Face Recognition Technology Research And Implementation

As one of the most successful applications of image analysis and understanding, face recognition has recently received significant attention, especially during the few years. The goal of our research is to develop a face detection and recognition system that can be used in a real-time face identifications.Two face detection methods are proposed in our prototype system, they are base on skin color model in HSV chrominance space and Haar-like feature. The face recognition approach that we proposed involves the use of Embedded Hidden Markov Models. The advantage of the HMM-based approach is its ability to handle variations in scale, which is a challenging problem for any face recognition system.In order to enhance the robustness of the system, we use dynamic histogram sampling and image normalization at face detection and recognition stages. Finally we discussed the approach to reduce the computational complexity of the Doubly Embedded Viterbi Algorithm.The face recognition system has been tested on the ORL database and Yale database. The recognition results are high as 89. 9%, 95%.