Optimization Online


Partial Convolution for Total Variation Deblurring and Denoising by New Linearized Alternating Direction Method of Multipliers with Extension Step

Yuan Shen(ocsiban***at***126.com)
Lei Ji(relbeek***at***126.com)

Abstract: In this paper, we propose a partial convolution model for image delburring and denoising. We also devise a new linearized alternating direction method of multipliers (ADMM) with extension step. On one hand, the computation of its subproblem is dominated by several FFTs, hence its per-iteration cost is low, on the other hand, the relaxed parameter condition together with the extra extension step inspired by Ye and Yuan's ADMM enables faster convergence than the original linearized ADMM. Preliminary experimental results show that our algorithm can produce a result with better quality than some existing efficient algorithms while the computation time is competitive. Specially, the advantage of our algorithm can be more evident when the noise ratio is high.

Keywords: convex optimization; proximal point algorithm; augmented Lagrangian; total variation; deblur and denoise; partial convolution

Category 1: Applications -- OR and Management Sciences

Citation: Tech Report 0702, Nanjing, May 2017

Download: [PDF]

Entry Submitted: 05/01/2017
Entry Accepted: 05/01/2017
Entry Last Modified: 05/01/2017

Modify/Update this entry

  Visitors Authors More about us Links
  Subscribe, Unsubscribe
Digest Archive
Search, Browse the Repository


Coordinator's Board
Classification Scheme
Give us feedback
Optimization Journals, Sites, Societies
Mathematical Optimization Society