A face detection method based on corner verifying

Face detection is the first step for automatic face recognition system and many surveillance systems, it has been widely applied in many fields. A face detection method based on corner verifying is presented. Firstly, most of the background area is quickly filtered out by skin color detection. Then faces are detected by AdaBoost face detection algorithm combination with skin color detection, but there are some non-faces to be detected. Finally, the faces are verified by corner distribution of human face after Harris corner detection to further reduce the error rate. Experimental results show that this method can efficiently and accurately detect the faces with multi-feature fusion.