Face detection in video sequence with complex background

Face detection is a key technique of face recognition using images in video sequences. In order to improve the accuracy of face location in video sequences, the paper presents a method for locating face in video sequences through analysis of face features in video sequences. Background removing, skin region detecting, horizontal brightness projection and face template with gradient calculating, feature component fitting are adopted for this method. The experiments demonstrate that this method can meet the requirements of actual application. In addition, face recognition-based access control system has been developed using this method.