scholarly journals Hyperspectral Anomaly Detection: Comparative Evaluation in Scenes with Diverse Complexity

2012 ◽  
Vol 2012 ◽  
pp. 1-16 ◽  
Author(s):  
Dirk Borghys ◽  
Ingebjørg Kåsen ◽  
Véronique Achard ◽  
Christiaan Perneel
Keyword(s):  
Anomaly Detection ◽  
Hyperspectral Data ◽  
Data Cube ◽  
Detection Methods ◽  
Multivariate Normal ◽  
Normal Probability ◽  
Local Methods ◽  
Anomaly Detector ◽  
The Difference ◽  
Rx Detector

Anomaly detection (AD) in hyperspectral data has received a lot of attention for various applications. The aim of anomaly detection is to detect pixels in the hyperspectral data cube whose spectra differ significantly from the background spectra. Many anomaly detectors have been proposed in the literature. They differ in the way the background is characterized and in the method used for determining the difference between the current pixel and the background. The most well-known anomaly detector is the RX detector that calculates the Mahalanobis distance between the pixel under test (PUT) and the background. Global RX characterizes the background of the complete scene by a single multivariate normal probability density function. In many cases, this model is not appropriate for describing the background. For that reason a variety of other anomaly detection methods have been developed. This paper examines three classes of anomaly detectors: subspace methods, local methods, and segmentation-based methods. Representative examples of each class are chosen and applied on a set of hyperspectral data with diverse complexity. The results are evaluated and compared.

scholarly journals HYPERSPECTRAL ANOMALY DETECTION IN URBAN SCENARIOS

2016 ◽  
Vol XLI-B7 ◽  
pp. 111-116 ◽  
Author(s):  
J. G. Rejas Ayuga ◽  
R. Martínez Marín ◽  
M. Marchamalo Sacristán ◽  
J. Bonatti ◽  
J. C. Ojeda
Keyword(s):  
Anomaly Detection ◽  
Near Infrared ◽  
Hyperspectral Data ◽  
Detection Methods ◽  
San Jose ◽  
First Results ◽  
Anomaly Detector ◽  
A New Technique ◽  
Spectral Ranges ◽  
Shortwave Infrared

We have studied the spectral features of reflectance and emissivity in the pattern recognition of urban materials in several single hyperspectral scenes through a comparative analysis of anomaly detection methods and their relationship with city surfaces with the aim to improve information extraction processes. Spectral ranges of the visible-near infrared (VNIR), shortwave infrared (SWIR) and thermal infrared (TIR) from hyperspectral data cubes of AHS sensor and HyMAP and MASTER of two cities, Alcalá de Henares (Spain) and San José (Costa Rica) respectively, have been used. <br><br> In this research it is assumed no prior knowledge of the targets, thus, the pixels are automatically separated according to their spectral information, significantly differentiated with respect to a background, either globally for the full scene, or locally by image segmentation. Several experiments on urban scenarios and semi-urban have been designed, analyzing the behaviour of the standard RX anomaly detector and different methods based on subspace, image projection and segmentation-based anomaly detection methods. A new technique for anomaly detection in hyperspectral data called DATB (Detector of Anomalies from Thermal Background) based on dimensionality reduction by projecting targets with unknown spectral signatures to a background calculated from thermal spectrum wavelengths is presented. First results and their consequences in non-supervised classification and extraction information processes are discussed.

scholarly journals HYPERSPECTRAL ANOMALY DETECTION IN URBAN SCENARIOS

2016 ◽  
Vol XLI-B7 ◽  
pp. 111-116
Author(s):  
J. G. Rejas Ayuga ◽  
R. Martínez Marín ◽  
M. Marchamalo Sacristán ◽  
J. Bonatti ◽  
J. C. Ojeda
Keyword(s):  
Anomaly Detection ◽  
Near Infrared ◽  
Hyperspectral Data ◽  
Detection Methods ◽  
San Jose ◽  
First Results ◽  
Anomaly Detector ◽  
A New Technique ◽  
Spectral Ranges ◽  
Shortwave Infrared

