Does EigenPalm work? A System and Evaluation Perspective

Author

Cheung, King-Hong ; Kong, Adams ; Zhang, David ; Kamel, Mohamed ; You, Jane

Author_Institution

Dept. of Comput., Hong Kong Polytech. Univ.

Volume

4

fYear

0

fDate

0-0 0

Firstpage

445

Lastpage

448

Abstract

Recently, there are keen interests in EigenPalm, which, collectively, refers to those methods that extract palmprint features directly from the appearance by means of principal component analysis (PCA) for (dis)similarity matching. Encouraging results have been reported with the use of EigenPalm. However, we find a different story under a system and evaluation perspective. In this paper, we would like to introduce three issues that should be considered: the effects of templates from two different sessions, the effects of identical twins and the effects of unseen subjects. They are missing in the previous studies of EigenPalm

Keywords

feature extraction; image matching; principal component analysis; EigenPalm; palmprint feature extraction; principal component analysis; similarity matching; Biometrics; Databases; Discrete transforms; Feature extraction; Karhunen-Loeve transforms; Machine intelligence; Pattern analysis; Pattern matching; Principal component analysis; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Pattern Recognition, 2006. ICPR 2006. 18th International Conference on

Conference_Location

Hong Kong

ISSN

1051-4651

Print_ISBN

0-7695-2521-0

Type

conf

DOI

10.1109/ICPR.2006.460

Filename

1699874