WorldWideScience

Sample records for learning detection system

  1. IMPROVING CAUSE DETECTION SYSTEMS WITH ACTIVE LEARNING

    Data.gov (United States)

    National Aeronautics and Space Administration — IMPROVING CAUSE DETECTION SYSTEMS WITH ACTIVE LEARNING ISAAC PERSING AND VINCENT NG Abstract. Active learning has been successfully applied to many natural language...

  2. An Android malware detection system based on machine learning

    Science.gov (United States)

    Wen, Long; Yu, Haiyang

    2017-08-01

    The Android smartphone, with its open source character and excellent performance, has attracted many users. However, the convenience of the Android platform also has motivated the development of malware. The traditional method which detects the malware based on the signature is unable to detect unknown applications. The article proposes a machine learning-based lightweight system that is capable of identifying malware on Android devices. In this system we extract features based on the static analysis and the dynamitic analysis, then a new feature selection approach based on principle component analysis (PCA) and relief are presented in the article to decrease the dimensions of the features. After that, a model will be constructed with support vector machine (SVM) for classification. Experimental results show that our system provides an effective method in Android malware detection.

  3. Aberrant Learning Achievement Detection Based on Person-Fit Statistics in Personalized e-Learning Systems

    Science.gov (United States)

    Liu, Ming-Tsung; Yu, Pao-Ta

    2011-01-01

    A personalized e-learning service provides learning content to fit learners' individual differences. Learning achievements are influenced by cognitive as well as non-cognitive factors such as mood, motivation, interest, and personal styles. This paper proposes the Learning Caution Indexes (LCI) to detect aberrant learning patterns. The philosophy…

  4. Thutmose - Investigation of Machine Learning-Based Intrusion Detection Systems

    Science.gov (United States)

    2016-06-01

    monitoring. This analyzed payload is within the application layer of the OSI model . The analysis tries to establish whether or not the payload is...24 3.2.5 Model Drift Experiments...ADVERSARIAL ENVIRONMENTS (SPIE DSS 2014) .................................................. 58 APPENDIX C - EVALUATING MODEL DRIFT IN MACHINE LEARNING

  5. A Multilingual System for Cyberbullying Detection: Arabic Content Detection using Machine Learning

    Directory of Open Access Journals (Sweden)

    Batoul Haidar

    2017-12-01

    Full Text Available With the abundance of Internet and electronic devices bullying has moved its place from schools and backyards into cyberspace; to be now known as Cyberbullying. Cyberbullying is affecting a lot of children around the world, especially Arab countries. Thus concerns from cyberbullying are rising. A lot of research is ongoing with the purpose of diminishing cyberbullying. The current research efforts are focused around detection and mitigation of cyberbullying. Previously, researches dealt with the psychological effects of cyberbullying on the victim and the predator. A lot of research work proposed solutions for detecting cyberbullying in English language and a few more languages, but none till now covered cyberbullying in Arabic language. Several techniques contribute in cyberbullying detection, mainly Machine Learning (ML and Natural Language Processing (NLP. This journal extends on a previous paper to elaborate on a solution for detecting and stopping cyberbullying. It first presents a thorough survey for the previous work done in cyberbullying detection. Then a solution that focuses on detecting cyberbullying in Arabic content is displayed and assessed.

  6. Video copy protection and detection framework (VPD) for e-learning systems

    Science.gov (United States)

    ZandI, Babak; Doustarmoghaddam, Danial; Pour, Mahsa R.

    2013-03-01

    This Article reviews and compares the copyright issues related to the digital video files, which can be categorized as contended based and Digital watermarking copy Detection. Then we describe how to protect a digital video by using a special Video data hiding method and algorithm. We also discuss how to detect the copy right of the file, Based on expounding Direction of the technology of the video copy detection, and Combining with the own research results, brings forward a new video protection and copy detection approach in terms of plagiarism and e-learning systems using the video data hiding technology. Finally we introduce a framework for Video protection and detection in e-learning systems (VPD Framework).

  7. Rapid Object Detection Systems, Utilising Deep Learning and Unmanned Aerial Systems (uas) for Civil Engineering Applications

    Science.gov (United States)

    Griffiths, D.; Boehm, J.

    2018-05-01

    With deep learning approaches now out-performing traditional image processing techniques for image understanding, this paper accesses the potential of rapid generation of Convolutional Neural Networks (CNNs) for applied engineering purposes. Three CNNs are trained on 275 UAS-derived and freely available online images for object detection of 3m2 segments of railway track. These includes two models based on the Faster RCNN object detection algorithm (Resnet and Incpetion-Resnet) as well as the novel onestage Focal Loss network architecture (Retinanet). Model performance was assessed with respect to three accuracy metrics. The first two consisted of Intersection over Union (IoU) with thresholds 0.5 and 0.1. The last assesses accuracy based on the proportion of track covered by object detection proposals against total track length. In under six hours of training (and two hours of manual labelling) the models detected 91.3 %, 83.1 % and 75.6 % of track in the 500 test images acquired from the UAS survey Retinanet, Resnet and Inception-Resnet respectively. We then discuss the potential for such applications of such systems within the engineering field for a range of scenarios.

  8. Metric Learning Method Aided Data-Driven Design of Fault Detection Systems

    Directory of Open Access Journals (Sweden)

    Guoyang Yan

    2014-01-01

    Full Text Available Fault detection is fundamental to many industrial applications. With the development of system complexity, the number of sensors is increasing, which makes traditional fault detection methods lose efficiency. Metric learning is an efficient way to build the relationship between feature vectors with the categories of instances. In this paper, we firstly propose a metric learning-based fault detection framework in fault detection. Meanwhile, a novel feature extraction method based on wavelet transform is used to obtain the feature vector from detection signals. Experiments on Tennessee Eastman (TE chemical process datasets demonstrate that the proposed method has a better performance when comparing with existing methods, for example, principal component analysis (PCA and fisher discriminate analysis (FDA.

  9. Incipient fault detection and identification in process systems using accelerating neural network learning

    International Nuclear Information System (INIS)

    Parlos, A.G.; Muthusami, J.; Atiya, A.F.

    1994-01-01

    The objective of this paper is to present the development and numerical testing of a robust fault detection and identification (FDI) system using artificial neural networks (ANNs), for incipient (slowly developing) faults occurring in process systems. The challenge in using ANNs in FDI systems arises because of one's desire to detect faults of varying severity, faults from noisy sensors, and multiple simultaneous faults. To address these issues, it becomes essential to have a learning algorithm that ensures quick convergence to a high level of accuracy. A recently developed accelerated learning algorithm, namely a form of an adaptive back propagation (ABP) algorithm, is used for this purpose. The ABP algorithm is used for the development of an FDI system for a process composed of a direct current motor, a centrifugal pump, and the associated piping system. Simulation studies indicate that the FDI system has significantly high sensitivity to incipient fault severity, while exhibiting insensitivity to sensor noise. For multiple simultaneous faults, the FDI system detects the fault with the predominant signature. The major limitation of the developed FDI system is encountered when it is subjected to simultaneous faults with similar signatures. During such faults, the inherent limitation of pattern-recognition-based FDI methods becomes apparent. Thus, alternate, more sophisticated FDI methods become necessary to address such problems. Even though the effectiveness of pattern-recognition-based FDI methods using ANNs has been demonstrated, further testing using real-world data is necessary

  10. VLSI Design of SVM-Based Seizure Detection System With On-Chip Learning Capability.

    Science.gov (United States)

    Feng, Lichen; Li, Zunchao; Wang, Yuanfa

    2018-02-01

    Portable automatic seizure detection system is very convenient for epilepsy patients to carry. In order to make the system on-chip trainable with high efficiency and attain high detection accuracy, this paper presents a very large scale integration (VLSI) design based on the nonlinear support vector machine (SVM). The proposed design mainly consists of a feature extraction (FE) module and an SVM module. The FE module performs the three-level Daubechies discrete wavelet transform to fit the physiological bands of the electroencephalogram (EEG) signal and extracts the time-frequency domain features reflecting the nonstationary signal properties. The SVM module integrates the modified sequential minimal optimization algorithm with the table-driven-based Gaussian kernel to enable efficient on-chip learning. The presented design is verified on an Altera Cyclone II field-programmable gate array and tested using the two publicly available EEG datasets. Experiment results show that the designed VLSI system improves the detection accuracy and training efficiency.

  11. SQL injection detection system

    OpenAIRE

    Vargonas, Vytautas

    2017-01-01

    SQL injection detection system Programmers do not always ensure security of developed systems. That is why it is important to look for solutions outside being reliant on developers. In this work SQL injection detection system is proposed. The system analyzes HTTP request parameters and detects intrusions. It is based on unsupervised machine learning. Trained by regular request data system detects outlier user parameters. Since training is not reliant on previous knowledge of SQL injections, t...

  12. Efficacy of a Deep Learning System for Detecting Glaucomatous Optic Neuropathy Based on Color Fundus Photographs.

    Science.gov (United States)

    Li, Zhixi; He, Yifan; Keel, Stuart; Meng, Wei; Chang, Robert T; He, Mingguang

    2018-03-02

    To assess the performance of a deep learning algorithm for detecting referable glaucomatous optic neuropathy (GON) based on color fundus photographs. A deep learning system for the classification of GON was developed for automated classification of GON on color fundus photographs. We retrospectively included 48 116 fundus photographs for the development and validation of a deep learning algorithm. This study recruited 21 trained ophthalmologists to classify the photographs. Referable GON was defined as vertical cup-to-disc ratio of 0.7 or more and other typical changes of GON. The reference standard was made until 3 graders achieved agreement. A separate validation dataset of 8000 fully gradable fundus photographs was used to assess the performance of this algorithm. The area under receiver operator characteristic curve (AUC) with sensitivity and specificity was applied to evaluate the efficacy of the deep learning algorithm detecting referable GON. In the validation dataset, this deep learning system achieved an AUC of 0.986 with sensitivity of 95.6% and specificity of 92.0%. The most common reasons for false-negative grading (n = 87) were GON with coexisting eye conditions (n = 44 [50.6%]), including pathologic or high myopia (n = 37 [42.6%]), diabetic retinopathy (n = 4 [4.6%]), and age-related macular degeneration (n = 3 [3.4%]). The leading reason for false-positive results (n = 480) was having other eye conditions (n = 458 [95.4%]), mainly including physiologic cupping (n = 267 [55.6%]). Misclassification as false-positive results amidst a normal-appearing fundus occurred in only 22 eyes (4.6%). A deep learning system can detect referable GON with high sensitivity and specificity. Coexistence of high or pathologic myopia is the most common cause resulting in false-negative results. Physiologic cupping and pathologic myopia were the most common reasons for false-positive results. Copyright © 2018 American Academy of Ophthalmology. Published by

  13. A Multilingual System for Cyberbullying Detection: Arabic Content Detection using Machine Learning

    OpenAIRE

    Batoul Haidar; Maroun Chamoun; Ahmed Serhrouchni

    2017-01-01

    With the abundance of Internet and electronic devices bullying has moved its place from schools and backyards into cyberspace; to be now known as Cyberbullying. Cyberbullying is affecting a lot of children around the world, especially Arab countries. Thus concerns from cyberbullying are rising. A lot of research is ongoing with the purpose of diminishing cyberbullying. The current research efforts are focused around detection and mitigation of cyberbullying. Previously, researches dealt with ...

  14. Deep Learning-Based Gaze Detection System for Automobile Drivers Using a NIR Camera Sensor

    Directory of Open Access Journals (Sweden)

    Rizwan Ali Naqvi

    2018-02-01

    Full Text Available A paradigm shift is required to prevent the increasing automobile accident deaths that are mostly due to the inattentive behavior of drivers. Knowledge of gaze region can provide valuable information regarding a driver’s point of attention. Accurate and inexpensive gaze classification systems in cars can improve safe driving. However, monitoring real-time driving behaviors and conditions presents some challenges: dizziness due to long drives, extreme lighting variations, glasses reflections, and occlusions. Past studies on gaze detection in cars have been chiefly based on head movements. The margin of error in gaze detection increases when drivers gaze at objects by moving their eyes without moving their heads. To solve this problem, a pupil center corneal reflection (PCCR-based method has been considered. However, the error of accurately detecting the pupil center and corneal reflection center is increased in a car environment due to various environment light changes, reflections on glasses surface, and motion and optical blurring of captured eye image. In addition, existing PCCR-based methods require initial user calibration, which is difficult to perform in a car environment. To address this issue, we propose a deep learning-based gaze detection method using a near-infrared (NIR camera sensor considering driver head and eye movement that does not require any initial user calibration. The proposed system is evaluated on our self-constructed database as well as on open Columbia gaze dataset (CAVE-DB. The proposed method demonstrated greater accuracy than the previous gaze classification methods.

  15. Deep Learning-Based Gaze Detection System for Automobile Drivers Using a NIR Camera Sensor.

    Science.gov (United States)

    Naqvi, Rizwan Ali; Arsalan, Muhammad; Batchuluun, Ganbayar; Yoon, Hyo Sik; Park, Kang Ryoung

    2018-02-03

    A paradigm shift is required to prevent the increasing automobile accident deaths that are mostly due to the inattentive behavior of drivers. Knowledge of gaze region can provide valuable information regarding a driver's point of attention. Accurate and inexpensive gaze classification systems in cars can improve safe driving. However, monitoring real-time driving behaviors and conditions presents some challenges: dizziness due to long drives, extreme lighting variations, glasses reflections, and occlusions. Past studies on gaze detection in cars have been chiefly based on head movements. The margin of error in gaze detection increases when drivers gaze at objects by moving their eyes without moving their heads. To solve this problem, a pupil center corneal reflection (PCCR)-based method has been considered. However, the error of accurately detecting the pupil center and corneal reflection center is increased in a car environment due to various environment light changes, reflections on glasses surface, and motion and optical blurring of captured eye image. In addition, existing PCCR-based methods require initial user calibration, which is difficult to perform in a car environment. To address this issue, we propose a deep learning-based gaze detection method using a near-infrared (NIR) camera sensor considering driver head and eye movement that does not require any initial user calibration. The proposed system is evaluated on our self-constructed database as well as on open Columbia gaze dataset (CAVE-DB). The proposed method demonstrated greater accuracy than the previous gaze classification methods.

  16. Developing an Automated Machine Learning Marine Oil Spill Detection System with Synthetic Aperture Radar

    Science.gov (United States)

    Pinales, J. C.; Graber, H. C.; Hargrove, J. T.; Caruso, M. J.

    2016-02-01

    Previous studies have demonstrated the ability to detect and classify marine hydrocarbon films with spaceborne synthetic aperture radar (SAR) imagery. The dampening effects of hydrocarbon discharges on small surface capillary-gravity waves renders the ocean surface "radar dark" compared with the standard wind-borne ocean surfaces. Given the scope and impact of events like the Deepwater Horizon oil spill, the need for improved, automated and expedient monitoring of hydrocarbon-related marine anomalies has become a pressing and complex issue for governments and the extraction industry. The research presented here describes the development, training, and utilization of an algorithm that detects marine oil spills in an automated, semi-supervised manner, utilizing X-, C-, or L-band SAR data as the primary input. Ancillary datasets include related radar-borne variables (incidence angle, etc.), environmental data (wind speed, etc.) and textural descriptors. Shapefiles produced by an experienced human-analyst served as targets (validation) during the training portion of the investigation. Training and testing datasets were chosen for development and assessment of algorithm effectiveness as well as optimal conditions for oil detection in SAR data. The algorithm detects oil spills by following a 3-step methodology: object detection, feature extraction, and classification. Previous oil spill detection and classification methodologies such as machine learning algorithms, artificial neural networks (ANN), and multivariate classification methods like partial least squares-discriminant analysis (PLS-DA) are evaluated and compared. Statistical, transform, and model-based image texture techniques, commonly used for object mapping directly or as inputs for more complex methodologies, are explored to determine optimal textures for an oil spill detection system. The influence of the ancillary variables is explored, with a particular focus on the role of strong vs. weak wind forcing.

  17. A framework for detection of malicious software in Android handheld systems using machine learning techniques

    OpenAIRE

    Torregrosa García, Blas

    2015-01-01

    The present study aims at designing and developing new approaches to detect malicious applications in Android-based devices. More precisely, MaLDroide (Machine Learning-based Detector for Android malware), a framework for detection of Android malware based on machine learning techniques, is introduced here. It is devised to identify malicious applications. Este trabajo tiene como objetivo el diseño y el desarrollo de nuevas formas de detección de aplicaciones maliciosas en los dispositivos...

  18. Intrusion detection system using Online Sequence Extreme Learning Machine (OS-ELM) in advanced metering infrastructure of smart grid.

    Science.gov (United States)

    Li, Yuancheng; Qiu, Rixuan; Jing, Sitong

    2018-01-01

    Advanced Metering Infrastructure (AMI) realizes a two-way communication of electricity data through by interconnecting with a computer network as the core component of the smart grid. Meanwhile, it brings many new security threats and the traditional intrusion detection method can't satisfy the security requirements of AMI. In this paper, an intrusion detection system based on Online Sequence Extreme Learning Machine (OS-ELM) is established, which is used to detecting the attack in AMI and carrying out the comparative analysis with other algorithms. Simulation results show that, compared with other intrusion detection methods, intrusion detection method based on OS-ELM is more superior in detection speed and accuracy.

  19. Intrusion detection system using Online Sequence Extreme Learning Machine (OS-ELM in advanced metering infrastructure of smart grid.

    Directory of Open Access Journals (Sweden)

    Yuancheng Li

    Full Text Available Advanced Metering Infrastructure (AMI realizes a two-way communication of electricity data through by interconnecting with a computer network as the core component of the smart grid. Meanwhile, it brings many new security threats and the traditional intrusion detection method can't satisfy the security requirements of AMI. In this paper, an intrusion detection system based on Online Sequence Extreme Learning Machine (OS-ELM is established, which is used to detecting the attack in AMI and carrying out the comparative analysis with other algorithms. Simulation results show that, compared with other intrusion detection methods, intrusion detection method based on OS-ELM is more superior in detection speed and accuracy.

  20. JACoW Model learning algorithms for anomaly detection in CERN control systems

    CERN Document Server

    Tilaro, Filippo; Gonzalez-Berges, Manuel; Roshchin, Mikhail; Varela, Fernando

    2018-01-01

    The CERN automation infrastructure consists of over 600 heterogeneous industrial control systems with around 45 million deployed sensors, actuators and control objects. Therefore, it is evident that the monitoring of such huge system represents a challenging and complex task. This paper describes three different mathematical approaches that have been designed and developed to detect anomalies in any of the CERN control systems. Specifically, one of these algorithms is purely based on expert knowledge; the other two mine the historical generated data to create a simple model of the system; this model is then used to detect faulty sensors measurements. The presented methods can be categorized as dynamic unsupervised anomaly detection; “dynamic” since the behaviour of the system and the evolution of its attributes are observed and changing in time. They are “unsupervised” because we are trying to predict faulty events without examples in the data history. So, the described strategies involve monitoring t...

  1. Simultaneous detection and classification of breast masses in digital mammograms via a deep learning YOLO-based CAD system.

    Science.gov (United States)

    Al-Masni, Mohammed A; Al-Antari, Mugahed A; Park, Jeong-Min; Gi, Geon; Kim, Tae-Yeon; Rivera, Patricio; Valarezo, Edwin; Choi, Mun-Taek; Han, Seung-Moo; Kim, Tae-Seong

    2018-04-01

    Automatic detection and classification of the masses in mammograms are still a big challenge and play a crucial role to assist radiologists for accurate diagnosis. In this paper, we propose a novel Computer-Aided Diagnosis (CAD) system based on one of the regional deep learning techniques, a ROI-based Convolutional Neural Network (CNN) which is called You Only Look Once (YOLO). Although most previous studies only deal with classification of masses, our proposed YOLO-based CAD system can handle detection and classification simultaneously in one framework. The proposed CAD system contains four main stages: preprocessing of mammograms, feature extraction utilizing deep convolutional networks, mass detection with confidence, and finally mass classification using Fully Connected Neural Networks (FC-NNs). In this study, we utilized original 600 mammograms from Digital Database for Screening Mammography (DDSM) and their augmented mammograms of 2,400 with the information of the masses and their types in training and testing our CAD. The trained YOLO-based CAD system detects the masses and then classifies their types into benign or malignant. Our results with five-fold cross validation tests show that the proposed CAD system detects the mass location with an overall accuracy of 99.7%. The system also distinguishes between benign and malignant lesions with an overall accuracy of 97%. Our proposed system even works on some challenging breast cancer cases where the masses exist over the pectoral muscles or dense regions. Copyright © 2018 Elsevier B.V. All rights reserved.

  2. Integration of a framework with a learning management system for detection, assessment and assistance of university students with reading difficulties

    Directory of Open Access Journals (Sweden)

    Carolina Mejía Corredor

    2015-12-01

    Full Text Available Rev.esc.adm.neg Dyslexia is a common learning disability in Spanish-speaking university students, and requires special attention from higher educational institutions in order to support affected individuals during their learning process. In previous studies, a framework to detect, assess and assist university students with reading difficulties related to dyslexia was developed. In this paper, the integration of this framework with a Learning Management System (LMS is presented. Two case studies were performed to test the functionality and the usability of this integration. The first case study was carried out with 20 students, while the second one with four teachers. The results show that both students and teachers were satisfied with the integration performed in Moodle.ce, among others.

  3. A Pathological Brain Detection System based on Extreme Learning Machine Optimized by Bat Algorithm.

    Science.gov (United States)

    Lu, Siyuan; Qiu, Xin; Shi, Jianping; Li, Na; Lu, Zhi-Hai; Chen, Peng; Yang, Meng-Meng; Liu, Fang-Yuan; Jia, Wen-Juan; Zhang, Yudong

    2017-01-01

    It is beneficial to classify brain images as healthy or pathological automatically, because 3D brain images can generate so much information which is time consuming and tedious for manual analysis. Among various 3D brain imaging techniques, magnetic resonance (MR) imaging is the most suitable for brain, and it is now widely applied in hospitals, because it is helpful in the four ways of diagnosis, prognosis, pre-surgical, and postsurgical procedures. There are automatic detection methods; however they suffer from low accuracy. Therefore, we proposed a novel approach which employed 2D discrete wavelet transform (DWT), and calculated the entropies of the subbands as features. Then, a bat algorithm optimized extreme learning machine (BA-ELM) was trained to identify pathological brains from healthy controls. A 10x10-fold cross validation was performed to evaluate the out-of-sample performance. The method achieved a sensitivity of 99.04%, a specificity of 93.89%, and an overall accuracy of 98.33% over 132 MR brain images. The experimental results suggest that the proposed approach is accurate and robust in pathological brain detection. Copyright© Bentham Science Publishers; For any queries, please email at epub@benthamscience.org.

  4. An overload behavior detection system for engineering transport vehicles based on deep learning

    Science.gov (United States)

    Zhou, Libo; Wu, Gang

    2018-04-01

    This paper builds an overloaded truck detect system called ITMD to help traffic department automatically identify the engineering transport vehicles (commonly known as `dirt truck') in CCTV and determine whether the truck is overloaded or not. We build the ITMD system based on the Single Shot MultiBox Detector (SSD) model. By constructing the image dataset of the truck and adjusting hyper-parameters of the original SSD neural network, we successfully trained a basic network model which the ITMD system depends on. The basic ITMD system achieves 83.01% mAP on classifying overload/non-overload truck, which is a not bad result. Still, some shortcomings of basic ITMD system have been targeted to enhance: it is easy for the ITMD system to misclassify other similar vehicle as truck. In response to this problem, we optimized the basic ITMD system, which effectively reduced basic model's false recognition rate. The optimized ITMD system achieved 86.18% mAP on the test set, which is better than the 83.01% mAP of the basic ITMD system.

  5. A machine learning system for automated whole-brain seizure detection

    Directory of Open Access Journals (Sweden)

    P. Fergus

    2016-01-01

    Full Text Available Epilepsy is a chronic neurological condition that affects approximately 70 million people worldwide. Characterised by sudden bursts of excess electricity in the brain, manifesting as seizures, epilepsy is still not well understood when compared with other neurological disorders. Seizures often happen unexpectedly and attempting to predict them has been a research topic for the last 30 years. Electroencephalograms have been integral to these studies, as the recordings that they produce can capture the brain’s electrical signals. The diagnosis of epilepsy is usually made by a neurologist, but can be difficult to make in the early stages. Supporting para-clinical evidence obtained from magnetic resonance imaging and electroencephalography may enable clinicians to make a diagnosis of epilepsy and instigate treatment earlier. However, electroencephalogram capture and interpretation is time consuming and can be expensive due to the need for trained specialists to perform the interpretation. Automated detection of correlates of seizure activity generalised across different regions of the brain and across multiple subjects may be a solution. This paper explores this idea further and presents a supervised machine learning approach that classifies seizure and non-seizure records using an open dataset containing 342 records (171 seizures and 171 non-seizures. Our approach posits a new method for generalising seizure detection across different subjects without prior knowledge about the focal point of seizures. Our results show an improvement on existing studies with 88% for sensitivity, 88% for specificity and 93% for the area under the curve, with a 12% global error, using the k-NN classifier.

  6. Development and Experimental Evaluation of Machine-Learning Techniques for an Intelligent Hairy Scalp Detection System

    Directory of Open Access Journals (Sweden)

    Wei-Chien Wang

    2018-05-01

    Full Text Available Deep learning has become the most popular research subject in the fields of artificial intelligence (AI and machine learning. In October 2013, MIT Technology Review commented that deep learning was a breakthrough technology. Deep learning has made progress in voice and image recognition, image classification, and natural language processing. Prior to deep learning, decision tree, linear discriminant analysis (LDA, support vector machines (SVM, k-nearest neighbors algorithm (K-NN, and ensemble learning were popular in solving classification problems. In this paper, we applied the previously mentioned and deep learning techniques to hairy scalp images. Hairy scalp problems are usually diagnosed by non-professionals in hair salons, and people with such problems may be advised by these non-professionals. Additionally, several common scalp problems are similar; therefore, non-experts may provide incorrect diagnoses. Hence, scalp problems have worsened. In this work, we implemented and compared the deep-learning method, the ImageNet-VGG-f model Bag of Words (BOW, with machine-learning classifiers, and histogram of oriented gradients (HOG/pyramid histogram of oriented gradients (PHOG with machine-learning classifiers. The tools from the classification learner apps were used for hairy scalp image classification. The results indicated that deep learning can achieve an accuracy of 89.77% when the learning rate is 1 × 10−4, and this accuracy is far higher than those achieved by BOW with SVM (80.50% and PHOG with SVM (53.0%.

  7. An Improved Pathological Brain Detection System Based on Two-Dimensional PCA and Evolutionary Extreme Learning Machine.

    Science.gov (United States)

    Nayak, Deepak Ranjan; Dash, Ratnakar; Majhi, Banshidhar

    2017-12-07

    Pathological brain detection has made notable stride in the past years, as a consequence many pathological brain detection systems (PBDSs) have been proposed. But, the accuracy of these systems still needs significant improvement in order to meet the necessity of real world diagnostic situations. In this paper, an efficient PBDS based on MR images is proposed that markedly improves the recent results. The proposed system makes use of contrast limited adaptive histogram equalization (CLAHE) to enhance the quality of the input MR images. Thereafter, two-dimensional PCA (2DPCA) strategy is employed to extract the features and subsequently, a PCA+LDA approach is used to generate a compact and discriminative feature set. Finally, a new learning algorithm called MDE-ELM is suggested that combines modified differential evolution (MDE) and extreme learning machine (ELM) for segregation of MR images as pathological or healthy. The MDE is utilized to optimize the input weights and hidden biases of single-hidden-layer feed-forward neural networks (SLFN), whereas an analytical method is used for determining the output weights. The proposed algorithm performs optimization based on both the root mean squared error (RMSE) and norm of the output weights of SLFNs. The suggested scheme is benchmarked on three standard datasets and the results are compared against other competent schemes. The experimental outcomes show that the proposed scheme offers superior results compared to its counterparts. Further, it has been noticed that the proposed MDE-ELM classifier obtains better accuracy with compact network architecture than conventional algorithms.

  8. Machine-learning-based real-bogus system for the HSC-SSP moving object detection pipeline

    Science.gov (United States)

    Lin, Hsing-Wen; Chen, Ying-Tung; Wang, Jen-Hung; Wang, Shiang-Yu; Yoshida, Fumi; Ip, Wing-Huen; Miyazaki, Satoshi; Terai, Tsuyoshi

    2018-01-01

    Machine-learning techniques are widely applied in many modern optical sky surveys, e.g., Pan-STARRS1, PTF/iPTF, and the Subaru/Hyper Suprime-Cam survey, to reduce human intervention in data verification. In this study, we have established a machine-learning-based real-bogus system to reject false detections in the Subaru/Hyper-Suprime-Cam Strategic Survey Program (HSC-SSP) source catalog. Therefore, the HSC-SSP moving object detection pipeline can operate more effectively due to the reduction of false positives. To train the real-bogus system, we use stationary sources as the real training set and "flagged" data as the bogus set. The training set contains 47 features, most of which are photometric measurements and shape moments generated from the HSC image reduction pipeline (hscPipe). Our system can reach a true positive rate (tpr) ˜96% with a false positive rate (fpr) ˜1% or tpr ˜99% at fpr ˜5%. Therefore, we conclude that stationary sources are decent real training samples, and using photometry measurements and shape moments can reject false positives effectively.

  9. Unsupervised machine-learning method for improving the performance of ambulatory fall-detection systems

    Directory of Open Access Journals (Sweden)

    Yuwono Mitchell

    2012-02-01

    Full Text Available Abstract Background Falls can cause trauma, disability and death among older people. Ambulatory accelerometer devices are currently capable of detecting falls in a controlled environment. However, research suggests that most current approaches can tend to have insufficient sensitivity and specificity in non-laboratory environments, in part because impacts can be experienced as part of ordinary daily living activities. Method We used a waist-worn wireless tri-axial accelerometer combined with digital signal processing, clustering and neural network classifiers. The method includes the application of Discrete Wavelet Transform, Regrouping Particle Swarm Optimization, Gaussian Distribution of Clustered Knowledge and an ensemble of classifiers including a multilayer perceptron and Augmented Radial Basis Function (ARBF neural networks. Results Preliminary testing with 8 healthy individuals in a home environment yields 98.6% sensitivity to falls and 99.6% specificity for routine Activities of Daily Living (ADL data. Single ARB and MLP classifiers were compared with a combined classifier. The combined classifier offers the greatest sensitivity, with a slight reduction in specificity for routine ADL and an increased specificity for exercise activities. In preliminary tests, the approach achieves 100% sensitivity on in-group falls, 97.65% on out-group falls, 99.33% specificity on routine ADL, and 96.59% specificity on exercise ADL. Conclusion The pre-processing and feature-extraction steps appear to simplify the signal while successfully extracting the essential features that are required to characterize a fall. The results suggest this combination of classifiers can perform better than MLP alone. Preliminary testing suggests these methods may be useful for researchers who are attempting to improve the performance of ambulatory fall-detection systems.

  10. Targeting safety improvements through identification of incident origination and detection in a near-miss incident learning system

    International Nuclear Information System (INIS)

    Novak, Avrey; Nyflot, Matthew J.; Ermoian, Ralph P.; Jordan, Loucille E.; Sponseller, Patricia A.; Kane, Gabrielle M.; Ford, Eric C.; Zeng, Jing

    2016-01-01

    Purpose: Radiation treatment planning involves a complex workflow that has multiple potential points of vulnerability. This study utilizes an incident reporting system to identify the origination and detection points of near-miss errors, in order to guide their departmental safety improvement efforts. Previous studies have examined where errors arise, but not where they are detected or applied a near-miss risk index (NMRI) to gauge severity. Methods: From 3/2012 to 3/2014, 1897 incidents were analyzed from a departmental incident learning system. All incidents were prospectively reviewed weekly by a multidisciplinary team and assigned a NMRI score ranging from 0 to 4 reflecting potential harm to the patient (no potential harm to potential critical harm). Incidents were classified by point of incident origination and detection based on a 103-step workflow. The individual steps were divided among nine broad workflow categories (patient assessment, imaging for radiation therapy (RT) planning, treatment planning, pretreatment plan review, treatment delivery, on-treatment quality management, post-treatment completion, equipment/software quality management, and other). The average NMRI scores of incidents originating or detected within each broad workflow area were calculated. Additionally, out of 103 individual process steps, 35 were classified as safety barriers, the process steps whose primary function is to catch errors. The safety barriers which most frequently detected incidents were identified and analyzed. Finally, the distance between event origination and detection was explored by grouping events by the number of broad workflow area events passed through before detection, and average NMRI scores were compared. Results: Near-miss incidents most commonly originated within treatment planning (33%). However, the incidents with the highest average NMRI scores originated during imaging for RT planning (NMRI = 2.0, average NMRI of all events = 1.5), specifically

  11. Targeting safety improvements through identification of incident origination and detection in a near-miss incident learning system

    Energy Technology Data Exchange (ETDEWEB)

    Novak, Avrey; Nyflot, Matthew J.; Ermoian, Ralph P.; Jordan, Loucille E.; Sponseller, Patricia A.; Kane, Gabrielle M.; Ford, Eric C.; Zeng, Jing, E-mail: jzeng13@uw.edu [Department of Radiation Oncology, University of Washington Medical Center, 1959 NE Pacific Street, Campus Box 356043, Seattle, Washington 98195 (United States)

    2016-05-15

    Purpose: Radiation treatment planning involves a complex workflow that has multiple potential points of vulnerability. This study utilizes an incident reporting system to identify the origination and detection points of near-miss errors, in order to guide their departmental safety improvement efforts. Previous studies have examined where errors arise, but not where they are detected or applied a near-miss risk index (NMRI) to gauge severity. Methods: From 3/2012 to 3/2014, 1897 incidents were analyzed from a departmental incident learning system. All incidents were prospectively reviewed weekly by a multidisciplinary team and assigned a NMRI score ranging from 0 to 4 reflecting potential harm to the patient (no potential harm to potential critical harm). Incidents were classified by point of incident origination and detection based on a 103-step workflow. The individual steps were divided among nine broad workflow categories (patient assessment, imaging for radiation therapy (RT) planning, treatment planning, pretreatment plan review, treatment delivery, on-treatment quality management, post-treatment completion, equipment/software quality management, and other). The average NMRI scores of incidents originating or detected within each broad workflow area were calculated. Additionally, out of 103 individual process steps, 35 were classified as safety barriers, the process steps whose primary function is to catch errors. The safety barriers which most frequently detected incidents were identified and analyzed. Finally, the distance between event origination and detection was explored by grouping events by the number of broad workflow area events passed through before detection, and average NMRI scores were compared. Results: Near-miss incidents most commonly originated within treatment planning (33%). However, the incidents with the highest average NMRI scores originated during imaging for RT planning (NMRI = 2.0, average NMRI of all events = 1.5), specifically

  12. Interorganizational learning systems

    DEFF Research Database (Denmark)

    Hjalager, Anne-Mette

    1999-01-01

    The occurrence of organizational and interorganizational learning processes is not only the result of management endeavors. Industry structures and market related issues have substantial spill-over effects. The article reviews literature, and it establishes a learning model in which elements from...... organizational environments are included into a systematic conceptual framework. The model allows four types of learning to be identified: P-learning (professional/craft systems learning), T-learning (technology embedded learning), D-learning (dualistic learning systems, where part of the labor force is exclude...... from learning), and S-learning (learning in social networks or clans). The situation related to service industries illustrates the typology....

  13. Detection and Learning of Unexpected Behaviors of Systems of Dynamical Systems by Using the Q2 Abstractions

    Science.gov (United States)

    2017-11-01

    implemented and detected such types of undesirable emergent behaviors in a multi-UAV system. Next sections present the methods and approaches we...so, we use qualitative methods to study the behaviors exhibited by EBS. The use of such methods will lower the computational complexity of the...behaviors. Figure 43. Number of Comparisons for N Qualitative Partitions Another aspect of our research was to gain more understanding of the

  14. Automatic Earthquake Detection by Active Learning

    Science.gov (United States)

    Bergen, K.; Beroza, G. C.

    2017-12-01

    In recent years, advances in machine learning have transformed fields such as image recognition, natural language processing and recommender systems. Many of these performance gains have relied on the availability of large, labeled data sets to train high-accuracy models; labeled data sets are those for which each sample includes a target class label, such as waveforms tagged as either earthquakes or noise. Earthquake seismologists are increasingly leveraging machine learning and data mining techniques to detect and analyze weak earthquake signals in large seismic data sets. One of the challenges in applying machine learning to seismic data sets is the limited labeled data problem; learning algorithms need to be given examples of earthquake waveforms, but the number of known events, taken from earthquake catalogs, may be insufficient to build an accurate detector. Furthermore, earthquake catalogs are known to be incomplete, resulting in training data that may be biased towards larger events and contain inaccurate labels. This challenge is compounded by the class imbalance problem; the events of interest, earthquakes, are infrequent relative to noise in continuous data sets, and many learning algorithms perform poorly on rare classes. In this work, we investigate the use of active learning for automatic earthquake detection. Active learning is a type of semi-supervised machine learning that uses a human-in-the-loop approach to strategically supplement a small initial training set. The learning algorithm incorporates domain expertise through interaction between a human expert and the algorithm, with the algorithm actively posing queries to the user to improve detection performance. We demonstrate the potential of active machine learning to improve earthquake detection performance with limited available training data.

  15. A Visual Detection Learning Model

    Science.gov (United States)

    Beard, Bettina L.; Ahumada, Albert J., Jr.; Trejo, Leonard (Technical Monitor)

    1998-01-01

    Our learning model has memory templates representing the target-plus-noise and noise-alone stimulus sets. The best correlating template determines the response. The correlations and the feedback participate in the additive template updating rule. The model can predict the relative thresholds for detection in random, fixed and twin noise.

  16. Cross-Language Plagiarism Detection System Using Latent Semantic Analysis and Learning Vector Quantization

    Directory of Open Access Journals (Sweden)

    Anak Agung Putri Ratna

    2017-06-01

    Full Text Available Computerized cross-language plagiarism detection has recently become essential. With the scarcity of scientific publications in Bahasa Indonesia, many Indonesian authors frequently consult publications in English in order to boost the quantity of scientific publications in Bahasa Indonesia (which is currently rising. Due to the syntax disparity between Bahasa Indonesia and English, most of the existing methods for automated cross-language plagiarism detection do not provide satisfactory results. This paper analyses the probability of developing Latent Semantic Analysis (LSA for a computerized cross-language plagiarism detector for two languages with different syntax. To improve performance, various alterations in LSA are suggested. By using a linear vector quantization (LVQ classifier in the LSA and taking into account the Frobenius norm, output has reached up to 65.98% in accuracy. The results of the experiments showed that the best accuracy achieved is 87% with a document size of 6 words, and the document definition size must be kept below 10 words in order to maintain high accuracy. Additionally, based on experimental results, this paper suggests utilizing the frequency occurrence method as opposed to the binary method for the term–document matrix construction.

  17. Portable modular detection system

    Science.gov (United States)

    Brennan, James S [Rodeo, CA; Singh, Anup [Danville, CA; Throckmorton, Daniel J [Tracy, CA; Stamps, James F [Livermore, CA

    2009-10-13

    Disclosed herein are portable and modular detection devices and systems for detecting electromagnetic radiation, such as fluorescence, from an analyte which comprises at least one optical element removably attached to at least one alignment rail. Also disclosed are modular detection devices and systems having an integrated lock-in amplifier and spatial filter and assay methods using the portable and modular detection devices.

  18. System modeling based on machine learning for anomaly detection and predictive maintenance in industrial plants

    OpenAIRE

    Kroll, Björn; Schaffranek, David; Schriegel, Sebastian; Niggemann, Oliver

    2014-01-01

    Electricity, water or air are some Industrial energy carriers which are struggling under the prices of primary energy carriers. The European Union for example used more 20.000.000 GWh electricity in 2011 based on the IEA Report [1]. Cyber Physical Production Systems (CPPS) are able to reduce this amount, but they also help to increase the efficiency of machines above expectations which results in a more cost efficient production. Especially in the field of improving industrial plants, one of ...

  19. Using transfer learning to detect galaxy mergers

    Science.gov (United States)

    Ackermann, Sandro; Schawinksi, Kevin; Zhang, Ce; Weigel, Anna K.; Turp, M. Dennis

    2018-05-01

    We investigate the use of deep convolutional neural networks (deep CNNs) for automatic visual detection of galaxy mergers. Moreover, we investigate the use of transfer learning in conjunction with CNNs, by retraining networks first trained on pictures of everyday objects. We test the hypothesis that transfer learning is useful for improving classification performance for small training sets. This would make transfer learning useful for finding rare objects in astronomical imaging datasets. We find that these deep learning methods perform significantly better than current state-of-the-art merger detection methods based on nonparametric systems like CAS and GM20. Our method is end-to-end and robust to image noise and distortions; it can be applied directly without image preprocessing. We also find that transfer learning can act as a regulariser in some cases, leading to better overall classification accuracy (p = 0.02). Transfer learning on our full training set leads to a lowered error rate from 0.0381 down to 0.0321, a relative improvement of 15%. Finally, we perform a basic sanity-check by creating a merger sample with our method, and comparing with an already existing, manually created merger catalogue in terms of colour-mass distribution and stellar mass function.

  20. Intrusion detection system elements

    International Nuclear Information System (INIS)

    Eaton, M.J.; Mangan, D.L.

    1980-09-01

    This report highlights elements required for an intrusion detection system and discusses problems which can be encountered in attempting to make the elements effective. Topics discussed include: sensors, both for exterior detection and interior detection; alarm assessment systems, with the discussion focused on video assessment; and alarm reporting systems, including alarm communication systems and dislay/console considerations. Guidance on careful planning and design of a new or to-be-improved system is presented

  1. Development of an e-learning system for teaching endoscopists how to diagnose early gastric cancer: basic principles for improving early detection.

    Science.gov (United States)

    Yao, Kenshi; Uedo, Noriya; Muto, Manabu; Ishikawa, Hideki

    2017-03-01

    We developed an internet e-learning system in order to improve the ability of endoscopists to diagnose gastric cancer at an early stage. The efficacy of this system at expanding knowledge and providing invaluable experience regarding the endoscopic detection of early gastric cancer was demonstrated through an international multicenter randomized controlled trial. However, the contents of the system have not yet been fully described in the literature. Accordingly, we herein introduce the contents and their principles, which comprise three main subjects: technique, knowledge, and experience. Since all the e-learning contents and principles are based on conventional white-light endoscopy alone, which is commonly available throughout the world, they should provide a good reference point for any endoscopist who wishes to devise learning materials and guidelines for improving their own clinical practice.

  2. Interior intrusion detection systems

    Energy Technology Data Exchange (ETDEWEB)

    Rodriguez, J.R.; Matter, J.C. (Sandia National Labs., Albuquerque, NM (United States)); Dry, B. (BE, Inc., Barnwell, SC (United States))

    1991-10-01

    The purpose of this NUREG is to present technical information that should be useful to NRC licensees in designing interior intrusion detection systems. Interior intrusion sensors are discussed according to their primary application: boundary-penetration detection, volumetric detection, and point protection. Information necessary for implementation of an effective interior intrusion detection system is presented, including principles of operation, performance characteristics and guidelines for design, procurement, installation, testing, and maintenance. A glossary of sensor data terms is included. 36 figs., 6 tabs.

  3. Interior intrusion detection systems

    International Nuclear Information System (INIS)

    Rodriguez, J.R.; Matter, J.C.; Dry, B.

    1991-10-01

    The purpose of this NUREG is to present technical information that should be useful to NRC licensees in designing interior intrusion detection systems. Interior intrusion sensors are discussed according to their primary application: boundary-penetration detection, volumetric detection, and point protection. Information necessary for implementation of an effective interior intrusion detection system is presented, including principles of operation, performance characteristics and guidelines for design, procurement, installation, testing, and maintenance. A glossary of sensor data terms is included. 36 figs., 6 tabs

  4. Using Dictionary Pair Learning for Seizure Detection.

    Science.gov (United States)

    Ma, Xin; Yu, Nana; Zhou, Weidong

    2018-02-13

    Automatic seizure detection is extremely important in the monitoring and diagnosis of epilepsy. The paper presents a novel method based on dictionary pair learning (DPL) for seizure detection in the long-term intracranial electroencephalogram (EEG) recordings. First, for the EEG data, wavelet filtering and differential filtering are applied, and the kernel function is performed to make the signal linearly separable. In DPL, the synthesis dictionary and analysis dictionary are learned jointly from original training samples with alternating minimization method, and sparse coefficients are obtained by using of linear projection instead of costly [Formula: see text]-norm or [Formula: see text]-norm optimization. At last, the reconstructed residuals associated with seizure and nonseizure sub-dictionary pairs are calculated as the decision values, and the postprocessing is performed for improving the recognition rate and reducing the false detection rate of the system. A total of 530[Formula: see text]h from 20 patients with 81 seizures were used to evaluate the system. Our proposed method has achieved an average segment-based sensitivity of 93.39%, specificity of 98.51%, and event-based sensitivity of 96.36% with false detection rate of 0.236/h.

  5. Recommender Systems for Learning

    CERN Document Server

    Manouselis, Nikos; Verbert, Katrien; Duval, Erik

    2013-01-01

    Technology enhanced learning (TEL) aims to design, develop and test sociotechnical innovations that will support and enhance learning practices of both individuals and organisations. It is therefore an application domain that generally covers technologies that support all forms of teaching and learning activities. Since information retrieval (in terms of searching for relevant learning resources to support teachers or learners) is a pivotal activity in TEL, the deployment of recommender systems has attracted increased interest. This brief attempts to provide an introduction to recommender systems for TEL settings, as well as to highlight their particularities compared to recommender systems for other application domains.

  6. Semiconductor radiation detection systems

    CERN Document Server

    2010-01-01

    Covers research in semiconductor detector and integrated circuit design in the context of medical imaging using ionizing radiation. This book explores other applications of semiconductor radiation detection systems in security applications such as luggage scanning, dirty bomb detection and border control.

  7. Learning Content Management Systems

    Directory of Open Access Journals (Sweden)

    Tache JURUBESCU

    2008-01-01

    Full Text Available The paper explains the evolution of e-Learning and related concepts and tools and its connection with other concepts such as Knowledge Management, Human Resources Management, Enterprise Resource Planning, and Information Technology. The paper also distinguished Learning Content Management Systems from Learning Management Systems and Content Management Systems used for general web-based content. The newest Learning Content Management System, very expensive and yet very little implemented is one of the best tools that helps us to cope with the realities of the 21st Century in what learning concerns. The debates over how beneficial one or another system is for an organization, can be driven by costs involved, efficiency envisaged, and availability of the product on the market.

  8. Facial detection using deep learning

    Science.gov (United States)

    Sharma, Manik; Anuradha, J.; Manne, H. K.; Kashyap, G. S. C.

    2017-11-01

    In the recent past, we have observed that Facebook has developed an uncanny ability to recognize people in photographs. Previously, we had to tag people in photos by clicking on them and typing their name. Now as soon as we upload a photo, Facebook tags everyone on its own. Facebook can recognize faces with 98% accuracy which is pretty much as good as humans can do. This technology is called Face Detection. Face detection is a popular topic in biometrics. We have surveillance cameras in public places for video capture as well as security purposes. The main advantages of this algorithm over other are uniqueness and approval. We need speed and accuracy to identify. But face detection is really a series of several related problems: First, look at a picture and find all the faces in it. Second, focus on each face and understand that even if a face is turned in a weird direction or in bad lighting, it is still the same person. Third select features which can be used to identify each face uniquely like size of the eyes, face etc. Finally, compare these features to data we have to find the person name. As a human, your brain is wired to do all of this automatically and instantly. In fact, humans are too good at recognizing faces. Computers are not capable of this kind of high-level generalization, so we must teach them how to do each step in this process separately. The growth of face detection is largely driven by growing applications such as credit card verification, surveillance video images, authentication for banking and security system access.

  9. System Detects Vibrational Instabilities

    Science.gov (United States)

    Bozeman, Richard J., Jr.

    1990-01-01

    Sustained vibrations at two critical frequencies trigger diagnostic response or shutdown. Vibration-analyzing electronic system detects instabilities of combustion in rocket engine. Controls pulse-mode firing of engine and identifies vibrations above threshold amplitude at 5.9 and/or 12kHz. Adapted to other detection and/or control schemes involving simultaneous real-time detection of signals above or below preset amplitudes at two or more specified frequencies. Potential applications include rotating machinery and encoders and decoders in security systems.

  10. π0 detection system

    International Nuclear Information System (INIS)

    Suzuki, Yoichiro

    1977-01-01

    A π-zero meson detection system used for the measurement of charge exchange reaction is described. The detection of π-zero is made by observing the coincidence events of two gamma-ray emission following the decay of π-zero meson. The angles of the emitted gamma-rays are detected with a wire spark chamber system, and the energies of the gamma-rays are measured with hodoscope type lead glass Cherenkov counters. In front of the π-zero counter system, a lead converter is set, and the incident gamma-rays convert to electron positron pairs, which can be detected with the wire spark chambers. The system is a multi-track detection system. The high voltage pulser of the wire spark chamber system is a charge line thyratron pulser, and the chamber itself is a transmission line type. Read-out can be made by a mag-line system. Wave forms and efficiencies were measured. The three-track efficiency was about 90% by the condenser method and 95% by the charge line method. (Kato, T.)

  11. Solar system fault detection

    Science.gov (United States)

    Farrington, R.B.; Pruett, J.C. Jr.

    1984-05-14

    A fault detecting apparatus and method are provided for use with an active solar system. The apparatus provides an indication as to whether one or more predetermined faults have occurred in the solar system. The apparatus includes a plurality of sensors, each sensor being used in determining whether a predetermined condition is present. The outputs of the sensors are combined in a pre-established manner in accordance with the kind of predetermined faults to be detected. Indicators communicate with the outputs generated by combining the sensor outputs to give the user of the solar system and the apparatus an indication as to whether a predetermined fault has occurred. Upon detection and indication of any predetermined fault, the user can take appropriate corrective action so that the overall reliability and efficiency of the active solar system are increased.

  12. Multivariate Statistics and Supervised Learning for Predictive Detection of Unintentional Islanding in Grid-Tied Solar PV Systems

    Directory of Open Access Journals (Sweden)

    Shashank Vyas

    2016-01-01

    Full Text Available Integration of solar photovoltaic (PV generation with power distribution networks leads to many operational challenges and complexities. Unintentional islanding is one of them which is of rising concern given the steady increase in grid-connected PV power. This paper builds up on an exploratory study of unintentional islanding on a modeled radial feeder having large PV penetration. Dynamic simulations, also run in real time, resulted in exploration of unique potential causes of creation of accidental islands. The resulting voltage and current data underwent dimensionality reduction using principal component analysis (PCA which formed the basis for the application of Q statistic control charts for detecting the anomalous currents that could island the system. For reducing the false alarm rate of anomaly detection, Kullback-Leibler (K-L divergence was applied on the principal component projections which concluded that Q statistic based approach alone is not reliable for detection of the symptoms liable to cause unintentional islanding. The obtained data was labeled and a K-nearest neighbor (K-NN binomial classifier was then trained for identification and classification of potential islanding precursors from other power system transients. The three-phase short-circuit fault case was successfully identified as statistically different from islanding symptoms.

  13. Idaho Explosives Detection System

    International Nuclear Information System (INIS)

    Reber, Edward L.; Blackwood, Larry G.; Edwards, Andrew J.; Jewell, J. Keith; Rohde, Kenneth W.; Seabury, Edward H.; Klinger, Jeffery B.

