Face Detection And Recognition System Based On Support Vector Machine Technology

Face recognition is a significiant challenge to the pattern recognition research community. It also becomes an active research topic recently. But to data, it is still far from satisfactorily solved. Human faces are very similar in structure with minor differences from person to person. Furthermore, lighting condition changes, facial expressions, and pose variations further complicate the face recognition task as one of the difficult problems in pattern analysis.The main contributions of this paper are as follows::1. A real-time face detection and recognition system based on SVMs is presented.2. A frame of multi-class classification based on the coding is presented. In this frame, any multi-class classification method based on 2-class classification is just one case of coding. Further more, the paper presents a simple multi-class classification method for face recognition.3. A simple method is presented to get virtual multi-view face images from single view face image. This method improves the correct rate of recognition when there is only one image for every person.