人脸检测技术在ATM自动识别功能拓展系统中应用  

The Application of Face Detection Techniques in ATM Automatic Identification Function Expands System

作　　者：张明慧[1] 张尧禹[1] 

机构地区：[1]中国科学院长春光学精密机械与物理研究所

出　　处：《微型电脑应用》2010年第4期24-25,共2页Microcomputer Applications

基　　金：长春市科技攻关项目(07163UC070)

摘　　要：针对ATM机存在的不安全因素,开发研制了ATM自动识别功能拓展系统。鉴于人脸特征的独一无二性,在现有的ATM机上增加人脸识别作为一种身份验证手段。而人脸检测是人脸识别的首要环节,采用了基于AdaBoost算法,构造了人脸检测系统。实验结果表明,该算法实现人脸检测时间在70ms左右,在环境较好的情况下,可有效检测单个人脸和多个人脸,满足了该系统的要求。The ATM automatic identification function expanding system is developed in the light of ATM machine safety factor.Seeing that people's face characteristic is unique,it increases the face recognition means on existing ATM as a kind of identity in order.And face detection is primary links of face recognition,the system adopts AdaBoost algorithm conforming face detection.Experiment results demonstrate that people's face detection time is about 70ms using this algorithm,and can effectively measure the single and many personal faces under better environment,meeting the demand of the system.

关 键 词：人脸检测 ATM 自动识别 ADABOOST算法 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]