Study Of Color And Template-based Face Detection Algorithm

Human face detection is to locate the faces and determine their sizes in an image. In recent years, face detection has been paid much attention and become a very active research branch in pattern recognition and computer vision application areas.In this paper, we present a multi-feature optimal fusion algorithm, inclusive of skin color, to detect one or multiple faces in color image with complex background. It is a hierarchical approach and integrates the skin color segmentation, face template matching and a neural network frontal face detector. With the elimination of false areas, the search area will become smaller and smaller, and the detection will be accomplished eventually. The main contributions are as follows:Firstly, the YCbCr information is used to extract the skin color area. On the basis of skin color model, a new skin color segmentation method is proposed in which the threshold will be optimized according to the Fisher criterion. The experimental results show that this method is feasible and effective.Secondly, according to the possessive area ratio and the ratio between height and width of the skin region, the error regions can be eliminated and the candidate face areas can be obtained.Finally, the face template matching and BP neural network are used to verify whether the probable face area is a face indeed, and the final detection result will be obtained.Experimental results show that the algorithm is feasible and effective.