Volume 3, Issue 4
Relaxed Alternating Minimization Algorithm for Separable Convex Programming with Applications to Imaging

Shuangshuang Wu, Yuchao Tang & Tieyong Zeng

CSIAM Trans. Appl. Math., 3 (2022), pp. 626-661.

Published online: 2022-11

Export citation
  • Abstract

We propose a relaxed alternating minimization algorithm for solving two-block separable convex minimization problems with linear equality constraints, where one block in the objective functions is strongly convex. We prove that the proposed algorithm converges to the optimal primal-dual solution of the original problem. Furthermore, the convergence rates of the proposed algorithm in both ergodic and nonergodic senses have also been studied. We apply the proposed algorithm to solve several composite convex minimization problems arising in image denoising and evaluate the numerical performance of the proposed algorithm on a novel image denoising model. Numerical results for both artificial and real noisy images demonstrate the efficiency and effectiveness of the proposed algorithm.

  • AMS Subject Headings

47H05, 90C25, 65K05

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{CSIAM-AM-3-626, author = {Wu , ShuangshuangTang , Yuchao and Zeng , Tieyong}, title = {Relaxed Alternating Minimization Algorithm for Separable Convex Programming with Applications to Imaging}, journal = {CSIAM Transactions on Applied Mathematics}, year = {2022}, volume = {3}, number = {4}, pages = {626--661}, abstract = {

We propose a relaxed alternating minimization algorithm for solving two-block separable convex minimization problems with linear equality constraints, where one block in the objective functions is strongly convex. We prove that the proposed algorithm converges to the optimal primal-dual solution of the original problem. Furthermore, the convergence rates of the proposed algorithm in both ergodic and nonergodic senses have also been studied. We apply the proposed algorithm to solve several composite convex minimization problems arising in image denoising and evaluate the numerical performance of the proposed algorithm on a novel image denoising model. Numerical results for both artificial and real noisy images demonstrate the efficiency and effectiveness of the proposed algorithm.

}, issn = {2708-0579}, doi = {https://doi.org/10.4208/csiam-am.SO-2021-0039}, url = {http://global-sci.org/intro/article_detail/csiam-am/21151.html} }
TY - JOUR T1 - Relaxed Alternating Minimization Algorithm for Separable Convex Programming with Applications to Imaging AU - Wu , Shuangshuang AU - Tang , Yuchao AU - Zeng , Tieyong JO - CSIAM Transactions on Applied Mathematics VL - 4 SP - 626 EP - 661 PY - 2022 DA - 2022/11 SN - 3 DO - http://doi.org/10.4208/csiam-am.SO-2021-0039 UR - https://global-sci.org/intro/article_detail/csiam-am/21151.html KW - Alternating minimization algorithm, strongly convex function, total variation, image denoising. AB -

We propose a relaxed alternating minimization algorithm for solving two-block separable convex minimization problems with linear equality constraints, where one block in the objective functions is strongly convex. We prove that the proposed algorithm converges to the optimal primal-dual solution of the original problem. Furthermore, the convergence rates of the proposed algorithm in both ergodic and nonergodic senses have also been studied. We apply the proposed algorithm to solve several composite convex minimization problems arising in image denoising and evaluate the numerical performance of the proposed algorithm on a novel image denoising model. Numerical results for both artificial and real noisy images demonstrate the efficiency and effectiveness of the proposed algorithm.

Shuangshuang Wu, Yuchao Tang & Tieyong Zeng. (2022). Relaxed Alternating Minimization Algorithm for Separable Convex Programming with Applications to Imaging. CSIAM Transactions on Applied Mathematics. 3 (4). 626-661. doi:10.4208/csiam-am.SO-2021-0039
Copy to clipboard
The citation has been copied to your clipboard