance on each cluster stemming from the classification. In order to determine the contribution of the classification to the background radiance estimation, we compared the two approaches on synthetic data and Telops Fourier Transform Spectrometer (FTS Imaging Hyper-Cam LW airborne acquisition above ethylene release. We finally show ethylene retrieved concentration map and estimate flow rate of the ethylene release.

  2. Hyperspectral image-based methods for spectral diversity

    Science.gov (United States)

    Sotomayor, Alejandro; Medina, Ollantay; Chinea, J. D.; Manian, Vidya

    2015-05-01

    Hyperspectral images are an important tool to assess ecosystem biodiversity. To obtain more precise analysis of biodiversity indicators that agree with indicators obtained using field data, analysis of spectral diversity calculated from images have to be validated with field based diversity estimates. The plant species richness is one of the most important indicators of biodiversity. This indicator can be measured in hyperspectral images considering the Spectral Variation Hypothesis (SVH) which states that the spectral heterogeneity is related to spatial heterogeneity and thus to species richness. The goal of this research is to capture spectral heterogeneity from hyperspectral images for a terrestrial neo tropical forest site using Vector Quantization (VQ) method and then use the result for prediction of plant species richness. The results are compared with that of Hierarchical Agglomerative Clustering (HAC). The validation of the process index is done calculating the Pearson correlation coefficient between the Shannon entropy from actual field data and the Shannon entropy computed in the images. One of the advantages of developing more accurate analysis tools would be the extension of the analysis to larger zones. Multispectral image with a lower spatial resolution has been evaluated as a prospective tool for spectral diversity.

  3. Face Recognition via Ensemble SIFT Matching of Uncorrelated Hyperspectral Bands and Spectral PCTs

    Science.gov (United States)

    2011-06-01

    Unsupervised Hyperspectral Target Detection Algorithm. Wright-Patterson AFB: Air Force Institute of Technology. Kirby, M., & Sirovich, L. (1990...Xu, Y., Hu, K., Tian, Y., & Peng, F. (2008). Classification of Hyperspectral imagery using SIFT for spectral matching. Image and Signal Processing...single hyperspectral band which also performs very well under temporal variation. 15. SUBJECT TERMS Face Recognition, SIFT, Ensemble Classification

  4. Hyperspectral Image Turbulence Measurements of the Atmosphere

    Science.gov (United States)

    Lane, Sarah E.; West, Leanne L.; Gimmestad, Gary G.; Kireev, Stanislav; Smith, William L., Sr.; Burdette, Edward M.; Daniels, Taumi; Cornman, Larry

    2012-01-01

    A Forward Looking Interferometer (FLI) sensor has the potential to be used as a means of detecting aviation hazards in flight. One of these hazards is mountain wave turbulence. The results from a data acquisition activity at the University of Colorado s Mountain Research Station will be presented here. Hyperspectral datacubes from a Telops Hyper-Cam are being studied to determine if evidence of a turbulent event can be identified in the data. These data are then being compared with D&P TurboFT data, which are collected at a much higher time resolution and broader spectrum.

  5. Nonlinear unmixing of hyperspectral images: models and algorithms

    CERN Document Server

    Dobigeon, Nicolas; Richard, Cédric; Bermudez, José C M; McLaughlin, Stephen; Hero, Alfred O

    2013-01-01

    When considering the problem of unmixing hyperspectral images, most of the literature in the geoscience and image processing areas rely on the widely acknowledged linear mixing model (LMM). However, in specific but common contexts, the LMM may be not valid and other nonlinear models should be invoked. Consequently, over the last few years, several significant contributions have been proposed to overcome the limitations inherent in the LMM. In this paper, we present an overview of recent advances that deal with the nonlinear unmixing problem. The main nonlinear models are introduced and their validity discussed. Then, we describe the main classes of unmixing strategies designed to solve the problem in supervised and unsupervised frameworks. Finally, the problem of detecting nonlinear mixtures in hyperspectral images is addressed.

  6. Lossless compression of hyperspectral images using hybrid context prediction.

    Science.gov (United States)

    Liang, Yuan; Li, Jianping; Guo, Ke

    2012-03-26

    In this letter a new algorithm for lossless compression of hyperspectral images using hybrid context prediction is proposed. Lossless compression algorithms are typically divided into two stages, a decorrelation stage and a coding stage. The decorrelation stage supports both intraband and interband predictions. The intraband (spatial) prediction uses the median prediction model, since the median predictor is fast and efficient. The interband prediction uses hybrid context prediction. The hybrid context prediction is the combination of a linear prediction (LP) and a context prediction. Finally, the residual image of hybrid context prediction is coded by the arithmetic coding. We compare the proposed lossless compression algorithm with some of the existing algorithms for hyperspectral images such as 3D-CALIC, M-CALIC, LUT, LAIS-LUT, LUT-NN, DPCM (C-DPCM), JPEG-LS. The performance of the proposed lossless compression algorithm is evaluated. Simulation results show that our algorithm achieves high compression ratios with low complexity and computational cost.

  7. An automatic fractional coefficient setting method of FODPSO for hyperspectral image segmentation

    Science.gov (United States)

    Xie, Weiying; Li, Yunsong

    2015-05-01

    In this paper, an automatic fractional coefficient setting method of fractional-order Darwinian particle swarm optimization (FODPSO) is proposed for hyperspectral image segmentation. The spectrum has been already taken into consideration by integrating various types of band selection algorithms, firstly. We provide a short overview of the hyperspectral image to select an appropriate set of bands by combining supervised, semi-supervised and unsupervised band selection algorithms. Some approaches are not limited in regards to their spectral dimension, but are limited with respect to their spatial dimension owing to low spatial resolution. The addition of spatial information will be focused on improving the performance of hyperspectral image segmentation for later fusion or classification. Many researchers have advocated that a large fractional coefficient should be in the exploration state while a small fractional coefficient should be in the exploitation, which does not mean the coefficient purely decrease with time. Due to such reasons, we propose an adaptive FODPSO by setting the fractional coefficient adaptively for the application of final hyperspectral image segmentation. In fact, the paper introduces an evolutionary factor to automatically control the fractional coefficient by using a sigmoid function. Therefore, fractional coefficient with large value will benefit the global search in the exploration state. Conversely, when the fractional coefficient has a small value, the exploitation state is detected. Hence, it can avoid optimization process get trapped into the local optima. Ultimately, the experimental segmentation results prove the validity and efficiency of our proposed automatic fractional coefficient setting method of FODPSO compared with traditional PSO, DPSO and FODPSO.

  8. Modelling the appearance of chromatic environment using hyperspectral imaging

    Science.gov (United States)

    Fomins, S.; Ozolinsh, M.

    2013-11-01

    Color of objects is a spectral composition of incident light source, reflection properties of the object itself, and spectral tuning of the eye. Light sources with different spectral characteristics can produce metameric representation of color; however most variable in this regard is vision. Pigments of color vision are continuously bleached by different stimuli and optical density of the pigment is changed, while continuous conditions provide an adaptation and perception of white. Special cases are color vision deficiencies which cover almost 8 % of male population in Europe. Hyperspectral imaging allows obtaining the spectra of the environment and modelling the performance of the dichromatic, anomalous trichromatic, as also normal trichromatic adapted behavior. First, CRI Nuance hyperspectral imaging system was spectrally calibrated for natural continuous spectral illumination of high color rendering index and narrow band fluorescent light sources. Full-scale images of color deficiency tests were acquired in the range of 420 to 720 nm to evaluate the modelling capacity for dichromatic and anomalous trichromatic vision. Hyperspectral images were turned to cone excitation images according to Stockman and Sharpe (2000) 1. Further, model was extended for anomalous trichromacy conditions. Cone sensitivity spectra were shifted by 4 nm according to each anomaly type. LWS and SWS cone signals were balanced in each condition to provide the appropriate appearance of colors in CIE system.

  9. Color measurement of tea leaves at different drying periods using hyperspectral imaging technique.

    Directory of Open Access Journals (Sweden)

    Chuanqi Xie

    Full Text Available This study investigated the feasibility of using hyperspectral imaging technique for nondestructive measurement of color components (ΔL*, Δa* and Δb* and classify tea leaves during different drying periods. Hyperspectral images of tea leaves at five drying periods were acquired in the spectral region of 380-1030 nm. The three color features were measured by the colorimeter. Different preprocessing algorithms were applied to select the best one in accordance with the prediction results of partial least squares regression (PLSR models. Competitive adaptive reweighted sampling (CARS and successive projections algorithm (SPA were used to identify the effective wavelengths, respectively. Different models (least squares-support vector machine [LS-SVM], PLSR, principal components regression [PCR] and multiple linear regression [MLR] were established to predict the three color components, respectively. SPA-LS-SVM model performed excellently with the correlation coefficient (rp of 0.929 for ΔL*, 0.849 for Δa*and 0.917 for Δb*, respectively. LS-SVM model was built for the classification of different tea leaves. The correct classification rates (CCRs ranged from 89.29% to 100% in the calibration set and from 71.43% to 100% in the prediction set, respectively. The total classification results were 96.43% in the calibration set and 85.71% in the prediction set. The result showed that hyperspectral imaging technique could be used as an objective and nondestructive method to determine color features and classify tea leaves at different drying periods.

  10. Classification of Hyperspectral or Trichromatic Measurements of Ocean Color Data into Spectral Classes

    Directory of Open Access Journals (Sweden)

    Dilip K. Prasad

    2016-03-01

    Full Text Available We propose a method for classifying radiometric oceanic color data measured by hyperspectral satellite sensors into known spectral classes, irrespective of the downwelling irradiance of the particular day, i.e., the illumination conditions. The focus is not on retrieving the inherent optical properties but to classify the pixels according to the known spectral classes of the reflectances from the ocean. The method compensates for the unknown downwelling irradiance by white balancing the radiometric data at the ocean pixels using the radiometric data of bright pixels (typically from clouds. The white-balanced data is compared with the entries in a pre-calibrated lookup table in which each entry represents the spectral properties of one class. The proposed approach is tested on two datasets of in situ measurements and 26 different daylight illumination spectra for medium resolution imaging spectrometer (MERIS, moderate-resolution imaging spectroradiometer (MODIS, sea-viewing wide field-of-view sensor (SeaWiFS, coastal zone color scanner (CZCS, ocean and land colour instrument (OLCI, and visible infrared imaging radiometer suite (VIIRS sensors. Results are also shown for CIMEL’s SeaPRISM sun photometer sensor used on-board field trips. Accuracy of more than 92% is observed on the validation dataset and more than 86% is observed on the other dataset for all satellite sensors. The potential of applying the algorithms to non-satellite and non-multi-spectral sensors mountable on airborne systems is demonstrated by showing classification results for two consumer cameras. Classification on actual MERIS data is also shown. Additional results comparing the spectra of remote sensing reflectance with level 2 MERIS data and chlorophyll concentration estimates of the data are included.

  11. Classification of Hyperspectral or Trichromatic Measurements of Ocean Color Data into Spectral Classes

    Science.gov (United States)

    Prasad, Dilip K.; Agarwal, Krishna

    2016-01-01

    We propose a method for classifying radiometric oceanic color data measured by hyperspectral satellite sensors into known spectral classes, irrespective of the downwelling irradiance of the particular day, i.e., the illumination conditions. The focus is not on retrieving the inherent optical properties but to classify the pixels according to the known spectral classes of the reflectances from the ocean. The method compensates for the unknown downwelling irradiance by white balancing the radiometric data at the ocean pixels using the radiometric data of bright pixels (typically from clouds). The white-balanced data is compared with the entries in a pre-calibrated lookup table in which each entry represents the spectral properties of one class. The proposed approach is tested on two datasets of in situ measurements and 26 different daylight illumination spectra for medium resolution imaging spectrometer (MERIS), moderate-resolution imaging spectroradiometer (MODIS), sea-viewing wide field-of-view sensor (SeaWiFS), coastal zone color scanner (CZCS), ocean and land colour instrument (OLCI), and visible infrared imaging radiometer suite (VIIRS) sensors. Results are also shown for CIMEL’s SeaPRISM sun photometer sensor used on-board field trips. Accuracy of more than 92% is observed on the validation dataset and more than 86% is observed on the other dataset for all satellite sensors. The potential of applying the algorithms to non-satellite and non-multi-spectral sensors mountable on airborne systems is demonstrated by showing classification results for two consumer cameras. Classification on actual MERIS data is also shown. Additional results comparing the spectra of remote sensing reflectance with level 2 MERIS data and chlorophyll concentration estimates of the data are included. PMID:27011185

  12. Classification of Hyperspectral or Trichromatic Measurements of Ocean Color Data into Spectral Classes.

    Science.gov (United States)

    Prasad, Dilip K; Agarwal, Krishna

    2016-03-22

    We propose a method for classifying radiometric oceanic color data measured by hyperspectral satellite sensors into known spectral classes, irrespective of the downwelling irradiance of the particular day, i.e., the illumination conditions. The focus is not on retrieving the inherent optical properties but to classify the pixels according to the known spectral classes of the reflectances from the ocean. The method compensates for the unknown downwelling irradiance by white balancing the radiometric data at the ocean pixels using the radiometric data of bright pixels (typically from clouds). The white-balanced data is compared with the entries in a pre-calibrated lookup table in which each entry represents the spectral properties of one class. The proposed approach is tested on two datasets of in situ measurements and 26 different daylight illumination spectra for medium resolution imaging spectrometer (MERIS), moderate-resolution imaging spectroradiometer (MODIS), sea-viewing wide field-of-view sensor (SeaWiFS), coastal zone color scanner (CZCS), ocean and land colour instrument (OLCI), and visible infrared imaging radiometer suite (VIIRS) sensors. Results are also shown for CIMEL's SeaPRISM sun photometer sensor used on-board field trips. Accuracy of more than 92% is observed on the validation dataset and more than 86% is observed on the other dataset for all satellite sensors. The potential of applying the algorithms to non-satellite and non-multi-spectral sensors mountable on airborne systems is demonstrated by showing classification results for two consumer cameras. Classification on actual MERIS data is also shown. Additional results comparing the spectra of remote sensing reflectance with level 2 MERIS data and chlorophyll concentration estimates of the data are included.

  13. The challenges of analysing blood stains with hyperspectral imaging

    Science.gov (United States)

    Kuula, J.; Puupponen, H.-H.; Rinta, H.; Pölönen, I.

    2014-06-01

    Hyperspectral imaging is a potential noninvasive technology for detecting, separating and identifying various substances. In the forensic and military medicine and other CBRNE related use it could be a potential method for analyzing blood and for scanning other human based fluids. For example, it would be valuable to easily detect whether some traces of blood are from one or more persons or if there are some irrelevant substances or anomalies in the blood. This article represents an experiment of separating four persons' blood stains on a white cotton fabric with a SWIR hyperspectral camera and FT-NIR spectrometer. Each tested sample includes standardized 75 _l of 100 % blood. The results suggest that on the basis of the amount of erythrocytes in the blood, different people's blood might be separable by hyperspectral analysis. And, referring to the indication given by erythrocytes, there might be a possibility to find some other traces in the blood as well. However, these assumptions need to be verified with wider tests, as the number of samples in the study was small. According to the study there also seems to be several biological, chemical and physical factors which affect alone and together on the hyperspectral analyzing results of blood on fabric textures, and these factors need to be considered before making any further conclusions on the analysis of blood on various materials.

  14. Linear Mixture Models and Partial Unmixing in Multi- and Hyperspectral Image Data

    DEFF Research Database (Denmark)

    Nielsen, Allan Aasbjerg

    1998-01-01

    As a supplement or an alternative to classification of hyperspectral image data the linear mixture model is considered in order to obtain estimates of abundance of each class or end-member in pixels with mixed membership. Full unmixing and the partial unmixing methods orthogonal subspace projection...... (OSP), constrained energy minimization (CEM) and an eigenvalue formulation alternative are dealt with. The solution to the eigenvalue formulation alternative proves to be identical to the CEM solution. The matrix inversion involved in CEM can be avoided by working on (a subset of) orthogonally...

  15. Optimized extreme learning machine for urban land cover classification using hyperspectral imagery

    Science.gov (United States)

    Su, Hongjun; Tian, Shufang; Cai, Yue; Sheng, Yehua; Chen, Chen; Najafian, Maryam

    2016-11-01

    This work presents a new urban land cover classification framework using the firefly algorithm (FA) optimized extreme learning machine (ELM). FA is adopted to optimize the regularization coefficient C and Gaussian kernel s for kernel ELM. Additionally, effectiveness of spectral features derived from an FA-based band selection algorithm is studied for the proposed classification task. Three sets of hyperspectral databases were recorded using different sensors, namely HYDICE, HyMap, and AVIRIS. Our study shows that the proposed method outperforms traditional classification algorithms such as SVM and reduces computational cost significantly.

  16. Identification of pesticide varieties by testing microalgae using Visible/Near Infrared Hyperspectral Imaging technology

    Science.gov (United States)

    Shao, Yongni; Jiang, Linjun; Zhou, Hong; Pan, Jian; He, Yong

    2016-04-01

    In our study, the feasibility of using visible/near infrared hyperspectral imaging technology to detect the changes of the internal components of Chlorella pyrenoidosa so as to determine the varieties of pesticides (such as butachlor, atrazine and glyphosate) at three concentrations (0.6 mg/L, 3 mg/L, 15 mg/L) was investigated. Three models (partial least squares discriminant analysis combined with full wavelengths, FW-PLSDA; partial least squares discriminant analysis combined with competitive adaptive reweighted sampling algorithm, CARS-PLSDA; linear discrimination analysis combined with regression coefficients, RC-LDA) were built by the hyperspectral data of Chlorella pyrenoidosa to find which model can produce the most optimal result. The RC-LDA model, which achieved an average correct classification rate of 97.0% was more superior than FW-PLSDA (72.2%) and CARS-PLSDA (84.0%), and it proved that visible/near infrared hyperspectral imaging could be a rapid and reliable technique to identify pesticide varieties. It also proved that microalgae can be a very promising medium to indicate characteristics of pesticides.

  17. [Lossless compression of hyperspectral image for space-borne application].

    Science.gov (United States)

    Li, Jin; Jin, Long-xu; Li, Guo-ning

    2012-08-01

    In order to resolve the difficulty in hardware implementation, lower compression ratio and time consuming for the whole hyperspectral image lossless compression algorithm based on the prediction, transform, vector quantization and their combination, a hyperspectral image lossless compression algorithm for space-borne application was proposed in the present paper. Firstly, intra-band prediction is used only for the first image along the spectral line using a median predictor. And inter- band prediction is applied to other band images. A two-step and bidirectional prediction algorithm is proposed for the inter-band prediction. In the first step prediction, a bidirectional and second order predictor proposed is used to obtain a prediction reference value. And a improved LUT prediction algorithm proposed is used to obtain four values of LUT prediction. Then the final prediction is obtained through comparison between them and the prediction reference. Finally, the verification experiments for the compression algorithm proposed using compression system test equipment of XX-X space hyperspectral camera were carried out. The experiment results showed that compression system can be fast and stable work. The average compression ratio reached 3.05 bpp. Compared with traditional approaches, the proposed method could improve the average compression ratio by 0.14-2.94 bpp. They effectively improve the lossless compression ratio and solve the difficulty of hardware implementation of the whole wavelet-based compression scheme.

