Hyperspectral Anomaly Detection With Kurtosis-Driven Local Covariance Matrix Corruption Mitigation
作者： Stefania Matteoli , Marco Diani , Giovanni Corsini
DOI: 10.1109/LGRS.2010.2090337
关键词:
摘要: Local background covariance matrix corruption due to outliers in the sample data may be one of major causes that limit detection performance those algorithms detect local anomalies hyperspectral images on basis Mahalanobis distance. In this letter, an original scheme is presented efficiently embeds mitigation. A kurtosis-based binary hypothesis test first applied each pixel quickly determine presence neighborhood. Rejection null triggers application a robust-to-outlier estimation technique. Results real exhibit good and robustness outliers. Contrary previous works, achieved without unnecessary increase procedural complexity.
sci-hub.se 参考文章(16)
Katrien Van Driessen, Peter J. Rousseeuw, A fast algorithm for the minimum covariance determinant estimator Technometrics. ,vol. 41, pp. 212- 223 ,(1999) , 10.2307/1270566
Edisanter Lo, John Ingram, Hyperspectral Anomaly Detection Based on Minimum Generalized Variance Method Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XIV. ,vol. 6966, pp. 696603- ,(2008) , 10.1117/12.778929
Heesung Kwon, Sandor Z Der, Nasser M Nasrabadi, Adaptive anomaly detection using subspace separation for hyperspectral imagery Optical Engineering. ,vol. 42, pp. 3342- 3351 ,(2003) , 10.1117/1.1614265
E. Madar, O. Kuybeda, D. Malah, M. Barzohar, Local-global background modeling for anomaly detection in hyperspectral images workshop on hyperspectral image and signal processing: evolution in remote sensing. pp. 1- 4 ,(2009) , 10.1109/WHISPERS.2009.5289036
Stefania Matteoli, Marco Diani, Giovanni Corsini, Improved estimation of local background covariance matrix for anomaly detection in hyperspectral images Optical Engineering. ,vol. 49, pp. 046201- ,(2010) , 10.1117/1.3386069
I.S. Reed, X. Yu, Adaptive multiple-band CFAR detection of an optical pattern with unknown spectral distribution IEEE Transactions on Acoustics, Speech, and Signal Processing. ,vol. 38, pp. 1760- 1770 ,(1990) , 10.1109/29.60107
D. Manolakis, G. Shaw, Detection algorithms for hyperspectral imaging applications IEEE Signal Processing Magazine. ,vol. 19, pp. 29- 43 ,(2002) , 10.1109/79.974724
Timothy E. Smetek, Kenneth W. Bauer, Finding Hyperspectral Anomalies Using Multivariate Outlier Detection ieee aerospace conference. pp. 1- 24 ,(2007) , 10.1109/AERO.2007.353062
F. J. ANSCOMBE, WILLIAM J. GLYNN, Distribution of the Kurtosis Statistic b2 for Normal Samples. Biometrika. ,vol. 70, pp. 227- 234 ,(1983) , 10.1093/BIOMET/70.1.227
RALPH D'AGOSTINO, E. S. PEARSON, Tests for departure from normality. Empirical results for the distributions of b2 and √b1 Biometrika. ,vol. 60, pp. 613- 622 ,(1973) , 10.1093/BIOMET/60.3.613