Selecting the Principal Feature Components in the Three-dimensional Parameter Space for Face Recognition

Author

Junbao, Li ; Shuchuan, Chu ; Jengshyang, Pan

Author_Institution

Harbin Inst. of Technol., Harbin

fYear

2007

Firstpage

42396

Lastpage

42400

Abstract

This paper presents a novel face recognition method of selecting the principal feature components in the 3D parameter space constructed using the dimensions of three subspaces, i.e., PCA subspace, LDA subspace and LPP subspace, as axes. The global, local and clustering structure information can be used fully to enhance the recognition performance by selecting the principal feature component in 3D parameter space. The feasibility of the proposed method is successfully tested on the ORL and Yale face databases.

Keywords

face recognition; principal component analysis; 3D parameter space; LDA subspace; LPP subspace; ORL face database; PCA subspace; Yale face database; face recognition; principal feature components; Automatic control; Automatic testing; Electronic equipment testing; Face recognition; Feature extraction; Information management; Instruments; Linear discriminant analysis; Principal component analysis; Space technology; Face recognition; linear discriminant analysis; locality preserving projection; principal component analysis; three-dimensional parameter space;

fLanguage

English

Publisher

ieee

Conference_Titel

Electronic Measurement and Instruments, 2007. ICEMI '07. 8th International Conference on

Conference_Location

Xi´an

Print_ISBN

978-1-4244-1136-8

Electronic_ISBN

978-1-4244-1136-8

Type

conf

DOI

10.1109/ICEMI.2007.4350440

Filename

4350440