    2005-01-01

    The Idaho Explosives Detection System was developed at the Idaho National Laboratory (INL) to respond to threats imposed by delivery trucks potentially carrying explosives into military bases. A full-scale prototype system has been built and is currently undergoing testing. The system consists of two racks, one on each side of a subject vehicle. Each rack includes a neutron generator and an array of NaI detectors. The two neutron generators are pulsed and synchronized. A laptop computer controls the entire system. The control software is easily operable by minimally trained staff. The system was developed to detect explosives in a medium size truck within a 5-min measurement time. System performance was successfully demonstrated with explosives at the INL in June 2004 and at Andrews Air Force Base in July 2004

  14. Idaho Explosives Detection System

    Energy Technology Data Exchange (ETDEWEB)

    Reber, Edward L. [Idaho National Laboratory, 2525 N. Freemont Ave., Idaho Falls, ID 83415-2114 (United States)]. E-mail: reber@inel.gov; Blackwood, Larry G. [Idaho National Laboratory, 2525 N. Freemont Ave., Idaho Falls, ID 83415-2114 (United States); Edwards, Andrew J. [Idaho National Laboratory, 2525 N. Freemont Ave., Idaho Falls, ID 83415-2114 (United States); Jewell, J. Keith [Idaho National Laboratory, 2525 N. Freemont Ave., Idaho Falls, ID 83415-2114 (United States); Rohde, Kenneth W. [Idaho National Laboratory, 2525 N. Freemont Ave., Idaho Falls, ID 83415-2114 (United States); Seabury, Edward H. [Idaho National Laboratory, 2525 N. Freemont Ave., Idaho Falls, ID 83415-2114 (United States); Klinger, Jeffery B. [Idaho National Laboratory, 2525 N. Freemont Ave., Idaho Falls, ID 83415-2114 (United States)

    2005-12-15

    The Idaho Explosives Detection System was developed at the Idaho National Laboratory (INL) to respond to threats imposed by delivery trucks potentially carrying explosives into military bases. A full-scale prototype system has been built and is currently undergoing testing. The system consists of two racks, one on each side of a subject vehicle. Each rack includes a neutron generator and an array of NaI detectors. The two neutron generators are pulsed and synchronized. A laptop computer controls the entire system. The control software is easily operable by minimally trained staff. The system was developed to detect explosives in a medium size truck within a 5-min measurement time. System performance was successfully demonstrated with explosives at the INL in June 2004 and at Andrews Air Force Base in July 2004.

  15. Machine learning systems

    Energy Technology Data Exchange (ETDEWEB)

    Forsyth, R

    1984-05-01

    With the dramatic rise of expert systems has come a renewed interest in the fuel that drives them-knowledge. For it is specialist knowledge which gives expert systems their power. But extracting knowledge from human experts in symbolic form has proved arduous and labour-intensive. So the idea of machine learning is enjoying a renaissance. Machine learning is any automatic improvement in the performance of a computer system over time, as a result of experience. Thus a learning algorithm seeks to do one or more of the following: cover a wider range of problems, deliver more accurate solutions, obtain answers more cheaply, and simplify codified knowledge. 6 references.

  16. Machine Learning Methods for Attack Detection in the Smart Grid.

    Science.gov (United States)

    Ozay, Mete; Esnaola, Inaki; Yarman Vural, Fatos Tunay; Kulkarni, Sanjeev R; Poor, H Vincent

    2016-08-01

    Attack detection problems in the smart grid are posed as statistical learning problems for different attack scenarios in which the measurements are observed in batch or online settings. In this approach, machine learning algorithms are used to classify measurements as being either secure or attacked. An attack detection framework is provided to exploit any available prior knowledge about the system and surmount constraints arising from the sparse structure of the problem in the proposed approach. Well-known batch and online learning algorithms (supervised and semisupervised) are employed with decision- and feature-level fusion to model the attack detection problem. The relationships between statistical and geometric properties of attack vectors employed in the attack scenarios and learning algorithms are analyzed to detect unobservable attacks using statistical learning methods. The proposed algorithms are examined on various IEEE test systems. Experimental analyses show that machine learning algorithms can detect attacks with performances higher than attack detection algorithms that employ state vector estimation methods in the proposed attack detection framework.

  17. Proximity detection system underground

    Energy Technology Data Exchange (ETDEWEB)

    Denis Kent [Mine Site Technologies (Australia)

    2008-04-15

    Mine Site Technologies (MST) with the support ACARP and Xstrata Coal NSW, as well as assistance from Centennial Coal, has developed a Proximity Detection System to proof of concept stage as per plan. The basic aim of the project was to develop a system to reduce the risk of the people coming into contact with vehicles in an uncontrolled manner (i.e. being 'run over'). The potential to extend the developed technology into other areas, such as controls for vehicle-vehicle collisions and restricting access of vehicle or people into certain zones (e.g. non FLP vehicles into Hazardous Zones/ERZ) was also assessed. The project leveraged off MST's existing Intellectual Property and experience gained with our ImPact TRACKER tagging technology, allowing the development to be fast tracked. The basic concept developed uses active RFID Tags worn by miners underground to be detected by vehicle mounted Readers. These Readers in turn provide outputs that can be used to alert a driver (e.g. by light and/or audible alarm) that a person (Tag) approaching within their vicinity. The prototype/test kit developed proved the concept and technology, the four main components being: Active RFID Tags to send out signals for detection by vehicle mounted receivers; Receiver electronics to detect RFID Tags approaching within the vicinity of the unit to create a long range detection system (60 m to 120 m); A transmitting/exciter device to enable inner detection zone (within 5 m to 20 m); and A software/hardware device to process & log incoming Tags reads and create certain outputs. Tests undertaken in the laboratory and at a number of mine sites, confirmed the technology path taken could form the basis of a reliable Proximity Detection/Alert System.

  18. Network anomaly detection a machine learning perspective

    CERN Document Server

    Bhattacharyya, Dhruba Kumar

    2013-01-01

    With the rapid rise in the ubiquity and sophistication of Internet technology and the accompanying growth in the number of network attacks, network intrusion detection has become increasingly important. Anomaly-based network intrusion detection refers to finding exceptional or nonconforming patterns in network traffic data compared to normal behavior. Finding these anomalies has extensive applications in areas such as cyber security, credit card and insurance fraud detection, and military surveillance for enemy activities. Network Anomaly Detection: A Machine Learning Perspective presents mach

  19. Gas detection system

    International Nuclear Information System (INIS)

    Allan, C.J.; Bayly, J.G.

    1975-01-01

    The gas detection system provides for the effective detection of gas leaks over a large area. It includes a laser which has a laser line corresponding to an absorption line of the gas to be detected. A He-Xe laser scans a number of retroreflectors which are strategically located around a D 2 O plant to detect H 2 S leaks. The reflected beam is focused by a telescope, filtered, and passed into an infrared detector. The laser may be made to emit two frequencies, one of which corresponds with an H 2 S absorption line; or it may be modulated on and off the H 2 S absorption line. The relative amplitude of the absorbed light will be a measure of the H 2 S present

  20. Remote detection system

    International Nuclear Information System (INIS)

    Nixon, K.V.; France, S.W.; Garcia, C.; Hastings, R.D.

    1981-05-01

    A newly designed remote detection system has been developed at Los Alamos that allows the collection of high-resolution gamma-ray spectra and neutron data from a remote location. The system consists of the remote unit and a command unit. The remote unit collects data in a potentially hostile environment while the operator controls the unit by either radio or wire link from a safe position. Both units are battery powered and are housed in metal carrying cases

  1. Remote Voice Detection System

    National Research Council Canada - National Science Library

    Blackmon, Fletcher A

    2007-01-01

    A device and system to remotely detect vocalizations of speech. The skin located on the throat region of a speaking person or a reflective layer on the skin on the throat region vibrates in response to vocalizations of speech by the person...

  2. Moving Sources Detection System

    International Nuclear Information System (INIS)

    Coulon, Romain; Kondrasovs, Vladimir; Boudergui, Karim; Normand, Stephane

    2013-06-01

    To monitor radioactivity passing through a pipe or in a given container such as a train or a truck, radiation detection systems are commonly employed. These detectors could be used in a network set along the source track to increase the overall detection efficiency. However detection methods are based on counting statistics analysis. The method usually implemented consists in trigging an alarm when an individual signal rises over a threshold initially estimated in regards to the natural background signal. The detection efficiency is then proportional to the number of detectors in use, due to the fact that each sensor is taken as a standalone sensor. A new approach is presented in this paper taking into account the temporal periodicity of the signals taken by all distributed sensors as a whole. This detection method is not based only on counting statistics but also on the temporal series analysis aspect. Therefore, a specific algorithm is then developed in our lab for this kind of applications and shows a significant improvement, especially in terms of detection efficiency and false alarms reduction. We also plan on extracting information from the source vector. This paper presents the theoretical approach and some preliminary results obtain in our laboratory. (authors)

  3. Learning to Detect Traffic Signs

    DEFF Research Database (Denmark)

    Møgelmose, Andreas; Trivedi, Mohan M.; Moeslund, Thomas B.

    2012-01-01

    This study compares the performance of sign detection based on synthetic training data to the performance of detection based on real-world training images. Viola-Jones detectors are created for 4 different traffic signs with both synthetic and real data, and varying numbers of training samples. T...

  4. Lung Nodule Detection via Deep Reinforcement Learning

    Directory of Open Access Journals (Sweden)

    Issa Ali

    2018-04-01

    Full Text Available Lung cancer is the most common cause of cancer-related death globally. As a preventive measure, the United States Preventive Services Task Force (USPSTF recommends annual screening of high risk individuals with low-dose computed tomography (CT. The resulting volume of CT scans from millions of people will pose a significant challenge for radiologists to interpret. To fill this gap, computer-aided detection (CAD algorithms may prove to be the most promising solution. A crucial first step in the analysis of lung cancer screening results using CAD is the detection of pulmonary nodules, which may represent early-stage lung cancer. The objective of this work is to develop and validate a reinforcement learning model based on deep artificial neural networks for early detection of lung nodules in thoracic CT images. Inspired by the AlphaGo system, our deep learning algorithm takes a raw CT image as input and views it as a collection of states, and output a classification of whether a nodule is present or not. The dataset used to train our model is the LIDC/IDRI database hosted by the lung nodule analysis (LUNA challenge. In total, there are 888 CT scans with annotations based on agreement from at least three out of four radiologists. As a result, there are 590 individuals having one or more nodules, and 298 having none. Our training results yielded an overall accuracy of 99.1% [sensitivity 99.2%, specificity 99.1%, positive predictive value (PPV 99.1%, negative predictive value (NPV 99.2%]. In our test, the results yielded an overall accuracy of 64.4% (sensitivity 58.9%, specificity 55.3%, PPV 54.2%, and NPV 60.0%. These early results show promise in solving the major issue of false positives in CT screening of lung nodules, and may help to save unnecessary follow-up tests and expenditures.

  5. Anomaly detection in wide area network mesh using two machine learning anomaly detection algorithms

    OpenAIRE

    Zhang, James; Vukotic, Ilija; Gardner, Robert

    2018-01-01

    Anomaly detection is the practice of identifying items or events that do not conform to an expected behavior or do not correlate with other items in a dataset. It has previously been applied to areas such as intrusion detection, system health monitoring, and fraud detection in credit card transactions. In this paper, we describe a new method for detecting anomalous behavior over network performance data, gathered by perfSONAR, using two machine learning algorithms: Boosted Decision Trees (BDT...

  6. Adapting Parameterized Motions using Iterative Learning and Online Collision Detection

    DEFF Research Database (Denmark)

    Laursen, Johan Sund; Sørensen, Lars Carøe; Schultz, Ulrik Pagh

    2018-01-01

    utilizing Gaussian Process learning. This allows for motion parameters to be optimized using real world trials which incorporate all uncertainties inherent in the assembly process without requiring advanced robot and sensor setups. The result is a simple and straightforward system which helps the user...... automatically find robust and uncertainty-tolerant motions. We present experiments for an assembly case showing both detection and learning in the real world and how these combine to a robust robot system....

  7. SIP threats detection system

    OpenAIRE

    Vozňák, Miroslav; Řezáč, Filip

    2010-01-01

    The paper deals with detection of threats in IP telephony, the authors developed a penetration testing system that is able to check up the level of protection from security threats in IP telephony. The SIP server is a key komponent of VoIP infrastructure and often becomes the aim of attacks and providers have to ensure the appropriate level of security. We have developed web-based penetration system which is able to check the SIP server if can face to the most common attacks.The d...

  8. Quantum machine learning for quantum anomaly detection

    Science.gov (United States)

    Liu, Nana; Rebentrost, Patrick

    2018-04-01

    Anomaly detection is used for identifying data that deviate from "normal" data patterns. Its usage on classical data finds diverse applications in many important areas such as finance, fraud detection, medical diagnoses, data cleaning, and surveillance. With the advent of quantum technologies, anomaly detection of quantum data, in the form of quantum states, may become an important component of quantum applications. Machine-learning algorithms are playing pivotal roles in anomaly detection using classical data. Two widely used algorithms are the kernel principal component analysis and the one-class support vector machine. We find corresponding quantum algorithms to detect anomalies in quantum states. We show that these two quantum algorithms can be performed using resources that are logarithmic in the dimensionality of quantum states. For pure quantum states, these resources can also be logarithmic in the number of quantum states used for training the machine-learning algorithm. This makes these algorithms potentially applicable to big quantum data applications.

  9. Gas Flow Detection System

    Science.gov (United States)

    Moss, Thomas; Ihlefeld, Curtis; Slack, Barry

    2010-01-01

    This system provides a portable means to detect gas flow through a thin-walled tube without breaking into the tubing system. The flow detection system was specifically designed to detect flow through two parallel branches of a manifold with only one inlet and outlet, and is a means for verifying a space shuttle program requirement that saves time and reduces the risk of flight hardware damage compared to the current means of requirement verification. The prototype Purge Vent and Drain Window Cavity Conditioning System (PVD WCCS) Flow Detection System consists of a heater and a temperature-sensing thermistor attached to a piece of Velcro to be attached to each branch of a WCCS manifold for the duration of the requirement verification test. The heaters and thermistors are connected to a shielded cable and then to an electronics enclosure, which contains the power supplies, relays, and circuit board to provide power, signal conditioning, and control. The electronics enclosure is then connected to a commercial data acquisition box to provide analog to digital conversion as well as digital control. This data acquisition box is then connected to a commercial laptop running a custom application created using National Instruments LabVIEW. The operation of the PVD WCCS Flow Detection System consists of first attaching a heater/thermistor assembly to each of the two branches of one manifold while there is no flow through the manifold. Next, the software application running on the laptop is used to turn on the heaters and to monitor the manifold branch temperatures. When the system has reached thermal equilibrium, the software application s graphical user interface (GUI) will indicate that the branch temperatures are stable. The operator can then physically open the flow control valve to initiate the test flow of gaseous nitrogen (GN2) through the manifold. Next, the software user interface will be monitored for stable temperature indications when the system is again at

  10. Single particle detecting telescope system

    International Nuclear Information System (INIS)

    Yamamoto, I.; Tomiyama, T.; Iga, Y.; Komatsubara, T.; Kanada, M.; Yamashita, Y.; Wada, T.; Furukawa, S.

    1981-01-01

    We constructed the single particle detecting telescope system for detecting a fractionally charged particle. The telescope consists of position detecting counters, wall-less multi-cell chambers, single detecting circuits and microcomputer system as data I/0 processor. Especially, a frequency of double particle is compared the case of the single particle detecting with the case of an ordinary measurement

  11. Arc fault detection system

    Science.gov (United States)

    Jha, K.N.

    1999-05-18

    An arc fault detection system for use on ungrounded or high-resistance-grounded power distribution systems is provided which can be retrofitted outside electrical switchboard circuits having limited space constraints. The system includes a differential current relay that senses a current differential between current flowing from secondary windings located in a current transformer coupled to a power supply side of a switchboard, and a total current induced in secondary windings coupled to a load side of the switchboard. When such a current differential is experienced, a current travels through a operating coil of the differential current relay, which in turn opens an upstream circuit breaker located between the switchboard and a power supply to remove the supply of power to the switchboard. 1 fig.

  12. Arc fault detection system

    Science.gov (United States)

    Jha, Kamal N.

    1999-01-01

    An arc fault detection system for use on ungrounded or high-resistance-grounded power distribution systems is provided which can be retrofitted outside electrical switchboard circuits having limited space constraints. The system includes a differential current relay that senses a current differential between current flowing from secondary windings located in a current transformer coupled to a power supply side of a switchboard, and a total current induced in secondary windings coupled to a load side of the switchboard. When such a current differential is experienced, a current travels through a operating coil of the differential current relay, which in turn opens an upstream circuit breaker located between the switchboard and a power supply to remove the supply of power to the switchboard.

  13. Counterbalanced radiation detection system

    International Nuclear Information System (INIS)

    Platz, W.

    1987-01-01

    A counterbalanced radiation detection system is described comprising: (a) a stand; (b) a first radiation detector; (c) a first radiation detector arm means for tiltably connecting the first radiation detector with the stand; (d) a second radiation detector; (e) a second radiation detector arm means for tiltably connecting the second radiation detector with the stand, whereby the tilting angles of the radiation detector arm means define a distance between the radiation detectors; and (f) a torque transforming means connected between the first and second radiation detector arm means for transforming the torque created by one of the radiation detectors in a sense opposed to the torque created by the other radiation detector

  14. Radiation detection system

    International Nuclear Information System (INIS)

    Haeuszer, F.A.

    1976-01-01

    A circuit is disclosed that detects radiation transients and provides a clamping signal in response to each transient. The clamping signal is present from the time the transient rises above a given threshold level and for a known duration thereafter. The system includes radiation sensors, a blocking oscillator that generates a pulse in response to each sensor signal, and an output pulse duration control circuit. The oscillator pulses are fed simultaneously to the output pulse duration control circuit and to an OR gate, the output of which comprises the system output. The output pulse duration is controlled by the time required to magnetize a magnetic core to saturation in first one direction and then the other

  15. Ferret Workflow Anomaly Detection System

    National Research Council Canada - National Science Library

    Smith, Timothy J; Bryant, Stephany

    2005-01-01

    The Ferret workflow anomaly detection system project 2003-2004 has provided validation and anomaly detection in accredited workflows in secure knowledge management systems through the use of continuous, automated audits...

  16. Supervised Learning for Dynamical System Learning.

    Science.gov (United States)

    Hefny, Ahmed; Downey, Carlton; Gordon, Geoffrey J

    2015-01-01

    Recently there has been substantial interest in spectral methods for learning dynamical systems. These methods are popular since they often offer a good tradeoff between computational and statistical efficiency. Unfortunately, they can be difficult to use and extend in practice: e.g., they can make it difficult to incorporate prior information such as sparsity or structure. To address this problem, we present a new view of dynamical system learning: we show how to learn dynamical systems by solving a sequence of ordinary supervised learning problems, thereby allowing users to incorporate prior knowledge via standard techniques such as L 1 regularization. Many existing spectral methods are special cases of this new framework, using linear regression as the supervised learner. We demonstrate the effectiveness of our framework by showing examples where nonlinear regression or lasso let us learn better state representations than plain linear regression does; the correctness of these instances follows directly from our general analysis.

  17. Introduction to detection systems

    DEFF Research Database (Denmark)

    Larsen, Jan

    Presentation of the information processing pipleline for detection including discussing of various issues and the use of mathematical modeling. A simple example of detection a signal in noise illustrated that simple modeling outperforms human visual and auditory perception. Particiants are going...... to discuss issues in detection which is followed by an auditory object recognition exercise. The results of the exercise and its relation to issues in the information processing pipleline is also discussed....

  18. Deep learning for automated drivetrain fault detection

    DEFF Research Database (Denmark)

    Bach-Andersen, Martin; Rømer-Odgaard, Bo; Winther, Ole

    2018-01-01

    A novel data-driven deep-learning system for large-scale wind turbine drivetrain monitoring applications is presented. It uses convolutional neural network processing on complex vibration signal inputs. The system is demonstrated to learn successfully from the actions of human diagnostic experts...... the fleet-wide diagnostic model performance. The analysis also explores the time dependence of the diagnostic performance, providing a detailed view of the timeliness and accuracy of the diagnostic outputs across the different architectures. Deep architectures are shown to outperform the human analyst...... as well as shallow-learning architectures, and the results demonstrate that when applied in a large-scale monitoring system, machine intelligence is now able to handle some of the most challenging diagnostic tasks related to wind turbines....

  19. Supervised Learning for Detection of Duplicates in Genomic Sequence Databases.

    Directory of Open Access Journals (Sweden)

    Qingyu Chen

    Full Text Available First identified as an issue in 1996, duplication in biological databases introduces redundancy and even leads to inconsistency when contradictory information appears. The amount of data makes purely manual de-duplication impractical, and existing automatic systems cannot detect duplicates as precisely as can experts. Supervised learning has the potential to address such problems by building automatic systems that learn from expert curation to detect duplicates precisely and efficiently. While machine learning is a mature approach in other duplicate detection contexts, it has seen only preliminary application in genomic sequence databases.We developed and evaluated a supervised duplicate detection method based on an expert curated dataset of duplicates, containing over one million pairs across five organisms derived from genomic sequence databases. We selected 22 features to represent distinct attributes of the database records, and developed a binary model and a multi-class model. Both models achieve promising performance; under cross-validation, the binary model had over 90% accuracy in each of the five organisms, while the multi-class model maintains high accuracy and is more robust in generalisation. We performed an ablation study to quantify the impact of different sequence record features, finding that features derived from meta-data, sequence identity, and alignment quality impact performance most strongly. The study demonstrates machine learning can be an effective additional tool for de-duplication of genomic sequence databases. All Data are available as described in the supplementary material.

  20. Supervised Learning for Detection of Duplicates in Genomic Sequence Databases.

    Science.gov (United States)

    Chen, Qingyu; Zobel, Justin; Zhang, Xiuzhen; Verspoor, Karin

    2016-01-01

    First identified as an issue in 1996, duplication in biological databases introduces redundancy and even leads to inconsistency when contradictory information appears. The amount of data makes purely manual de-duplication impractical, and existing automatic systems cannot detect duplicates as precisely as can experts. Supervised learning has the potential to address such problems by building automatic systems that learn from expert curation to detect duplicates precisely and efficiently. While machine learning is a mature approach in other duplicate detection contexts, it has seen only preliminary application in genomic sequence databases. We developed and evaluated a supervised duplicate detection method based on an expert curated dataset of duplicates, containing over one million pairs across five organisms derived from genomic sequence databases. We selected 22 features to represent distinct attributes of the database records, and developed a binary model and a multi-class model. Both models achieve promising performance; under cross-validation, the binary model had over 90% accuracy in each of the five organisms, while the multi-class model maintains high accuracy and is more robust in generalisation. We performed an ablation study to quantify the impact of different sequence record features, finding that features derived from meta-data, sequence identity, and alignment quality impact performance most strongly. The study demonstrates machine learning can be an effective additional tool for de-duplication of genomic sequence databases. All Data are available as described in the supplementary material.

  1. On Combining Multiple-Instance Learning and Active Learning for Computer-Aided Detection of Tuberculosis

    NARCIS (Netherlands)

    Melendez Rodriguez, J.C.; Ginneken, B. van; Maduskar, P.; Philipsen, R.H.H.M.; Ayles, H.; Sanchez, C.I.

    2016-01-01

    The major advantage of multiple-instance learning (MIL) applied to a computer-aided detection (CAD) system is that it allows optimizing the latter with case-level labels instead of accurate lesion outlines as traditionally required for a supervised approach. As shown in previous work, a MIL-based

  2. Antigen detection systems

    Science.gov (United States)

    Infectious agents or their constituent parts (antigens or nucleic acids) can be detected in fresh, frozen, or fixed tissues or other specimens, using a variety of direct or indirect assays. The assays can be modified to yield the greatest sensitivity and specificity but in most cases a particular m...

  3. Detection and counting systems

    International Nuclear Information System (INIS)

    Abreu, M.A.N. de

    1976-01-01

    Detection devices based on gaseous ionization are analysed, such as: electroscopes ionization chambers, proportional counters and Geiger-Mueller counters. Scintillation methods are also commented. A revision of the basic concepts in electronics is done and the main equipment for counting is detailed. In the study of gama spectrometry, scintillation and semiconductor detectors are analysed [pt

  4. Rapid deployment intrusion detection system

    International Nuclear Information System (INIS)

    Graham, R.H.

    1997-01-01

    A rapidly deployable security system is one that provides intrusion detection, assessment, communications, and annunciation capabilities; is easy to install and configure; can be rapidly deployed, and is reusable. A rapidly deployable intrusion detection system (RADIDS) has many potential applications within the DOE Complex: back-up protection for failed zones in a perimeter intrusion detection and assessment system, intrusion detection and assessment capabilities in temporary locations, protection of assets during Complex reconfiguration, and protection in hazardous locations, protection of assets during Complex reconfiguration, and protection in hazardous locations. Many DOE user-need documents have indicated an interest in a rapidly deployable intrusion detection system. The purpose of the RADIDS project is to design, develop, and implement such a system. 2 figs

  5. Modeling learning technology systems as business systems

    NARCIS (Netherlands)

    Avgeriou, Paris; Retalis, Symeon; Papaspyrou, Nikolaos

    2003-01-01

    The design of Learning Technology Systems, and the Software Systems that support them, is largely conducted on an intuitive, ad hoc basis, thus resulting in inefficient systems that defectively support the learning process. There is now justifiable, increasing effort in formalizing the engineering

  6. Automatic Emergence Detection in Complex Systems

    Directory of Open Access Journals (Sweden)

    Eugene Santos

    2017-01-01

    Full Text Available Complex systems consist of multiple interacting subsystems, whose nonlinear interactions can result in unanticipated (emergent system events. Extant systems analysis approaches fail to detect such emergent properties, since they analyze each subsystem separately and arrive at decisions typically through linear aggregations of individual analysis results. In this paper, we propose a quantitative definition of emergence for complex systems. We also propose a framework to detect emergent properties given observations of its subsystems. This framework, based on a probabilistic graphical model called Bayesian Knowledge Bases (BKBs, learns individual subsystem dynamics from data, probabilistically and structurally fuses said dynamics into a single complex system dynamics, and detects emergent properties. Fusion is the central element of our approach to account for situations when a common variable may have different probabilistic distributions in different subsystems. We evaluate our detection performance against a baseline approach (Bayesian Network ensemble on synthetic testbeds from UCI datasets. To do so, we also introduce a method to simulate and a metric to measure discrepancies that occur with shared/common variables. Experiments demonstrate that our framework outperforms the baseline. In addition, we demonstrate that this framework has uniform polynomial time complexity across all three learning, fusion, and reasoning procedures.

  7. Particle detection systems and methods

    Science.gov (United States)

    Morris, Christopher L.; Makela, Mark F.

    2010-05-11

    Techniques, apparatus and systems for detecting particles such as muons and neutrons. In one implementation, a particle detection system employs a plurality of drift cells, which can be for example sealed gas-filled drift tubes, arranged on sides of a volume to be scanned to track incoming and outgoing charged particles, such as cosmic ray-produced muons. The drift cells can include a neutron sensitive medium to enable concurrent counting of neutrons. The system can selectively detect devices or materials, such as iron, lead, gold, uranium, plutonium, and/or tungsten, occupying the volume from multiple scattering of the charged particles passing through the volume and can concurrently detect any unshielded neutron sources occupying the volume from neutrons emitted therefrom. If necessary, the drift cells can be used to also detect gamma rays. The system can be employed to inspect occupied vehicles at border crossings for nuclear threat objects.

  8. Rapid detection of small oscillation faults via deterministic learning.

    Science.gov (United States)

    Wang, Cong; Chen, Tianrui

    2011-08-01

    Detection of small faults is one of the most important and challenging tasks in the area of fault diagnosis. In this paper, we present an approach for the rapid detection of small oscillation faults based on a recently proposed deterministic learning (DL) theory. The approach consists of two phases: the training phase and the test phase. In the training phase, the system dynamics underlying normal and fault oscillations are locally accurately approximated through DL. The obtained knowledge of system dynamics is stored in constant radial basis function (RBF) networks. In the diagnosis phase, rapid detection is implemented. Specially, a bank of estimators are constructed using the constant RBF neural networks to represent the training normal and fault modes. By comparing the set of estimators with the test monitored system, a set of residuals are generated, and the average L(1) norms of the residuals are taken as the measure of the differences between the dynamics of the monitored system and the dynamics of the training normal mode and oscillation faults. The occurrence of a test oscillation fault can be rapidly detected according to the smallest residual principle. A rigorous analysis of the performance of the detection scheme is also given. The novelty of the paper lies in that the modeling uncertainty and nonlinear fault functions are accurately approximated and then the knowledge is utilized to achieve rapid detection of small oscillation faults. Simulation studies are included to demonstrate the effectiveness of the approach.

  9. Radiation detecting system

    International Nuclear Information System (INIS)

    1975-01-01

    In spectrophotometry systems, a usual arrangement for modulating the radiation is a rotating disc having one or more sectors removed. A beam of radiation may be blocked by the disc except when a cut-away sector is in the path of the beam. With a double-beam system, a cut-away sector of 180 0 may be used so that when the first path is blocked, the second is allowed through, and vice versa. One or both sides of the disc may be formed as mirrors to facilitate beam switching and to allow use of more than two beams for background compensation purposes or for analysis of more than one substance within a sample. (G.T.H.)

  10. Radiation detection system

    Science.gov (United States)

    Whited, R.C.

    A system for obtaining improved resolution in relatively thick semiconductor radiation detectors, such as HgI/sub 2/, which exhibit significant hole trapping. Two amplifiers are used: the first measures the charge collected and the second the contribution of the electrons to the charge collected. The outputs of the two amplifiers are utilized to unfold the total charge generated within the detector in response to a radiation event.

  11. An environmental monitoring detection system

    International Nuclear Information System (INIS)

    Leli Yuniarsari; Istofa; Sukandar

    2015-01-01

    Is part of radiation detection of the nuclear facilities engineering activities within nuclear facilities. The system comprised of gamma-ray radiation detector and weather detection which includes anemometer to detect the wind direction and speed, as well as rain gauge to measure the rainfall in a period of time. Data acquisition of the output is processed by Arduino Uno system which transformed the data into a particular standard and then displayed online in the website. The radiation detection system uses gamma-ray detector of NaI(Tl) and GM which convert the radiation detected into electric pulse to be fed into a pre-amp and amplifier and modified into square pulse. The weather detection system on the other hand works based on switch principle. For example, the wind with a certain speed could turn on a switch in the system and produce a voltage or pulse which can be measured. This value will then be interpreted as the wind direction and speed. Likewise for the rainfall gauge, the volume of water entering the bucket will turn the switch on, at the same time producing 1 pulse. The result of the experiment shows that for radiation detection system the output is a square pulse 4 volts by using detector NaI(Tl) and 4.4 volts by using detector GM. For weather detection system, basically was able to detect the wind direction, wind speed and rainfall just to find out further research is needed accuracy and the results compared with the standard tools available in BMKG. (author)

  12. Ferromagnetic Objects Magnetovision Detection System.

    Science.gov (United States)

    Nowicki, Michał; Szewczyk, Roman

    2013-12-02

    This paper presents the application of a weak magnetic fields magnetovision scanning system for detection of dangerous ferromagnetic objects. A measurement system was developed and built to study the magnetic field vector distributions. The measurements of the Earth's field distortions caused by various ferromagnetic objects were carried out. The ability for passive detection of hidden or buried dangerous objects and the determination of their location was demonstrated.

  13. Ferromagnetic Objects Magnetovision Detection System

    Directory of Open Access Journals (Sweden)

    Michał Nowicki

    2013-12-01

    Full Text Available This paper presents the application of a weak magnetic fields magnetovision scanning system for detection of dangerous ferromagnetic objects. A measurement system was developed and built to study the magnetic field vector distributions. The measurements of the Earth’s field distortions caused by various ferromagnetic objects were carried out. The ability for passive detection of hidden or buried dangerous objects and the determination of their location was demonstrated.

  14. Recommendation System for Adaptive Learning.

    Science.gov (United States)

    Chen, Yunxiao; Li, Xiaoou; Liu, Jingchen; Ying, Zhiliang

    2018-01-01

    An adaptive learning system aims at providing instruction tailored to the current status of a learner, differing from the traditional classroom experience. The latest advances in technology make adaptive learning possible, which has the potential to provide students with high-quality learning benefit at a low cost. A key component of an adaptive learning system is a recommendation system, which recommends the next material (video lectures, practices, and so on, on different skills) to the learner, based on the psychometric assessment results and possibly other individual characteristics. An important question then follows: How should recommendations be made? To answer this question, a mathematical framework is proposed that characterizes the recommendation process as a Markov decision problem, for which decisions are made based on the current knowledge of the learner and that of the learning materials. In particular, two plain vanilla systems are introduced, for which the optimal recommendation at each stage can be obtained analytically.

  15. Learning Companion Systems, Social Learning Systems, and the Global Social Learning Club.

    Science.gov (United States)

    Chan, Tak-Wai

    1996-01-01

    Describes the development of learning companion systems and their contributions to the class of social learning systems that integrate artificial intelligence agents and use machine learning to tutor and interact with students. Outlines initial social learning projects, their programming languages, and weakness. Future improvements will include…

  16. Adaptive and accelerated tracking-learning-detection

    Science.gov (United States)

    Guo, Pengyu; Li, Xin; Ding, Shaowen; Tian, Zunhua; Zhang, Xiaohu

    2013-08-01

    An improved online long-term visual tracking algorithm, named adaptive and accelerated TLD (AA-TLD) based on Tracking-Learning-Detection (TLD) which is a novel tracking framework has been introduced in this paper. The improvement focuses on two aspects, one is adaption, which makes the algorithm not dependent on the pre-defined scanning grids by online generating scale space, and the other is efficiency, which uses not only algorithm-level acceleration like scale prediction that employs auto-regression and moving average (ARMA) model to learn the object motion to lessen the detector's searching range and the fixed number of positive and negative samples that ensures a constant retrieving time, but also CPU and GPU parallel technology to achieve hardware acceleration. In addition, in order to obtain a better effect, some TLD's details are redesigned, which uses a weight including both normalized correlation coefficient and scale size to integrate results, and adjusts distance metric thresholds online. A contrastive experiment on success rate, center location error and execution time, is carried out to show a performance and efficiency upgrade over state-of-the-art TLD with partial TLD datasets and Shenzhou IX return capsule image sequences. The algorithm can be used in the field of video surveillance to meet the need of real-time video tracking.

  17. Security Enrichment in Intrusion Detection System Using Classifier Ensemble

    Directory of Open Access Journals (Sweden)

    Uma R. Salunkhe

    2017-01-01

    Full Text Available In the era of Internet and with increasing number of people as its end users, a large number of attack categories are introduced daily. Hence, effective detection of various attacks with the help of Intrusion Detection Systems is an emerging trend in research these days. Existing studies show effectiveness of machine learning approaches in handling Intrusion Detection Systems. In this work, we aim to enhance detection rate of Intrusion Detection System by using machine learning technique. We propose a novel classifier ensemble based IDS that is constructed using hybrid approach which combines data level and feature level approach. Classifier ensembles combine the opinions of different experts and improve the intrusion detection rate. Experimental results show the improved detection rates of our system compared to reference technique.

  18. A system for learning statistical motion patterns.

    Science.gov (United States)

    Hu, Weiming; Xiao, Xuejuan; Fu, Zhouyu; Xie, Dan; Tan, Tieniu; Maybank, Steve

    2006-09-01

    Analysis of motion patterns is an effective approach for anomaly detection and behavior prediction. Current approaches for the analysis of motion patterns depend on known scenes, where objects move in predefined ways. It is highly desirable to automatically construct object motion patterns which reflect the knowledge of the scene. In this paper, we present a system for automatically learning motion patterns for anomaly detection and behavior prediction based on a proposed algorithm for robustly tracking multiple objects. In the tracking algorithm, foreground pixels are clustered using a fast accurate fuzzy K-means algorithm. Growing and prediction of the cluster centroids of foreground pixels ensure that each cluster centroid is associated with a moving object in the scene. In the algorithm for learning motion patterns, trajectories are clustered hierarchically using spatial and temporal information and then each motion pattern is represented with a chain of Gaussian distributions. Based on the learned statistical motion patterns, statistical methods are used to detect anomalies and predict behaviors. Our system is tested using image sequences acquired, respectively, from a crowded real traffic scene and a model traffic scene. Experimental results show the robustness of the tracking algorithm, the efficiency of the algorithm for learning motion patterns, and the encouraging performance of algorithms for anomaly detection and behavior prediction.

  19. CLASSIFICATION OF LEARNING MANAGEMENT SYSTEMS

    Directory of Open Access Journals (Sweden)

    Yu. B. Popova

    2016-01-01

    Full Text Available Using of information technologies and, in particular, learning management systems, increases opportunities of teachers and students in reaching their goals in education. Such systems provide learning content, help organize and monitor training, collect progress statistics and take into account the individual characteristics of each user. Currently, there is a huge inventory of both paid and free systems are physically located both on college servers and in the cloud, offering different features sets of different licensing scheme and the cost. This creates the problem of choosing the best system. This problem is partly due to the lack of comprehensive classification of such systems. Analysis of more than 30 of the most common now automated learning management systems has shown that a classification of such systems should be carried out according to certain criteria, under which the same type of system can be considered. As classification features offered by the author are: cost, functionality, modularity, keeping the customer’s requirements, the integration of content, the physical location of a system, adaptability training. Considering the learning management system within these classifications and taking into account the current trends of their development, it is possible to identify the main requirements to them: functionality, reliability, ease of use, low cost, support for SCORM standard or Tin Can API, modularity and adaptability. According to the requirements at the Software Department of FITR BNTU under the guidance of the author since 2009 take place the development, the use and continuous improvement of their own learning management system.

  20. Fall detection using supervised machine learning algorithms: A comparative study

    KAUST Repository

    Zerrouki, Nabil; Harrou, Fouzi; Houacine, Amrane; Sun, Ying

    2017-01-01

    Fall incidents are considered as the leading cause of disability and even mortality among older adults. To address this problem, fall detection and prevention fields receive a lot of intention over the past years and attracted many researcher efforts. We present in the current study an overall performance comparison between fall detection systems using the most popular machine learning approaches which are: Naïve Bayes, K nearest neighbor, neural network, and support vector machine. The analysis of the classification power associated to these most widely utilized algorithms is conducted on two fall detection databases namely FDD and URFD. Since the performance of the classification algorithm is inherently dependent on the features, we extracted and used the same features for all classifiers. The classification evaluation is conducted using different state of the art statistical measures such as the overall accuracy, the F-measure coefficient, and the area under ROC curve (AUC) value.

  1. Fall detection using supervised machine learning algorithms: A comparative study

    KAUST Repository

    Zerrouki, Nabil

    2017-01-05

    Fall incidents are considered as the leading cause of disability and even mortality among older adults. To address this problem, fall detection and prevention fields receive a lot of intention over the past years and attracted many researcher efforts. We present in the current study an overall performance comparison between fall detection systems using the most popular machine learning approaches which are: Naïve Bayes, K nearest neighbor, neural network, and support vector machine. The analysis of the classification power associated to these most widely utilized algorithms is conducted on two fall detection databases namely FDD and URFD. Since the performance of the classification algorithm is inherently dependent on the features, we extracted and used the same features for all classifiers. The classification evaluation is conducted using different state of the art statistical measures such as the overall accuracy, the F-measure coefficient, and the area under ROC curve (AUC) value.

  2. Integrated multisensor perimeter detection systems

    Science.gov (United States)

    Kent, P. J.; Fretwell, P.; Barrett, D. J.; Faulkner, D. A.

    2007-10-01

    The report describes the results of a multi-year programme of research aimed at the development of an integrated multi-sensor perimeter detection system capable of being deployed at an operational site. The research was driven by end user requirements in protective security, particularly in threat detection and assessment, where effective capability was either not available or prohibitively expensive. Novel video analytics have been designed to provide robust detection of pedestrians in clutter while new radar detection and tracking algorithms provide wide area day/night surveillance. A modular integrated architecture based on commercially available components has been developed. A graphical user interface allows intuitive interaction and visualisation with the sensors. The fusion of video, radar and other sensor data provides the basis of a threat detection capability for real life conditions. The system was designed to be modular and extendable in order to accommodate future and legacy surveillance sensors. The current sensor mix includes stereoscopic video cameras, mmWave ground movement radar, CCTV and a commercially available perimeter detection cable. The paper outlines the development of the system and describes the lessons learnt after deployment in a pilot trial.

  3. Intelligent fractions learning system: implementation

    CSIR Research Space (South Africa)

    Smith, Andrew C

    2011-05-01

    Full Text Available Conference Proceedings Paul Cunningham and Miriam Cunningham (Eds) IIMC International Information Management Corporation, 2011 ISBN: 978-1-905824-24-3 An Intelligent Fractions Learning System: Implementation Andrew Cyrus SMITH1, Teemu H. LAINE2 1CSIR... to fractions. Our aim with the current research project is to extend the existing UFractions learning system to incorporate automatic data capturing. ?Intelligent UFractions? allows a teacher to remotely monitor the children?s progress during...

  4. Overcoming complexities: Damage detection using dictionary learning framework

    Science.gov (United States)

    Alguri, K. Supreet; Melville, Joseph; Deemer, Chris; Harley, Joel B.

    2018-04-01

    For in situ damage detection, guided wave structural health monitoring systems have been widely researched due to their ability to evaluate large areas and their ability detect many types of damage. These systems often evaluate structural health by recording initial baseline measurements from a pristine (i.e., undamaged) test structure and then comparing later measurements with that baseline. Yet, it is not always feasible to have a pristine baseline. As an alternative, substituting the baseline with data from a surrogate (nearly identical and pristine) structure is a logical option. While effective in some circumstance, surrogate data is often still a poor substitute for pristine baseline measurements due to minor differences between the structures. To overcome this challenge, we present a dictionary learning framework to adapt surrogate baseline data to better represent an undamaged test structure. We compare the performance of our framework with two other surrogate-based damage detection strategies: (1) using raw surrogate data for comparison and (2) using sparse wavenumber analysis, a precursor to our framework for improving the surrogate data. We apply our framework to guided wave data from two 108 mm by 108 mm aluminum plates. With 20 measurements, we show that our dictionary learning framework achieves a 98% accuracy, raw surrogate data achieves a 92% accuracy, and sparse wavenumber analysis achieves a 57% accuracy.

  5. An FPGA-Based People Detection System

    Directory of Open Access Journals (Sweden)

    James J. Clark

    2005-05-01

    Full Text Available This paper presents an FPGA-based system for detecting people from video. The system is designed to use JPEG-compressed frames from a network camera. Unlike previous approaches that use techniques such as background subtraction and motion detection, we use a machine-learning-based approach to train an accurate detector. We address the hardware design challenges involved in implementing such a detector, along with JPEG decompression, on an FPGA. We also present an algorithm that efficiently combines JPEG decompression with the detection process. This algorithm carries out the inverse DCT step of JPEG decompression only partially. Therefore, it is computationally more efficient and simpler to implement, and it takes up less space on the chip than the full inverse DCT algorithm. The system is demonstrated on an automated video surveillance application and the performance of both hardware and software implementations is analyzed. The results show that the system can detect people accurately at a rate of about 2.5 frames per second on a Virtex-II 2V1000 using a MicroBlaze processor running at 75 MHz, communicating with dedicated hardware over FSL links.

  6. An efficient flow-based botnet detection using supervised machine learning

    DEFF Research Database (Denmark)

    Stevanovic, Matija; Pedersen, Jens Myrup

    2014-01-01

    Botnet detection represents one of the most crucial prerequisites of successful botnet neutralization. This paper explores how accurate and timely detection can be achieved by using supervised machine learning as the tool of inferring about malicious botnet traffic. In order to do so, the paper...... introduces a novel flow-based detection system that relies on supervised machine learning for identifying botnet network traffic. For use in the system we consider eight highly regarded machine learning algorithms, indicating the best performing one. Furthermore, the paper evaluates how much traffic needs...... to accurately and timely detect botnet traffic using purely flow-based traffic analysis and supervised machine learning. Additionally, the results show that in order to achieve accurate detection traffic flows need to be monitored for only a limited time period and number of packets per flow. This indicates...

  7. Basis UST leak detection systems

    International Nuclear Information System (INIS)

    Silveria, V.

    1992-01-01

    This paper reports that gasoline and other petroleum products are leaking from underground storage tanks (USTs) at an alarming rate, seeping into soil and groundwater. Buried pipes are an even greater culprit, accounting for most suspected and detected leaks according to Environmental Protection Agency (EPA) estimates. In response to this problem, the EPA issued regulations setting standards for preventing, detecting, reporting, and cleaning up leaks, as well as fiscal responsibility. However, federal regulations are only a minimum; some states have cracked down even harder Plant managers and engineers have a big job ahead of them. The EPA estimates that there are more than 75,000 fuel USTs at US industrial facilities. When considering leak detection systems, the person responsible for making the decision has five primary choices: inventory reconciliation combined with regular precision tightness tests; automatic tank gauging; groundwater monitoring; interstitial monitoring of double containment systems; and vapor monitoring

  8. Wide range neutron detection system

    International Nuclear Information System (INIS)

    Todt, W.H. Sr.

    1978-01-01

    A neutron detection system for reactor control is described which is operable over a wide range of neutron flux levels. The system includes a fission type ionization chamber neutron detector, means for gamma and alpha signal compensation, and means for operating the neutron detector in the pulse counting mode for low neutron flux levels, and in the direct current mode for high neutron flux levels

  9. Bat detective-Deep learning tools for bat acoustic signal detection.

    Science.gov (United States)

    Mac Aodha, Oisin; Gibb, Rory; Barlow, Kate E; Browning, Ella; Firman, Michael; Freeman, Robin; Harder, Briana; Kinsey, Libby; Mead, Gary R; Newson, Stuart E; Pandourski, Ivan; Parsons, Stuart; Russ, Jon; Szodoray-Paradi, Abigel; Szodoray-Paradi, Farkas; Tilova, Elena; Girolami, Mark; Brostow, Gabriel; Jones, Kate E

    2018-03-01

    Passive acoustic sensing has emerged as a powerful tool for quantifying anthropogenic impacts on biodiversity, especially for echolocating bat species. To better assess bat population trends there is a critical need for accurate, reliable, and open source tools that allow the detection and classification of bat calls in large collections of audio recordings. The majority of existing tools are commercial or have focused on the species classification task, neglecting the important problem of first localizing echolocation calls in audio which is particularly problematic in noisy recordings. We developed a convolutional neural network based open-source pipeline for detecting ultrasonic, full-spectrum, search-phase calls produced by echolocating bats. Our deep learning algorithms were trained on full-spectrum ultrasonic audio collected along road-transects across Europe and labelled by citizen scientists from www.batdetective.org. When compared to other existing algorithms and commercial systems, we show significantly higher detection performance of search-phase echolocation calls with our test sets. As an example application, we ran our detection pipeline on bat monitoring data collected over five years from Jersey (UK), and compared results to a widely-used commercial system. Our detection pipeline can be used for the automatic detection and monitoring of bat populations, and further facilitates their use as indicator species on a large scale. Our proposed pipeline makes only a small number of bat specific design decisions, and with appropriate training data it could be applied to detecting other species in audio. A crucial novelty of our work is showing that with careful, non-trivial, design and implementation considerations, state-of-the-art deep learning methods can be used for accurate and efficient monitoring in audio.

  10. Road Anomalies Detection System Evaluation.

    Science.gov (United States)

    Silva, Nuno; Shah, Vaibhav; Soares, João; Rodrigues, Helena

    2018-06-21

    Anomalies on road pavement cause discomfort to drivers and passengers, and may cause mechanical failure or even accidents. Governments spend millions of Euros every year on road maintenance, often causing traffic jams and congestion on urban roads on a daily basis. This paper analyses the difference between the deployment of a road anomalies detection and identification system in a “conditioned” and a real world setup, where the system performed worse compared to the “conditioned” setup. It also presents a system performance analysis based on the analysis of the training data sets; on the analysis of the attributes complexity, through the application of PCA techniques; and on the analysis of the attributes in the context of each anomaly type, using acceleration standard deviation attributes to observe how different anomalies classes are distributed in the Cartesian coordinates system. Overall, in this paper, we describe the main insights on road anomalies detection challenges to support the design and deployment of a new iteration of our system towards the deployment of a road anomaly detection service to provide information about roads condition to drivers and government entities.

  11. Tape Cassette Bacteria Detection System

    Science.gov (United States)

    1973-01-01

    The design, fabrication, and testing of an automatic bacteria detection system with a zero-g capability and based on the filter-capsule approach is described. This system is intended for monitoring the sterility of regenerated water in a spacecraft. The principle of detection is based on measuring the increase in chemiluminescence produced by the action of bacterial porphyrins (i.e., catalase, cytochromes, etc.) on a luminol-hydrogen peroxide mixture. Since viable as well as nonviable organisms initiate this luminescence, viable organisms are detected by comparing the signal of an incubated water sample with an unincubated control. Higher signals for the former indicate the presence of viable organisms. System features include disposable sealed sterile capsules, each containing a filter membrane, for processing discrete water samples and a tape transport for moving these capsules through a processing sequence which involves sample concentration, nutrient addition, incubation, a 4 Molar Urea wash and reaction with luminol-hydrogen peroxide in front of a photomultiplier tube. Liquids are introduced by means of a syringe needle which pierces a rubber septum contained in the wall of the capsule. Detection thresholds obtained with this unit towards E. coli and S. marcescens assuming a 400 ml water sample are indicated.

  12. LONS: Learning Object Negotiation System

    Science.gov (United States)

    García, Antonio; García, Eva; de-Marcos, Luis; Martínez, José-Javier; Gutiérrez, José-María; Gutiérrez, José-Antonio; Barchino, Roberto; Otón, Salvador; Hilera, José-Ramón

    This system comes up as a result of the increase of e-learning systems. It manages all relevant modules in this context, such as the association of digital rights with the contents (courses), management and payment processing on rights. There are three blocks:

  13. Learning in Artificial Neural Systems

    Science.gov (United States)

    Matheus, Christopher J.; Hohensee, William E.

    1987-01-01

    This paper presents an overview and analysis of learning in Artificial Neural Systems (ANS's). It begins with a general introduction to neural networks and connectionist approaches to information processing. The basis for learning in ANS's is then described, and compared with classical Machine learning. While similar in some ways, ANS learning deviates from tradition in its dependence on the modification of individual weights to bring about changes in a knowledge representation distributed across connections in a network. This unique form of learning is analyzed from two aspects: the selection of an appropriate network architecture for representing the problem, and the choice of a suitable learning rule capable of reproducing the desired function within the given network. The various network architectures are classified, and then identified with explicit restrictions on the types of functions they are capable of representing. The learning rules, i.e., algorithms that specify how the network weights are modified, are similarly taxonomized, and where possible, the limitations inherent to specific classes of rules are outlined.

  14. Lightning Protection and Detection System

    Science.gov (United States)

    Dudley, Kenneth L. (Inventor); Szatkowski, George N. (Inventor); Woodard, Marie (Inventor); Nguyen, Truong X. (Inventor); Ely, Jay J. (Inventor); Wang, Chuantong (Inventor); Mielnik, John J. (Inventor); Koppen, Sandra V. (Inventor); Smith, Laura J. (Inventor)

    2017-01-01

    A lightning protection and detection system includes a non-conductive substrate material of an apparatus; a sensor formed of a conductive material and deposited on the non-conductive substrate material of the apparatus. The sensor includes a conductive trace formed in a continuous spiral winding starting at a first end at a center region of the sensor and ending at a second end at an outer corner region of the sensor, the first and second ends being open and unconnected. An electrical measurement system is in communication with the sensor and receives a resonant response from the sensor, to perform detection, in real-time, of lightning strike occurrences and damage therefrom to the sensor and the non-conductive substrate material.

  15. Optical detection in microfluidic systems

    DEFF Research Database (Denmark)

    Mogensen, Klaus Bo; Kutter, Jörg Peter

    2009-01-01

    Optical detection schemes continue to be favoured for measurements in microfluidic systems. A selection of the latest progress mainly within the last two years is critically reviewed. Emphasis is on integrated solutions, such as planar waveguides, coupling schemes to the outside world, evanescent...... to ease commercialisation of the devices. This work will hopefully result in more commercial products that benefit from integrated optics, because the impact on commercial devices so far has been modest....

