Edge detection with multiscale products for SAR image despeckling

A despeckling algorithm of a synthetic aperture radar (SAR) image is proposed, which can effectively preserve edges while cleanly despeckling. The proposed algorithm is robust, simple, and effective, without any a priori information, which employs multiscale products in stationary bandlet transform (SBT) domain to detect edges, and utilises SBT to denoise the edge-removed image. Experimental results show that the visual quality and evaluation indexes of the proposed algorithm outperform that of Lee filtering.