基于均值平移的自动人脸检测与跟踪算法  

Algorithm for Auto Face Detection and Tracking Based on Mean Shift

作　　者：王展青[1] 李健[1] 童恒庆[1] 

机构地区：[1]武汉理工大学理学院,湖北武汉430070

出　　处：《武汉理工大学学报（信息与管理工程版）》2009年第1期5-7,15,共4页Journal of Wuhan University of Technology：Information & Management Engineering

摘　　要：针对彩色图像序列的人脸跟踪问题,提出了一种基于均值平移的自动人脸检测与跟踪算法。对每帧图像采用基于肤色和支持向量机的算法检测人脸,可以减少计算量。对于检测到的人脸,判断是否含有新增人脸,有则将新的人脸加入跟踪序列,否则继续跟踪原有人脸序列。然后采用均值平移算法对人脸序列进行跟踪,其计算量小,跟踪稳定可靠,从而实现自动的人脸检测与跟踪。实验结果表明,该算法可以很好地进行人脸的自动检测跟踪。A method of human face detection and tracking based on Mean - Shift algorithm in color image sequences was presented. The human face for each frame was detected by the algorithm based on skin color and SVM. The cost of calculation was reduced. The judgment on whether any new faces are contained in the detected faces depends on whether there exists one. If the answer is yes,then the new face area should be inserted into the tracking sequence ,otherwise the original sequence is remained. This tracking algorithm based on Mean -Shift has less calculation and reliable performance,which thereby implements the automatic human face tracking. The experiment results show that this algorithm could deal with the automatic human face detecting and tracking efficiently.

关 键 词：人脸检测 人脸跟踪 均值平移 肤色判别 支持向量机 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]