SAR speckle reduction by weighted filtering

Abstract

The weighting filter is a new non-iinear algorithm designed to reduce speckle noise in synthetic aperture radar (SAR) images and is presented in this article. Intrinsic similarities between this and the well documented sigma filter allow a useful quantitative comparison to be made that helps appraise the new technique. Implications drawn from these results, in both the real and Fourier domains, suggest that the new weighting filter may be a useful alternative when considering speckle suppression. © 1993 Taylor & Francis Group, LLC.