Research And Access Control System Based On Face Recognition Algorithm Design And Realization

Face recognition is very important value for research and application. It is one of the hot domain nowadays. As an important component of face recognition, Facial feature extraction affects the effectiveness of face recognition systems to a large extent. In the thesis, one type of face recognition algorithms was discussed, which was subspace face recognition.The main work of this thesis includes the following aspects:(1) Firstly, the research background and significance of face recognition were depicted; secondly, the history and current status of live feature recognition was discussed. Thirdly, the contents of face recognition research and difficulties were introduced. And then, different types of face recognition algorithm were given and comparison was done on them. Finally, several different face databases that were used in face recognition were introduced.(2)Some frequent image preprocessing algorithms were discussed, such as geometric correction,median filtering, histogram equalization, image zoom.Their important roles were given. Preprocessing algorithms were the base of our following recognition algorithm.(3)Face recognition algorithm based on PCA was researched. First, its basic procedure was depicted. Second, detail face recognition algorithm based on PCA was given. Third, experiments were done on ORL database. Experiment results show the correctness of our algorithm. Finally, several key problems were researched.(4)Face recognition algorithm based on LDA was researched. Fist, its basic procedure was depicted. Second, detail face recognition algorithm based on LDA was given. Third, experiments are done on ORL database. Experiments results show the correctness of our algorithm. Finally, several key problems are researched.(5)A access control system was realized based on our algorithms discussed in the thesis.