POLARIMETRIC SAR IMAGE DECOMPOSITION FOR NON-REFLECTION SYMMETRY CONDITION

Yamaguchi Yoshio,Motoi Ishido,Moriyama Toshifumi,Jian Yang,Hiroyoshi Yamada
2005-01-01
IEICE Proceeding Series
Abstract:Terrain and land use classification is one of the most important applications of Polarimetric Synthetic Aperture Radar (POLSAR) imaging. A four-component scattering model is proposed to decompose POLSAR images. Covariance matrix approach is used to deal with the non-reflection symmetry scattering case. This scheme includes and extends the three-component decomposition method [1] dealing with the reflection symmetry condition that the co-pol and the cross-pol correlations are close to zero. Circular polarization power term is added as the fourth component to the three component scattering model which describes surface, double bounce, and volume scattering. This circular polarization term is taken into account for the co-pol and the cross-pol correlations which regularly appear in complex urban area scattering and disappear in natural distributed target scatterer. This term is relevant for describing man-made targets in urban area scattering. 2. Covariance Matrix for the Non-Reflection Symmetry Condition To derive polarimetric scattering characteristics contained in POLSAR image, it is necessary to evaluate the second order statistics of its scattering matrices. Here, we follow the scheme [1] and present covariance matrix approach to derive a four-component scattering model mathematically. The general covariance matrix is defined as
What problem does this paper attempt to address?