We have studied the spectral features of reflectance and emissivity in the pattern recognition of urban materials in several single hyperspectral scenes through a comparative analysis of anomaly detection methods and their relationship with city surfaces with the aim to improve information extraction processes. Spectral ranges of the visible-near infrared (VNIR), shortwave infrared (SWIR) and thermal infrared (TIR) from hyperspectral data cubes of AHS sensor and HyMAP and MASTER of two cities, Alcalá de Henares (Spain) and San José (Costa Rica) respectively, have been used. <br><br> In this research it is assumed no prior knowledge of the targets, thus, the pixels are automatically separated according to their spectral information, significantly differentiated with respect to a background, either globally for the full scene, or locally by image segmentation. Several experiments on urban scenarios and semi-urban have been designed, analyzing the behaviour of the standard RX anomaly detector and different methods based on subspace, image projection and segmentation-based anomaly detection methods. A new technique for anomaly detection in hyperspectral data called DATB (Detector of Anomalies from Thermal Background) based on dimensionality reduction by projecting targets with unknown spectral signatures to a background calculated from thermal spectrum wavelengths is presented. First results and their consequences in non-supervised classification and extraction information processes are discussed.

scholarly journals A Hybrid Semi-Supervised Anomaly Detection Model for High-Dimensional Data

2017 ◽  
Vol 2017 ◽  
pp. 1-9 ◽  
Author(s):  
Hongchao Song ◽  
Zhuqing Jiang ◽  
Aidong Men ◽  
Bo Yang
Keyword(s):  
Anomaly Detection ◽  
Nearest Neighbor ◽  
Dimensional Space ◽  
High Dimensional Data ◽  
Real Life ◽  
Detection Methods ◽  
High Dimensional ◽  
Detection Accuracy ◽  
Detection Model ◽  
Anomaly Detector

Anomaly detection, which aims to identify observations that deviate from a nominal sample, is a challenging task for high-dimensional data. Traditional distance-based anomaly detection methods compute the neighborhood distance between each observation and suffer from the curse of dimensionality in high-dimensional space; for example, the distances between any pair of samples are similar and each sample may perform like an outlier. In this paper, we propose a hybrid semi-supervised anomaly detection model for high-dimensional data that consists of two parts: a deep autoencoder (DAE) and an ensemble k-nearest neighbor graphs- (K-NNG-) based anomaly detector. Benefiting from the ability of nonlinear mapping, the DAE is first trained to learn the intrinsic features of a high-dimensional dataset to represent the high-dimensional data in a more compact subspace. Several nonparametric KNN-based anomaly detectors are then built from different subsets that are randomly sampled from the whole dataset. The final prediction is made by all the anomaly detectors. The performance of the proposed method is evaluated on several real-life datasets, and the results confirm that the proposed hybrid model improves the detection accuracy and reduces the computational complexity.

Study on Anomaly Detection Methods in Hyperspectral Image

2014 ◽  
Vol 631-632 ◽  
pp. 631-635
Author(s):  
Yi Ting Wang ◽  
Shi Qi Huang ◽  
Hong Xia Wang ◽  
Dai Zhi Liu
Keyword(s):  
Anomaly Detection ◽  
Target Detection ◽  
Hyperspectral Image ◽  
Detection Methods ◽  
Imaging Data ◽  
Detection Algorithms ◽  
Sensing Technology ◽  
Anomaly Detector ◽  
And Performance ◽  
Spectral Diagnosis

Hyperspectral remote sensing technology can be used to make a correct spectral diagnosis on substances. So it is widely used in the field of target detection and recognition. However, it is very difficult to gather accurate prior information for target detect since the spectral uncertainty of objects is pervasive in existence. An anomaly detector can enable one to detect targets whose signatures are spectrally distinct from their surroundings with no prior knowledge. It becomes a focus in the field of target detection. Therefore, we study four anomaly detection algorithms and conclude with empirical results that use hyperspectral imaging data to illustrate the operation and performance of various detectors.

scholarly journals Unsupervised Anomaly Detection Approach for Time-Series in Multi-Domains Using Deep Reconstruction Error

