Projective robust nonnegative factorization

CORC > 西安光学精密机械研究所 > 中国科学院西安光学精密机械研究所 > 光学影像学习与分析中心

	Projective robust nonnegative factorization
	Lu, Yuwu 1; Lai, Zhihui 2; Xu, Yong 3; You, Jane 4; Li, Xuelong5 ; Yuan, Chun 1
刊名	information sciences
	2016-10-10
卷号	364 页码:16-32
关键词	Robust Nonnegative matrix factorization Graph regularization Face recognition
ISSN号	0020-0255
产权排序	5
英文摘要	nonnegative matrix factorization (nmf) has been successfully used in many fields as a low-dimensional representation method. projective nonnegative matrix factorization (pnmf) is a variant of nmf that was proposed to learn a subspace for feature extraction. however, both original nmf and pnmf are sensitive to noise and are unsuitable for feature extraction if data is grossly corrupted. in order to improve the robustness of nmf, a framework named projective robust nonnegative factorization (prnf) is proposed in this paper for robust image feature extraction and classification. since learned projections can weaken noise disturbances, prnf is more suitable for classification and feature extraction. in order to preserve the geometrical structure of original data, prnf introduces a graph regularization term which encodes geometrical structure. in the prnf framework, three algorithms are proposed that add a sparsity constraint on the noise matrix based on l-1/2 norm, l-1 norm, and l-2,l-1 norm, respectively. robustness and classification performance of the three proposed algorithms are verified with experiments on four face image databases and results are compared with state-of-the-art robust nmf-based algorithms. experimental results demonstrate the robustness and effectiveness of the algorithms for image classification and feature extraction. (c) 2016 elsevier inc. all rights reserved.
WOS标题词	science & technology ; technology
类目[WOS]	computer science, information systems
研究领域[WOS]	computer science
关键词[WOS]	matrix factorization ; image representation ; algorithm ; parts
收录类别	SCI ; EI
语种	英语
WOS记录号	WOS:000378969400002
内容类型	期刊论文
源URL	[http://ir.opt.ac.cn/handle/181661/28168]
专题	西安光学精密机械研究所_光学影像学习与分析中心
作者单位	1.Tsinghua Univ, Tsinghua CUHK Joint Res Ctr Media Sci Technol & S, Grad Sch Shenzhen, Beijing, Peoples R China 2.Shenzhen Univ, Coll Comp Sci & Software Engn, Shenzhen, Peoples R China 3.Harbin Inst Technol, Shenzhen Grad Sch, Biocomp Res Ctr, Harbin, Peoples R China 4.Hong Kong Polytech Univ, Dept Comp, Hong Kong, Hong Kong, Peoples R China 5.Chinese Acad Sci, State Key Lab Transient Opt & Photon, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Lu, Yuwu,Lai, Zhihui,Xu, Yong,et al. Projective robust nonnegative factorization[J]. information sciences,2016,364:16-32.
APA	Lu, Yuwu,Lai, Zhihui,Xu, Yong,You, Jane,Li, Xuelong,&Yuan, Chun.(2016).Projective robust nonnegative factorization.information sciences,364,16-32.
MLA	Lu, Yuwu,et al."Projective robust nonnegative factorization".information sciences 364(2016):16-32.