Together with the growing interest in the development of human and computer
interface and biometric identification, human face recognition has become
an active research area since early 1990. Nowadays, principal component ana
lysis (PCA) has been widely adopted as the most promising face recognition
algorithm. Yet still. traditional PCA approach has its limitations: poor di
scriminatory power and large computational load. In view of these limitatio
ns. this article proposed a subband approach in using PCA-apply PCA on wave
let subband. Traditionally, to represent the human face, PGA is performed o
n the whole facial image. In the proposed method, wavelet transform is used
to decompose an image into different frequency subbands, and a midrange fr
equency subband is used for PCA representation. In comparison with the trad
itional use of PCA, the proposed method gives better recognition accuracy a
nd discriminatory power; further the proposed method reduces the computatio
nal bad significantly when the image database is large, with more than 256
training images. This article details the design and implementation of the
proposed method, and presents the encouraging experimental results. (C) 200
0 SPIE and IS&T. [S1017-9909(00)01702-5].