Mixed Gaussian-Impulse Noise Removal from Highly Corrupted Images via Adaptive Local and Nonlocal Statistical Priors

TL;DR

This paper introduces De-JASP, a novel variational framework utilizing adaptive statistical priors for effective removal of mixed Gaussian and impulse noise from highly corrupted images, outperforming current methods.

Contribution

It proposes a new adaptive curvelet thresholding and a joint statistical prior (JASP) for simultaneous local and nonlocal noise suppression in a unified variational scheme.

Findings

De-JASP outperforms state-of-the-art methods in mixed noise removal.

The adaptive curvelet thresholding improves denoising accuracy.

The joint statistical prior effectively enforces local and nonlocal consistency.

Abstract

The motivation of this paper is to introduce a novel framework for the restoration of images corrupted by mixed Gaussian-impulse noise. To this aim, first, an adaptive curvelet thresholding criterion is proposed which tries to adaptively remove the perturbations appeared during denoising process. Then, a new statistical regularization term, called joint adaptive statistical prior (JASP), is established which enforces both the local and nonlocal statistical consistencies, simultaneously, in a unified manner. Furthermore, a novel technique for mixed Gaussian plus impulse noise removal using JASP in a variational scheme is developed--we refer to it as De-JASP. To efficiently solve the above variational scheme, an efficient alternating minimization algorithm based on split Bregman iterative framework is developed. Extensive experimental results manifest the effectiveness of the proposed method comparing with the current state-of-the-art methods in mixed Gaussian-impulse noise removal.