  16. Social software: E-learning beyond learning management systems

    DEFF Research Database (Denmark)

    Dalsgaard, Christian

    2006-01-01

    The article argues that it is necessary to move e-learning beyond learning management systems and engage students in an active use of the web as a resource for their self-governed, problem-based and collaborative activities. The purpose of the article is to discuss the potential of social software...... to move e-learning beyond learning management systems. An approach to use of social software in support of a social constructivist approach to e-learning is presented, and it is argued that learning management systems do not support a social constructivist approach which emphasizes self-governed learning...... activities of students. The article suggests a limitation of the use of learning management systems to cover only administrative issues. Further, it is argued that students' self-governed learning processes are supported by providing students with personal tools and engaging them in different kinds of social...

  17. Detecting Attacks in CyberManufacturing Systems: Additive Manufacturing Example

    Directory of Open Access Journals (Sweden)

    Wu Mingtao

    2017-01-01

    Full Text Available CyberManufacturing System is a vision for future manufacturing where physical components are fully integrated with computational processes in a connected environment. However, realizing the vision requires that its security be adequately ensured. This paper presents a vision-based system to detect intentional attacks on additive manufacturing processes, utilizing machine learning techniques. Particularly, additive manufacturing systems have unique vulnerabilities to malicious attacks, which can result in defective infills but without affecting the exterior. In order to detect such infill defects, the research uses simulated 3D printing process images as well as actual 3D printing process images to compare accuracies of machine learning algorithms in classifying, clustering and detecting anomalies on different types of infills. Three algorithms - (i random forest, (ii k nearest neighbor, and (iii anomaly detection - have been adopted in the research and shown to be effective in detecting such defects.

  18. Unsupervised learning for robust bitcoin fraud detection

    CSIR Research Space (South Africa)

    Monamo, Patrick

    2016-08-01

    Full Text Available The rampant absorption of Bitcoin as a cryptographic currency, along with rising cybercrime activities, warrants utilization of anomaly detection to identify potential fraud. Anomaly detection plays a pivotal role in data mining since most outlying...

  19. Nucleic acid detection system and method for detecting influenza

    Science.gov (United States)

    Cai, Hong; Song, Jian

    2015-03-17

    The invention provides a rapid, sensitive and specific nucleic acid detection system which utilizes isothermal nucleic acid amplification in combination with a lateral flow chromatographic device, or DNA dipstick, for DNA-hybridization detection. The system of the invention requires no complex instrumentation or electronic hardware, and provides a low cost nucleic acid detection system suitable for highly sensitive pathogen detection. Hybridization to single-stranded DNA amplification products using the system of the invention provides a sensitive and specific means by which assays can be multiplexed for the detection of multiple target sequences.

  20. Learning a Novel Detection Metric for the Detection of O’Connell Effect Eclipsing Binaries

    Science.gov (United States)

    Johnston, Kyle; Haber, Rana; Knote, Matthew; Caballero-Nieves, Saida Maria; Peter, Adrian; Petit, Véronique

    2018-01-01

    With the advent of digital astronomy, new benefits and new challenges have been presented to the modern day astronomer. No longer can the astronomer rely on manual processing, instead the profession as a whole has begun to adopt more advanced computational means. Here we focus on the construction and application of a novel time-domain signature extraction methodology and the development of a supporting supervised pattern detection algorithm for the targeted identification of eclipsing binaries which demonstrate a feature known as the O’Connell Effect. A methodology for the reduction of stellar variable observations (time-domain data) into Distribution Fields (DF) is presented. Push-Pull metric learning, a variant of LMNN learning, is used to generate a learned distance metric for the specific detection problem proposed. The metric will be trained on a set of a labelled Kepler eclipsing binary data, in particular systems showing the O’Connell effect. Performance estimates will be presented, as well the results of the detector applied to an unlabeled Kepler EB data set; this work is a crucial step in the upcoming era of big data from the next generation of big telescopes, such as LSST.

  1. Machine learning for the automatic detection of anomalous events

    Science.gov (United States)

    Fisher, Wendy D.

    In this dissertation, we describe our research contributions for a novel approach to the application of machine learning for the automatic detection of anomalous events. We work in two different domains to ensure a robust data-driven workflow that could be generalized for monitoring other systems. Specifically, in our first domain, we begin with the identification of internal erosion events in earth dams and levees (EDLs) using geophysical data collected from sensors located on the surface of the levee. As EDLs across the globe reach the end of their design lives, effectively monitoring their structural integrity is of critical importance. The second domain of interest is related to mobile telecommunications, where we investigate a system for automatically detecting non-commercial base station routers (BSRs) operating in protected frequency space. The presence of non-commercial BSRs can disrupt the connectivity of end users, cause service issues for the commercial providers, and introduce significant security concerns. We provide our motivation, experimentation, and results from investigating a generalized novel data-driven workflow using several machine learning techniques. In Chapter 2, we present results from our performance study that uses popular unsupervised clustering algorithms to gain insights to our real-world problems, and evaluate our results using internal and external validation techniques. Using EDL passive seismic data from an experimental laboratory earth embankment, results consistently show a clear separation of events from non-events in four of the five clustering algorithms applied. Chapter 3 uses a multivariate Gaussian machine learning model to identify anomalies in our experimental data sets. For the EDL work, we used experimental data from two different laboratory earth embankments. Additionally, we explore five wavelet transform methods for signal denoising. The best performance is achieved with the Haar wavelets. We achieve up to 97

  2. Compensated intruder-detection systems

    Science.gov (United States)

    McNeilly, David R.; Miller, William R.

    1984-01-01

    Intruder-detection systems in which intruder-induced signals are transmitted through a medium also receive spurious signals induced by changes in a climatic condition affecting the medium. To combat this, signals received from the detection medium are converted to a first signal. The system also provides a reference signal proportional to climate-induced changes in the medium. The first signal and the reference signal are combined for generating therefrom an output signal which is insensitive to the climatic changes in the medium. An alarm is energized if the output signal exceeds a preselected value. In one embodiment, an acoustic cable is coupled to a fence to generate a first electrical signal proportional to movements thereof. False alarms resulting from wind-induced movements of the fence (detection medium) are eliminated by providing an anemometer-driven voltage generator to provide a reference voltage proportional to the velocity of wind incident on the fence. An analog divider receives the first electrical signal and the reference signal as its numerator and denominator inputs, respectively, and generates therefrom an output signal which is insensitive to the wind-induced movements in the fence.

  3. Detecting Learning Style through Biometric Technology for Mobile GBL

    Science.gov (United States)

    Mehigan, Tracey J.; Pitt, Ian

    2012-01-01

    Adaptive learning systems tailor content delivery to meet specific needs of the individual for improved learning-outcomes. Learning-styles and personalities are usually determined through the completion of questionnaires. There are a number of models available for this purpose including the Myer-Briggs Model (MBTI), the Big Five Model, and the…

  4. Ionization detection system for aerosols

    International Nuclear Information System (INIS)

    Jacobs, M.E.

    1977-01-01

    This invention relates to an improved smoke-detection system of the ionization-chamber type. In the preferred embodiment, the system utilizes a conventional detector head comprising a measuring ionization chamber, a reference ionization chamber, and a normally non-conductive gas triode for discharging when a threshold concentration of airborne particulates is present in the measuring chamber. The improved system utilizes a measuring ionization chamber which is modified to minimize false alarms and reductions in sensitivity resulting from changes in ambient temperature. In the preferred form of the modification, an annular radiation shield is mounted about the usual radiation source provided to effect ionization in the measuring chamber. The shield is supported by a bimetallic strip which flexes in response to changes in ambient temperature, moving the shield relative to the source so as to vary the radiative area of the source in a manner offsetting temperature-induced variations in the sensitivity of the chamber. 8 claims, 7 figures

  5. Microcontroller based driver alertness detection systems to detect drowsiness

    Science.gov (United States)

    Adenin, Hasibah; Zahari, Rahimi; Lim, Tiong Hoo

    2018-04-01

    The advancement of embedded system for detecting and preventing drowsiness in a vehicle is a major challenge for road traffic accident systems. To prevent drowsiness while driving, it is necessary to have an alert system that can detect a decline in driver concentration and send a signal to the driver. Studies have shown that traffc accidents usually occur when the driver is distracted while driving. In this paper, we have reviewed a number of detection systems to monitor the concentration of a car driver and propose a portable Driver Alertness Detection System (DADS) to determine the level of concentration of the driver based on pixelated coloration detection technique using facial recognition. A portable camera will be placed at the front visor to capture facial expression and the eye activities. We evaluate DADS using 26 participants and have achieved 100% detection rate with good lighting condition and a low detection rate at night.

  6. Learning Multimodal Deep Representations for Crowd Anomaly Event Detection

    Directory of Open Access Journals (Sweden)

    Shaonian Huang

    2018-01-01

    Full Text Available Anomaly event detection in crowd scenes is extremely important; however, the majority of existing studies merely use hand-crafted features to detect anomalies. In this study, a novel unsupervised deep learning framework is proposed to detect anomaly events in crowded scenes. Specifically, low-level visual features, energy features, and motion map features are simultaneously extracted based on spatiotemporal energy measurements. Three convolutional restricted Boltzmann machines are trained to model the mid-level feature representation of normal patterns. Then a multimodal fusion scheme is utilized to learn the deep representation of crowd patterns. Based on the learned deep representation, a one-class support vector machine model is used to detect anomaly events. The proposed method is evaluated using two available public datasets and compared with state-of-the-art methods. The experimental results show its competitive performance for anomaly event detection in video surveillance.

  7. An Intelligent System for Determining Learning Style

    Science.gov (United States)

    Ozdemir, Ali; Alaybeyoglu, Aysegul; Mulayim, Naciye; Uysal, Muhammed

    2018-01-01

    In this study, an intelligent system which determines learning style of the students is developed to increase success in effective and easy learning. The importance of the proposed software system is to determine convenience degree of the student's learning style. Personal information form and Dunn Learning Style Preference Survey are used to…

  8. Interference and deception detection technology of satellite navigation based on deep learning

    Science.gov (United States)

    Chen, Weiyi; Deng, Pingke; Qu, Yi; Zhang, Xiaoguang; Li, Yaping

    2017-10-01

    Satellite navigation system plays an important role in people's daily life and war. The strategic position of satellite navigation system is prominent, so it is very important to ensure that the satellite navigation system is not disturbed or destroyed. It is a critical means to detect the jamming signal to avoid the accident in a navigation system. At present, the detection technology of jamming signal in satellite navigation system is not intelligent , mainly relying on artificial decision and experience. For this issue, the paper proposes a method based on deep learning to monitor the interference source in a satellite navigation. By training the interference signal data, and extracting the features of the interference signal, the detection sys tem model is constructed. The simulation results show that, the detection accuracy of our detection system can reach nearly 70%. The method in our paper provides a new idea for the research on intelligent detection of interference and deception signal in a satellite navigation system.

  9. Learning Management Systems and Comparison of Open Source Learning Management Systems and Proprietary Learning Management Systems

    Directory of Open Access Journals (Sweden)

    Yücel Yılmaz

    2016-04-01

    Full Text Available The concept of learning has been increasingly gaining importance for individuals, businesses and communities in the age of information. On the other hand, developments in information and communication technologies take effect in the field of learning activities. With these technologies, barriers of time and space against the learning activities largely disappear and these technologies make it easier to carry out these activities more effectively. There remain a lot of questions regarding selection of learning management system (LMS to be used for the management of e-learning processes by all organizations conducing educational practices including universities, companies, non-profit organizations, etc. The main questions are as follows: Shall we choose open source LMS or commercial LMS? Can the selected LMS meet existing needs and future potential needs for the organization? What are the possibilities of technical support in the management of LMS? What kind of problems may be experienced in the use of LMS and how can these problems be solved? How much effective can officials in the organization be in the management of LMS? In this study, primarily e-learning and the concept of LMS will be discussed, and in the next section, as for answers to these questions, open source LMSs and centrally developed LMSs will be examined and their advantages and disadvantages relative to each other will be discussed.

  10. Detection of Hypertension Retinopathy Using Deep Learning and Boltzmann Machines

    Science.gov (United States)

    Triwijoyo, B. K.; Pradipto, Y. D.

    2017-01-01

    hypertensive retinopathy (HR) in the retina of the eye is disturbance caused by high blood pressure disease, where there is a systemic change of arterial in the blood vessels of the retina. Most heart attacks occur in patients caused by high blood pressure symptoms of undiagnosed. Hypertensive retinopathy Symptoms such as arteriolar narrowing, retinal haemorrhage and cotton wool spots. Based on this reasons, the early diagnosis of the symptoms of hypertensive retinopathy is very urgent to aim the prevention and treatment more accurate. This research aims to develop a system for early detection of hypertension retinopathy stage. The proposed method is to determine the combined features artery and vein diameter ratio (AVR) as well as changes position with Optic Disk (OD) in retinal images to review the classification of hypertensive retinopathy using Deep Neural Networks (DNN) and Boltzmann Machines approach. We choose this approach of because based on previous research DNN models were more accurate in the image pattern recognition, whereas Boltzmann machines selected because It requires speedy iteration in the process of learning neural network. The expected results from this research are designed a prototype system early detection of hypertensive retinopathy stage and analysed the effectiveness and accuracy of the proposed methods.

  11. VIRTUAL LABORATORY IN DISTANCE LEARNING SYSTEM

    Directory of Open Access Journals (Sweden)

    Е. Kozlovsky

    2011-11-01

    Full Text Available Questions of designing and a choice of technologies of creation of virtual laboratory for the distance learning system are considered. Distance learning system «Kherson Virtual University» is used as illustration.

  12. Deep Learning Approach for Car Detection in UAV Imagery

    Directory of Open Access Journals (Sweden)

    Nassim Ammour

    2017-03-01

    Full Text Available This paper presents an automatic solution to the problem of detecting and counting cars in unmanned aerial vehicle (UAV images. This is a challenging task given the very high spatial resolution of UAV images (on the order of a few centimetres and the extremely high level of detail, which require suitable automatic analysis methods. Our proposed method begins by segmenting the input image into small homogeneous regions, which can be used as candidate locations for car detection. Next, a window is extracted around each region, and deep learning is used to mine highly descriptive features from these windows. We use a deep convolutional neural network (CNN system that is already pre-trained on huge auxiliary data as a feature extraction tool, combined with a linear support vector machine (SVM classifier to classify regions into “car” and “no-car” classes. The final step is devoted to a fine-tuning procedure which performs morphological dilation to smooth the detected regions and fill any holes. In addition, small isolated regions are analysed further using a few sliding rectangular windows to locate cars more accurately and remove false positives. To evaluate our method, experiments were conducted on a challenging set of real UAV images acquired over an urban area. The experimental results have proven that the proposed method outperforms the state-of-the-art methods, both in terms of accuracy and computational time.

  13. In-situ trainable intrusion detection system

    Energy Technology Data Exchange (ETDEWEB)

    Symons, Christopher T.; Beaver, Justin M.; Gillen, Rob; Potok, Thomas E.

    2016-11-15

    A computer implemented method detects intrusions using a computer by analyzing network traffic. The method includes a semi-supervised learning module connected to a network node. The learning module uses labeled and unlabeled data to train a semi-supervised machine learning sensor. The method records events that include a feature set made up of unauthorized intrusions and benign computer requests. The method identifies at least some of the benign computer requests that occur during the recording of the events while treating the remainder of the data as unlabeled. The method trains the semi-supervised learning module at the network node in-situ, such that the semi-supervised learning modules may identify malicious traffic without relying on specific rules, signatures, or anomaly detection.

  14. Automatic Microaneurysms Detection Based on Multifeature Fusion Dictionary Learning

    Directory of Open Access Journals (Sweden)

    Wei Zhou

    2017-01-01

    Full Text Available Recently, microaneurysm (MA detection has attracted a lot of attention in the medical image processing community. Since MAs can be seen as the earliest lesions in diabetic retinopathy, their detection plays a critical role in diabetic retinopathy diagnosis. In this paper, we propose a novel MA detection approach named multifeature fusion dictionary learning (MFFDL. The proposed method consists of four steps: preprocessing, candidate extraction, multifeature dictionary learning, and classification. The novelty of our proposed approach lies in incorporating the semantic relationships among multifeatures and dictionary learning into a unified framework for automatic detection of MAs. We evaluate the proposed algorithm by comparing it with the state-of-the-art approaches and the experimental results validate the effectiveness of our algorithm.

  15. Learning to Detect Human-Object Interactions

    KAUST Repository

    Chao, Yu-Wei; Liu, Yunfan; Liu, Xieyang; Zeng, Huayi; Deng, Jia

    2017-01-01

    In this paper we study the problem of detecting human-object interactions (HOI) in static images, defined as predicting a human and an object bounding box with an interaction class label that connects them. HOI detection is a fundamental problem in computer vision as it provides semantic information about the interactions among the detected objects. We introduce HICO-DET, a new large benchmark for HOI detection, by augmenting the current HICO classification benchmark with instance annotations. We propose Human-Object Region-based Convolutional Neural Networks (HO-RCNN), a novel DNN-based framework for HOI detection. At the core of our HO-RCNN is the Interaction Pattern, a novel DNN input that characterizes the spatial relations between two bounding boxes. We validate the effectiveness of our HO-RCNN using HICO-DET. Experiments demonstrate that our HO-RCNN, by exploiting human-object spatial relations through Interaction Patterns, significantly improves the performance of HOI detection over baseline approaches.

  16. Learning to Detect Human-Object Interactions

    KAUST Repository

    Chao, Yu-Wei

    2017-02-17

    In this paper we study the problem of detecting human-object interactions (HOI) in static images, defined as predicting a human and an object bounding box with an interaction class label that connects them. HOI detection is a fundamental problem in computer vision as it provides semantic information about the interactions among the detected objects. We introduce HICO-DET, a new large benchmark for HOI detection, by augmenting the current HICO classification benchmark with instance annotations. We propose Human-Object Region-based Convolutional Neural Networks (HO-RCNN), a novel DNN-based framework for HOI detection. At the core of our HO-RCNN is the Interaction Pattern, a novel DNN input that characterizes the spatial relations between two bounding boxes. We validate the effectiveness of our HO-RCNN using HICO-DET. Experiments demonstrate that our HO-RCNN, by exploiting human-object spatial relations through Interaction Patterns, significantly improves the performance of HOI detection over baseline approaches.

  17. Nuclear detection systems in traffic

    International Nuclear Information System (INIS)

    Farkas, T.; Pernicka, L.; Svec, A.

    2005-01-01

    Illicit trafficking in nuclear materials (nuclear criminality) has become a problem, due to the circulation of a high number of radioactive sources caused by the changes of the organisational infrastructures to supervise these material within the successor states of the former Soviet Union. Aim of this paper is to point out the technical requirements and the practicability of an useful monitoring system at preselected traffic check points (railway and highway border crossings, industrial sites entry gates, international airports). The ITRAP lab test was designed to work as strict benchmark to qualify border monitoring systems 67 with very low false alarm rates, in addition the minimum sensitivity to give an alarm has been defined for fix-installed systems, pocket type and hand held instruments. For the neutron tests a special prepared Californium source ( 252 Cf) was used to simulate the weapons plutonium. The source is shielded against gamma radiation, use a moderator and provides the required neutron rate of 20000 n/s at 2 rn distance. To test the false alarm rate (rate of false positive ) the same test facility , under the same background conditions, was used but without a radioactive test source. The ITRAP lab tests for the fix-installed systems started at May 1998 and first results were given in September 1998. Only 2 of 14 fix-installed monitoring systems could fulfil the minimum requirement for neutron detection. 7 of 14 fix-installed monitoring systems (50%) passed the ITRAP lab test. The analytical method developed and used for certification of installed radiation monitors in the Slovak Institute of Metrology consists in measurement of radiation activity of selected radionuclide in defined conditions. (authors)

  18. A machine learning approach for indirect human presence detection using IoT devices

    OpenAIRE

    Madeira, Rui Nuno Neves

    2016-01-01

    The recent increased democratization of technology led to the appearance of new devices dedicated to the improvement of our daily living and working spaces, capable of being remotely controlled through the internet and interoperability with other systems. In this context, human presence detection is fundamental for several purposes, such has: further automization, usage pattern learning, problem detection (illness, or intruder), etc. Current intrusion detection devices usual...

  19. Thermal animal detection system (TADS)

    Energy Technology Data Exchange (ETDEWEB)

    Desholm, M.

    2003-03-01

    This report presents data from equipment tests and software development for the Thermal Animal Detection System (TADS) development project: 'Development of a method for estimating collision frequency between migrating birds and offshore wind turbines'. The technical tests were performed to investigate the performance of remote controlling, video file compression tool and physical stress of the thermal camera when operating outdoors and under the real time vibration conditions at a 2 MW turbine. Furthermore, experimental tests on birds were performed to describe the decreasing detectability with distance on free flying birds, the performance of the thermal camera during poor visibility, and finally, the performance of the thermal sensor software developed for securing high -quality data. In general, it can be concluded that the thermal camera and its related hardware and software, the TADS, are capable of recording migrating birds approaching the rotating blades of a turbine, even under conditions with poor visibility. If the TADS is used in a vertical viewing scenario it would comply with the requirements for a setup used for estimating the avian collision frequency at offshore wind turbines. (au)

  20. Thermal animal detection system (TADS)

    Energy Technology Data Exchange (ETDEWEB)

    Desholm, M

    2003-03-01

    This report presents data from equipment tests and software development for the Thermal Animal Detection System (TADS) development project: 'Development of a method for estimating collision frequency between migrating birds and offshore wind turbines'. The technical tests were performed to investigate the performance of remote controlling, video file compression tool and physical stress of the thermal camera when operating outdoors and under the real time vibration conditions at a 2 MW turbine. Furthermore, experimental tests on birds were performed to describe the decreasing detectability with distance on free flying birds, the performance of the thermal camera during poor visibility, and finally, the performance of the thermal sensor software developed for securing high -quality data. In general, it can be concluded that the thermal camera and its related hardware and software, the TADS, are capable of recording migrating birds approaching the rotating blades of a turbine, even under conditions with poor visibility. If the TADS is used in a vertical viewing scenario it would comply with the requirements for a setup used for estimating the avian collision frequency at offshore wind turbines. (au)

  1. Technological learning in bioenergy systems

    International Nuclear Information System (INIS)

    Junginger, Martin; Visser, Erika de; Hjort-Gregersen, Kurt; Koornneef, Joris; Raven, Rob; Faaij, Andre; Turkenburg, Wim

    2006-01-01

    The main goal of this article is to determine whether cost reductions in different bioenergy systems can be quantified using the experience curve approach, and how specific issues (arising from the complexity of biomass energy systems) can be addressed. This is pursued by case studies on biofuelled combined heat and power (CHP) plants in Sweden, global development of fluidized bed boilers and Danish biogas plants. As secondary goal, the aim is to identify learning mechanisms behind technology development and cost reduction for the biomass energy systems investigated. The case studies reveal large difficulties to devise empirical experience curves for investment costs of biomass-fuelled power plants. To some extent, this is due to lack of (detailed) data. The main reason, however, are varying plant costs due to differences in scale, fuel type, plant layout, region etc. For fluidized bed boiler plants built on a global level, progress ratios (PRs) for the price of entire plants lies approximately between 90-93% (which is typical for large plant-like technologies). The costs for the boiler section alone was found to decline much faster. The experience curve approach delivers better results, when the production costs of the final energy carrier are analyzed. Electricity from biofuelled CHP-plants yields PRs of 91-92%, i.e. an 8-9% reduction of electricity production costs with each cumulative doubling of electricity production. The experience curve for biogas production displays a PR of 85% from 1984 to the beginning of 1990, and then levels to approximately 100% until 2002. For technologies developed on a local level (e.g. biogas plants), learning-by-using and learning-by-interacting are important learning mechanism, while for CHP plants utilizing fluidized bed boilers, upscaling is probably one of the main mechanisms behind cost reductions

  2. Cosmic String Detection with Tree-Based Machine Learning

    Science.gov (United States)

    Vafaei Sadr, A.; Farhang, M.; Movahed, S. M. S.; Bassett, B.; Kunz, M.

    2018-05-01

    We explore the use of random forest and gradient boosting, two powerful tree-based machine learning algorithms, for the detection of cosmic strings in maps of the cosmic microwave background (CMB), through their unique Gott-Kaiser-Stebbins effect on the temperature anisotropies. The information in the maps is compressed into feature vectors before being passed to the learning units. The feature vectors contain various statistical measures of the processed CMB maps that boost cosmic string detectability. Our proposed classifiers, after training, give results similar to or better than claimed detectability levels from other methods for string tension, Gμ. They can make 3σ detection of strings with Gμ ≳ 2.1 × 10-10 for noise-free, 0.9΄-resolution CMB observations. The minimum detectable tension increases to Gμ ≳ 3.0 × 10-8 for a more realistic, CMB S4-like (II) strategy, improving over previous results.

  3. Advanced Ground Systems Maintenance Anomaly Detection

    Data.gov (United States)

    National Aeronautics and Space Administration — The Inductive Monitoring System (IMS) software utilizes techniques from the fields of model-based reasoning, machine learning, and data mining to build system...

  4. Localization-Aware Active Learning for Object Detection

    OpenAIRE

    Kao, Chieh-Chi; Lee, Teng-Yok; Sen, Pradeep; Liu, Ming-Yu

    2018-01-01

    Active learning - a class of algorithms that iteratively searches for the most informative samples to include in a training dataset - has been shown to be effective at annotating data for image classification. However, the use of active learning for object detection is still largely unexplored as determining informativeness of an object-location hypothesis is more difficult. In this paper, we address this issue and present two metrics for measuring the informativeness of an object hypothesis,...

  5. Deep Learning for Hate Speech Detection in Tweets

    OpenAIRE

    Badjatiya, Pinkesh; Gupta, Shashank; Gupta, Manish; Varma, Vasudeva

    2017-01-01

    Hate speech detection on Twitter is critical for applications like controversial event extraction, building AI chatterbots, content recommendation, and sentiment analysis. We define this task as being able to classify a tweet as racist, sexist or neither. The complexity of the natural language constructs makes this task very challenging. We perform extensive experiments with multiple deep learning architectures to learn semantic word embeddings to handle this complexity. Our experiments on a ...

  6. Learning Management Systems on Blended Learning Courses

    DEFF Research Database (Denmark)

    Kuran, Mehmet Şükrü; Pedersen, Jens Myrup; Elsner, Raphael

    2017-01-01

    LMSes, Moodle, Blackboard Learn, Canvas, and Stud.IP with respect to these. We explain how these features were utilized to increase the efficiency, tractability, and quality of experience of the course. We found that an LMS with advanced features such as progress tracking, modular course support...

  7. Generalized Detectability for Discrete Event Systems

    Science.gov (United States)

    Shu, Shaolong; Lin, Feng

    2011-01-01

    In our previous work, we investigated detectability of discrete event systems, which is defined as the ability to determine the current and subsequent states of a system based on observation. For different applications, we defined four types of detectabilities: (weak) detectability, strong detectability, (weak) periodic detectability, and strong periodic detectability. In this paper, we extend our results in three aspects. (1) We extend detectability from deterministic systems to nondeterministic systems. Such a generalization is necessary because there are many systems that need to be modeled as nondeterministic discrete event systems. (2) We develop polynomial algorithms to check strong detectability. The previous algorithms are based on observer whose construction is of exponential complexity, while the new algorithms are based on a new automaton called detector. (3) We extend detectability to D-detectability. While detectability requires determining the exact state of a system, D-detectability relaxes this requirement by asking only to distinguish certain pairs of states. With these extensions, the theory on detectability of discrete event systems becomes more applicable in solving many practical problems. PMID:21691432

  8. E-Learning Systems, Environments and Approaches

    OpenAIRE

    Isaias, P.; Spector, J.M.; Ifenthaler, D.; Sampson, D.G.

    2015-01-01

    The volume consists of twenty-five chapters selected from among peer-reviewed papers presented at the CELDA (Cognition and Exploratory Learning in the Digital Age) 2013 Conference held in Fort Worth, Texas, USA, in October 2013 and also from world class scholars in e-learning systems, environments and approaches. The following sub-topics are included: Exploratory Learning Technologies (Part I), e-Learning social web design (Part II), Learner communities through e-Learning implementations (Par...

  9. Fluorescence detection system for microfluidic droplets

    Science.gov (United States)

    Chen, Binyu; Han, Xiaoming; Su, Zhen; Liu, Quanjun

    2018-05-01

    In microfluidic detection technology, because of the universality of optical methods in laboratory, optical detection is an attractive solution for microfluidic chip laboratory equipment. In addition, the equipment with high stability and low cost can be realized by integrating appropriate optical detection technology on the chip. This paper reports a detection system for microfluidic droplets. Photomultiplier tubes (PMT) is used as a detection device to improve the sensitivity of detection. This system improves the signal to noise ratio by software filtering and spatial filter. The fluorescence intensity is proportional to the concentration of the fluorescence and intensity of the laser. The fluorescence micro droplets of different concentrations can be distinguished by this system.

  10. Detecting Vessels Carrying Migrants Using Machine Learning

    Science.gov (United States)

    Sfyridis, A.; Cheng, T.; Vespe, M.

    2017-10-01

    Political instability, conflicts and inequalities result into significant flows of people worldwide, moving to different countries in search of a better life, safety or to be reunited with their families. Irregular crossings into Europe via sea routes, despite not being new, have recently increased together with the loss of lives of people in the attempt to reach EU shores. This highlights the need to find ways to improve the understanding of what is happening at sea. This paper, intends to expand the knowledge available on practices among smugglers and contribute to early warning and maritime situational awareness. By identifying smuggling techniques and based on anomaly detection methods, behaviours of interest are modelled and one class support vector machines are used to classify unlabelled data and detect potential smuggling vessels. Nine vessels are identified as potentially carrying irregular migrants and refugees. Though, further inspection of the results highlights possible misclassifications caused by data gaps and limited knowledge on smuggling tactics. Accepted classifications are considered subject to further investigation by the authorities.

  11. DETECTING VESSELS CARRYING MIGRANTS USING MACHINE LEARNING

    Directory of Open Access Journals (Sweden)

    A. Sfyridis

    2017-10-01

    Full Text Available Political instability, conflicts and inequalities result into significant flows of people worldwide, moving to different countries in search of a better life, safety or to be reunited with their families. Irregular crossings into Europe via sea routes, despite not being new, have recently increased together with the loss of lives of people in the attempt to reach EU shores. This highlights the need to find ways to improve the understanding of what is happening at sea. This paper, intends to expand the knowledge available on practices among smugglers and contribute to early warning and maritime situational awareness. By identifying smuggling techniques and based on anomaly detection methods, behaviours of interest are modelled and one class support vector machines are used to classify unlabelled data and detect potential smuggling vessels. Nine vessels are identified as potentially carrying irregular migrants and refugees. Though, further inspection of the results highlights possible misclassifications caused by data gaps and limited knowledge on smuggling tactics. Accepted classifications are considered subject to further investigation by the authorities.

  12. Nuclear fuel element leak detection system

    International Nuclear Information System (INIS)

    John, C.D. Jr.

    1978-01-01

    Disclosed is a leak detection system integral with a wall of a building used to fabricate nuclear fuel elements for detecting radiation leakage from the nuclear fuel elements as the fuel elements exit the building. The leak detecting system comprises a shielded compartment constructed to withstand environmental hazards extending into a similarly constructed building and having sealed doors on both ends along with leak detecting apparatus connected to the compartment. The leak detecting system provides a system for removing a nuclear fuel element from its fabrication building while testing for radiation leaks in the fuel element

  13. Baby Cry Detection in Domestic Environment using Deep Learning

    OpenAIRE

    Ijzerman, Hans

    2017-01-01

    Automatic detection of a baby cry in audio signals is an essential step in applications such as remote baby monitoring. It is also important for researchers, who study the relation between baby cry patterns and various health or developmental parameters. In this paper, we propose two machine-learning algorithms for automatic detection of baby cry in audio recordings. The first algorithm is a low-complexity logistic regression classifier, used as a reference. To train this classifier, we extra...

  14. Critical Points in Distance Learning System

    Directory of Open Access Journals (Sweden)

    Airina Savickaitė

    2013-08-01

    Full Text Available Purpose – This article presents the results of distance learning system analysis, i.e. the critical elements of the distance learning system. The critical points of distance learning are a part of distance education online environment interactivity/community process model. The most important is the fact that the critical point is associated with distance learning participants. Design/methodology/approach – Comparative review of articles and analysis of distance learning module. Findings – A modern man is a lifelong learner and distance learning is a way to be a modern person. The focus on a learner and feedback is the most important thing of learning distance system. Also, attention should be paid to the lecture-appropriate knowledge and ability to convey information. Distance system adaptation is the way to improve the learner’s learning outcomes. Research limitations/implications – Different learning disciplines and learning methods may have different critical points. Practical implications – The information of analysis could be important for both lecturers and students, who studies distance education systems. There are familiar critical points which may deteriorate the quality of learning. Originality/value – The study sought to develop remote systems for applications in order to improve the quality of knowledge. Keywords: distance learning, process model, critical points. Research type: review of literature and general overview.

  15. NIST Special Publication on Intrusion Detection Systems

    National Research Council Canada - National Science Library

    Bace, Rebecca Gurley

    2001-01-01

    Intrusion detection systems (IDSs) are software or hardware systems that automate the process of monitoring the events occurring in a computer system or network, analyzing them for signs of security problems...

  16. Real-Time Barcode Detection and Classification Using Deep Learning

    DEFF Research Database (Denmark)

    Hansen, Daniel Kold; Nasrollahi, Kamal; Rasmussen, Christoffer Bøgelund

    2017-01-01

    Barcodes, in their different forms, can be found on almost any packages available in the market. Detecting and then decoding of barcodes have therefore great applications. We describe how to adapt the state-of-the- art deep learning-based detector of You Only Look Once (YOLO) for the purpose...

  17. Detection of player learning curve in a car driving game

    NARCIS (Netherlands)

    Bontchev, Boyan; Vassileva, Dessislava

    2018-01-01

    Detection of learning curves of player metrics is very important for the serious (or so called applied) games, because it provides an indicator representing how players master the game tasks by acquiring cognitive abilities, knowledge, and necessary skills for solving the game challenges. Real

  18. Defeating crypsis: detection and learning of camouflage strategies.

    Directory of Open Access Journals (Sweden)

    Jolyon Troscianko

    Full Text Available Camouflage is perhaps the most widespread defence against predators in nature and an active area of interdisciplinary research. Recent work has aimed to understand what camouflage types exist (e.g. background matching, disruptive, and distractive patterns and their effectiveness. However, work has almost exclusively focused on the efficacy of these strategies in preventing initial detection, despite the fact that predators often encounter the same prey phenotype repeatedly, affording them opportunities to learn to find those prey more effectively. The overall value of a camouflage strategy may, therefore, reflect both its ability to prevent detection by predators and resist predator learning. We conducted four experiments with humans searching for hidden targets of different camouflage types (disruptive, distractive, and background matching of various contrast levels over a series of touch screen trials. As with previous work, disruptive coloration was the most successful method of concealment overall, especially with relatively high contrast patterns, whereas potentially distractive markings were either neutral or costly. However, high contrast patterns incurred faster decreases in detection times over trials compared to other stimuli. In addition, potentially distractive markings were sometimes learnt more slowly than background matching markings, despite being found more readily overall. Finally, learning effects were highly dependent upon the experimental paradigm, including the number of prey types seen and whether subjects encountered targets simultaneously or sequentially. Our results show that the survival advantage of camouflage strategies reflects both their ability to avoid initial detection (sensory mechanisms and predator learning (perceptual mechanisms.

  19. Defeating crypsis: detection and learning of camouflage strategies.

    Science.gov (United States)

    Troscianko, Jolyon; Lown, Alice E; Hughes, Anna E; Stevens, Martin

    2013-01-01

    Camouflage is perhaps the most widespread defence against predators in nature and an active area of interdisciplinary research. Recent work has aimed to understand what camouflage types exist (e.g. background matching, disruptive, and distractive patterns) and their effectiveness. However, work has almost exclusively focused on the efficacy of these strategies in preventing initial detection, despite the fact that predators often encounter the same prey phenotype repeatedly, affording them opportunities to learn to find those prey more effectively. The overall value of a camouflage strategy may, therefore, reflect both its ability to prevent detection by predators and resist predator learning. We conducted four experiments with humans searching for hidden targets of different camouflage types (disruptive, distractive, and background matching of various contrast levels) over a series of touch screen trials. As with previous work, disruptive coloration was the most successful method of concealment overall, especially with relatively high contrast patterns, whereas potentially distractive markings were either neutral or costly. However, high contrast patterns incurred faster decreases in detection times over trials compared to other stimuli. In addition, potentially distractive markings were sometimes learnt more slowly than background matching markings, despite being found more readily overall. Finally, learning effects were highly dependent upon the experimental paradigm, including the number of prey types seen and whether subjects encountered targets simultaneously or sequentially. Our results show that the survival advantage of camouflage strategies reflects both their ability to avoid initial detection (sensory mechanisms) and predator learning (perceptual mechanisms).

  20. Fusion of Heterogeneous Intrusion Detection Systems for Network Attack Detection

    Directory of Open Access Journals (Sweden)

    Jayakumar Kaliappan

    2015-01-01

    Full Text Available An intrusion detection system (IDS helps to identify different types of attacks in general, and the detection rate will be higher for some specific category of attacks. This paper is designed on the idea that each IDS is efficient in detecting a specific type of attack. In proposed Multiple IDS Unit (MIU, there are five IDS units, and each IDS follows a unique algorithm to detect attacks. The feature selection is done with the help of genetic algorithm. The selected features of the input traffic are passed on to the MIU for processing. The decision from each IDS is termed as local decision. The fusion unit inside the MIU processes all the local decisions with the help of majority voting rule and makes the final decision. The proposed system shows a very good improvement in detection rate and reduces the false alarm rate.

  1. Fully automatic AI-based leak detection system

    Energy Technology Data Exchange (ETDEWEB)

    Tylman, Wojciech; Kolczynski, Jakub [Dept. of Microelectronics and Computer Science, Technical University of Lodz in Poland, ul. Wolczanska 221/223, Lodz (Poland); Anders, George J. [Kinectrics Inc., 800 Kipling Ave., Toronto, Ontario M8Z 6C4 (Canada)

    2010-09-15

    This paper presents a fully automatic system intended to detect leaks of dielectric fluid in underground high-pressure, fluid-filled (HPFF) cables. The system combines a number of artificial intelligence (AI) and data processing techniques to achieve high detection capabilities for various rates of leaks, including leaks as small as 15 l per hour. The system achieves this level of precision mainly thanks to a novel auto-tuning procedure, enabling learning of the Bayesian network - the decision-making component of the system - using simulated leaks of various rates. Significant new developments extending the capabilities of the original leak detection system described in and form the basis of this paper. Tests conducted on the real-life HPFF cable system in New York City are also discussed. (author)

  2. Authoring Systems Delivering Reusable Learning Objects

    Directory of Open Access Journals (Sweden)

    George Nicola Sammour

    2009-10-01

    Full Text Available A three layer e-learning course development model has been defined based on a conceptual model of learning content object. It starts by decomposing the learning content into small chunks which are initially placed in a hierarchic structure of units and blocks. The raw content components, being the atomic learning objects (ALO, were linked to the blocks and are structured in the database. We set forward a dynamic generation of LO's using re-usable e-learning raw materials or ALO’s In that view we need a LO authoring/ assembling system fitting the requirements of interoperability and reusability and starting from selecting the raw learning content from the learning materials content database. In practice authoring systems are used to develop e-learning courses. The company EDUWEST has developed an authoring system that is database based and will be SCORM compliant in the near future.

  3. Establishment of a Learning Management System

    International Nuclear Information System (INIS)

    Han, K. W.; Kim, Y. T.; Lee, E. J.; Min, B. J.

    2006-01-01

    A web-based learning management system (LMS) has been established to address the need of customized education and training of Nuclear Training Center (NTC) of KAERI. The LMS is designed to deal with various learning types (e.g. on-line, off-line and blended) and a practically comprehensive learning activity cycle (e.g. course preparation, registration, learning, and postlearning) as well as to be user-friendly. A test with an example course scenario on the established system has shown its satisfactory performance. This paper discusses details of the established webbased learning management system in terms of development approach and functions of the LMS

  4. System for detecting nuclear explosions

    International Nuclear Information System (INIS)

    Rawls, L.E.

    1978-01-01

    Apparatus for detecting underground nuclear explosions is described that is comprised of an antenna located in the dielectric substance of a deep waveguide in the earth and adapted to detect low frequency electromagnetic waves generated by a nuclear explosion, the deep waveguide comprising the high conductivity upper sedimentary layers of the earth, the dielectric basement rock, and a high conductivity layer of basement rock due to the increased temperature thereof at great depths, and means for receiving the electromagnetic waves detected by said antenna means

  5. Automated Inattention and Fatigue Detection System in Distance Education for Elementary School Students

    Science.gov (United States)

    Hwang, Kuo-An; Yang, Chia-Hao

    2009-01-01

    Most courses based on distance learning focus on the cognitive domain of learning. Because students are sometimes inattentive or tired, they may neglect the attention goal of learning. This study proposes an auto-detection and reinforcement mechanism for the distance-education system based on the reinforcement teaching strategy. If a student is…

  6. Challenges, issues and trends in fall detection systems

    Science.gov (United States)

    2013-01-01

    Since falls are a major public health problem among older people, the number of systems aimed at detecting them has increased dramatically over recent years. This work presents an extensive literature review of fall detection systems, including comparisons among various kinds of studies. It aims to serve as a reference for both clinicians and biomedical engineers planning or conducting field investigations. Challenges, issues and trends in fall detection have been identified after the reviewing work. The number of studies using context-aware techniques is still increasing but there is a new trend towards the integration of fall detection into smartphones as well as the use of machine learning methods in the detection algorithm. We have also identified challenges regarding performance under real-life conditions, usability, and user acceptance as well as issues related to power consumption, real-time operations, sensing limitations, privacy and record of real-life falls. PMID:23829390

  7. Homodyne detection of holographic memory systems

    Science.gov (United States)

    Urness, Adam C.; Wilson, William L.; Ayres, Mark R.

    2014-09-01

    We present a homodyne detection system implemented for a page-wise holographic memory architecture. Homodyne detection by holographic memory systems enables phase quadrature multiplexing (doubling address space), and lower exposure times (increasing read transfer rates). It also enables phase modulation, which improves signal-to-noise ratio (SNR) to further increase data capacity. We believe this is the first experimental demonstration of homodyne detection for a page-wise holographic memory system suitable for a commercial design.

  8. Repetitive learning control of continuous chaotic systems

    International Nuclear Information System (INIS)

    Chen Maoyin; Shang Yun; Zhou Donghua

    2004-01-01

    Combining a shift method and the repetitive learning strategy, a repetitive learning controller is proposed to stabilize unstable periodic orbits (UPOs) within chaotic attractors in the sense of least mean square. If nonlinear parts in chaotic systems satisfy Lipschitz condition, the proposed controller can be simplified into a simple proportional repetitive learning controller

  9. Expert Students in Social Learning Management Systems

    Science.gov (United States)

    Avogadro, Paolo; Calegari, Silvia; Dominoni, Matteo Alessandro

    2016-01-01

    Purpose: A social learning management system (social LMS) is a tool which favors social interactions and allows scholastic institutions to supervise and guide the learning process. The inclusion of the social feature to a "normal" LMS leads to the creation of educational social networks (EduSN), where the students interact and learn. The…

  10. A deep learning approach for fetal QRS complex detection.

    Science.gov (United States)

    Zhong, Wei; Liao, Lijuan; Guo, Xuemei; Wang, Guoli

    2018-04-20

    Non-invasive foetal electrocardiography (NI-FECG) has the potential to provide more additional clinical information for detecting and diagnosing fetal diseases. We propose and demonstrate a deep learning approach for fetal QRS complex detection from raw NI-FECG signals by using a convolutional neural network (CNN) model. The main objective is to investigate whether reliable fetal QRS complex detection performance can still be obtained from features of single-channel NI-FECG signals, without canceling maternal ECG (MECG) signals. A deep learning method is proposed for recognizing fetal QRS complexes. Firstly, we collect data from set-a of the PhysioNet/computing in Cardiology Challenge database. The sample entropy method is used for signal quality assessment. Part of the bad quality signals is excluded in the further analysis. Secondly, in the proposed method, the features of raw NI-FECG signals are normalized before they are fed to a CNN classifier to perform fetal QRS complex detection. We use precision, recall, F-measure and accuracy as the evaluation metrics to assess the performance of fetal QRS complex detection. The proposed deep learning method can achieve relatively high precision (75.33%), recall (80.54%), and F-measure scores (77.85%) compared with three other well-known pattern classification methods, namely KNN, naive Bayes and SVM. the proposed deep learning method can attain reliable fetal QRS complex detection performance from the raw NI-FECG signals without canceling MECG signals. In addition, the influence of different activation functions and signal quality assessment on classification performance are evaluated, and results show that Relu outperforms the Sigmoid and Tanh on this particular task, and better classification performance is obtained with the signal quality assessment step in this study.

  11. Learning Rich Features from RGB-D Images for Object Detection and Segmentation

    OpenAIRE

    Gupta, Saurabh; Girshick, Ross; Arbeláez, Pablo; Malik, Jitendra

    2014-01-01

    In this paper we study the problem of object detection for RGB-D images using semantically rich image and depth features. We propose a new geocentric embedding for depth images that encodes height above ground and angle with gravity for each pixel in addition to the horizontal disparity. We demonstrate that this geocentric embedding works better than using raw depth images for learning feature representations with convolutional neural networks. Our final object detection system achieves an av...

  12. Automated sleep stage detection with a classical and a neural learning algorithm--methodological aspects.

    Science.gov (United States)

    Schwaibold, M; Schöchlin, J; Bolz, A

    2002-01-01

    For classification tasks in biosignal processing, several strategies and algorithms can be used. Knowledge-based systems allow prior knowledge about the decision process to be integrated, both by the developer and by self-learning capabilities. For the classification stages in a sleep stage detection framework, three inference strategies were compared regarding their specific strengths: a classical signal processing approach, artificial neural networks and neuro-fuzzy systems. Methodological aspects were assessed to attain optimum performance and maximum transparency for the user. Due to their effective and robust learning behavior, artificial neural networks could be recommended for pattern recognition, while neuro-fuzzy systems performed best for the processing of contextual information.

  13. Accuracy comparison among different machine learning techniques for detecting malicious codes

    Science.gov (United States)

    Narang, Komal

    2016-03-01

    In this paper, a machine learning based model for malware detection is proposed. It can detect newly released malware i.e. zero day attack by analyzing operation codes on Android operating system. The accuracy of Naïve Bayes, Support Vector Machine (SVM) and Neural Network for detecting malicious code has been compared for the proposed model. In the experiment 400 benign files, 100 system files and 500 malicious files have been used to construct the model. The model yields the best accuracy 88.9% when neural network is used as classifier and achieved 95% and 82.8% accuracy for sensitivity and specificity respectively.

  14. Linear System Control Using Stochastic Learning Automata

    Science.gov (United States)

    Ziyad, Nigel; Cox, E. Lucien; Chouikha, Mohamed F.

    1998-01-01

    This paper explains the use of a Stochastic Learning Automata (SLA) to control switching between three systems to produce the desired output response. The SLA learns the optimal choice of the damping ratio for each system to achieve a desired result. We show that the SLA can learn these states for the control of an unknown system with the proper choice of the error criteria. The results of using a single automaton are compared to using multiple automata.

  15. Micro Learning: A Modernized Education System

    Directory of Open Access Journals (Sweden)

    Omer Jomah

    2016-03-01

    Full Text Available Learning is an understanding of how the human brain is wired to learning rather than to an approach or a system. It is one of the best and most frequent approaches for the 21st century learners. Micro learning is more interesting due to its way of teaching and learning the content in a small, very specific burst. Here the learners decide what and when to learn. Content, time, curriculum, form, process, mediality, and learning type are the dimensions of micro learning. Our paper will discuss about micro learning and about the micro-content management system. The study will reflect the views of different users, and will analyze the collected data. Finally, it will be concluded with its pros and cons. 

  16. Exploiting ensemble learning for automatic cataract detection and grading.

    Science.gov (United States)

    Yang, Ji-Jiang; Li, Jianqiang; Shen, Ruifang; Zeng, Yang; He, Jian; Bi, Jing; Li, Yong; Zhang, Qinyan; Peng, Lihui; Wang, Qing

    2016-02-01

    Cataract is defined as a lenticular opacity presenting usually with poor visual acuity. It is one of the most common causes of visual impairment worldwide. Early diagnosis demands the expertise of trained healthcare professionals, which may present a barrier to early intervention due to underlying costs. To date, studies reported in the literature utilize a single learning model for retinal image classification in grading cataract severity. We present an ensemble learning based approach as a means to improving diagnostic accuracy. Three independent feature sets, i.e., wavelet-, sketch-, and texture-based features, are extracted from each fundus image. For each feature set, two base learning models, i.e., Support Vector Machine and Back Propagation Neural Network, are built. Then, the ensemble methods, majority voting and stacking, are investigated to combine the multiple base learning models for final fundus image classification. Empirical experiments are conducted for cataract detection (two-class task, i.e., cataract or non-cataractous) and cataract grading (four-class task, i.e., non-cataractous, mild, moderate or severe) tasks. The best performance of the ensemble classifier is 93.2% and 84.5% in terms of the correct classification rates for cataract detection and grading tasks, respectively. The results demonstrate that the ensemble classifier outperforms the single learning model significantly, which also illustrates the effectiveness of the proposed approach. Copyright © 2015 Elsevier Ireland Ltd. All rights reserved.

  17. Machine Learning Based Classifier for Falsehood Detection

    Science.gov (United States)

    Mallikarjun, H. M.; Manimegalai, P., Dr.; Suresh, H. N., Dr.

    2017-08-01

    The investigation of physiological techniques for Falsehood identification tests utilizing the enthusiastic aggravations started as a part of mid 1900s. The need of Falsehood recognition has been a piece of our general public from hundreds of years back. Different requirements drifted over the general public raising the need to create trick evidence philosophies for Falsehood identification. The established similar addressing tests have been having a tendency to gather uncertain results against which new hearty strategies are being explored upon for acquiring more productive Falsehood discovery set up. Electroencephalography (EEG) is a non-obtrusive strategy to quantify the action of mind through the anodes appended to the scalp of a subject. Electroencephalogram is a record of the electric signs produced by the synchronous activity of mind cells over a timeframe. The fundamental goal is to accumulate and distinguish the important information through this action which can be acclimatized for giving surmising to Falsehood discovery in future analysis. This work proposes a strategy for Falsehood discovery utilizing EEG database recorded on irregular people of various age gatherings and social organizations. The factual investigation is directed utilizing MATLAB v-14. It is a superior dialect for specialized registering which spares a considerable measure of time with streamlined investigation systems. In this work center is made on Falsehood Classification by Support Vector Machine (SVM). 72 Samples are set up by making inquiries from standard poll with a Wright and wrong replies in a diverse era from the individual in wearable head unit. 52 samples are trained and 20 are tested. By utilizing Bluetooth based Neurosky’s Mindwave kit, brain waves are recorded and qualities are arranged appropriately. In this work confusion matrix is derived by matlab programs and accuracy of 56.25 % is achieved.

  18. ENGINEERING OF UNIVERSITY INTELLIGENT LEARNING SYSTEMS

    Directory of Open Access Journals (Sweden)

    Vasiliy M. Trembach

    2016-01-01

    Full Text Available In the article issues of engineering intelligent tutoring systems of University with adaptation are considered. The article also dwells on some modern approaches to engineering of information systems. It shows the role of engineering e-learning devices (systems in system engineering. The article describes the basic principles of system engineering and these principles are expanded regarding to intelligent information systems. The structure of intelligent learning systems with adaptation of the individual learning environments based on services is represented in the article.

  19. Learning and the Instructional System

    Science.gov (United States)

    Kozma, Robert B.

