基于图像特征点的考勤管理系统

Attendance management system based on image feature points

文中针对传统考勤管理系统识别率低、安全性差、考勤速度慢等问题,提出了基于人脸图像特征点的考勤管理系统。采用AdaBoost人脸检测算法检测图像中的面部信息,并通过HOG局部特征提取算法得到图像中人脸的特征。然后与信息库进行比对,进而确定人员信息。对考勤管理系统的性能测试结果表明,在光线充足、角度正的情况下识别率达到94%;光线较弱、角度偏的情况下,识别率较低为65%。光线强度与人脸角度对识别率有较大影响。

Aiming at the problems of low recognition rate,low security and slow attendance speed of traditional attendance management system,this paper proposes a attendance management system based on face image feature points. HOG local feature extraction algorithm is adopted to obtain the features of the face in the image. Then,it is compared with the information base to determine the personnel information. According to the performance test of the attendance management system,the recognition rate reaches 94% when the light is sufficient and the Angle is positive. When the light is weak and the Angle is off,the recognition rate is 65%. Light intensity and face Angle have great influence on recognition rate.