5.00
Hdl Handle:
http://hdl.handle.net/10547/224106
Title:
Multi-scale blind motion deblurring using local minimum
Authors:
Wang, Chao; Sun, Li-Feng; Chen, ZhuoYuan; Zhang, JianWei; Yang, ShiQiang
Abstract:
Blind deconvolution, a chronic inverse problem, is the recovery of the latent sharp image from a blurred one when the blur kernel is unknown. Recent algorithms based on the MAP approach encounter failures since the global minimum of the negative MAP scores really favors the blurry image. The goal of this paper is to demonstrate that the sharp image can be obtained from the local minimum by using the MAP approach. We first propose a cross-scale constraint to make the sharp image correspond to a good local minimum. Then the cross-scale initialization, iterative likelihood update and the iterative residual deconvolution are adopted to trap the MAP approach in the desired local minimum. These techniques result in our cross-scale blind deconvolution approach which constrains the solution from coarse to fine. We test our approach on the standard dataset and many other challenging images. The experimental results suggest that our approach outperforms all existing alternatives.
Affiliation:
Department of Computer Science and Technology, Tsinghua University, Beijing, People's Republic of China; Key Laboratory of Media and Networking, MOE-Microsoft, Tsinghua University, Beijing, People's Republic of China; Department of Informatics, Hamburg University, Germany
Citation:
Multi-scale blind motion deblurring using local minimum 2010, 26 (1):015003 Inverse Problems
Publisher:
Institute of Physics
Journal:
Inverse Problems
Issue Date:
Jan-2010
URI:
http://hdl.handle.net/10547/224106
DOI:
10.1088/0266-5611/26/1/015003
Additional Links:
http://stacks.iop.org/0266-5611/26/i=1/a=015003?key=crossref.70ae8b5d92ab06879ed93848482958a2
Type:
Article
Language:
en
ISSN:
0266-5611; 1361-6420
Appears in Collections:
Centre for Computer Graphics and Visualisation (CCGV)

Full metadata record

DC FieldValue Language
dc.contributor.authorWang, Chaoen_GB
dc.contributor.authorSun, Li-Fengen_GB
dc.contributor.authorChen, ZhuoYuanen_GB
dc.contributor.authorZhang, JianWeien_GB
dc.contributor.authorYang, ShiQiangen_GB
dc.date.accessioned2012-05-16T11:36:17Z-
dc.date.available2012-05-16T11:36:17Z-
dc.date.issued2010-01-
dc.identifier.citationMulti-scale blind motion deblurring using local minimum 2010, 26 (1):015003 Inverse Problemsen_GB
dc.identifier.issn0266-5611-
dc.identifier.issn1361-6420-
dc.identifier.doi10.1088/0266-5611/26/1/015003-
dc.identifier.urihttp://hdl.handle.net/10547/224106-
dc.description.abstractBlind deconvolution, a chronic inverse problem, is the recovery of the latent sharp image from a blurred one when the blur kernel is unknown. Recent algorithms based on the MAP approach encounter failures since the global minimum of the negative MAP scores really favors the blurry image. The goal of this paper is to demonstrate that the sharp image can be obtained from the local minimum by using the MAP approach. We first propose a cross-scale constraint to make the sharp image correspond to a good local minimum. Then the cross-scale initialization, iterative likelihood update and the iterative residual deconvolution are adopted to trap the MAP approach in the desired local minimum. These techniques result in our cross-scale blind deconvolution approach which constrains the solution from coarse to fine. We test our approach on the standard dataset and many other challenging images. The experimental results suggest that our approach outperforms all existing alternatives.en_GB
dc.language.isoenen
dc.publisherInstitute of Physicsen_GB
dc.relation.urlhttp://stacks.iop.org/0266-5611/26/i=1/a=015003?key=crossref.70ae8b5d92ab06879ed93848482958a2en_GB
dc.rightsArchived with thanks to Inverse Problemsen_GB
dc.titleMulti-scale blind motion deblurring using local minimumen
dc.typeArticleen
dc.contributor.departmentDepartment of Computer Science and Technology, Tsinghua University, Beijing, People's Republic of Chinaen_GB
dc.contributor.departmentKey Laboratory of Media and Networking, MOE-Microsoft, Tsinghua University, Beijing, People's Republic of Chinaen
dc.contributor.departmentDepartment of Informatics, Hamburg University, Germanyen
dc.identifier.journalInverse Problemsen_GB
All Items in UOBREP are protected by copyright, with all rights reserved, unless otherwise indicated.