NUMEROUS face recognition methods, such as principal
component analysis (PCA) , independent component
analysis (ICA) , and linear discriminant analysis
(LDA) , have been introduced to achieve successful face
recognition, which attempt to estimate a low-dimensional subspace
for dimensionality reduction and have been demonstrated
effectiveness for face recognition. Moreover, an unsupervised
feature extraction plus supervised classification, that is, kernel
PCA plus LDA (KPCA+LDA) , has been introduced for
classification. Sparse representation classification (SRC)
has been introduced for face recognition, which is now an
important tool for computer vision.