    1977-01-01

    Faculty members can use information about six components of the learning situation to increase student learning. The nature, function, and interrelationships of the following elements are described: instructor, content, medium, student, evaluation, environment, and implementation. (Editor/LBH)

  20. Incidental orthographic learning during a color detection task.

    Science.gov (United States)

    Protopapas, Athanassios; Mitsi, Anna; Koustoumbardis, Miltiadis; Tsitsopoulou, Sofia M; Leventi, Marianna; Seitz, Aaron R

    2017-09-01

    Orthographic learning refers to the acquisition of knowledge about specific spelling patterns forming words and about general biases and constraints on letter sequences. It is thought to occur by strengthening simultaneously activated visual and phonological representations during reading. Here we demonstrate that a visual perceptual learning procedure that leaves no time for articulation can result in orthographic learning evidenced in improved reading and spelling performance. We employed task-irrelevant perceptual learning (TIPL), in which the stimuli to be learned are paired with an easy task target. Assorted line drawings and difficult-to-spell words were presented in red color among sequences of other black-colored words and images presented in rapid succession, constituting a fast-TIPL procedure with color detection being the explicit task. In five experiments, Greek children in Grades 4-5 showed increased recognition of words and images that had appeared in red, both during and after the training procedure, regardless of within-training testing, and also when targets appeared in blue instead of red. Significant transfer to reading and spelling emerged only after increased training intensity. In a sixth experiment, children in Grades 2-3 showed generalization to words not presented during training that carried the same derivational affixes as in the training set. We suggest that reinforcement signals related to detection of the target stimuli contribute to the strengthening of orthography-phonology connections beyond earlier levels of visually-based orthographic representation learning. These results highlight the potential of perceptual learning procedures for the reinforcement of higher-level orthographic representations. Copyright © 2017 The Authors. Published by Elsevier B.V. All rights reserved.

  1. Damage Detection and Deteriorating Structural Systems

    DEFF Research Database (Denmark)

    Long, Lijia; Thöns, Sebastian; Döhler, Michael

    2017-01-01

    This paper addresses the quantification of the value of damage detection system and algorithm information on the basis of Value of Information (VoI) analysis to enhance the benefit of damage detection information by providing the basis for its optimization before it is performed and implemented....... The approach of the quantification the value of damage detection information builds upon the Bayesian decision theory facilitating the utilization of damage detection performance models, which describe the information and its precision on structural system level, facilitating actions to ensure the structural...... detection information is determined utilizing Bayesian updating. The damage detection performance is described with the probability of indication for different component and system damage states taking into account type 1 and type 2 errors. The value of damage detection information is then calculated...

  2. Ship Detection Using Transfer Learned Single Shot Multi Box Detector

    Directory of Open Access Journals (Sweden)

    Nie Gu-Hong

    2017-01-01

    Full Text Available Ship detection in satellite images is a challenging task. In this paper, we introduce a transfer learned Single Shot MultiBox Detector (SSD for ship detection. To this end, a state-of-the-art object detection model pre-trained from a large number of natural images was transfer learned for ship detection with limited labeled satellite images. To the best of our knowledge, this could be one of the first studies which introduce SSD into ship detection on satellite images. Experiments demonstrated that our method could achieve 87.9% AP at 47 FPS using NVIDIA TITAN X. In comparison with Faster R-CNN, 6.7% AP improvement could be achieved. Effects of the observation resolution has also been studied with the changing input sizes among 300 × 300, 600 × 600 and 900 × 900. It has been noted that the detection accuracy declined sharply with the decreasing resolution that is mainly caused by the missing small ships.

  3. US Army Nuclear Burst Detection System (NBDS)

    International Nuclear Information System (INIS)

    Glaser, R.F.

    1980-07-01

    The Nuclear Burst Detection System (NBDS) was developed to meet the Army requirements of an unattended, automatic nuclear burst reporting system. It provides pertinent data for battlefield commanders on a timely basis with high reliability

  4. Apriori-based network intrusion detection system

    International Nuclear Information System (INIS)

    Wang Wenjin; Liu Junrong; Liu Baoxu

    2012-01-01

    With the development of network communication technology, more and more social activities run by Internet. In the meantime, the network information security is getting increasingly serious. Intrusion Detection System (IDS) has greatly improved the general security level of whole network. But there are still many problem exists in current IDS, e.g. high leak rate detection/false alarm rates and feature library need frequently upgrade. This paper presents an association-rule based IDS. This system can detect unknown attack by generate rules from training data. Experiment in last chapter proved the system has great accuracy on unknown attack detection. (authors)

  5. Counterfeit Electronics Detection Using Image Processing and Machine Learning

    Science.gov (United States)

    Asadizanjani, Navid; Tehranipoor, Mark; Forte, Domenic

    2017-01-01

    Counterfeiting is an increasing concern for businesses and governments as greater numbers of counterfeit integrated circuits (IC) infiltrate the global market. There is an ongoing effort in experimental and national labs inside the United States to detect and prevent such counterfeits in the most efficient time period. However, there is still a missing piece to automatically detect and properly keep record of detected counterfeit ICs. Here, we introduce a web application database that allows users to share previous examples of counterfeits through an online database and to obtain statistics regarding the prevalence of known defects. We also investigate automated techniques based on image processing and machine learning to detect different physical defects and to determine whether or not an IC is counterfeit.

  6. Counterfeit Electronics Detection Using Image Processing and Machine Learning

    International Nuclear Information System (INIS)

    Asadizanjani, Navid; Tehranipoor, Mark; Forte, Domenic

    2017-01-01

    Counterfeiting is an increasing concern for businesses and governments as greater numbers of counterfeit integrated circuits (IC) infiltrate the global market. There is an ongoing effort in experimental and national labs inside the United States to detect and prevent such counterfeits in the most efficient time period. However, there is still a missing piece to automatically detect and properly keep record of detected counterfeit ICs. Here, we introduce a web application database that allows users to share previous examples of counterfeits through an online database and to obtain statistics regarding the prevalence of known defects. We also investigate automated techniques based on image processing and machine learning to detect different physical defects and to determine whether or not an IC is counterfeit. (paper)

  7. USE OF FACIAL EMOTION RECOGNITION IN E-LEARNING SYSTEMS

    Directory of Open Access Journals (Sweden)

    Uğur Ayvaz

    2017-09-01

    Full Text Available Since the personal computer usage and internet bandwidth are increasing, e-learning systems are also widely spreading. Although e-learning has some advantages in terms of information accessibility, time and place flexibility compared to the formal learning, it does not provide enough face-to-face interactivity between an educator and learners. In this study, we are proposing a hybrid information system, which is combining computer vision and machine learning technologies for visual and interactive e-learning systems. The proposed information system detects emotional states of the learners and gives feedback to an educator about their instant and weighted emotional states based on facial expressions. In this way, the educator will be aware of the general emotional state of the virtual classroom and the system will create a formal learning-like interactive environment. Herein, several classification algorithms were applied to learn instant emotional state and the best accuracy rates were obtained using kNN and SVM algorithms.

  8. Development of the environmental neutron detection system

    International Nuclear Information System (INIS)

    Kume, Kyo

    2002-03-01

    Environmental neutron detection system was proposed and developed. The main goal of this system was set to detect fast and thermal neutrons with the identical detectors setup without degraders. This system consists of a 10 B doped liquid scintillator for n detection and CsI scintillators for simultaneous γ emission from 10 B doped in the liquid scintillator after the n capture reaction. The first setup was optimized for the thermal n detection, while the second setup was for the fast n detection. It was shown that the thermal n flux was obtained in the first setup by using the method of the γ coincidence method with the help of the Monte Carlo calculation. The second setup was designed to improve the detection efficiency for the fast n, and was shown qualitatively that both the pulse shape discrimination and the coincidence methods are efficient. There will be more improvements, particularly for the quantitative discussion. (author)

  9. Deep Recurrent Neural Networks for seizure detection and early seizure detection systems

    Energy Technology Data Exchange (ETDEWEB)

    Talathi, S. S. [Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)

    2017-06-05

    Epilepsy is common neurological diseases, affecting about 0.6-0.8 % of world population. Epileptic patients suffer from chronic unprovoked seizures, which can result in broad spectrum of debilitating medical and social consequences. Since seizures, in general, occur infrequently and are unpredictable, automated seizure detection systems are recommended to screen for seizures during long-term electroencephalogram (EEG) recordings. In addition, systems for early seizure detection can lead to the development of new types of intervention systems that are designed to control or shorten the duration of seizure events. In this article, we investigate the utility of recurrent neural networks (RNNs) in designing seizure detection and early seizure detection systems. We propose a deep learning framework via the use of Gated Recurrent Unit (GRU) RNNs for seizure detection. We use publicly available data in order to evaluate our method and demonstrate very promising evaluation results with overall accuracy close to 100 %. We also systematically investigate the application of our method for early seizure warning systems. Our method can detect about 98% of seizure events within the first 5 seconds of the overall epileptic seizure duration.

  10. Recommender Systems in Technology Enhanced Learning

    NARCIS (Netherlands)

    Manouselis, Nikos; Drachsler, Hendrik; Verbert, Katrien; Santos, Olga

    2010-01-01

    Manouselis, N., Drachsler, H., Verbert, K., & Santos, C. S. (Eds.) (2010). Recommender System in Technology Enhanced Learning. Elsevier Procedia Computer Science: Volume 1, Issue 2. Proceedings of the 1st Workshop on Recommender Systems for Technology Enhanced Learning (RecSysTEL). September, 29-30,

  11. Panorama of Recommender Systems to Support Learning

    NARCIS (Netherlands)

    Drachsler, Hendrik; Verbert, Katrien; Santos, Olga C.; Manouselis, Nikos

    2015-01-01

    This chapter presents an analysis of recommender systems in TechnologyEnhanced Learning along their 15 years existence (2000-2014). All recommender systems considered for the review aim to support educational stakeholders by personalising the learning process. In this meta-review 82 recommender

  12. Reconceptualizing Learning as a Dynamical System.

    Science.gov (United States)

    Ennis, Catherine D.

    1992-01-01

    Dynamical systems theory can increase our understanding of the constantly evolving learning process. Current research using experimental and interpretive paradigms focuses on describing the attractors and constraints stabilizing the educational process. Dynamical systems theory focuses attention on critical junctures in the learning process as…

  13. Divulging Personal Information within Learning Analytics Systems

    Science.gov (United States)

    Ifenthaler, Dirk; Schumacher, Clara

    2015-01-01

    The purpose of this study was to investigate if students are prepared to release any personal data in order to inform learning analytics systems. Besides the well-documented benefits of learning analytics, serious concerns and challenges are associated with the application of these data driven systems. Most notably, empirical evidence regarding…

  14. Online reinforcement learning control for aerospace systems

    NARCIS (Netherlands)

    Zhou, Y.

    2018-01-01

    Reinforcement Learning (RL) methods are relatively new in the field of aerospace guidance, navigation, and control. This dissertation aims to exploit RL methods to improve the autonomy and online learning of aerospace systems with respect to the a priori unknown system and environment, dynamical

  15. Adaptive e-learning system using ontology

    OpenAIRE

    Yarandi, Maryam; Tawil, Abdel-Rahman; Jahankhani, Hossein

    2011-01-01

    This paper proposes an innovative ontological approach to design a personalised e-learning system which creates a tailored workflow for individual learner. Moreover, the learning content and sequencing logic is separated into content model and pedagogical model to increase the reusability and flexibility of the system.

  16. An Evaluation of Machine Learning Methods to Detect Malicious SCADA Communications

    Energy Technology Data Exchange (ETDEWEB)

    Beaver, Justin M [ORNL; Borges, Raymond Charles [ORNL; Buckner, Mark A [ORNL

    2013-01-01

    Critical infrastructure Supervisory Control and Data Acquisition (SCADA) systems were designed to operate on closed, proprietary networks where a malicious insider posed the greatest threat potential. The centralization of control and the movement towards open systems and standards has improved the efficiency of industrial control, but has also exposed legacy SCADA systems to security threats that they were not designed to mitigate. This work explores the viability of machine learning methods in detecting the new threat scenarios of command and data injection. Similar to network intrusion detection systems in the cyber security domain, the command and control communications in a critical infrastructure setting are monitored, and vetted against examples of benign and malicious command traffic, in order to identify potential attack events. Multiple learning methods are evaluated using a dataset of Remote Terminal Unit communications, which included both normal operations and instances of command and data injection attack scenarios.

  17. Real-time petroleum spill detection system

    International Nuclear Information System (INIS)

    Dakin, D.T.

    2001-01-01

    A real-time autonomous oil and fuel spill detection system has been developed to rapidly detect of a wide range of petroleum products floating on, or suspended in water. The system consists of an array of spill detection buoys distributed within the area to be monitored. The buoys are composed of a float and a multispectral fluorometer, which looks up through the top 5 cm of water to detect floating and suspended petroleum products. The buoys communicate to a base station computer that controls the sampling of the buoys and analyses the data from each buoy to determine if a spill has occurred. If statistically significant background petroleum levels are detected, the system raises an oil spill alarm. The system is useful because early detection of a marine oil spill allows for faster containment, thereby minimizing the contaminated area and reducing cleanup costs. This paper also provided test results for biofouling, various petroleum product detection, water turbidity and wave tolerance. The technology has been successfully demonstrated. The UV light source keeps the optic window free from biofouling, and the electronics are fully submerged so there is no risk that the unit could ignite the vapours of a potential oil spill. The system can also tolerate moderately turbid waters and can therefore be used in many rivers, harbours, water intakes and sumps. The system can detect petroleum products with an average thickness of less than 3 micrometers floating on the water surface. 3 refs., 15 figs

  18. Recommender Systems in Technology Enhanced Learning

    NARCIS (Netherlands)

    Manouselis, Nikos; Drachsler, Hendrik; Vuorikari, Riina; Hummel, Hans; Koper, Rob

    2010-01-01

    Manouselis, N., Drachsler, H., Vuorikari, R., Hummel, H. G. K., & Koper, R. (2011). Recommender Systems in Technology Enhanced Learning. In P. B. Kantor, F. Ricci, L. Rokach, & B. Shapira (Eds.), Recommender Systems Handbook (pp. 387-415). Berlin: Springer.

  19. Anomaly Detection for Complex Systems

    Data.gov (United States)

    National Aeronautics and Space Administration — In performance maintenance in large, complex systems, sensor information from sub-components tends to be readily available, and can be used to make predictions about...

  20. Monocular perceptual learning of contrast detection facilitates binocular combination in adults with anisometropic amblyopia

    OpenAIRE

    Chen, Zidong; Li, Jinrong; Liu, Jing; Cai, Xiaoxiao; Yuan, Junpeng; Deng, Daming; Yu, Minbin

    2016-01-01

    Perceptual learning in contrast detection improves monocular visual function in adults with anisometropic amblyopia; however, its effect on binocular combination remains unknown. Given that the amblyopic visual system suffers from pronounced binocular functional loss, it is important to address how the amblyopic visual system responds to such training strategies under binocular viewing conditions. Anisometropic amblyopes (n?=?13) were asked to complete two psychophysical supra-threshold binoc...

  1. Imbalance aware lithography hotspot detection: a deep learning approach

    Science.gov (United States)

    Yang, Haoyu; Luo, Luyang; Su, Jing; Lin, Chenxi; Yu, Bei

    2017-07-01

    With the advancement of very large scale integrated circuits (VLSI) technology nodes, lithographic hotspots become a serious problem that affects manufacture yield. Lithography hotspot detection at the post-OPC stage is imperative to check potential circuit failures when transferring designed patterns onto silicon wafers. Although conventional lithography hotspot detection methods, such as machine learning, have gained satisfactory performance, with the extreme scaling of transistor feature size and layout patterns growing in complexity, conventional methodologies may suffer from performance degradation. For example, manual or ad hoc feature extraction in a machine learning framework may lose important information when predicting potential errors in ultra-large-scale integrated circuit masks. We present a deep convolutional neural network (CNN) that targets representative feature learning in lithography hotspot detection. We carefully analyze the impact and effectiveness of different CNN hyperparameters, through which a hotspot-detection-oriented neural network model is established. Because hotspot patterns are always in the minority in VLSI mask design, the training dataset is highly imbalanced. In this situation, a neural network is no longer reliable, because a trained model with high classification accuracy may still suffer from a high number of false negative results (missing hotspots), which is fatal in hotspot detection problems. To address the imbalance problem, we further apply hotspot upsampling and random-mirror flipping before training the network. Experimental results show that our proposed neural network model achieves comparable or better performance on the ICCAD 2012 contest benchmark compared to state-of-the-art hotspot detectors based on deep or representative machine leaning.

  2. A DDoS Attack Detection Method Based on Hybrid Heterogeneous Multiclassifier Ensemble Learning

    Directory of Open Access Journals (Sweden)

    Bin Jia

    2017-01-01

    Full Text Available The explosive growth of network traffic and its multitype on Internet have brought new and severe challenges to DDoS attack detection. To get the higher True Negative Rate (TNR, accuracy, and precision and to guarantee the robustness, stability, and universality of detection system, in this paper, we propose a DDoS attack detection method based on hybrid heterogeneous multiclassifier ensemble learning and design a heuristic detection algorithm based on Singular Value Decomposition (SVD to construct our detection system. Experimental results show that our detection method is excellent in TNR, accuracy, and precision. Therefore, our algorithm has good detective performance for DDoS attack. Through the comparisons with Random Forest, k-Nearest Neighbor (k-NN, and Bagging comprising the component classifiers when the three algorithms are used alone by SVD and by un-SVD, it is shown that our model is superior to the state-of-the-art attack detection techniques in system generalization ability, detection stability, and overall detection performance.

  3. System tuning and measurement error detection testing

    International Nuclear Information System (INIS)

    Krejci, Petr; Machek, Jindrich

    2008-09-01

    The project includes the use of the PEANO (Process Evaluation and Analysis by Neural Operators) system to verify the monitoring of the status of dependent measurements with a view to early measurement fault detection and estimation of selected signal levels. At the present stage, the system's capabilities of detecting measurement errors was assessed and the quality of the estimates was evaluated for various system configurations and the formation of empiric models, and rules were sought for system training at chosen process data recording parameters and operating modes. The aim was to find a suitable system configuration and to document the quality of the tuned system on artificial failures

  4. Panorama of recommender systems to support learning

    OpenAIRE

    Drachsler, Hendrik; Verbert, Katrien; Santos, Olga; Manouselis, Nikos

    2015-01-01

    This chapter presents an analysis of recommender systems in Technology-Enhanced Learning along their 15 years existence (2000-2014). All recommender systems considered for the review aim to support educational stakeholders by personalising the learning process. In this meta-review 82 recommender systems from 35 different countries have been investigated and categorised according to a given classification framework. The reviewed systems have been classified into 7 clusters according to their c...

  5. Learning Management Systems and E-Learning within Cyprus Universities

    Directory of Open Access Journals (Sweden)

    Amirkhanpour, Monaliz

    2011-01-01

    Full Text Available This paper presents an extensive research study and results on the use of existing open-source Learning Management Systems, or LMS within the public and private universities of Cyprus. The most significant objective of this research is the identification of the different types of E-Learning, i.e. Computer-Based Training (CBT, Technology-Based Learning (TBL, and Web-Based Training (WBT within Cyprus universities. The paper identifies the benefits and limitations of the main learning approaches used in higher educational institutions, i.e. synchronous and asynchronous learning, investigates the open-source LMS used in the Cypriot universities and compares their features with regards to students’ preferences for a collaborative E-Learning environment. The required data for this research study were collected from undergraduate and graduate students, alumni, faculty members, and IT professionals who currently work and/or study at the public and private universities of Cyprus. The most noteworthy recommendation of this study is the clear indication that most of the undergraduate students that extensively use the specific E-Learning platform of their university do not have a clear picture of the differences between an LMS and a VLE. This gap has to be gradually diminished in order to make optimum use of the different features offered by the specific E-Learning platform.

  6. Web-Based Learning Support System

    Science.gov (United States)

    Fan, Lisa

    Web-based learning support system offers many benefits over traditional learning environments and has become very popular. The Web is a powerful environment for distributing information and delivering knowledge to an increasingly wide and diverse audience. Typical Web-based learning environments, such as Web-CT, Blackboard, include course content delivery tools, quiz modules, grade reporting systems, assignment submission components, etc. They are powerful integrated learning management systems (LMS) that support a number of activities performed by teachers and students during the learning process [1]. However, students who study a course on the Internet tend to be more heterogeneously distributed than those found in a traditional classroom situation. In order to achieve optimal efficiency in a learning process, an individual learner needs his or her own personalized assistance. For a web-based open and dynamic learning environment, personalized support for learners becomes more important. This chapter demonstrates how to realize personalized learning support in dynamic and heterogeneous learning environments by utilizing Adaptive Web technologies. It focuses on course personalization in terms of contents and teaching materials that is according to each student's needs and capabilities. An example of using Rough Set to analyze student personal information to assist students with effective learning and predict student performance is presented.

  7. Intrusion Detection Systems with Live Knowledge System

    Science.gov (United States)

    2016-05-31

    people try to reveal sensitive information of Internet users, also called as phishing. Phishing detection has received great attention but there has...node. Figure 3 describes the result of modified nodes from the original RDR rule tree. Red- coloured ‘X’ sign represents the stopping rule, and the...green- coloured boxes describe the refined rule. However, when human knowledge is applied to those incorrectly classified data, not all of the

  8. Learning Markov models for stationary system behaviors

    DEFF Research Database (Denmark)

    Chen, Yingke; Mao, Hua; Jaeger, Manfred

    2012-01-01

    to a single long observation sequence, and in these situations existing automatic learning methods cannot be applied. In this paper, we adapt algorithms for learning variable order Markov chains from a single observation sequence of a target system, so that stationary system properties can be verified using......Establishing an accurate model for formal verification of an existing hardware or software system is often a manual process that is both time consuming and resource demanding. In order to ease the model construction phase, methods have recently been proposed for automatically learning accurate...... the learned model. Experiments demonstrate that system properties (formulated as stationary probabilities of LTL formulas) can be reliably identified using the learned model....

  9. Learning-Based Detection of Harmful Data in Mobile Devices

    Directory of Open Access Journals (Sweden)

    Seok-Woo Jang

    2016-01-01

    Full Text Available The Internet has supported diverse types of multimedia content flowing freely on smart phones and tablet PCs based on its easy accessibility. However, multimedia content that can be emotionally harmful for children is also easily spread, causing many social problems. This paper proposes a method to assess the harmfulness of input images automatically based on an artificial neural network. The proposed method first detects human face areas based on the MCT features from the input images. Next, based on color characteristics, this study identifies human skin color areas along with the candidate areas of nipples, one of the human body parts representing harmfulness. Finally, the method removes nonnipple areas among the detected candidate areas using the artificial neural network. The experimental results show that the suggested neural network learning-based method can determine the harmfulness of various types of images more effectively by detecting nipple regions from input images robustly.

  10. Embedded Systems - Missile Detection/Interception

    Directory of Open Access Journals (Sweden)

    Luis Cintron

    2010-01-01

    Full Text Available Missile defense systems are often related to major military resources aimed at shielding a specific region from incoming attacks. They are intended to detect, track, intercept, and destruct incoming enemy missiles. These systems vary in cost, efficiency, dependability, and technology. In present times, the possession of these types of systems is associated with large capacity military countries. Demonstrated here are the mathematical techniques behind missile systems which calculate trajectories of incoming missiles and potential intercept positions after initial missile detection. This procedure involved the use of vector-valued functions, systems of equations, and knowledge of projectile motion concepts.

  11. A failure detection and isolation system simulator

    International Nuclear Information System (INIS)

    Assumpcao Filho, E.O.; Nakata, H.

    1990-04-01

    A failure detection and isolation system (FDI) simulation program has been developed for IBM-PC microcomputers. The program, based on the sequential likelihood ratio testing method developed by A. Wald, was implemented with the Monte-Carlo technique. The calculated failure detection rate was favorably compared against the wind-tunnel experimental redundant temperature sensors. (author) [pt

  12. Statistical fault detection in photovoltaic systems

    KAUST Repository

    Garoudja, Elyes; Harrou, Fouzi; Sun, Ying; Kara, Kamel; Chouder, Aissa; Silvestre, Santiago

    2017-01-01

    and efficiency. Here, an innovative model-based fault-detection approach for early detection of shading of PV modules and faults on the direct current (DC) side of PV systems is proposed. This approach combines the flexibility, and simplicity of a one-diode model

  13. Statistical fault detection in photovoltaic systems

    KAUST Repository

    Garoudja, Elyes

    2017-05-08

    Faults in photovoltaic (PV) systems, which can result in energy loss, system shutdown or even serious safety breaches, are often difficult to avoid. Fault detection in such systems is imperative to improve their reliability, productivity, safety and efficiency. Here, an innovative model-based fault-detection approach for early detection of shading of PV modules and faults on the direct current (DC) side of PV systems is proposed. This approach combines the flexibility, and simplicity of a one-diode model with the extended capacity of an exponentially weighted moving average (EWMA) control chart to detect incipient changes in a PV system. The one-diode model, which is easily calibrated due to its limited calibration parameters, is used to predict the healthy PV array\\'s maximum power coordinates of current, voltage and power using measured temperatures and irradiances. Residuals, which capture the difference between the measurements and the predictions of the one-diode model, are generated and used as fault indicators. Then, the EWMA monitoring chart is applied on the uncorrelated residuals obtained from the one-diode model to detect and identify the type of fault. Actual data from the grid-connected PV system installed at the Renewable Energy Development Center, Algeria, are used to assess the performance of the proposed approach. Results show that the proposed approach successfully monitors the DC side of PV systems and detects temporary shading.

  14. [Development of operation patient security detection system].

    Science.gov (United States)

    Geng, Shu-Qin; Tao, Ren-Hai; Zhao, Chao; Wei, Qun

    2008-11-01

    This paper describes a patient security detection system developed with two dimensional bar codes, wireless communication and removal storage technique. Based on the system, nurses and correlative personnel check code wait operation patient to prevent the defaults. The tests show the system is effective. Its objectivity and currency are more scientific and sophisticated than current traditional method in domestic hospital.

  15. Global scene layout modulates contextual learning in change detection.

    Science.gov (United States)

    Conci, Markus; Müller, Hermann J

    2014-01-01

    Change in the visual scene often goes unnoticed - a phenomenon referred to as "change blindness." This study examined whether the hierarchical structure, i.e., the global-local layout of a scene can influence performance in a one-shot change detection paradigm. To this end, natural scenes of a laid breakfast table were presented, and observers were asked to locate the onset of a new local object. Importantly, the global structure of the scene was manipulated by varying the relations among objects in the scene layouts. The very same items were either presented as global-congruent (typical) layouts or as global-incongruent (random) arrangements. Change blindness was less severe for congruent than for incongruent displays, and this congruency benefit increased with the duration of the experiment. These findings show that global layouts are learned, supporting detection of local changes with enhanced efficiency. However, performance was not affected by scene congruency in a subsequent control experiment that required observers to localize a static discontinuity (i.e., an object that was missing from the repeated layouts). Our results thus show that learning of the global layout is particularly linked to the local objects. Taken together, our results reveal an effect of "global precedence" in natural scenes. We suggest that relational properties within the hierarchy of a natural scene are governed, in particular, by global image analysis, reducing change blindness for local objects through scene learning.

  16. Alumina Concentration Detection Based on the Kernel Extreme Learning Machine.

    Science.gov (United States)

    Zhang, Sen; Zhang, Tao; Yin, Yixin; Xiao, Wendong

    2017-09-01

    The concentration of alumina in the electrolyte is of great significance during the production of aluminum. The amount of the alumina concentration may lead to unbalanced material distribution and low production efficiency and affect the stability of the aluminum reduction cell and current efficiency. The existing methods cannot meet the needs for online measurement because industrial aluminum electrolysis has the characteristics of high temperature, strong magnetic field, coupled parameters, and high nonlinearity. Currently, there are no sensors or equipment that can detect the alumina concentration on line. Most companies acquire the alumina concentration from the electrolyte samples which are analyzed through an X-ray fluorescence spectrometer. To solve the problem, the paper proposes a soft sensing model based on a kernel extreme learning machine algorithm that takes the kernel function into the extreme learning machine. K-fold cross validation is used to estimate the generalization error. The proposed soft sensing algorithm can detect alumina concentration by the electrical signals such as voltages and currents of the anode rods. The predicted results show that the proposed approach can give more accurate estimations of alumina concentration with faster learning speed compared with the other methods such as the basic ELM, BP, and SVM.

  17. Global scene layout modulates contextual learning in change detection

    Directory of Open Access Journals (Sweden)

    Markus eConci

    2014-02-01

    Full Text Available Change in the visual scene often goes unnoticed – a phenomenon referred to as ‘change blindness’. This study examined whether the hierarchical structure, i.e., the global-local layout of a scene can influence performance in a one-shot change detection paradigm. To this end, natural scenes of a laid breakfast table were presented, and observers were asked to locate the onset of a new local object. Importantly, the global structure of the scene was manipulated by varying the relations among objects in the scene layouts. The very same items were either presented as global-congruent (typical layouts or as global-incongruent (random arrangements. Change blindness was less severe for congruent than for incongruent displays, and this congruency benefit increased with the duration of the experiment. These findings show that global layouts are learned, supporting detection of local changes with enhanced efficiency. However, performance was not affected by scene congruency in a subsequent control experiment that required observers to localize a static discontinuity (i.e., an object that was missing from the repeated layouts. Our results thus show that learning of the global layout is particularly linked to the local objects. Taken together, our results reveal an effect of global precedence in natural scenes. We suggest that relational properties within the hierarchy of a natural scene are governed, in particular, by global image analysis, reducing change blindness for local objects through scene learning.

  18. Skin Lesion Analysis towards Melanoma Detection Using Deep Learning Network

    Directory of Open Access Journals (Sweden)

    Yuexiang Li

    2018-02-01

    Full Text Available Skin lesions are a severe disease globally. Early detection of melanoma in dermoscopy images significantly increases the survival rate. However, the accurate recognition of melanoma is extremely challenging due to the following reasons: low contrast between lesions and skin, visual similarity between melanoma and non-melanoma lesions, etc. Hence, reliable automatic detection of skin tumors is very useful to increase the accuracy and efficiency of pathologists. In this paper, we proposed two deep learning methods to address three main tasks emerging in the area of skin lesion image processing, i.e., lesion segmentation (task 1, lesion dermoscopic feature extraction (task 2 and lesion classification (task 3. A deep learning framework consisting of two fully convolutional residual networks (FCRN is proposed to simultaneously produce the segmentation result and the coarse classification result. A lesion index calculation unit (LICU is developed to refine the coarse classification results by calculating the distance heat-map. A straight-forward CNN is proposed for the dermoscopic feature extraction task. The proposed deep learning frameworks were evaluated on the ISIC 2017 dataset. Experimental results show the promising accuracies of our frameworks, i.e., 0.753 for task 1, 0.848 for task 2 and 0.912 for task 3 were achieved.

  19. Skin Lesion Analysis towards Melanoma Detection Using Deep Learning Network.

    Science.gov (United States)

    Li, Yuexiang; Shen, Linlin

    2018-02-11

    Skin lesions are a severe disease globally. Early detection of melanoma in dermoscopy images significantly increases the survival rate. However, the accurate recognition of melanoma is extremely challenging due to the following reasons: low contrast between lesions and skin, visual similarity between melanoma and non-melanoma lesions, etc. Hence, reliable automatic detection of skin tumors is very useful to increase the accuracy and efficiency of pathologists. In this paper, we proposed two deep learning methods to address three main tasks emerging in the area of skin lesion image processing, i.e., lesion segmentation (task 1), lesion dermoscopic feature extraction (task 2) and lesion classification (task 3). A deep learning framework consisting of two fully convolutional residual networks (FCRN) is proposed to simultaneously produce the segmentation result and the coarse classification result. A lesion index calculation unit (LICU) is developed to refine the coarse classification results by calculating the distance heat-map. A straight-forward CNN is proposed for the dermoscopic feature extraction task. The proposed deep learning frameworks were evaluated on the ISIC 2017 dataset. Experimental results show the promising accuracies of our frameworks, i.e., 0.753 for task 1, 0.848 for task 2 and 0.912 for task 3 were achieved.

  20. Skin Lesion Analysis towards Melanoma Detection Using Deep Learning Network

    Science.gov (United States)

    2018-01-01

    Skin lesions are a severe disease globally. Early detection of melanoma in dermoscopy images significantly increases the survival rate. However, the accurate recognition of melanoma is extremely challenging due to the following reasons: low contrast between lesions and skin, visual similarity between melanoma and non-melanoma lesions, etc. Hence, reliable automatic detection of skin tumors is very useful to increase the accuracy and efficiency of pathologists. In this paper, we proposed two deep learning methods to address three main tasks emerging in the area of skin lesion image processing, i.e., lesion segmentation (task 1), lesion dermoscopic feature extraction (task 2) and lesion classification (task 3). A deep learning framework consisting of two fully convolutional residual networks (FCRN) is proposed to simultaneously produce the segmentation result and the coarse classification result. A lesion index calculation unit (LICU) is developed to refine the coarse classification results by calculating the distance heat-map. A straight-forward CNN is proposed for the dermoscopic feature extraction task. The proposed deep learning frameworks were evaluated on the ISIC 2017 dataset. Experimental results show the promising accuracies of our frameworks, i.e., 0.753 for task 1, 0.848 for task 2 and 0.912 for task 3 were achieved. PMID:29439500

  1. Caltrans fog detection and warning system.

    Science.gov (United States)

    2009-01-01

    The California Department of Transportation (Caltrans) has implemented a fog detection and warning system on Highway 99 near Fresno. The entire central valley region is susceptible to Tule fog, which can reduce visibility tremendously, sometimes to n...

  2. Detection of cardiovascular anomalies: Hybrid systems approach

    KAUST Repository

    Ledezma, Fernando; Laleg-Kirati, Taous-Meriem

    2012-01-01

    In this paper, we propose a hybrid interpretation of the cardiovascular system. Based on a model proposed by Simaan et al. (2009), we study the problem of detecting cardiovascular anomalies that can be caused by variations in some physiological

  3. Adaptive polymeric system for Hebbian type learning

    OpenAIRE

    2011-01-01

    Abstract We present the experimental realization of an adaptive polymeric system displaying a ?learning behaviour?. The system consists on a statistically organized networks of memristive elements (memory-resitors) based on polyaniline. In a such network the path followed by the current increments its conductivity, a property which makes the system able to mimic Hebbian type learning and have application in hardware neural networks. After discussing the working principle of ...

  4. Detection and intelligent systems for homeland security

    CERN Document Server

    Voeller, John G

    2014-01-01

    Detection and Intelligent Systems for Homeland Security features articles from the Wiley Handbook of Science and Technology for Homeland Security covering advanced technology for image and video interpretation systems used for surveillance, which help in solving such problems as identifying faces from live streaming or stored videos. Biometrics for human identification, including eye retinas and irises, and facial patterns are also presented. The book then provides information on sensors for detection of explosive and radioactive materials and methods for sensing chemical

  5. Doubler system quench detection threshold

    International Nuclear Information System (INIS)

    Kuepke, K.; Kuchnir, M.; Martin, P.

    1983-01-01

    The experimental study leading to the determination of the sensitivity needed for protecting the Fermilab Doubler from damage during quenches is presented. The quench voltage thresholds involved were obtained from measurements made on Doubler cable of resistance x temperature and voltage x time during quenches under several currents and from data collected during operation of the Doubler Quench Protection System as implemented in the B-12 string of 20 magnets. At 4kA, a quench voltage threshold in excess of 5.OV will limit the peak Doubler cable temperature to 452K for quenches originating in the magnet coils whereas a threshold of 0.5V is required for quenches originating outside of coils

  6. Intelligent Web-Based Learning System with Personalized Learning Path Guidance

    Science.gov (United States)

    Chen, C. M.

    2008-01-01

    Personalized curriculum sequencing is an important research issue for web-based learning systems because no fixed learning paths will be appropriate for all learners. Therefore, many researchers focused on developing e-learning systems with personalized learning mechanisms to assist on-line web-based learning and adaptively provide learning paths…

  7. Patterns for Designing Learning Management Systems

    NARCIS (Netherlands)

    Avgeriou, Paris; Retalis, Symeon; Papasalouros, Andreas

    2003-01-01

    Learning Management Systems are sophisticated web-based applications that are being engineered today in increasing numbers by numerous institutions and companies that want to get involved in e-learning either for providing services to third parties, or for educating and training their own people.

  8. Process Systems Engineering Education: Learning by Research

    Science.gov (United States)

    Abbas, A.; Alhammadi, H. Y.; Romagnoli, J. A.

    2009-01-01

    In this paper, we discuss our approach in teaching the final-year course Process Systems Engineering. Students are given ownership of the course by transferring to them the responsibility of learning. A project-based group environment stimulates learning while solving a real engineering problem. We discuss postgraduate student involvement and how…

  9. Learning and Organizational Effectiveness: A Systems Perspective

    Science.gov (United States)

    Andreadis, Nicholas

    2009-01-01

    The challenge for leaders today is to create and develop the capability of their organization. Leaders must perceive and manage their organization as a dynamic, open system where learning is the core competence underlying innovation, growth, and sustainability. Creating a culture of learning is the first work of leadership. This article presents a…

  10. A Distributed Intelligent E-Learning System

    Science.gov (United States)

    Kristensen, Terje

    2016-01-01

    An E-learning system based on a multi-agent (MAS) architecture combined with the Dynamic Content Manager (DCM) model of E-learning, is presented. We discuss the benefits of using such a multi-agent architecture. Finally, the MAS architecture is compared with a pure service-oriented architecture (SOA). This MAS architecture may also be used within…

  11. Intrusion Detection amp Prevention Systems - Sourcefire Snort

    Directory of Open Access Journals (Sweden)

    Rajesh Vuppala

    2015-08-01

    Full Text Available Information security is a challenging issue for all business organizations today amidst increasing cyber threats. While there are many alternative intrusion detection amp prevention systems available to choose from selecting the best solution to implement to detect amp prevent cyber-attacks is a difficult task. The best solution is of the one that gets the best reviews and suits the organizations needs amp budget. In this review paper we summarize various classes of intrusion detection and prevention systems compare features of alternative solutions and make recommendation for implementation of one as the best solution for business organization in Fiji.

  12. Active fault detection in MIMO systems

    DEFF Research Database (Denmark)

    Niemann, Hans Henrik; Poulsen, Niels Kjølstad

    2014-01-01

    The focus in this paper is on active fault detection (AFD) for MIMO systems with parametric faults. The problem of design of auxiliary inputs with respect to detection of parametric faults is investigated. An analysis of the design of auxiliary inputs is given based on analytic transfer functions...... from auxiliary input to residual outputs. The analysis is based on a singular value decomposition of these transfer functions Based on this analysis, it is possible to design auxiliary input as well as design of the associated residual vector with respect to every single parametric fault in the system...... such that it is possible to detect these faults....

  13. Portable reconfigurable detection and assessment system

    International Nuclear Information System (INIS)

    Blattman, D.A.

    1991-01-01

    Rapidly changing geopolitical issues throughout the world have made the ability to effectively respond to political, military, terrorist and peace-keeping requirements increasingly important. Recent Middle East events indicate a continuing escalation in these activities. These activities are defining the requirements for a rapidly deployable, portable, real-time detection and assessment operational security system that is reconfigurable to site specific threats. This paper describes such a system Mobile Operational Detection and Assessment system (MODAS); a commercially-off-the shelf (COTS) integrated and reconfigurable hardware/software system solution for the ever-changing geopolitical security issues of the Nineties

  14. Building machine learning systems with Python

    CERN Document Server

    Richert, Willi

    2013-01-01

    This is a tutorial-driven and practical, but well-grounded book showcasing good Machine Learning practices. There will be an emphasis on using existing technologies instead of showing how to write your own implementations of algorithms. This book is a scenario-based, example-driven tutorial. By the end of the book you will have learnt critical aspects of Machine Learning Python projects and experienced the power of ML-based systems by actually working on them.This book primarily targets Python developers who want to learn about and build Machine Learning into their projects, or who want to pro

  15. Flat Surface Damage Detection System (FSDDS)

    Science.gov (United States)

    Williams, Martha; Lewis, Mark; Gibson, Tracy; Lane, John; Medelius, Pedro; Snyder, Sarah; Ciarlariello, Dan; Parks, Steve; Carrejo, Danny; Rojdev, Kristina

    2013-01-01

    The Flat Surface Damage Detection system (FSDDS} is a sensory system that is capable of detecting impact damages to surfaces utilizing a novel sensor system. This system will provide the ability to monitor the integrity of an inflatable habitat during in situ system health monitoring. The system consists of three main custom designed subsystems: the multi-layer sensing panel, the embedded monitoring system, and the graphical user interface (GUI). The GUI LABVIEW software uses a custom developed damage detection algorithm to determine the damage location based on the sequence of broken sensing lines. It estimates the damage size, the maximum depth, and plots the damage location on a graph. Successfully demonstrated as a stand alone technology during 2011 D-RATS. Software modification also allowed for communication with HDU avionics crew display which was demonstrated remotely (KSC to JSC} during 2012 integration testing. Integrated FSDDS system and stand alone multi-panel systems were demonstrated remotely and at JSC, Mission Operations Test using Space Network Research Federation (SNRF} network in 2012. FY13, FSDDS multi-panel integration with JSC and SNRF network Technology can allow for integration with other complementary damage detection systems.

  16. The immune system, adaptation, and machine learning

    Science.gov (United States)

    Farmer, J. Doyne; Packard, Norman H.; Perelson, Alan S.

    1986-10-01

    The immune system is capable of learning, memory, and pattern recognition. By employing genetic operators on a time scale fast enough to observe experimentally, the immune system is able to recognize novel shapes without preprogramming. Here we describe a dynamical model for the immune system that is based on the network hypothesis of Jerne, and is simple enough to simulate on a computer. This model has a strong similarity to an approach to learning and artificial intelligence introduced by Holland, called the classifier system. We demonstrate that simple versions of the classifier system can be cast as a nonlinear dynamical system, and explore the analogy between the immune and classifier systems in detail. Through this comparison we hope to gain insight into the way they perform specific tasks, and to suggest new approaches that might be of value in learning systems.

  17. A new digital correlation flaw detection system

    International Nuclear Information System (INIS)

    Lee, B.B.; Furgason, E.S.

    1981-01-01

    A new portable digital random signal flaw detection system is described which uses a digital delay line to replace the acoustic delay line of the original random signal system. Using this new system, a comparison was made between the two types of transmit signals which have been used in previous systems--m-sequences and random signals. This comparison has not been possible with these previous correlation flaw detection systems. Results indicated that for high-speed short code operation, the m-sequences produced slightly lower range sidelobes than typical samples of a clipped random signal. For normal long code operation, results indicated that system performance is essentially equivalent in resolution and signal-to-noise ratio using either m-sequences or clipped and sampled random signals. Further results also showed that for normal long code operation, the system produces outputs equivalent in resolution to pulse-echo systems, but with the added benefit of signal-to-noise ratio enhancement

  18. An anomaly detection and isolation scheme with instance-based learning and sequential analysis

    International Nuclear Information System (INIS)

    Yoo, T. S.; Garcia, H. E.

    2006-01-01

    This paper presents an online anomaly detection and isolation (FDI) technique using an instance-based learning method combined with a sequential change detection and isolation algorithm. The proposed method uses kernel density estimation techniques to build statistical models of the given empirical data (null hypothesis). The null hypothesis is associated with the set of alternative hypotheses modeling the abnormalities of the systems. A decision procedure involves a sequential change detection and isolation algorithm. Notably, the proposed method enjoys asymptotic optimality as the applied change detection and isolation algorithm is optimal in minimizing the worst mean detection/isolation delay for a given mean time before a false alarm or a false isolation. Applicability of this methodology is illustrated with redundant sensor data set and its performance. (authors)

  19. SYSTEM APPROACH TO THE BLENDED LEARNING

    Directory of Open Access Journals (Sweden)

    Vladimir Kukharenko

    2015-10-01

    Full Text Available Currently, much attention is paid to the development of learning sour cream – a combination of traditional and distance (30-70% of training. Such training is sometimes called hybrid and referred to disruptive technologies. Purpose – to show that the use of systemic campaign in blended learning provides a high quality of education, and the technology can be devastating. The subject of the study – blended learning, object of study – Mixed learning process. The analysis results show that the combined training increases the motivation of students, qualification of teachers, personalized learning process. At the same time there are no reliable methods of assessing the quality of education and training standards. It is important that blended learning strategy to support the institutional goals and had an effective organizational model for support.

  20. Unsupervised behaviour-specific dictionary learning for abnormal event detection

    DEFF Research Database (Denmark)

    Ren, Huamin; Liu, Weifeng; Olsen, Søren Ingvor

    2015-01-01

    the training data is only a small proportion of the surveillance data. Therefore, we propose behavior-specific dictionaries (BSD) through unsupervised learning, pursuing atoms from the same type of behavior to represent one behavior dictionary. To further improve the dictionary by introducing information from...... potential infrequent normal patterns, we refine the dictionary by searching ‘missed atoms’ that have compact coefficients. Experimental results show that our BSD algorithm outperforms state-of-the-art dictionaries in abnormal event detection on the public UCSD dataset. Moreover, BSD has less false alarms...

  1. Unsupervised obstacle detection in driving environments using deep-learning-based stereovision

    KAUST Repository

    Dairi, Abdelkader; Harrou, Fouzi; Senouci, Mohamed; Sun, Ying

    2017-01-01

    A vision-based obstacle detection system is a key enabler for the development of autonomous robots and vehicles and intelligent transportation systems. This paper addresses the problem of urban scene monitoring and tracking of obstacles based on unsupervised, deep-learning approaches. Here, we design an innovative hybrid encoder that integrates deep Boltzmann machines (DBM) and auto-encoders (AE). This hybrid auto-encode (HAE) model combines the greedy learning features of DBM with the dimensionality reduction capacity of AE to accurately and reliably detect the presence of obstacles. We combine the proposed hybrid model with the one-class support vector machines (OCSVM) to visually monitor an urban scene. We also propose an efficient approach to estimating obstacles location and track their positions via scene densities. Specifically, we address obstacle detection as an anomaly detection problem. If an obstacle is detected by the OCSVM algorithm, then localization and tracking algorithm is executed. We validated the effectiveness of our approach by using experimental data from two publicly available dataset, the Malaga stereovision urban dataset (MSVUD) and the Daimler urban segmentation dataset (DUSD). Results show the capacity of the proposed approach to reliably detect obstacles.

  2. Unsupervised obstacle detection in driving environments using deep-learning-based stereovision

    KAUST Repository

    Dairi, Abdelkader

    2017-12-06

    A vision-based obstacle detection system is a key enabler for the development of autonomous robots and vehicles and intelligent transportation systems. This paper addresses the problem of urban scene monitoring and tracking of obstacles based on unsupervised, deep-learning approaches. Here, we design an innovative hybrid encoder that integrates deep Boltzmann machines (DBM) and auto-encoders (AE). This hybrid auto-encode (HAE) model combines the greedy learning features of DBM with the dimensionality reduction capacity of AE to accurately and reliably detect the presence of obstacles. We combine the proposed hybrid model with the one-class support vector machines (OCSVM) to visually monitor an urban scene. We also propose an efficient approach to estimating obstacles location and track their positions via scene densities. Specifically, we address obstacle detection as an anomaly detection problem. If an obstacle is detected by the OCSVM algorithm, then localization and tracking algorithm is executed. We validated the effectiveness of our approach by using experimental data from two publicly available dataset, the Malaga stereovision urban dataset (MSVUD) and the Daimler urban segmentation dataset (DUSD). Results show the capacity of the proposed approach to reliably detect obstacles.

  3. Indirect learning control for nonlinear dynamical systems

    Science.gov (United States)

    Ryu, Yeong Soon; Longman, Richard W.

    1993-01-01

    In a previous paper, learning control algorithms were developed based on adaptive control ideas for linear time variant systems. The learning control methods were shown to have certain advantages over their adaptive control counterparts, such as the ability to produce zero tracking error in time varying systems, and the ability to eliminate repetitive disturbances. In recent years, certain adaptive control algorithms have been developed for multi-body dynamic systems such as robots, with global guaranteed convergence to zero tracking error for the nonlinear system euations. In this paper we study the relationship between such adaptive control methods designed for this specific class of nonlinear systems, and the learning control problem for such systems, seeking to converge to zero tracking error in following a specific command repeatedly, starting from the same initial conditions each time. The extension of these methods from the adaptive control problem to the learning control problem is seen to be trivial. The advantages and disadvantages of using learning control based on such adaptive control concepts for nonlinear systems, and the use of other currently available learning control algorithms are discussed.

  4. systemic approach to teaching and learning chemistry

    African Journals Online (AJOL)

    unesco

    2National Core Group in Chemistry, H.E.J Research Institute of Chemistry,. University of ... innovative way of teaching and learning through systemic approach (SATL) has been .... available to do useful work in a thermodynamic process.

  5. On Textual Analysis and Machine Learning for Cyberstalking Detection.

    Science.gov (United States)

    Frommholz, Ingo; Al-Khateeb, Haider M; Potthast, Martin; Ghasem, Zinnar; Shukla, Mitul; Short, Emma

    2016-01-01

    Cyber security has become a major concern for users and businesses alike. Cyberstalking and harassment have been identified as a growing anti-social problem. Besides detecting cyberstalking and harassment, there is the need to gather digital evidence, often by the victim. To this end, we provide an overview of and discuss relevant technological means, in particular coming from text analytics as well as machine learning, that are capable to address the above challenges. We present a framework for the detection of text-based cyberstalking and the role and challenges of some core techniques such as author identification, text classification and personalisation. We then discuss PAN, a network and evaluation initiative that focusses on digital text forensics, in particular author identification.

  6. DMIA: A MALWARE DETECTION SYSTEM ON IOS PLATFORM

    OpenAIRE

    Hongliang Liang; Yilun Xie; Yan Song

    2016-01-01

    iOS is a popular operating system on Apple’s smartphones, and recent security events have shown the possibility of stealing the users' privacy in iOS without being detected, such as XcodeGhost. So, we present the design and implementation of a malware vetting system, called DMIA. DMIA first collects runtime information of an app and then distinguish between malicious and normal apps by a novel machine learning model. We evaluated DMIA with 1000 apps from the official App Store. The results of...

  7. Learning Visual Representations for Perception-Action Systems

    DEFF Research Database (Denmark)

    Piater, Justus; Jodogne, Sebastien; Detry, Renaud

    2011-01-01

    and RLJC, our second method learns structural object models for robust object detection and pose estimation by probabilistic inference. To these models, the method associates grasp experiences autonomously learned by trial and error. These experiences form a nonparametric representation of grasp success......We discuss vision as a sensory modality for systems that effect actions in response to perceptions. While the internal representations informed by vision may be arbitrarily complex, we argue that in many cases it is advantageous to link them rather directly to action via learned mappings....... These arguments are illustrated by two examples of our own work. First, our RLVC algorithm performs reinforcement learning directly on the visual input space. To make this very large space manageable, RLVC interleaves the reinforcement learner with a supervised classification algorithm that seeks to split...

  8. Transformative Learning: Patterns of Psychophysiologic Response and Technology-Enabled Learning and Intervention Systems

    Science.gov (United States)

    2008-09-01

    Psychophysiologic Response and Technology -Enabled Learning and Intervention Systems PRINCIPAL INVESTIGATOR: Leigh W. Jerome, Ph.D...NUMBER Transformative Learning : Patterns of Psychophysiologic Response and Technology - Enabled Learning and Intervention Systems 5b. GRANT NUMBER...project entitled “Transformative Learning : Patterns of Psychophysiologic Response in Technology Enabled Learning and Intervention Systems.” The

  9. Learning approach to the detection of gravitational wave transients

    International Nuclear Information System (INIS)

    Chassande-Mottin, E.

