OPT OpenIR  > 光学影像学习与分析中心
Low-Rank 2-D Neighborhood Preserving Projection for Enhanced Robust Image Representation
Lu, Yuwu1; Lai, Zhihui1,2; Li, Xuelong3,4,5; Wong, Wai Keung2; Yuan, Chun6; Zhang, David7
Department光学影像学习与分析中心
2019-05
Source PublicationIEEE TRANSACTIONS ON CYBERNETICS
ISSN2168-2267;2168-2275
Volume49Issue:5Pages:1859–1872
Contribution Rank5
Abstract

2-D neighborhood preserving projection (2DNPP) uses 2-D images as feature input instead of 1-D vectors used by neighborhood preserving projection (NPP). 2DNPP requires less computation time than NPP. However, both NPP and 2DNPP use the L2 norm as a metric, which is sensitive to noise in data. In this paper, we proposed a novel NPP method called low-rank 2DNPP (LR-2DNPP). This method divided the input data into a component part that encoded low-rank features, and an error part that ensured the noise was sparse. Then, a nearest neighbor graph was learned from the clean data using the same procedure as 2DNPP. To ensure that the features learned by LR-2DNPP were optimal for classification, we combined the structurally incoherent learning and low-rank learning with NPP to form a unified model called discriminative LR-2DNPP (DLR-2DNPP). By encoding the structural incoherence of the learned clean data, DLR-2DNPP could enhance the discriminative ability for feature extraction. Theoretical analyses on the convergence and computational complexity of LR-2DNPP and DLR-2DNPP were presented in details. We used seven public image databases to verify the performance of the proposed methods. The experimental results showed the effectiveness of our methods for robust image representation.

Keyword2-D neighborhood preserving projection (2DNPP) image representation low-rank robust feature extraction
DOI10.1109/TCYB.2018.2815559
Indexed BySCI
Language英语
WOS IDWOS:000460667400026
PublisherIEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Citation statistics
Cited Times:4[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Identifierhttp://ir.opt.ac.cn/handle/181661/31330
Collection光学影像学习与分析中心
Corresponding AuthorLu, Yuwu
Affiliation1.Shenzhen Univ, Coll Comp Sci & Software Engn, Shenzhen 518055, Peoples R China
2.Hong Kong Polytech Univ, Inst Text & Clothing, Hong Kong, Peoples R China
3.Northwestern Polytech Univ, Sch Comp Sci, Xian 710072, Shaanxi, Peoples R China
4.Northwestern Polytech Univ, Ctr OPT IMagery Anal & Learning OPTIMAL, Xian 710072, Shaanxi, Peoples R China
5.Chinese Acad Sci, Xian Inst Opt & Precis Mech, Xian 710119, Shaanxi, Peoples R China
6.Tsinghua Univ, Shenzhen Grad Sch, Shenzhen 518055, Peoples R China
7.Hong Kong Polytech Univ, Biometr Res Ctr, Hong Kong, Peoples R China
Recommended Citation
GB/T 7714
Lu, Yuwu,Lai, Zhihui,Li, Xuelong,et al. Low-Rank 2-D Neighborhood Preserving Projection for Enhanced Robust Image Representation[J]. IEEE TRANSACTIONS ON CYBERNETICS,2019,49(5):1859–1872.
APA Lu, Yuwu,Lai, Zhihui,Li, Xuelong,Wong, Wai Keung,Yuan, Chun,&Zhang, David.(2019).Low-Rank 2-D Neighborhood Preserving Projection for Enhanced Robust Image Representation.IEEE TRANSACTIONS ON CYBERNETICS,49(5),1859–1872.
MLA Lu, Yuwu,et al."Low-Rank 2-D Neighborhood Preserving Projection for Enhanced Robust Image Representation".IEEE TRANSACTIONS ON CYBERNETICS 49.5(2019):1859–1872.
Files in This Item:
File Name/Size DocType Version Access License
Low-Rank 2-D Neighbo(1718KB)期刊论文出版稿开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Lu, Yuwu]'s Articles
[Lai, Zhihui]'s Articles
[Li, Xuelong]'s Articles
Baidu academic
Similar articles in Baidu academic
[Lu, Yuwu]'s Articles
[Lai, Zhihui]'s Articles
[Li, Xuelong]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Lu, Yuwu]'s Articles
[Lai, Zhihui]'s Articles
[Li, Xuelong]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Low-Rank 2-D Neighborhood Preserving Projection for Enhanced Robust Image Representation.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

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