A Four-Component Decomposition Model for PolSAR Data Using Asymmetric Scattering Component

This paper describes a polarimetric information extraction method using PolSAR data. The proposed method is based on the covariance matrix and intended to be applicable to the general nonreflection symmetric scattering case that the copolar and cross-polar correlations are not close to zero. Asymmetric scattering term, which can entirely represent the asymmetric information of PolSAR data, is proposed to describe the nonreflection symmetric scattering caused by complicated shape targets or man-made structures in urban areas. Using this asymmetric scattering term, together with surface, double-bounce, and volume scattering terms, the new four-component scattering power decomposition method is then proposed, which can fully utilize all the parameters contained in the PolSAR data. It is found that asymmetric scattering is predominant in urban areas and close to zero in natural distributed areas. Although designed for nonreflection symmetric scattering case, the decomposition method can also deal with the reflection symmetry case, since it automatically becomes Freeman three-component decomposition method in natural areas. The E-SAR L-band full-polarized data acquired over Oberpfaffenhofen area in Germany and the CONVAIR-SAR C-band data acquired over Ottawa area in Canada are applied to validate the proposed decomposition method. The results of the experiments show that the proposed method has better performance than the four-component decomposition model in distinguishing different types of terrains.