    2003-01-01

    We investigate the class of quadratic detectors (i.e., the statistic is a bilinear function of the data) for the detection of poorly modeled gravitational transients of short duration. We point out that all such detection methods are equivalent to passing the signal through a filter bank and linearly combining the output energy. Existing methods for the choice of the filter bank and of the weight parameters (to be multiplied by the output energy of each filter before summation) rely essentially on the two following ideas: (i) the use of the likelihood function based on a (possibly noninformative) statistical model of the signal and the noise; (ii) the use of Monte Carlo simulations for the tuning of parametric filters to get the best detection probability while keeping the false alarm rate fixed. We propose a third approach according to which the filter bank is 'learned' from a set of training data. By-products of this viewpoint are that, contrarily to previous methods, (i) there is no requirement of an explicit description of the probability density function of the data when the signal is present and (ii) the filters we use are nonparametric. The learning procedure may be described as a two step process: first, estimate the mean and covariance of the signal with the training data; second, find the filters which maximize a contrast criterion referred to as the deflection between the 'noise only' and 'signal + noise' hypotheses. The deflection is homogeneous to the signal-to-noise ratio and it uses the quantities estimated at the first step. We apply this original method to the problem of the detection of supernovae core collapses. We use the catalog of waveforms provided recently by Dimmelmeier et al. to train our algorithm. We expect such a detector to have better performances in this particular problem provided that the reference signals are reliable

  10. Capacitive system detects and locates fluid leaks

    Science.gov (United States)

    1966-01-01

    Electronic monitoring system automatically detects and locates minute leaks in seams of large fluid storage tanks and pipelines covered with thermal insulation. The system uses a capacitive tape-sensing element that is adhesively bonded over seams where fluid leaks are likely to occur.

  11. Improved biosensor-based detection system

    DEFF Research Database (Denmark)

    2015-01-01

    Described is a new biosensor-based detection system for effector compounds, useful for in vivo applications in e.g. screening and selecting of cells which produce a small molecule effector compound or which take up a small molecule effector compound from its environment. The detection system...... comprises a protein or RNA-based biosensor for the effector compound which indirectly regulates the expression of a reporter gene via two hybrid proteins, providing for fewer false signals or less 'noise', tuning of sensitivity or other advantages over conventional systems where the biosensor directly...

  12. Hydrogen detection systems leak response codes

    International Nuclear Information System (INIS)

    Desmas, T.; Kong, N.; Maupre, J.P.; Schindler, P.; Blanc, D.

    1990-01-01

    A loss in tightness of a water tube inside a Steam Generator Unit of a Fast Reactor is usually monitored by hydrogen detection systems. Such systems have demonstrated in the past their ability to detect a leak in a SGU. However, the increase in size of the SGU or the choice of ferritic material entails improvement of these systems in order to avoid secondary leak or to limit damages to the tube bundle. The R and D undertaken in France on this subject is presented. (author). 11 refs, 10 figs

  13. New designing of E-Learning systems with using network learning

    OpenAIRE

    Malayeri, Amin Daneshmand; Abdollahi, Jalal

    2010-01-01

    One of the most applied learning in virtual spaces is using E-Learning systems. Some E-Learning methodologies has been introduced, but the main subject is the most positive feedback from E-Learning systems. In this paper, we introduce a new methodology of E-Learning systems entitle "Network Learning" with review of another aspects of E-Learning systems. Also, we present benefits and advantages of using these systems in educating and fast learning programs. Network Learning can be programmable...

  14. Multisignal detecting system of pile integrity testing

    Science.gov (United States)

    Liu, Zuting; Luo, Ying; Yu, Shihai

    2002-05-01

    The low strain reflection wave method plays a principal rule in the integrating detection of base piles. However, there are some deficiencies with this method. For example, there is a blind area of detection on top of the tested pile; it is difficult to recognize the defects at deep-seated parts of the pile; there is still the planar of 3D domino effect, etc. It is very difficult to solve these problems only with the single-transducer pile integrity testing system. A new multi-signal piles integrity testing system is proposed in this paper, which is able to impulse and collect signals on multiple points on top of the pile. By using the multiple superposition data processing method, the detecting system can effectively restrain the interference and elevate the precision and SNR of pile integrity testing. The system can also be applied to the evaluation of engineering structure health.

  15. Feedback Design Patterns for Math Online Learning Systems

    Science.gov (United States)

    Inventado, Paul Salvador; Scupelli, Peter; Heffernan, Cristina; Heffernan, Neil

    2017-01-01

    Increasingly, computer-based learning systems are used by educators to facilitate learning. Evaluations of several math learning systems show that they result in significant student learning improvements. Feedback provision is one of the key features in math learning systems that contribute to its success. We have recently been uncovering feedback…

  16. Detection of Hail Storms in Radar Imagery Using Deep Learning

    Science.gov (United States)

    Pullman, Melinda; Gurung, Iksha; Ramachandran, Rahul; Maskey, Manil

    2017-01-01

    In 2016, hail was responsible for 3.5 billion and 23 million dollars in damage to property and crops, respectively, making it the second costliest weather phenomenon in the United States. In an effort to improve hail-prediction techniques and reduce the societal impacts associated with hail storms, we propose a deep learning technique that leverages radar imagery for automatic detection of hail storms. The technique is applied to radar imagery from 2011 to 2016 for the contiguous United States and achieved a precision of 0.848. Hail storms are primarily detected through the visual interpretation of radar imagery (Mrozet al., 2017). With radars providing data every two minutes, the detection of hail storms has become a big data task. As a result, scientists have turned to neural networks that employ computer vision to identify hail-bearing storms (Marzbanet al., 2001). In this study, we propose a deep Convolutional Neural Network (ConvNet) to understand the spatial features and patterns of radar echoes for detecting hailstorms.

  17. Image analysis and machine learning for detecting malaria.

    Science.gov (United States)

    Poostchi, Mahdieh; Silamut, Kamolrat; Maude, Richard J; Jaeger, Stefan; Thoma, George

    2018-04-01

    Malaria remains a major burden on global health, with roughly 200 million cases worldwide and more than 400,000 deaths per year. Besides biomedical research and political efforts, modern information technology is playing a key role in many attempts at fighting the disease. One of the barriers toward a successful mortality reduction has been inadequate malaria diagnosis in particular. To improve diagnosis, image analysis software and machine learning methods have been used to quantify parasitemia in microscopic blood slides. This article gives an overview of these techniques and discusses the current developments in image analysis and machine learning for microscopic malaria diagnosis. We organize the different approaches published in the literature according to the techniques used for imaging, image preprocessing, parasite detection and cell segmentation, feature computation, and automatic cell classification. Readers will find the different techniques listed in tables, with the relevant articles cited next to them, for both thin and thick blood smear images. We also discussed the latest developments in sections devoted to deep learning and smartphone technology for future malaria diagnosis. Published by Elsevier Inc.

  18. Network Intrusion Detection System using Apache Storm

    Directory of Open Access Journals (Sweden)

    Muhammad Asif Manzoor

    2017-06-01

    Full Text Available Network security implements various strategies for the identification and prevention of security breaches. Network intrusion detection is a critical component of network management for security, quality of service and other purposes. These systems allow early detection of network intrusion and malicious activities; so that the Network Security infrastructure can react to mitigate these threats. Various systems are proposed to enhance the network security. We are proposing to use anomaly based network intrusion detection system in this work. Anomaly based intrusion detection system can identify the new network threats. We also propose to use Real-time Big Data Stream Processing Framework, Apache Storm, for the implementation of network intrusion detection system. Apache Storm can help to manage the network traffic which is generated at enormous speed and size and the network traffic speed and size is constantly increasing. We have used Support Vector Machine in this work. We use Knowledge Discovery and Data Mining 1999 (KDD’99 dataset to test and evaluate our proposed solution.

  19. Systems Thinking, Lean Production and Action Learning

    Science.gov (United States)

    Seddon, John; Caulkin, Simon

    2007-01-01

    Systems thinking underpins "lean" management and is best understood through action-learning as the ideas are counter-intuitive. The Toyota Production System is just that--a system; the failure to appreciate that starting-place and the advocacy of "tools" leads many to fail to grasp what is, without doubt, a significant…

  20. The organization of an autonomous learning system

    Science.gov (United States)

    Kanerva, Pentti

    1988-01-01

    The organization of systems that learn from experience is examined, human beings and animals being prime examples of such systems. How is their information processing organized. They build an internal model of the world and base their actions on the model. The model is dynamic and predictive, and it includes the systems' own actions and their effects. In modeling such systems, a large pattern of features represents a moment of the system's experience. Some of the features are provided by the system's senses, some control the system's motors, and the rest have no immediate external significance. A sequence of such patterns then represents the system's experience over time. By storing such sequences appropriately in memory, the system builds a world model based on experience. In addition to the essential function of memory, fundamental roles are played by a sensory system that makes raw information about the world suitable for memory storage and by a motor system that affects the world. The relation of sensory and motor systems to the memory is discussed, together with how favorable actions can be learned and unfavorable actions can be avoided. Results in classical learning theory are explained in terms of the model, more advanced forms of learning are discussed, and the relevance of the model to the frame problem of robotics is examined.

  1. Obstacle detection system for underground mining vehicles

    Energy Technology Data Exchange (ETDEWEB)

    Cohen, P.; Polotski, V.; Piotte, M.; Melamed, F. [Ecole Polytechnique de Montreal, Montreal, PQ (Canada)

    1998-01-01

    A device for detecting obstacles by autonomous vehicles navigating in mine drifts is described. The device is based upon structured lighting and the extraction of relevant features from images of obstacles. The system uses image profile changes, ground and wall irregularities, disturbances of the vehicle`s trajectory, and impaired visibility to detect obstacles, rather than explicit three-dimensional scene reconstruction. 7 refs., 5 figs.

  2. An ensemble deep learning based approach for red lesion detection in fundus images.

    Science.gov (United States)

    Orlando, José Ignacio; Prokofyeva, Elena; Del Fresno, Mariana; Blaschko, Matthew B

    2018-01-01

    Diabetic retinopathy (DR) is one of the leading causes of preventable blindness in the world. Its earliest sign are red lesions, a general term that groups both microaneurysms (MAs) and hemorrhages (HEs). In daily clinical practice, these lesions are manually detected by physicians using fundus photographs. However, this task is tedious and time consuming, and requires an intensive effort due to the small size of the lesions and their lack of contrast. Computer-assisted diagnosis of DR based on red lesion detection is being actively explored due to its improvement effects both in clinicians consistency and accuracy. Moreover, it provides comprehensive feedback that is easy to assess by the physicians. Several methods for detecting red lesions have been proposed in the literature, most of them based on characterizing lesion candidates using hand crafted features, and classifying them into true or false positive detections. Deep learning based approaches, by contrast, are scarce in this domain due to the high expense of annotating the lesions manually. In this paper we propose a novel method for red lesion detection based on combining both deep learned and domain knowledge. Features learned by a convolutional neural network (CNN) are augmented by incorporating hand crafted features. Such ensemble vector of descriptors is used afterwards to identify true lesion candidates using a Random Forest classifier. We empirically observed that combining both sources of information significantly improve results with respect to using each approach separately. Furthermore, our method reported the highest performance on a per-lesion basis on DIARETDB1 and e-ophtha, and for screening and need for referral on MESSIDOR compared to a second human expert. Results highlight the fact that integrating manually engineered approaches with deep learned features is relevant to improve results when the networks are trained from lesion-level annotated data. An open source implementation of our

  3. Device for detecting failure of reactor system

    International Nuclear Information System (INIS)

    Miyazawa, Tatsuo.

    1979-01-01

    Purpose: To make it possible to rapidly detect any failure in a reactor system prior to the leakage of coolants. Constitution: The dose of beta line is computed from the difference between the power of a detector for reacting with both beta and gamma lines and a detector for reacting only with gamma line to detect the failure of a reactor system, thereby to raise the detection speed and improve the detection accuracy. More specifically, a radiation detector A detects gamma and beta lines by means of piezoelectric elements. A radiation detector B caused the opening of the detector A to be covered with a metal, and detects only gamma line. The detected values of detectors A and B are amplified by an amplifier and applied to a rate meter and a counter, the values being converted into DC and introduced into a comparison circuit, where the outputs of the rate meter are compared with each other. When the difference is more than the predetermined range, it is supplied as output to an alarm circuit where an alarm signal is produced. (Nakamura, S.)

  4. Edge detection techniques for iris recognition system

    International Nuclear Information System (INIS)

    Tania, U T; Motakabber, S M A; Ibrahimy, M I

    2013-01-01

    Nowadays security and authentication are the major parts of our daily life. Iris is one of the most reliable organ or part of human body which can be used for identification and authentication purpose. To develop an iris authentication algorithm for personal identification, this paper examines two edge detection techniques for iris recognition system. Between the Sobel and the Canny edge detection techniques, the experimental result shows that the Canny's technique has better ability to detect points in a digital image where image gray level changes even at slow rate

  5. Evaluating E-Learning Systems: An Empirical Investigation on Students' Perception in Higher Education Area

    Directory of Open Access Journals (Sweden)

    Muneer Abbad

    2014-06-01

    Full Text Available In search of better, traditional learning universities have expanded their ways to deliver knowledge and integrate cost effective e-learning systems. Universities' use of information and communication technologies has grown tremendously over the last decade. To ensure efficient use of the e-learning system, the Arab Open University (AOU in Bahrain was the first to use e-learning system there, aimed to evaluate the good and bad practices, detect errors and determine areas for further improvements in usage. This study critically evaluated the students' perception of the elearning system in Bahrain and recommended changes to improve students' e-learning usage. Results of the study indicated that, in general, students have favourable perceptions toward using the e-learning system. This study has shown that technology acceptance is the most variable, factor that contributes to students' perception and satisfaction of the e-learning system.

  6. A Mobile Gamification Learning System for Improving the Learning Motivation and Achievements

    Science.gov (United States)

    Su, C-H.; Cheng, C-H.

    2015-01-01

    This paper aims to investigate how a gamified learning approach influences science learning, achievement and motivation, through a context-aware mobile learning environment, and explains the effects on motivation and student learning. A series of gamified learning activities, based on MGLS (Mobile Gamification Learning System), was developed and…

  7. Maximum Temperature Detection System for Integrated Circuits

    Science.gov (United States)

    Frankiewicz, Maciej; Kos, Andrzej

    2015-03-01

    The paper describes structure and measurement results of the system detecting present maximum temperature on the surface of an integrated circuit. The system consists of the set of proportional to absolute temperature sensors, temperature processing path and a digital part designed in VHDL. Analogue parts of the circuit where designed with full-custom technique. The system is a part of temperature-controlled oscillator circuit - a power management system based on dynamic frequency scaling method. The oscillator cooperates with microprocessor dedicated for thermal experiments. The whole system is implemented in UMC CMOS 0.18 μm (1.8 V) technology.

  8. Credit Card Fraud Detection: A Realistic Modeling and a Novel Learning Strategy.

    Science.gov (United States)

    Dal Pozzolo, Andrea; Boracchi, Giacomo; Caelen, Olivier; Alippi, Cesare; Bontempi, Gianluca

    2017-09-14

    Detecting frauds in credit card transactions is perhaps one of the best testbeds for computational intelligence algorithms. In fact, this problem involves a number of relevant challenges, namely: concept drift (customers' habits evolve and fraudsters change their strategies over time), class imbalance (genuine transactions far outnumber frauds), and verification latency (only a small set of transactions are timely checked by investigators). However, the vast majority of learning algorithms that have been proposed for fraud detection rely on assumptions that hardly hold in a real-world fraud-detection system (FDS). This lack of realism concerns two main aspects: 1) the way and timing with which supervised information is provided and 2) the measures used to assess fraud-detection performance. This paper has three major contributions. First, we propose, with the help of our industrial partner, a formalization of the fraud-detection problem that realistically describes the operating conditions of FDSs that everyday analyze massive streams of credit card transactions. We also illustrate the most appropriate performance measures to be used for fraud-detection purposes. Second, we design and assess a novel learning strategy that effectively addresses class imbalance, concept drift, and verification latency. Third, in our experiments, we demonstrate the impact of class unbalance and concept drift in a real-world data stream containing more than 75 million transactions, authorized over a time window of three years.

  9. Fast learning method for convolutional neural networks using extreme learning machine and its application to lane detection.

    Science.gov (United States)

    Kim, Jihun; Kim, Jonghong; Jang, Gil-Jin; Lee, Minho

    2017-03-01

    Deep learning has received significant attention recently as a promising solution to many problems in the area of artificial intelligence. Among several deep learning architectures, convolutional neural networks (CNNs) demonstrate superior performance when compared to other machine learning methods in the applications of object detection and recognition. We use a CNN for image enhancement and the detection of driving lanes on motorways. In general, the process of lane detection consists of edge extraction and line detection. A CNN can be used to enhance the input images before lane detection by excluding noise and obstacles that are irrelevant to the edge detection result. However, training conventional CNNs requires considerable computation and a big dataset. Therefore, we suggest a new learning algorithm for CNNs using an extreme learning machine (ELM). The ELM is a fast learning method used to calculate network weights between output and hidden layers in a single iteration and thus, can dramatically reduce learning time while producing accurate results with minimal training data. A conventional ELM can be applied to networks with a single hidden layer; as such, we propose a stacked ELM architecture in the CNN framework. Further, we modify the backpropagation algorithm to find the targets of hidden layers and effectively learn network weights while maintaining performance. Experimental results confirm that the proposed method is effective in reducing learning time and improving performance. Copyright © 2016 Elsevier Ltd. All rights reserved.

  10. Night-Time Vehicle Detection Algorithm Based on Visual Saliency and Deep Learning

    Directory of Open Access Journals (Sweden)

    Yingfeng Cai

    2016-01-01

    Full Text Available Night vision systems get more and more attention in the field of automotive active safety field. In this area, a number of researchers have proposed far-infrared sensor based night-time vehicle detection algorithm. However, existing algorithms have low performance in some indicators such as the detection rate and processing time. To solve this problem, we propose a far-infrared image vehicle detection algorithm based on visual saliency and deep learning. Firstly, most of the nonvehicle pixels will be removed with visual saliency computation. Then, vehicle candidate will be generated by using prior information such as camera parameters and vehicle size. Finally, classifier trained with deep belief networks will be applied to verify the candidates generated in last step. The proposed algorithm is tested in around 6000 images and achieves detection rate of 92.3% and processing time of 25 Hz which is better than existing methods.

  11. IAEA eLearning Program: The Use of Radiation Detection Instruments

    International Nuclear Information System (INIS)

    2010-01-01

    This CD-ROM contains a computer based training on Radiation Detection Techniques for Nuclear Security Applications. The IAEA Nuclear Security eLearning tool offers computer based training to Frontline Officers to improve their understanding about key elements of the use of radiation detection instruments. The eLearning program prepares Frontline Officers for the IAEA Detection and Response Frontline Officer course

  12. Object tracking by occlusion detection via structured sparse learning

    KAUST Repository

    Zhang, Tianzhu

    2013-06-01

    Sparse representation based methods have recently drawn much attention in visual tracking due to good performance against illumination variation and occlusion. They assume the errors caused by image variations can be modeled as pixel-wise sparse. However, in many practical scenarios these errors are not truly pixel-wise sparse but rather sparsely distributed in a structured way. In fact, pixels in error constitute contiguous regions within the object\\'s track. This is the case when significant occlusion occurs. To accommodate for non-sparse occlusion in a given frame, we assume that occlusion detected in previous frames can be propagated to the current one. This propagated information determines which pixels will contribute to the sparse representation of the current track. In other words, pixels that were detected as part of an occlusion in the previous frame will be removed from the target representation process. As such, this paper proposes a novel tracking algorithm that models and detects occlusion through structured sparse learning. We test our tracker on challenging benchmark sequences, such as sports videos, which involve heavy occlusion, drastic illumination changes, and large pose variations. Experimental results show that our tracker consistently outperforms the state-of-the-art. © 2013 IEEE.

  13. Occlusion detection via structured sparse learning for robust object tracking

    KAUST Repository

    Zhang, Tianzhu

    2014-01-01

    Sparse representation based methods have recently drawn much attention in visual tracking due to good performance against illumination variation and occlusion. They assume the errors caused by image variations can be modeled as pixel-wise sparse. However, in many practical scenarios, these errors are not truly pixel-wise sparse but rather sparsely distributed in a structured way. In fact, pixels in error constitute contiguous regions within the object’s track. This is the case when significant occlusion occurs. To accommodate for nonsparse occlusion in a given frame, we assume that occlusion detected in previous frames can be propagated to the current one. This propagated information determines which pixels will contribute to the sparse representation of the current track. In other words, pixels that were detected as part of an occlusion in the previous frame will be removed from the target representation process. As such, this paper proposes a novel tracking algorithm that models and detects occlusion through structured sparse learning. We test our tracker on challenging benchmark sequences, such as sports videos, which involve heavy occlusion, drastic illumination changes, and large pose variations. Extensive experimental results show that our proposed tracker consistently outperforms the state-of-the-art trackers.

  14. Learning to Control Advanced Life Support Systems

    Science.gov (United States)

    Subramanian, Devika

    2004-01-01

    Advanced life support systems have many interacting processes and limited resources. Controlling and optimizing advanced life support systems presents unique challenges. In particular, advanced life support systems are nonlinear coupled dynamical systems and it is difficult for humans to take all interactions into account to design an effective control strategy. In this project. we developed several reinforcement learning controllers that actively explore the space of possible control strategies, guided by rewards from a user specified long term objective function. We evaluated these controllers using a discrete event simulation of an advanced life support system. This simulation, called BioSim, designed by Nasa scientists David Kortenkamp and Scott Bell has multiple, interacting life support modules including crew, food production, air revitalization, water recovery, solid waste incineration and power. They are implemented in a consumer/producer relationship in which certain modules produce resources that are consumed by other modules. Stores hold resources between modules. Control of this simulation is via adjusting flows of resources between modules and into/out of stores. We developed adaptive algorithms that control the flow of resources in BioSim. Our learning algorithms discovered several ingenious strategies for maximizing mission length by controlling the air and water recycling systems as well as crop planting schedules. By exploiting non-linearities in the overall system dynamics, the learned controllers easily out- performed controllers written by human experts. In sum, we accomplished three goals. We (1) developed foundations for learning models of coupled dynamical systems by active exploration of the state space, (2) developed and tested algorithms that learn to efficiently control air and water recycling processes as well as crop scheduling in Biosim, and (3) developed an understanding of the role machine learning in designing control systems for

  15. Coaxial direct-detection lidar-system

    DEFF Research Database (Denmark)

    2014-01-01

    The invention relates to a coaxial direct-detection LIDAR system for measuring velocity, temperature and/or particulate density. The system comprises a laser source for emitting a laser light beam having a lasing center frequency along an emission path. The system further comprises an optical....... Finally, the system comprises a detector system arranged to receive the return signal from the optical delivery system, the detector system comprising a narrowband optical filter and a detector, the narrowband optical filter having a filter center frequency of a pass-band, wherein the center lasing...... frequency and/or the center filter frequency may be scanned. The invention further relates to an aircraft airspeed measurement device, and a wind turbine airspeed measurement device comprising the LIDAR system....

  16. A Mobile Outdoor Augmented Reality Method Combining Deep Learning Object Detection and Spatial Relationships for Geovisualization.

    Science.gov (United States)

    Rao, Jinmeng; Qiao, Yanjun; Ren, Fu; Wang, Junxing; Du, Qingyun

    2017-08-24

    The purpose of this study was to develop a robust, fast and markerless mobile augmented reality method for registration, geovisualization and interaction in uncontrolled outdoor environments. We propose a lightweight deep-learning-based object detection approach for mobile or embedded devices; the vision-based detection results of this approach are combined with spatial relationships by means of the host device's built-in Global Positioning System receiver, Inertial Measurement Unit and magnetometer. Virtual objects generated based on geospatial information are precisely registered in the real world, and an interaction method based on touch gestures is implemented. The entire method is independent of the network to ensure robustness to poor signal conditions. A prototype system was developed and tested on the Wuhan University campus to evaluate the method and validate its results. The findings demonstrate that our method achieves a high detection accuracy, stable geovisualization results and interaction.

  17. A Mobile Outdoor Augmented Reality Method Combining Deep Learning Object Detection and Spatial Relationships for Geovisualization

    Directory of Open Access Journals (Sweden)

    Jinmeng Rao

    2017-08-01

    Full Text Available The purpose of this study was to develop a robust, fast and markerless mobile augmented reality method for registration, geovisualization and interaction in uncontrolled outdoor environments. We propose a lightweight deep-learning-based object detection approach for mobile or embedded devices; the vision-based detection results of this approach are combined with spatial relationships by means of the host device’s built-in Global Positioning System receiver, Inertial Measurement Unit and magnetometer. Virtual objects generated based on geospatial information are precisely registered in the real world, and an interaction method based on touch gestures is implemented. The entire method is independent of the network to ensure robustness to poor signal conditions. A prototype system was developed and tested on the Wuhan University campus to evaluate the method and validate its results. The findings demonstrate that our method achieves a high detection accuracy, stable geovisualization results and interaction.

  18. A Mobile Outdoor Augmented Reality Method Combining Deep Learning Object Detection and Spatial Relationships for Geovisualization

    Science.gov (United States)

    Rao, Jinmeng; Qiao, Yanjun; Ren, Fu; Wang, Junxing; Du, Qingyun

    2017-01-01

    The purpose of this study was to develop a robust, fast and markerless mobile augmented reality method for registration, geovisualization and interaction in uncontrolled outdoor environments. We propose a lightweight deep-learning-based object detection approach for mobile or embedded devices; the vision-based detection results of this approach are combined with spatial relationships by means of the host device’s built-in Global Positioning System receiver, Inertial Measurement Unit and magnetometer. Virtual objects generated based on geospatial information are precisely registered in the real world, and an interaction method based on touch gestures is implemented. The entire method is independent of the network to ensure robustness to poor signal conditions. A prototype system was developed and tested on the Wuhan University campus to evaluate the method and validate its results. The findings demonstrate that our method achieves a high detection accuracy, stable geovisualization results and interaction. PMID:28837096

  19. The Joint Lessons Learned System and Interoperability

    Science.gov (United States)

    1989-06-02

    Learned: 1988-1989 As mentioned in the introduction to this chaoter, the Organizacion of the JcinC Chiefs cf Staff .OJCS) ueren significant transformatioi...Organization and Functions Manual . Washington, D.C.: HQDA, Office of the Deputy Chief 0f Staff for Operations and Plans, June 1984. ’..S. Army. Concept...U.S. Department of Defense. Joint Universal Lessons Learned System (JULLS) User’s Manual . Orlando, Florida: University of Central Florida, Institute

  20. Methods and systems for detection of radionuclides

    Science.gov (United States)

    Coates, Jr., John T.; DeVol, Timothy A.

    2010-05-25

    Disclosed are materials and systems useful in determining the existence of radionuclides in an aqueous sample. The materials provide the dual function of both extraction and scintillation to the systems. The systems can be both portable and simple to use, and as such can beneficially be utilized to determine presence and optionally concentration of radionuclide contamination in an aqueous sample at any desired location and according to a relatively simple process without the necessity of complicated sample handling techniques. The disclosed systems include a one-step process, providing simultaneous extraction and detection capability, and a two-step process, providing a first extraction step that can be carried out in a remote field location, followed by a second detection step that can be carried out in a different location.

  1. How Instructional Systems Will Manage Learning

    Science.gov (United States)

    Flanagan, John C.

    1970-01-01

    Discusses trends toward the systems approach in education including the development of effective instructional systems in government and industry; the introduction of teaching machines, programed learning, and computer- assisted instruction; and the increase in both the amount and sophistication of educational research and development. (JF)

  2. An Architecture for Open Learning Management Systems

    NARCIS (Netherlands)

    Avgeriou, Paris; Retalis, Simos; Skordalakis, Manolis

    2003-01-01

    There exists an urgent demand on defining architectures for Learning Management Systems, so that high-level frameworks for understanding these systems can be discovered, and quality attributes like portability, interoperability, reusability and modifiability can be achieved. In this paper we propose

  3. Digital detection systems for projection radiography

    International Nuclear Information System (INIS)

    Schulz, R.F.

    2001-01-01

    This article gives a survey of digital X-ray detection systems for projection radiography. The different principles are compared and some general characteristics are derived. The basic conversion mechanisms in the absorption layers are described. The basic principles of solid state X-ray detectors and their general characteristics are elucidated as well as some similarities with detectors for computed tomography. Some important application and system aspects are considered. An outlook on further possible developments in this field is given. (orig.) [de

  4. Statistics of multi-tube detecting systems

    International Nuclear Information System (INIS)

    Grau Carles, P.; Grau Malonda, A.

    1994-01-01

    In this paper three new statistical theorems are demonstrated and applied. These theorems simplify very much the obtention of the formulae to compute the counting efficiency when the detection system is formed by several photomultipliers associated in coincidence and sume. These theorems are applied to several photomultiplier arrangements in order to show their potential and the application. way

  5. The deep sea Acoustic Detection system AMADEUS

    International Nuclear Information System (INIS)

    Naumann, Christopher Lindsay

    2008-01-01

    As a part of the ANTARES neutrino telescope, the AMADEUS (ANTARES Modules for Acoustic Detection Under the Sea) system is an array of acoustical sensors designed to investigate the possibilities of acoustic detection of ultra-high energy neutrinos in the deep sea. The complete system will comprise a total of 36 acoustic sensors in six clusters on two of the ANTARES detector lines. With an inter-sensor spacing of about one metre inside the clusters and between 15 and 340 metres between the different clusters, it will cover a wide range of distances as will as provide a considerable lever arm for point source triangulation. Three of these clusters have already been deployed in 2007 and have been in operation since, currently yielding around 2GB of acoustic data per day. The remaining three clusters are scheduled to be deployed in May 2008 together with the final ANTARES detector line. Apart from proving the feasibility of operating an acoustic detection system in the deep sea, the main aim of this project is an in-depth survey of both the acoustic properties of the sea water and the acoustic background present at the detector site. It will also serve as a platform for the development and refinement of triggering, filtering and reconstruction algorithms for acoustic particle detection. In this presentation, a description of the acoustic sensor and read-out system is given, together with examples for the reconstruction and evaluation of the acoustic data.

  6. Detection of cardiovascular anomalies: Hybrid systems approach

    KAUST Repository

    Ledezma, Fernando

    2012-06-06

    In this paper, we propose a hybrid interpretation of the cardiovascular system. Based on a model proposed by Simaan et al. (2009), we study the problem of detecting cardiovascular anomalies that can be caused by variations in some physiological parameters, using an observerbased approach. We present the first numerical results obtained. © 2012 IFAC.

  7. Statistics of multi-tube detecting systems

    International Nuclear Information System (INIS)

    Grau Carles, P.; Grau Malonda, A.

    1994-01-01

    In this paper three new statistical theorems are demonstrated and applied. These theorems simplify very much the obtention of the formulae to compute the counting efficiency when the detection system is formed by several photomultipliers associated in coincidence and sum. These theorems are applied to several photomultiplier arrangements in order to show their potential and the application way. (Author) 6 refs

  8. Transitioning from learning healthcare systems to learning health care communities.

    Science.gov (United States)

    Mullins, C Daniel; Wingate, La'Marcus T; Edwards, Hillary A; Tofade, Toyin; Wutoh, Anthony

    2018-02-26

    The learning healthcare system (LHS) model framework has three core, foundational components. These include an infrastructure for health-related data capture, care improvement targets and a supportive policy environment. Despite progress in advancing and implementing LHS approaches, low levels of participation from patients and the public have hampered the transformational potential of the LHS model. An enhanced vision of a community-engaged LHS redesign would focus on the provision of health care from the patient and community perspective to complement the healthcare system as the entity that provides the environment for care. Addressing the LHS framework implementation challenges and utilizing community levers are requisite components of a learning health care community model, version two of the LHS archetype.

  9. Detecting, anticipating, and predicting critical transitions in spatially extended systems.

    Science.gov (United States)

    Kwasniok, Frank

    2018-03-01

    A data-driven linear framework for detecting, anticipating, and predicting incipient bifurcations in spatially extended systems based on principal oscillation pattern (POP) analysis is discussed. The dynamics are assumed to be governed by a system of linear stochastic differential equations which is estimated from the data. The principal modes of the system together with corresponding decay or growth rates and oscillation frequencies are extracted as the eigenvectors and eigenvalues of the system matrix. The method can be applied to stationary datasets to identify the least stable modes and assess the proximity to instability; it can also be applied to nonstationary datasets using a sliding window approach to track the changing eigenvalues and eigenvectors of the system. As a further step, a genuinely nonstationary POP analysis is introduced. Here, the system matrix of the linear stochastic model is time-dependent, allowing for extrapolation and prediction of instabilities beyond the learning data window. The methods are demonstrated and explored using the one-dimensional Swift-Hohenberg equation as an example, focusing on the dynamics of stochastic fluctuations around the homogeneous stable state prior to the first bifurcation. The POP-based techniques are able to extract and track the least stable eigenvalues and eigenvectors of the system; the nonstationary POP analysis successfully predicts the timing of the first instability and the unstable mode well beyond the learning data window.

  10. Structural damage detection using deep learning of ultrasonic guided waves

    Science.gov (United States)

    Melville, Joseph; Alguri, K. Supreet; Deemer, Chris; Harley, Joel B.

    2018-04-01

    Structural health monitoring using ultrasonic guided waves relies on accurate interpretation of guided wave propagation to distinguish damage state indicators. However, traditional physics based models do not provide an accurate representation, and classic data driven techniques, such as a support vector machine, are too simplistic to capture the complex nature of ultrasonic guide waves. To address this challenge, this paper uses a deep learning interpretation of ultrasonic guided waves to achieve fast, accurate, and automated structural damaged detection. To achieve this, full wavefield scans of thin metal plates are used, half from the undamaged state and half from the damaged state. This data is used to train our deep network to predict the damage state of a plate with 99.98% accuracy given signals from just 10 spatial locations on the plate, as compared to that of a support vector machine (SVM), which achieved a 62% accuracy.

  11. Chemical detection, identification, and analysis system

    International Nuclear Information System (INIS)

    Morel, R.S.; Gonzales, D.; Mniszewski, S.

    1990-01-01

    The chemical detection, identification, and analysis system (CDIAS) has three major goals. The first is to display safety information regarding chemical environment before personnel entry. The second is to archive personnel exposure to the environment. Third, the system assists users in identifying the stage of a chemical process in progress and suggests safety precautions associated with that process. In addition to these major goals, the system must be sufficiently compact to provide transportability, and it must be extremely simple to use in order to keep user interaction at a minimum. The system created to meet these goals includes several pieces of hardware and the integration of four software packages. The hardware consists of a low-oxygen, carbon monoxide, explosives, and hydrogen sulfide detector; an ion mobility spectrometer for airborne vapor detection; and a COMPAQ 386/20 portable computer. The software modules are a graphics kernel, an expert system shell, a data-base management system, and an interface management system. A supervisory module developed using the interface management system coordinates the interaction of the other software components. The system determines the safety of the environment using conventional data acquisition and analysis techniques. The low-oxygen, carbon monoxide, hydrogen sulfide, explosives, and vapor detectors are monitored for hazardous levels, and warnings are issued accordingly

  12. Comparison of machine learning algorithms for detecting coral reef

    Directory of Open Access Journals (Sweden)

    Eduardo Tusa

    2014-09-01

    Full Text Available (Received: 2014/07/31 - Accepted: 2014/09/23This work focuses on developing a fast coral reef detector, which is used for an autonomous underwater vehicle, AUV. A fast detection secures the AUV stabilization respect to an area of reef as fast as possible, and prevents devastating collisions. We use the algorithm of Purser et al. (2009 because of its precision. This detector has two parts: feature extraction that uses Gabor Wavelet filters, and feature classification that uses machine learning based on Neural Networks. Due to the extensive time of the Neural Networks, we exchange for a classification algorithm based on Decision Trees. We use a database of 621 images of coral reef in Belize (110 images for training and 511 images for testing. We implement the bank of Gabor Wavelets filters using C++ and the OpenCV library. We compare the accuracy and running time of 9 machine learning algorithms, whose result was the selection of the Decision Trees algorithm. Our coral detector performs 70ms of running time in comparison to 22s executed by the algorithm of Purser et al. (2009.

  13. Multi-sensor explosive detection system

    International Nuclear Information System (INIS)

    Gozani, T.; Shea, P.M.; Sawa, Z.P.

    1992-01-01

    This patent describes an explosive detection system. It comprises a source of neutrons; a detector array comprising a plurality of gamma ray detectors, each of the gamma ray detectors providing a detection signal in the event a gamma ray is captured by the detector, and at least one neutron detector, the neutron detector providing a neutron detection signal in the event a neutron is captured by the neutron detector; means for irradiating an object being examined with neutrons from the neutron source and for positioning the detector array relative to the object so that gamma rays emitted from the elements within the object as a result of the neutron irradiation are detected by the gamma ray detectors of the detector array; and parallel distributed processing means responsive to the detection signals of the detector array for discriminating between objects carrying explosives and objects not carrying explosives, the parallel distributed processing means including an artificial neural system (ANS), the ANS having a parallel network of processors, each processor of the parallel network of processors, each processor of the parallel network of processors including means for receiving at least one input signal, and means for generating an output signal as a function of the at least one input signal

  14. Evaluating Usability of E-Learning Systems in Universities

    OpenAIRE

    Nicholas Kipkurui Kiget; Professor G. Wanyembi; Anselemo Ikoha Peters

    2014-01-01

    The use of e-learning systems has increased significantly in the recent times. E-learning systems are supplementing teaching and learning in universities globally. Kenyan universities have adopted e-learning technologies as means for delivering course content. However despite adoption of these systems, there are considerable challenges facing the usability of the systems. Lecturers and students have different perceptions in regard to the usability of e-learning systems. The aim of this study ...

  15. [Detection and specific studies in procedural learning difficulties].

    Science.gov (United States)

    Magallón, S; Narbona, J

    2009-02-27

    The main disabilities in non-verbal learning disorder (NLD) are: the acquisition and automating of motor and cognitive processes, visual spatial integration, motor coordination, executive functions, difficulty in comprehension of the context, and social skills. AIMS. To review the research to date on NLD, and to discuss whether the term 'procedural learning disorder' (PLD) would be more suitable to refer to NLD. A considerable amount of research suggests a neurological correlate of PLD with dysfunctions in the 'posterior' attention system, or the right hemisphere, or the cerebellum. Even if it is said to be difficult the delimitation between NLD and other disorders or syndromes like Asperger syndrome, certain characteristics contribute to differential diagnosis. Intervention strategies for the PLD must lead to the development of motor automatisms and problem solving strategies, including social skills. The basic dysfunction in NLD affects to implicit learning of routines, automating of motor skills and cognitive strategies that spare conscious resources in daily behaviours. These limitations are partly due to a dysfunction in non-declarative procedural memory. Various dimensions of language are also involved: context comprehension, processing of the spatial and emotional indicators of verbal language, language inferences, prosody, organization of the inner speech, use of language and non-verbal communication; this is why the diagnostic label 'PLD' would be more appropriate, avoiding the euphemistic adjective 'non-verbal'.

  16. Improving the Accuracy of Cloud Detection Using Machine Learning

    Science.gov (United States)

    Craddock, M. E.; Alliss, R. J.; Mason, M.

    2017-12-01

    show 97% accuracy during the daytime, 94% accuracy at night, and 95% accuracy for all times. The total time to train, tune and test was approximately one week. The improved performance and reduced time to produce results is testament to improved computer technology and the use of machine learning as a more efficient and accurate methodology of cloud detection.

  17. Assisted Learning Systems in e-Education

    Directory of Open Access Journals (Sweden)

    Gabriel ZAMFIR

    2014-01-01

    Full Text Available Human society, analyzed as a learning environment, presumes different languages in order to know, to understand or to develop it. This statement results as a default application of the cog-nitive domain in the educational scientific research, and it highlights a key feature: each essen-tial discovery was available for the entire language compatible society. E-Society is constructed as an application of E-Science in social services, and it is going to reveal a learning system for each application of the information technology developed for a compatible society. This article is proposed as a conceptual one focused on scientific research and the interrelationship be-tween the building blocks of research, defined as an engine for any designed learning system applied in the cognitive domain. In this approach, educational research become a learning sys-tem in e-Education. The purpose of this analysis is to configure the teacher assisted learning system and to expose its main principles which could be integrated in standard assisted instruc-tion applications, available in e-Classroom, supporting the design of specific didactic activities.

  18. 46 CFR 108.405 - Fire detection system.

    Science.gov (United States)

    2010-10-01

    ... 46 Shipping 4 2010-10-01 2010-10-01 false Fire detection system. 108.405 Section 108.405 Shipping... EQUIPMENT Fire Extinguishing Systems § 108.405 Fire detection system. (a) Each fire detection system and each smoke detection system on a unit must— (1) Be approved by the Commandant; and (2) Have a visual...

  19. Machine learning paradigms applications in recommender systems

    CERN Document Server

    Lampropoulos, Aristomenis S

    2015-01-01

    This timely book presents Applications in Recommender Systems which are making recommendations using machine learning algorithms trained via examples of content the user likes or dislikes. Recommender systems built on the assumption of availability of both positive and negative examples do not perform well when negative examples are rare. It is exactly this problem that the authors address in the monograph at hand. Specifically, the books approach is based on one-class classification methodologies that have been appearing in recent machine learning research. The blending of recommender systems and one-class classification provides a new very fertile field for research, innovation and development with potential applications in “big data” as well as “sparse data” problems. The book will be useful to researchers, practitioners and graduate students dealing with problems of extensive and complex data. It is intended for both the expert/researcher in the fields of Pattern Recognition, Machine Learning and ...

  20. Deep Learning-Based Data Forgery Detection in Automatic Generation Control

    Energy Technology Data Exchange (ETDEWEB)

    Zhang, Fengli [Univ. of Arkansas, Fayetteville, AR (United States); Li, Qinghua [Univ. of Arkansas, Fayetteville, AR (United States)

    2017-10-09

    Automatic Generation Control (AGC) is a key control system in the power grid. It is used to calculate the Area Control Error (ACE) based on frequency and tie-line power flow between balancing areas, and then adjust power generation to maintain the power system frequency in an acceptable range. However, attackers might inject malicious frequency or tie-line power flow measurements to mislead AGC to do false generation correction which will harm the power grid operation. Such attacks are hard to be detected since they do not violate physical power system models. In this work, we propose algorithms based on Neural Network and Fourier Transform to detect data forgery attacks in AGC. Different from the few previous work that rely on accurate load prediction to detect data forgery, our solution only uses the ACE data already available in existing AGC systems. In particular, our solution learns the normal patterns of ACE time series and detects abnormal patterns caused by artificial attacks. Evaluations on the real ACE dataset show that our methods have high detection accuracy.

  1. Development of a portable radon detection system

    International Nuclear Information System (INIS)

    Allen, J.W.

    1976-09-01

    The presence of radon-222 in soil gas and ground water can indicate the existence of nearby uranium deposits even when heavy overburdens completely absorb the associated gamma radiation. Techniques to detect and measure radon have evolved during the past several years to the point where radon prospecting is routinely employed in a number of countries. A program to develop and field test a prototype system for measuring radon from soil gas and water is described. A prototype system employing a flow through scintillation detector was designed and constructed, utilizing standard commercial components, to provide a fieldworthy unit for testing the system concepts. Laboratory and preliminary field tests of this unit indicate that it can detect anomalous radon levels of less than 10 picoCuries per liter (pCi/l) in soil gas and ground water

  2. Evaluation of pipeline leak detection systems

    International Nuclear Information System (INIS)

    Glauz, W.D.; Flora, J.D.; Hennon, G.J.

    1993-01-01

    Leaking underground storage tank system presents an environmental concern and a potential health hazard. It is well known that leaks in the piping associated with these systems account for a sizeable fraction of the leaks. EPA has established performance standards for pipeline leak detection systems, and published a document presenting test protocols for evaluating these systems against the standards. This paper discusses a number of facets and important features of evaluating such systems, and presents results from tests of several systems. The importance of temperature differences between the ground and the product in the line is shown both in theory and with test data. The impact of the amount of soil moisture present is addressed, along with the effect of frozen soil. These features are addressed both for line tightness test systems, which must detect leaks of 0.10 gal/h (0.38 L/h) at 150% of normal line pressure, or 0.20 gal/h (0.76 L/h) at normal line pressure, and for automatic line leak detectors that must detect leaks of 3 gal/h (11 L/h) at 10 psi (69 kPa) within an hour of the occurrence of the leak. This paper also addresses some statistical aspects of the evaluation of these systems. Reasons for keeping the evaluation process ''blind'' to the evaluated company are given, along with methods for assuring that the tests are blind. Most importantly, a test procedure is presented for evaluating systems that report a flow rate (not just a pass/fail decision) that is much more efficient than the procedure presented in the EPA protocol, and is just as stringent

  3. Multidimensional Learner Model In Intelligent Learning System

    Science.gov (United States)

    Deliyska, B.; Rozeva, A.

    2009-11-01

    The learner model in an intelligent learning system (ILS) has to ensure the personalization (individualization) and the adaptability of e-learning in an online learner-centered environment. ILS is a distributed e-learning system whose modules can be independent and located in different nodes (servers) on the Web. This kind of e-learning is achieved through the resources of the Semantic Web and is designed and developed around a course, group of courses or specialty. An essential part of ILS is learner model database which contains structured data about learner profile and temporal status in the learning process of one or more courses. In the paper a learner model position in ILS is considered and a relational database is designed from learner's domain ontology. Multidimensional modeling agent for the source database is designed and resultant learner data cube is presented. Agent's modules are proposed with corresponding algorithms and procedures. Multidimensional (OLAP) analysis guidelines on the resultant learner module for designing dynamic learning strategy have been highlighted.

  4. Leak rate measurements and detection systems

    International Nuclear Information System (INIS)

    Kupperman, D.; Shack, W.J.; Claytor, T.

    1983-10-01

    A research program is under way to evaluate and develop improve leak detection systems. The primary focus of the work has been on acoustic emission detection of leaks. Leaks from artificial flaws, laboratory-generated IGSCCs and thermal fatigue cracks, and field-induced intergranular stress corrosion cracks (IGSCCs) from reactor piping have been examined. The effects of pressure, temperature, and leak rate and geometry on the acoustic signature are under study. The use of cross-correlation techniques for leak location and pattern recognition and autocorrelation for source discrimination is also being considered

  5. Bat detective—Deep learning tools for bat acoustic signal detection

    Science.gov (United States)

    Barlow, Kate E.; Firman, Michael; Freeman, Robin; Harder, Briana; Kinsey, Libby; Mead, Gary R.; Newson, Stuart E.; Pandourski, Ivan; Russ, Jon; Szodoray-Paradi, Abigel; Tilova, Elena; Girolami, Mark; Jones, Kate E.

    2018-01-01

    Passive acoustic sensing has emerged as a powerful tool for quantifying anthropogenic impacts on biodiversity, especially for echolocating bat species. To better assess bat population trends there is a critical need for accurate, reliable, and open source tools that allow the detection and classification of bat calls in large collections of audio recordings. The majority of existing tools are commercial or have focused on the species classification task, neglecting the important problem of first localizing echolocation calls in audio which is particularly problematic in noisy recordings. We developed a convolutional neural network based open-source pipeline for detecting ultrasonic, full-spectrum, search-phase calls produced by echolocating bats. Our deep learning algorithms were trained on full-spectrum ultrasonic audio collected along road-transects across Europe and labelled by citizen scientists from www.batdetective.org. When compared to other existing algorithms and commercial systems, we show significantly higher detection performance of search-phase echolocation calls with our test sets. As an example application, we ran our detection pipeline on bat monitoring data collected over five years from Jersey (UK), and compared results to a widely-used commercial system. Our detection pipeline can be used for the automatic detection and monitoring of bat populations, and further facilitates their use as indicator species on a large scale. Our proposed pipeline makes only a small number of bat specific design decisions, and with appropriate training data it could be applied to detecting other species in audio. A crucial novelty of our work is showing that with careful, non-trivial, design and implementation considerations, state-of-the-art deep learning methods can be used for accurate and efficient monitoring in audio. PMID:29518076

  6. Subsea leak detection systems - recommended practice

    Energy Technology Data Exchange (ETDEWEB)

    Berg, Kristin Nergaard

    2010-07-01

    It is known in the industry that occasional leakages occur in subsea production systems. In spite of efforts to apply subsea leak detectors, the experience is that most leakages are either detected by ROV during routine inspections or interventions or as oil slicks on the surface . Operators and authority awareness towards the environmental impact of oil and gas production is increasing. The regulatory bodies in Norway, EU and USA specify requirements for detection of acute pollution. This paper presents the development of a Recommended Practice (RP) sponsored by OLF (The Norwegian Oil Industry Association). The JIP includes several major oil and gas operators. The objective of the RP is to serve as a technical and practical reference in the field of subsea leak detection. (Author)

  7. Surface blemish detection from passive imagery using learned fuzzy set concepts

    International Nuclear Information System (INIS)

    Gurbuz, S.; Carver, A.; Schalkoff, R.

    1997-12-01

    An image analysis method for real-time surface blemish detection using passive imagery and fuzzy set concepts is described. The method develops an internal knowledge representation for surface blemish characteristics on the basis of experience, thus facilitating autonomous learning based upon positive and negative exemplars. The method incorporates fuzzy set concepts in the learning subsystem and image segmentation algorithms, thereby mimicking human visual perception. This enables a generic solution for color image segmentation. This method has been applied in the development of ARIES (Autonomous Robotic Inspection Experimental System), designed to inspect DOE warehouse waste storage drums for rust. In this project, the ARIES vision system is used to acquire drum surface images under controlled conditions and subsequently perform visual inspection leading to the classification of the drum as acceptable or suspect

  8. Improved detection of chemical substances from colorimetric sensor data using probabilistic machine learning

    Science.gov (United States)

    Mølgaard, Lasse L.; Buus, Ole T.; Larsen, Jan; Babamoradi, Hamid; Thygesen, Ida L.; Laustsen, Milan; Munk, Jens Kristian; Dossi, Eleftheria; O'Keeffe, Caroline; Lässig, Lina; Tatlow, Sol; Sandström, Lars; Jakobsen, Mogens H.

    2017-05-01

    We present a data-driven machine learning approach to detect drug- and explosives-precursors using colorimetric sensor technology for air-sampling. The sensing technology has been developed in the context of the CRIM-TRACK project. At present a fully- integrated portable prototype for air sampling with disposable sensing chips and automated data acquisition has been developed. The prototype allows for fast, user-friendly sampling, which has made it possible to produce large datasets of colorimetric data for different target analytes in laboratory and simulated real-world application scenarios. To make use of the highly multi-variate data produced from the colorimetric chip a number of machine learning techniques are employed to provide reliable classification of target analytes from confounders found in the air streams. We demonstrate that a data-driven machine learning method using dimensionality reduction in combination with a probabilistic classifier makes it possible to produce informative features and a high detection rate of analytes. Furthermore, the probabilistic machine learning approach provides a means of automatically identifying unreliable measurements that could produce false predictions. The robustness of the colorimetric sensor has been evaluated in a series of experiments focusing on the amphetamine pre-cursor phenylacetone as well as the improvised explosives pre-cursor hydrogen peroxide. The analysis demonstrates that the system is able to detect analytes in clean air and mixed with substances that occur naturally in real-world sampling scenarios. The technology under development in CRIM-TRACK has the potential as an effective tool to control trafficking of illegal drugs, explosive detection, or in other law enforcement applications.

  9. Multiple systems for motor skill learning.

    Science.gov (United States)

    Clark, Dav; Ivry, Richard B

    2010-07-01

    Motor learning is a ubiquitous feature of human competence. This review focuses on two particular classes of model tasks for studying skill acquisition. The serial reaction time (SRT) task is used to probe how people learn sequences of actions, while adaptation in the context of visuomotor or force field perturbations serves to illustrate how preexisting movements are recalibrated in novel environments. These tasks highlight important issues regarding the representational changes that occur during the course of motor learning. One important theme is that distinct mechanisms vary in their information processing costs during learning and performance. Fast learning processes may require few trials to produce large changes in performance but impose demands on cognitive resources. Slower processes are limited in their ability to integrate complex information but minimally demanding in terms of attention or processing resources. The representations derived from fast systems may be accessible to conscious processing and provide a relatively greater measure of flexibility, while the representations derived from slower systems are more inflexible and automatic in their behavior. In exploring these issues, we focus on how multiple neural systems may interact and compete during the acquisition and consolidation of new behaviors. Copyright © 2010 John Wiley & Sons, Ltd. This article is categorized under: Psychology > Motor Skill and Performance. Copyright © 2010 John Wiley & Sons, Ltd.

