Using cross-entropy for polarimetric SAR image classification

Author

Xu, Jun-Yi ; Yang, Jian ; Peng, Ying-Ning ; Wang, Chao

Author_Institution

Dept. of Electron. Eng., Tsinghua Univ., Beijing, China

Volume

3

fYear

2002

fDate

24-28 June 2002

Firstpage

1917

Abstract

In this paper, the concept of cross-entropy is introduced for measuring the difference between two scatterers in polarimetric SAR image classification. The difference between two scatterers is composed of three parts, i.e., the difference of average scattering mechanism, the difference of scattering randomness and the difference of the two spans of scattering matrices. The minimum cross-entropy principle is adopted to make the classification decision. The capability of this method is demonstrated by the experimental results.

Keywords

entropy codes; geophysical signal processing; geophysical techniques; image classification; radar imaging; radar polarimetry; remote sensing by radar; synthetic aperture radar; terrain mapping; SAR; cross entropy method; geophysical measurement technique; image classification; land surface; minimum cross-entropy principle; polarimetric SAR method; radar imaging; radar polarimetry; radar remote sensing; randomness; scattering mechanism; synthetic aperture radar; terrain mapping; Chaos; Covariance matrix; Eigenvalues and eigenfunctions; Gaussian distribution; Hydrogen; Image classification; Pixel; Radar scattering; Remote sensing; Robustness;

fLanguage

English

Publisher

ieee

Conference_Titel

Geoscience and Remote Sensing Symposium, 2002. IGARSS '02. 2002 IEEE International

Print_ISBN

0-7803-7536-X

Type

conf

DOI

10.1109/IGARSS.2002.1026299

Filename

1026299