Face recognition is one of the methods of biometrics. Currently in pattern recognition and image processing it is a quite active research topic. In the process of face recognition, feature extraction is an important part for the recognition result. On the basis of the traditional principal component analysis and improved 2DPCA, this article proposes a new method of weighted modular 2DPCA. In this method face image will be divided into three parts and implemented face extraction separately. Considering the condition that the amount of information included in each part of face image is unequal, in the classification endow different parts with different weights. And then, use weighted modular 2DPCA and Bayesian method for face recognition and achieves better recognition results. |