人脸考勤系统中人脸视觉特征融合跟踪算法  

Tracking Algorithm of Face Visual Features Integration in Face Attendance System

作　　者：史万莉[1] 曾光宇[1] 高建中[1] 

机构地区：[1]中北大学信息与通信工程学院,山西太原030051

出　　处：《电视技术》2010年第6期100-101,108,共3页Video Engineering

摘　　要：基于人脸考勤系统提出了一种人脸视觉特征融合的跟踪算法,并将其扩展到多人脸跟踪场合。该算法充分利用了人脸的颜色信息和轮廓梯度信息,实现了两种特征的互补。最后,给出了人脸跟踪仿真结果。仿真结果表明,该人脸跟踪算法有效且稳健性较好,能够很好地处理人脸旋转、遮挡等问题,具有很好的应用前景。A face tracking algorithm of face visual features integration is proposed based on human face attendance system, and it is expanded to multi-face tracking occasions. The algorithm makes good use of the color information of human face and profile gradient information to achieve complement of these two features. Finally, a human face tracking simulation results are given. The experimental results show that the proposed algorithm is effective and robust well, and it is able to handle the problems as face rotation and occlusion. Thus, the algorhhn^l has a good prospect of application.

关 键 词：人脸考勤 人脸跟踪 特征融合 

分 类 号：TP391.8[自动化与计算机技术—计算机应用技术]