Separated Component-Based Restoration of Speckled SAR Images

The coherent nature of the synthetic aperture radar (SAR) imaginary process, detected images suffer from the multiplicative noise, commonly referred to as speckle. This noise often makes the interpretation of the data difficult for both automated algorithms and human observers. Several methods have been proposed for speckle reduction is to use a dictionary that can represent the features in the speckled images. However, such methods fail to capture important salient features such as texture. Namely, we would like to separate the image to its structure and texture components based on the algorithm suggested for SAR images in the following paper. In this paper we present a speckle reduction algorithm for restoration problem of SAR images so that the structure and texture components can be separately estimated with different dictionaries. To solve this problem, an iterative algorithm based on surrogate functional is proposed. This paper shows better results than state-of-the-art speckle reduction methods.