UM  > 科技學院  > 電腦及資訊科學系
Adaptive weighted orthogonal matching pursuit for robust pattern classification
YU-LONG WANG2; YUAN YAN TANG2; LI-NA YANG1
2016-11-02
Conference NameProceedings of the 2016 International Conference on Wavelet Analysis and Pattern Recognition
Source PublicationInternational Conference on Wavelet Analysis and Pattern Recognition
Volume2016-November
Pages174-178
Conference Date10-13 July 2016
Conference PlaceJeju
CountrySouth Korea
PublisherIEEE
Abstract

Orthogonal Matching Pursuit has been a popular sparse representation algorithm due to its simplicity and competitive performance. However, the original orthogonal matching pursuit algorithm and most of its variant utilize the mean square error (MSE) criterion and assign the same weight to each entry of the data. In the presence of large impulsive noise, their performance may severely degrade due to the effect of corrupted entries with large noises. In this work, we propose a novel adaptive weighted orthogonal matching pursuit method for sparse representation with application to robust classification. We show that the proposed method can adaptively assign large weights on clean entries of data and small weights on severely corrupted ones. The experiments on benchmark real-world databases demonstrate the efficacy of the proposed method for handwritten digit and face recognition.

KeywordDigit Recognition Face Recognition Orthogonal Matching Pursuit Sparse Representation
DOIhttps://doi.org/10.1109/ICWAPR.2016.7731639
URLView the original
Indexed BySCI
Language英语
WOS Research AreaComputer Science ; Engineering
WOS SubjectComputer Science, Interdisciplinary Applications ; Engineering, Electrical & Electronic
WOS IDWOS:000387487900020
Fulltext Access
Citation statistics
Document TypeConference paper
CollectionDEPARTMENT OF COMPUTER AND INFORMATION SCIENCE
Affiliation1.School of Computer, Electronics and Information, Guangxi University
2.Faculty of Science and Technology, University of Macau, Macau 999078, China
First Author AffilicationFaculty of Science and Technology
Recommended Citation
GB/T 7714
YU-LONG WANG,YUAN YAN TANG,LI-NA YANG. Adaptive weighted orthogonal matching pursuit for robust pattern classification[C]:IEEE,2016:174-178.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[YU-LONG WANG]'s Articles
[YUAN YAN TANG]'s Articles
[LI-NA YANG]'s Articles
Baidu academic
Similar articles in Baidu academic
[YU-LONG WANG]'s Articles
[YUAN YAN TANG]'s Articles
[LI-NA YANG]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[YU-LONG WANG]'s Articles
[YUAN YAN TANG]'s Articles
[LI-NA YANG]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.