基于OpenCV双摄像头活体人脸判定及实时跟踪系统.docVIP

基于OpenCV双摄像头活体人脸判定及实时跟踪系统 　　[摘 要]活体检测及人脸跟踪系统的构建已成为现代计算机视觉及OpenCV研究的热点。提出一种基于OpenCV的双摄像头活体人脸判定实时人脸跟踪系统。一种新的双摄像头活体人脸判定的方法,利用基于HAAR特征人脸检测算法,并判断距离,编写Windows程序调用串口API和下位单片机STC12C5410通信,采用CH341A芯片将RS232串口数据加载到USB协议数据栈中实现USB传输,单片机控制FUTABA舵机,获取上位机人脸检测坐标信息,实时调整舵机转动角度,使人脸始终处于图像中间290~310阈值范围内,达到人脸跟踪。该系统结构精简,可作为模块用于门禁安保系统及构建其他机器人研究平台。 　　[关键词]OpenCV 活体检测 人脸跟踪 CH341A 　　[中图分类号]TP391 [文献标识码]A [文章编号]1007-9416(2010)07-0094-02 　　 　　The Dual Cameras-Vivo-Face Recognition and Real-time Tracking System based on OpenCV 　　 　　SONG Hong-jia1, HAO Ji-jun1, WANG Jian-jun2, Wang Jing-xuan1, ZHUANG Fei-fei3, WANG Jin3 　　(1. School of Information Science and Engineering,Shandong University,Jinan 250100,China 2. School of Mechanical Engineering,Shandong University,Jinan 250061,China 3.School of Control Science and Engineering,Shandong University,Jinan 250061,China) 　　 　　[Abstract] Vivo detection and face tracking system has become a hotspot of modern computer vision and OpenCV research.. The dual cameras-vivo-face recognition and real-time tracking system based on OpenCV is presented. A new method of dual cameras-vivo-face recognition, account of the HAAR face detection algorithm which can judge distance, is applied. Serial API and microcontroller STC12C5410 can achieve the call communication by writing program of windows, adding the RS232 serial data to the USB protocol stack with CH341A chip to implement USB data transmission. The microprocessor controls the FUTABA Servo to obtain the coordinate information of face detection from the PC, with real-time adjustment of the servo rotating angle to guarantee the face is always in the middle of the image within a certain threshold. This lean system can be used as a module in access security systems and the construction of robots research platform. 　　[Key words] OpenCV, Vivo detection, Face tracking,CH341A 　　 　　0 引言 　　随着现代技术的提高,暴露出利用各种手段破解认证系统的问题。运动目标检测是现代智能视觉监控系统的重要组成部分,在小区安保门禁系统中,人脸追踪在有效提高破案率方面有较广泛应用,加入活体检测程序可以有效防范造假情况发生。目前图像采集采用PCI图像采