A nonlocal model for image restoration with gamma distributed multiplicative noise

Abstract

The question of eliminating the multiplicative noise has attracted much interest in many research studies. In this work, we are interested with the one that are follows the Gamma distribution. The rationale of this paper is to shed light on a brief comparative study of some local and nonlocal models, for denoising images contaminated with the noise of this type. The improved method of Split Bregman is used to implement those models. The totality of experiments indicates that the proposed nonlocal method gives better results than some other methods.