Application of the UPRE method to optimal parameter selection for large scale regularization problems

被引:8
|
作者
Lin, Youzuo [1 ]
Wohlberg, Brendt [2 ]
机构
[1] Arizona State Univ, Dept Math & Stat, Tempe, AZ 85287 USA
[2] Los Alamos Natl Lab, Math Modeling & Anal, Los Alamos, NM 87545 USA
来源
2008 IEEE SOUTHWEST SYMPOSIUM ON IMAGE ANALYSIS & INTERPRETATION | 2008年
关键词
parameter selection; large scale problem; inverse problem; Tikhonov regularization; total variation regularization;
D O I
10.1109/SSIAI.2008.4512292
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Regularization is an important method for solving a wide variety of inverse problems in image processing. In order to optimize the reconstructed image, it is important to choose the optimal regularization parameter. The Unbiased Predictive Risk Estimator (UPRE) has been shown to give a very good estimate of this parameter. Applying the traditional UPRE is impractical, however, in the case of inverse problems such as deblurring, due to the large scale of the associated linear problem. We propose an approach to reducing the large scale problem to a small problem, significantly reducing computational requirements while providing a good approximation to the original problem.
引用
收藏
页码:89 / +
页数:2
相关论文
共 50 条
  • [31] Optimized Parameter Search for Large Datasets of the Regularization Parameter and Feature Selection for Ridge Regression
    Pieter Buteneers
    Ken Caluwaerts
    Joni Dambre
    David Verstraeten
    Benjamin Schrauwen
    Neural Processing Letters, 2013, 38 : 403 - 416
  • [32] Scale recognition, regularization parameter selection, and Meyer's G norm in total variation regularization
    Strong, David M.
    Aujol, Jean-Francois
    Chan, Tony F.
    MULTISCALE MODELING & SIMULATION, 2006, 5 (01): : 273 - 303
  • [33] ON THE CHOICE OF THE OPTIMAL PARAMETER FOR TIKHONOV REGULARIZATION OF ILL-POSED PROBLEMS
    LI, H
    CHINESE SCIENCE BULLETIN, 1992, 37 (21): : 1770 - 1773
  • [34] ON THE CHOICE OF THE OPTIMAL PARAMETER FOR TIKHONOV REGULARIZATION OF ILL-POSED PROBLEMS
    李浩
    Chinese Science Bulletin, 1992, (21) : 1770 - 1773
  • [35] INVESTIGATION OF AN ADAPTIVE REGULARIZATION PARAMETER SELECTION METHOD IN BIOLUMINESCENCE TOMOGRAPHY
    Feng, Jinchao
    Yang, MingJie
    Jia, Kebin
    2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI), 2014, : 181 - 184
  • [36] REGULARIZATION OF BELLMANS METHOD IN OPTIMAL QUICK ACTION PROBLEMS
    KUN, LA
    PRONOZIN, YF
    DOKLADY AKADEMII NAUK SSSR, 1971, 200 (06): : 1294 - &
  • [37] Selection of regularization parameter in GMM based image denoising method
    Zheng, Yuhui
    Li, Min
    Zhang, Jianwei
    Wang, Jin
    MULTIMEDIA TOOLS AND APPLICATIONS, 2018, 77 (22) : 30121 - 30134
  • [38] Selection of regularization parameter in GMM based image denoising method
    Yuhui Zheng
    Min Li
    Jianwei Zhang
    Jin Wang
    Multimedia Tools and Applications, 2018, 77 : 30121 - 30134
  • [39] A NEW REGULARIZATION METHOD FOR PARAMETER-IDENTIFICATION IN ELLIPTIC PROBLEMS
    TAUTENHAHN, U
    INVERSE PROBLEMS, 1990, 6 (03) : 465 - 477
  • [40] ON A CLASS OF STOCHASTIC IMPULSIVE OPTIMAL PARAMETER SELECTION PROBLEMS
    Liu, Chun Min
    Feng, Zhi Guo
    Teo, Kok Lay
    INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL, 2009, 5 (04): : 1043 - 1054