Title :
Statistical assessment of eigenvector-based target decomposition theorems in radar polarimetry
Author :
Lopez-Martinez, Carlos ; Pottier, Eric
Author_Institution :
IETR Lab., Rennes I Univ.
Abstract :
This paper concerns the analytical study of the eigen decomposition of hermitian, positive semidefinite matrices applied to PolSAR data analysis. Based on the Gaussian scattering assumption for multidimensional SAR data, the joint distribution of the sample eigenvalues of the coherency, or covariance, matrices is derived for a general case. The distribution is particularized for PolSAR data, and the moments of the sample eigenvalues, the entropy (H), and the anisotropy (A) are analyzed
Keywords :
Gaussian distribution; Hermitian matrices; eigenvalues and eigenfunctions; geophysical signal processing; radar polarimetry; synthetic aperture radar; Gaussian scattering; Hermitian matrices; PolSAR data analysis; eigenvector-based target decomposition theorem; entropy-anisotropy analysis; multidimensional SAR data; polarimetry synthetic aperture radar; radar polarimetry; statistical assessment; Anisotropic magnetoresistance; Covariance matrix; Eigenvalues and eigenfunctions; Entropy; Matrix decomposition; Multidimensional systems; Radar polarimetry; Radar scattering; Remote sensing; Speckle;
Conference_Titel :
Geoscience and Remote Sensing Symposium, 2004. IGARSS '04. Proceedings. 2004 IEEE International
Conference_Location :
Anchorage, AK
Print_ISBN :
0-7803-8742-2
DOI :
10.1109/IGARSS.2004.1368992