  10. Automated Detection of Microaneurysms Using Scale-Adapted Blob Analysis and Semi-Supervised Learning

    Energy Technology Data Exchange (ETDEWEB)

    Adal, Kedir M. [Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States); Sidebe, Desire [Univ. of Burgundy, Dijon (France); Ali, Sharib [Univ. of Burgundy, Dijon (France); Chaum, Edward [Univ. of Tennessee, Knoxville, TN (United States); Karnowski, Thomas Paul [Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States); Meriaudeau, Fabrice [Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)

    2014-01-07

    Despite several attempts, automated detection of microaneurysm (MA) from digital fundus images still remains to be an open issue. This is due to the subtle nature of MAs against the surrounding tissues. In this paper, the microaneurysm detection problem is modeled as finding interest regions or blobs from an image and an automatic local-scale selection technique is presented. Several scale-adapted region descriptors are then introduced to characterize these blob regions. A semi-supervised based learning approach, which requires few manually annotated learning examples, is also proposed to train a classifier to detect true MAs. The developed system is built using only few manually labeled and a large number of unlabeled retinal color fundus images. The performance of the overall system is evaluated on Retinopathy Online Challenge (ROC) competition database. A competition performance measure (CPM) of 0.364 shows the competitiveness of the proposed system against state-of-the art techniques as well as the applicability of the proposed features to analyze fundus images.

  11. Automated detection of microaneurysms using scale-adapted blob analysis and semi-supervised learning.

    Science.gov (United States)

    Adal, Kedir M; Sidibé, Désiré; Ali, Sharib; Chaum, Edward; Karnowski, Thomas P; Mériaudeau, Fabrice

    2014-04-01

    Despite several attempts, automated detection of microaneurysm (MA) from digital fundus images still remains to be an open issue. This is due to the subtle nature of MAs against the surrounding tissues. In this paper, the microaneurysm detection problem is modeled as finding interest regions or blobs from an image and an automatic local-scale selection technique is presented. Several scale-adapted region descriptors are introduced to characterize these blob regions. A semi-supervised based learning approach, which requires few manually annotated learning examples, is also proposed to train a classifier which can detect true MAs. The developed system is built using only few manually labeled and a large number of unlabeled retinal color fundus images. The performance of the overall system is evaluated on Retinopathy Online Challenge (ROC) competition database. A competition performance measure (CPM) of 0.364 shows the competitiveness of the proposed system against state-of-the art techniques as well as the applicability of the proposed features to analyze fundus images. Copyright © 2013 Elsevier Ireland Ltd. All rights reserved.

  12. Clone tag detection in distributed RFID systems

    Science.gov (United States)

    Kamaludin, Hazalila; Mahdin, Hairulnizam

    2018-01-01

    Although Radio Frequency Identification (RFID) is poised to displace barcodes, security vulnerabilities pose serious challenges for global adoption of the RFID technology. Specifically, RFID tags are prone to basic cloning and counterfeiting security attacks. A successful cloning of the RFID tags in many commercial applications can lead to many serious problems such as financial losses, brand damage, safety and health of the public. With many industries such as pharmaceutical and businesses deploying RFID technology with a variety of products, it is important to tackle RFID tag cloning problem and improve the resistance of the RFID systems. To this end, we propose an approach for detecting cloned RFID tags in RFID systems with high detection accuracy and minimal overhead thus overcoming practical challenges in existing approaches. The proposed approach is based on consistency of dual hash collisions and modified count-min sketch vector. We evaluated the proposed approach through extensive experiments and compared it with existing baseline approaches in terms of execution time and detection accuracy under varying RFID tag cloning ratio. The results of the experiments show that the proposed approach outperforms the baseline approaches in cloned RFID tag detection accuracy. PMID:29565982

  13. The synchronous active neutron detection assay system

    International Nuclear Information System (INIS)

    Pickrell, M.M.; Kendall, P.K.

    1994-01-01

    We have begun to develop a novel technique for active neutron assay of fissile material in spent nuclear fuel. This approach will exploit a 14-MeV neutron generator developed by Schlumberger. The technique, termed synchronous active neutron detection (SAND), follows a method used routinely in other branches of physics to detect very small signals in presence of large backgrounds. Synchronous detection instruments are widely available commercially and are termed ''lock-in'' amplifiers. We have implemented a digital lock-in amplifier in conjunction with the Schlumberger neutron generator to explore the possibility of synchronous detection with active neutrons. The Schlumberger system can operate at up to a 50% duty factor, in effect, a square wave of neutron yield. Results are preliminary but promising. The system is capable of resolving the fissile material contained in a small fraction of the fuel rods in a cold fuel assembly; it also appears resilient to background neutron interference. The interrogating neutrons appear to be non-thermal and penetrating. Work remains to fully explore relevant physics and optimize instrument design

  14. Clone tag detection in distributed RFID systems.

    Science.gov (United States)

    Kamaludin, Hazalila; Mahdin, Hairulnizam; Abawajy, Jemal H

    2018-01-01

    Although Radio Frequency Identification (RFID) is poised to displace barcodes, security vulnerabilities pose serious challenges for global adoption of the RFID technology. Specifically, RFID tags are prone to basic cloning and counterfeiting security attacks. A successful cloning of the RFID tags in many commercial applications can lead to many serious problems such as financial losses, brand damage, safety and health of the public. With many industries such as pharmaceutical and businesses deploying RFID technology with a variety of products, it is important to tackle RFID tag cloning problem and improve the resistance of the RFID systems. To this end, we propose an approach for detecting cloned RFID tags in RFID systems with high detection accuracy and minimal overhead thus overcoming practical challenges in existing approaches. The proposed approach is based on consistency of dual hash collisions and modified count-min sketch vector. We evaluated the proposed approach through extensive experiments and compared it with existing baseline approaches in terms of execution time and detection accuracy under varying RFID tag cloning ratio. The results of the experiments show that the proposed approach outperforms the baseline approaches in cloned RFID tag detection accuracy.

  15. Handheld microwave bomb-detecting imaging system

    Science.gov (United States)

    Gorwara, Ashok; Molchanov, Pavlo

    2017-05-01

    Proposed novel imaging technique will provide all weather high-resolution imaging and recognition capability for RF/Microwave signals with good penetration through highly scattered media: fog, snow, dust, smoke, even foliage, camouflage, walls and ground. Image resolution in proposed imaging system is not limited by diffraction and will be determined by processor and sampling frequency. Proposed imaging system can simultaneously cover wide field of view, detect multiple targets and can be multi-frequency, multi-function. Directional antennas in imaging system can be close positioned and installed in cell phone size handheld device, on small aircraft or distributed around protected border or object. Non-scanning monopulse system allows dramatically decrease in transmitting power and at the same time provides increased imaging range by integrating 2-3 orders more signals than regular scanning imaging systems.

  16. 3D Game-Based Learning System for Improving Learning Achievement in Software Engineering Curriculum

    Science.gov (United States)

    Su,Chung-Ho; Cheng, Ching-Hsue

    2013-01-01

    The advancement of game-based learning has encouraged many related studies, such that students could better learn curriculum by 3-dimension virtual reality. To enhance software engineering learning, this paper develops a 3D game-based learning system to assist teaching and assess the students' motivation, satisfaction and learning achievement. A…

  17. An integrated system for interactive continuous learning of categorical knowledge

    Science.gov (United States)

    Skočaj, Danijel; Vrečko, Alen; Mahnič, Marko; Janíček, Miroslav; Kruijff, Geert-Jan M.; Hanheide, Marc; Hawes, Nick; Wyatt, Jeremy L.; Keller, Thomas; Zhou, Kai; Zillich, Michael; Kristan, Matej

    2016-09-01

    This article presents an integrated robot system capable of interactive learning in dialogue with a human. Such a system needs to have several competencies and must be able to process different types of representations. In this article, we describe a collection of mechanisms that enable integration of heterogeneous competencies in a principled way. Central to our design is the creation of beliefs from visual and linguistic information, and the use of these beliefs for planning system behaviour to satisfy internal drives. The system is able to detect gaps in its knowledge and to plan and execute actions that provide information needed to fill these gaps. We propose a hierarchy of mechanisms which are capable of engaging in different kinds of learning interactions, e.g. those initiated by a tutor or by the system itself. We present the theory these mechanisms are build upon and an instantiation of this theory in the form of an integrated robot system. We demonstrate the operation of the system in the case of learning conceptual models of objects and their visual properties.

  18. Early detection of incipient faults in power plants using accelerated neural network learning

    International Nuclear Information System (INIS)

    Parlos, A.G.; Jayakumar, M.; Atiya, A.

    1992-01-01

    An important aspect of power plant automation is the development of computer systems able to detect and isolate incipient (slowly developing) faults at the earliest possible stages of their occurrence. In this paper, the development and testing of such a fault detection scheme is presented based on recognition of sensor signatures during various failure modes. An accelerated learning algorithm, namely adaptive backpropagation (ABP), has been developed that allows the training of a multilayer perceptron (MLP) network to a high degree of accuracy, with an order of magnitude improvement in convergence speed. An artificial neural network (ANN) has been successfully trained using the ABP algorithm, and it has been extensively tested with simulated data to detect and classify incipient faults of various types and severity and in the presence of varying sensor noise levels

  19. Towards synergy between learning management systems and educational server applications

    OpenAIRE

    Hartog, R.J.M.; Schaaf, van der, H.; Kassahun, A.

    2008-01-01

    Most well-known Learning Management Systems (LMS) are based on a paradigm of learning objects to be uploaded into the system. Most formulations of this paradigm implicitly assume that the learning objects are self contained learning objects such as FLASH objects or JAVA applets or presentational learning objects such as slide presentations. These are typically client side objects. However, a demand for learning support that activates the student can often be satisfied better with a server app...

  20. Indirect Tire Monitoring System - Machine Learning Approach

    Science.gov (United States)

    Svensson, O.; Thelin, S.; Byttner, S.; Fan, Y.

    2017-10-01

    The heavy vehicle industry has today no requirement to provide a tire pressure monitoring system by law. This has created issues surrounding unknown tire pressure and thread depth during active service. There is also no standardization for these kind of systems which means that different manufacturers and third party solutions work after their own principles and it can be hard to know what works for a given vehicle type. The objective is to create an indirect tire monitoring system that can generalize a method that detect both incorrect tire pressure and thread depth for different type of vehicles within a fleet without the need for additional physical sensors or vehicle specific parameters. The existing sensors that are connected communicate through CAN and are interpreted by the Drivec Bridge hardware that exist in the fleet. By using supervised machine learning a classifier was created for each axle where the main focus was the front axle which had the most issues. The classifier will classify the vehicles tires condition and will be implemented in Drivecs cloud service where it will receive its data. The resulting classifier is a random forest implemented in Python. The result from the front axle with a data set consisting of 9767 samples of buses with correct tire condition and 1909 samples of buses with incorrect tire condition it has an accuracy of 90.54% (0.96%). The data sets are created from 34 unique measurements from buses between January and May 2017. This classifier has been exported and is used inside a Node.js module created for Drivecs cloud service which is the result of the whole implementation. The developed solution is called Indirect Tire Monitoring System (ITMS) and is seen as a process. This process will predict bad classes in the cloud which will lead to warnings. The warnings are defined as incidents. They contain only the information needed and the bandwidth of the incidents are also controlled so incidents are created within an

  1. An Instructional and Collaborative Learning System with Content Recommendation

    Science.gov (United States)

    Zheng, Xiang-wei; Ma, Hong-wei; Li, Yan

    2013-01-01

    With the rapid development of Internet, e-learning has become a new teaching and learning mode. However, lots of e-learning systems deployed on Internet are just electronic learning materials with very limited interactivity and diagnostic capability. This paper presents an integrated e-learning environment named iCLSR. Firstly, iCLSR provides an…

  2. Learning through the taste system

    Directory of Open Access Journals (Sweden)

    Thomas R. Scott

    2011-11-01

    Full Text Available Taste is the final arbiter of which chemicals from the environment will be admitted to the body. The action of swallowing a substance leads to a physiological consequence of which the taste system should be informed. Accordingly, taste neurons in the central nervous system are closely allied with those that receive input from the viscera so as to monitor the impact of a recently ingested substance. There is behavioral, anatomical, electrophysiological, gene expression, and neurochemical evidence that the consequences of ingestion influence subsequent food selection through development of either a conditioned taste aversion (if illness ensues or a conditioned taste preference (if satiety. This ongoing communication between taste and the viscera permits the animal to tailor its taste system to its individual needs over a lifetime.

  3. 29 CFR 1910.164 - Fire detection systems.

    Science.gov (United States)

    2010-07-01

    ... 29 Labor 5 2010-07-01 2010-07-01 false Fire detection systems. 1910.164 Section 1910.164 Labor... detection systems. (a) Scope and application. This section applies to all automatic fire detection systems... detection systems and components to normal operating condition as promptly as possible after each test or...

  4. Multichannel Digital Emulator of Radiation Detection Systems

    International Nuclear Information System (INIS)

    Abba, A.; Caponio, F.; Geraci, A.

    2013-06-01

    A digital system for emulating in real time signals from generic setups for radiation detection is presented. The instrument is not a pulse generator of recorded shapes but a synthesizer of random pulses compliant to programmable statistics for energy and occurrence time. Completely programmable procedures for emulation of noise, disturbances and reference level variation can be implemented. The instrument has been realized and fully tested. (authors)

  5. Perimeter intrusion detection and assessment system

    International Nuclear Information System (INIS)

    Eaton, M.J.; Jacobs, J.; McGovern, D.E.

    1977-11-01

    To obtain an effective perimeter intrusion detection system requires careful sensor selection, procurement, and installation. The selection process involves a thorough understanding of the unique site features and how these features affect the performance of each type of sensor. It is necessary to develop procurement specifications to establish acceptable sensor performance limits. Careful explanation and inspection of critical installation dimensions is required during on-site construction. The implementation of these activities at a particular site is discussed

  6. Information Systems in University Learning

    Directory of Open Access Journals (Sweden)

    Gheorghe SABAU

    2010-01-01

    Full Text Available The authors of this article are going to bring into light the significance, the place and the role of information systems in the university education process. At the same time they define the objectives and the target group of the subject named Economic Information Systems and state the competence gained by students by studying this subject. Special attention is given to the curriculum to be taught to students and to a suggestive enumeration of a series of economic applications that can be themes for laboratory practice and for students’ dissertation (graduation thesis.

  7. In-Situ Wire Damage Detection System

    Science.gov (United States)

    Williams, Martha K. (Inventor); Roberson, Luke B. (Inventor); Tate, Lanetra C. (Inventor); Smith, Trent M. (Inventor); Gibson, Tracy L. (Inventor); Jolley, Scott T. (Inventor); Medelius, Pedro J. (Inventor)

    2014-01-01

    An in-situ system for detecting damage in an electrically conductive wire. The system includes a substrate at least partially covered by a layer of electrically conductive material forming a continuous or non-continuous electrically conductive layer connected to an electrical signal generator adapted to delivering electrical signals to the electrically conductive layer. Data is received and processed to identify damage to the substrate or electrically conductive layer. The electrically conductive material may include metalized carbon fibers, a thin metal coating, a conductive polymer, carbon nanotubes, metal nanoparticles or a combination thereof.

  8. The remote atmospheric and ionospheric detection system

    International Nuclear Information System (INIS)

    McCoy, R.P.; Wolfram, K.D.; Meier, R.R.

    1986-01-01

    The Remote Atmospheric and Ionospheric Detection System (RAIDS) experiment, to fly on a TIROS spacecraft in the late 1980's, consists of a comprehensive set of one limb imaging and seven limb scanning optical sensors. These eight instruments span the spectral range from the extreme ultraviolet to the near infrared, allowing simultaneous observations of the neutral and ion composition on the day and night side as well as in the auroral region. The primary objective of RAIDS is to demonstrate a system for remote sensing of the ionosphere to produce global maps of the electron density, peak altitude and critical frequency

  9. Detection Learning Style Vark For Out Of School Children (OSC)

    Science.gov (United States)

    Amran, Ali; Desiani, Anita; Hasibuan, MS

    2017-04-01

    Learning style is different for every learner especially for out of school children or OSC. They are not like formal students, they are learners but they don’t have a teacher as a guide for learning. E-learning is one of the solutions to help OSC to get education. E-learning should have preferred learning styles of learners. Data for identifying the learning style in this study were collected with a VARK questionnaire from 25 OSC in junior high school level from 5 municipalities in Palembang. The validity of the questionnaire was considered on basis of experts’ views and its reliability was calculated by using Cronbach’s alpha coefficients (α=0.68). Overall, 55% preferred to use a single learning style (Uni-modal). Of these, 27,76% preferred Aural, 20,57% preferred Reading Writing, 33,33% preferred Kinaesthetic and 23,13% preferred Visual. 45% of OSC preferred more than one style, 30% chose two-modes (bimodal), and 15% chose three-modes (tri-modal). The Most preferred Learning style of OSC is kinaesthetic learning. Kinaesthetic learning requires body movements, interactivities, and direct contacts with learning materials, these things can be difficult to implement in eLearning, but E-learning should be able to adopt any learning styles which are flexible in terms of time, period, curriculum, pedagogy, location, and language.

  10. Implementing the learning health care system.

    NARCIS (Netherlands)

    Verheij, R.; Barten, D.J.; Hek, K.; Nielen, M.; Prins, M.; Zwaanswijk, M.; Bakker, D. de

    2014-01-01

    Background: As computerisation of primary care facilities is rapidly increasing, a wealth of data is created in routinely recorded electronic health records (EHRs). This data can be used to create a true learning health care system, in which routinely available data are processed and analysed in

  11. Distance Learning Delivery Systems: Instructional Options.

    Science.gov (United States)

    Steele, Ray L.

    1993-01-01

    Discusses the availability of satellite and cable programing to provide distance education opportunities in school districts. Various delivery systems are described, including telephones with speakers, personal computers, and satellite dishes; and a sidebar provides a directory of distance learning opportunities, including telecommunications…

  12. Learning System Center App Controller

    CERN Document Server

    Naeem, Nasir

    2015-01-01

    This book is intended for IT professionals working with Hyper-V, Azure cloud, VMM, and private cloud technologies who are looking for a quick way to get up and running with System Center 2012 R2 App Controller. To get the most out of this book, you should be familiar with Microsoft Hyper-V technology. Knowledge of Virtual Machine Manager is helpful but not mandatory.

  13. Learning to teach secondary mathematics using an online learning system

    Science.gov (United States)

    Cavanagh, Michael; Mitchelmore, Michael

    2011-12-01

    We report the results of a classroom study of three secondary mathematics teachers who had no prior experience teaching with technology as they began to use an online mathematics learning system in their lessons. We gave the teachers only basic instruction on how to operate the system and then observed them intensively over four school terms as they taught using it. We documented changes in the teachers' Pedagogical Technology Knowledge and subsequently classified their various roles as technology bystanders, adopters, adaptors and innovators. Results show that all teachers made some progress toward using the system in more sophisticated ways, but the improvements were not uniform across the teachers. We suggest possible reasons to explain the variation and discuss some implications for teacher professional development.

  14. Multiple-instance learning for computer-aided detection of tuberculosis

    Science.gov (United States)

    Melendez, J.; Sánchez, C. I.; Philipsen, R. H. H. M.; Maduskar, P.; van Ginneken, B.

    2014-03-01

    Detection of tuberculosis (TB) on chest radiographs (CXRs) is a hard problem. Therefore, to help radiologists or even take their place when they are not available, computer-aided detection (CAD) systems are being developed. In order to reach a performance comparable to that of human experts, the pattern recognition algorithms of these systems are typically trained on large CXR databases that have been manually annotated to indicate the abnormal lung regions. However, manually outlining those regions constitutes a time-consuming process that, besides, is prone to inconsistencies and errors introduced by interobserver variability and the absence of an external reference standard. In this paper, we investigate an alternative pattern classi cation method, namely multiple-instance learning (MIL), that does not require such detailed information for a CAD system to be trained. We have applied this alternative approach to a CAD system aimed at detecting textural lesions associated with TB. Only the case (or image) condition (normal or abnormal) was provided in the training stage. We compared the resulting performance with those achieved by several variations of a conventional system trained with detailed annotations. A database of 917 CXRs was constructed for experimentation. It was divided into two roughly equal parts that were used as training and test sets. The area under the receiver operating characteristic curve was utilized as a performance measure. Our experiments show that, by applying the investigated MIL approach, comparable results as with the aforementioned conventional systems are obtained in most cases, without requiring condition information at the lesion level.

  15. Automating Vendor Fraud Detection in Enterprise Systems

    Directory of Open Access Journals (Sweden)

    Kishore Singh

    2013-06-01

    Full Text Available Fraud is a multi-billion dollar industry that continues to grow annually. Many organisations are poorly prepared to prevent and detect fraud. Fraud detection strategies are intended to quickly and efficiently identify fraudulent activities that circumvent preventative measures. In this paper we adopt a Design-Science methodological framework to develop a model for detection of vendor fraud based on analysis of patterns or signatures identified in enterprise system audit trails. The concept is demonstrated be developing prototype software. Verification of the prototype is achieved by performing a series of experiments. Validation is achieved by independent reviews from auditing practitioners. Key findings of this study are: i automating routine data analytics improves auditor productivity and reduces time taken to identify potential fraud, and ii visualisations assist in promptly identifying potentially fraudulent user activities. The study makes the following contributions: i a model for proactive fraud detection, ii methods for visualising user activities in transaction data, iii a stand-alone MCL-based prototype.

  16. Principles of e-learning systems engineering

    CERN Document Server

    Gilbert, Lester

    2008-01-01

    The book integrates the principles of software engineering with the principles of educational theory, and applies them to the problems of e-learning development, thus establishing the discipline of E-learning systems engineering. For the first time, these principles are collected and organised into the coherent framework that this book provides. Both newcomers to and established practitioners in the field are provided with integrated and grounded advice on theory and practice. The book presents strong practical and theoretical frameworks for the design and development of technology-based mater

  17. Detecting false positive sequence homology: a machine learning approach.

    Science.gov (United States)

    Fujimoto, M Stanley; Suvorov, Anton; Jensen, Nicholas O; Clement, Mark J; Bybee, Seth M

    2016-02-24

    Accurate detection of homologous relationships of biological sequences (DNA or amino acid) amongst organisms is an important and often difficult task that is essential to various evolutionary studies, ranging from building phylogenies to predicting functional gene annotations. There are many existing heuristic tools, most commonly based on bidirectional BLAST searches that are used to identify homologous genes and combine them into two fundamentally distinct classes: orthologs and paralogs. Due to only using heuristic filtering based on significance score cutoffs and having no cluster post-processing tools available, these methods can often produce multiple clusters constituting unrelated (non-homologous) sequences. Therefore sequencing data extracted from incomplete genome/transcriptome assemblies originated from low coverage sequencing or produced by de novo processes without a reference genome are susceptible to high false positive rates of homology detection. In this paper we develop biologically informative features that can be extracted from multiple sequence alignments of putative homologous genes (orthologs and paralogs) and further utilized in context of guided experimentation to verify false positive outcomes. We demonstrate that our machine learning method trained on both known homology clusters obtained from OrthoDB and randomly generated sequence alignments (non-homologs), successfully determines apparent false positives inferred by heuristic algorithms especially among proteomes recovered from low-coverage RNA-seq data. Almost ~42 % and ~25 % of predicted putative homologies by InParanoid and HaMStR respectively were classified as false positives on experimental data set. Our process increases the quality of output from other clustering algorithms by providing a novel post-processing method that is both fast and efficient at removing low quality clusters of putative homologous genes recovered by heuristic-based approaches.

  18. Component-Based Approach in Learning Management System Development

    Science.gov (United States)

    Zaitseva, Larisa; Bule, Jekaterina; Makarov, Sergey

    2013-01-01

    The paper describes component-based approach (CBA) for learning management system development. Learning object as components of e-learning courses and their metadata is considered. The architecture of learning management system based on CBA being developed in Riga Technical University, namely its architecture, elements and possibilities are…

  19. A Situated Cultural Festival Learning System Based on Motion Sensing

    Science.gov (United States)

    Chang, Yi-Hsing; Lin, Yu-Kai; Fang, Rong-Jyue; Lu, You-Te

    2017-01-01

    A situated Chinese cultural festival learning system based on motion sensing is developed in this study. The primary design principle is to create a highly interactive learning environment, allowing learners to interact with Kinect through natural gestures in the designed learning situation to achieve efficient learning. The system has the…

  20. Machine Learning Techniques for Optical Performance Monitoring from Directly Detected PDM-QAM Signals

    DEFF Research Database (Denmark)

    Thrane, Jakob; Wass, Jesper; Piels, Molly

    2017-01-01

    Linear signal processing algorithms are effective in dealing with linear transmission channel and linear signal detection, while the nonlinear signal processing algorithms, from the machine learning community, are effective in dealing with nonlinear transmission channel and nonlinear signal...... detection. In this paper, a brief overview of the various machine learning methods and their application in optical communication is presented and discussed. Moreover, supervised machine learning methods, such as neural networks and support vector machine, are experimentally demonstrated for in-band optical...

  1. Neural Network Based Intrusion Detection System for Critical Infrastructures

    Energy Technology Data Exchange (ETDEWEB)

    Todd Vollmer; Ondrej Linda; Milos Manic

    2009-07-01

    Resiliency and security in control systems such as SCADA and Nuclear plant’s in today’s world of hackers and malware are a relevant concern. Computer systems used within critical infrastructures to control physical functions are not immune to the threat of cyber attacks and may be potentially vulnerable. Tailoring an intrusion detection system to the specifics of critical infrastructures can significantly improve the security of such systems. The IDS-NNM – Intrusion Detection System using Neural Network based Modeling, is presented in this paper. The main contributions of this work are: 1) the use and analyses of real network data (data recorded from an existing critical infrastructure); 2) the development of a specific window based feature extraction technique; 3) the construction of training dataset using randomly generated intrusion vectors; 4) the use of a combination of two neural network learning algorithms – the Error-Back Propagation and Levenberg-Marquardt, for normal behavior modeling. The presented algorithm was evaluated on previously unseen network data. The IDS-NNM algorithm proved to be capable of capturing all intrusion attempts presented in the network communication while not generating any false alerts.

  2. Incipient fault detection and power system protection for spaceborne systems

    Science.gov (United States)

    Russell, B. Don; Hackler, Irene M.

    1987-01-01

    A program was initiated to study the feasibility of using advanced terrestrial power system protection techniques for spacecraft power systems. It was designed to enhance and automate spacecraft power distribution systems in the areas of safety, reliability and maintenance. The proposed power management/distribution system is described as well as security assessment and control, incipient and low current fault detection, and the proposed spaceborne protection system. It is noted that the intelligent remote power controller permits the implementation of digital relaying algorithms with both adaptive and programmable characteristics.

  3. DESIGNING MOTIVATIONAL LEARNING SYSTEMS IN DISTANCE EDUCATION

    Directory of Open Access Journals (Sweden)

    Jale BALABAN-SALI

    2008-07-01

    Full Text Available ABSTRACT The designing of instruction, when considered as a process, is the determination of instructional requirements of the learner and development of functional learning systems in order to meet these requirements. In fact, as a consequence of studies on the development of effective learning systems some instructional design theories have emerged. Among these theories the motivational design theory points out that instructional processes are required to be configured with the strategies which increases the attention, relevance, confidence and satisfaction of the students for an instructional design which ensures the continuity of learning motivation. The studies indicate that the systems which are developed on the basis of mentioned strategies raise the attention of the student during instruction, develop a relevance to the students’ requirements, create a positive expectation for success and help having a satisfaction by reinforcing success. In this article, the empirical studies related with this subject and the suggestions for presenting more effective motivational instructional designs in distance learning are summarized.

  4. A complete low cost radon detection system

    International Nuclear Information System (INIS)

    Bayrak, A.; Barlas, E.; Emirhan, E.; Kutlu, Ç.; Ozben, C.S.

    2013-01-01

    Monitoring the 222 Rn activity through the 1200 km long Northern Anatolian fault line, for the purpose of earthquake precursory, requires large number of cost effective radon detectors. We have designed, produced and successfully tested a low cost radon detection system (a radon monitor). In the detector circuit of this monitor, First Sensor PS100-7-CER-2 windowless PIN photodiode and a custom made transempedence/shaping amplifier were used. In order to collect the naturally ionized radon progeny to the surface of the PIN photodiode, a potential of 3500 V was applied between the conductive hemi-spherical shell and the PIN photodiode. In addition to the count rate of the radon progeny, absolute pressure, humidity and temperature were logged during the measurements. A GSM modem was integrated to the system for transferring the measurements from the remote locations to the data process center. - Author-Highlights: • Low cost radon detection. • Integrated GSM modem for early warning of radon anomalies. • Radon detection in environment

  5. BENCHMARKING MACHINE LEARNING TECHNIQUES FOR SOFTWARE DEFECT DETECTION

    OpenAIRE

    Saiqa Aleem; Luiz Fernando Capretz; Faheem Ahmed

    2015-01-01

    Machine Learning approaches are good in solving problems that have less information. In most cases, the software domain problems characterize as a process of learning that depend on the various circumstances and changes accordingly. A predictive model is constructed by using machine learning approaches and classified them into defective and non-defective modules. Machine learning techniques help developers to retrieve useful information after the classification and enable them to analyse data...

  6. What Learning Systems do Intelligent Agents Need? Complementary Learning Systems Theory Updated.

    Science.gov (United States)

    Kumaran, Dharshan; Hassabis, Demis; McClelland, James L

    2016-07-01

    We update complementary learning systems (CLS) theory, which holds that intelligent agents must possess two learning systems, instantiated in mammalians in neocortex and hippocampus. The first gradually acquires structured knowledge representations while the second quickly learns the specifics of individual experiences. We broaden the role of replay of hippocampal memories in the theory, noting that replay allows goal-dependent weighting of experience statistics. We also address recent challenges to the theory and extend it by showing that recurrent activation of hippocampal traces can support some forms of generalization and that neocortical learning can be rapid for information that is consistent with known structure. Finally, we note the relevance of the theory to the design of artificial intelligent agents, highlighting connections between neuroscience and machine learning. Copyright © 2016 Elsevier Ltd. All rights reserved.

  7. Machine learning approach to detect intruders in database based on hexplet data structure

    Directory of Open Access Journals (Sweden)

    Saad M. Darwish

    2016-09-01

    Full Text Available Most of valuable information resources for any organization are stored in the database; it is a serious subject to protect this information against intruders. However, conventional security mechanisms are not designed to detect anomalous actions of database users. An intrusion detection system (IDS, delivers an extra layer of security that cannot be guaranteed by built-in security tools, is the ideal solution to defend databases from intruders. This paper suggests an anomaly detection approach that summarizes the raw transactional SQL queries into a compact data structure called hexplet, which can model normal database access behavior (abstract the user's profile and recognize impostors specifically tailored for role-based access control (RBAC database system. This hexplet lets us to preserve the correlation among SQL statements in the same transaction by exploiting the information in the transaction-log entry with the aim to improve detection accuracy specially those inside the organization and behave strange behavior. The model utilizes naive Bayes classifier (NBC as the simplest supervised learning technique for creating profiles and evaluating the legitimacy of a transaction. Experimental results show the performance of the proposed model in the term of detection rate.

  8. Vision-Based Parking-Slot Detection: A Benchmark and A Learning-Based Approach

    Directory of Open Access Journals (Sweden)

    Lin Zhang

    2018-03-01

    Full Text Available Recent years have witnessed a growing interest in developing automatic parking systems in the field of intelligent vehicles. However, how to effectively and efficiently locating parking-slots using a vision-based system is still an unresolved issue. Even more seriously, there is no publicly available labeled benchmark dataset for tuning and testing parking-slot detection algorithms. In this paper, we attempt to fill the above-mentioned research gaps to some extent and our contributions are twofold. Firstly, to facilitate the study of vision-based parking-slot detection, a large-scale parking-slot image database is established. This database comprises 8600 surround-view images collected from typical indoor and outdoor parking sites. For each image in this database, the marking-points and parking-slots are carefully labeled. Such a database can serve as a benchmark to design and validate parking-slot detection algorithms. Secondly, a learning-based parking-slot detection approach, namely P S D L , is proposed. Using P S D L , given a surround-view image, the marking-points will be detected first and then the valid parking-slots can be inferred. The efficacy and efficiency of P S D L have been corroborated on our database. It is expected that P S D L can serve as a baseline when the other researchers develop more sophisticated methods.

  9. Dynamic analysis methods for detecting anomalies in asynchronously interacting systems

    Energy Technology Data Exchange (ETDEWEB)

    Kumar, Akshat; Solis, John Hector; Matschke, Benjamin

    2014-01-01

    Detecting modifications to digital system designs, whether malicious or benign, is problematic due to the complexity of the systems being analyzed. Moreover, static analysis techniques and tools can only be used during the initial design and implementation phases to verify safety and liveness properties. It is computationally intractable to guarantee that any previously verified properties still hold after a system, or even a single component, has been produced by a third-party manufacturer. In this paper we explore new approaches for creating a robust system design by investigating highly-structured computational models that simplify verification and analysis. Our approach avoids the need to fully reconstruct the implemented system by incorporating a small verification component that dynamically detects for deviations from the design specification at run-time. The first approach encodes information extracted from the original system design algebraically into a verification component. During run-time this component randomly queries the implementation for trace information and verifies that no design-level properties have been violated. If any deviation is detected then a pre-specified fail-safe or notification behavior is triggered. Our second approach utilizes a partitioning methodology to view liveness and safety properties as a distributed decision task and the implementation as a proposed protocol that solves this task. Thus the problem of verifying safety and liveness properties is translated to that of verifying that the implementation solves the associated decision task. We develop upon results from distributed systems and algebraic topology to construct a learning mechanism for verifying safety and liveness properties from samples of run-time executions.

  10. Accent detection and social cognition: evidence of protracted learning.

    Science.gov (United States)

    Creel, Sarah C

    2018-03-01

    How and when do children become aware that speakers have different accents? While adults readily make a variety of subtle social inferences based on speakers' accents, findings from children are more mixed: while one line of research suggests that even infants may be acutely sensitive to accent unfamiliarity, other studies suggest that 5-year-olds have difficulty identifying accents as different from their own. In an attempt to resolve this paradox, the current study assesses American children's sensitivity to American vs. Dutch accents in two situations. First, in an eye-tracked sentence processing paradigm where children have previously shown sensitivity to a salient social distinction (gender) from voice cues, 3-5-year-old children showed no sensitivity to accent differences. Second, in a social decision-making task where accent sensitivity has been found in 5-year-olds, an age gradient appeared, suggesting that familiar accent preferences emerge slowly between 3 and 7 years. Counter to claims that accent is an early, salient signal of social group, results are more consistent with a protracted learning hypothesis that children need extended exposure to native-language sound patterns in order to detect that an accent deviates from their own. A video abstract of this article can be viewed at: https://www.youtube.com/watch?v=BQAgy3IFYXA. © 2017 John Wiley & Sons Ltd.

  11. Inductive learning as a fusion engine for mine detection

    Energy Technology Data Exchange (ETDEWEB)

    Kercel, S.W.; Dress, W.B.

    1997-08-01

    Semiotics is defined by some researchers as {open_quotes}the study of the appearance (visual or otherwise) meaning, and use of symbols and symbol systems.{close_quotes} Semiotic fusion of data from multiple sensory sources is a potential solution to the problem of landmine detection. This turns out to be significant, because notwithstanding the diversity of sensor technologies being used to attack the problem, there is no single effective landmine sensor technology. The only practical, general-purpose mine detector presently available is the trained dog. Most research into mine-detection technology seeds to emulate the dog`s seemingly uncanny abilities. An ideal data-fusion system would mimic animal reaction, with the brain`s perceptive power melding multiple sensory cues into an awareness of the size and location of a mine. Furthermore, the fusion process should be adaptive, with the skill at combining cues into awareness improving with experience. Electronic data-fusion systems reported in the countermine literature use conventional vector-based pattern recognition methods. Although neural nets are popular, they have never satisfactorily met the challenge. Despite years of investigation, nobody has ever found a vector space representation that reliably characterizes mine identity. This strongly suggests that the features have not been found because researchers have been looking for the wrong characteristics. It is worth considering that dogs probably do not represent data as mathematical number lists, but they almost certainly represent data via semiotic structures.

  12. High sensitivity neutron bursts detecting system

    International Nuclear Information System (INIS)

    Shyam, A.; Kaushik, T.C.; Srinivasan, M.; Kulkarni, L.V.

    1993-01-01

    Technique and instrumentation to detect multiplicity of fast neutrons, emitted in sharp bursts, has been developed. A bank of 16 BF 3 detectors, in an appropriate thermalising assembly, efficiency ∼ 16%, is used to detect neutron bursts. The output from this setup, through appropriate electronics, is divided into two paths. The first is directly connected to a computer controlled scalar. The second is connected to another similar scalar through a delay time unit (DTU). The DTU design is such that once it is triggered by a count pulse than it does not allow any counts to be recorded for a fixed dead time set at ∼ 100 μs. The difference in counts recorded directly and through DTU gives the total number of neutrons produced in bursts. This setup is being used to study lattice cracking, anomalous effects in solid deuterium systems and various reactor physics experiments. (author). 3 refs., 1 fig

  13. System and method for anomaly detection

    Science.gov (United States)

    Scherrer, Chad

    2010-06-15

    A system and method for detecting one or more anomalies in a plurality of observations is provided. In one illustrative embodiment, the observations are real-time network observations collected from a stream of network traffic. The method includes performing a discrete decomposition of the observations, and introducing derived variables to increase storage and query efficiencies. A mathematical model, such as a conditional independence model, is then generated from the formatted data. The formatted data is also used to construct frequency tables which maintain an accurate count of specific variable occurrence as indicated by the model generation process. The formatted data is then applied to the mathematical model to generate scored data. The scored data is then analyzed to detect anomalies.

  14. Collision Detection for Underwater ROV Manipulator Systems

    Directory of Open Access Journals (Sweden)

    Satja Sivčev

    2018-04-01

    Full Text Available Work-class ROVs equipped with robotic manipulators are extensively used for subsea intervention operations. Manipulators are teleoperated by human pilots relying on visual feedback from the worksite. Operating in a remote environment, with limited pilot perception and poor visibility, manipulator collisions which may cause significant damage are likely to happen. This paper presents a real-time collision detection algorithm for marine robotic manipulation. The proposed collision detection mechanism is developed, integrated into a commercial ROV manipulator control system, and successfully evaluated in simulations and experimental setup using a real industry standard underwater manipulator. The presented collision sensing solution has a potential to be a useful pilot assisting tool that can reduce the task load, operational time, and costs of subsea inspection, repair, and maintenance operations.

  15. Deep learning of contrast-coated serrated polyps for computer-aided detection in CT colonography

    Science.gov (United States)

    Näppi, Janne J.; Pickhardt, Perry; Kim, David H.; Hironaka, Toru; Yoshida, Hiroyuki

    2017-03-01

    Serrated polyps were previously believed to be benign lesions with no cancer potential. However, recent studies have revealed a novel molecular pathway where also serrated polyps can develop into colorectal cancer. CT colonography (CTC) can detect serrated polyps using the radiomic biomarker of contrast coating, but this requires expertise from the reader and current computer-aided detection (CADe) systems have not been designed to detect the contrast coating. The purpose of this study was to develop a novel CADe method that makes use of deep learning to detect serrated polyps based on their contrast-coating biomarker in CTC. In the method, volumetric shape-based features are used to detect polyp sites over soft-tissue and fecal-tagging surfaces of the colon. The detected sites are imaged using multi-angular 2D image patches. A deep convolutional neural network (DCNN) is used to review the image patches for the presence of polyps. The DCNN-based polyp-likelihood estimates are merged into an aggregate likelihood index where highest values indicate the presence of a polyp. For pilot evaluation, the proposed DCNN-CADe method was evaluated with a 10-fold cross-validation scheme using 101 colonoscopy-confirmed cases with 144 biopsy-confirmed serrated polyps from a CTC screening program, where the patients had been prepared for CTC with saline laxative and fecal tagging by barium and iodine-based diatrizoate. The average per-polyp sensitivity for serrated polyps >=6 mm in size was 93+/-7% at 0:8+/-1:8 false positives per patient on average. The detection accuracy was substantially higher that of a conventional CADe system. Our results indicate that serrated polyps can be detected automatically at high accuracy in CTC.

  16. 46 CFR 108.413 - Fusible element fire detection system.

    Science.gov (United States)

    2010-10-01

    ... 46 Shipping 4 2010-10-01 2010-10-01 false Fusible element fire detection system. 108.413 Section... UNITS DESIGN AND EQUIPMENT Fire Extinguishing Systems § 108.413 Fusible element fire detection system. (a) A fusible element fire detection system may be installed. (b) The arrangements for the system...

  17. 46 CFR 108.411 - Smoke detection system.

    Science.gov (United States)

    2010-10-01

    ... 46 Shipping 4 2010-10-01 2010-10-01 false Smoke detection system. 108.411 Section 108.411 Shipping... EQUIPMENT Fire Extinguishing Systems § 108.411 Smoke detection system. Each smoke accumulator in a smoke detection system must be located on the overhead of the compartment protected by the system in a location...

  18. Fault Detection for Nonlinear Process With Deterministic Disturbances: A Just-In-Time Learning Based Data Driven Method.

    Science.gov (United States)

    Yin, Shen; Gao, Huijun; Qiu, Jianbin; Kaynak, Okyay

    2017-11-01

    Data-driven fault detection plays an important role in industrial systems due to its applicability in case of unknown physical models. In fault detection, disturbances must be taken into account as an inherent characteristic of processes. Nevertheless, fault detection for nonlinear processes with deterministic disturbances still receive little attention, especially in data-driven field. To solve this problem, a just-in-time learning-based data-driven (JITL-DD) fault detection method for nonlinear processes with deterministic disturbances is proposed in this paper. JITL-DD employs JITL scheme for process description with local model structures to cope with processes dynamics and nonlinearity. The proposed method provides a data-driven fault detection solution for nonlinear processes with deterministic disturbances, and owns inherent online adaptation and high accuracy of fault detection. Two nonlinear systems, i.e., a numerical example and a sewage treatment process benchmark, are employed to show the effectiveness of the proposed method.

  19. Intelligent Flamefinder Detection and Alert System (IFDAS), Phase II

    Data.gov (United States)

    National Aeronautics and Space Administration — Current hydrogen flame detection systems exhibit shortcomings ranging from limited detection range, to localization inaccuracy, limited sensitivity, false alarms,...

  20. Multiple Kernel Learning for Heterogeneous Anomaly Detection: Algorithm and Aviation Safety Case Study

    Science.gov (United States)

    Das, Santanu; Srivastava, Ashok N.; Matthews, Bryan L.; Oza, Nikunj C.

    2010-01-01

    The world-wide aviation system is one of the most complex dynamical systems ever developed and is generating data at an extremely rapid rate. Most modern commercial aircraft record several hundred flight parameters including information from the guidance, navigation, and control systems, the avionics and propulsion systems, and the pilot inputs into the aircraft. These parameters may be continuous measurements or binary or categorical measurements recorded in one second intervals for the duration of the flight. Currently, most approaches to aviation safety are reactive, meaning that they are designed to react to an aviation safety incident or accident. In this paper, we discuss a novel approach based on the theory of multiple kernel learning to detect potential safety anomalies in very large data bases of discrete and continuous data from world-wide operations of commercial fleets. We pose a general anomaly detection problem which includes both discrete and continuous data streams, where we assume that the discrete streams have a causal influence on the continuous streams. We also assume that atypical sequence of events in the discrete streams can lead to off-nominal system performance. We discuss the application domain, novel algorithms, and also discuss results on real-world data sets. Our algorithm uncovers operationally significant events in high dimensional data streams in the aviation industry which are not detectable using state of the art methods

  1. Implementation of a General Real-Time Visual Anomaly Detection System Via Soft Computing

    Science.gov (United States)

    Dominguez, Jesus A.; Klinko, Steve; Ferrell, Bob; Steinrock, Todd (Technical Monitor)

    2001-01-01

    The intelligent visual system detects anomalies or defects in real time under normal lighting operating conditions. The application is basically a learning machine that integrates fuzzy logic (FL), artificial neural network (ANN), and generic algorithm (GA) schemes to process the image, run the learning process, and finally detect the anomalies or defects. The system acquires the image, performs segmentation to separate the object being tested from the background, preprocesses the image using fuzzy reasoning, performs the final segmentation using fuzzy reasoning techniques to retrieve regions with potential anomalies or defects, and finally retrieves them using a learning model built via ANN and GA techniques. FL provides a powerful framework for knowledge representation and overcomes uncertainty and vagueness typically found in image analysis. ANN provides learning capabilities, and GA leads to robust learning results. An application prototype currently runs on a regular PC under Windows NT, and preliminary work has been performed to build an embedded version with multiple image processors. The application prototype is being tested at the Kennedy Space Center (KSC), Florida, to visually detect anomalies along slide basket cables utilized by the astronauts to evacuate the NASA Shuttle launch pad in an emergency. The potential applications of this anomaly detection system in an open environment are quite wide. Another current, potentially viable application at NASA is in detecting anomalies of the NASA Space Shuttle Orbiter's radiator panels.

  2. Radon detection system, design, test and performance

    International Nuclear Information System (INIS)

    Balcazar, M.; Chavez, A.; Pina-Villalpando, G.; Navarrete, M.

    1999-01-01

    A portable radon detection system (α-Inin) has been designed and constructed for using it in adverse environmental conditions where humidity, temperature and chemical vaporous are present. The minimum integration time is in periods of 15 min during 41 days. A 12 V battery and a photovoltaic module allow the α-Inin autonomy in field measurements. Data is collected by means of a laptop computer where data processing and α-Inin programming are carried out. α-Inin performance was simultaneously tested in a controlled radon chamber, together with a commercial α-Meter

  3. Saliency Detection and Deep Learning-Based Wildfire Identification in UAV Imagery.

    Science.gov (United States)

    Zhao, Yi; Ma, Jiale; Li, Xiaohui; Zhang, Jie

    2018-02-27

    An unmanned aerial vehicle (UAV) equipped with global positioning systems (GPS) can provide direct georeferenced imagery, mapping an area with high resolution. So far, the major difficulty in wildfire image classification is the lack of unified identification marks, the fire features of color, shape, texture (smoke, flame, or both) and background can vary significantly from one scene to another. Deep learning (e.g., DCNN for Deep Convolutional Neural Network) is very effective in high-level feature learning, however, a substantial amount of training images dataset is obligatory in optimizing its weights value and coefficients. In this work, we proposed a new saliency detection algorithm for fast location and segmentation of core fire area in aerial images. As the proposed method can effectively avoid feature loss caused by direct resizing; it is used in data augmentation and formation of a standard fire image dataset 'UAV_Fire'. A 15-layered self-learning DCNN architecture named 'Fire_Net' is then presented as a self-learning fire feature exactor and classifier. We evaluated different architectures and several key parameters (drop out ratio, batch size, etc.) of the DCNN model regarding its validation accuracy. The proposed architecture outperformed previous methods by achieving an overall accuracy of 98%. Furthermore, 'Fire_Net' guarantied an average processing speed of 41.5 ms per image for real-time wildfire inspection. To demonstrate its practical utility, Fire_Net is tested on 40 sampled images in wildfire news reports and all of them have been accurately identified.

  4. Ionizing radiation detection meter systems usability evaluation. Technical memorandum

    International Nuclear Information System (INIS)

    Severs, Y.; Hughes, R.; Tack, D.; Bossi, L.L.M.

    2002-12-01

    In support of the Canadian Forces Nuclear Detection, Identification and Dosimetry project a usability trial of three Ionizing Radiation Detection Meter Systems (IRDMS) was conducted 18-22 June 2001 at Connaught Ranges. The systems evaluated: System 1: ANPDR 77 Operational System with PDR 77 Alpha/Beta probes and GPS (Global Positioning System) Training System; System 2: ADM 300C Operational System with ABP 100 Alpha/Beta probes and GPS Training System, and System 3: RAM R200 Operational System with PA 100M and RG12 Alpha/Beta probes and GPS Training System. Twelve military personnel (Private to Captain) determined compliance with the bid evaluation criteria. The counter-balanced study, based on a Latin square design, was divided into two assessment phases, an operational phase and a training system phase, which rotated three groups (of four participants each) concurrently through each IRDMS. Groups were trained in set-up, maintenance and operation of each bid IRDMS prior to evaluation. The operational evaluation consisted of a test/maintenance stand, an indoor/outdoor operational stand, and an indoor,/outdoor training system stand. Participants rated the utility and usability of each system, by measuring case of use/maintenance under a range of relevant operational and clothing conditions, compatibility with soldier tasks, compatibility with clothing, compatibility with equipment and operational environments, and case of learning (including an assessment of initial and refresher training implications). A 7-point Likert-like acceptability rating scale, which covered a range from completely unacceptable (1) to completely acceptable (7), was used to evaluate the measures. The combined weighted mean scores were 584.3 for System 1 (70 o /b), 682.0 for System 2 (81 %) and 756.9 for System 3 (90%) out of a total possible score of 840. Participants reported mean scores for System 1 between barely unacceptable and borderline for both indoor and outdoor operations. System 2

  5. Autonomous system for pathogen detection and identification

    International Nuclear Information System (INIS)

    Belgrader, P.; Benett, W.; Langlois, R.; Long, G.; Mariella, R.; Milanovich, F.; Miles, R.; Nelson, W.; Venkateswaran, K.

    1998-01-01

    This purpose of this project is to build a prototype instrument that will, running unattended, detect, identify, and quantify BW agents. In order to accomplish this, we have chosen to start with the world s leading, proven, assays for pathogens: surface-molecular recognition assays, such as antibody-based assays, implemented on a high-performance, identification (ID)-capable flow cytometer, and the polymerase chain reaction (PCR) for nucleic-acid based assays. With these assays, we must integrate the capability to: l collect samples from aerosols, water, or surfaces; l perform sample preparation prior to the assays; l incubate the prepared samples, if necessary, for a period of time; l transport the prepared, incubated samples to the assays; l perform the assays; l interpret and report the results of the assays. Issues such as reliability, sensitivity and accuracy, quantity of consumables, maintenance schedule, etc. must be addressed satisfactorily to the end user. The highest possible sensitivity and specificity of the assay must be combined with no false alarms. Today, we have assays that can, in under 30 minutes, detect and identify stimulants for BW agents at concentrations of a few hundred colony-forming units per ml of solution. If the bio-aerosol sampler of this system collects 1000 Ymin and concentrates the respirable particles into 1 ml of solution with 70% processing efficiency over a period of 5 minutes, then this translates to a detection/ID capability of under 0.1 agent-containing particle/liter of air

  6. Mass detection in digital breast tomosynthesis data using convolutional neural networks and multiple instance learning.

    Science.gov (United States)

    Yousefi, Mina; Krzyżak, Adam; Suen, Ching Y

    2018-05-01

    Digital breast tomosynthesis (DBT) was developed in the field of breast cancer screening as a new tomographic technique to minimize the limitations of conventional digital mammography breast screening methods. A computer-aided detection (CAD) framework for mass detection in DBT has been developed and is described in this paper. The proposed framework operates on a set of two-dimensional (2D) slices. With plane-to-plane analysis on corresponding 2D slices from each DBT, it automatically learns complex patterns of 2D slices through a deep convolutional neural network (DCNN). It then applies multiple instance learning (MIL) with a randomized trees approach to classify DBT images based on extracted information from 2D slices. This CAD framework was developed and evaluated using 5040 2D image slices derived from 87 DBT volumes. The empirical results demonstrate that this proposed CAD framework achieves much better performance than CAD systems that use hand-crafted features and deep cardinality-restricted Bolzmann machines to detect masses in DBTs. Copyright © 2018 Elsevier Ltd. All rights reserved.

