Block Principal Component Analysis With Nongreedy l1-Norm Maximization

被引：27

作者：

Li, Bing Nan ^{[1
]}

Yu, Qiang ^{[2
]}

Wang, Rong ^{[2
]}

Xiang, Kui ^{[3
]}

Wang, Meng ^{[4
]}

Li, Xuelong ^{[5
]}

机构：

[1] Hefei Univ Technol, Dept Biomed Engn, Hefei 230009, Peoples R China

[2] Xian Res Inst Hitech, Xian 710025, Peoples R China

[3] Wuhan Univ Technol, Sch Automat, Wuhan 430070, Peoples R China

[4] Hefei Univ Technol, Sch Comp Sci & Informat Engn, Hefei 230009, Peoples R China

[5] Chinese Acad Sci, Xian Inst Opt & Precis Mech, State Key Lab Transient Opt & Photon, Ctr OPT IMagery Anal & Learning OPTIMAL, Xian 710119, Peoples R China

来源：

IEEE TRANSACTIONS ON CYBERNETICS | 2016年 / 46卷 / 11期

基金：

中国博士后科学基金; 中国国家自然科学基金;

关键词：

Block principal component analysis (BPCA); dimensionality reduction; l(1)-norm; nongreedy strategy; outliers; FACE RECOGNITION; 2-DIMENSIONAL PCA; REPRESENTATION; L1-NORM;

D O I：

10.1109/TCYB.2015.2479645

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Block principal component analysis with l(1)-norm (BPCA-L1) has demonstrated its effectiveness in a lot of visual classification and data mining tasks. However, the greedy strategy for solving the l(1)-norm maximization problem is prone to being struck in local solutions. In this paper, we propose a BPCA with nongreedy l(1)-norm maximization, which obtains better solutions than BPCA-L1 with all the projection directions optimized simultaneously. Other than BPCA-L1, the new algorithm has been evaluated against some popular principal component analysis (PCA) algorithms including PCA-L1 and 2-D PCA-L1 on a variety of benchmark data sets. The results demonstrate the effectiveness of the proposed method.

引用

页码：2543 / 2547

页数：5

共 21 条

[1]

[Anonymous], P IEEE INT C COMP VI

[2] Orthogonal laplacianfaces for face recognition [J].

Cai, Deng ;

He, Xiaofei ;

Han, Jiawei ;

Zhang, Hong-Jiang .

IEEE TRANSACTIONS ON IMAGE PROCESSING, 2006, 15 (11) :3608-3614

[3]

Ding C, 2006, P 23 INT C MACH LEAR, P281, DOI DOI 10.1145/1143844.1143880

[4] Is two-dimensional PCA equivalent to a special case of modular PCA? [J].

Gao, Quan-xue .

PATTERN RECOGNITION LETTERS, 2007, 28 (10) :1250-1251

[5] An improved face recognition technique based on modular PCA approach [J].

Gottumukkal, R ;

Asari, VK .

PATTERN RECOGNITION LETTERS, 2004, 25 (04) :429-436

[6] Face recognition using Laplacianfaces [J].

He, XF ;

Yan, SC ;

Hu, YX ;

Niyogi, P ;

Zhang, HJ .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2005, 27 (03) :328-340

[7] TPSLVM: A Dimensionality Reduction Algorithm Based On Thin Plate Splines [J].

Jiang, Xinwei ;

Gao, Junbin ;

Wang, Tianjiang ;

Shi, Daming .

IEEE TRANSACTIONS ON CYBERNETICS, 2014, 44 (10) :1795-1807

[8] Image covariance-based subspace method for face recognition [J].

Kim, Chunghoon ;

Choi, Chong-Ho .

PATTERN RECOGNITION, 2007, 40 (05) :1592-1604

[9] Principal component analysis based on L1-norm maximization [J].

Kwak, Nojun .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2008, 30 (09) :1672-1680

[10] Principal Component Analysis by Lp-Norm Maximization [J].

Kwak, Nojun .

IEEE TRANSACTIONS ON CYBERNETICS, 2014, 44 (05) :594-609

← 1 2 3 →