Performance metrics for SAR speckle-suppression filters

Abstract

Because speckle complicates the application of automatic image classification algorithms, several adaptive filters have been proposed for suppressing the speckle while simultaneously preserving the true underlying reflectivity structure of the image. This paper proposes a new set of tools and metrics for evaluating the performance of such filters, and by way of illustration, they are applied to the refined filter introduced by Lee and to the edge-preserving filter introduced by Hagg and Sties.