Novel supervised classification approach for multifrequency polarimetric SAR data

摘要

A novel method is proposed for the supervised classification of multifrequency polarimetric synthetic aperture radar (PolSAR) images. The coherency matrices in P-, L-, and C-bands are mapped onto a 9×9 matrix Ω based on the eigenvalue decomposition of the coherency matrix of each band. A boxcar filter is then performed on the matrix Ω. The filtered data are put into a complex Wishart classifier. Finally, the effectiveness of the proposed method is demonstrated with JPL/AIRSAR multifrequency PolSAR data acquired over the Flevoland area.