首页 | 本学科首页   官方微博 | 高级检索  
     检索      

人脸识别考勤技术研究
引用本文:王晓亚,封筠.人脸识别考勤技术研究[J].河北省科学院学报,2014(2):54-59.
作者姓名:王晓亚  封筠
作者单位:石家庄铁道大学信息科学与技术学院;
摘    要:人脸识别考勤系统因其便捷性、准确性、易于管理性,取代打卡机成为一种新兴的考勤方式。本文提出一种人脸识别考勤系统的设计方案,包括人脸识别子系统和考勤管理子系统。人脸识别子系统从获取的视频中利用Adaboost算法检测出人脸区域,对检测出的人脸区域进行归一化、去噪等处理,对处理好的图像用HOG+PCA+LDA特征提取,把提取出的人脸特征向量与注册形成的人脸特征向量利用最近邻方法比对,将识别结果放入考勤数据库中。考勤管理子系统主要实现的有考勤规则的制定、对考勤数据做统计并生成报表及系统维护等功能。

关 键 词:人脸识别  考勤机  HOG  特征降维

Research of attendance technology on face recognition
WANG Xiao-ya,FENG Jun.Research of attendance technology on face recognition[J].Journal of The Hebei Academy of Sciences,2014(2):54-59.
Authors:WANG Xiao-ya  FENG Jun
Institution:(School of Information Science and Technology ,Shijiazhuang Tiedao University ,Shijiazhuang Hebei 050043 ,China)
Abstract:This paper proposes a design of attendance system based on face recognition.This system include two subsystems,face recognition subsystems,and attendance management subsystem.The former is realized by using algorithm of Adaboost to detect and preprocess the face region from the acquired video.The features are extracted by HOG+PCA+LDA.The face features are compared with feature database using nearest neighbor method.Then the results are added to attendance database.The main functions of attendance management subsystem include attendance rules,statistics on attendance data,generating reports and system maintenance.
Keywords:Face recognition  Attendance machine  HOG  Feature reduction
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号