  7. A polarization system for persistent chemical detection

    Science.gov (United States)

    Craven-Jones, Julia; Appelhans, Leah; Couphos, Eric; Embree, Todd; Finnegan, Patrick; Goldstein, Dennis; Karelitz, David; LaCasse, Charles; Luk, Ting S.; Mahamat, Adoum; Massey, Lee; Tanbakuchi, Anthony; Washburn, Cody; Vigil, Steven

    2015-09-01

    We report on the development of a prototype polarization tag based system for detecting chemical vapors. The system primarily consists of two components, a chemically sensitive tag that experiences a change in its optical polarization properties when exposed to a specific chemical of interest, and an optical imaging polarimeter that is used to measure the polarization properties of the tags. Although the system concept could be extended to other chemicals, for the initial system prototype presented here the tags were developed to be sensitive to hydrogen fluoride (HF) vapors. HF is used in many industrial processes but is highly toxic and thus monitoring for its presence and concentration is often of interest for personnel and environmental safety. The tags are periodic multilayer structures that are produced using standard photolithographic processes. The polarimetric imager has been designed to measure the degree of linear polarization reflected from the tags in the short wave infrared. By monitoring the change in the reflected polarization signature from the tags, the polarimeter can be used to determine if the tag was exposed to HF gas. In this paper, a review of the system development effort and preliminary test results are presented and discussed, as well as our plan for future work.

  8. Acoustic leak detection in piping systems, 4

    International Nuclear Information System (INIS)

    Kitajima, Akira; Naohara, Nobuyuki; Aihara, Akihiko

    1983-01-01

    To monitor a high-pressure piping of nuclear power plants, a possibility of acoustic leak detection method has been experimentally studied in practical field tests and laboratory tests. Characteristics of background noise in field test and the results of experiment are summarized as follows: (1) The level of background noise in primary loop (PWR) was almost constant under actual plant operation. But it is possible that it rises at the condition of the pressure in primary loop. (2) Based on many experience of laboratory tests and practical field tests. The leak monitoring system for practical field was designed and developed. To improve the reliability, a judgment of leak on this system is used three factors of noise level, duration time of phenomena and frequency spectrum of noise signal emitted from the leak point. (author)

  9. Reactor coolant pressure boundary leakage detection system

    International Nuclear Information System (INIS)

    Dissing, E.; Svansson, L.

    1980-01-01

    This study deals with a system for monitoring the leakage of reactor coolant. This system is based primarily on the detection of the 13 N content in the containment atmosphere. 13 N is produced from the oxygen of the reactor water via the recoil proton nuclear process Hl+016/yields/ 13 N+ 4 He. The generation is therefore independent of fuel element leakage and of the corrosion product content in the water. It is solely related to the neutron flux level in the reactor core. Typical figures for the equilibrium 13 N concentration in the containment atmosphere following a 4 kg/minute coolant leakage are 5 kBq m/sup -3/ and 7 kBq m/sup -3/ for BWR and PWR respectively. These levels are readily measured with a 10 liter Ge(Li) flow detector assembly operated at elevated pressure. 8 refs

  10. Reactor coolant pressure boundary leakage detection system

    International Nuclear Information System (INIS)

    Dissing, E.; Svansson, L.

    1980-01-01

    This study deals with a system for monitoring the leakage of reactor coolant. This system is based primarily on the detection of the N13 content in the containment atmosphere. N13 is produced from the oxygen of the reactor water via the recoil proton nuclear process Hl+016/yields/Nl3+He4. The generation is therefore independent of fuel element leakage and of the corrosion product content in the water. It is solely related to the neutron flux level in the reactor core. Typical figures for the equilibrium N13 concentration in the containment atmosphere following a 4 kg/minute coolant leakage are 5 kBq m/sup -3/ and 7 kBq m/sup -3/ for BWR and PWR respectively. These levels are readily measured with a 10 liter Ge(Li) flow detector assembly operated at elevated pressure. 8 refs

  11. Reactor coolant pressure boundary leakage detection system

    International Nuclear Information System (INIS)

    Dissing, E.; Svansson, L.

    1979-08-01

    The present paper deals with a system for monitoring the leakage of reactor coolant. This system is based primarily on the detection of the N13 content in the containment atmosphere. N13 is produced from the oxygen of the reactor water via the recoil proton nuclear process H1+016 → N13+He4. The generation is therefore independent of fuel element leakage and of the corrosion product content in the water. It is solely related to the neutron flux level in the reactor core. Typical figures for the equilibrium N13 concentration in the containment atmosphere following a 4 kg/minute coolant leakage are 5 kBq m -3 and 7 kBq m -3 for BWR and PWR respectively. These levels are readily measured with a 10 liter Ge (Li) flow detector assembly operated at elevated pressure. (Auth.)

  12. The Environics Mobile CBRN Detection System

    International Nuclear Information System (INIS)

    Jaakkola, T.

    2007-01-01

    Environics Oy has developed a novel monitoring system for detection of Chemical, Biological, Radiological and Nuclear compounds. The system is portable and rapidly installed into a monitoring location. It will allow real time monitoring and alarming of: Chemical Warfare agents (Nerve, Blister, Blood), Toxic industrial chemicals (General toxic-alarm), Biological warfare agents (Bacteria, viruses, toxins), Radiological agents such as alpha and gamma radiation. Monitoring Station makes continuously measurements. Sensor data is processed and stored to local database by the Master Module (MM) that is located within the station. The MM sends the data to the Control Centers by using communication network. The Control Center receives and logs the data and shows it in real time on a map interface. The status of each sensor and detector can be seen in real time. (author)

  13. Simulation Model of Mobile Detection Systems

    International Nuclear Information System (INIS)

    Edmunds, T.; Faissol, D.; Yao, Y.

    2009-01-01

    In this paper, we consider a mobile source that we attempt to detect with man-portable, vehicle-mounted or boat-mounted radiation detectors. The source is assumed to transit an area populated with these mobile detectors, and the objective is to detect the source before it reaches a perimeter. We describe a simulation model developed to estimate the probability that one of the mobile detectors will come in to close proximity of the moving source and detect it. We illustrate with a maritime simulation example. Our simulation takes place in a 10 km by 5 km rectangular bay patrolled by boats equipped with 2-inch x 4-inch x 16-inch NaI detectors. Boats to be inspected enter the bay and randomly proceed to one of seven harbors on the shore. A source-bearing boat enters the mouth of the bay and proceeds to a pier on the opposite side. We wish to determine the probability that the source is detected and its range from target when detected. Patrol boats select the nearest in-bound boat for inspection and initiate an intercept course. Once within an operational range for the detection system, a detection algorithm is started. If the patrol boat confirms the source is not present, it selects the next nearest boat for inspection. Each run of the simulation ends either when a patrol successfully detects a source or when the source reaches its target. Several statistical detection algorithms have been implemented in the simulation model. First, a simple k-sigma algorithm, which alarms with the counts in a time window exceeds the mean background plus k times the standard deviation of background, is available to the user. The time window used is optimized with respect to the signal-to-background ratio for that range and relative speed. Second, a sequential probability ratio test [Wald 1947] is available, and configured in this simulation with a target false positive probability of 0.001 and false negative probability of 0.1. This test is utilized when the mobile detector maintains

  14. Simulation Model of Mobile Detection Systems

    Energy Technology Data Exchange (ETDEWEB)

    Edmunds, T; Faissol, D; Yao, Y

    2009-01-27

    In this paper, we consider a mobile source that we attempt to detect with man-portable, vehicle-mounted or boat-mounted radiation detectors. The source is assumed to transit an area populated with these mobile detectors, and the objective is to detect the source before it reaches a perimeter. We describe a simulation model developed to estimate the probability that one of the mobile detectors will come in to close proximity of the moving source and detect it. We illustrate with a maritime simulation example. Our simulation takes place in a 10 km by 5 km rectangular bay patrolled by boats equipped with 2-inch x 4-inch x 16-inch NaI detectors. Boats to be inspected enter the bay and randomly proceed to one of seven harbors on the shore. A source-bearing boat enters the mouth of the bay and proceeds to a pier on the opposite side. We wish to determine the probability that the source is detected and its range from target when detected. Patrol boats select the nearest in-bound boat for inspection and initiate an intercept course. Once within an operational range for the detection system, a detection algorithm is started. If the patrol boat confirms the source is not present, it selects the next nearest boat for inspection. Each run of the simulation ends either when a patrol successfully detects a source or when the source reaches its target. Several statistical detection algorithms have been implemented in the simulation model. First, a simple k-sigma algorithm, which alarms with the counts in a time window exceeds the mean background plus k times the standard deviation of background, is available to the user. The time window used is optimized with respect to the signal-to-background ratio for that range and relative speed. Second, a sequential probability ratio test [Wald 1947] is available, and configured in this simulation with a target false positive probability of 0.001 and false negative probability of 0.1. This test is utilized when the mobile detector maintains

  15. Mass detection in digital breast tomosynthesis: Deep convolutional neural network with transfer learning from mammography.

    Science.gov (United States)

    Samala, Ravi K; Chan, Heang-Ping; Hadjiiski, Lubomir; Helvie, Mark A; Wei, Jun; Cha, Kenny

    2016-12-01

    Develop a computer-aided detection (CAD) system for masses in digital breast tomosynthesis (DBT) volume using a deep convolutional neural network (DCNN) with transfer learning from mammograms. A data set containing 2282 digitized film and digital mammograms and 324 DBT volumes were collected with IRB approval. The mass of interest on the images was marked by an experienced breast radiologist as reference standard. The data set was partitioned into a training set (2282 mammograms with 2461 masses and 230 DBT views with 228 masses) and an independent test set (94 DBT views with 89 masses). For DCNN training, the region of interest (ROI) containing the mass (true positive) was extracted from each image. False positive (FP) ROIs were identified at prescreening by their previously developed CAD systems. After data augmentation, a total of 45 072 mammographic ROIs and 37 450 DBT ROIs were obtained. Data normalization and reduction of non-uniformity in the ROIs across heterogeneous data was achieved using a background correction method applied to each ROI. A DCNN with four convolutional layers and three fully connected (FC) layers was first trained on the mammography data. Jittering and dropout techniques were used to reduce overfitting. After training with the mammographic ROIs, all weights in the first three convolutional layers were frozen, and only the last convolution layer and the FC layers were randomly initialized again and trained using the DBT training ROIs. The authors compared the performances of two CAD systems for mass detection in DBT: one used the DCNN-based approach and the other used their previously developed feature-based approach for FP reduction. The prescreening stage was identical in both systems, passing the same set of mass candidates to the FP reduction stage. For the feature-based CAD system, 3D clustering and active contour method was used for segmentation; morphological, gray level, and texture features were extracted and merged with a

  16. Extensible Adaptive System for STEM Learning

    Science.gov (United States)

    2013-07-16

    Copyright 2013 Raytheon BBN Technologies Corp. All Rights Reserved ONR STEM Grand Challenge Extensible Adaptive System for STEM Learning ...Contract # N00014-12-C-0535 Raytheon BBN Technologies Corp. (BBN) Reference # 14217 In partial fulfillment of contract deliverable item # A001...Quarterly Progress Report #2 April 7, 2013 –July 6, 2013 Submitted July 16, 2013 BBN Technical POC: John Makhoul Raytheon BBN Technologies

  17. 46 CFR 108.404 - Selection of fire detection system.

    Science.gov (United States)

    2010-10-01

    ... 46 Shipping 4 2010-10-01 2010-10-01 false Selection of fire detection system. 108.404 Section 108... DESIGN AND EQUIPMENT Fire Extinguishing Systems § 108.404 Selection of fire detection system. (a) If a... space. (b) The fire detection system must be designed to minimize false alarms. ...

  18. Broad Learning System: An Effective and Efficient Incremental Learning System Without the Need for Deep Architecture.

    Science.gov (United States)

    Chen, C L Philip; Liu, Zhulin

    2018-01-01

    Broad Learning System (BLS) that aims to offer an alternative way of learning in deep structure is proposed in this paper. Deep structure and learning suffer from a time-consuming training process because of a large number of connecting parameters in filters and layers. Moreover, it encounters a complete retraining process if the structure is not sufficient to model the system. The BLS is established in the form of a flat network, where the original inputs are transferred and placed as "mapped features" in feature nodes and the structure is expanded in wide sense in the "enhancement nodes." The incremental learning algorithms are developed for fast remodeling in broad expansion without a retraining process if the network deems to be expanded. Two incremental learning algorithms are given for both the increment of the feature nodes (or filters in deep structure) and the increment of the enhancement nodes. The designed model and algorithms are very versatile for selecting a model rapidly. In addition, another incremental learning is developed for a system that has been modeled encounters a new incoming input. Specifically, the system can be remodeled in an incremental way without the entire retraining from the beginning. Satisfactory result for model reduction using singular value decomposition is conducted to simplify the final structure. Compared with existing deep neural networks, experimental results on the Modified National Institute of Standards and Technology database and NYU NORB object recognition dataset benchmark data demonstrate the effectiveness of the proposed BLS.

  19. Learning a Transferable Change Rule from a Recurrent Neural Network for Land Cover Change Detection

    Directory of Open Access Journals (Sweden)

    Haobo Lyu

    2016-06-01

    Full Text Available When exploited in remote sensing analysis, a reliable change rule with transfer ability can detect changes accurately and be applied widely. However, in practice, the complexity of land cover changes makes it difficult to use only one change rule or change feature learned from a given multi-temporal dataset to detect any other new target images without applying other learning processes. In this study, we consider the design of an efficient change rule having transferability to detect both binary and multi-class changes. The proposed method relies on an improved Long Short-Term Memory (LSTM model to acquire and record the change information of long-term sequence remote sensing data. In particular, a core memory cell is utilized to learn the change rule from the information concerning binary changes or multi-class changes. Three gates are utilized to control the input, output and update of the LSTM model for optimization. In addition, the learned rule can be applied to detect changes and transfer the change rule from one learned image to another new target multi-temporal image. In this study, binary experiments, transfer experiments and multi-class change experiments are exploited to demonstrate the superiority of our method. Three contributions of this work can be summarized as follows: (1 the proposed method can learn an effective change rule to provide reliable change information for multi-temporal images; (2 the learned change rule has good transferability for detecting changes in new target images without any extra learning process, and the new target images should have a multi-spectral distribution similar to that of the training images; and (3 to the authors’ best knowledge, this is the first time that deep learning in recurrent neural networks is exploited for change detection. In addition, under the framework of the proposed method, changes can be detected under both binary detection and multi-class change detection.

  20. Supporting Teachers in Identifying Students' Learning Styles in Learning Management Systems: An Automatic Student Modelling Approach

    Science.gov (United States)

    Graf, Sabine; Kinshuk; Liu, Tzu-Chien

    2009-01-01

    In learning management systems (LMSs), teachers have more difficulties to notice and know how individual students behave and learn in a course, compared to face-to-face education. Enabling teachers to know their students' learning styles and making students aware of their own learning styles increases teachers' and students' understanding about…

  1. Lessons learned on digital systems safety

    International Nuclear Information System (INIS)

    Sivertsen, Terje

    2005-06-01

    A decade ago, in 1994, lessons learned from Halden research activities on digital systems safety were summarized in the reports HWR-374 and HWR-375, under the title 'A Lessons Learned Report on Software Dependability'. The reports reviewed all activities made at the Halden Project in this field since 1977. As such, the reports provide a wealth of information on Halden research. At the same time, the lessons learned from the different activities are made more accessible to the reader by being summarized in terms of results, conclusions and recommendations. The present report provides a new lessons learned report, covering the Halden Project research activities in this area from 1994 to medio 2005. As before, the emphasis is on the results, conclusions and recommendations made from these activities, in particular how they can be utilized by different types of organisations, such as licensing authorities, safety assessors, power companies, and software developers. The contents of the report have been edited on the basis of input from a large number of Halden work reports, involving many different authors. Brief summaries of these reports are included in the last part of the report. (Author)

  2. Smart sensor systems for outdoor intrusion detection

    International Nuclear Information System (INIS)

    Lynn, J.K.

    1988-01-01

    A major improvement in outdoor perimeter security system probability of detection (PD) and reduction in false alarm rate (FAR) and nuisance alarm rate (NAR) may be obtained by analyzing the indications immediately preceding an event which might be interpreted as an intrusion. Existing systems go into alarm after crossing a threshold. Very slow changes, which accumulate until the threshold is reached, may be assessed falsely as an intrusion. A hierarchial program has begun at Stellar to develop a modular, expandable Smart Sensor system which may be interfaced to most types of sensor and alarm reporting systems. A major upgrade to the SSI Test Site is in progress so that intrusions may be simulated in a controlled and repeatable manner. A test platform is being constructed which will operate in conduction with a mobile instrumentation center with CCTVB, lighting control, weather and data monitoring and remote control of the test platform and intrusion simulators. Additional testing was contracted with an independent test facility to assess the effects of severe winter weather conditions

  3. An Architecture for Online Laboratory E-Learning System

    Science.gov (United States)

    Duan, Bing; Hosseini, Habib Mir M.; Ling, Keck Voon; Gay, Robert Kheng Leng

    2006-01-01

    Internet-based learning systems, or e-learning, are widely available in institutes, universities, and industrial companies, hosting regular or continuous education programs. The dream of teaching and learning from anywhere and at anytime becomes a reality due to the construction of e-learning infrastructure. Traditional teaching materials and…

  4. Systemic Assessment as a new tool for assessing students learning ...

    African Journals Online (AJOL)

    Systemic Assessment [SA] has been shown to be highly effective new tool in raising the level of students academic achievements, improve their ability to learn by enhancing the process of teaching and learning, and converts students from surface to deep learning. It also allow teacher to monitor students learning ...

  5. Detecting anomalous nuclear materials accounting transactions: Applying machine learning to plutonium processing facilities

    International Nuclear Information System (INIS)

    Vaccaro, H.S.

    1989-01-01

    Nuclear materials accountancy is the only safeguards measure that provides direct evidence of the status of nuclear materials. Of the six categories that gives rise to inventory differences, the technical capability is now in place to implement the technical innovations necessary to reduce the human error categories. There are really three main approaches to detecting anomalies in materials control and accountability (MC ampersand A) data: (1) Statistical: numeric methods such as the Page's Test, CUSUM, CUMUF, SITMUF, etc., can detect anomalies in metric (numeric) data. (2) Expert systems: Human expert's rules can be encoded into software systems such as ART, KEE, or Prolog. (3) Machine learning: Training data, such as historical MC ampersand A records, can be fed to a classifier program or neutral net or other machine learning algorithm. The Wisdom ampersand Sense (W ampersand S) software is a combination of approaches 2 and 3. The W ampersand S program includes full features for adding administrative rules and expert judgment rules to the rule base. if desired, the software can enforce consistency among all rules in the rule base

  6. Learning from unbalanced data: a cascade-based approach for detecting clustered microcalcifications.

    Science.gov (United States)

    Bria, A; Karssemeijer, N; Tortorella, F

    2014-02-01

    Finding abnormalities in diagnostic images is a difficult task even for expert radiologists because the normal tissue locations largely outnumber those with suspicious signs which may thus be missed or incorrectly interpreted. For the same reason the design of a Computer-Aided Detection (CADe) system is very complex because the large predominance of normal samples in the training data may hamper the ability of the classifier to recognize the abnormalities on the images. In this paper we present a novel approach for computer-aided detection which faces the class imbalance with a cascade of boosting classifiers where each node is trained by a learning algorithm based on ranking instead of classification error. Such approach is used to design a system (CasCADe) for the automated detection of clustered microcalcifications (μCs), which is a severely unbalanced classification problem because of the vast majority of image locations where no μC is present. The proposed approach was evaluated with a dataset of 1599 full-field digital mammograms from 560 cases and compared favorably with the Hologic R2CAD ImageChecker, one of the most widespread commercial CADe systems. In particular, at the same lesion sensitivity of R2CAD (90%) on biopsy proven malignant cases, CasCADe and R2CAD detected 0.13 and 0.21 false positives per image (FPpi), respectively (p-value=0.09), whereas at the same FPpi of R2CAD (0.21), CasCADe and R2CAD detected 93% and 90% of true lesions respectively (p-value=0.11) thus showing that CasCADe can compete with high-end CADe commercial systems. Copyright © 2013 Elsevier B.V. All rights reserved.

  7. INFORMATION-ANALYTICAL LEARNING MANAGEMENT SYSTEM UNIVERSITIES

    Directory of Open Access Journals (Sweden)

    Yu. Tryus

    2016-12-01

    Full Text Available We consider conceptual approaches to creation of information systems, learning management school, which uses modern methods of decision-making and simulational modeling, web-technologies. The main criteria for the selection of development tools of the system are: openness, free of charge, easy to use and independence from system software and hardware. The chosen technology and the system itself satisfies such requirements as: focus on national and international standards in the field of higher education, adherence to service-oriented architecture, ensuring stable operation with a large number of users, support for a clear division of user rights to obtain and change information resources, software modularity the final product and its ability to integrate into the corporate information system of the university

  8. LAN attack detection using Discrete Event Systems.

    Science.gov (United States)

    Hubballi, Neminath; Biswas, Santosh; Roopa, S; Ratti, Ritesh; Nandi, Sukumar

    2011-01-01

    Address Resolution Protocol (ARP) is used for determining the link layer or Medium Access Control (MAC) address of a network host, given its Internet Layer (IP) or Network Layer address. ARP is a stateless protocol and any IP-MAC pairing sent by a host is accepted without verification. This weakness in the ARP may be exploited by malicious hosts in a Local Area Network (LAN) by spoofing IP-MAC pairs. Several schemes have been proposed in the literature to circumvent these attacks; however, these techniques either make IP-MAC pairing static, modify the existing ARP, patch operating systems of all the hosts etc. In this paper we propose a Discrete Event System (DES) approach for Intrusion Detection System (IDS) for LAN specific attacks which do not require any extra constraint like static IP-MAC, changing the ARP etc. A DES model is built for the LAN under both a normal and compromised (i.e., spoofed request/response) situation based on the sequences of ARP related packets. Sequences of ARP events in normal and spoofed scenarios are similar thereby rendering the same DES models for both the cases. To create different ARP events under normal and spoofed conditions the proposed technique uses active ARP probing. However, this probing adds extra ARP traffic in the LAN. Following that a DES detector is built to determine from observed ARP related events, whether the LAN is operating under a normal or compromised situation. The scheme also minimizes extra ARP traffic by probing the source IP-MAC pair of only those ARP packets which are yet to be determined as genuine/spoofed by the detector. Also, spoofed IP-MAC pairs determined by the detector are stored in tables to detect other LAN attacks triggered by spoofing namely, man-in-the-middle (MiTM), denial of service etc. The scheme is successfully validated in a test bed. Copyright © 2010 ISA. Published by Elsevier Ltd. All rights reserved.

  9. Learning Faults Detection by AIS Techniques in CSCL Environments

    Science.gov (United States)

    Zedadra, Amina; Lafifi, Yacine

    2015-01-01

    By the increase of e-learning platforms, huge data sets are made from different kinds of the collected traces. These traces differ from one learner to another according to their characteristics (learning styles, preferences, performed actions, etc.). Learners' traces are very heterogeneous and voluminous, so their treatments and exploitations are…

  10. A deep-learning-based emergency alert system

    Directory of Open Access Journals (Sweden)

    Byungseok Kang

    2016-06-01

    Full Text Available Emergency alert systems serve as a critical link in the chain of crisis communication, and they are essential to minimize loss during emergencies. Acts of terrorism and violence, chemical spills, amber alerts, nuclear facility problems, weather-related emergencies, flu pandemics, and other emergencies all require those responsible such as government officials, building managers, and university administrators to be able to quickly and reliably distribute emergency information to the public. This paper presents our design of a deep-learning-based emergency warning system. The proposed system is considered suitable for application in existing infrastructure such as closed-circuit television and other monitoring devices. The experimental results show that in most cases, our system immediately detects emergencies such as car accidents and natural disasters.

  11. Learning towards system innovation: Evaluating a systemic instrument

    NARCIS (Netherlands)

    Mierlo, B. van; Leeuwis, C.; Smits, R.; Klein Woolthuis, R.J.A.

    2010-01-01

    In this paper we develop an analytical framework for studying learning processes in the context of efforts to bring about system innovation by building new networks of actors who are willing to work on a change towards sustainable development. We then use it to evaluate two specific intervention

  12. A Novel Approach for Enhancing Lifelong Learning Systems by Using Hybrid Recommender System

    Science.gov (United States)

    Kardan, Ahmad A.; Speily, Omid R. B.; Modaberi, Somayyeh

    2011-01-01

    The majority of current web-based learning systems are closed learning environments where courses and learning materials are fixed, and the only dynamic aspect is the organization of the material that can be adapted to allow a relatively individualized learning environment. In this paper, we propose an evolving web-based learning system which can…

  13. The design method and research status of vehicle detection system based on geomagnetic detection principle

    Science.gov (United States)

    Lin, Y. H.; Bai, R.; Qian, Z. H.

    2018-03-01

    Vehicle detection systems are applied to obtain real-time information of vehicles, realize traffic control and reduce traffic pressure. This paper reviews geomagnetic sensors as well as the research status of the vehicle detection system. Presented in the paper are also our work on the vehicle detection system, including detection algorithms and experimental results. It is found that the GMR based vehicle detection system has a detection accuracy up to 98% with a high potential for application in the road traffic control area.

  14. Model-based fault detection algorithm for photovoltaic system monitoring

    KAUST Repository

    Harrou, Fouzi; Sun, Ying; Saidi, Ahmed

    2018-01-01

    Reliable detection of faults in PV systems plays an important role in improving their reliability, productivity, and safety. This paper addresses the detection of faults in the direct current (DC) side of photovoltaic (PV) systems using a

  15. Wireless Falling Detection System Based on Community.

    Science.gov (United States)

    Xia, Yun; Wu, Yanqi; Zhang, Bobo; Li, Zhiyang; He, Nongyue; Li, Song

    2015-06-01

    The elderly are more likely to suffer the aches or pains from the accidental falls, and both the physiology and psychology of patients would subject to a long-term disturbance, especially when the emergency treatment was not given timely and properly. Although many methods and devices have been developed creatively and shown their efficiency in experiments, few of them are suitable for commercial applications routinely. Here, we design a wearable falling detector as a mobile terminal, and utilize the wireless technology to transfer and monitor the activity data of the host in a relatively small community. With the help of the accelerometer sensor and the Google Mapping service, information of the location and the activity data will be send to the remote server for the downstream processing. The experimental result has shown that SA (Sum-vector of all axes) value of 2.5 g is the threshold value to distinguish the falling from other activities. A three-stage detection algorithm was adopted to increase the accuracy of the real alarm, and the accuracy rate of our system was more than 95%. With the further improvement, the falling detecting device which is low-cost, accurate and user-friendly would become more and more common in everyday life.

  16. Optical fiber-applied radiation detection system

    International Nuclear Information System (INIS)

    Nishiura, Ryuichi; Uranaka, Yasuo; Izumi, Nobuyuki

    2001-01-01

    A technique to measure radiation by using plastic scintillation fibers doped radiation fluorescent (scintillator) to plastic optical fiber for a radiation sensor, was developed. The technique contains some superiority such as high flexibility due to using fibers, relatively easy large area due to detecting portion of whole of fibers, and no electromagnetic noise effect due to optical radiation detection and signal transmission. Measurable to wide range of and continuous radiation distribution along optical fiber cable at a testing portion using scintillation fiber and flight time method, the optical fiber-applied radiation sensing system can effectively monitor space radiation dose or apparatus operation condition monitoring. And, a portable type scintillation optical fiber body surface pollution monitor can measure pollution concentration of radioactive materials attached onto body surface by arranging scintillation fiber processed to a plate with small size and flexibility around a man to be tested. Here were described on outline and fundamental properties of various application products using these plastic scintillation fiber. (G.K.)

  17. Automatic system for detecting pornographic images

    Science.gov (United States)

    Ho, Kevin I. C.; Chen, Tung-Shou; Ho, Jun-Der

    2002-09-01

    Due to the dramatic growth of network and multimedia technology, people can more easily get variant information by using Internet. Unfortunately, it also makes the diffusion of illegal and harmful content much easier. So, it becomes an important topic for the Internet society to protect and safeguard Internet users from these content that may be encountered while surfing on the Net, especially children. Among these content, porno graphs cause more serious harm. Therefore, in this study, we propose an automatic system to detect still colour porno graphs. Starting from this result, we plan to develop an automatic system to search porno graphs or to filter porno graphs. Almost all the porno graphs possess one common characteristic that is the ratio of the size of skin region and non-skin region is high. Based on this characteristic, our system first converts the colour space from RGB colour space to HSV colour space so as to segment all the possible skin-colour regions from scene background. We also apply the texture analysis on the selected skin-colour regions to separate the skin regions from non-skin regions. Then, we try to group the adjacent pixels located in skin regions. If the ratio is over a given threshold, we can tell if the given image is a possible porno graph. Based on our experiment, less than 10% of non-porno graphs are classified as pornography, and over 80% of the most harmful porno graphs are classified correctly.

  18. Decision algorithms in fire detection systems

    Directory of Open Access Journals (Sweden)

    Ristić Jovan D.

    2011-01-01

    Full Text Available Analogue (and addressable fire detection systems enables a new quality in improving sensitivity to real fires and reducing susceptibility to nuisance alarm sources. Different decision algorithms types were developed with intention to improve sensitivity and reduce false alarm occurrence. At the beginning, it was free alarm level adjustment based on preset level. Majority of multi-criteria decision work was based on multi-sensor (multi-signature decision algorithms - using different type of sensors on the same location or, rather, using different aspects (level and rise of one sensor measured value. Our idea is to improve sensitivity and reduce false alarm occurrence by forming groups of sensors that work in similar conditions (same world side in the building, same or similar technology or working time. Original multi-criteria decision algorithms based on level, rise and difference of level and rise from group average are discussed in this paper.

  19. Distributed gas detection system and method

    Science.gov (United States)

    Challener, William Albert; Palit, Sabarni; Karp, Jason Harris; Kasten, Ansas Matthias; Choudhury, Niloy

    2017-11-21

    A distributed gas detection system includes one or more hollow core fibers disposed in different locations, one or more solid core fibers optically coupled with the one or more hollow core fibers and configured to receive light of one or more wavelengths from a light source, and an interrogator device configured to receive at least some of the light propagating through the one or more solid core fibers and the one or more hollow core fibers. The interrogator device is configured to identify a location of a presence of a gas-of-interest by examining absorption of at least one of the wavelengths of the light at least one of the hollow core fibers.

  20. Detection system of sodium oxide vapors

    International Nuclear Information System (INIS)

    Hundal, Rolv.

    1976-01-01

    The description is given of a sodium oxide vapor detection system which comprises a containment, a light source located to send a light beam into and through this containment and a photodetector located to intercept the light from the source after it has covered a given path through the containment. In response to the intensity of the incident light, the photodetector produces an output signal representative of it. The feature of this device is a first polarizer located near the light source, along the path of the light coming from it and designed to polarize the light projected through the containment in a given plane, and a second polarizer located near the photodetector, along the path of the polarized light and designed virtually to prevent all the light rays whose orientation differs from the given polarization plane from reaching the photodetector [fr

  1. Liquid chromatography detection unit, system, and method

    Science.gov (United States)

    Derenzo, Stephen E.; Moses, William W.

    2015-10-27

    An embodiment of a liquid chromatography detection unit includes a fluid channel and a radiation detector. The radiation detector is operable to image a distribution of a radiolabeled compound as the distribution travels along the fluid channel. An embodiment of a liquid chromatography system includes an injector, a separation column, and a radiation detector. The injector is operable to inject a sample that includes a radiolabeled compound into a solvent stream. The position sensitive radiation detector is operable to image a distribution of the radiolabeled compound as the distribution travels along a fluid channel. An embodiment of a method of liquid chromatography includes injecting a sample that comprises radiolabeled compounds into a solvent. The radiolabeled compounds are then separated. A position sensitive radiation detector is employed to image distributions of the radiolabeled compounds as the radiolabeled compounds travel along a fluid channel.

  2. A Self-Learning Sensor Fault Detection Framework for Industry Monitoring IoT

    Directory of Open Access Journals (Sweden)

    Yu Liu

    2013-01-01

    Full Text Available Many applications based on Internet of Things (IoT technology have recently founded in industry monitoring area. Thousands of sensors with different types work together in an industry monitoring system. Sensors at different locations can generate streaming data, which can be analyzed in the data center. In this paper, we propose a framework for online sensor fault detection. We motivate our technique in the context of the problem of the data value fault detection and event detection. We use the Statistics Sliding Windows (SSW to contain the recent sensor data and regress each window by Gaussian distribution. The regression result can be used to detect the data value fault. Devices on a production line may work in different workloads and the associate sensors will have different status. We divide the sensors into several status groups according to different part of production flow chat. In this way, the status of a sensor is associated with others in the same group. We fit the values in the Status Transform Window (STW to get the slope and generate a group trend vector. By comparing the current trend vector with history ones, we can detect a rational or irrational event. In order to determine parameters for each status group we build a self-learning worker thread in our framework which can edit the corresponding parameter according to the user feedback. Group-based fault detection (GbFD algorithm is proposed in this paper. We test the framework with a simulation dataset extracted from real data of an oil field. Test result shows that GbFD detects 95% sensor fault successfully.

  3. Abnormality detection of mammograms by discriminative dictionary learning on DSIFT descriptors.

    Science.gov (United States)

    Tavakoli, Nasrin; Karimi, Maryam; Nejati, Mansour; Karimi, Nader; Reza Soroushmehr, S M; Samavi, Shadrokh; Najarian, Kayvan

    2017-07-01

    Detection and classification of breast lesions using mammographic images are one of the most difficult studies in medical image processing. A number of learning and non-learning methods have been proposed for detecting and classifying these lesions. However, the accuracy of the detection/classification still needs improvement. In this paper we propose a powerful classification method based on sparse learning to diagnose breast cancer in mammograms. For this purpose, a supervised discriminative dictionary learning approach is applied on dense scale invariant feature transform (DSIFT) features. A linear classifier is also simultaneously learned with the dictionary which can effectively classify the sparse representations. Our experimental results show the superior performance of our method compared to existing approaches.

  4. 49 CFR 1544.213 - Use of explosives detection systems.

    Science.gov (United States)

    2010-10-01

    ... 49 Transportation 9 2010-10-01 2010-10-01 false Use of explosives detection systems. 1544.213...: AIR CARRIERS AND COMMERCIAL OPERATORS Operations § 1544.213 Use of explosives detection systems. (a... explosives detection system approved by TSA to screen checked baggage on international flights. (b) Signs and...

  5. 46 CFR 154.1350 - Flammable gas detection system.

    Science.gov (United States)

    2010-10-01

    ... 46 Shipping 5 2010-10-01 2010-10-01 false Flammable gas detection system. 154.1350 Section 154... Equipment Instrumentation § 154.1350 Flammable gas detection system. (a) The vessel must have a fixed flammable gas detection system that has sampling points in: (1) Each cargo pump room; (2) Each cargo...

  6. 75 FR 5009 - Proximity Detection Systems for Underground Mines

    Science.gov (United States)

    2010-02-01

    ... Proximity Detection Systems for Underground Mines AGENCY: Mine Safety and Health Administration, Labor... information regarding whether the use of proximity detection systems would reduce the risk of accidents where... . Information on MSHA-approved proximity detection systems is available on the Internet at http://www.msha.gov...

  7. A Studi on High Plant Systems Course with Active Learning in Higher Education Through Outdoor Learning to Increase Student Learning Activities

    OpenAIRE

    Nur Rokhimah Hanik, Anwari Adi Nugroho

    2015-01-01

    Biology learning especially high plant system courses needs to be applied to active learning centered on the student (Active Learning In Higher Education) to enhance the students' learning activities so that the quality of learning for the better. Outdoor Learning is one of the active learning invites students to learn outside of the classroom by exploring the surrounding environment. This research aims to improve the students' learning activities in the course of high plant systems through t...

  8. Harnessing the Power of Learning Management Systems: An E-Learning Approach for Professional Development.

    Science.gov (United States)

    White, Meagan; Shellenbarger, Teresa

    E-learning provides an alternative approach to traditional professional development activities. A learning management system may help nursing professional development practitioners deliver content more efficiently and effectively; however, careful consideration is needed during planning and implementation. This article provides essential information in the selection and use of a learning management system for professional development.

  9. Resolving the Problem of Intelligent Learning Content in Learning Management Systems

    Science.gov (United States)

    Rey-Lopez, Marta; Brusilovsky, Peter; Meccawy, Maram; Diaz-Redondo, Rebeca; Fernandez-Vilas, Ana; Ashman, Helen

    2008-01-01

    Current e-learning standardization initiatives have put much effort into easing interoperability between systems and the reusability of contents. For this to be possible, one of the most relevant areas is the definition of a run-time environment, which allows Learning Management Systems to launch, track and communicate with learning objects.…

  10. Cultural impacts on e-learning systems' success

    OpenAIRE

    Aparicio, M.; Bação, F.; Oliveira, T.

    2016-01-01

    WOS:000383295100007 (Nº de Acesso Web of Science) E-learning systems are enablers in the learning process, strengthening their importance as part of the educational strategy. Understanding the determinants of e-learning success is crucial for defining instructional strategies. Several authors have studied e-learning implementation and adoption, and various studies have addressed e-learning success from different perspectives. However, none of these studies have verified whether students' c...

  11. Detecting Structural Metadata with Decision Trees and Transformation-Based Learning

    National Research Council Canada - National Science Library

    Kim, Joungbum; Schwarm, Sarah E; Ostendorf, Mari

    2004-01-01

    .... Specifically, combinations of decision trees and language models are used to predict sentence ends and interruption points and given these events transformation based learning is used to detect edit...

  12. A novel framework for intelligent surveillance system based on abnormal human activity detection in academic environments.

    Science.gov (United States)

    Al-Nawashi, Malek; Al-Hazaimeh, Obaida M; Saraee, Mohamad

    2017-01-01

    Abnormal activity detection plays a crucial role in surveillance applications, and a surveillance system that can perform robustly in an academic environment has become an urgent need. In this paper, we propose a novel framework for an automatic real-time video-based surveillance system which can simultaneously perform the tracking, semantic scene learning, and abnormality detection in an academic environment. To develop our system, we have divided the work into three phases: preprocessing phase, abnormal human activity detection phase, and content-based image retrieval phase. For motion object detection, we used the temporal-differencing algorithm and then located the motions region using the Gaussian function. Furthermore, the shape model based on OMEGA equation was used as a filter for the detected objects (i.e., human and non-human). For object activities analysis, we evaluated and analyzed the human activities of the detected objects. We classified the human activities into two groups: normal activities and abnormal activities based on the support vector machine. The machine then provides an automatic warning in case of abnormal human activities. It also embeds a method to retrieve the detected object from the database for object recognition and identification using content-based image retrieval. Finally, a software-based simulation using MATLAB was performed and the results of the conducted experiments showed an excellent surveillance system that can simultaneously perform the tracking, semantic scene learning, and abnormality detection in an academic environment with no human intervention.

  13. End-to-End Airplane Detection Using Transfer Learning in Remote Sensing Images

    Directory of Open Access Journals (Sweden)

    Zhong Chen

    2018-01-01

    Full Text Available Airplane detection in remote sensing images remains a challenging problem due to the complexity of backgrounds. In recent years, with the development of deep learning, object detection has also obtained great breakthroughs. For object detection tasks in natural images, such as the PASCAL (Pattern Analysis, Statistical Modelling and Computational Learning VOC (Visual Object Classes Challenge, the major trend of current development is to use a large amount of labeled classification data to pre-train the deep neural network as a base network, and then use a small amount of annotated detection data to fine-tune the network for detection. In this paper, we use object detection technology based on deep learning for airplane detection in remote sensing images. In addition to using some characteristics of remote sensing images, some new data augmentation techniques have been proposed. We also use transfer learning and adopt a single deep convolutional neural network and limited training samples to implement end-to-end trainable airplane detection. Classification and positioning are no longer divided into multistage tasks; end-to-end detection attempts to combine them for optimization, which ensures an optimal solution for the final stage. In our experiment, we use remote sensing images of airports collected from Google Earth. The experimental results show that the proposed algorithm is highly accurate and meaningful for remote sensing object detection.

  14. Occlusion detection via structured sparse learning for robust object tracking

    KAUST Repository

    Zhang, Tianzhu; Ghanem, Bernard; Xu, Changsheng; Ahuja, Narendra

    2014-01-01

    occlusion through structured sparse learning. We test our tracker on challenging benchmark sequences, such as sports videos, which involve heavy occlusion, drastic illumination changes, and large pose variations. Extensive experimental results show that our

  15. Object tracking by occlusion detection via structured sparse learning

    KAUST Repository

    Zhang, Tianzhu; Ghanem, Bernard; Xu, Changsheng; Ahuja, Narendra

    2013-01-01

    occlusion through structured sparse learning. We test our tracker on challenging benchmark sequences, such as sports videos, which involve heavy occlusion, drastic illumination changes, and large pose variations. Experimental results show that our tracker

  16. Framework for Designing Context-Aware Learning Systems

    Science.gov (United States)

    Tortorella, Richard A. W.; Kinshuk; Chen, Nian-Shing

    2018-01-01

    Today people learn in many diverse locations and contexts, beyond the confines of classical brick and mortar classrooms. This trend is ever increasing, progressing hand-in-hand with the progress of technology. Context-aware learning systems are systems which adapt to the learner's context, providing tailored learning for a particular learning…

  17. An Intelligent Computer Assisted Language Learning System for Arabic Learners

    Science.gov (United States)

    Shaalan, Khaled F.

    2005-01-01

    This paper describes the development of an intelligent computer-assisted language learning (ICALL) system for learning Arabic. This system could be used for learning Arabic by students at primary schools or by learners of Arabic as a second or foreign language. It explores the use of Natural Language Processing (NLP) techniques for learning…

  18. Which Recommender System Can Best Fit Social Learning Platforms?

    NARCIS (Netherlands)

    Fazeli, Soude; Loni, Babak; Drachsler, Hendrik; Sloep, Peter

    2014-01-01

    This study aims to develop a recommender system for social learning platforms that combine traditional learning management systems with commercial social networks like Facebook. We therefore take into account social interactions of users to make recommendations on learning resources. We propose to

  19. Design of detection module for smart ligthting system

    OpenAIRE

    Matveev, I. G.; Goponenko, A. S.

    2015-01-01

    The paper considers a smart lighting system based on Beaglebone microcomputer. The analysis of existing motion and presence sensors was carried out and then used as a basis for design of a detection system. The detection system and the corresponding connection solution for a smart lighting system were developed. Using the designed smart lighting system, experimental studies were carried out.

  20. Towards Stable Adversarial Feature Learning for LiDAR based Loop Closure Detection

    OpenAIRE

    Xu, Lingyun; Yin, Peng; Luo, Haibo; Liu, Yunhui; Han, Jianda

    2017-01-01

    Stable feature extraction is the key for the Loop closure detection (LCD) task in the simultaneously localization and mapping (SLAM) framework. In our paper, the feature extraction is operated by using a generative adversarial networks (GANs) based unsupervised learning. GANs are powerful generative models, however, GANs based adversarial learning suffers from training instability. We find that the data-code joint distribution in the adversarial learning is a more complex manifold than in the...

  1. Monocular perceptual learning of contrast detection facilitates binocular combination in adults with anisometropic amblyopia.

    Science.gov (United States)

    Chen, Zidong; Li, Jinrong; Liu, Jing; Cai, Xiaoxiao; Yuan, Junpeng; Deng, Daming; Yu, Minbin

    2016-02-01

    Perceptual learning in contrast detection improves monocular visual function in adults with anisometropic amblyopia; however, its effect on binocular combination remains unknown. Given that the amblyopic visual system suffers from pronounced binocular functional loss, it is important to address how the amblyopic visual system responds to such training strategies under binocular viewing conditions. Anisometropic amblyopes (n = 13) were asked to complete two psychophysical supra-threshold binocular summation tasks: (1) binocular phase combination and (2) dichoptic global motion coherence before and after monocular training to investigate this question. We showed that these participants benefited from monocular training in terms of binocular combination. More importantly, the improvements observed with the area under log CSF (AULCSF) were found to be correlated with the improvements in binocular phase combination.

  2. Using YOLO based deep learning network for real time detection and localization of lung nodules from low dose CT scans

    Science.gov (United States)

    Ramachandran S., Sindhu; George, Jose; Skaria, Shibon; V. V., Varun

    2018-02-01

    Lung cancer is the leading cause of cancer related deaths in the world. The survival rate can be improved if the presence of lung nodules are detected early. This has also led to more focus being given to computer aided detection (CAD) and diagnosis of lung nodules. The arbitrariness of shape, size and texture of lung nodules is a challenge to be faced when developing these detection systems. In the proposed work we use convolutional neural networks to learn the features for nodule detection, replacing the traditional method of handcrafting features like geometric shape or texture. Our network uses the DetectNet architecture based on YOLO (You Only Look Once) to detect the nodules in CT scans of lung. In this architecture, object detection is treated as a regression problem with a single convolutional network simultaneously predicting multiple bounding boxes and class probabilities for those boxes. By performing training using chest CT scans from Lung Image Database Consortium (LIDC), NVIDIA DIGITS and Caffe deep learning framework, we show that nodule detection using this single neural network can result in reasonably low false positive rates with high sensitivity and precision.

  3. Toward A Dual-Learning Systems Model of Speech Category Learning

    Directory of Open Access Journals (Sweden)

    Bharath eChandrasekaran

    2014-07-01

    Full Text Available More than two decades of work in vision posits the existence of dual-learning systems of category learning. The reflective system uses working memory to develop and test rules for classifying in an explicit fashion, while the reflexive system operates by implicitly associating perception with actions that lead to reinforcement. Dual-learning systems models hypothesize that in learning natural categories, learners initially use the reflective system and, with practice, transfer control to the reflexive system. The role of reflective and reflexive systems in auditory category learning and more specifically in speech category learning has not been systematically examined. In this article we describe a neurobiologically-constrained dual-learning systems theoretical framework that is currently being developed in speech category learning and review recent applications of this framework. Using behavioral and computational modeling approaches, we provide evidence that speech category learning is predominantly mediated by the reflexive learning system. In one application, we explore the effects of normal aging on non-speech and speech category learning. We find an age related deficit in reflective-optimal but not reflexive-optimal auditory category learning. Prominently, we find a large age-related deficit in speech learning. The computational modeling suggests that older adults are less likely to transition from simple, reflective, uni-dimensional rules to more complex, reflexive, multi-dimensional rules. In a second application we summarize a recent study examining auditory category learning in individuals with elevated depressive symptoms. We find a deficit in reflective-optimal and an enhancement in reflexive-optimal auditory category learning. Interestingly, individuals with elevated depressive symptoms also show an advantage in learning speech categories. We end with a brief summary and description of a number of future directions.

  4. Distributed Impact Detection System, Phase II

    Data.gov (United States)

    National Aeronautics and Space Administration — Automated impact detection and characterization on manned spacecraft has been an elusive goal due to the transitory nature of the detectable high-frequency signals....

  5. Reliability of leak detection systems in LWRs

    International Nuclear Information System (INIS)

    Kupperman, D.S.

    1986-10-01

    In this paper, NRC guidelines for leak detection will be reviewed, current practices described, potential safety-related problems discussed, and potential improvements in leak detection technology (with emphasis on acoustic methods) evaluated

  6. Effects of Semantic Ambiguity Detection Training on Reading Comprehension Achievement of English Learners with Learning Difficulties

    Science.gov (United States)

    Jozwik, Sara L.; Douglas, Karen H.

    2016-01-01

    This study examined how explicit instruction in semantic ambiguity detection affected the reading comprehension and metalinguistic awareness of five English learners (ELs) with learning difficulties (e.g., attention deficit/hyperactivity disorder, specific learning disability). A multiple probe across participants design (Gast & Ledford, 2010)…

  7. Detecting and Understanding the Impact of Cognitive and Interpersonal Conflict in Computer Supported Collaborative Learning Environments

    Science.gov (United States)

    Prata, David Nadler; Baker, Ryan S. J. d.; Costa, Evandro d. B.; Rose, Carolyn P.; Cui, Yue; de Carvalho, Adriana M. J. B.

    2009-01-01

    This paper presents a model which can automatically detect a variety of student speech acts as students collaborate within a computer supported collaborative learning environment. In addition, an analysis is presented which gives substantial insight as to how students' learning is associated with students' speech acts, knowledge that will…

  8. Digital case-based learning system in school.

    Science.gov (United States)

    Gu, Peipei; Guo, Jiayang

    2017-01-01

    With the continuing growth of multi-media learning resources, it is important to offer methods helping learners to explore and acquire relevant learning information effectively. As services that organize multi-media learning materials together to support programming learning, the digital case-based learning system is needed. In order to create a case-oriented e-learning system, this paper concentrates on the digital case study of multi-media resources and learning processes with an integrated framework. An integration of multi-media resources, testing and learning strategies recommendation as the learning unit is proposed in the digital case-based learning framework. The learning mechanism of learning guidance, multi-media materials learning and testing feedback is supported in our project. An improved personalized genetic algorithm which incorporates preference information and usage degree into the crossover and mutation process is proposed to assemble the personalized test sheet for each learner. A learning strategies recommendation solution is proposed to recommend learning strategies for learners to help them to learn. The experiments are conducted to prove that the proposed approaches are capable of constructing personalized sheets and the effectiveness of the framework.

  9. Digital case-based learning system in school.

    Directory of Open Access Journals (Sweden)

    Peipei Gu

    Full Text Available With the continuing growth of multi-media learning resources, it is important to offer methods helping learners to explore and acquire relevant learning information effectively. As services that organize multi-media learning materials together to support programming learning, the digital case-based learning system is needed. In order to create a case-oriented e-learning system, this paper concentrates on the digital case study of multi-media resources and learning processes with an integrated framework. An integration of multi-media resources, testing and learning strategies recommendation as the learning unit is proposed in the digital case-based learning framework. The learning mechanism of learning guidance, multi-media materials learning and testing feedback is supported in our project. An improved personalized genetic algorithm which incorporates preference information and usage degree into the crossover and mutation process is proposed to assemble the personalized test sheet for each learner. A learning strategies recommendation solution is proposed to recommend learning strategies for learners to help them to learn. The experiments are conducted to prove that the proposed approaches are capable of constructing personalized sheets and the effectiveness of the framework.

  10. A Web-Based Learning Support System for Inquiry-Based Learning

    Science.gov (United States)

    Kim, Dong Won; Yao, Jingtao

    The emergence of the Internet and Web technology makes it possible to implement the ideals of inquiry-based learning, in which students seek truth, information, or knowledge by questioning. Web-based learning support systems can provide a good framework for inquiry-based learning. This article presents a study on a Web-based learning support system called Online Treasure Hunt. The Web-based learning support system mainly consists of a teaching support subsystem, a learning support subsystem, and a treasure hunt game. The teaching support subsystem allows instructors to design their own inquiry-based learning environments. The learning support subsystem supports students' inquiry activities. The treasure hunt game enables students to investigate new knowledge, develop ideas, and review their findings. Online Treasure Hunt complies with a treasure hunt model. The treasure hunt model formalizes a general treasure hunt game to contain the learning strategies of inquiry-based learning. This Web-based learning support system empowered with the online-learning game and founded on the sound learning strategies furnishes students with the interactive and collaborative student-centered learning environment.

  11. A Teaching System To Learn Programming: the Programmer's Learning Machine

    OpenAIRE

    Quinson , Martin; Oster , Gérald

    2015-01-01

    International audience; The Programmer's Learning Machine (PLM) is an interactive exerciser for learning programming and algorithms. Using an integrated and graphical environment that provides a short feedback loop, it allows students to learn in a (semi)-autonomous way. This generic platform also enables teachers to create specific programming microworlds that match their teaching goals. This paper discusses our design goals and motivations, introduces the existing material and the proposed ...

  12. Towards an intelligent learning management system under blended learning trends, profiles and modeling perspectives

    CERN Document Server

    Dias, Sofia B; Hadjileontiadis, Leontios J

    2013-01-01

    This book offers useful information that evokes initiatives towards rethinking of the value, efficiency, inclusiveness, effectiveness and personalization of the intelligent learning management systems-based blended-learning environment.

