Consideration of the Correlation between Beta-angle and Lineament Patterns by Using Polarimetric SAR Images

This paper describes the relationship between a polarimetric parameter, the beta-angle, and the lineament features based on the eigenvalues and eigenvectors of the coherency matrix of the polarimetric SAR (POLSAR) data. First, coherency matrix was calculated from the Pauli vector of the POLSAR image set. Then, eigenvalues and corresponding eigenvectors of the coherency matrix were calculated. Subsequently, beta-angle, derived from the eigenvalue and the corresponding eigenvectors. However, the beta-angle has the potential to represent the orientation of the scatterer about the radar line of sight. In addition, alpha-angle has been obtained using the same eigenvector above. We validate them using fully polarimetric L-band SAR data. The relationship between these two angles is also discussed, and a two-dimensional feature space is constructed (beta-alpha space). We then show initial results of applying the beta-alpha space to the Japanese airborne polarimetric and interferometric SAR (Pi-SAR) to extract features of lineaments as one of the geological features of the land surface. The results demonstrate that the effectiveness of the proposed extraction algorithm.