A study of singular value decomposition of face image matrix

Author

Junying Gan ; Zhang, Youwei

Author_Institution

Sch. of Inf., Wuyi Univ., Guangdong, China

Volume

1

fYear

2003

fDate

14-17 Dec. 2003

Firstpage

197

Abstract

In this paper, singular value decomposition (SVD) and its stability, proportion invariance, and rotation invariance of image matrix are discussed in detail. Optimal approximation of the rank reduction matrix is studied. As for the Olivetti Research Laboratory (ORL) face database, experimental results demonstrate that it is effective and feasible that singular values of the face image matrix are used as recognition features in face recognition, and that singular value features take on separability, stability and independence, and are valid in feature extraction of face image.

Keywords

face recognition; feature extraction; singular value decomposition; stability; Olivetti Research Laboratory face database; face image matrix; face recognition; feature extraction; proportion invariance; rank reduction matrix; rotation invariance; singular value decomposition; stability; Face recognition; Feature extraction; Image databases; Image recognition; Interference; Laboratories; Matrix decomposition; Singular value decomposition; Spatial databases; Stability;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks and Signal Processing, 2003. Proceedings of the 2003 International Conference on

Conference_Location

Nanjing

Print_ISBN

0-7803-7702-8

Type

conf

DOI

10.1109/ICNNSP.2003.1279245

Filename

1279245