An algorithm combining coordinate descent and split Bregman iterative for fractional image denoising model. (August 2021)

• Record Type: Journal Article
• Authors: Zhao, Donghong; Fan, Yonghua; Liu, Haoyu; Yang, Yafeng
• Is Part Of: Journal of algorithms & computational technology. Volume 15(2021)
• Publication Date: 2021-08
• Subjects: Image denoising -- coordinate descent method -- fractional total variation -- split Bregman algorithm; Computer algorithms -- Periodicals; Numerical calculations -- Periodicals; Computer algorithms; Numerical calculations; Periodicals; 518.1
• DOI: 10.1177/17483026211031167
• Languages: English
• ISSNs: 1748-3018

Abstract

The split Bregman algorithm and the coordinate descent method are efficient tools for solving optimization problems, which have been proven to be effective for the total variation model. We propose an algorithm for fractional total variation model in this paper, and employ the coordinate descent method to decompose the fractional-order minimization problem into scalar sub-problems, then solve the sub-problem by using split Bregman algorithm. Numerical results are presented in the end to demonstrate the superiority of the proposed algorithm.