Speckle filtering algorithm for polarimetric SAR based on mean shift

In the area of speckle filtering, proper averaging of samples with similar scattering characteristics is of great importance. However, existing filtering algorithms are either lack of similarity judgment of scattering characteristics or using only intensity information for similarity judgment. In this paper, a novel polarimetric SAR (PolSAR) speckle filtering algorithm based on mean shift theory is proposed. By using range-spatial joint mean shift theory in Riemannian manifold, the pixels chosen for averaging are ensured to be close not only in scattering characteristics but also in spatial domain. Experiment results show that the proposed speckle filtering algorithm achieves good performance in terms of speckle filtering, edge protection as well as polarimetric characteristics preservation.