The Research And Implementation Of Face Detection Based On Color And Template Matching

Human face detection is to locate the faces and determine their sizes inan image. In recent years, face detection has been paid much attention andbecome a very active research branch in pattern recognition and computervision application areas. In this paper, one novel algorithm for face detection incolor images by skin model, feature of face structure and principal component analysisis proposed. Firstly, skin color model is constructed by using a large database oflabeled skin pixels in YCbCr space. Then the images are segmented to face candidatesand background using the skin model and the face candidates are validated usingstructure features of faces. Finally, model match method with principal componentanalysis is used to the face candidates. Experimental results with VC++ show that theproposed method can detect single face from complex background in real time.