  18. Miniaturized hyperspectral imager calibration and UAV flight campaigns

    Science.gov (United States)

    Saari, Heikki; Pölönen, Ilkka; Salo, Heikki; Honkavaara, Eija; Hakala, Teemu; Holmlund, Christer; Mäkynen, Jussi; Mannila, Rami; Antila, Tapani; Akujärvi, Altti

    2013-10-01

    VTT Technical Research Centre of Finland has developed Tunable Fabry-Perot Interferometer (FPI) based miniaturized hyperspectral imager which can be operated from light weight Unmanned Aerial Vehicles (UAV). The concept of the hyperspectral imager has been published in the SPIE Proc. 7474, 8174 and 8374. This instrument requires dedicated laboratory and on-board calibration procedures which are described. During summer 2012 extensive UAV Hyperspectral imaging campaigns in the wavelength range 400 - 900 nm at resolution range 10 - 40 nm @ FWHM were performed to study forest inventory, crop biomass and nitrogen distributions and environmental status of natural water applications. The instrument includes spectral band limiting filters which can be used for the on-board wavelength scale calibration by scanning the FPI pass band center wavelength through the low and high edge of the operational wavelength band. The procedure and results of the calibration tests will be presented. A short summary of the performed extensive UAV imaging campaign during summer 2012 will be presented.

  19. Random Forest (RF) Wrappers for Waveband Selection and Classification of Hyperspectral Data.

    Science.gov (United States)

    Poona, Nitesh Keshavelal; van Niekerk, Adriaan; Nadel, Ryan Leslie; Ismail, Riyad

    2016-02-01

    Hyperspectral data collected using a field spectroradiometer was used to model asymptomatic stress in Pinus radiata and Pinus patula seedlings infected with the pathogen Fusarium circinatum. Spectral data were analyzed using the random forest algorithm. To improve the classification accuracy of the model, subsets of wavebands were selected using three feature selection algorithms: (1) Boruta; (2) recursive feature elimination (RFE); and (3) area under the receiver operating characteristic curve of the random forest (AUC-RF). Results highlighted the robustness of the above feature selection methods when used in conjunction with the random forest algorithm for analyzing hyperspectral data. Overall, the Boruta feature selection algorithm provided the best results. When discriminating F. circinatum stress in Pinus radiata seedlings, Boruta selected wavebands (n = 69) yielded the best overall classification accuracies (training error of 17.00%, independent test error of 17.00% and an AUC value of 0.91). Classification results were, however, significantly lower for P. patula seedlings, with a training error of 24.00%, independent test error of 38.00%, and an AUC value of 0.65. A hybrid selection method that utilizes combinations of wavebands selected from the three feature selection algorithms was also tested. The hybrid method showed an improvement in classification accuracies for P. patula, and no improvement for P. radiata. The results of this study provide impetus towards implementing a hyperspectral framework for detecting stress within nursery environments.

  20. Imaging of blood cells based on snapshot Hyper-Spectral Imaging systems

    Science.gov (United States)

    Robison, Christopher J.; Kolanko, Christopher; Bourlai, Thirimachos; Dawson, Jeremy M.

    2015-05-01

    Snapshot Hyper-Spectral imaging systems are capable of capturing several spectral bands simultaneously, offering coregistered images of a target. With appropriate optics, these systems are potentially able to image blood cells in vivo as they flow through a vessel, eliminating the need for a blood draw and sample staining. Our group has evaluated the capability of a commercial Snapshot Hyper-Spectral imaging system, the Arrow system from Rebellion Photonics, in differentiating between white and red blood cells on unstained blood smear slides. We evaluated the imaging capabilities of this hyperspectral camera; attached to a microscope at varying objective powers and illumination intensity. Hyperspectral data consisting of 25, 443x313 hyperspectral bands with ~3nm spacing were captured over the range of 419 to 494nm. Open-source hyper-spectral data cube analysis tools, used primarily in Geographic Information Systems (GIS) applications, indicate that white blood cells features are most prominent in the 428-442nm band for blood samples viewed under 20x and 50x magnification over a varying range of illumination intensities. These images could potentially be used in subsequent automated white blood cell segmentation and counting algorithms for performing in vivo white blood cell counting.

  1. Characterization of Chromobacterium violaceum pigment through a hyperspectral imaging system

    OpenAIRE

    Gallardo, Maria J; Staforelli, Juan P; Meza, Pablo; Bordeu, Ignacio; Torres, Sergio

    2014-01-01

    In this paper, a comprehensive spatio-spectral and temporal analysis for Chromobacterium violaceum colonies is reported. A hyperspectral imaging (HSI) system is used to recover the spectral signatures of pigment production in a non-homogeneous media with high spectral resolution and high sensitivity in vivo, without destructing the sample. This non-contact sensing technique opens avenues to study the temporal growing of a specific section in the bacterial colony. Further, from a 580 [nm] and ...

  2. Assessing pasture quality and degradation status using hyperspectral imaging: a case study from western Tibet

    Science.gov (United States)

    Lehnert, Lukas W.; Meyer, Hanna; Meyer, Nele; Reudenbach, Christoph; Bendix, Jörg

    2013-10-01

    Alpine grasslands on the Tibetan Plateau (TP) are suffering from pasture degradation induced by over-grazing, climate change and improper livestock management. Meanwhile, the status of pastures is largely unknown especially in poor accessible western parts on the TP. The aim of this case study was to assess the suitability of hyperspectral imaging to predict quality and amount of forage on the western TP. Therefore, 18 ground- based hyperspectral images taken along two transects on a winter pasture were used to estimate leaf chlorophyll content, photosynthetic-active vegetation cover (PV) and proportion of grasses. For calibration and validation purposes, chlorophyll content of 20 grass plants was measured in situ. From the images reference spectra of grass and non-grass species were collected. PV was assessed from similarity of images to mean vegetation spectra using spectral angle mapper and threshold classifications. A set of 48 previously published hyperspectral vegetation indices (VI) was used as predictors to estimate chlorophyll content and to discriminate grass and non-grass pixels. Separation into grass and non-grass species was performed using partial least squares (PLS) discriminant analysis and chlorophyll content was estimated with PLS regression. The accuracy of the models was assessed with leave-one-out cross validation and normalised root mean square errors (nRMSE) for chlorophyll and contingency matrices for grass classification and total PV separation. Highest error rates were observed for discrimination between vegetated and non-vegetated parts (Overall accuracy = 0.85), whilst accuracies of grass and non grass separation (Overall accuracy = 0.98) and chlorophyll estimation were higher (nRMSE = 10.7).

  3. Optimizing Hyperspectral Imagery Anomaly Detection through Robust Parameter Design

    Science.gov (United States)

    2011-10-01

    Geoscience and Remote Sensing, IEEE Transactions on, 44 (8):2282 –2291 (2006). [8] Benediktsson, J.A., et al. “ Classification of hyperspectral data...1873 vol.4. 2001. [18] Chiang, S.S et al. “ Unsupervised Hyperspectral Image Analysis Using Indepen- dent Component Analysis.”. [19] Copeland, K.A.F. and...3):337 – 352 (2004). [34] Harsanyi, J.C. and C.I. Chang. “ Hyperspectral image classification and dimen- sionality reduction: an orthogonal subspace

  4. A Novel Spatial-Spectral Sparse Representation for Hyperspectral Image Classification Based on Neighborhood Segmentation%基于邻域分割的空谱联合稀疏表示高光谱图像分类技术研究

    Institute of Scientific and Technical Information of China (English)

    王彩玲; 王洪伟; 胡炳樑; 温佳; 徐君; 李湘眷

    2016-01-01

    Traditional hyperspectral image classification algorithms focus on spectral information application,however,with the increase of spatial resolution of hyperspectral remote sensing images,hyperspectral imaging presents clustering properties on spatial domain for the same category.It is critical for hyperspectral image classification algorithms to use spatial information in order to improve the classification accuracy.However,the marginal differences of different categories display more obviously.If it is introduced directly into the spatial-spectral sparse representation for image classification without the selection of neighbor-hood pixels,the classification error and the computation time will increase.This paper presents a spatial-spectral joint sparse representation classification algorithm based on neighborhood segmentation.The algorithm calculates the similarity with spectral angel in order to choose proper neighborhood pixel into spatial-spectral joint sparse representation model.With simultaneous subspace pursuit and simultaneous orthogonal matching pursuit to solve the model,the classification is determined by computing the minimum reconstruction error between testing samples and training pixels.Two typical hyperspectral images from AVIRIS and ROSIS are chosen for simulation experiment and results display that the classification accuracy of two images both improves as neighborhood segmentation threshold increasing.It concludes that neighborhood segmentation is necessary for joint sparse representation classification.%传统的高光谱遥感影像分类算法侧重于光谱信息的应用。随着高光谱遥感影像的空间分辨率的增加,高光谱影像中相同类别的地物在空间分布上呈现聚类特性,将空间特性有效地应用于高光谱遥感影像分类算法对分类精度的提升非常关键。但是,高光谱影像的高分辨率提供空间聚类特性的同时,在不同地物边缘处表现出的差异性更加明显

  5. Maximum relevance, minimum redundancy band selection based on neighborhood rough set for hyperspectral data classification

    Science.gov (United States)

    Liu, Yao; Chen, Yuehua; Tan, Kezhu; Xie, Hong; Wang, Liguo; Yan, Xiaozhen; Xie, Wu; Xu, Zhen

    2016-12-01

    Band selection is considered to be an important processing step in handling hyperspectral data. In this work, we selected informative bands according to the maximal relevance minimal redundancy (MRMR) criterion based on neighborhood mutual information. Two measures MRMR difference and MRMR quotient were defined and a forward greedy search for band selection was constructed. The performance of the proposed algorithm, along with a comparison with other methods (neighborhood dependency measure based algorithm, genetic algorithm and uninformative variable elimination algorithm), was studied using the classification accuracy of extreme learning machine (ELM) and random forests (RF) classifiers on soybeans’ hyperspectral datasets. The results show that the proposed MRMR algorithm leads to promising improvement in band selection and classification accuracy.

  6. Pixelated camouflage patterns from the perspective of hyperspectral imaging

    Science.gov (United States)

    Racek, František; Jobánek, Adam; Baláž, Teodor; Krejčí, Jaroslav

    2016-10-01

    Pixelated camouflage patterns fulfill the role of both principles the matching and the disrupting that are exploited for blending the target into the background. It means that pixelated pattern should respect natural background in spectral and spatial characteristics embodied in micro and macro patterns. The HS imaging plays the similar, however the reverse role in the field of reconnaissance systems. The HS camera fundamentally records and extracts both the spectral and spatial information belonging to the recorded scenery. Therefore, the article deals with problems of hyperspectral (HS) imaging and subsequent processing of HS images of pixelated camouflage patterns which are among others characterized by their specific spatial frequency heterogeneity.

  7. A new approach for discrimination of objects on hyperspectral images

    DEFF Research Database (Denmark)

    Kucheryavskiy, Sergey

    2013-01-01

    A new approach for discrimination of objects on hyperspectral images, which combines state-of-art image processing methods and multivariate image analysis, is proposed. The basic idea of the approach is to build a joint principal component space for all objects' pixels, detect patterns, pixels from...... a particular object shared in this space, and use quantitative evaluation of the patterns as the objects' features. The approach was particularly developed for dealing with challenging cases, when objects from different classes have many similar pixels. It has been tested on several real cases and showed very...

  8. An application based on the decision tree to classify the marbling of beef by hyperspectral imaging.

    Science.gov (United States)

    Velásquez, Lía; Cruz-Tirado, J P; Siche, Raúl; Quevedo, Roberto

    2017-11-01

    The aim of this study was to develop a system to classify the marbling of beef using the hyperspectral imaging technology. The Japanese standard classification of the degree of marbling of beef was used as reference and twelve standards were digitized to obtain the parameters of shape and spatial distribution of marbling of each class. A total of 35 samples M. longissmus dorsi muscle were scanned by the hyperspectral imaging system of 400-1000 nm in reflectance mode. The wavelength of 528nm was selected to segment the sample and the background, and 440nm was used for classified the samples. Processing algorithms on image, based on decision tree method, were used in the region of interest obtaining a classification error of 0.08% in the building stage. The results showed that the proposed technique has a great potential, as a non-destructive and fast technique, that can be used to classify beef with respect to the degree of marbling. Copyright © 2017 Elsevier Ltd. All rights reserved.

  9. Aflatoxin contaminated chili pepper detection by hyperspectral imaging and machine learning

    Science.gov (United States)

    Atas, Musa; Yardimci, Yasemin; Temizel, Alptekin

    2011-06-01

    Mycotoxins are toxic secondary metabolites produced by fungi. They have been demonstrated to cause various health problems in humans, including immunosuppression and cancer. A class of mycotoxins, aflatoxins, has been studied extensively because they have caused many deaths particularly in developing countries. Chili pepper is also prone to aflatoxin contamination during harvesting, production and storage periods. Chemical methods to detect aflatoxins are quite accurate but expensive and destructive in nature. Hyperspectral and multispectral imaging are becoming increasingly important for rapid and nondestructive testing for the presence of such contaminants. We propose a compact machine vision system based on hyperspectral imaging and machine learning for detection of aflatoxin contaminated chili peppers. We used the difference images of consecutive spectral bands along with individual band energies to classify chili peppers into aflatoxin contaminated and uncontaminated classes. Both UV and halogen illumination sources were used in the experiments. The significant bands that provide better discrimination were selected based on their neural network connection weights. Higher classification rates were achieved with fewer numbers of spectral bands. This selection scheme was compared with an information-theoretic approach and it demonstrated robust performance with higher classification accuracy.

  10. Airborne Radiative Transfer Spectral Scanner: A new airborne hyperspectral imager for hyperspectral volcano observations

    Science.gov (United States)

    Jitsufuchi, T.

    2007-12-01

    In 2006, a new airborne hyperspectral imager, the Airborne Radiative Transfer Spectral Scanner (ARTS), was developed for hyperspectral volcano observations. ARTS provides hyperspectral images to support developing algorithms for the remote sensing of the geothermal distribution, the ash fall areas, and the volcanic gasses columnar content from the air. ARTS will be used mainly to assess volcanic activity and to mitigate volcanic disasters. ARTS is a pushbroom imaging spectrometer covering wavelengths from 380 to 2450nm and 8000 to 11500nm with 421 bands. The ARTS imaging spectrometer consists of three sensor head units (SHUs). These SHUs are the visible - near infrared (VNIR) SHU, the shortwave infrared (SWIR) SHU, and the long-wave infrared (LWIR) SHU. These sensor head units operate as a line scanner in the pushbroom mode from an aircraft. The VNIR SHU covers wavelengths from 380 to 1050nm with 288 spectrum bands. The field of view (FOV) is 40 degrees, and the image of this SHU is 1500 pixels wide cross-track, making the instantaneous field of view (IFOV) 0.49mrad. The SWIR SHU covers wavelengths from 900 to 2450nm with 101 spectrum bands. The LWIR SHU covers wavelengths from 8000 to 11500nm with 32 spectrum bands. SWIR SHU and LWIR SHU have FOVs of 40 degrees and 600-pixel-wide images cross-track, giving them an IFOV of 1.2mrad. ARTS has precise position and attitude measurement systems (GPS/IMU). Direct, accurate geo-corrections of each SHU image can be made using the GPS/IMU systems. ARTS will be used for the operational volcano observation beginning in 2008. We are now validating the in-flight performance of this sensor. In this study, we describe the ARTS optical, electrical, and mechanical systems; its data acquisition and system design; and present some preliminary in- flight performance test results obtained from measurements acquired aboard the Beechcraft King Air B200 aircraft. The validation results indicate that the geo-correction accuracy is

  11. The use of airborne hyperspectral data for tree species classification in a species-rich Central European forest area

    Science.gov (United States)

    Richter, Ronny; Reu, Björn; Wirth, Christian; Doktor, Daniel; Vohland, Michael

    2016-10-01

    The success of remote sensing approaches to assess tree species diversity in a heterogeneously mixed forest stand depends on the availability of both appropriate data and suitable classification algorithms. To separate the high number of in total ten broadleaf tree species in a small structured floodplain forest, the Leipzig Riverside Forest, we introduce a majority based classification approach for Discriminant Analysis based on Partial Least Squares (PLS-DA), which was tested against Random Forest (RF) and Support Vector Machines (SVM). The classifier performance was tested on different sets of airborne hyperspectral image data (AISA DUAL) that were acquired on single dates in August and September and also stacked to a composite product. Shadowed gaps and shadowed crown parts were eliminated via spectral mixture analysis (SMA) prior to the pixel-based classification. Training and validation sets were defined spectrally with the conditioned Latin hypercube method as a stratified random sampling procedure. In the validation, PLS-DA consistently outperformed the RF and SVM approaches on all datasets. The additional use of spectral variable selection (CARS, "competitive adaptive reweighted sampling") combined with PLS-DA further improved classification accuracies. Up to 78.4% overall accuracy was achieved for the stacked dataset. The image recorded in August provided slightly higher accuracies than the September image, regardless of the applied classifier.

  12. Using hyperspectral imaging to determine germination of native Australian plant seeds.

    Science.gov (United States)

    Nansen, Christian; Zhao, Genpin; Dakin, Nicole; Zhao, Chunhui; Turner, Shane R

    2015-04-01

    We investigated the ability to accurately and non-destructively determine the germination of three native Australian tree species, Acacia cowleana Tate (Fabaceae), Banksia prionotes L.F. (Proteaceae), and Corymbia calophylla (Lindl.) K.D. Hill & L.A.S. Johnson (Myrtaceae) based on hyperspectral imaging data. While similar studies have been conducted on agricultural and horticultural seeds, we are unaware of any published studies involving reflectance-based assessments of the germination of tree seeds. Hyperspectral imaging data (110 narrow spectral bands from 423.6nm to 878.9nm) were acquired of individual seeds after 0, 1, 2, 5, 10, 20, 30, and 50days of standardized rapid ageing. At each time point, seeds were subjected to hyperspectral imaging to obtain reflectance profiles from individual seeds. A standard germination test was performed, and we predicted that loss of germination was associated with a significant change in seed coat reflectance profiles. Forward linear discriminant analysis (LDA) was used to select the 10 spectral bands with the highest contribution to classifications of the three species. In all species, germination decreased from over 90% to below 20% in about 10-30days of experimental ageing. P50 values (equal to 50% germination) for each species were 19.3 (A. cowleana), 7.0 (B. prionotes) and 22.9 (C. calophylla) days. Based on independent validation of classifications of hyperspectral imaging data, we found that germination of Acacia and Corymbia seeds could be classified with over 85% accuracy, while it was about 80% for Banksia seeds. The selected spectral bands in each LDA-based classification were located near known pigment peaks involved in photosynthesis and/or near spectral bands used in published indices to predict chlorophyll or nitrogen content in leaves. The results suggested that seed germination may be successfully classified (predicted) based on reflectance in narrow spectral bands associated with the primary metabolism

  13. Subsurface classification of objects under turbid waters by means of regularization techniques applied to real hyperspectral data

    Science.gov (United States)

    Carpena, Emmanuel; Jiménez, Luis O.; Arzuaga, Emmanuel; Fonseca, Sujeily; Reyes, Ernesto; Figueroa, Juan

    2017-05-01

    Improved benthic habitat mapping is needed to monitor coral reefs around the world and to assist coastal zones management programs. A fundamental challenge to remotely sensed mapping of coastal shallow waters is due to the significant disparity in the optical properties of the water column caused by the interaction between the coast and the sea. The objects to be classified have weak signals that interact with turbid waters that include sediments. In real scenarios, the absorption and backscattering coefficients are unknown with different sources of variability (river discharges and coastal interactions). Under normal circumstances, another unknown variable is the depth of shallow waters. This paper presents the development of algorithms for retrieving information and its application to the classification and mapping of objects under coastal shallow waters with different unknown concentrations of sediments. A mathematical model that simplifies the radiative transfer equation was used to quantify the interaction between the object of interest, the medium and the sensor. The retrieval of information requires the development of mathematical models and processing tools in the area of inversion, image reconstruction and classification of hyperspectral data. The algorithms developed were applied to one set of real hyperspectral imagery taken in a tank filled with water and TiO2 that emulates turbid coastal shallow waters. Tikhonov method of regularization was used in the inversion process to estimate the bottom albedo of the water tank using a priori information in the form of stored spectral signatures, previously measured, of objects of interest.

