An advanced BPNN face recognition based on curvelet transform and 2DPCA

The paper proposed an advanced face recognition method of BPNN based on Curvelet transform and 2DPCA to increase the face recognition rate.. Firstly, we used Curvelet transform to process the face images, after that we got higher dimension feature of face images, then used 2DPCA to reduce the dimension, finally we got some feature vectors, which could represent the face images. We used ORL face database to conduct the experiment, some of the feature vectors to train the BPNN classifier, the others to test. we repeatedly made some of the less important feature vectors value replaced by zeros, and made the recognition result depend on big probability event . We obtained a better result In this way compared with traditional BPNN face recognition, this method can not only improve the face recognition rate, but also reduce the recognition time.