Producing computationally efficient KPCA-based feature extraction for classification problems

被引：10

作者：

Xu, Y. ^{[1
]}

Lin, C. ^{[1
]}

Zhao, W. ^{[2
]}

机构：

[1] Harbin Inst Technol, Shenzhen Grad Sch, Biocomp Res Ctr, Shenzhen 518055, Peoples R China

[2] Wuhan Univ Technol, Dept Math, Wuhan 430070, Hubei, Peoples R China

来源：

ELECTRONICS LETTERS | 2010年 / 46卷 / 06期

关键词：

KERNEL FISHER DISCRIMINANT;

D O I：

10.1049/el.2010.2814

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

An improvement to kernel principal component analysis (KPCA) to produce computationally efficient KPCA-based feature extraction is proposed. This improvement is applicable to all cases no matter whether the samples in the feature space have zero mean or not. Experiments on several benchmark datasets show that the improvement performs well in classification problems.

引用

页码：452 / U100

页数：2

共 5 条

[1] TWINING C, 2001, BRIT MACH VIS C MANC, V1, P23
[2] An efficient renovation on kernel Fisher discriminant analysis and face recognition experiments
Xu, Y
Yang, JY
Lu, JF
Yu, DJ
[J]. PATTERN RECOGNITION, 2004, 37 (10) : 2091 - 2094
[3] A reformative kernel Fisher discriminant analysis
Xu, Y
Yang, JY
Yang, J
[J]. PATTERN RECOGNITION, 2004, 37 (06) : 1299 - 1302
[4] A method for speeding up feature extraction based on KPCA
Xu, Yong
Zhang, David
Song, Fengxi
Yang, Jing-Yu
Jing, Zhong
Li, Miao
[J]. NEUROCOMPUTING, 2007, 70 (4-6) : 1056 - 1061
[5] Essence of kernel Fisher discriminant: KPCA plus LDA
Yang, J
Jin, Z
Yang, JY
Zhang, D
Frangi, AF
[J]. PATTERN RECOGNITION, 2004, 37 (10) : 2097 - 2100

← 1 →