摘要： 给出了一种视频图像中的实时人脸检测方法,该方法综合了彩色视频图像的运动信息和颜色信息,可以快速地对图像中的人脸区域进行定位。算法通过对视频图像序列中每连续3帧图像进行对称差分,提取出运动区域;然后基于肤色聚类模型,再对运动区域进行肤色检测,经过候选人脸验证,最终定位图像中的人脸。实验表明,提出的方法检测速度快,实现简单、高效,满足实时系统的要求。

关键词: 人脸检测, 对称差分, 运动区域提取, 肤色聚类模型

Abstract: A real-time algorithm for detecting faces in video images is presented, which is composed of the motion information and color information. First, three continuous video frames symmetrical differences are used to acquire the motion area of the moving objects. Then, on the base of the skin-color clustering model, search the faces in the motion area. The candidate face regions are examined with knowledge of human face. The experimental results show that the proposed algorithm has high speed and high efficiency, so it can be used in the real-time system.

中图分类号: 

• TP391.41