A Statistical PCA Method for Face Recognition

The standard PCA algorithm has two mainly disadvantages: one is computing complexity, The other is it can only process the faces have the same face expression. In order to solve these problems, a new face recognition method called SPCA( Statistical Principal Component Analysis Method) is proposed in this paper. First, an improved PCA algorithm is used to compute the eigen-vector and eigen-values of the face. Second, Bayesian rule is used to design the classification designer. The experimental result shows that the method introduced in this paper has the advantages of simple computation and high recognition rate. It can also process the faces have different expression, the recognition rate is up to 95.08%.