  14. Hyperspectral image analysis for CARS, SRS, and Raman data

    Science.gov (United States)

    Karuna, Arnica; Borri, Paola; Langbein, Wolfgang

    2015-01-01

    In this work, we have significantly enhanced the capabilities of the hyperspectral image analysis (HIA) first developed by Masia et al. 1 The HIA introduced a method to factorize the hyperspectral data into the product of component concentrations and spectra for quantitative analysis of the chemical composition of the sample. The enhancements shown here comprise (1) a spatial weighting to reduce the spatial variation of the spectral error, which improves the retrieval of the chemical components with significant local but small global concentrations; (2) a new selection criterion for the spectra used when applying sparse sampling2 to speed up sequential hyperspectral imaging; and (3) a filter for outliers in the data using singular value decomposition, suited e.g. to suppress motion artifacts. We demonstrate the enhancements on coherent anti‐Stokes Raman scattering, stimulated Raman scattering, and spontaneous Raman data. We provide the HIA software as executable for public use. © 2015 The Authors. Journal of Raman Spectroscopy published by John Wiley & Sons, Ltd. PMID:27478301

  15. Study on Development of Non-Destructive Measurement Technique for Viability of Lettuce Seed (Lactuca sativa L) Using Hyperspectral Reflectance Imaging

    Energy Technology Data Exchange (ETDEWEB)

    Ahn, Chi Kook; Cho, Byoung Kwan [College of Agriculture and Life Science, Chungnam National University, Daejeon (Korea, Republic of); Mo, Chang Yeon [National Acadamy of Agricultural Science, Daejeon (Korea, Republic of); Kim, Moon S. [Environmental Microbial and Food Safety Laboratory, Animal and Natural Resources Institute, Agricultural Research Service, United States Department of Agriculture, Washington (United States)

    2012-10-15

    In this study, the feasibility of hyperspectral reflectance imaging technique was investigated for the discrimination of viable and non-viable lettuce seeds. The spectral data of hyperspectral reflectance images with the spectral range between 750 nm and 1000 nm were used to develop PLS-DA model for the classification of viable and non-viable lettuce seeds. The discrimination accuracy of the calibration set was 81.6% and that of the test set was 81.2%. The image analysis method was developed to construct the discriminant images of non-viable seeds with the developed PLS-DA model. The discrimination accuracy obtained from the resultant image were 91%, which showed the feasibility of hyperspectral reflectance imaging technique for the mass discrimination of non-viable lettuce seeds from viable ones.

  16. Hyperspectral Imaging and Chemometric Modeling of Echinacea — A Novel Approach in the Quality Control of Herbal Medicines

    Directory of Open Access Journals (Sweden)

    Maxleene Sandasi

    2014-08-01

    Full Text Available Echinacea species are popularly included in various formulations to treat upper respiratory tract infections. These products are of commercial importance, with a collective sales figure of $132 million in 2009. Due to their close taxonomic alliance it is difficult to distinguish between the three Echinacea species and incidences of incorrectly labeled commercial products have been reported. The potential of hyperspectral imaging as a rapid quality control method for raw material and products containing Echinacea species was investigated. Hyperspectral images of root and leaf material of authentic Echinacea species (E. angustifolia, E. pallida and E. purpurea were acquired using a sisuChema shortwave infrared (SWIR hyperspectral pushbroom imaging system with a spectral range of 920–2514 nm. Principal component analysis (PCA plots showed a clear distinction between the root and leaf samples of the three Echinacea species and further differentiated the roots of different species. A classification model with a high coefficient of determination was constructed to predict the identity of the species included in commercial products. The majority of products (12 out of 20 were convincingly predicted as containing E. purpurea, E. angustifolia or both. The use of ultra performance liquid chromatography-mass spectrometry (UPLC-MS in the differentiation of the species presented a challenge due to chemical similarities between the solvent extracts. The results show that hyperspectral imaging is an objective and non-destructive quality control method for authenticating raw material.

  17. One-Dimensional Convolutional Neural Network Land-Cover Classification of Multi-Seasonal Hyperspectral Imagery in the San Francisco Bay Area, California

    Directory of Open Access Journals (Sweden)

    Daniel Guidici

    2017-06-01

    Full Text Available In this study, a 1-D Convolutional Neural Network (CNN architecture was developed, trained and utilized to classify single (summer and three seasons (spring, summer, fall of hyperspectral imagery over the San Francisco Bay Area, California for the year 2015. For comparison, the Random Forests (RF and Support Vector Machine (SVM classifiers were trained and tested with the same data. In order to support space-based hyperspectral applications, all analyses were performed with simulated Hyperspectral Infrared Imager (HyspIRI imagery. Three-season data improved classifier overall accuracy by 2.0% (SVM, 1.9% (CNN to 3.5% (RF over single-season data. The three-season CNN provided an overall classification accuracy of 89.9%, which was comparable to overall accuracy of 89.5% for SVM. Both three-season CNN and SVM outperformed RF by over 7% overall accuracy. Analysis and visualization of the inner products for the CNN provided insight to distinctive features within the spectral-temporal domain. A method for CNN kernel tuning was presented to assess the importance of learned features. We concluded that CNN is a promising candidate for hyperspectral remote sensing applications because of the high classification accuracy and interpretability of its inner products.

  18. A new deblurring morphological filter for hyperspectral images

    Science.gov (United States)

    Abdelkawy, Ezz Eldin F.; Mahmoud, Tarek A.; Hussein, Wesam M.

    2011-06-01

    Hyperspectral imaging becomes an important technique that increases the valuable information enclosed within the image. Spectral cube produced by this type of imaging introduces a new material signature known as "spectral signature". This signature is unique for each material as it depends on the molecular composition of the material surface. To produce the spectral cube, a spectrometer should be used in the imagery device to split the electromagnetic energy at different wavelengths before its projection on the imaging array. This spectrometer may be a dispersive element, such as prism and grating, or an electronically tuneable filter. Some of dispersive spectrometers, such as Fourier transform interferometer (FTIR) and image multi-spectral imaging (IMSS), are based on sliding the lenses, or mirrors, along the optical axis which may result in a slightly out-of-focus blurring. Blind deconvolution techniques have been successfully used to decrease this blurring but at the expense of edge sharpening which may be a problem in some applications such as target detection and recognition. In this paper, we introduce a new method to deblurr the hyperspectral images keeping edges as sharp as possible. This is done by firstly detecting the edges locations and then applying a class of morphological filtering. Motivated by the success of threshold decomposition, gradient-based operators are used to detect the locations of these edges followed by an adaptive morphological filter to sharpen these detected edges. Experimental results demonstrate that the performance of the proposed deblurring filter is superior to that of the blind deconvolution methods.

  19. Development of non-destructive sorting technique for viability of watermelon seed by using hyperspectral image processing

    Energy Technology Data Exchange (ETDEWEB)

    Bae, Hyun Jin; Seo, Young Wook; Lohumi, Santosh; Park, Eun Soo; Cho, Byoung Kwan [Biosystems Machinery Engineering, Chungnam National University, Daejeon (Korea, Republic of); Kim, Dae Yong [Logistics institude, CJ Korea Express, Seoul (Korea, Republic of)

    2016-02-15

    Seed viability is one of the most important parameters that is directly related with seed germination performance and seedling emergence. In this study, a hyperspectral imaging (HSI) system having a range of 1000 –2500 nm was used to classify viable watermelon seeds from nonviable seeds. In order to obtain nonviable watermelon seeds, a total of 96 seeds were artificially aged by immersing the seeds in hot water (25°C) for 15 days. Further, hyperspectral images for 192 seeds (96 normal and 96 aged) were acquired using the developed HSI system. A germination test was performed for all the 192 seeds in order to confirm their viability. Spectral data from the hyperspectral images of the seeds were extracted by selecting pixels from the region of interest. Each seed spectrum was averaged and preprocessed to develop a classification model of partial least square discriminant analysis (PLS-DA). The developed PLS-DA model showed a classification accuracy of 94.7% for the calibration set, and 84.2% for the validation set. The results demonstrate that the proposed technique can classify viable and nonviable watermelon seeds with a reasonable accuracy, and can be further converted into an online sorting system for rapid and nondestructive classification of watermelon seeds with regard to viability.

  20. Characterizing pigments with hyperspectral imaging variable false-color composites

    Science.gov (United States)

    Hayem-Ghez, Anita; Ravaud, Elisabeth; Boust, Clotilde; Bastian, Gilles; Menu, Michel; Brodie-Linder, Nancy

    2015-11-01

    Hyperspectral imaging has been used for pigment characterization on paintings for the last 10 years. It is a noninvasive technique, which mixes the power of spectrophotometry and that of imaging technologies. We have access to a visible and near-infrared hyperspectral camera, ranging from 400 to 1000 nm in 80-160 spectral bands. In order to treat the large amount of data that this imaging technique generates, one can use statistical tools such as principal component analysis (PCA). To conduct the characterization of pigments, researchers mostly use PCA, convex geometry algorithms and the comparison of resulting clusters to database spectra with a specific tolerance (like the Spectral Angle Mapper tool on the dedicated software ENVI). Our approach originates from false-color photography and aims at providing a simple tool to identify pigments thanks to imaging spectroscopy. It can be considered as a quick first analysis to see the principal pigments of a painting, before using a more complete multivariate statistical tool. We study pigment spectra, for each kind of hue (blue, green, red and yellow) to identify the wavelength maximizing spectral differences. The case of red pigments is most interesting because our methodology can discriminate the red pigments very well—even red lakes, which are always difficult to identify. As for the yellow and blue categories, it represents a good progress of IRFC photography for pigment discrimination. We apply our methodology to study the pigments on a painting by Eustache Le Sueur, a French painter of the seventeenth century. We compare the results to other noninvasive analysis like X-ray fluorescence and optical microscopy. Finally, we draw conclusions about the advantages and limits of the variable false-color image method using hyperspectral imaging.

  1. Nonlinear spectral unmixing of hyperspectral images using Gaussian processes

    CERN Document Server

    Altmann, Yoann; McLaughlin, Steve; Tourneret, Jean-Yves

    2012-01-01

    This paper presents an unsupervised algorithm for nonlinear unmixing of hyperspectral images. The proposed model assumes that the pixel reflectances result from a nonlinear function of the abundance vectors associated with the pure spectral components. We assume that the spectral signatures of the pure components and the nonlinear function are unknown. The first step of the proposed method consists of the Bayesian estimation of the abundance vectors for all the image pixels and the nonlinear function relating the abundance vectors to the observations. The endmembers are subsequently estimated using Gaussian process regression. The performance of the unmixing strategy is evaluated with simulations conducted on synthetic and real data.

  2. Standoff Hyperspectral Imaging of Explosives Residues Using Broadly Tunable External Cavity Quantum Cascade Laser Illumination

    Energy Technology Data Exchange (ETDEWEB)

    Bernacki, Bruce E.; Phillips, Mark C.

    2010-05-01

    We describe experimental results on the detection of explosives residues using active hyperspectral imaging by illumination of the target surface using an external cavity quantum cascade laser (ECQCL) and imaging using a room temperature microbolometer camera. The active hyperspectral imaging technique forms an image hypercube by recording one image for each tuning step of the ECQCL. The resulting hyperspectral image contains the full absorption spectrum produced by the illumination laser at each pixel in the image which can then be used to identify the explosive type and relative quantity using spectral identification approaches developed initially in the remote sensing community.

  3. Image enhancement based on in vivo hyperspectral gastroscopic images: a case study

    Science.gov (United States)

    Gu, Xiaozhou; Han, Zhimin; Yao, Liqing; Zhong, Yunshi; Shi, Qiang; Fu, Ye; Liu, Changsheng; Wang, Xiguang; Xie, Tianyu

    2016-10-01

    Hyperspectral imaging (HSI) has been recognized as a powerful tool for noninvasive disease detection in the gastrointestinal field. However, most of the studies on HSI in this field have involved ex vivo biopsies or resected tissues. We proposed an image enhancement method based on in vivo hyperspectral gastroscopic images. First, we developed a flexible gastroscopy system capable of obtaining in vivo hyperspectral images of different types of stomach disease mucosa. Then, depending on a specific object, an appropriate band selection algorithm based on dependence of information was employed to determine a subset of spectral bands that would yield useful spatial information. Finally, these bands were assigned to be the color components of an enhanced image of the object. A gastric ulcer case study demonstrated that our method yields higher color tone contrast, which enhanced the displays of the gastric ulcer regions, and that it will be valuable in clinical applications.

  4. Hyperspectral imaging system for disease scanning on banana plants

    Science.gov (United States)

    Ochoa, Daniel; Cevallos, Juan; Vargas, German; Criollo, Ronald; Romero, Dennis; Castro, Rodrigo; Bayona, Oswaldo

    2016-05-01

    Black Sigatoka (BS) is a banana plant disease caused by the fungus Mycosphaerella fijiensis. BS symptoms can be observed at late infection stages. By that time, BS has probably spread to other plants. In this paper, we present our current work on building an hyper-spectral (HS) imaging system aimed at in-vivo detection of BS pre-symptomatic responses in banana leaves. The proposed imaging system comprises a motorized stage, a high-sensitivity VIS-NIR camera and an optical spectrograph. To capture images of the banana leaf, the stage's speed and camera's frame rate must be computed to reduce motion blur and to obtain the same resolution along both spatial dimensions of the resulting HS cube. Our continuous leaf scanning approach allows imaging leaves of arbitrary length with minimum frame loss. Once the images are captured, a denoising step is performed to improve HS image quality and spectral profile extraction.

  5. Optimal Decision Fusion for Urban Land-Use/Land-Cover Classification Based on Adaptive Differential Evolution Using Hyperspectral and LiDAR Data

    Directory of Open Access Journals (Sweden)

    Yanfei Zhong

    2017-08-01

    Full Text Available Hyperspectral images and light detection and ranging (LiDAR data have, respectively, the high spectral resolution and accurate elevation information required for urban land-use/land-cover (LULC classification. To combine the respective advantages of hyperspectral and LiDAR data, this paper proposes an optimal decision fusion method based on adaptive differential evolution, namely ODF-ADE, for urban LULC classification. In the ODF-ADE framework the normalized difference vegetation index (NDVI, gray-level co-occurrence matrix (GLCM and digital surface model (DSM are extracted to form the feature map. The three different classifiers of the maximum likelihood classifier (MLC, support vector machine (SVM and multinomial logistic regression (MLR are used to classify the extracted features. To find the optimal weights for the different classification maps, weighted voting is used to obtain the classification result and the weights of each classification map are optimized by the differential evolution algorithm which uses a self-adaptive strategy to obtain the parameter adaptively. The final classification map is obtained after post-processing based on conditional random fields (CRF. The experimental results confirm that the proposed algorithm is very effective in urban LULC classification.

  6. On-Line Hyperspectral Transmittance Imaging for Internal Defect Detection of Pickling Cucumbers

    Science.gov (United States)

    Hyperspectral imaging technique under transmittance mode was investigated for detection of internal defect in pickling cucumbers such as carpel suture separation or hollow cucumbers caused by mechanical stress. A prototype of on-line hyperspectral transmittance imaging system was developed for real...

  7. Determination of germination quality of cucumber (Cucumis sativus) seed by LED-induced hyperspectral reflectance imaging

    Science.gov (United States)

    Purpose: We developed a viability evaluation method for cucumber (Cucumis sativus) seed using hyperspectral reflectance imaging. Methods: Reflectance spectra of cucumber seeds in the 400 to 1000 nm range were collected from hyperspectral reflectance images obtained using blue, green, and red LED ill...

  8. Hyperspectral microscope imaging methods to classify gram-positive and gram-negative foodborne pathogenic bacteria

    Science.gov (United States)

    An acousto-optic tunable filter-based hyperspectral microscope imaging method has potential for identification of foodborne pathogenic bacteria from microcolony rapidly with a single cell level. We have successfully developed the method to acquire quality hyperspectral microscopic images from variou...

  9. Recent Advances in Compressed Sensing: Discrete Uncertainty Principles and Fast Hyperspectral Imaging

    Science.gov (United States)

    2015-03-26

    Recent Advances in Compressed Sensing : Discrete Uncertainty Principles and Fast Hyperspectral Imaging THESIS MARCH 2015 Megan E. Lewis, Second...IN COMPRESSED SENSING : DISCRETE UNCERTAINTY PRINCIPLES AND FAST HYPERSPECTRAL IMAGING THESIS Presented to the Faculty Department of Mathematics and...MARCH 2015 DISTRIBUTION STATEMENT A APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED. AFIT-ENC–MS-15-M-002 RECENT ADVANCES IN COMPRESSED SENSING

  10. Hyperspectral imaging based on diffused laser light for prediction of astaxanthin coating concentration

    DEFF Research Database (Denmark)

    Ljungqvist, Martin Georg; Nielsen, Otto Højager Attermann; Frosch, Stina

    2014-01-01

    We present a study on predicting the concentration level of synthetic astaxanthin in fish feed pellet coating using multi- and hyperspectral image analysis. This was done in parallel using two different vision systems. A new instrument for hyperspectral imaging, the SuperK setup, using a super...

  11. Manifold learning based feature extraction for classification of hyperspectral data

    CSIR Research Space (South Africa)

    Lunga, D

    2014-01-01

    Full Text Available Interest in manifold learning for representing the topology of large, high dimensional nonlinear data sets in lower, but still meaningful dimensions for visualization and classification has grown rapidly over the past decade, and particularly...

  12. Calibratio performance evaluation of the spaceborne hyperspectral CHRIS imager

    Institute of Scientific and Technical Information of China (English)

    ZHANG; Xia; ZHANG; Bing; HU; Fangchao; TONG; Qingxi

    2006-01-01

    The CHRIS (Compact High Resolution Imaging Spectrometer) is a new imaging spectrometer, carried on board a new space platform called PROBA (Project for On Board Autonomy). The satellite was successfully launched in October 2001 by the European Space Agency (ESA). CHRIS operates over the visible/near infrared band (400-1050 nm). It has five work modes and can reach a maximum of 62 spectral bands. In this research, atmospheric correction based on hyperspectral images was performed on CHRIS images by using the popular radiance transfer code ACORN (Atmospheric Correction Now) and empirical algorithms. ACORN was also used to evaluate the calibration performance of CHRIS by the retrieved spectra of typical vegetation and soil. As a result,the maize reflectance spectrum corrected by ACORN could characterize vegetation reflectance in the range of 498-750 nm, but gave a fairly large deviation beyond 750 nm,showing the deficiency of spectral calibration beyond 750 nm. The ACORN-derived soil reflectance decreased beyond 800 nm, which was quite inconsistent with field-spectrummeasurement and showed that the calibration accuracy couldn't meet the requirements of ACORN for spectral and radiometric calibration within a certain spectral range. In addition,the stripes on the retrieved water-vapor content map indicated that the radiance-calibration performance of CHRIS is not perfect. As the first spaceborne hyperspectral imager of ESA, the calibration performance of CHRIS needs to be improved.

  13. Application of hyperspectral imaging for characterization of intramuscular fat distribution in beef

    Science.gov (United States)

    Lohumi, Santosh; Lee, Sangdae; Lee, Hoonsoo; Kim, Moon S.; Lee, Wang-Hee; Cho, Byoung-Kwan

    2016-01-01

    In this study, a hyperspectral imaging system in the spectral region of 400-1000 nm was used for visualization and determination of intramuscular fat concentration in beef samples. Hyperspectral images were acquired for beef samples, and spectral information was then extracted from each single sample from the fat and non-fat regions. The intramuscular fat content was chemically extracted and quantified for the same samples. Chemometrics including analysis of variance (ANOVA) and spectral similarity measures involving spectral angle measure (SAM), and Euclidian distance measure (EDM) were then used to analyze the data. An ANOVA analysis indicates that the two selected spectral variables (e.g., 650.4-736.4 nm) are effective to generate ratio image for visualization of the intramuscular fat distribution in beef. The spectral similarity analysis methods, which is based on the quantifying the spectral similarities by using predetermined endmember spectrum vector, provided comparable results for characterization and detection of intramuscular fat in beef. In term of overall classification accuracy, spectral similarity measure methods outperformed the ratio image of selected bands based on the result of ANOVA analysis. The results demonstrate that proposed technique has a potential for fast and nondestructive determination of intramuscular fat in beef.

