A class of robust estimates for detection in hyperspectral images using elliptical distributions background

When dealing with impulsive background echoes, Gaussian model is no longer pertinent. We study in this paper the class of elliptically contoured (EC) distributions. They provide a multivariate location-scatter family of distributions that primarily serve as long tailed alternatives to the multivariate normal model. They are proven to represent a more accurate characterization of HSI data than models based on the multivariate Gaussian assumption. For data in ℝk, robust proposals for the sample covariance estimate are the M-estimators. We have also analyzed the performance of an adaptive non- Gaussian detector built with these improved estimators. Constant False Alarm Rate (CFAR) is pursued to allow the detector independence of nuisance parameters and false alarm regulation.

[1]  Dimitris G. Manolakis,et al.  Detection algorithms for hyperspectral imaging applications , 2002, IEEE Signal Process. Mag..

[2]  Philippe Forster,et al.  Covariance Structure Maximum-Likelihood Estimates in Compound Gaussian Noise: Existence and Algorithm Analysis , 2008, IEEE Transactions on Signal Processing.

[3]  Philippe Forster,et al.  Theoretical analysis of an improved covariance matrix estimator in non-Gaussian noise [radar detection applications] , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[4]  Louis L. Scharf,et al.  Adaptive subspace detectors , 2001, IEEE Trans. Signal Process..

[5]  Giuseppe Ricci,et al.  Recursive estimation of the covariance matrix of a compound-Gaussian process and its application to adaptive CFAR detection , 2002, IEEE Trans. Signal Process..

[6]  E. Ollila,et al.  Influence functions for array covariance matrix estimators , 2004, IEEE Workshop on Statistical Signal Processing, 2003.

[7]  E. J. Kelly.  An Adaptive Detection Algorithm , 1986, IEEE Transactions on Aerospace and Electronic Systems.

[8]  S Matteoli,et al.  A tutorial overview of anomaly detection in hyperspectral images , 2010, IEEE Aerospace and Electronic Systems Magazine.

[9]  David E. Tyler.  A Distribution-Free $M$-Estimator of Multivariate Scatter , 1987 .

[10]  David Declercq,et al.  BORD: bayesian optimum radar detector , 2003, Signal Process..

[11]  Fulvio Gini,et al.  Covariance matrix estimation for CFAR detection in correlated heavy tailed clutter , 2002, Signal Process..

[12]  F. Gini.  Sub-optimum coherent radar detection in a mixture of K-distributed and Gaussian clutter , 1997 .

[13]  Sze Kim Pang,et al.  Robust detection using the SIRV background modelling for hyperspectral imaging , 2011, 2011 IEEE International Geoscience and Remote Sensing Symposium.

[14]  Frederick R. Forst,et al.  On robust estimation of the location parameter , 1980 .

[15]  Philippe Forster,et al.  Performance Analysis of Covariance Matrix Estimates in Impulsive Noise , 2008, IEEE Transactions on Signal Processing.

[16]  R. Maronna.  Robust $M$-Estimators of Multivariate Location and Scatter , 1976 .