| Face Detection is an important research content in modern image processing, and which has great value in many fields. Such as Identity Verification, security access control, Visual Surveillance and other applications fields. The target of Face Detection is to judge whether there are faces in images. and if so, then output the face's location and scope.In recent years, the method based on the face color has taken more attention from researchers, proved to be a significant and effective feature,especially in the Fast Face Detection and Tracking System. The method based on face color is one of face detection methods,which needs deferent improvement in real application.The researches in this paper aim at face detection based on color images with complex background. The mostly job is as follows:1 .researching a suitable illumination compensateon method. In order to reduce the disbenefit factor of illumination to face detection, bring forward a method of illumination compensateon named intensity reference by aimed at the character of that face color is sensitive to illumination.2.setting up a Skin Gaussian Model adapted to alterable light. The Gaussian Model skin color segmentation under YCbCr chrominance space method is choose by the thesis to construct the Gaussian Model for skin color under YCbCr and get the likely region of face.3.designing mouth(eye) deformable template.The average mouth(eye) template is used to verify whether the probable mouth(eye) area is a mouth(eye) indeed. In this phase to solve the problem that the really mouth(eye) have some inclination and size is not always same, this thesis adjust the stand template thought calculate candidate mouth(eye) areas angle and area size. It can promote the veracity of template matching. At same time it avoid use different size template to try multi-times, the algorithm have higher efficiency.4.based on above detection methods,developing a face detection system in video sequence image. |
PDF Full Text Request