  14. Optimization and applications of an excitation-scanning hyperspectral imaging system

    Science.gov (United States)

    Mayes, Sam A.; Klomkaew, Phiwat; Leavesley, Silas J.; Rich, Thomas C.

    2017-02-01

    Currently, the majority of microscopic and endoscopic technologies utilize white light illumination. For a number of applications, hyper-spectral imaging can be shown to have significant improvements over standard white-light imaging techniques. This is true for both microscopy and in vivo imaging. However, hyperspectral imaging methods have suffered from slow application times. Often, minutes are required to gather a full imaging stack. Here we will describe the system and evaluate optimizations and applications of a novel excitation-scanning hyperspectral imaging system. We have developed and are optimizing a novel approach called excitation-scanning hyperspectral imaging that provides an order of magnitude increased signal strength. Optimization of the light path, optical components and illumination sources have allowed us to achieve high speed image acquisition. This high speed allows for potential live video acquisition. This excitation-scanning hyperspectral imaging technology has potential to impact a range of applications. The current system allows triggering of up to 16 wavelengths at less than 1 millisecond per image using digital strobing. Analog intensity control is also provided for a fully customizable excitation profile. A significant advantage of excitation scanning hyperspectral imaging is can identify multiple targets simultaneously in real time. We are optimizing the system to compare sensitivity and specificity of excitation-scanning hyperspectral imaging with pathology techniques. Finally, we are exploring utilizing this technology to measure cAMP distribution in three dimensions within a cell.

  15. [The linear hyperspectral camera rotating scan imaging geometric correction based on the precise spectral sampling].

    Science.gov (United States)

    Wang, Shu-min; Zhang, Ai-wu; Hu, Shao-xing; Wang, Jing-meng; Meng, Xian-gang; Duan, Yi-hao; Sun, Wei-dong

    2015-02-01

    As the rotation speed of ground based hyperspectral imaging system is too fast in the image collection process, which exceeds the speed limitation, there is data missed in the rectified image, it shows as the_black lines. At the same time, there is serious distortion in the collected raw images, which effects the feature information classification and identification. To solve these problems, in this paper, we introduce the each component of the ground based hyperspectral imaging system at first, and give the general process of data collection. The rotation speed is controlled in data collection process, according to the image cover area of each frame and the image collection speed of the ground based hyperspectral imaging system, And then the spatial orientation model is deduced in detail combining with the star scanning angle, stop scanning angle and the minimum distance between the sensor and the scanned object etc. The oriented image is divided into grids and resampled with new spectral. The general flow of distortion image corrected is presented in this paper. Since the image spatial resolution is different between the adjacent frames, and in order to keep the highest image resolution of corrected image, the minimum ground sampling distance is employed as the grid unit to divide the geo-referenced image. Taking the spectral distortion into account caused by direct sampling method when the new uniform grids and the old uneven grids are superimposed to take the pixel value, the precise spectral sampling method based on the position distribution is proposed. The distortion image collected in Lao Si Cheng ruin which is in the Zhang Jiajie town Hunan province is corrected through the algorithm proposed on above. The features keep the original geometric characteristics. It verifies the validity of the algorithm. And we extract the spectral of different features to compute the correlation coefficient. The results show that the improved spectral sampling method is

  16. Parallel optimization of pixel purity index algorithm for massive hyperspectral images in cloud computing environment

    Science.gov (United States)

    Chen, Yufeng; Wu, Zebin; Sun, Le; Wei, Zhihui; Li, Yonglong

    2016-04-01

    With the gradual increase in the spatial and spectral resolution of hyperspectral images, the size of image data becomes larger and larger, and the complexity of processing algorithms is growing, which poses a big challenge to efficient massive hyperspectral image processing. Cloud computing technologies distribute computing tasks to a large number of computing resources for handling large data sets without the limitation of memory and computing resource of a single machine. This paper proposes a parallel pixel purity index (PPI) algorithm for unmixing massive hyperspectral images based on a MapReduce programming model for the first time in the literature. According to the characteristics of hyperspectral images, we describe the design principle of the algorithm, illustrate the main cloud unmixing processes of PPI, and analyze the time complexity of serial and parallel algorithms. Experimental results demonstrate that the parallel implementation of the PPI algorithm on the cloud can effectively process big hyperspectral data and accelerate the algorithm.

  17. 多线性局部与全局保持嵌入在高光谱遥感影像分类中的应用%Multi-linear Local and Global Preserving Embedding and its Application in Hyperspectral Remote Sensing Image Classification

    Institute of Scientific and Technical Information of China (English)

    王立志; 黄鸿; 冯海亮

    2012-01-01

    Traditional manifold learning methods assume that hyperspectral data may reside on one single manifold, but data from different classes may reside on different manifolds of possible different intrinsic dimensions. In order to explore multiple low-dimensional manifolds in hyperspectral images, a multi-manifold learning algorithm based on local and global preserving embedding ( LLGPE) is proposed. First, the manifolds of different classes are learned by LLGPE for each class separately, and the data are projected onto low-dimensional spaces. Then, the optimal dimensionality of each class is founded by genetic algorithm (GA) from the viewpoint of classification. At last, classification is performed under a minimum reconstruction error based classifier. The experimental results on the HYDICE hyperspectral data show the effectiveness of the proposed algorithm, when 2, 4 and 6 samples of each class are randomly selected for training and 90 samples of each class for testing, the overall accuracy of the proposed algorithm is improved by 3. 5%, 6. 9% and 7. 2% respectively, as compared with other methods.%针对一般流形学习算法在学习高光谱数据的多流形结构时存在的不足,提出一种基于线性局部与全局保持嵌入(LLGPE)的多流形学习算法.对于分布在不同流形上的高维观测数据,利用LLGPE算法学习每类分组数据的内蕴特征;然后通过遗传算法搜索每类数据的本质维数;最后根据重构误差最小化准则确定样本所属的类别.在HYDICE高光谱数据集上的分类识别实验结果表明,文中算法能够有效地揭示高维空间中数据的内蕴几何结构;在每类随机选取2,4,6个训练样本的情况下,该算法的总体分类精度比其他流形学习算法分别提高了约3.5%,6.9%和7.2%,且分类精度也有明显的提高.

  18. NIR hyperspectral imaging to evaluate degradation in captopril commercial tablets.

    Science.gov (United States)

    França, Leandro de Moura; Pimentel, Maria Fernanda; Simões, Simone da Silva; Grangeiro, Severino; Prats-Montalbán, José M; Ferrer, Alberto

    2016-07-01

    Pharmaceutical quality control is important for improving the effectiveness, purity and safety of drugs, as well as for the prevention or control of drug degradation. In the present work, near infrared hyperspectral images (HSI-NIR) of tablets with different expiration dates were employed to evaluate the degradation of captopril into captopril disulfide in different layers, on the top and on the bottom surfaces of the tablets. Multivariate curve resolution (MCR) models were used to extract the concentration distribution maps from the hyperspectral images. Afterward, multivariate image techniques were applied to the concentration distribution maps (CDMs), to extract features and build models relating the main characteristics of the images to their corresponding manufacturing dates. Resolution methods followed by extracting features were able to estimate the tablet manufacture date with a prediction error of 120days. The model developed could be useful to evaluate whether a sample shows a degradation pattern consistent with the date of manufacturing or to detect abnormal behaviors in the natural degradation process of the sample. The information provided by the HIS-NIR is important for the development of the process (QbD), looking inside the formulation, revealing the behavior of the active pharmaceutical ingredient (API) during the product's shelf life. Copyright © 2016 Elsevier B.V. All rights reserved.

  19. Raman hyperspectral imaging of iron transport across membranes in cells

    Science.gov (United States)

    Das, Anupam; Costa, Xavier Felipe; Khmaladze, Alexander; Barroso, Margarida; Sharikova, Anna

    2016-09-01

    Raman scattering microscopy is a powerful imaging technique used to identify chemical composition, structural and conformational state of molecules of complex samples in biology, biophysics, medicine and materials science. In this work, we have shown that Raman techniques allow the measurement of the iron content in protein mixtures and cells. Since the mechanisms of iron acquisition, storage, and excretion by cells are not completely understood, improved knowledge of iron metabolism can offer insight into many diseases in which iron plays a role in the pathogenic process, such as diabetes, neurodegenerative diseases, cancer, and metabolic syndrome. Understanding of the processes involved in cellular iron metabolism will improve our knowledge of cell functioning. It will also have a big impact on treatment of diseases caused by iron deficiency (anemias) and iron overload (hereditary hemochromatosis). Previously, Raman studies have shown substantial differences in spectra of transferrin with and without bound iron, thus proving that it is an appropriate technique to determine the levels of bound iron in the protein mixture. We have extended these studies to obtain hyperspectral images of transferrin in cells. By employing a Raman scanning microscope together with spectral detection by a highly sensitive back-illuminated cooled CCD camera, we were able to rapidly acquire and process images of fixed cells with chemical selectivity. We discuss and compare various methods of hyperspectral Raman image analysis and demonstrate the use of these methods to characterize cellular iron content without the need for dye labeling.

  20. Automatic Hierarchical Color Image Classification

    Directory of Open Access Journals (Sweden)

    Jing Huang

    2003-02-01

    Full Text Available Organizing images into semantic categories can be extremely useful for content-based image retrieval and image annotation. Grouping images into semantic classes is a difficult problem, however. Image classification attempts to solve this hard problem by using low-level image features. In this paper, we propose a method for hierarchical classification of images via supervised learning. This scheme relies on using a good low-level feature and subsequently performing feature-space reconfiguration using singular value decomposition to reduce noise and dimensionality. We use the training data to obtain a hierarchical classification tree that can be used to categorize new images. Our experimental results suggest that this scheme not only performs better than standard nearest-neighbor techniques, but also has both storage and computational advantages.

  1. Fluorescent marker-based and marker-free discrimination between healthy and cancerous human tissues using hyper-spectral imaging

    Science.gov (United States)

    Arnold, Thomas; De Biasio, Martin; Leitner, Raimund

    2015-06-01

    Two problems are addressed in this paper (i) the fluorescent marker-based and the (ii) marker-free discrimination between healthy and cancerous human tissues. For both applications the performance of hyper-spectral methods are quantified. Fluorescent marker-based tissue classification uses a number of fluorescent markers to dye specific parts of a human cell. The challenge is that the emission spectra of the fluorescent dyes overlap considerably. They are, furthermore disturbed by the inherent auto-fluorescence of human tissue. This results in ambiguities and decreased image contrast causing difficulties for the treatment decision. The higher spectral resolution introduced by tunable-filter-based spectral imaging in combination with spectral unmixing techniques results in an improvement of the image contrast and therefore more reliable information for the physician to choose the treatment decision. Marker-free tissue classification is based solely on the subtle spectral features of human tissue without the use of artificial markers. The challenge in this case is that the spectral differences between healthy and cancerous tissues are subtle and embedded in intra- and inter-patient variations of these features. The contributions of this paper are (i) the evaluation of hyper-spectral imaging in combination with spectral unmixing techniques for fluorescence marker-based tissue classification, (ii) the evaluation of spectral imaging for marker-free intra surgery tissue classification. Within this paper, we consider real hyper-spectral fluorescence and endoscopy data sets to emphasize the practical capability of the proposed methods. It is shown that the combination of spectral imaging with multivariate statistical methods can improve the sensitivity and specificity of the detection and the staging of cancerous tissues compared to standard procedures.

  2. GPU implementation issues for fast unmixing of hyperspectral images

    Science.gov (United States)

    Legendre, Maxime; Capriotti, Luca; Schmidt, Frédéric; Moussaoui, Saïd; Schmidt, Albrecht

    2013-04-01

    Space missions usually use hyperspectral imaging techniques to analyse the composition of planetary surfaces. Missions such as ESA's Mars Express and Venus Express generate extensive datasets whose processing demands so far have exceeded the resources available to many researchers. To overcome this limitation, the challenge is to develop numerical methods allowing to exploit the potential of modern calculation tools. The processing of a hyperspectral image consists of the identification of the observed surface components and eventually the assessment of their fractional abundances inside each pixel area. In this latter case, the problem is referred to as spectral unmixing. This work focuses on a supervised unmixing approach where the relevant component spectra are supposed to be part of an available spectral library. Therefore, the question addressed here is reduced to the estimation of the fractional abundances, or abundance maps. It requires the solution of a large-scale optimization problem subject to linear constraints; positivity of the abundances and their partial/full additivity (sum less/equal to one). Conventional approaches to such a problem usually suffer from a high computational overhead. Recently, an interior-point optimization using a primal-dual approach has been proven an efficient method to solve this spectral unmixing problem at reduced computational cost. This is achieved with a parallel implementation based on Graphics Processing Units (GPUs). Several issues are discussed such as the data organization in memory and the strategy used to compute efficiently one global quantity from a large dataset in a parallel fashion. Every step of the algorithm is optimized to be GPU-efficient. Finally, the main steps of the global system for the processing of a large number of hyperspectral images are discussed. The advantage of using a GPU is demonstrated by unmixing a large dataset consisting of 1300 hyperspectral images from Mars Express' OMEGA instrument

  3. Kernel based orthogonalization for change detection in hyperspectral images

    DEFF Research Database (Denmark)

    Nielsen, Allan Aasbjerg

    Kernel versions of principal component analysis (PCA) and minimum noise fraction (MNF) analysis are applied to change detection in hyperspectral image (HyMap) data. The kernel versions are based on so-called Q-mode analysis in which the data enter into the analysis via inner products in the Gram...... the kernel function and then performing a linear analysis in that space. An example shows the successful application of (kernel PCA and) kernel MNF analysis to change detection in HyMap data covering a small agricultural area near Lake Waging-Taching, Bavaria, in Southern Germany. In the change detection...

  4. Full-frame, programmable hyperspectral imager

    Energy Technology Data Exchange (ETDEWEB)

    Love, Steven P.; Graff, David L.

    2017-07-25

    A programmable, many-band spectral imager based on addressable spatial light modulators (ASLMs), such as micro-mirror-, micro-shutter- or liquid-crystal arrays, is described. Capable of collecting at once, without scanning, a complete two-dimensional spatial image with ASLM spectral processing applied simultaneously to the entire image, the invention employs optical assemblies wherein light from all image points is forced to impinge at the same angle onto the dispersing element, eliminating interplay between spatial position and wavelength. This is achieved, as examples, using telecentric optics to image light at the required constant angle, or with micro-optical array structures, such as micro-lens- or capillary arrays, that aim the light on a pixel-by-pixel basis. Light of a given wavelength then emerges from the disperser at the same angle for all image points, is collected at a unique location for simultaneous manipulation by the ASLM, then recombined with other wavelengths to form a final spectrally-processed image.

  5. Jeffries Matusita based mixed-measure for improved spectral matching in hyperspectral image analysis

    Science.gov (United States)

    Padma, S.; Sanjeevi, S.

    2014-10-01

    This paper proposes a novel hyperspectral matching technique by integrating the Jeffries-Matusita measure (JM) and the Spectral Angle Mapper (SAM) algorithm. The deterministic Spectral Angle Mapper and stochastic Jeffries-Matusita measure are orthogonally projected using the sine and tangent functions to increase their spectral ability. The developed JM-SAM algorithm is implemented in effectively discriminating the landcover classes and cover types in the hyperspectral images acquired by PROBA/CHRIS and EO-1 Hyperion sensors. The reference spectra for different land-cover classes were derived from each of these images. The performance of the proposed measure is compared with the performance of the individual SAM and JM approaches. From the values of the relative spectral discriminatory probability (RSDPB) and relative discriminatory entropy value (RSDE), it is inferred that the hybrid JM-SAM approach results in a high spectral discriminability than the SAM and JM measures. Besides, the use of the improved JM-SAM algorithm for supervised classification of the images results in 92.9% and 91.47% accuracy compared to 73.13%, 79.41%, and 85.69% of minimum-distance, SAM and JM measures. It is also inferred that the increased spectral discriminability of JM-SAM measure is contributed by the JM distance. Further, it is seen that the proposed JM-SAM measure is compatible with varying spectral resolutions of PROBA/CHRIS (62 bands) and Hyperion (242 bands).

  6. Melanoma detection using smartphone and multimode hyperspectral imaging

    Science.gov (United States)

    MacKinnon, Nicholas; Vasefi, Fartash; Booth, Nicholas; Farkas, Daniel L.

    2016-04-01

    This project's goal is to determine how to effectively implement a technology continuum from a low cost, remotely deployable imaging device to a more sophisticated multimode imaging system within a standard clinical practice. In this work a smartphone is used in conjunction with an optical attachment to capture cross-polarized and collinear color images of a nevus that are analyzed to quantify chromophore distribution. The nevus is also imaged by a multimode hyperspectral system, our proprietary SkinSpect™ device. Relative accuracy and biological plausibility of the two systems algorithms are compared to assess aspects of feasibility of in-home or primary care practitioner smartphone screening prior to rigorous clinical analysis via the SkinSpect.

  7. Hyperspectral image analysis for the determination of alteration minerals in geothermal fields: Çürüksu (Denizli) Graben, Turkey

    Science.gov (United States)

    Uygur, Merve; Karaman, Muhittin; Kumral, Mustafa

    2016-04-01

    Çürüksu (Denizli) Graben hosts various geothermal fields such as Kızıldere, Yenice, Gerali, Karahayıt, and Tekkehamam. Neotectonic activities, which are caused by extensional tectonism, and deep circulation in sub-volcanic intrusions are heat sources of hydrothermal solutions. The temperature of hydrothermal solutions is between 53 and 260 degree Celsius. Phyllic, argillic, silicic, and carbonatization alterations and various hydrothermal minerals have been identified in various research studies of these areas. Surfaced hydrothermal alteration minerals are one set of potential indicators of geothermal resources. Developing the exploration tools to define the surface indicators of geothermal fields can assist in the recognition of geothermal resources. Thermal and hyperspectral imaging and analysis can be used for defining the surface indicators of geothermal fields. This study tests the hypothesis that hyperspectral image analysis based on EO-1 Hyperion images can be used for the delineation and definition of surfaced hydrothermal alteration in geothermal fields. Hyperspectral image analyses were applied to images covering the geothermal fields whose alteration characteristic are known. To reduce data dimensionality and identify spectral endmembers, Kruse's multi-step process was applied to atmospherically and geometrically-corrected hyperspectral images. Minimum Noise Fraction was used to reduce the spectral dimensions and isolate noise in the images. Extreme pixels were identified from high order MNF bands using the Pixel Purity Index. n-Dimensional Visualization was utilized for unique pixel identification. Spectral similarities between pixel spectral signatures and known endmember spectrum (USGS Spectral Library) were compared with Spectral Angle Mapper Classification. EO-1 Hyperion hyperspectral images and hyperspectral analysis are sensitive to hydrothermal alteration minerals, as their diagnostic spectral signatures span the visible and shortwave

  8. Using Hyperspectral Frame Images from Unmanned Airborne Vehicle for Detailed Measurement of Boreal Forest 3D Structure

    Science.gov (United States)

    de Oliveira, Raquel A.; Tommaselli, Antonio M. G.; Honkavaara, Eija

    2016-10-01

    Objective of this work was to investigate the feasibility of using multi-image matching and information extracted from image classification to improve strategies in generation of point clouds of 3D forest scene. Image data sets were collected by a Fabry-Pérot interferometer (FPI) based hyperspectral frame camera on-board a UAV in a boreal forest area. The results of the new method are analysed and compared with commercial software and LiDAR data. Experiments showed that the point clouds generated with the proposed algorithm fitted better with the LiDAR data at the ground level, which is favourable for digital terrain model (DTM) extraction.

