Research of multi-scale PCA algorithm for face recognition

Aiming at the problems of complex calculation and low recognition rate of the traditional principle component analysis (PCA) face recognition algorithm etc, a new multiscale PCA algorithm for face recognition is proposed in this paper. First of all, facial images are analyzed by the multiscale algorithm so that the images with low resolution ratio and small size are got, which can reduce the image dimensions and at the same time remove the redundant information, also preferably keep the overall information of the image. Then the features are extracted by using PCA algorithm, so that the dimensions of the facial images are reduced further and the main features are highlighted. Finally a support vector machine (SVM) is used for classification and recognition. The results of the simulation experiment using ORL facial image database show that the calculation amount of the algorithm is small and its recognition effect is obviously superior to the traditional face recognition algorithms.