Face classification using a multiresolution principal component analysis

Author

Brennan, Vic ; Principe, Jose

Author_Institution

Dept. of Electr. Eng., Florida Univ., Gainesville, FL, USA

fYear

1998

fDate

31 Aug-2 Sep 1998

Firstpage

506

Lastpage

515

Abstract

Multiresolution principal component analysis (M-PCA) uses principal component analysis (PCA) to obtain multiresolution features for a signal. Bischof (1995) and Bischof and Hornik (1996) used 3-layer networks to train principal component pyramids for image compression. M-PCA uses a single computational layer adaptive linear network trained with the generalized Hebbian algorithm (GHA). The multiresolution features were applied to automatic face recognition and tested against the Olivetti Research Lab database. Classification with multiresolution had an average (over 10 runs) error rate of 2.4%

Keywords

Hebbian learning; face recognition; image classification; neural nets; Olivetti Research Lab database; face classification; generalized Hebbian algorithm; multiresolution principal component analysis; single computational layer adaptive linear network; Adaptive systems; Automatic testing; Computer networks; Error analysis; Face recognition; Image coding; Image databases; Principal component analysis; Signal resolution; Spatial databases;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks for Signal Processing VIII, 1998. Proceedings of the 1998 IEEE Signal Processing Society Workshop

Conference_Location

Cambridge

ISSN

1089-3555

Print_ISBN

0-7803-5060-X

Type

conf

DOI

10.1109/NNSP.1998.710681

Filename

710681