A color image fusion model by saturation-value total variation

Abstract

In this paper, we propose and develop a novel color image fusion model by using saturation-value total variation. In the proposed model, we develop a variational approach containing an energy functional to determine the weighting mask functions and the fused image together. The objective fused image is modeled by using the saturation-value total variation regularization. The data-fitting term is formulated based on the weighting L2 norm of the objective fused image and the input observed images. The objective weighting mask functions are modeled by using total variation for the piecewise-smoothness purpose. Meanwhile, we add another fidelity term to capture more texture and color information from the input images by forcing the weighting local mean images to be close to the clear input image. The existence of a solution of the proposed minimizing model is shown. Numerically, we apply alternating direction method of multipliers (ADMM) to solve the proposed minimization problem, meanwhile, we give the convergence analysis of the proposed algorithm. Numerical examples are presented to demonstrate that the performance of the proposed color image fusion model is better than that of other testing methods in terms of visual quality and some criterias such as structure similarity (SSIM), average local contrast (ALC), discrete entropy (DE), natural image quality evaluator (NIQE) and perception-based image quality evaluator (PIQE).