A Neurodynamic Approach to Noisy Image Fusion Based on a Linear Constrained Least Square Method

Abstract

In this paper, a neurodynamic approach is proposed to denoise Gaussian noise through fusing noisy images. A constrained optimization problem based on a linearly constrained least square (LCLS) method is introduced for image fusion in RGB channels. Moreover, a neurodynamic model is introduced to solve the optimization problem in three channels respectively. Experimental results substantiate the efficacy of the proposed approach.