摘要
提出了一种用于智能访客系统的人脸检测及人脸方向自动识别的算法。算法首先利用皮肤颜色过滤法和背景去除技术提取出类似皮肤的区域,再利用区域填充和基于高度落差的方法将人脸分割出来,最后,利用人眼的位置信息确定人脸方向。实验结果证明了此算法的可行性和实时性。
An effective and robust face tracking algorithm is proposed to locate and track the user＇s face in video sequence, The algorithm relies on a two - step process in which skin- like regions are first segmented by using both skin color filter and background removal technique. Then the face can be detected utilizing region filling processing and image analysis technique. In the end of this paper, a face direction distinguishing method based on the former face detection result, which can automatically recognize the direction of face through the location of people＇s eyes is presented. The experiment results have demonstrated the feasibility and near real time performance of the proposed algorithm.
出处
《计算机应用与软件》
CSCD
北大核心
2007年第5期160-162,共3页
Computer Applications and Software
关键词
智能访客系统
人脸检测
人脸方向判别
Intelligent reception system Face detection Face direction recognition