Symmetry ◽  
2020 ◽  
Vol 12 (8) ◽  
pp. 1251 ◽  
Author(s):  
Tsatsral Amarbayasgalan ◽  
Van Huy Pham ◽  
Nipon Theera-Umpon ◽  
Keun Ho Ryu
Keyword(s):  
Time Series ◽  
Anomaly Detection ◽  
Real Time ◽  
Reconstruction Error ◽  
Ar Model ◽  
Detection Methods ◽  
Series Data ◽  
Detection Approach ◽  
Unsupervised Anomaly Detection ◽  
Anomaly Detector

Automatic anomaly detection for time-series is critical in a variety of real-world domains such as fraud detection, fault diagnosis, and patient monitoring. Current anomaly detection methods detect the remarkably low proportion of the actual abnormalities correctly. Furthermore, most of the datasets do not provide data labels, and require unsupervised approaches. By focusing on these problems, we propose a novel deep learning-based unsupervised anomaly detection approach (RE-ADTS) for time-series data, which can be applicable to batch and real-time anomaly detections. RE-ADTS consists of two modules including the time-series reconstructor and anomaly detector. The time-series reconstructor module uses the autoregressive (AR) model to find an optimal window width and prepares the subsequences for further analysis according to the width. Then, it uses a deep autoencoder (AE) model to learn the data distribution, which is then used to reconstruct a time-series close to the normal. For anomalies, their reconstruction error (RE) was higher than that of the normal data. As a result of this module, RE and compressed representation of the subsequences were estimated. Later, the anomaly detector module defines the corresponding time-series as normal or an anomaly using a RE based anomaly threshold. For batch anomaly detection, the combination of the density-based clustering technique and anomaly threshold is employed. In the case of real-time anomaly detection, only the anomaly threshold is used without the clustering process. We conducted two types of experiments on a total of 52 publicly available time-series benchmark datasets for the batch and real-time anomaly detections. Experimental results show that the proposed RE-ADTS outperformed the state-of-the-art publicly available anomaly detection methods in most cases.

scholarly journals NOVEL HYPERSPECTRAL ANOMALY DETECTION METHODS BASED ON UNSUPERVISED NEAREST REGULARIZED SUBSPACE

2018 ◽  
Vol XLII-3 ◽  
pp. 539-546
Author(s):  
Z. Hou ◽  
Y. Chen ◽  
K. Tan ◽  
P. Du
Keyword(s):  
Anomaly Detection ◽  
Spatial Information ◽  
Hyperspectral Imagery ◽  
Detection Methods ◽  
Detection Accuracy ◽  
Outlier Removal ◽  
Residual Image ◽  
Affect Detection ◽  
Anomaly Detector ◽  
Dual Window

Anomaly detection has been of great interest in hyperspectral imagery analysis. Most conventional anomaly detectors merely take advantage of spectral and spatial information within neighboring pixels. In this paper, two methods of Unsupervised Nearest Regularized Subspace-based with Outlier Removal Anomaly Detector (UNRSORAD) and Local Summation UNRSORAD (LSUNRSORAD) are proposed, which are based on the concept that each pixel in background can be approximately represented by its spatial neighborhoods, while anomalies cannot. Using a dual window, an approximation of each testing pixel is a representation of surrounding data via a linear combination. The existence of outliers in the dual window will affect detection accuracy. Proposed detectors remove outlier pixels that are significantly different from majority of pixels. In order to make full use of various local spatial distributions information with the neighboring pixels of the pixels under test, we take the local summation dual-window sliding strategy. The residual image is constituted by subtracting the predicted background from the original hyperspectral imagery, and anomalies can be detected in the residual image. Experimental results show that the proposed methods have greatly improved the detection accuracy compared with other traditional detection method.