  9. Biogeography based Satellite Image Classification

    CERN Document Server

    Panchal, V K; Kaur, Navdeep; Kundra, Harish

    2009-01-01

    Biogeography is the study of the geographical distribution of biological organisms. The mindset of the engineer is that we can learn from nature. Biogeography Based Optimization is a burgeoning nature inspired technique to find the optimal solution of the problem. Satellite image classification is an important task because it is the only way we can know about the land cover map of inaccessible areas. Though satellite images have been classified in past by using various techniques, the researchers are always finding alternative strategies for satellite image classification so that they may be prepared to select the most appropriate technique for the feature extraction task in hand. This paper is focused on classification of the satellite image of a particular land cover using the theory of Biogeography based Optimization. The original BBO algorithm does not have the inbuilt property of clustering which is required during image classification. Hence modifications have been proposed to the original algorithm and...

  10. Hyperspectral Image Projector with Polarization Capability

    Data.gov (United States)

    National Aeronautics and Space Administration — The goal of this proposal is improve testing and calibration of imaging sensors used on remote sensing platforms through the development of calibrated scene...

  11. Hyperspectral Single Pixel Image Sensor Project

    Data.gov (United States)

    National Aeronautics and Space Administration — This is a drastic enhancement to the current prototype which only allows us to collect visible light and reconstruct a single wavelength image. This approach is a...

  12. Advanced Airborne Hyperspectral Imaging System (AAHIS): an imaging spectrometer for maritime applications

    Science.gov (United States)

    Voelker, Mark A.; Resmini, Ronald G.; Mooradian, Gregory C.; McCord, Thomas B.; Warren, Christopher P.; Fene, Michael W.; Coyle, Christopher C.; Anderson, Richard

    1995-06-01

    The Advanced Airborne Hyperspectral Imaging System (AAHIS) is a compact, lightweight visible and near IR pushbroom hyperspectral imaging spectrometer flown on a Piper Aztec aircraft. AAHIS is optimized for use in shallow water, littoral, and vegetation remote sensing. Data are collected at up to 55 frames/second and may be displayed and analyzed inflight or recorded for post-flight processing. Swath width is 200 meters at a flight altitude of 1 km. Each image pixel contains hyperspectral data simultaneously recorded in up to 288 contiguous spectral channels covering the 432 to 832 nm spectral region. Pixel binning typically yields pixels 1.0 meter square with a spectral channel width of 5.5 nm. Design and performance of the AAHIS is presented, including processed imagery demonstrating feature detection and materials discrimination on land and underwater at depths up to 27 meters.

  13. a Spatio-Spectral Camera for High Resolution Hyperspectral Imaging

    Science.gov (United States)

    Livens, S.; Pauly, K.; Baeck, P.; Blommaert, J.; Nuyts, D.; Zender, J.; Delauré, B.

    2017-08-01

    Imaging with a conventional frame camera from a moving remotely piloted aircraft system (RPAS) is by design very inefficient. Less than 1 % of the flying time is used for collecting light. This unused potential can be utilized by an innovative imaging concept, the spatio-spectral camera. The core of the camera is a frame sensor with a large number of hyperspectral filters arranged on the sensor in stepwise lines. It combines the advantages of frame cameras with those of pushbroom cameras. By acquiring images in rapid succession, such a camera can collect detailed hyperspectral information, while retaining the high spatial resolution offered by the sensor. We have developed two versions of a spatio-spectral camera and used them in a variety of conditions. In this paper, we present a summary of three missions with the in-house developed COSI prototype camera (600-900 nm) in the domains of precision agriculture (fungus infection monitoring in experimental wheat plots), horticulture (crop status monitoring to evaluate irrigation management in strawberry fields) and geology (meteorite detection on a grassland field). Additionally, we describe the characteristics of the 2nd generation, commercially available ButterflEYE camera offering extended spectral range (475-925 nm), and we discuss future work.

  14. Calibration methodology and performance characterization of a polarimetric hyperspectral imager

    Science.gov (United States)

    Holder, Joel G.; Martin, Jacob A.; Pitz, Jeremey; Pezzaniti, Joseph L.; Gross, Kevin C.

    2014-05-01

    Polarimetric hyperspectral imaging (P-HSI) has the potential to improve target detection, material identification, and background characterization over conventional hyperspectral imaging and polarimetric imaging. To fully exploit the spectro-polarimetric signatures captured by such an instrument, a careful calibration process is required to remove the spectrally- and polarimetrically-dependent system response (gain). Calibration of instruments operating in the long-wave infrared (LWIR, 8μm to 12 μm) is further complicated by the polarized spectral radiation generated within the instrument (offset). This paper presents a calibration methodology developed for a LWIR Telops Hyper-Cam modified for polarimetry by replacing the entrance window with a rotatable holographic wire-grid polarizer (4000 line/mm, ZnSe substrate, 350:1 extinction ratio). A standard Fourier-transform spectrometer (FTS) spectro-radiometric calibration is modified to include a Mueller-matrix approach to account for polarized transmission through and polarized selfemission from each optical interface. It is demonstrated that under the ideal polarizer assumption, two distinct blackbody measurements at polarizer angles of 0°, 45°, 90°, and 135° are sufficient to calibrate the system for apparent degree-of-linear-polarization (DoLP) measurements. Noise-equivalent s1, s2, and DoLP are quantified using a wide-area blackbody. A polarization-state generator is used to determine the Mueller deviation matrix. Finally, a realistic scene involving buildings, cars, sky radiance, and natural vegetation is presented.

  15. NIR DLP hyperspectral imaging system for medical applications

    Science.gov (United States)

    Wehner, Eleanor; Thapa, Abhas; Livingston, Edward; Zuzak, Karel

    2011-03-01

    DLP® hyperspectral reflectance imaging in the visible range has been previously shown to quantify hemoglobin oxygenation in subsurface tissues, 1 mm to 2 mm deep. Extending the spectral range into the near infrared reflects biochemical information from deeper subsurface tissues. Unlike any other illumination method, the digital micro-mirror device, DMD, chip is programmable, allowing the user to actively illuminate with precisely predetermined spectra of illumination with a minimum bandpass of approximately 10 nm. It is possible to construct active spectral-based illumination that includes but is not limited to containing sharp cutoffs to act as filters or forming complex spectra, varying the intensity of light at discrete wavelengths. We have characterized and tested a pure NIR, 760 nm to 1600 nm, DLP hyperspectral reflectance imaging system. In its simplest application, the NIR system can be used to quantify the percentage of water in a subject, enabling edema visualization. It can also be used to map vein structure in a patient in real time. During gall bladder surgery, this system could be invaluable in imaging bile through fatty tissue, aiding surgeons in locating the common bile duct in real time without injecting any contrast agents.

  16. Detecting pits in tart cherries by hyperspectral transmission imaging

    Science.gov (United States)

    Qin, Jianwei; Lu, Renfu

    2004-11-01

    The presence of pits in processed cherry products causes safety concerns for consumers and imposes potential liability for the food industry. The objective of this research was to investigate a hyperspectral transmission imaging technique for detecting the pit in tart cherries. A hyperspectral imaging system was used to acquire transmission images from individual cherry fruit for four orientations before and after pits were removed over the spectral region between 450 nm and 1,000 nm. Cherries of three size groups (small, intermediate, and large), each with two color classes (light red and dark red) were used for determining the effect of fruit orientation, size, and color on the pit detection accuracy. Additional cherries were studied for the effect of defect (i.e., bruises) on the pit detection. Computer algorithms were developed using the neural network (NN) method to classify the cherries with and without the pit. Two types of data inputs, i.e., single spectra and selected regions of interest (ROIs), were compared. The spectral region between 690 nm and 850 nm was most appropriate for cherry pit detection. The NN with inputs of ROIs achieved higher pit detection rates ranging from 90.6% to 100%, with the average correct rate of 98.4%. Fruit orientation and color had a small effect (less than 1%) on pit detection. Fruit size and defect affected pit detection and their effect could be minimized by training the NN with properly selected cherry samples.

  17. A novel highly parallel algorithm for linearly unmixing hyperspectral images

    Science.gov (United States)

    Guerra, Raúl; López, Sebastián.; Callico, Gustavo M.; López, Jose F.; Sarmiento, Roberto

    2014-10-01

    Endmember extraction and abundances calculation represent critical steps within the process of linearly unmixing a given hyperspectral image because of two main reasons. The first one is due to the need of computing a set of accurate endmembers in order to further obtain confident abundance maps. The second one refers to the huge amount of operations involved in these time-consuming processes. This work proposes an algorithm to estimate the endmembers of a hyperspectral image under analysis and its abundances at the same time. The main advantage of this algorithm is its high parallelization degree and the mathematical simplicity of the operations implemented. This algorithm estimates the endmembers as virtual pixels. In particular, the proposed algorithm performs the descent gradient method to iteratively refine the endmembers and the abundances, reducing the mean square error, according with the linear unmixing model. Some mathematical restrictions must be added so the method converges in a unique and realistic solution. According with the algorithm nature, these restrictions can be easily implemented. The results obtained with synthetic images demonstrate the well behavior of the algorithm proposed. Moreover, the results obtained with the well-known Cuprite dataset also corroborate the benefits of our proposal.

  18. Platforms for hyperspectral imaging, in-situ optical and acoustical imaging in urbanized regions

    Science.gov (United States)

    Bostater, Charles R.; Oney, Taylor

    2016-10-01

    Hyperspectral measurements of the water surface of urban coastal waters are presented. Oblique bidirectional reflectance factor imagery was acquired made in a turbid coastal sub estuary of the Indian River Lagoon, Florida and along coastal surf zone waters of the nearby Atlantic Ocean. Imagery was also collected using a pushbroom hyperspectral imager mounted on a fixed platform with a calibrated circular mechatronic rotation stage. Oblique imagery of the shoreline and subsurface features clearly shows subsurface bottom features and rip current features within the surf zone water column. In-situ hyperspectral optical signatures were acquired from a vessel as a function of depth to determine the attenuation spectrum in Palm Bay. A unique stationary platform methodology to acquire subsurface acoustic images showing the presence of moving bottom boundary nephelometric layers passing through the acoustic fan beam. The acoustic fan beam imagery indicated the presence of oscillatory subsurface waves in the urbanized coastal estuary. Hyperspectral imaging using the fixed platform techniques are being used to collect hyperspectral bidirectional reflectance factor (BRF) measurements from locations at buildings and bridges in order to provide new opportunities to advance our scientific understanding of aquatic environments in urbanized regions.

  19. Spatial-Spectral Classification Based on the Unsupervised Convolutional Sparse Auto-Encoder for Hyperspectral Remote Sensing Imagery

    Science.gov (United States)

    Han, Xiaobing; Zhong, Yanfei; Zhang, Liangpei

    2016-06-01

    Current hyperspectral remote sensing imagery spatial-spectral classification methods mainly consider concatenating the spectral information vectors and spatial information vectors together. However, the combined spatial-spectral information vectors may cause information loss and concatenation deficiency for the classification task. To efficiently represent the spatial-spectral feature information around the central pixel within a neighbourhood window, the unsupervised convolutional sparse auto-encoder (UCSAE) with window-in-window selection strategy is proposed in this paper. Window-in-window selection strategy selects the sub-window spatial-spectral information for the spatial-spectral feature learning and extraction with the sparse auto-encoder (SAE). Convolution mechanism is applied after the SAE feature extraction stage with the SAE features upon the larger outer window. The UCSAE algorithm was validated by two common hyperspectral imagery (HSI) datasets - Pavia University dataset and the Kennedy Space Centre (KSC) dataset, which shows an improvement over the traditional hyperspectral spatial-spectral classification methods.

  20. Development of a Hyperspectral Imaging System for Online Quality Inspection of Pickling Cucumbers

    Science.gov (United States)

    This paper reports on the development of a hyperspectral imaging prototype for evaluation of external and internal quality of pickling cucumbers. The prototype consisted of a two-lane round belt conveyor, two illumination sources (one for reflectance and one for transmittance), and a hyperspectral i...

  1. Fast and robust pushbroom hyperspectral imaging via DMD-based scanning

    CERN Document Server

    Arablouei, Reza; Gensemer, Stephen; Kusy, Branislav

    2016-01-01

    We describe a new pushbroom hyperspectral imaging device that has no macro moving part. The main components of the proposed hyperspectral imager are a digital micromirror device (DMD), a CMOS image sensor with no filter as the spectral sensor, a CMOS color (RGB) image sensor as the auxiliary image sensor, and a diffraction grating. Using the image sensor pair, the device can simultaneously capture hyperspectral data as well as RGB images of the scene. The RGB images captured by the auxiliary image sensor can facilitate geometric co-registration of the hyperspectral image slices captured by the spectral sensor. In addition, the information discernible from the RGB images can lead to capturing the spectral data of only the regions of interest within the scene. The proposed hyperspectral imaging architecture is cost-effective, fast, and robust. It also enables a trade-off between resolution and speed. We have built an initial prototype based on the proposed design. The prototype can capture a hyperspectral image...

  2. Time series hyperspectral chemical imaging data: challenges, solutions and applications.

    Science.gov (United States)

    Gowen, A A; Marini, F; Esquerre, C; O'Donnell, C; Downey, G; Burger, J

    2011-10-31

    Hyperspectral chemical imaging (HCI) integrates imaging and spectroscopy resulting in three-dimensional data structures, hypercubes, with two spatial and one wavelength dimension. Each spatial image pixel in a hypercube contains a spectrum with >100 datapoints. While HCI facilitates enhanced monitoring of multi-component systems; time series HCI offers the possibility of a more comprehensive understanding of the dynamics of such systems and processes. This implies a need for modeling strategies that can cope with the large multivariate data structures generated in time series HCI experiments. The challenges posed by such data include dimensionality reduction, temporal morphological variation of samples and instrumental drift. This article presents potential solutions to these challenges, including multiway analysis, object tracking, multivariate curve resolution and non-linear regression. Several real world examples of time series HCI data are presented to illustrate the proposed solutions. Copyright © 2011 Elsevier B.V. All rights reserved.

  3. Compressive Fluorescence Microscopy for Biological and Hyperspectral Imaging

    CERN Document Server

    Studer, Vincent; Chahid, Makhlad; Moussavi, Hamed; Candes, Emmanuel; Dahan, Maxime

    2012-01-01

    The mathematical theory of compressed sensing (CS) asserts that one can acquire signals from measurements whose rate is much lower than the total bandwidth. Whereas the CS theory is now well developed, challenges concerning hardware implementations of CS-based acquisition devices---especially in optics---have only started being addressed. This paper presents an implementation of compressive sensing in fluorescence microscopy and its applications to biomedical imaging. Our CS microscope combines a dynamic structured wide-field illumination and a fast and sensitive single-point fluorescence detection to enable reconstructions of images of fluorescent beads, cells and tissues with undersampling ratios (between the number of pixels and number of measurements) up to 32. We further demonstrate a hyperspectral mode and record images with 128 spectral channels and undersampling ratios up to 64, illustrating the potential benefits of CS acquisition for higher dimensional signals which typically exhibits extreme redund...

  4. Methods for gas detection using stationary hyperspectral imaging sensors

    Science.gov (United States)

    Conger, James L [San Ramon, CA; Henderson, John R [Castro Valley, CA

    2012-04-24

    According to one embodiment, a method comprises producing a first hyperspectral imaging (HSI) data cube of a location at a first time using data from a HSI sensor; producing a second HSI data cube of the same location at a second time using data from the HSI sensor; subtracting on a pixel-by-pixel basis the second HSI data cube from the first HSI data cube to produce a raw difference cube; calibrating the raw difference cube to produce a calibrated raw difference cube; selecting at least one desired spectral band based on a gas of interest; producing a detection image based on the at least one selected spectral band and the calibrated raw difference cube; examining the detection image to determine presence of the gas of interest; and outputting a result of the examination. Other methods, systems, and computer program products for detecting the presence of a gas are also described.

  5. Adaptive interference hyperspectral image compression with spectrum distortion control

    Institute of Scientific and Technical Information of China (English)

    Jing Ma; Yunsong Li; Chengke Wu; Dong Chen

    2009-01-01

    As one of the next generation imaging spectrometers,interferential spectrometer has been paid much attention.With traditional spectrum compression methods,the hyperspectral images generated by interferential spectrometer can only be protected with better visual quality in spatial domain,but its optical applications in Fourier domain are often ignored.So the relation between the distortion in Fourier domain and the compression in spatial domain is analyzed in this letter.Based on this analysis,a novel coding scheme is proposed,which can compress data in spatial domain while reducing the distortion in Fourier domain.The bitstream of set partitioning in hierarchical trees (SPIHT) is truncated by adaptively lifting the rate-distortion slopes of zerotrees according to the priorities of optical path difference (OPD) based on rate-distortion optimization theory.Experimental results show that the proposed scheme can achieve better performance in Fourier domain while maintaining the image quality in spatial domain.

  6. Active Volcano Monitoring using a Space-based Hyperspectral Imager

    Science.gov (United States)

    Cipar, J. J.; Dunn, R.; Cooley, T.

    2010-12-01

    Active volcanoes occur on every continent, often in close proximity to heavily populated areas. While ground-based studies are essential for scientific research and disaster mitigation, remote sensing from space can provide rapid and continuous monitoring of active and potentially active volcanoes [Ramsey and Flynn, 2004]. In this paper, we report on hyperspectral measurements of Kilauea volcano, Hawaii. Hyperspectral images obtained by the US Air Force TacSat-3/ARTEMIS sensor [Lockwood et al, 2006] are used to obtain estimates of the surface temperatures for the volcano. ARTEMIS measures surface-reflected light in the visible, near-infrared, and short-wave infrared bands (VNIR-SWIR). The SWIR bands are known to be sensitive to thermal radiation [Green, 1996]. For example, images from the NASA Hyperion hyperspectral sensor have shown the extent of wildfires and active volcanoes [Young, 2009]. We employ the methodology described by Dennison et al, (2006) to obtain an estimate of the temperature of the active region of Kilauea. Both day and night-time images were used in the analysis. To improve the estimate, we aggregated neighboring pixels. The active rim of the lava lake is clearly discernable in the temperature image, with a measured temperature exceeding 1100o C. The temperature decreases markedly on the exterior of the summit crater. While a long-wave infrared (LWIR) sensor would be ideal for volcano monitoring, we have shown that the thermal state of an active volcano can be monitored using the SWIR channels of a reflective hyperspectral imager. References: Dennison, Philip E., Kraivut Charoensiri, Dar A. Roberts, Seth H. Peterson, and Robert O. Green (2006). Wildfire temperature and land cover modeling using hyperspectral data, Remote Sens. Environ., vol. 100, pp. 212-222. Green, R. O. (1996). Estimation of biomass fire temperature and areal extent from calibrated AVIRIS spectra, in Summaries of the 6th Annual JPL Airborne Earth Science Workshop, Pasadena, CA

  7. Band selection for change detection from hyperspectral images

    Science.gov (United States)

    Liu, Sicong; Du, Qian; Tong, Xiaohua

    2017-05-01

    In this paper, we propose to apply unsupervised band selection to improve the performance of change detection in multitemporal hyperspectral images (HSI-CD). By reducing data dimensionality through finding the most distinctive and informative bands in the difference image, foreground changes may be better detected. Band selection-based dimensionality reduction (BS-DR) technique is considered to investigate in details the following sub-problems in HSI-CD including: 1) the estimated number of multi-class changes; 2) the binary CD; 3) the multiple CD; 4) the change discriminability; 5) the optimal number of selected bands. Thus it contributes at first time a quantitative analysis of the BS-DR approach impacting on the HSI-CD performance. Due to the difficulty of having training samples in an unknown environment, unsupervised band selection and change detection are considered. A pair of real multitemporal hyperspectral Hyperion data set has been used to validate the proposed approach. Experimental results confirmed the effectiveness of selecting a band subset to obtain a satisfactory CD result, comparing with the one using original full bands. In addition, the results also demonstrated that the reduced feature space is capable to maintain sufficient information for detecting the occurred spectrally significant changes. CD performance is enhanced with respect to the increasing of change representative and discriminable capabilities.

