An adaptive multiresolution method for speckle noise reduction in synthetic aperture radar images

Abstract

This paper introduces a method for speckle noise reduction based on multiresolution algorithms and a new adaptive approach. It combines the characteristics for additive noise reduction of soft thresholding with the localization of information in the coarsest levels of the wavelet transform, its main contribution being to adapt the effective threshold to the level of resolution in the wavelet transform and to the local features of the image. The method has been applied successfully to the elimination of speckle noise of simulated SAR images. It has been applied to an optical image corrupted with multiplicative noise that simulates SAR statistics and it has been compared to other well-known noise reduction methods.