Adaptive Fractional Order Total Variation Image Denoising via the Alternating Direction Method of Multipliers

Due to the importance and priority, image denoising has various applications in image processing. How to mitigate the effect of staircase artifacts on denoised results and adjusts regularization parameters based on surrounding features are two main issues in the image denoising regularization process,. In this paper, an adaptive fractional order total variation l(1) regularization (AFOTV-l(1)) model is proposed. Regularization parameters are adaptively adjusted by fractional order alpha. Alternating direction method of multipliers (ADMM) algorithm is introduced for solving AFOTV-l(1) model, which provides a feasible method to solve the optimization problems of image denoising by introducing auxiliary variables. Two experiments have been conducted with standard images "Lena" and "Man" to verify the effectiveness of the proposed method. Experimental results demonstrate that the proposed method has good denoising effect and fast convergence rate for image denoising.