  8. Hyperspectral imaging for dermal hemoglobin spectroscopy

    Science.gov (United States)

    Dwyer, Peter J.; DiMarzio, Charles A.

    1999-10-01

    It has been shown previously that images collected at selected wavelengths in a sufficiently narrow bandwidth can be used to produce maps of the oxygen saturation of hemoglobin in the dermis. A four-wavelength algorithm has been developed based on a two-layer model of the skin, in which the blood is contained in the lower layer (dermis), while the upper layer attenuates some of the reflection and adds a clutter term. In the present work, the algorithm is compared analytically to simpler algorithms using three wavelengths and based on a single-layer model. It is shown through Monte-Carlo models that, for typical skin, the single-layer model is adequate to analyze data from fiber-optical reflectance spectroscopy, but the two-layer model produces better results for imaging systems. Although the model does not address the full complexity of reflectance of a two-layer skin, it has proven to be sufficient to recover the oxygen saturation, and perhaps other medically relevant information. The algorithm is demonstrated on a suction blister, where the epidermis is removed to reveal the underlying dermis. Applications for this imaging modality exist in dermatology, in surgery, and in developing treatment plans for various diseases.

  9. Identification of different varieties of sesame oil using near-infrared hyperspectral imaging and chemometrics algorithms.

    Directory of Open Access Journals (Sweden)

    Chuanqi Xie

    Full Text Available This study investigated the feasibility of using near infrared hyperspectral imaging (NIR-HSI technique for non-destructive identification of sesame oil. Hyperspectral images of four varieties of sesame oil were obtained in the spectral region of 874-1734 nm. Reflectance values were extracted from each region of interest (ROI of each sample. Competitive adaptive reweighted sampling (CARS, successive projections algorithm (SPA and x-loading weights (x-LW were carried out to identify the most significant wavelengths. Based on the sixty-four, seven and five wavelengths suggested by CARS, SPA and x-LW, respectively, two classified models (least squares-support vector machine, LS-SVM and linear discriminant analysis,LDA were established. Among the established models, CARS-LS-SVM and CARS-LDA models performed well with the highest classification rate (100% in both calibration and prediction sets. SPA-LS-SVM and SPA-LDA models obtained better results (95.59% and 98.53% of classification rate in prediction set with only seven wavelengths (938, 1160, 1214, 1406, 1656, 1659 and 1663 nm. The x-LW-LS-SVM and x-LW-LDA models also obtained satisfactory results (>80% of classification rate in prediction set with the only five wavelengths (921, 925, 995, 1453 and 1663 nm. The results showed that NIR-HSI technique could be used to identify the varieties of sesame oil rapidly and non-destructively, and CARS, SPA and x-LW were effective wavelengths selection methods.

  10. Identification of early cancerous lesion of esophagus with endoscopic images by hyperspectral image technique (Conference Presentation)

    Science.gov (United States)

    Huang, Shih-Wei; Chen, Shih-Hua; Chen, Weichung; Wu, I.-Chen; Wu, Ming Tsang; Kuo, Chie-Tong; Wang, Hsiang-Chen

    2016-03-01

    This study presents a method to identify early esophageal cancer within endoscope using hyperspectral imaging technology. The research samples are three kinds of endoscopic images including white light endoscopic, chromoendoscopic, and narrow-band endoscopic images with different stages of pathological changes (normal, dysplasia, dysplasia - esophageal cancer, and esophageal cancer). Research is divided into two parts: first, we analysis the reflectance spectra of endoscopic images with different stages to know the spectral responses by pathological changes. Second, we identified early cancerous lesion of esophagus by principal component analysis (PCA) of the reflectance spectra of endoscopic images. The results of this study show that the identification of early cancerous lesion is possible achieve from three kinds of images. In which the spectral characteristics of NBI endoscopy images of a gray area than those without the existence of the problem the first two, and the trend is very clear. Therefore, if simply to reflect differences in the degree of spectral identification, chromoendoscopic images are suitable samples. The best identification of early esophageal cancer is using the NBI endoscopic images. Based on the results, the use of hyperspectral imaging technology in the early endoscopic esophageal cancer lesion image recognition helps clinicians quickly diagnose. We hope for the future to have a relatively large amount of endoscopic image by establishing a hyperspectral imaging database system developed in this study, so the clinician can take this repository more efficiently preliminary diagnosis.

  11. Application of hyperspectral imaging in food safety inspection and control: a review.

    Science.gov (United States)

    Feng, Yao-Ze; Sun, Da-Wen

    2012-01-01

    Food safety is a great public concern, and outbreaks of food-borne illnesses can lead to disturbance to the society. Consequently, fast and nondestructive methods are required for sensing the safety situation of produce. As an emerging technology, hyperspectral imaging has been successfully employed in food safety inspection and control. After presenting the fundamentals of hyperspectral imaging, this paper provides a comprehensive review on its application in determination of physical, chemical, and biological contamination on food products. Additionally, other studies, including detecting meat and meat bone in feedstuffs as well as organic residue on food processing equipment, are also reported due to their close relationship with food safety control. With these applications, it can be demonstrated that miscellaneous hyperspectral imaging techniques including near-infrared hyperspectral imaging, fluorescence hyperspectral imaging, and Raman hyperspectral imaging or their combinations are powerful tools for food safety surveillance. Moreover, it is envisaged that hyperspectral imaging can be considered as an alternative technique for conventional methods in realizing inspection automation, leading to the elimination of the occurrence of food safety problems at the utmost.

  12. Evaluation of the effectiveness of laser crust removal on granites by means of hyperspectral imaging techniques

    Energy Technology Data Exchange (ETDEWEB)

    Pozo-Antonio, J.S., E-mail: santiago.pozo@udc.es [Laboratorio de Aplicacións Industriais do Láser, Centro de Investigacións Tecnolóxicas (CIT), Departamento de Enxeñaría Industrial II, Escola Politécnica Superior, Universidade de Coruña (UDC), Campus Ferrol, 15403 Ferrol (Spain); Fiorucci, M.P.; Ramil, A.; López, A.J. [Laboratorio de Aplicacións Industriais do Láser, Centro de Investigacións Tecnolóxicas (CIT), Departamento de Enxeñaría Industrial II, Escola Politécnica Superior, Universidade de Coruña (UDC), Campus Ferrol, 15403 Ferrol (Spain); Rivas, T. [Departamento de Enxeñaría dos Recursos Naturais e Medioambiente, Escola Superior de Minas, Universidade de Vigo, 36310 Vigo (Spain)

    2015-08-30

    Highlights: • Hyperspectral imaging techniques for determining the degree of crust removal on granites used in Cultural Heritage. • Hyperspectral imaging techniques allow to in situ evaluate of the effectiveness of the laser cleaning. • Hyperspectral imaging data are consistent with the information obtained by conventional techniques about the cleaning effectiveness. - Abstract: In this paper, we present a study of the application of the hyperspectral imaging technique in order to non-destructively evaluate the laser cleaning of the biogenic patina and the sulphated black crust developed on a fine-grained granite used in the construction of Cultural Heritage in NW Spain. The grained polymineral texture of the granite hinders the adjustment of laser irradiation parameters during the cleaning, and therefore the in situ process control. The cleaning was performed with a nanosecond pulsed Nd:YVO{sub 4} laser at 355 nm. A hyperspectral camera was used to in situ assess the effectiveness of cleaning by recording images of the rock surfaces before and during the laser treatment. Different analytical techniques were used to test the ability of the hyperspectral imaging technique to evaluate the cleaning process of the granite samples: optical microscopy, scanning electron microscopy with energy dispersive X-ray spectrometry (SEM - EDX), Fourier transform infrared spectroscopy (FTIR) and spectrophotometer colour measurements. The results indicated that hyperspectral imaging technique is a reliable and more affordable technique to in situ evaluate the process of laser cleaning of the biogenic patina and the sulphated black crust in fine-grained granites.

  13. Detection of cold injury in peaches by hyperspectral reflectance imaging and artificial neural network.

    Science.gov (United States)

    Pan, Leiqing; Zhang, Qiang; Zhang, Wei; Sun, Ye; Hu, Pengcheng; Tu, Kang

    2016-02-01

    Peaches in cold storage may develop chill damage, as symptomized by deteriorated texture and lack of juice. To examine fruit quality, we established a hyperspectral imaging system to detect cold injury, and an artificial neural network (ANN) model was developed for which eight optimal wavelengths were selected. Between normal and chill-damaged peaches, significant differences in fruit quality parameters and the spectral response to correlating selected wavelengths were observed. Evidencing this relationship, the correlation coefficients between quality parameters and the respective spectral response of eight selected wavelengths were -0.587 to -0.700, 0.393 to 0.552, 0.510 to 0.751, and 0.574 to 0.773. With optimal representative wavelengths as inputs for the ANN model, the overall classification accuracy of chill damage was 95.8% for all cold-stored samples. The ANN prediction models for quality parameters performed well, with correlation coefficients from 0.6979 to 0.9026. This research demonstrates feasibility of hyperspectral reflectance imaging technique for detecting cold injury.

  14. Superpixel Segmentation for Endmember Detection in Hyperspectral Images

    Science.gov (United States)

    Thompson, D. R.; de Granville, C.; Gilmore, M. S.; Castano, R.

    2009-12-01

    "Superpixel segmentation" is a novel approach to facilitate statistical analyses of hyperspectral image data with high spatial resolution and subtle spectral features. The method oversegments the image into homogeneous regions each comprised of several contiguous pixels. This can reduce noise by exploiting scene features' spatial contiguity: isolated spectral features are likely to be noise, but spectral features that appear in several adjacent pixels probably indicate real materials in the scene. The mean spectra from each superpixel define a smaller, noise-reduced dataset. This preprocessing step improves endmember detection for the images in our study. Our endmember detection approach presumes a linear (geographic) mixing model for image spectra. We generate superpixels with the Felzenszwalb/Huttenlocher graph-based segmentation [1] with a Euclidean distance metric. This segmentation shatters the image into thousands of superpixels, each with an area of approximately 20 image pixels. We then apply Symmetric Maximum Angle Convex Cone (SMACC) endmember detection algorithm to the data set consisting of the mean spectrum from all superpixels. We evaluated the approach for several images from the Compact Reconnaissance Imaging Spectrometer (CRISM) [2]. We used the 1000-2500nm wavelengths of images frt00003e12 and frt00003fb9. We cleaned the images with atmospheric correction based on Olympus Mons spectra [3] and preprocessed with a radius-1 median filter in the spectral domain. Endmembers produced with and without the superpixel reduction are compared to the representative (mean) spectra of five representative mineral classes identified in an expert analysis of each scene. Expert-identified minerals include mafic minerals and phyllosilicate deposits that in some cases subtended just a few tens of pixels. Only the endmembers from the superpixel approach reflected all major mineral constituents in the images. Additionally, the superpixel endmembers are more

  15. [Detection of Hawthorn Fruit Defects Using Hyperspectral Imaging].

    Science.gov (United States)

    Liu, De-hua; Zhang, Shu-juan; Wang, Bin; Yu, Ke-qiang; Zhao, Yan-ru; He, Yong

    2015-11-01

    Hyperspectral imaging technology covered the range of 380-1000 nm was employed to detect defects (bruise and insect damage) of hawthorn fruit. A total of 134 samples were collected, which included damage fruit of 46, pest fruit of 30, injure and pest fruit of 10 and intact fruit of 48. Because calyx · s⁻¹ tem-end and bruise/insect damage regions offered a similar appearance characteristic in RGB images, which could produce easily confusion between them. Hence, five types of defects including bruise, insect damage, sound, calyx, and stem-end were collected from 230 hawthorn fruits. After acquiring hyperspectral images of hawthorn fruits, the spectral data were extracted from region of interest (ROI). Then, several pretreatment methods of standard normalized variate (SNV), savitzky golay (SG), median filter (MF) and multiplicative scatter correction (MSC) were used and partial least squares method(PLS) model was carried out to obtain the better performance. Accordingly to their results, SNV pretreatment methods assessed by PLS was viewed as best pretreatment method. Lastly, SNV was chosen as the pretreatment method. Spectral features of five different regions were combined with Regression coefficients(RCs) of partial least squares-discriminant analysis (PLS-DA) model was used to identify the important wavelengths and ten wavebands at 483, 563, 645, 671, 686, 722, 777, 819, 837 and 942 nm were selected from all of the wavebands. Using Kennard-Stone algorithm, all kinds of samples were randomly divided into training set (173) and test set (57) according to the proportion of 3:1. And then, least squares-support vector machine (LS-SVM) discriminate model was established by using the selected wavebands. The results showed that the discriminate accuracy of the method was 91.23%. In the other hand, images at ten important wavebands were executed to Principal component analysis (PCA). Using "Sobel" operator and region growing algrorithm "Regiongrow", the edge and defect

  16. Classification Technique for Hyperspectral Image Based on Subspace of Bands Feature Extraction and LS-SVM%基于波段子集特征提取的最小二乘支持向量机高光谱图像分类技术

    Institute of Scientific and Technical Information of China (English)

    高恒振; 万建伟; 朱珍珍; 王力宝; 粘永健

    2011-01-01

    The present paper proposes a novel hyperspectral image classification algorithm based on LS-SVM (least squares support vector machine). The LS-SVM uses the features extracted from subspace of bands (SOB). The maximum noise fraction (MNF) method is adopted as the feature extraction method. The spectral correlations of the hyperspectral image are used in order to divide the feature space into several SOBs. Then the MNF is used to extract characteristic features of the SOBs. The extracted features are combined into the feature vector for classification. So the strong bands correlation is avoided and the spectral redundancies are reduced. The LS-SVM classifier is adopted, which replaces inequality constraints in SVM by equality constraints. So the computation consumption is reduced and the learning performance is improved. The proposed method optimizes spectral information by feature extraction and reduces the spectral noise. The classifier performance is improved. Experimental results show the superiorities of the proposed algorithm.%针对高光谱图像分类,文章提出一种基于波段子集最大噪声分量特征提取的最小二乘支持向量机的高光谱图像分类算法.利用高光谱图像的谱间相关性将原始光谱波段划分为若干个波段子集,并在各个子集上采用最大噪声分量方法进行特征提取,将提取的特征合成为分类的组合特征矢量,避免了高光谱图像较强的波段相关性,减少了谱间冗余.并且采用了最小二乘支持向量机,用等式约束取代了支持向量机中的不等式约束,降低了运算量,提高了学习效率.该方法利用特征提取优化了光谱信息,降低了谱间噪声,提高了分类器的性能.实验结果证明了本文算法的优越性.

  17. Hyperspectral imaging simulation of object under sea-sky background

    Science.gov (United States)

    Wang, Biao; Lin, Jia-xuan; Gao, Wei; Yue, Hui

    2016-10-01

    Remote sensing image simulation plays an important role in spaceborne/airborne load demonstration and algorithm development. Hyperspectral imaging is valuable in marine monitoring, search and rescue. On the demand of spectral imaging of objects under the complex sea scene, physics based simulation method of spectral image of object under sea scene is proposed. On the development of an imaging simulation model considering object, background, atmosphere conditions, sensor, it is able to examine the influence of wind speed, atmosphere conditions and other environment factors change on spectral image quality under complex sea scene. Firstly, the sea scattering model is established based on the Philips sea spectral model, the rough surface scattering theory and the water volume scattering characteristics. The measured bi directional reflectance distribution function (BRDF) data of objects is fit to the statistical model. MODTRAN software is used to obtain solar illumination on the sea, sky brightness, the atmosphere transmittance from sea to sensor and atmosphere backscattered radiance, and Monte Carlo ray tracing method is used to calculate the sea surface object composite scattering and spectral image. Finally, the object spectrum is acquired by the space transformation, radiation degradation and adding the noise. The model connects the spectrum image with the environmental parameters, the object parameters, and the sensor parameters, which provide a tool for the load demonstration and algorithm development.

  18. HELICoiD project: a new use of hyperspectral imaging for brain cancer detection in real-time during neurosurgical operations

    Science.gov (United States)

    Fabelo, Himar; Ortega, Samuel; Kabwama, Silvester; Callico, Gustavo M.; Bulters, Diederik; Szolna, Adam; Pineiro, Juan F.; Sarmiento, Roberto

    2016-05-01

    Hyperspectral images allow obtaining large amounts of information about the surface of the scene that is captured by the sensor. Using this information and a set of complex classification algorithms is possible to determine which material or substance is located in each pixel. The HELICoiD (HypErspectraL Imaging Cancer Detection) project is a European FET project that has the goal to develop a demonstrator capable to discriminate, with high precision, between normal and tumour tissues, operating in real-time, during neurosurgical operations. This demonstrator could help the neurosurgeons in the process of brain tumour resection, avoiding the excessive extraction of normal tissue and unintentionally leaving small remnants of tumour. Such precise delimitation of the tumour boundaries will improve the results of the surgery. The HELICoiD demonstrator is composed of two hyperspectral cameras obtained from Headwall. The first one in the spectral range from 400 to 1000 nm (visible and near infrared) and the second one in the spectral range from 900 to 1700 nm (near infrared). The demonstrator also includes an illumination system that covers the spectral range from 400 nm to 2200 nm. A data processing unit is in charge of managing all the parts of the demonstrator, and a high performance platform aims to accelerate the hyperspectral image classification process. Each one of these elements is installed in a customized structure specially designed for surgical environments. Preliminary results of the classification algorithms offer high accuracy (over 95%) in the discrimination between normal and tumour tissues.

  19. Excitation-scanning hyperspectral imaging system for microscopic and endoscopic applications

    Science.gov (United States)

    Mayes, Sam A.; Leavesley, Silas J.; Rich, Thomas C.

    2016-04-01

    Current microscopic and endoscopic technologies for cancer screening utilize white-light illumination sources. Hyper-spectral imaging has been shown to improve sensitivity while retaining specificity when compared to white-light imaging in both microscopy and in vivo imaging. However, hyperspectral imaging methods have historically suffered from slow acquisition times due to the narrow bandwidth of spectral filters. Often minutes are required to gather a full image stack. We have developed a novel approach called excitation-scanning hyperspectral imaging that provides 2-3 orders of magnitude increased signal strength. This reduces acquisition times significantly, allowing for live video acquisition. Here, we describe a preliminary prototype excitation-scanning hyperspectral imaging system that can be coupled with endoscopes or microscopes for hyperspectral imaging of tissues and cells. Our system is comprised of three subsystems: illumination, transmission, and imaging. The illumination subsystem employs light-emitting diode arrays to illuminate at different wavelengths. The transmission subsystem utilizes a unique geometry of optics and a liquid light guide. Software controls allow us to interface with and control the subsystems and components. Digital and analog signals are used to coordinate wavelength intensity, cycling and camera triggering. Testing of the system shows it can cycle 16 wavelengths at as fast as 1 ms per cycle. Additionally, more than 18% of the light transmits through the system. Our setup should allow for hyperspectral imaging of tissue and cells in real time.

