Abstract:

In order to solve the problem of low recognition rate caused by the interference of irresistible factors， a system is constructed to collect local and global features of face images and fuse them for recognition. The principal component analysis （PCA） and local binary pattern （LBP） are used to extract the global feature matrix and LBP local feature spectrum of the image， respectively， to construct an associated set of face image sets and extract face feature values. The recognition effects of a single PCA algorithm and a PCA+LBP algorithm in different dimensions and different sample numbers are compared through the ORL database face image set. At the same time， a GUI interface based on support vector machine algorithm and Matlab is designed to open， train， recognize and classify face images. The results show that the accuracy of face recognition can be improved by the fusion of local and global extracted feature values， and the face recognition system of GUI interface can meet the application requirements.

Key words: face recognition, principal component analysis, local binarization, vector machine classifier