Realization Of Face Detecting System Based On Complexion And Graphical Hybrid Modeling Method

As the rapid increasingly demand since the 1990s, the discussions of Face Detecting become more and more enthusiastic. As the feature of direct, friendly convenient, fleet and non-invasive Face Detecting became the most acceptable Method of biological detection. Because Face Detecting is a very important part of face recognition, face image retrieval, videoconference, access-exit control and human-computer interaction, Face Detecting became a valuable and potential independent research topic.Due to face is a complex construction nature object configuration, Face Detecting is a very valuable and challenging science domain. The challenge lies in:(1) Face has the variable pattern feature because the different of appearance, expression and complexion; (2) There likely to be some screening like spectacles or beard; (3) There will be obviously illumination effect of Face because it is a three dimensional body.If Face Detecting System is successfully constructed, it will provide a important enlightenment to the research of other complex pattern detection.In the part of face detecting position and face feature extraction and recognition introduction in this thesis, the existent research findings have been detailed summarized and classified. Based on the research of existent results, this thesis presents a Complexion and Graphical Hybrid Modeling Method and achieves all the algorithms with C++program. The characteristic of Complexion and Graphical Hybrid Modeling Method raise rate of accuracy and reduce system overhead.