  20. Near-Infrared Hyper-spectral Image Analysis of Astaxanthin Concentration in Fish Feed Coating

    DEFF Research Database (Denmark)

    Ljungqvist, Martin Georg; Ersbøll, Bjarne Kjær; Kobayashi, K.;

    2012-01-01

    The aim of this study was to investigate the possibility of predicting concentration levels of synthetic astaxanthin coating of aquaculture feed pellets by hyper-spectral image analysis in the near infra-red (NIR) range and optical filter design. The imaging devices used were a Videometer...... for prediction of the concentration level. The results show that it is possible to predict the level of synthetic astaxanthin coating using either hyper-spectral imaging or three bandpass filters (BPF)....

  1. Objected-oriented remote sensing image classification method based on geographic ontology model

    Science.gov (United States)

    Chu, Z.; Liu, Z. J.; Gu, H. Y.

    2016-11-01

    Nowadays, with the development of high resolution remote sensing image and the wide application of laser point cloud data, proceeding objected-oriented remote sensing classification based on the characteristic knowledge of multi-source spatial data has been an important trend on the field of remote sensing image classification, which gradually replaced the traditional method through improving algorithm to optimize image classification results. For this purpose, the paper puts forward a remote sensing image classification method that uses the he characteristic knowledge of multi-source spatial data to build the geographic ontology semantic network model, and carries out the objected-oriented classification experiment to implement urban features classification, the experiment uses protégé software which is developed by Stanford University in the United States, and intelligent image analysis software—eCognition software as the experiment platform, uses hyperspectral image and Lidar data that is obtained through flight in DaFeng City of JiangSu as the main data source, first of all, the experiment uses hyperspectral image to obtain feature knowledge of remote sensing image and related special index, the second, the experiment uses Lidar data to generate nDSM(Normalized DSM, Normalized Digital Surface Model),obtaining elevation information, the last, the experiment bases image feature knowledge, special index and elevation information to build the geographic ontology semantic network model that implement urban features classification, the experiment results show that, this method is significantly higher than the traditional classification algorithm on classification accuracy, especially it performs more evidently on the respect of building classification. The method not only considers the advantage of multi-source spatial data, for example, remote sensing image, Lidar data and so on, but also realizes multi-source spatial data knowledge integration and application

  2. Bayer Filter Snapshot Hyperspectral Fundus Camera for Human Retinal Imaging.

    Science.gov (United States)

    Kaluzny, Joel; Li, Hao; Liu, Wenzhong; Nesper, Peter; Park, Justin; Zhang, Hao F; Fawzi, Amani A

    2017-04-01

    To demonstrate the versatility and performance of a compact Bayer filter snapshot hyperspectral fundus camera for in-vivo clinical applications including retinal oximetry and macular pigment optical density measurements. 12 healthy volunteers were recruited under an Institutional Review Board (IRB) approved protocol. Fundus images were taken with a custom hyperspectral camera with a spectral range of 460-630 nm. We determined retinal vascular oxygen saturation (sO2) for the healthy population using the captured spectra by least squares curve fitting. Additionally, macular pigment optical density was localized and visualized using multispectral reflectometry from selected wavelengths. We successfully determined the mean sO2 of arteries and veins of each subject (ages 21-80) with excellent intrasubject repeatability (1.4% standard deviation). The mean arterial sO2 for all subjects was 90.9% ± 2.5%, whereas the mean venous sO2 for all subjects was 64.5% ± 3.5%. The mean artery-vein (A-V) difference in sO2 varied between 20.5% and 31.9%. In addition, we were able to reveal and quantify macular pigment optical density. We demonstrated a single imaging tool capable of oxygen saturation and macular pigment density measurements in vivo. The unique combination of broad spectral range, high spectral-spatial resolution, rapid and robust imaging capability, and compact design make this system a valuable tool for multifunction spectral imaging that can be easily performed in a clinic setting.

  3. Detection of hypercholesterolemia using hyperspectral imaging of human skin

    Science.gov (United States)

    Milanic, Matija; Bjorgan, Asgeir; Larsson, Marcus; Strömberg, Tomas; Randeberg, Lise L.

    2015-07-01

    Hypercholesterolemia is characterized by high blood levels of cholesterol and is associated with increased risk of atherosclerosis and cardiovascular disease. Xanthelasma is a subcutaneous lesion appearing in the skin around the eyes. Xanthelasma is related to hypercholesterolemia. Identifying micro-xanthelasma can thereforeprovide a mean for early detection of hypercholesterolemia and prevent onset and progress of disease. The goal of this study was to investigate spectral and spatial characteristics of hypercholesterolemia in facial skin. Optical techniques like hyperspectral imaging (HSI) might be a suitable tool for such characterization as it simultaneously provides high resolution spatial and spectral information. In this study a 3D Monte Carlo model of lipid inclusions in human skin was developed to create hyperspectral images in the spectral range 400-1090 nm. Four lesions with diameters 0.12-1.0 mm were simulated for three different skin types. The simulations were analyzed using three algorithms: the Tissue Indices (TI), the two layer Diffusion Approximation (DA), and the Minimum Noise Fraction transform (MNF). The simulated lesions were detected by all methods, but the best performance was obtained by the MNF algorithm. The results were verified using data from 11 volunteers with known cholesterol levels. The face of the volunteers was imaged by a LCTF system (400- 720 nm), and the images were analyzed using the previously mentioned algorithms. The identified features were then compared to the known cholesterol levels of the subjects. Significant correlation was obtained for the MNF algorithm only. This study demonstrates that HSI can be a promising, rapid modality for detection of hypercholesterolemia.

  4. Application of hyperspectral fluorescence lifetime imaging to tissue autofluorescence: arthritis

    Science.gov (United States)

    Talbot, C. B.; Benninger, R. K. P.; de Beule, P.; Requejo-Isidro, J.; Elson, D. S.; Dunsby, C.; Munro, I.; Neil, M. A.; Sandison, A.; Sofat, N.; Nagase, H.; French, P. M. W.; Lever, M. J.

    2005-08-01

    Tissue contains many natural fluorophores and therefore by exploiting autofluorescence, we can obtain information from tissue with less interference than conventional histological techniques. However, conventional intensity imaging is prone to artifacts since it is an absolute measurement. Fluorescence lifetime and spectral measurements are relative measurements and therefore allow for better measurements. We have applied FLIM and hyperspectral FLIM to the study of articular cartilage and its disease arthritis. We have analyzed normal human articular cartilage and cartilage which was in the early stages of disease. In this case, it was found that FLIM was able to detect changes in the diseased tissue that were not detectable with the conventional diagnosis. Specifically, the fluorescence lifetimes (FL) of the cells were different between the two samples. We have also applied hyperspectral FLIM to degraded cartilage through treatment with interleukin-1. In this case, it was found that there was a shift in the emission spectrum with treatment and that the lifetime had also increased. We also showed that there was greater contrast between the cells and the extracellular matrix (ECM) at longer wavelengths.

  5. Framelet-Based Sparse Unmixing of Hyperspectral Images.

    Science.gov (United States)

    Zhang, Guixu; Xu, Yingying; Fang, Faming

    2016-04-01

    Spectral unmixing aims at estimating the proportions (abundances) of pure spectrums (endmembers) in each mixed pixel of hyperspectral data. Recently, a semi-supervised approach, which takes the spectral library as prior knowledge, has been attracting much attention in unmixing. In this paper, we propose a new semi-supervised unmixing model, termed framelet-based sparse unmixing (FSU), which promotes the abundance sparsity in framelet domain and discriminates the approximation and detail components of hyperspectral data after framelet decomposition. Due to the advantages of the framelet representations, e.g., images have good sparse approximations in framelet domain, and most of the additive noises are included in the detail coefficients, the FSU model has a better antinoise capability, and accordingly leads to more desirable unmixing performance. The existence and uniqueness of the minimizer of the FSU model are then discussed, and the split Bregman algorithm and its convergence property are presented to obtain the minimal solution. Experimental results on both simulated data and real data demonstrate that the FSU model generally performs better than the compared methods.

  6. Pigment identification in pictorial layers by HyperSpectral Imaging

    Science.gov (United States)

    Capobianco, Giuseppe; Bonifazi, Giuseppe; Prestileo, Fernanda; Serranti, Silvia

    2014-05-01

    The use of Hyper-Spectral Imaging (HSI) as a diagnostic tool in the field of cultural heritage is of great interest presenting high potentialities. This analysis, in fact, is non-destructive, non-invasive and portable. Furthermore, the possibility to couple hyperspectral data with chemometric techniques allows getting qualitative and/or quantitative information on the nature and physical-chemical characteristics of the investigated materials. A study was carried out to explore the possibilities offered by this approach to identify pigments in paintings. More in detail, six pigments have been selected and they have been then mixed with four different binders and applied to a wood support. The resulting reference samples were acquired by HSI in the SWIR wavelength range (1000-2500 nm). Data were processed adopting a chemometric approach based on the PLS Toolbox (Eigenvector Research, Inc.) running inside Matlab® (The Mathworks, Inc.). The aim of the study was to verify, according to the information acquired in the investigated wavelength region, the correlation existing between collected spectral signatures and sample characteristics related to the different selected pigments and binders. Results were very good showing as correlations exist. New scenarios can thus be envisaged for analysis, characterization, conservation and restoration of paintings, considering that the developed approach allows to obtain, just "in one shot", information, not only on the type of pigment, but also on the utilized binder and support.

  7. Non-destructive quality evaluation of pepper (Capsicum annuum L.) seeds using LED-induced hyperspectral reflectance imaging.

    Science.gov (United States)

    Mo, Changyeun; Kim, Giyoung; Lee, Kangjin; Kim, Moon S; Cho, Byoung-Kwan; Lim, Jongguk; Kang, Sukwon

    2014-04-24

    In this study, we developed a viability evaluation method for pepper (Capsicum annuum L.) seeds based on hyperspectral reflectance imaging. The reflectance spectra of pepper seeds in the 400-700 nm range are collected from hyperspectral reflectance images obtained using blue, green, and red LED illumination. A partial least squares-discriminant analysis (PLS-DA) model is developed to classify viable and non-viable seeds. Four spectral ranges generated with four types of LEDs (blue, green, red, and RGB), which were pretreated using various methods, are investigated to develop the classification models. The optimal PLS-DA model based on the standard normal variate for RGB LED illumination (400-700 nm) yields discrimination accuracies of 96.7% and 99.4% for viable seeds and nonviable seeds, respectively. The use of images based on the PLS-DA model with the first-order derivative of a 31.5-nm gap for red LED illumination (600-700 nm) yields 100% discrimination accuracy for both viable and nonviable seeds. The results indicate that a hyperspectral imaging technique based on LED light can be potentially applied to high-quality pepper seed sorting.

  8. Non-Destructive Quality Evaluation of Pepper (Capsicum annuum L. Seeds Using LED-Induced Hyperspectral Reflectance Imaging

    Directory of Open Access Journals (Sweden)

    Changyeun Mo

    2014-04-01

    Full Text Available In this study, we developed a viability evaluation method for pepper (Capsicum annuum L. seeds based on hyperspectral reflectance imaging. The reflectance spectra of pepper seeds in the 400–700 nm range are collected from hyperspectral reflectance images obtained using blue, green, and red LED illumination. A partial least squares–discriminant analysis (PLS-DA model is developed to classify viable and non-viable seeds. Four spectral ranges generated with four types of LEDs (blue, green, red, and RGB, which were pretreated using various methods, are investigated to develop the classification models. The optimal PLS-DA model based on the standard normal variate for RGB LED illumination (400–700 nm yields discrimination accuracies of 96.7% and 99.4% for viable seeds and nonviable seeds, respectively. The use of images based on the PLS-DA model with the first-order derivative of a 31.5-nm gap for red LED illumination (600–700 nm yields 100% discrimination accuracy for both viable and nonviable seeds. The results indicate that a hyperspectral imaging technique based on LED light can be potentially applied to high-quality pepper seed sorting.

  9. Hyperspectral Imaging and Association Phenomenology of Pedestrians in a Cluttered Urban Environment

    Science.gov (United States)

    Herweg, Jared A.

    Remote hyperspectral imaging (HSI) has shown promise in several applications such as object detection and tracking. Typically research has focused on large objects, such as vehicles, for tracking due to the spatial resolution of current operational HSI systems. This research seeks to extend the utility of applying HSI to human pedestrian detection using the reflective solar spectral range between 400 - 2500 nm. A phenomenological investigation of a novel scheme to differentiate between pedestrians is studied. By applying the basics of detection theory, this research focuses on being able to differentiate between pedestrians, as well as background materials. Specifically, this research explores the likelihood of detecting and differentiating pedestrians based on four defined subregions comprised of the exposed hair, skin, and the fabrics used for shirts and trousers. The scope of this work encompassed detecting a pedestrian of interest outdoors among other pedestrians in an urban environment consisting of a mixture of asphalt, concrete, grass, and trees. Two unique datasets were created during the course of this effort. One dataset was a collection of fully ground-truthed hyperspectral images of pedestrians in an urban environment. A second dataset was a synthetic rendering of the real-world ground truthed pedestrian scene developed using the Digital Imaging and Remote Sensing Image Generation (DIRSIG) model. Subregion separability analysis results, using spectral reflectance data, provided strong evidence that combining the observable spectral features of detectable subregions is a viable means of distinguishing between pedestrians. Further analysis using real-world HSI data demonstrated that the detection and classification of the pedestrian subregions when changes in illumination, location, and background occur within the field of view of a hyperspectral sensor is achievable with a greater than 60% accuracy. In addition to the direct detection and association

  10. Seasonal Effect on Tree Species Classification in an Urban Environment Using Hyperspectral Data, LiDAR, and an Object- Oriented Approach.

    Science.gov (United States)

    Voss, Matthew; Sugumaran, Ramanathan

    2008-05-06

    The objective of the current study was to analyze the seasonal effect on differentiating tree species in an urban environment using multi-temporal hyperspectral data, Light Detection And Ranging (LiDAR) data, and a tree species database collected from the field. Two Airborne Imaging Spectrometer for Applications (AISA) hyperspectral images were collected, covering the Summer and Fall seasons. In order to make both datasets spatially and spectrally compatible, several preprocessing steps, including band reduction and a spatial degradation, were performed. An object-oriented classification was performed on both images using training data collected randomly from the tree species database. The seven dominant tree species (Gleditsia triacanthos, Acer saccharum, Tilia Americana, Quercus palustris, Pinus strobus and Picea glauca) were used in the classification. The results from this analysis did not show any major difference in overall accuracy between the two seasons. Overall accuracy was approximately 57% for the Summer dataset and 56% for the Fall dataset. However, the Fall dataset provided more consistent results for all tree species while the Summer dataset had a few higher individual class accuracies. Further, adding LiDAR into the classification improved the results by 19% for both fall and summer. This is mainly due to the removal of shadow effect and the addition of elevation data to separate low and high vegetation.

  11. Seasonal Effect on Tree Species Classification in an Urban Environment Using Hyperspectral Data, LiDAR, and an Object- Oriented Approach

    Directory of Open Access Journals (Sweden)

    Ramanathan Sugumaran

    2008-05-01

    Full Text Available The objective of the current study was to analyze the seasonal effect on differentiating tree species in an urban environment using multi-temporal hyperspectral data, Light Detection And Ranging (LiDAR data, and a tree species database collected from the field. Two Airborne Imaging Spectrometer for Applications (AISA hyperspectral images were collected, covering the Summer and Fall seasons. In order to make both datasets spatially and spectrally compatible, several preprocessing steps, including band reduction and a spatial degradation, were performed. An object-oriented classification was performed on both images using training data collected randomly from the tree species database. The seven dominant tree species (Gleditsia triacanthos, Acer saccharum, Tilia Americana, Quercus palustris, Pinus strobus and Picea glauca were used in the classification. The results from this analysis did not show any major difference in overall accuracy between the two seasons. Overall accuracy was approximately 57% for the Summer dataset and 56% for the Fall dataset. However, the Fall dataset provided more consistent results for all tree species while the Summer dataset had a few higher individual class accuracies. Further, adding LiDAR into the classification improved the results by 19% for both fall and summer. This is mainly due to the removal of shadow effect and the addition of elevation data to separate low and high vegetation.

  12. Classification of remotely sensed images

    CSIR Research Space (South Africa)

    Dudeni, N

    2008-10-01

    Full Text Available (s)) is the data vector for a pixel located at s θ(s) is an unknown ground class to which pixel s belongs Objective is to classify the pixel at location s to the one of the k clusters Classification of remotely sensed images N. Dudeni, P. Debba...(s) is an unknown ground class to which pixel s belongs Objective is to classify the pixel at location s to the one of the k clusters Classification of remotely sensed images N. Dudeni, P. Debba Introduction to Remote Sensing Introduction to Image...

  13. A Novel Clustering-Based Feature Representation for the Classification of Hyperspectral Imagery

    Directory of Open Access Journals (Sweden)

    Qikai Lu

    2014-06-01

    Full Text Available In this study, a new clustering-based feature extraction algorithm is proposed for the spectral-spatial classification of hyperspectral imagery. The clustering approach is able to group the high-dimensional data into a subspace by mining the salient information and suppressing the redundant information. In this way, the relationship between neighboring pixels, which was hidden in the original data, can be extracted more effectively. Specifically, in the proposed algorithm, a two-step process is adopted to make use of the clustering-based information. A clustering approach is first used to produce the initial clustering map, and, subsequently, a multiscale cluster histogram (MCH is proposed to represent the spatial information around each pixel. In order to evaluate the robustness of the proposed MCH, four clustering techniques are employed to analyze the influence of the clustering methods. Meanwhile, the performance of the MCH is compared to three other widely used spatial features: the gray-level co-occurrence matrix (GLCM, the 3D wavelet texture, and differential morphological profiles (DMPs. The experiments conducted on four well-known hyperspectral datasets verify that the proposed MCH can significantly improve the classification accuracy, and it outperforms other commonly used spatial features.

  14. Semi-supervised hyperspectral classification from a small number of training samples using a co-training approach

    Science.gov (United States)

    Romaszewski, Michał; Głomb, Przemysław; Cholewa, Michał

    2016-11-01

    We present a novel semi-supervised algorithm for classification of hyperspectral data from remote sensors. Our method is inspired by the Tracking-Learning-Detection (TLD) framework, originally applied for tracking objects in a video stream. TLD introduced the co-training approach called P-N learning, making use of two independent 'experts' (or learners) that scored samples in different feature spaces. In a similar fashion, we formulated the hyperspectral classification task as a co-training problem, that can be solved with the P-N learning scheme. Our method uses both spatial and spectral features of data, extending a small set of initial labelled samples during the process of region growing. We show that this approach is stable and achieves very good accuracy even for small training sets. We analyse the algorithm's performance on several publicly available hyperspectral data sets.

  15. Hyperspectral datum classification using kernel method based on mutual information of neighbor bands

    Institute of Scientific and Technical Information of China (English)

    ZHANG Miao; SHEN Yi; WANG Qiang

    2009-01-01

    Under the framework of support vector machines, this paper proposes a new kernel method based on neighbor bands mutual information for hyperspectral datum classification. This algorithm assigns weights to different bands in the kernel function according to the amount of useful information that they contain, which makes the band with more useful informa-tion play more important role in the classification. Our research has shown that the band with greater mutual information between neighbor bands contains more useful information, and hence we use the mutual information of each band and its neighbor bands as the weights of the proposed kernel method. The experimental results show that for the support vector machines based on polynomial and radial basis function, after introducing the proposed kernel function, the average accu-racy is increased more than 1.2% without using any reference map or increasing much more computational time.

  16. Shrinkage-divergence-proximity locally linear embedding algorithm for dimensionality reduction of hyperspectral image

    Institute of Scientific and Technical Information of China (English)

    Qin Luo; Zheng Tian; Zhixiang Zhao

    2008-01-01

    Existing manifold learning algorithms use Euclidean distance to measure the proximity of data points. However, in high-dimensional space, Minkowski metrics are no longer stable because the ratio of distance of nearest and farthest neighbors to a given query is almost unit. It will degrade the performance of manifold learning algorithms when applied to dimensionality reduction of high-dimensional data. We introduce a new distance function named shrinkage-divergence-proximity (SDP) to manifold learning, which is meaningful in any high-dimensional space. An improved locally linear embedding (LLE) algorithm named SDP-LLE is proposed in light of the theoretical result. Experiments are conducted on a hyperspectral data set and an image segmentation data set. Experimental results show that the proposed method can efficiently reduce the dimensionality while getting higher classification accuracy.

