Presentation 2001/3/16
Parameter Optimization for Image Restoration Filters by Subspace Information Criterion
Daisuke Imaizumi, Masashi Sugiyama, Hidemitsu Ogawa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Most of the image restoration filters proposed so far include parameters which control the characteristics of the filters. For obtaining fine restoration images, values of the parameters should be determined appropriately. In this paper, we propose a method for optimizing the parameters by using the model selection criterion called the subspace information criterion (SIC). SIC is an estimate of the mean squared error between restoration image and unknown original image, and it agrees with the expected mean squared error over the noise for any linear filters. We apply SIC to the moving-average filter, and derive an analytic form of the optimal parameter value. Finally, the effectiveness of the proposed method is demonstrated through computer simulations.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Image restoration / denoising / mean squared error / subspace information criterion (SIC) / linear filter / moving-average filter
Paper # PRMU2000-243
Date of Issue

Conference Information
Committee PRMU
Conference Date 2001/3/16(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Parameter Optimization for Image Restoration Filters by Subspace Information Criterion
Sub Title (in English)
Keyword(1) Image restoration
Keyword(2) denoising
Keyword(3) mean squared error
Keyword(4) subspace information criterion (SIC)
Keyword(5) linear filter
Keyword(6) moving-average filter
1st Author's Name Daisuke Imaizumi
1st Author's Affiliation Department of Computer Science, Graduate School of Information Science and Engineering, Tokyo Institute of Technology()
2nd Author's Name Masashi Sugiyama
2nd Author's Affiliation Department of Computer Science, Graduate School of Information Science and Engineering, Tokyo Institute of Technology
3rd Author's Name Hidemitsu Ogawa
3rd Author's Affiliation Department of Computer Science, Graduate School of Information Science and Engineering, Tokyo Institute of Technology
Date 2001/3/16
Paper # PRMU2000-243
Volume (vol) vol.100
Number (no) 702
Page pp.pp.-
#Pages 8
Date of Issue