Un-Supervised approach to backorder prediction using deep autoencoder

2019 ◽  
Vol 12 ◽  
Author(s):  
Gunjan Saraogi ◽  
Deepa Gupta ◽  
Lavanya Sharma ◽  
Ajay Rana
Keyword(s):  
Anomaly Detection ◽  
Predictive Model ◽  
Inventory Management ◽  
Supervised Classification ◽  
Operating Characteristic ◽  
Imbalanced Data ◽  
Bullwhip Effect ◽  
Detection Methods ◽  
Good Representation ◽  
Operational Management

Background: Backorders are an accepted abnormality affecting accumulation alternation and logistics, sales, chump service, and manufacturing, which generally leads to low sales and low chump satisfaction. A predictive archetypal can analyse which articles are best acceptable to acquaintance backorders giving the alignment advice and time to adjust, thereby demography accomplishes to aerate their profit. Objective: To address the issue of predicting backorders, this paper has proposed an un-supervised approach to backorder prediction using Deep Autoencoder. Method: In this paper, artificial intelligence paradigms are researched in order to introduce a predictive model for the present unbalanced data issues, where the number of products going on backorder is rare. Result: Un-supervised anomaly detection using deep auto encoders has shown better Area under the Receiver Operating Characteristic and precision-recall curves than supervised classification techniques employed with resampling techniques for imbalanced data problems. Conclusion: We demonstrated that Un-supervised anomaly detection methods specifically deep auto-encoders can be used to learn a good representation of the data. The method can be used as predictive model for inventory management and help to reduce bullwhip effect, raise customer satisfaction as well as improve operational management in the organization. This technology is expected to create the sentient supply chain of the future – able to feel, perceive and react to situations at an extraordinarily granular level

scholarly journals OutlierNets: Highly Compact Deep Autoencoder Network Architectures for On-Device Acoustic Anomaly Detection

Sensors ◽  
2021 ◽  
Vol 21 (14) ◽  
pp. 4805
Author(s):  
Saad Abbasi ◽  
Mahmoud Famouri ◽  
Mohammad Javad Shafiee ◽  
Alexander Wong
Keyword(s):  
Machine Learning ◽  
Anomaly Detection ◽  
Detection Methods ◽  
Detection Accuracy ◽  
Network Architectures ◽  
Design Exploration ◽  
Convolutional Autoencoder ◽  
Acoustic Anomaly ◽  
Human Operators ◽  
Computational Resources

Human operators often diagnose industrial machinery via anomalous sounds. Given the new advances in the field of machine learning, automated acoustic anomaly detection can lead to reliable maintenance of machinery. However, deep learning-driven anomaly detection methods often require an extensive amount of computational resources prohibiting their deployment in factories. Here we explore a machine-driven design exploration strategy to create OutlierNets, a family of highly compact deep convolutional autoencoder network architectures featuring as few as 686 parameters, model sizes as small as 2.7 KB, and as low as 2.8 million FLOPs, with a detection accuracy matching or exceeding published architectures with as many as 4 million parameters. The architectures are deployed on an Intel Core i5 as well as a ARM Cortex A72 to assess performance on hardware that is likely to be used in industry. Experimental results on the model’s latency show that the OutlierNet architectures can achieve as much as 30x lower latency than published networks.

Elimination of outlier measurements for damage detection of structures under changing environmental conditions

2021 ◽  
pp. 147592172199847
Author(s):  
William Soo Lon Wah ◽  
Yining Xia
Keyword(s):  
Damage Detection ◽  
Multiple Regression ◽  
Natural Frequencies ◽  
Detection Method ◽  
Detection Methods ◽  
Structure Model ◽  
Bridge Structure ◽  
Beam Structure ◽  
Dependent Variables ◽  
The Difference

Damage detection methods developed in the literature are affected by the presence of outlier measurements. These measurements can prevent small levels of damage to be detected. Therefore, a method to eliminate the effects of outlier measurements is proposed in this article. The method uses the difference in fits to examine how deleting an observation affects the predicted value of a model. This allows the observations that have a large influence on the model created, to be identified. These observations are the outlier measurements and they are eliminated from the database before the application of damage detection methods. Eliminating the outliers before the application of damage detection methods allows the normal procedures to detect damage, to be implemented. A multiple-regression-based damage detection method, which uses the natural frequencies as both the independent and dependent variables, is also developed in this article. A beam structure model and an experimental wooden bridge structure are analysed using the multiple-regression-based damage detection method with and without the application of the method proposed to eliminate the effects of outliers. The results obtained demonstrate that smaller levels of damage can be detected when the effects of outlier measurements are eliminated using the method proposed in this article.

Sign in / Sign up

Export Citation Format

Share Document