  17. [Use of Near-Infrared Hyperspectral Images to Differentiate Architectural Coatings with Different Qualities].

    Science.gov (United States)

    Jiang, Jin-bao; Qiao, Xiao-jun; He, Ru-yan; Tian, Fen-min

    2016-02-01

    Architectural coatings sold in market fall into many categories which mean different models and qualities. The research plans to differentiate different kinds of architectural coatings in quality using hyperspectral technology. Near-Infrared hyperspectral images of four kinds of architectural coatings (in a descending quality order of brand A, B, C, and D) in same color were acquired. The optimal wavelengths were selected at 1283 and 2447 nm to differentiate the four kinds of coatings through ANOVA (Analysis of Variance) method. The band ratio index of R₁₂₈₃/R₂₄₄₇ was built and the results were segmented into the corresponding coatings, and the accuracies of segmentation were compared with that from Maximum Likely Classification (MLC). The results indicated all J-M distances are more than 1.8 except between C and D; the lowest accuracy of 87.54% in segmentation and 95.63% in MLC were both from brand D, and others' accuracies all were over 90% in both ratio index and MLC. Therefore, the ratio index R₁₂₈₃/R₂₄₄₇ could be used to distinguish different kinds of architectural coatings. Also, the research could provide support for identification, quality acceptance, as well as conformity assessment of architectural coatings.

  18. Multi- and hyperspectral UAV imaging system for forest and agriculture applications

    Science.gov (United States)

    Mäkynen, Jussi; Saari, Heikki; Holmlund, Christer; Mannila, Rami; Antila, Tapani

    2012-06-01

    VTT Technical Research Centre of Finland has developed a Fabry-Perot Interferometer (FPI) based hyperspectral imager compatible with light weight UAV (Unmanned Aerial Vehicle) platforms (SPIE Proc. 74741, 8186B2). The FPI based hyperspectral imager was used in a UAV imaging campaign for forest and agriculture tests during the summer 2011 (SPIE Proc. 81743). During these tests high spatial resolution Color-Infrared (CIR) images and hyperspectral images were recorded on separate flights. The spectral bands of the CIR camera were 500 - 580 nm for the green band, 580 - 700 nm for the red band and 700 - 1000 nm for the near infrared band. For the summer 2012 flight campaign a new hyperspectral imager is currently being developed. A custom made CIR camera will also be used. The system which includes both the high spatial resolution Color-Infrared camera and a light weight hyperspectral imager can provide all necessary data with just one UAV flight over the target area. The new UAV imaging system contains a 4 Megapixel CIR camera which is used for the generation of the digital surface models and CIR mosaics. The hyperspectral data can be recorded in the wavelength range 500 - 900 nm at a resolution of 10 - 30 nm at FWHM. The resolution can be selected from approximate values of 10, 15, 20 or 30 nm at FWHM.

  19. A Symmetric Sparse Representation Based Band Selection Method for Hyperspectral Imagery Classification

    Directory of Open Access Journals (Sweden)

    Weiwei Sun

    2016-03-01

    Full Text Available A novel Symmetric Sparse Representation (SSR method has been presented to solve the band selection problem in hyperspectral imagery (HSI classification. The method assumes that the selected bands and the original HSI bands are sparsely represented by each other, i.e., symmetrically represented. The method formulates band selection into a famous problem of archetypal analysis and selects the representative bands by finding the archetypes in the minimal convex hull containing the HSI band points (i.e., one band corresponds to a band point in the high-dimensional feature space. Without any other parameter tuning work except the size of band subset, the SSR optimizes the band selection program using the block-coordinate descent scheme. Four state-of-the-art methods are utilized to make comparisons with the SSR on the Indian Pines and PaviaU HSI datasets. Experimental results illustrate that SSR outperforms all four methods in classification accuracies (i.e., Average Classification Accuracy (ACA and Overall Classification Accuracy (OCA and three quantitative evaluation results (i.e., Average Information Entropy (AIE, Average Correlation Coefficient (ACC and Average Relative Entropy (ARE, whereas it takes the second shortest computational time. Therefore, the proposed SSR is a good alternative method for band selection of HSI classification in realistic applications.

  20. An Ultra-Compact High-Definition Hyperspectral Imaging System Project

    Data.gov (United States)

    National Aeronautics and Space Administration — The innovation is a powerful Ultra-Compact High-Definition Hyperspectral Imaging System (UC-HDHIS) for UAV deployment. UC-HDHIS concurrently acquires pushbroom...

  1. Probability Density and CFAR Threshold Estimation for Hyperspectral Imaging

    Energy Technology Data Exchange (ETDEWEB)

    Clark, G A

    2004-09-21

    The work reported here shows the proof of principle (using a small data set) for a suite of algorithms designed to estimate the probability density function of hyperspectral background data and compute the appropriate Constant False Alarm Rate (CFAR) matched filter decision threshold for a chemical plume detector. Future work will provide a thorough demonstration of the algorithms and their performance with a large data set. The LASI (Large Aperture Search Initiative) Project involves instrumentation and image processing for hyperspectral images of chemical plumes in the atmosphere. The work reported here involves research and development on algorithms for reducing the false alarm rate in chemical plume detection and identification algorithms operating on hyperspectral image cubes. The chemical plume detection algorithms to date have used matched filters designed using generalized maximum likelihood ratio hypothesis testing algorithms [1, 2, 5, 6, 7, 12, 10, 11, 13]. One of the key challenges in hyperspectral imaging research is the high false alarm rate that often results from the plume detector [1, 2]. The overall goal of this work is to extend the classical matched filter detector to apply Constant False Alarm Rate (CFAR) methods to reduce the false alarm rate, or Probability of False Alarm P{sub FA} of the matched filter [4, 8, 9, 12]. A detector designer is interested in minimizing the probability of false alarm while simultaneously maximizing the probability of detection P{sub D}. This is summarized by the Receiver Operating Characteristic Curve (ROC) [10, 11], which is actually a family of curves depicting P{sub D} vs. P{sub FA}parameterized by varying levels of signal to noise (or clutter) ratio (SNR or SCR). Often, it is advantageous to be able to specify a desired P{sub FA} and develop a ROC curve (P{sub D} vs. decision threshold r{sub 0}) for that case. That is the purpose of this work. Specifically, this work develops a set of algorithms and MATLAB

  2. Study of consolidating materials applied on wood by hyperspectral imaging

    Science.gov (United States)

    Bonifazi, G.; Serranti, S.; Capobianco, G.; Agresti, G.; Calienno, L.; Picchio, R.; Lo Monaco, A.; Santamaria, U.; Pelosi, C.

    2016-05-01

    The focus of this study was addressed to investigate the potentiality of HyperSpectral Imaging (HSI) in the monitoring of commercial consolidant products applied on wood samples. Poplar (Populus Sp.) and walnut (Juglans Regia L.) were chosen for the consolidant application. Both traditional and innovative products were selected, based on acrylic, epoxy and aliphatic compounds. Wood samples were stresses by freeze/thaw cycles in order to cause material degradation. Then the consolidants were applied under vacuum. The samples were finally artificially aged for 168 hours in a solar box chamber. The samples were acquired in the SWIR (1000-2500 nm) range by SISUChema XL™ device (Specim, Finland) after 168 hours of irradiation. As comparison, color measurement was also used as economic, simple and noninvasive technique to evaluate the deterioration and consolidation effects on wood. All data were then processed adopting a chemometric approach finalized to define correlation models, HSI based, between consolidating materials, wood species and short time ageing effects.

  3. The Influence of Spectral Wavelength on the Quality of Pansharpened Image Simulated Using Hyperspectral Data

    Science.gov (United States)

    Matsuoka, M.

    2012-07-01

    Preservation of the spectral characteristics in multispectral images is important in the development of pansharpening methods because it affects the accuracy of subsequent applications, such as visual interpretation, land cover classification, and change detection. The combinations of the spectral properties (observation wavelength and width of spectral bands) of multispectral and panchromatic images affect both the spatial and spectral quality of pansharpened images. Therefore, the clarification of the relations between spectral bands and quality of pansharpened image is important for improving our understanding of pansharpening methods, and for developing better schemes for image fusion. This study investigated the influence of the spectral waveband of panchromatic images on the image quality of multispectral (MS) images using simulated images produced from hyperspectral data. Panchromatic images with different spectral band position and multispectral images with degraded spatial resolution were generated from airborne visible/infrared imaging spectrometer (AVIRIS) images and pansharpened using seven methods: additive wavelet intensity, additive wavelet principal component, generalized Laplacian pyramid with spectral distortion minimization, generalized intensity-huesaturation (GIHS) transform, GIHS adaptive, Gram-Schmidt spectral sharpening, and block-based synthetic variable ratio. The pansharpened near-infrared band was visually and statistically compared with the non-degraded image. Wide variation in quality was identified visually within and between methods depending on the spectral wavelengths of the panchromatic images. Quantitative evaluations using three frequently used indices, the correlation coefficient, erreur relative globale adimensionnelle de synthèse (ERGAS), and the Q index, showed the individual behaviors of the pansharpening methods in terms of the spectral similarity in panchromatic and near-infrared, though all methods had similar qualities

  4. A low cost thermal infrared hyperspectral imager for small satellites

    Science.gov (United States)

    Crites, S. T.; Lucey, P. G.; Wright, R.; Garbeil, H.; Horton, K. A.

    2011-06-01

    The traditional model for space-based earth observations involves long mission times, high cost, and long development time. Because of the significant time and monetary investment required, riskier instrument development missions or those with very specific scientific goals are unlikely to successfully obtain funding. However, a niche for earth observations exploiting new technologies in focused, short lifetime missions is opening with the growth of the small satellite market and launch opportunities for these satellites. These low-cost, short-lived missions provide an experimental platform for testing new sensor technologies that may transition to larger, more long-lived platforms. The low costs and short lifetimes also increase acceptable risk to sensors, enabling large decreases in cost using commercial off the shelf (COTS) parts and allowing early-career scientists and engineers to gain experience with these projects. We are building a low-cost long-wave infrared spectral sensor, funded by the NASA Experimental Project to Stimulate Competitive Research program (EPSCOR), to demonstrate the ways in which a university's scientific and instrument development programs can fit into this niche. The sensor is a low-mass, power efficient thermal hyperspectral imager with electronics contained in a pressure vessel to enable the use of COTS electronics, and will be compatible with small satellite platforms. The sensor, called Thermal Hyperspectral Imager (THI), is based on a Sagnac interferometer and uses an uncooled 320x256 microbolometer array. The sensor will collect calibrated radiance data at long-wave infrared (LWIR, 8-14 microns) wavelengths in 230-meter pixels with 20 wavenumber spectral resolution from a 400-km orbit.

  5. A low cost thermal infrared hyperspectral imager for small satellites

    Science.gov (United States)

    Crites, S. T.; Lucey, P. G.; Wright, R.; Garbeil, H.; Horton, K. A.; Wood, M.

    2012-06-01

    The growth of the small satellite market and launch opportunities for these satellites is creating a new niche for earth observations that contrasts with the long mission durations, high costs, and long development times associated with traditional space-based earth observations. Low-cost, short-lived missions made possible by this new approach provide an experimental platform for testing new sensor technologies that may transition to larger, more long-lived platforms. The low costs and short lifetimes also increase acceptable risk to sensors, enabling large decreases in cost using commercial off-the-shelf (COTS) parts and allowing early-career scientists and engineers to gain experience with these projects. We are building a low-cost long-wave infrared spectral sensor, funded by the NASA Experimental Project to Stimulate Competitive Research program (EPSCoR), to demonstrate ways in which a university's scientific and instrument development programs can fit into this niche. The sensor is a low-mass, power-efficient thermal hyperspectral imager with electronics contained in a pressure vessel to enable use of COTS electronics and will be compatible with small satellite platforms. The sensor, called Thermal Hyperspectral Imager (THI), is based on a Sagnac interferometer and uses an uncooled 320x256 microbolometer array. The sensor will collect calibrated radiance data at long-wave infrared (LWIR, 8-14 microns) wavelengths in 230 meter pixels with 20 wavenumber spectral resolution from a 400 km orbit. We are currently in the laboratory and airborne testing stage in order to demonstrate the spectro-radiometric quality of data that the instrument provides.

  6. Target detection algorithm for airborne thermal hyperspectral data

    OpenAIRE

    Marwaha, R.; Kumar, A.; Raju, P.L.N.; Y. V. N. Krishna Murthy

    2014-01-01

    Airborne hyperspectral imaging is constantly being used for classification purpose. But airborne thermal hyperspectral image usually is a challenge for conventional classification approaches. The Telops Hyper-Cam sensor is an interferometer-based imaging system that helps in the spatial and spectral analysis of targets utilizing a single sensor. It is based on the technology of Fourier-transform which yields high spectral resolution and enables high accuracy radiometric calibration. ...

  7. Development and processing of hyperspectral images in optical-electronic remote sensing systems

    Science.gov (United States)

    Kozinov, I. A.; Maltsev, G. N.

    2016-12-01

    The development and processing of three-dimensional images as a "hypercube" of spectral data in hyperspectral optical-electronic remote sensing systems are described in a formalized manner. The correlation identification of observed objects on the basis of spectral features is considered. The criterion for determining of similarity between vectors of recorded and reference spectral images of objects is based on their cross-correlation. Taking into the fact that the total spectral data array recorded by currently applicable hyperspectrometers is excessive for the solution of many issues related to remote sensing of the Earth, this paper proposes a method making it possible to reduce spectral data redundancy by selection of the most informative spectral channels. The essential dimension of the spectral data makes it possible to solve issues related to identification and classification of objects by spectral features through a limited number of very informative spectral channels selected in the areas where the function describing a spectral image of the observed object undergoes well-defined changes in behavior. The algorithm for selection of the most informative spectral channels, which is based on the determination of jump coordinates (major changes) of a spectral image, is substantiated. The selected channels meet the maximum likelihood criterion. The obtained experimental research data on object identification quality with involvement of real hyperspectral data of aerospace Earth remote sensing systems are reported. Five to twenty spectral readouts are needed to provide identification by a limited number of very informative spectral channels. This confirms the idea of existing essential dimensionality of the spectral data.

  8. Reflectance Prediction Modelling for Residual-Based Hyperspectral Image Coding

    Science.gov (United States)

    Xiao, Rui; Gao, Junbin; Bossomaier, Terry

    2016-01-01

    A Hyperspectral (HS) image provides observational powers beyond human vision capability but represents more than 100 times the data compared to a traditional image. To transmit and store the huge volume of an HS image, we argue that a fundamental shift is required from the existing “original pixel intensity”-based coding approaches using traditional image coders (e.g., JPEG2000) to the “residual”-based approaches using a video coder for better compression performance. A modified video coder is required to exploit spatial-spectral redundancy using pixel-level reflectance modelling due to the different characteristics of HS images in their spectral and shape domain of panchromatic imagery compared to traditional videos. In this paper a novel coding framework using Reflectance Prediction Modelling (RPM) in the latest video coding standard High Efficiency Video Coding (HEVC) for HS images is proposed. An HS image presents a wealth of data where every pixel is considered a vector for different spectral bands. By quantitative comparison and analysis of pixel vector distribution along spectral bands, we conclude that modelling can predict the distribution and correlation of the pixel vectors for different bands. To exploit distribution of the known pixel vector, we estimate a predicted current spectral band from the previous bands using Gaussian mixture-based modelling. The predicted band is used as the additional reference band together with the immediate previous band when we apply the HEVC. Every spectral band of an HS image is treated like it is an individual frame of a video. In this paper, we compare the proposed method with mainstream encoders. The experimental results are fully justified by three types of HS dataset with different wavelength ranges. The proposed method outperforms the existing mainstream HS encoders in terms of rate-distortion performance of HS image compression. PMID:27695102

  9. [Prediction of Encapsulation Temperatures of Copolymer Films in Photovoltaic Cells Using Hyperspectral Imaging Techniques and Chemometrics].

    Science.gov (United States)

    Lin, Ping; Chen, Yong-ming; Yao, Zhi-lei

    2015-11-01

    A novel method of combination of the chemometrics and the hyperspectral imaging techniques was presented to detect the temperatures of Ethylene-Vinyl Acetate copolymer (EVA) films in photovoltaic cells during the thermal encapsulation process. Four varieties of the EVA films which had been heated at the temperatures of 128, 132, 142 and 148 °C during the photovoltaic cells production process were used for investigation in this paper. These copolymer encapsulation films were firstly scanned by the hyperspectral imaging equipment (Spectral Imaging Ltd. Oulu, Finland). The scanning band range of hyperspectral equipemnt was set between 904.58 and 1700.01 nm. The hyperspectral dataset of copolymer films was randomly divided into two parts for the training and test purpose. Each type of the training set and test set contained 90 and 10 instances, respectively. The obtained hyperspectral images of EVA films were dealt with by using the ENVI (Exelis Visual Information Solutions, USA) software. The size of region of interest (ROI) of each obtained hyperspectral image of EVA film was set as 150 x 150 pixels. The average of reflectance hyper spectra of all the pixels in the ROI was used as the characteristic curve to represent the instance. There kinds of chemometrics methods including partial least squares regression (PLSR), multi-class support vector machine (SVM) and large margin nearest neighbor (LMNN) were used to correlate the characteristic hyper spectra with the encapsulation temperatures of of copolymer films. The plot of weighted regression coefficients illustrated that both bands of short- and long-wave near infrared hyperspectral data contributed to enhancing the prediction accuracy of the forecast model. Because the attained reflectance hyperspectral data of EVA materials displayed the strong nonlinearity, the prediction performance of linear modeling method of PLSR declined and the prediction precision only reached to 95%. The kernel-based forecast models were

  10. Hyperspectral Imaging for Determining Pigment Contents in Cucumber Leaves in Response to Angular Leaf Spot Disease.

    Science.gov (United States)

    Zhao, Yan-Ru; Li, Xiaoli; Yu, Ke-Qiang; Cheng, Fan; He, Yong

    2016-06-10

    Hyperspectral imaging technique was employed to determine spatial distributions of chlorophyll (Chl), and carotenoid (Car) contents in cucumber leaves in response to angular leaf spot (ALS). Altogether, 196 hyperspectral images of cucumber leaves with five infection severities of ALS were captured by a hyperspectral imaging system in the range of 380-1,030 nm covering 512 wavebands. Mean spectrum were extracted from regions of interest (ROIs) in the hyperspectral images. Partial least square regression (PLSR) models were used to develop quantitative analysis between the spectra and the pigment contents measured by biochemical analyses. In addition, regression coefficients (RCs) in PLSR models were employed to select important wavelengths (IWs) for modelling. It was found that the PLSR models developed by the IWs provided the optimal measurement results with correlation coefficient (R) of prediction of 0.871 and 0.876 for Chl and Car contents, respectively. Finally, Chl and Car distributions in cucumber leaves with the ALS infection were mapped by applying the optimal models pixel-wise to the hyperspectral images. The results proved the feasibility of hyperspectral imaging for visualizing the pigment distributions in cucumber leaves in response to ALS.

  11. Discrimination of Rice Varieties using LS-SVM Classification Algorithms and Hyperspectral Data

    Directory of Open Access Journals (Sweden)

    Jin Xiaming

    2015-03-01

    Full Text Available Fast discrimination of rice varieties plays a key role in the rice processing industry and benefits the management of rice in the supermarket. In order to discriminate rice varieties in a fast and nondestructiv