Comparative Analysis of Covariance Matrix Estimation for Anomaly Detection in Hyperspectral Images

被引:12
作者
Velasco-Forero, Santiago [1 ]
Chen, Marcus [2 ]
Goh, Alvina [2 ]
Pang, Sze Kim [3 ]
机构
[1] PSL Res Univ, MINES Paristech, CMM, F-75006 Paris, France
[2] Nanyang Technol Univ, Dept Math, Singapore 639798, Singapore
[3] DSO Natl Labs, Singapore 118230, Singapore
关键词
Remote sensing; hyperspectral imaging; covariance matrices; CLASSIFICATION; NUMBER;
D O I
10.1109/JSTSP.2015.2442213
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Covariance matrix estimation is fundamental for anomaly detection, especially for the Reed and Xiaoli Yu (RX) detector. Anomaly detection is challenging in hyperspectral images because the data has a high correlation among dimensions, heavy tailed distributions and multiple clusters. This paper comparatively evaluates modern techniques of covariance matrix estimation based on the performance and volume the RX detector. To address the different challenges, experiments were designed to systematically examine the robustness and effectiveness of various estimation techniques. In the experiments, three factors were considered, namely, sample size, outlier size, and modification in the distribution of the sample.
引用
收藏
页码:1061 / 1073
页数:13
相关论文
共 66 条
[1]  
Abramovich Y. I., 2007, P IEEEE ICASSP, V3, P1111
[2]   Regularized Covariance Matrix Estimation in Complex Elliptically Symmetric Distributions Using the Expected Likelihood Approach-Part 1: The Over-Sampled Case [J].
Abramovich, Yuri I. ;
Besson, Olivier .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2013, 61 (23) :5807-5818
[3]  
AITCHISON J, 1982, J ROY STAT SOC B, V44, P139
[4]  
Anderson G. W., 2010, An introduction to random matrices
[5]  
Anderson T.W., 1958, An introduction to multivariate statistical analysis, V2
[6]  
[Anonymous], INT J NEUROPSYCHOPHA
[7]  
[Anonymous], 1936, P NATL I SCI INDIA, DOI DOI 10.1007/S13171-019-00164-5
[8]  
[Anonymous], P IEEE 3 WORKSH HYP
[9]   Geodesic Paths for Time-Dependent Covariance Matrices in a Riemannian Manifold [J].
Ben-David, Avishai ;
Marks, Justin .
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2014, 11 (09) :1499-1503
[10]   Eigenvalue Estimation of Hyperspectral Wishart Covariance Matrices From Limited Number of Samples [J].
Ben-David, Avishai ;
Davidson, Charles E. .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2012, 50 (11) :4384-4396