Learning optimal spatially-dependent regularization parameters in total variation image denoising

被引：20
|
作者：
Van Chung, Cao [1 ,3 ]
De los Reyes, J. C. [1 ]
Schonlieb, C. B. [2 ]
机构：
[1] Escuela Politec Nacl, Res Ctr Math Modelling MODEMAT, Quito, Ecuador
[2] Univ Cambridge, Dept Appl Math & Theoret Phys, Cambridge, England
[3] Hanoi Univ Sci, Ctr High Performance Comp, Hanoi, Vietnam
来源：
INVERSE PROBLEMS | 2017年 / 33卷 / 07期
基金：
英国工程与自然科学研究理事会;
关键词：
optimization-based learning in imaging; bilevel optimization; PDE-constrained optimization; semismooth Newton method; Schwarz domain decomposition method; OPTIMIZATION; SELECTION;
D O I：
10.1088/1361-6420/33/7/074005
中图分类号：
O29 [应用数学];
学科分类号：
070104 ;
摘要：
We consider a bilevel optimization approach in function space for the choice of spatially dependent regularization parameters in TV image denoising models. First- and second-order optimality conditions for the bilevel problem are studied when the spatially-dependent parameter belongs to the Sobolev space H-1(Omega). A combined Schwarz domain decomposition-semismooth Newton method is proposed for the solution of the full optimality system and local superlinear convergence of the semismooth Newton method is verified. Exhaustive numerical computations are finally carried out to show the suitability of the approach.
引用
收藏
页数：31
相关论文
共 50 条
  • [1] Spatially dependent regularization parameter selection for total generalized variation-based image denoising
    Tian-Hui Ma
    Ting-Zhu Huang
    Xi-Le Zhao
    Computational and Applied Mathematics, 2018, 37 : 277 - 296
  • [2] Spatially dependent regularization parameter selection for total generalized variation-based image denoising
    Ma, Tian-Hui
    Huang, Ting-Zhu
    Zhao, Xi-Le
    COMPUTATIONAL & APPLIED MATHEMATICS, 2018, 37 (01): : 277 - 296
  • [3] Ultrasound Image Denoising via Dictionary Learning and Total Variation Regularization
    Li, Shuai
    Zhao, Ximei
    PROCEEDINGS OF THE 4TH INTERNATIONAL CONFERENCE ON MECHATRONICS, MATERIALS, CHEMISTRY AND COMPUTER ENGINEERING 2015 (ICMMCCE 2015), 2015, 39 : 410 - 416
  • [4] Selection of regularization parameters for total variation denoising
    Macquarie Univ, Sydney
    ICASSP IEEE Int Conf Acoust Speech Signal Process Proc, (1653-1656):
  • [5] New Regularization Models for Image Denoising with a Spatially Dependent Regularization Parameter
    Ma, Tian-Hui
    Huang, Ting-Zhu
    Zhao, Xi-Le
    ABSTRACT AND APPLIED ANALYSIS, 2013,
  • [6] Optimal spatially-dependent diffusion tensors
    Oakley, Bryan W.
    Bonner, Gage
    Thiffeault, Jean-Luc
    NONLINEARITY, 2023, 36 (12) : 6777 - 6797
  • [7] Generalized total variation-based MRI Rician denoising model with spatially adaptive regularization parameters
    Liu, Ryan Wen
    Shi, Lin
    Huang, Wenhua
    Xu, Jing
    Yu, Simon Chun Ho
    Wang, Defeng
    MAGNETIC RESONANCE IMAGING, 2014, 32 (06) : 702 - 720
  • [8] Spatially dependent regularization parameter selection in total generalized variation models for image restoration
    Bredies, Kristian
    Dong, Yiqiu
    Hintermueller, Michael
    INTERNATIONAL JOURNAL OF COMPUTER MATHEMATICS, 2013, 90 (01) : 109 - 123
  • [9] Total Variation Regularization for Image Denoising, III. Examples
    Allard, William K.
    SIAM JOURNAL ON IMAGING SCIENCES, 2009, 2 (02): : 532 - 568
  • [10] A new nonlocal total variation regularization algorithm for image denoising
    Liu, Xinwu
    Huang, Lihong
    MATHEMATICS AND COMPUTERS IN SIMULATION, 2014, 97 : 224 - 233
12345