中文摘要：

人脸是视频内容中的重要信息,对人脸的检测与跟踪是智能视频分析的一个重要分支.实现了基于OpenCV的人脸检测与跟踪方法.首先计算视频图像的Haar-like特征,然后利用AdaBoost级联分类器进行人脸检测,最后使用基于颜色的Camshift算法实现人脸跟踪.在VS 2010开发平台上调用OpenCV函数库进行C＋＋编程;并使用MFC框架实现了人脸检测与跟踪方法.仿真实验结果证明方法思路合理、计算复杂度较低、鲁棒性较好.

英文摘要：

Face is an important information in video,and dynamic face detection and tracking is an important part of video analysis and image recognition.This paper has achieved a method of the face detection and tracking based on OpenCV.The AdaBoost method is used to detect dynamic face,and the Camshift algorithm based on color is used to track dynamic face.The proposed method is programmed by using OpenCV library and C ＋＋ on VS 2010 development platform,and MFC framework is used to design the system interface.Simulating experimental results show that the face detection and tracking method in this paper is reasonable,which has a lower computational complexity and better robustness.