International Journal of Applied Mathematics and Machine LearningAuthor's: Daoling Lin and Wen Wang
Pages: [1] - [21]
Received Date: February 28, 2018; Revised March 12, 2018
Submitted by: Omer Faruk Ertugrul
DOI: http://dx.doi.org/10.18642/ijamml_7100121934
In this paper, we modify the second accelerated proximal gradient (APG) method proposed by Nesterov in [12] (and named by Tseng in [13]) for solving the total variation (TV) image deblurring problems. We also use the variables in [1] to decouple the problems so that we need not to solve the correlative subproblems in each iteration, borrow the idea of [19] to replace the Lipschitz constant by an appropriate positive-definite matrix, and use the continuation technique in [15] to provide good initial solutions for our algorithm.
image deblurring, total variation, the second APG (the second accelerated proximal gradient), auxiliary variables, continuation technique.
