Study. Based On Ehmm Model Of Face Recognition Algorithm

Automated face recognition is an important part of pattern recognition. It is one of the most active and challenging tasks for computer vision and pattern recognition. Approaches to real time frontal face detection and recognition are mainly researched in this dissertation, and an automated face recognition simulation system is established in the end. The essential contents can be listed as follows:1. A hidden Markov model method (HMM) is used to recognize face in our automated recognition system. Firstly how to establish the model of face is researched. The intensity characteristic values of human face and the interdependence are related with the identical state transfer model, not using the numerical characteristic of each organ. Then the coefficient of Discrete Cosine Transform (DCT) is applied to observation vector. Owing to the rapid calculation and compressibility of DCT, the HMM of this dissertation outperforms the ones using the intensity values. The experiments verified the consideration. Finally, the simulation of this face recognition is accomplished.2. A method based on Embedded Hidden Markov Model (EHMM) and DCT feature is presented. EHMM increases the states in horizontal direction and is a simplified 2D HMM. The EHMM can describe the face more elaborately and more completely. The experiment using the ORL face database has acquired a recognition rate of over 99.3% To our knowledge, this is also the better result using this database up to now.3. The algorithmof face recognition and design of interface are completed under Visual C++ 6.0 and OpenCV. OpenCV makes researchers devote more to the arithmetic itself, not considering how to program to complete it.