Face Recognition Combining 3-channel Filter Bank and 2DLDA

Author

Zhang, Yuhua ; Wang, Xin

Author_Institution

Sch. of Med., Dept. of Biomed. Eng., Shandong Univeristy, Jinan

Volume

2

fYear

2009

fDate

22-24 Jan. 2009

Firstpage

84

Lastpage

87

Abstract

In this paper, a novel 3-channel orthogonal filter bank was proposed at first, and then an efficient face recognition method combining the filter bank and 2DLDA (two-dimensional linear discriminant analysis) was presented. The framework of the proposed method consists of two steps: First, each face image was decomposed into nine sub images by using the 3-channel filter bank, and then 2DLDA was used to extract the features for face classification from the sub images. Better performance of the proposed method is confirmed by the Yale face database.

Keywords

face recognition; feature extraction; filters; image classification; 2D linear discriminant analysis; 3-channel orthogonal filter bank; face classification; face image decomposition; face recognition; feature extraction; Biomedical engineering; Covariance matrix; Discrete wavelet transforms; Face recognition; Feature extraction; Filter bank; Image reconstruction; Linear discriminant analysis; Low pass filters; Principal component analysis; face recognition; multi-channel filter bank; two-dimensional LDA; wavelet transform;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Engineering and Technology, 2009. ICCET '09. International Conference on

Conference_Location

Singapore

Print_ISBN

978-1-4244-3334-6

Type

conf

DOI

10.1109/ICCET.2009.82

Filename

4769563