  13. Intelligent data analysis for e-learning enhancing security and trustworthiness in online learning systems

    CERN Document Server

    Miguel, Jorge; Xhafa, Fatos

    2016-01-01

    Intelligent Data Analysis for e-Learning: Enhancing Security and Trustworthiness in Online Learning Systems addresses information security within e-Learning based on trustworthiness assessment and prediction. Over the past decade, many learning management systems have appeared in the education market. Security in these systems is essential for protecting against unfair and dishonest conduct-most notably cheating-however, e-Learning services are often designed and implemented without considering security requirements. This book provides functional approaches of trustworthiness analysis, modeling, assessment, and prediction for stronger security and support in online learning, highlighting the security deficiencies found in most online collaborative learning systems. The book explores trustworthiness methodologies based on collective intelligence than can overcome these deficiencies. It examines trustworthiness analysis that utilizes the large amounts of data-learning activities generate. In addition, as proc...

  14. Assessing segmentation processes by click detection: online measure of statistical learning, or simple interference?

    Science.gov (United States)

    Franco, Ana; Gaillard, Vinciane; Cleeremans, Axel; Destrebecqz, Arnaud

    2015-12-01

    Statistical learning can be used to extract the words from continuous speech. Gómez, Bion, and Mehler (Language and Cognitive Processes, 26, 212-223, 2011) proposed an online measure of statistical learning: They superimposed auditory clicks on a continuous artificial speech stream made up of a random succession of trisyllabic nonwords. Participants were instructed to detect these clicks, which could be located either within or between words. The results showed that, over the length of exposure, reaction times (RTs) increased more for within-word than for between-word clicks. This result has been accounted for by means of statistical learning of the between-word boundaries. However, even though statistical learning occurs without an intention to learn, it nevertheless requires attentional resources. Therefore, this process could be affected by a concurrent task such as click detection. In the present study, we evaluated the extent to which the click detection task indeed reflects successful statistical learning. Our results suggest that the emergence of RT differences between within- and between-word click detection is neither systematic nor related to the successful segmentation of the artificial language. Therefore, instead of being an online measure of learning, the click detection task seems to interfere with the extraction of statistical regularities.

  15. Perceptual strategies of pigeons to detect a rotational centre--a hint for star compass learning?

    Directory of Open Access Journals (Sweden)

    Bianca Alert

    Full Text Available Birds can rely on a variety of cues for orientation during migration and homing. Celestial rotation provides the key information for the development of a functioning star and/or sun compass. This celestial compass seems to be the primary reference for calibrating the other orientation systems including the magnetic compass. Thus, detection of the celestial rotational axis is crucial for bird orientation. Here, we use operant conditioning to demonstrate that homing pigeons can principally learn to detect a rotational centre in a rotating dot pattern and we examine their behavioural response strategies in a series of experiments. Initially, most pigeons applied a strategy based on local stimulus information such as movement characteristics of single dots. One pigeon seemed to immediately ignore eccentric stationary dots. After special training, all pigeons could shift their attention to more global cues, which implies that pigeons can learn the concept of a rotational axis. In our experiments, the ability to precisely locate the rotational centre was strongly dependent on the rotational velocity of the dot pattern and it crashed at velocities that were still much faster than natural celestial rotation. We therefore suggest that the axis of the very slow, natural, celestial rotation could be perceived by birds through the movement itself, but that a time-delayed pattern comparison should also be considered as a very likely alternative strategy.

  16. Hybrid Intrusion Detection System for DDoS Attacks

    Directory of Open Access Journals (Sweden)

    Özge Cepheli

    2016-01-01

    Full Text Available Distributed denial-of-service (DDoS attacks are one of the major threats and possibly the hardest security problem for today’s Internet. In this paper we propose a hybrid detection system, referred to as hybrid intrusion detection system (H-IDS, for detection of DDoS attacks. Our proposed detection system makes use of both anomaly-based and signature-based detection methods separately but in an integrated fashion and combines the outcomes of both detectors to enhance the overall detection accuracy. We apply two distinct datasets to our proposed system in order to test the detection performance of H-IDS and conclude that the proposed hybrid system gives better results than the systems based on nonhybrid detection.

  17. Cognitive Models for Learning to Control Dynamic Systems

    National Research Council Canada - National Science Library

    Eberhart, Russ; Hu, Xiaohui; Chen, Yaobin

    2008-01-01

    Report developed under STTR contract for topic "Cognitive models for learning to control dynamic systems" demonstrated a swarm intelligence learning algorithm and its application in unmanned aerial vehicle (UAV) mission planning...

  18. Introduction of a learning management system at the Kilimanjaro ...

    African Journals Online (AJOL)

    Background. Medical schools in Africa face daunting challenges including faculty shortages, growing class sizes, and inadequate resources. Learning management systems (LMS) may be powerful tools for organising and presenting curricular learning materials, with the potential for monitoring and evaluation functions.

  19. Planet Detectability in the Alpha Centauri System

    Science.gov (United States)

    Zhao, Lily; Fischer, Debra A.; Brewer, John; Giguere, Matt; Rojas-Ayala, Bárbara

    2018-01-01

    We use more than a decade of radial-velocity measurements for α {Cen} A, B, and Proxima Centauri from the High Accuracy Radial Velocity Planet Searcher, CTIO High Resolution Spectrograph, and the Ultraviolet and Visual Echelle Spectrograph to identify the M\\sin i and orbital periods of planets that could have been detected if they existed. At each point in a mass–period grid, we sample a simulated, Keplerian signal with the precision and cadence of existing data and assess the probability that the signal could have been produced by noise alone. Existing data places detection thresholds in the classically defined habitable zones at about M\\sin i of 53 {M}\\oplus for α {Cen} A, 8.4 {M}\\oplus for α {Cen} B, and 0.47 {M}\\oplus for Proxima Centauri. Additionally, we examine the impact of systematic errors, or “red noise” in the data. A comparison of white- and red-noise simulations highlights quasi-periodic variability in the radial velocities that may be caused by systematic errors, photospheric velocity signals, or planetary signals. For example, the red-noise simulations show a peak above white-noise simulations at the period of Proxima Centauri b. We also carry out a spectroscopic analysis of the chemical composition of the α {Centauri} stars. The stars have super-solar metallicity with ratios of C/O and Mg/Si that are similar to the Sun, suggesting that any small planets in the α {Cen} system may be compositionally similar to our terrestrial planets. Although the small projected separation of α {Cen} A and B currently hampers extreme-precision radial-velocity measurements, the angular separation is now increasing. By 2019, α {Cen} A and B will be ideal targets for renewed Doppler planet surveys.

  20. Safe Detection System for Hydrogen Leaks

    Energy Technology Data Exchange (ETDEWEB)

    Lieberman, Robert A. [Intelligent Optical Systems, Inc., Torrance, CA (United States); Beshay, Manal [Intelligent Optical Systems, Inc., Torrance, CA (United States)

    2012-02-29

    Hydrogen is an "environmentally friendly" fuel for future transportation and other applications, since it produces only pure ("distilled") water when it is consumed. Thus, hydrogen-powered vehicles are beginning to proliferate, with the total number of such vehicles expected to rise to nearly 100,000 within the next few years. However, hydrogen is also an odorless, colorless, highly flammable gas. Because of this, there is an important need for hydrogen safety monitors that can warn of hazardous conditions in vehicles, storage facilities, and hydrogen production plants. To address this need, IOS has developed a unique intrinsically safe optical hydrogen sensing technology, and has embodied it in detector systems specifically developed for safety applications. The challenge of using light to detect a colorless substance was met by creating chemically-sensitized optical materials whose color changes in the presence of hydrogen. This reversible reaction provides a sensitive, reliable, way of detecting hydrogen and measuring its concentration using light from low-cost LEDs. Hydrogen sensors based on this material were developed in three completely different optical formats: point sensors ("optrodes"), integrated optic sensors ("optical chips"), and optical fibers ("distributed sensors") whose entire length responds to hydrogen. After comparing performance, cost, time-to-market, and relative market need for these sensor types, the project focused on designing a compact optrode-based single-point hydrogen safety monitor. The project ended with the fabrication of fifteen prototype units, and the selection of two specific markets: fuel cell enclosure monitoring, and refueling/storage safety. Final testing and development of control software for these markets await future support.

  1. Computation in the Learning System of Cephalopods.

    Science.gov (United States)

    Young, J Z

    1991-04-01

    The memory mechanisms of cephalopods consist of a series of matrices of intersecting axes, which find associations between the signals of input events and their consequences. The tactile memory is distributed among eight such matrices, and there is also some suboesophageal learning capacity. The visual memory lies in the optic lobe and four matrices, with some re-exciting pathways. In both systems, damage to any part reduces proportionally the effectiveness of the whole memory. These matrices are somewhat like those in mammals, for instance those in the hippocampus. The first matrix in both visual and tactile systems receives signals of vision and taste, and its output serves to increase the tendency to attack or to take with the arms. The second matrix provides for the correlation of groups of signals on its neurons, which pass signals to the third matrix. Here large cells find clusters in the sets of signals. Their output re-excites those of the first lobe, unless pain occurs. In that case, this set of cells provides a record that ensures retreat. There is experimental evidence that these distributed memory systems allow for the identification of categories of visual and tactile inputs, for generalization, and for decision on appropriate behavior in the light of experience. The evidence suggests that learning in cephalopods is not localized to certain layers or "grandmother cells" but is distributed with high redundance in serial networks, with recurrent circuits.

  2. Adaptive Hypermedia Systems for E-Learning

    Directory of Open Access Journals (Sweden)

    Aammou Souhaib

    2010-11-01

    Full Text Available The domain of traditional hypermedia is revolutionized by the arrival of the concept of adaptation. Currently the domain of Adaptive Hypermedia Systems (AHS is constantly growing. A major goal of current research is to provide a personalized educational experience that meets the needs specific to each learner (knowledge level, goals, motivation etc.... In this article we have studied the possibility of implementing traditional features of adaptive hypermedia in an open environment, and discussed the standards for describing learning objects and architectural models based on the use of ontologies as a prerequisite for such an adaptation.

  3. System of Consciousness Contextual as Learning Support

    Directory of Open Access Journals (Sweden)

    Jorge Eliseo Gómez Gómez

    2016-08-01

    Full Text Available With the inclusion of ubiquitous computing in education, it is intended that the student is an active agent in its formation process and interacts with its context. This paper presents the design and implementation of architecture for ubiquitous learning environments in which they integrate physical spaces with applications which are then executed by users. The system supports objects augmented with RFID tags, NFC and QR Code. Each tag contains data that uniquely identifies the resource. To validate the proposed architecture is developed experimentally tested a prototype at the end of it is done to verify that the proposed architecture improves the academic performance of students.

  4. Detection of contamination of municipal water distribution systems

    Science.gov (United States)

    Cooper, John F [Oakland, CA

    2012-01-17

    A system for the detection of contaminates of a fluid in a conduit. The conduit is part of a fluid distribution system. A chemical or biological sensor array is connected to the conduit. The sensor array produces an acoustic signal burst in the fluid upon detection of contaminates in the fluid. A supervisory control system connected to the fluid and operatively connected to the fluid distribution system signals the fluid distribution system upon detection of contaminates in the fluid.

  5. Recommender systems for technology enhanced learning research trends and applications

    CERN Document Server

    Manouselis, Nikos; Verbert, Katrien

    2014-01-01

    Presents cutting edge research from leading experts in the growing field of Recommender Systems for Technology Enhanced Learning (RecSys TEL) International contributions are included to demonstrate the merging of various efforts and communities Topics include: Linked Data and the Social Web as Facilitators for TEL Recommender Systems in Research and Practice, Personalised Learning-Plan Recommendations in Game-Based Learning and Recommendations from Heterogeneous Sources in a Technology Enhanced Learning Ecosystem

  6. A model of positive and negative learning : Learning demands and resources, learning engagement, critical thinking, and fake news detection

    NARCIS (Netherlands)

    Dormann, Christian; Demerouti, Eva; Bakker, Arnold; Zlatkin-Troitschanskaia, O.; Wittum, G.; Dengel, A.

    2018-01-01

    This chapter proposes a model of positive and negative learning (PNL model). We use the term negative learning when stress among students occurs, and when knowledge and abilities are not properly developed. We use the term positive learning if motivation is high and active learning occurs. The PNL

  7. Research on IPv6 intrusion detection system Snort-based

    Science.gov (United States)

    Shen, Zihao; Wang, Hui

    2010-07-01

    This paper introduces the common intrusion detection technologies, discusses the work flow of Snort intrusion detection system, and analyzes IPv6 data packet encapsulation and protocol decoding technology. We propose the expanding Snort architecture to support IPv6 intrusion detection in accordance with CIDF standard combined with protocol analysis technology and pattern matching technology, and present its composition. The research indicates that the expanding Snort system can effectively detect various intrusion attacks; it is high in detection efficiency and detection accuracy and reduces false alarm and omission report, which effectively solves the problem of IPv6 intrusion detection.

  8. The Office Software Learning and Examination System Design Based on Fragmented Learning Idea

    Directory of Open Access Journals (Sweden)

    Xu Ling

    2016-01-01

    Full Text Available Fragmented learning is that through the segmentation of learning content or learning time, make learners can use the fragmented time for learning fragmentated content, have the characteristics of time flexibility, learning targeted and high learning efficiency. Based on the fragmented learning ideas, combined with the teaching idea of micro class and interactive teaching, comprehensive utilization of flash animation design software, .NET development platform, VSTO technology, multimedia development technology and so on, design and develop a system integrated with learning, practice and examination of the Office software, which is not only conducive to the effective and personalized learning of students, but also conducive to the understanding the students’ situation of teachers, and liberate teachers from the heavy labor of mechanical, focus on promoting the formation of students’ knowledge system.

  9. A Plane Target Detection Algorithm in Remote Sensing Images based on Deep Learning Network Technology

    Science.gov (United States)

    Shuxin, Li; Zhilong, Zhang; Biao, Li

    2018-01-01

    Plane is an important target category in remote sensing targets and it is of great value to detect the plane targets automatically. As remote imaging technology developing continuously, the resolution of the remote sensing image has been very high and we can get more detailed information for detecting the remote sensing targets automatically. Deep learning network technology is the most advanced technology in image target detection and recognition, which provided great performance improvement in the field of target detection and recognition in the everyday scenes. We combined the technology with the application in the remote sensing target detection and proposed an algorithm with end to end deep network, which can learn from the remote sensing images to detect the targets in the new images automatically and robustly. Our experiments shows that the algorithm can capture the feature information of the plane target and has better performance in target detection with the old methods.

  10. Distributed Fault Detection for a Class of Nonlinear Stochastic Systems

    Directory of Open Access Journals (Sweden)

    Bingyong Yan

    2014-01-01

    Full Text Available A novel distributed fault detection strategy for a class of nonlinear stochastic systems is presented. Different from the existing design procedures for fault detection, a novel fault detection observer, which consists of a nonlinear fault detection filter and a consensus filter, is proposed to detect the nonlinear stochastic systems faults. Firstly, the outputs of the nonlinear stochastic systems act as inputs of a consensus filter. Secondly, a nonlinear fault detection filter is constructed to provide estimation of unmeasurable system states and residual signals using outputs of the consensus filter. Stability analysis of the consensus filter is rigorously investigated. Meanwhile, the design procedures of the nonlinear fault detection filter are given in terms of linear matrix inequalities (LMIs. Taking the influence of the system stochastic noises into consideration, an outstanding feature of the proposed scheme is that false alarms can be reduced dramatically. Finally, simulation results are provided to show the feasibility and effectiveness of the proposed fault detection approach.

  11. Thermal History Devices, Systems For Thermal History Detection, And Methods For Thermal History Detection

    KAUST Repository

    Caraveo Frescas, Jesus Alfonso; Alshareef, Husam N.

    2015-01-01

    Embodiments of the present disclosure include nanowire field-effect transistors, systems for temperature history detection, methods for thermal history detection, a matrix of field effect transistors, and the like.

  12. Thermal History Devices, Systems For Thermal History Detection, And Methods For Thermal History Detection

    KAUST Repository

    Caraveo Frescas, Jesus Alfonso

    2015-05-28

    Embodiments of the present disclosure include nanowire field-effect transistors, systems for temperature history detection, methods for thermal history detection, a matrix of field effect transistors, and the like.

  13. STRATEGY FOR EVALUATION AND SELECTION OF SYSTEMS FOR ELECTRONIC LEARNING

    OpenAIRE

    Dubravka Mandušić; Lucija Blašković

    2012-01-01

    Today`s technology supported and accelerated learning time requires constant and continuous acquisition of new knowledge. On the other hand, it does not leave enough time for additional education. Increasing number of E-learning systems, withdraws a need for precise evaluation of functionality that those systems provide; so they could be reciprocally compared. While implementing new systems for electronic learning, it is very important to pre-evaluate existing systems in order to ...

  14. Development of a Relational Database for Learning Management Systems

    Science.gov (United States)

    Deperlioglu, Omer; Sarpkaya, Yilmaz; Ergun, Ertugrul

    2011-01-01

    In today's world, Web-Based Distance Education Systems have a great importance. Web-based Distance Education Systems are usually known as Learning Management Systems (LMS). In this article, a database design, which was developed to create an educational institution as a Learning Management System, is described. In this sense, developed Learning…

  15. Adaptive Learning Systems: Beyond Teaching Machines

    Science.gov (United States)

    Kara, Nuri; Sevim, Nese

    2013-01-01

    Since 1950s, teaching machines have changed a lot. Today, we have different ideas about how people learn, what instructor should do to help students during their learning process. We have adaptive learning technologies that can create much more student oriented learning environments. The purpose of this article is to present these changes and its…

  16. ACPR upgrade fuel motion detection system

    International Nuclear Information System (INIS)

    Kelly, J.G.; Stalker, K.T.

    1976-01-01

    Coded aperture imaging techniques are being applied to the problem of fuel motion detection in core disruptive safety experiments in the LMFBR. The upgraded Sandia Annular Core Pulse Reactor will be used for fuel motion studies, and a system is now being developed to image fuel motion through a slot in the core in l- and 7-pin experiments. The γ-rays emitted by the fuel will be modulated by a coded aperture and will form a pseudohologram on a scintillator. This shadowgram will be lens coupled first to an optical image intensifier, and then to a rotating prism framing camera. The goal is to obtain quality images of fuel with approximately 1 mm spatial resolution at about 10 4 frames/sec. Some of the feasibility milestones which have been reached thus far are the following: (1) A fuel pin shaped 252 Cf fission gamma ray source has been imaged through 1.27 cm of steel (simulating reactor containment). (2) A 252 Cf source has been imaged through 1.27 cm of steel with x-ray image intensifier amplification. (3) Collimation and shielding tests indicate that at ACPR it should be possible to obtain adequate signal to noise at the detector when the fuel is observed through a slot in the core

  17. Detection systems for radioactive ion beams

    International Nuclear Information System (INIS)

    Savajols, H.

    2002-01-01

    Two main methods are used to produce radioactive ion beams: -) the ISOL method (isotope separation on-line) in which the stable beam interacts with a thick target, the reaction products diffuse outside the target and are transferred to a source where they are ionized, a mass separator and a post-accelerator drive the selected radioactive ions to the right energy; -) the in-flight fragmentation method in which the stable beam interacts with a thin target, the reaction products are emitted from the target with a restricted angular distribution and a velocity close to that of the incident beam, the experimenter has to take advantage from the reaction kinetics to get the right particle beam. Characteristic time is far longer with the ISOL method but the beam intensity is much better because of the use of a post-accelerator. In both cases, the beam intensity is lower by several orders of magnitude than in the case of a stable beam. This article presents all the constraints imposed by radioactive beams to the detection systems of the reaction products and gives new technical solutions according to the type of nuclear reaction studied. (A.C.)

  18. Fast-neutron detecting system with n, γ discrimination

    International Nuclear Information System (INIS)

    Ouyang Xiaoping; Huang Bao; Cao Jinyun

    1997-11-01

    In the present work, a new type neutron detecting system is reported, which can absolutely measure neutron parameters in n + γ mixed fields and has a long continuance of static high vacuum of 10 -4 Pa. The detecting system, with middle neutron-detecting sensitivity, short time response and big linear current output, has applied successfully in pulsed neutron beam measurement

  19. Sensorimotor learning configures the human mirror system.

    Science.gov (United States)

    Catmur, Caroline; Walsh, Vincent; Heyes, Cecilia

    2007-09-04

    Cells in the "mirror system" fire not only when an individual performs an action but also when one observes the same action performed by another agent [1-4]. The mirror system, found in premotor and parietal cortices of human and monkey brains, is thought to provide the foundation for social understanding and to enable the development of theory of mind and language [5-9]. However, it is unclear how mirror neurons acquire their mirror properties -- how they derive the information necessary to match observed with executed actions [10]. We address this by showing that it is possible to manipulate the selectivity of the human mirror system, and thereby make it operate as a countermirror system, by giving participants training to perform one action while observing another. Before this training, participants showed event-related muscle-specific responses to transcranial magnetic stimulation over motor cortex during observation of little- and index-finger movements [11-13]. After training, this normal mirror effect was reversed. These results indicate that the mirror properties of the mirror system are neither wholly innate [14] nor fixed once acquired; instead they develop through sensorimotor learning [15, 16]. Our findings indicate that the human mirror system is, to some extent, both a product and a process of social interaction.

  20. Learning modalities in artificial intelligence systems: a framework and review

    Energy Technology Data Exchange (ETDEWEB)

    Araya, A A

    1982-01-01

    Intelligent systems should possess two fundamental capabilities: problem solving and learning. Problem solving capabilities allow an intelligent system to cope with problems in a given domain. Learning capabilities make possible for an intelligent system to improve the solution to the problems within its current reach or to cope with new problems. This paper examines research in artificial intelligence from the perspective of learning with the purpose of: 1) developing and understanding of the problem of learning from the AI point of view, and II) characterizing the current state of the art on learning in AI. 35 references.

  1. Systems and Methods for Automated Water Detection Using Visible Sensors

    Science.gov (United States)

    Rankin, Arturo L. (Inventor); Matthies, Larry H. (Inventor); Bellutta, Paolo (Inventor)

    2016-01-01

    Systems and methods are disclosed that include automated machine vision that can utilize images of scenes captured by a 3D imaging system configured to image light within the visible light spectrum to detect water. One embodiment includes autonomously detecting water bodies within a scene including capturing at least one 3D image of a scene using a sensor system configured to detect visible light and to measure distance from points within the scene to the sensor system, and detecting water within the scene using a processor configured to detect regions within each of the at least one 3D images that possess at least one characteristic indicative of the presence of water.

  2. Detection technique of targets for missile defense system

    Science.gov (United States)

    Guo, Hua-ling; Deng, Jia-hao; Cai, Ke-rong

    2009-11-01

    Ballistic missile defense system (BMDS) is a weapon system for intercepting enemy ballistic missiles. It includes ballistic-missile warning system, target discrimination system, anti-ballistic-missile guidance systems, and command-control communication system. Infrared imaging detection and laser imaging detection are widely used in BMDS for surveillance, target detection, target tracking, and target discrimination. Based on a comprehensive review of the application of target-detection techniques in the missile defense system, including infrared focal plane arrays (IRFPA), ground-based radar detection technology, 3-dimensional imaging laser radar with a photon counting avalanche photodiode (APD) arrays and microchip laser, this paper focuses on the infrared and laser imaging detection techniques in missile defense system, as well as the trends for their future development.

  3. Saliency Detection and Deep Learning-Based Wildfire Identification in UAV Imagery

    Directory of Open Access Journals (Sweden)

    Yi Zhao

    2018-02-01

    Full Text Available An unmanned aerial vehicle (UAV equipped with global positioning systems (GPS can provide direct georeferenced imagery, mapping an area with high resolution. So far, the major difficulty in wildfire image classification is the lack of unified identification marks, the fire features of color, shape, texture (smoke, flame, or both and background can vary significantly from one scene to another. Deep learning (e.g., DCNN for Deep Convolutional Neural Network is very effective in high-level feature learning, however, a substantial amount of training images dataset is obligatory in optimizing its weights value and coefficients. In this work, we proposed a new saliency detection algorithm for fast location and segmentation of core fire area in aerial images. As the proposed method can effectively avoid feature loss caused by direct resizing; it is used in data augmentation and formation of a standard fire image dataset ‘UAV_Fire’. A 15-layered self-learning DCNN architecture named ‘Fire_Net’ is then presented as a self-learning fire feature exactor and classifier. We evaluated different architectures and several key parameters (drop out ratio, batch size, etc. of the DCNN model regarding its validation accuracy. The proposed architecture outperformed previous methods by achieving an overall accuracy of 98%. Furthermore, ‘Fire_Net’ guarantied an average processing speed of 41.5 ms per image for real-time wildfire inspection. To demonstrate its practical utility, Fire_Net is tested on 40 sampled images in wildfire news reports and all of them have been accurately identified.

  4. Single vs. dual color fire detection systems: operational tradeoffs

    Science.gov (United States)

    Danino, Meir; Danan, Yossef; Sinvani, Moshe

    2017-10-01

    In attempt to supply a reasonable fire plume detection, multinational cooperation with significant capital is invested in the development of two major Infra-Red (IR) based fire detection alternatives, single-color IR (SCIR) and dual-color IR (DCIR). False alarm rate was expected to be high not only as a result of real heat sources but mainly due to the IR natural clutter especially solar reflections clutter. SCIR uses state-of-the-art technology and sophisticated algorithms to filter out threats from clutter. On the other hand, DCIR are aiming at using additional spectral band measurements (acting as a guard), to allow the implementation of a simpler and more robust approach for performing the same task. In this paper we present the basics of SCIR & DCIR architecture and the main differences between them. In addition, we will present the results from a thorough study conducted for the purpose of learning about the added value of the additional data available from the second spectral band. Here we consider the two CO2 bands of 4-5 micron and of 2.5-3 micron band as well as off peak band (guard). The findings of this study refer also to Missile warning systems (MWS) efficacy, in terms of operational value. We also present a new approach for tunable filter to such sensor.

  5. Transductive and matched-pair machine learning for difficult target detection problems

    Science.gov (United States)

    Theiler, James

    2014-06-01

    This paper will describe the application of two non-traditional kinds of machine learning (transductive machine learning and the more recently proposed matched-pair machine learning) to the target detection problem. The approach combines explicit domain knowledge to model the target signal with a more agnostic machine-learning approach to characterize the background. The concept is illustrated with simulated data from an elliptically-contoured background distribution, on which a subpixel target of known spectral signature but unknown spatial extent has been implanted.

  6. Hybrid image representation learning model with invariant features for basal cell carcinoma detection

    Science.gov (United States)

    Arevalo, John; Cruz-Roa, Angel; González, Fabio A.

    2013-11-01

    This paper presents a novel method for basal-cell carcinoma detection, which combines state-of-the-art methods for unsupervised feature learning (UFL) and bag of features (BOF) representation. BOF, which is a form of representation learning, has shown a good performance in automatic histopathology image classi cation. In BOF, patches are usually represented using descriptors such as SIFT and DCT. We propose to use UFL to learn the patch representation itself. This is accomplished by applying a topographic UFL method (T-RICA), which automatically learns visual invariance properties of color, scale and rotation from an image collection. These learned features also reveals these visual properties associated to cancerous and healthy tissues and improves carcinoma detection results by 7% with respect to traditional autoencoders, and 6% with respect to standard DCT representations obtaining in average 92% in terms of F-score and 93% of balanced accuracy.

  7. A review for detecting gene-gene interactions using machine learning methods in genetic epidemiology.

    Science.gov (United States)

    Koo, Ching Lee; Liew, Mei Jing; Mohamad, Mohd Saberi; Salleh, Abdul Hakim Mohamed

    2013-01-01

    Recently, the greatest statistical computational challenge in genetic epidemiology is to identify and characterize the genes that interact with other genes and environment factors that bring the effect on complex multifactorial disease. These gene-gene interactions are also denoted as epitasis in which this phenomenon cannot be solved by traditional statistical method due to the high dimensionality of the data and the occurrence of multiple polymorphism. Hence, there are several machine learning methods to solve such problems by identifying such susceptibility gene which are neural networks (NNs), support vector machine (SVM), and random forests (RFs) in such common and multifactorial disease. This paper gives an overview on machine learning methods, describing the methodology of each machine learning methods and its application in detecting gene-gene and gene-environment interactions. Lastly, this paper discussed each machine learning method and presents the strengths and weaknesses of each machine learning method in detecting gene-gene interactions in complex human disease.

  8. Machine learning for network-based malware detection

    DEFF Research Database (Denmark)

    Stevanovic, Matija

    and based on different, mutually complementary, principles of traffic analysis. The proposed approaches rely on machine learning algorithms (MLAs) for automated and resource-efficient identification of the patterns of malicious network traffic. We evaluated the proposed methods through extensive evaluations...

  9. Improving EEG signal peak detection using feature weight learning ...

    Indian Academy of Sciences (India)

    Therefore, we aimed to develop a general procedure for eye event-related applications based on feature weight learning (FWL), through the use of a neural network with random weights (NNRW) as the classifier. The FWL is performed using a particle swarm optimization algorithm, applied to the well-studied Dumpala, Acir, ...

  10. Learning Analytics for Supporting Seamless Language Learning Using E-Book with Ubiquitous Learning System

    Science.gov (United States)

    Mouri, Kousuke; Uosaki, Noriko; Ogata, Hiroaki

    2018-01-01

    Seamless learning has been recognized as an effective learning approach across various dimensions including formal and informal learning contexts, individual and social learning, and physical world and cyberspace. With the emergence of seamless learning, the majority of the current research focuses on realizing a seamless learning environment at…

  11. Improving buried threat detection in ground-penetrating radar with transfer learning and metadata analysis

    Science.gov (United States)

    Colwell, Kenneth A.; Torrione, Peter A.; Morton, Kenneth D.; Collins, Leslie M.

    2015-05-01

    Ground-penetrating radar (GPR) technology has proven capable of detecting buried threats. The system relies on a binary classifier that is trained to distinguish between two classes: a target class, encompassing many types of buried threats and their components; and a nontarget class, which includes false alarms from the system prescreener. Typically, the training process involves a simple partition of the data into these two classes, which allows for straightforward application of standard classifiers. However, since training data is generally collected in fully controlled environments, it includes auxiliary information about each example, such as the specific type of threat, its purpose, its components, and its depth. Examples from the same specific or general type may be expected to exhibit similarities in their GPR data, whereas examples from different types may differ greatly. This research aims to leverage this additional information to improve overall classification performance by fusing classifier concepts for multiple groups, and to investigate whether structure in this information can be further utilized for transfer learning, such that the amount of expensive training data necessary to learn a new, previously-unseen target type may be reduced. Methods for accomplishing these goals are presented with results from a dataset containing a variety of target types.

  12. Active learning approach for detection of hard exudates, cotton wool spots, and drusen in retinal images

    Science.gov (United States)

    Sánchez, Clara I.; Niemeijer, Meindert; Kockelkorn, Thessa; Abràmoff, Michael D.; van Ginneken, Bram

    2009-02-01

    Computer-aided Diagnosis (CAD) systems for the automatic identification of abnormalities in retinal images are gaining importance in diabetic retinopathy screening programs. A huge amount of retinal images are collected during these programs and they provide a starting point for the design of machine learning algorithms. However, manual annotations of retinal images are scarce and expensive to obtain. This paper proposes a dynamic CAD system based on active learning for the automatic identification of hard exudates, cotton wool spots and drusen in retinal images. An uncertainty sampling method is applied to select samples that need to be labeled by an expert from an unlabeled set of 4000 retinal images. It reduces the number of training samples needed to obtain an optimum accuracy by dynamically selecting the most informative samples. Results show that the proposed method increases the classification accuracy compared to alternative techniques, achieving an area under the ROC curve of 0.87, 0.82 and 0.78 for the detection of hard exudates, cotton wool spots and drusen, respectively.

  13. Feature learning and change feature classification based on deep learning for ternary change detection in SAR images

    Science.gov (United States)

    Gong, Maoguo; Yang, Hailun; Zhang, Puzhao

    2017-07-01

    Ternary change detection aims to detect changes and group the changes into positive change and negative change. It is of great significance in the joint interpretation of spatial-temporal synthetic aperture radar images. In this study, sparse autoencoder, convolutional neural networks (CNN) and unsupervised clustering are combined to solve ternary change detection problem without any supervison. Firstly, sparse autoencoder is used to transform log-ratio difference image into a suitable feature space for extracting key changes and suppressing outliers and noise. And then the learned features are clustered into three classes, which are taken as the pseudo labels for training a CNN model as change feature classifier. The reliable training samples for CNN are selected from the feature maps learned by sparse autoencoder with certain selection rules. Having training samples and the corresponding pseudo labels, the CNN model can be trained by using back propagation with stochastic gradient descent. During its training procedure, CNN is driven to learn the concept of change, and more powerful model is established to distinguish different types of changes. Unlike the traditional methods, the proposed framework integrates the merits of sparse autoencoder and CNN to learn more robust difference representations and the concept of change for ternary change detection. Experimental results on real datasets validate the effectiveness and superiority of the proposed framework.

  14. Machine learning techniques applied to system characterization and equalization

    DEFF Research Database (Denmark)

    Zibar, Darko; Thrane, Jakob; Wass, Jesper

    2016-01-01

    Linear signal processing algorithms are effective in combating linear fibre channel impairments. We demonstrate the ability of machine learning algorithms to combat nonlinear fibre channel impairments and perform parameter extraction from directly detected signals.......Linear signal processing algorithms are effective in combating linear fibre channel impairments. We demonstrate the ability of machine learning algorithms to combat nonlinear fibre channel impairments and perform parameter extraction from directly detected signals....

  15. Exploring machine-learning-based control plane intrusion detection techniques in software defined optical networks

    Science.gov (United States)

    Zhang, Huibin; Wang, Yuqiao; Chen, Haoran; Zhao, Yongli; Zhang, Jie

    2017-12-01

    In software defined optical networks (SDON), the centralized control plane may encounter numerous intrusion threatens which compromise the security level of provisioned services. In this paper, the issue of control plane security is studied and two machine-learning-based control plane intrusion detection techniques are proposed for SDON with properly selected features such as bandwidth, route length, etc. We validate the feasibility and efficiency of the proposed techniques by simulations. Results show an accuracy of 83% for intrusion detection can be achieved with the proposed machine-learning-based control plane intrusion detection techniques.

  16. Moving object detection in video satellite image based on deep learning

    Science.gov (United States)

    Zhang, Xueyang; Xiang, Junhua

    2017-11-01

    Moving object detection in video satellite image is studied. A detection algorithm based on deep learning is proposed. The small scale characteristics of remote sensing video objects are analyzed. Firstly, background subtraction algorithm of adaptive Gauss mixture model is used to generate region proposals. Then the objects in region proposals are classified via the deep convolutional neural network. Thus moving objects of interest are detected combined with prior information of sub-satellite point. The deep convolution neural network employs a 21-layer residual convolutional neural network, and trains the network parameters by transfer learning. Experimental results about video from Tiantuo-2 satellite demonstrate the effectiveness of the algorithm.

  17. Adaptive E- Learning System Based on Personalized Learning Style

    African Journals Online (AJOL)

    pc

    2018-03-05

    Mar 5, 2018 ... motivation to this research is to improve the learner performance and achieve the ... valuable factor for enhancing learning process by adopting an effective .... Video. Reflective Intuitive. Primer Test. Verbal Sequential. Tutorial.

  18. The Effectiveness of the Gesture-Based Learning System (GBLS and Its Impact on Learning Experience

    Directory of Open Access Journals (Sweden)

    Moamer Ali Shakroum

    2016-06-01

    Full Text Available Several studies and experiments have been conducted in recent years to examine the value and the advantage of using the Gesture-Based Learning System (GBLS.The investigation of the influence of the GBLS mode on the learning outcomes is still scarce. Most previous studies did not address more than one category of learning outcomes (cognitive, affective outcomes, etc. at the same time when used to understand the impact of GBLS. Moreover, none of these studies considered the difference in students’ characteristics such as learning styles and spatial abilities. Therefore, a comprehensive empirical research on the impact of the GBLS mode on learning outcomes is needed. The purpose of this paper is to fill in the gap and to investigate the effectiveness of the GBLS mode on learning using Technology Mediated Learning (TML models. This study revealed that the GBLS mode has greater positive impact on students’ learning outcomes (cognitive and affective outcomes when compared with other two learning modes that are classified as Computer Simulation Software Learning (CSSL mode and conventional learning mode. In addition, this study also found that the GBLS mode is capable of serving all students with different learning styles and spatial ability levels. The results of this study revealed that the GBLS mode outperformed the existing learning methods by providing a unique learning experience that considers the differences between students. The results have also shown that the Kinect user interface can create an interactive and an enjoyable learning experience.

  19. Learning to Support Learning Together: An Experience with the Soft Systems Methodology

    Science.gov (United States)

    Sanchez, Adolfo; Mejia, Andres

    2008-01-01

    An action research approach called soft systems methodology (SSM) was used to foster organisational learning in a school regarding the role of the learning support department within the school and its relation with the normal teaching-learning activities. From an initial situation of lack of coordination as well as mutual misunderstanding and…

  20. Courseware Development with Animated Pedagogical Agents in Learning System to Improve Learning Motivation

    Science.gov (United States)

    Chin, Kai-Yi; Hong, Zeng-Wei; Huang, Yueh-Min; Shen, Wei-Wei; Lin, Jim-Min

    2016-01-01

    The addition of animated pedagogical agents (APAs) in computer-assisted learning (CAL) systems could successfully enhance students' learning motivation and engagement in learning activities. Conventionally, the APA incorporated multimedia materials are constructed through the cooperation of teachers and software programmers. However, the thinking…

  1. The Impacts of System and Human Factors on Online Learning Systems Use and Learner Satisfaction

    Science.gov (United States)

    Alshare, Khaled A.; Freeze, Ronald D.; Lane, Peggy L.; Wen, H. Joseph

    2011-01-01

    Success in an online learning environment is tied to both human and system factors. This study illuminates the unique contributions of human factors (comfort with online learning, self-management of learning, and perceived Web self-efficacy) to online learning system success, which is measured in terms of usage and satisfaction. The research model…

  2. Research progress in machine learning methods for gene-gene interaction detection.

    Science.gov (United States)

    Peng, Zhe-Ye; Tang, Zi-Jun; Xie, Min-Zhu

    2018-03-20

    Complex diseases are results of gene-gene and gene-environment interactions. However, the detection of high-dimensional gene-gene interactions is computationally challenging. In the last two decades, machine-learning approaches have been developed to detect gene-gene interactions with some successes. In this review, we summarize the progress in research on machine learning methods, as applied to gene-gene interaction detection. It systematically examines the principles and limitations of the current machine learning methods used in genome wide association studies (GWAS) to detect gene-gene interactions, such as neural networks (NN), random forest (RF), support vector machines (SVM) and multifactor dimensionality reduction (MDR), and provides some insights on the future research directions in the field.

  3. A Comprehensive Review and meta-analysis on Applications of Machine Learning Techniques in Intrusion Detection

    Directory of Open Access Journals (Sweden)

    Manojit Chattopadhyay

    2018-05-01

    Full Text Available Securing a machine from various cyber-attacks has been of serious concern for researchers, statutory bodies such as governments, business organizations and users in both wired and wireless media. However, during the last decade, the amount of data handling by any device, particularly servers, has increased exponentially and hence the security of these devices has become a matter of utmost concern. This paper attempts to examine the challenges in the application of machine learning techniques to intrusion detection. We review different inherent issues in defining and applying the machine learning techniques to intrusion detection. We also attempt to identify the best technological solution for changing usage pattern by comparing different machine learning techniques on different datasets and summarizing their performance using various performance metrics. This paper highlights the research challenges and future trends of intrusion detection in dynamic scenarios of intrusion detection problems in diverse network technologies.

  4. Detecting Mental States by Machine Learning Techniques: The Berlin Brain-Computer Interface

    Science.gov (United States)

    Blankertz, Benjamin; Tangermann, Michael; Vidaurre, Carmen; Dickhaus, Thorsten; Sannelli, Claudia; Popescu, Florin; Fazli, Siamac; Danóczy, Márton; Curio, Gabriel; Müller, Klaus-Robert

    The Berlin Brain-Computer Interface Brain-Computer Interface (BBCI) uses a machine learning approach to extract user-specific patterns from high-dimensional EEG-features optimized for revealing the user's mental state. Classical BCI applications are brain actuated tools for patients such as prostheses (see Section 4.1) or mental text entry systems ([1] and see [2-5] for an overview on BCI). In these applications, the BBCI uses natural motor skills of the users and specifically tailored pattern recognition algorithms for detecting the user's intent. But beyond rehabilitation, there is a wide range of possible applications in which BCI technology is used to monitor other mental states, often even covert ones (see also [6] in the fMRI realm). While this field is still largely unexplored, two examples from our studies are exemplified in Sections 4.3 and 4.4.

  5. DIDACTIC ENGINEERING: DESIGNING NEW GENERATION LEARNING SYSTEMS

    Directory of Open Access Journals (Sweden)

    Nail K. Nuriyev

    2016-09-01

    Full Text Available Introduction: the article deals with the organisation of training activities in the man-made environment. Didactic engineering is seen as a methodology within which problems of didactics are solved with application of pedagogical, psychological, engineering methods. It is obvious that in order to implement the training of future engineers in a competence-based format (according to educational standard a new type of teaching system is needed, with new capacities (properties. These systems should set each student towards the development of professionally significant (key abilities, taking into account his/her psychological characteristics; ensure training on the verge of permissible difficulties (developing training, and thereby achieve rapid development of key skills, through his/her zone of “immediate development”; to diagnose the quality of possession of a competence in the academic sense. For the objectivity and reliability of assessment of the level and depth of learned knowledge it is necessary to generate this evaluation in a metric format. As a result, we created a didactic system, which combines all the listed properties and the properties of classical systems. This allowed us to construct a new generation of didactic systems. Materials and Methods: the research is based on a systematic analysis of the activity of an engineer; on models of “zones of immediate development” by L. S. Vygotsky; on “developmental education” by L. N. Zankova; on the use of pedagogical and psychological patterns as well as taxonomic methods, didactic engineering, theory of probability and mathematical statistics. Results: constructed is a model for training engineers in the metric format of competence, which envisages a rapid development of students project and constructive abilit ies based on their knowledge learned. Discussion and Conclusions: the parameters defining the probability of engineer’s success have been described; the taxonomic scale

  6. Inclusive E-Learning - Towards an Integrated System Design.

    Science.gov (United States)

    Patzer, Yasmin; Pinkwart, Niels

    2017-01-01

    At first sight there seem to be issues combining technical accessibility guidelines and educational needs when designing inclusive E-Learning. Furthermore Universal Design for Learning seems to contradict individualization. In this paper we address both issues with an inclusive E-Learning design for the LAYA system, which targets disabled and non-disabled learners.

  7. Collaborative Learning in Advanced Supply Systems: The KLASS Pilot Project.

    Science.gov (United States)

    Rhodes, Ed; Carter, Ruth

    2003-01-01

    The Knowledge and Learning in Advanced Supply Systems (KLASS) project developed collaborative learning networks of suppliers in the British automotive and aerospace industries. Methods included face-to-face and distance learning, work toward National Vocational Qualifications, and diagnostic workshops for senior managers on improving quality,…

  8. Collaborative Learning Framework in Business Management Systems

    Directory of Open Access Journals (Sweden)

    Vladimir GRIGORE

    2008-01-01

    Full Text Available This paper presents a solution based on collaboration with experts and practitioner from university and ERP companies involved in process learning by training and learning by working. The solution uses CPI test to establish proper team for framework modules: Real-Time Chat Room, Discussion Forum, E-mail Support and Learning through Training. We define novice, practitioner and expert competence level based on CORONET train methodology. ERP companies have own roles for mentoring services to knowledge workers and evaluate the performance of learning process with teachers’ cooperation in learning by teaching and learning by working module.

  9. Intelligent e-Learning Systems: An Educational Paradigm Shift

    Directory of Open Access Journals (Sweden)

    Suman Bhattacharya

    2016-12-01

    Full Text Available Learning is the long process of transforming information as well as experience into knowledge, skills, attitude and behaviors. To make up the wide gap between the demand of increasing higher education and comparatively limited resources, more and more educational institutes are looking into instructional technology. Use of online resources not only reduces the cost of education but also meet the needs of society. Intelligent e-learning has become one of the important channels to reach out to students exceeding geographic boundaries. Besides this, the characteristics of e-learning have complicated the process of education, and have brought challenges to both instructors and students. This paper will focus on the discussion of different discipline of intelligent e-learning like scaffolding based e-learning, personalized e-learning, confidence based e-learning, intelligent tutoring system, etc. to illuminate the educational paradigm shift in intelligent e-learning system.

  10. System Quality Characteristics for Selecting Mobile Learning Applications

    Directory of Open Access Journals (Sweden)

    Mohamed SARRAB

    2015-10-01

    Full Text Available The majority of M-learning (Mobile learning applications available today are developed for the formal learning and education environment. These applications are characterized by the improvement in the interaction between learners and instructors to provide high interaction and flexibility to the learning process. M-learning is gaining increased recognition and adoption by different organizations. With the high number of M-learning applications available today, making the right decision about which, application to choose can be quite challenging. To date there is no complete and well defined set of system characteristics for such M-learning applications. This paper presents system quality characteristics for selecting M-learning applications based on the result of a systematic review conducted in this domain.

  11. Teach Them How They Learn: Learning Styles and Information Systems Education

    Science.gov (United States)

    Cegielski, Casey G.; Hazen, Benjamin T.; Rainer, R. Kelly

    2011-01-01

    The rich, interdisciplinary tradition of learning styles is markedly absent in information systems-related research. The current study applies the framework of learning styles to a common educational component of many of today's information systems curricula--object-oriented systems development--in an effort to answer the question as to whether…

  12. Detection of eardrum abnormalities using ensemble deep learning approaches

    Science.gov (United States)

    Senaras, Caglar; Moberly, Aaron C.; Teknos, Theodoros; Essig, Garth; Elmaraghy, Charles; Taj-Schaal, Nazhat; Yua, Lianbo; Gurcan, Metin N.

    2018-02-01

    In this study, we proposed an approach to report the condition of the eardrum as "normal" or "abnormal" by ensembling two different deep learning architectures. In the first network (Network 1), we applied transfer learning to the Inception V3 network by using 409 labeled samples. As a second network (Network 2), we designed a convolutional neural network to take advantage of auto-encoders by using additional 673 unlabeled eardrum samples. The individual classification accuracies of the Network 1 and Network 2 were calculated as 84.4%(+/- 12.1%) and 82.6% (+/- 11.3%), respectively. Only 32% of the errors of the two networks were the same, making it possible to combine two approaches to achieve better classification accuracy. The proposed ensemble method allows us to achieve robust classification because it has high accuracy (84.4%) with the lowest standard deviation (+/- 10.3%).

  13. Evaluating Recommender Systems for Technology Enhanced Learning: A Quantitative Survey

    Science.gov (United States)

    Erdt, Mojisola; Fernandez, Alejandro; Rensing, Christoph

    2015-01-01

    The increasing number of publications on recommender systems for Technology Enhanced Learning (TEL) evidence a growing interest in their development and deployment. In order to support learning, recommender systems for TEL need to consider specific requirements, which differ from the requirements for recommender systems in other domains like…

  14. A Simple and Effective Remedial Learning System with a Fuzzy Expert System

    Science.gov (United States)

    Lin, C.-C.; Guo, K.-H.; Lin, Y.-C.

    2016-01-01

    This study aims at implementing a simple and effective remedial learning system. Based on fuzzy inference, a remedial learning material selection system is proposed for a digital logic course. Two learning concepts of the course have been used in the proposed system: number systems and combinational logic. We conducted an experiment to validate…

  15. Mlifdect: Android Malware Detection Based on Parallel Machine Learning and Information Fusion

    Directory of Open Access Journals (Sweden)

    Xin Wang

    2017-01-01

    Full Text Available In recent years, Android malware has continued to grow at an alarming rate. More recent malicious apps’ employing highly sophisticated detection avoidance techniques makes the traditional machine learning based malware detection methods far less effective. More specifically, they cannot cope with various types of Android malware and have limitation in detection by utilizing a single classification algorithm. To address this limitation, we propose a novel approach in this paper that leverages parallel machine learning and information fusion techniques for better Android malware detection, which is named Mlifdect. To implement this approach, we first extract eight types of features from static analysis on Android apps and build two kinds of feature sets after feature selection. Then, a parallel machine learning detection model is developed for speeding up the process of classification. Finally, we investigate the probability analysis based and Dempster-Shafer theory based information fusion approaches which can effectively obtain the detection results. To validate our method, other state-of-the-art detection works are selected for comparison with real-world Android apps. The experimental results demonstrate that Mlifdect is capable of achieving higher detection accuracy as well as a remarkable run-time efficiency compared to the existing malware detection solutions.

  16. MOODLE – COMPUTERIZED DISTANT LEARNING SUPPORT SYSTEM

    Directory of Open Access Journals (Sweden)

    V. A. Tunda

    2014-01-01

    Full Text Available Moodle (Modular Object-Oriented Dynamic Learning Environment is the system developed by the English-speaking community around the world for more than 10 years, supports both, offline and online training. In most cases, Moodle is used to support and connect learning face-to-face with online training, as well as with other types of learning. Moodle allows you to ask all sorts of questions and assess responses in a variety of ways. The central concept of Moodle is a course in which one or more teachers offer students resources (such as files, folders, Web pages and participate in interactive activities (such as forums, wikis, blogging, lessons, seminars, assignments, examinations. Students and teachers can change roles, mutually assess each other, share knowledge on the topics of study in the glossary and database system.In the English-speaking world, there are public sites with detailed documentation on Moodle, which is constantly verified and modified in accordance with emerging new versions of the system; Russian versions do not have public sites with detailed translation of the English-language documentation. On what and to whom you want to perform actions in the daily practice of the modern versions, only the barest of outlines on paid seminars and presentations to those who started the implementation of Moodle in your organization are given. And this despite the fact that the system has a great variety of Moodle (over 500 settings with different levels of functionality to be performed, to maintain and develop specially organized by the team of specialists. The question is not only about creating a training course in Moodle. It's about maintaining the health of the system within the institution, such as a University. Under the "health maintenance" means: availability and preservation of up-to-date documentation on system and manuals on Moodle separately for administrators, managers, teachers and students at their level, training consultants

  17. The Role of a Commander in Military Lessons Learned Systems

    Directory of Open Access Journals (Sweden)

    Zenon Waliński

    2015-06-01

    Full Text Available The aim of the paper is to investigate the role of a commander in military Lessons Learned systems. In order to achieve the aim, the paper presents (1 the architecture of the Lessons Learned capabilities in the U.S. Army, NATO and the Polish Armed Forces, (2 the commander’s role in the Lessons Learned process (3 the commander’s role in fostering Lessons Learned organisation culture. The paper is based on multiple case study analysis including Lessons Learned systems in NATO, the U.S. Army and the Polish Armed Forces.

  18. Intensitas Perilaku Pengguna E-learning System dengan Model Utaut

    OpenAIRE

    Sari, Fatma; Purnamasari, Susan Dian

    2013-01-01

    This study aims to determine behavioral intention in the use of e-learning system using models UTAUT. The phenomenon underlying the research is: It is not yet optimal use of e-learning by students information systems in the learning process, not yet optimal socialization of the existence of e-learning, so that is not maximized and yet utilization measurability of the impact of using e-learning for lecturers.This study is limited in its scope: analysis of the influence of performance expectanc...

  19. ABOUT SYSTEM OF DISTANCE LEARNING IN OPEN ONLINE COURSE

    Directory of Open Access Journals (Sweden)

    V.М. Kukharenko

    2012-03-01

    Full Text Available This paper describes the first part of the open online course "E-Learning from A to Z", dedicated to the creation and development of system of distance learning (university or corporation. The results of the learning process and discussion on the workshop at NTU "KPI" in 2012 is shown the interest of teachers in a new form of online course and lack of development of personal learning environment. The open online courses can contribute to society practice.

  20. Implementation of standards within eLearning information systems

    OpenAIRE

    Roman Malo

    2007-01-01

    Nowadays, eLearning standards' support within eLearning systems is much discussed problem. In this problem domain especially the reference model SCORM must be considered. This de-facto standard is a package of common standards and specifications used for the standardization of eLearning activities as eLearning content preparation, using e-course, communication etc. Implementation of standards itself is a process with great difficulty and time requests. Interesting and considerable approach to...