人脸识别在远程智能监控系统中的研究与实现
|
摘要
文中设计一种远程智能视频监控系统,通过加入人脸识别技术,对视频中感兴趣的人脸进行提取并识别。考虑到使用便利性以及成本节约,系统采用嵌入式微处理器平台,并使用Linux作为操作系统。在底层利用V4L2来驱动摄像头输入,利用TCP/IP网络通信协议进行传输。系统采用的是B/S架构,通过浏览器的形式可以查看实时视频监控画面。为了改进普通视频监控的局限性,加入人脸识别技术使其可以自主地筛选出视频中出现人员的信息。采用Adaboost算法检测出人脸,然后通过训练人脸模型库,利用LBP算法对检测出的人脸进行识别。传统的被动视频监控系统存在视频数据利用率不高、资源浪费、增加人工成本等问题。与之相比,基于人脸识别的视频监控系统能够主动地识别视频中人员信息,保存识别的人脸数据,用户可以方便查看视频中出现的人员信息,有着较好的实用性。
A remote intelligent video monitoring system is designed in this paper to extract and recognize the interested human faces in the video by adding the human face recognition technology. Taking the usage convenience and cost savings into account,the embedded microprocessor platform and operating system Linux are adopted for the system. In the bottom layer,the V4 L2 is used to drive the camera input,and the TCP/IP network communication protocol is used for transmission. The B/S architecture is adopted for the system. The real-time video monitoring screen can be viewed in the form of browser. The human face recognition technology is added to make it possible to autonomously filter the personnel information appearing in the video,so as to improve the limitations of ordinary video monitoring. The human faces are detected by using the Adaboost algorithm,and then recognized by training the human face model library and using the LBP algorithm. In comparison with the traditional passive video monitoring system which has the problems of low utilization rate of video data,waste of resources,and increase of labor costs,the video monitoring system based on human face recognition can actively recognize the personnel information in the video,save the recognized human face data,and make users conveniently view the personnel information appearing in the video,which has a good practicality.
A remote intelligent video monitoring system is designed in this paper to extract and recognize the interested human faces in the video by adding the human face recognition technology. Taking the usage convenience and cost savings into account,the embedded microprocessor platform and operating system Linux are adopted for the system. In the bottom layer,the V4 L2 is used to drive the camera input,and the TCP/IP network communication protocol is used for transmission. The B/S architecture is adopted for the system. The real-time video monitoring screen can be viewed in the form of browser. The human face recognition technology is added to make it possible to autonomously filter the personnel information appearing in the video,so as to improve the limitations of ordinary video monitoring. The human faces are detected by using the Adaboost algorithm,and then recognized by training the human face model library and using the LBP algorithm. In comparison with the traditional passive video monitoring system which has the problems of low utilization rate of video data,waste of resources,and increase of labor costs,the video monitoring system based on human face recognition can actively recognize the personnel information in the video,save the recognized human face data,and make users conveniently view the personnel information appearing in the video,which has a good practicality.
引文
[1] WOJEK C,DOLLAR P,SCHIELE B,et al. Pedestrian detection:an evaluation of the state of the art[J]. IEEE transactions on software engineering,2012,34(4):743-761.
[2]黄凯奇,陈晓棠,康运锋,等.智能视频监控技术综述[J].计算机学报,2015,38(6):1093-1118.HUANG Kaiqi,CHEN Xiaotang,KANG Yunfeng,et al. Intelligent visual surveillance:a review[J]. Chinese journal of computers,2015,38(6):1093-1118.
[3]吴群,王田,王汉武,等.现代智能视频监控研究综述[J].计算机应用研究,2016,33(6):1601-1606.WU Qun,WANG Tian,WANG Hanwu,et al. Survey on modern intelligent video surveillance[J]. Application research of computers,2016,33(6):1601-1606.
[4] Markets and Markets. Video surveillance market-global forecast&analysis[EB/OL].[2014-11-01]. http://www.marketsandmarkets.com/Market-Reports/video-surveillance-market-645.html.
[5]王海峰.传统报警与视频智能结合及报警联网发展方向探讨[J].中国安防,2017(3):87-91.WANG Haifeng. Discussion on intelligent combination of traditional alarm and video and development direction of alarm networking[J]. China security&protection,2017(3):87-91.
[6]叶剑锋,王化明.AdaBoost检测结合SOM的自动人脸识别方法[J].哈尔滨工程大学学报,2018,39(1):129-134.YE Jianfeng,WANG Huaming. An automatic face recognition method using AdaBoost detection and SOM[J]. Journal of Harbin Engineering University,2018,39(1):129-134.
[7]朱黎,胡涛,罗锋,等.基于自适应LBP人脸识别的身份验证[J].计算机工程与科学,2014,36(11):2217-2222.ZHU Li,HU Tao,LUO Feng,et al. Identity authentication based on adaptive LBP face recognition[J]. Computer engineering&science,2014,36(11):2217-2222.
[8]曹洁,朱晶晶,李伟,等.融合LSH和LoG特征的人脸识别[J].计算机工程与应用,2018,54(17):158-163.CAO Jie,ZHU Jingjing,LI Wei,et al. Face recognition based on fusing LSH and LoG feature[J]. Computer engineering and applications,2018,54(17):158-163.
[9] LI H. Design and implementation of mobile video surveillance system based on V4L2[J]. International journal of signal processing,image processing and pattern recognition,2016,9(1):409-416.
[10]陈金超,谢东亮.无线网络TCP拥塞控制算法研究综述[J].软件,2015,36(1):82-87.CHEN Jinchao,XIE Dongliang. Survey on TCP optimization in wireless network[J]. Computer engineering&software,2015,36(1):82-87.
[2]黄凯奇,陈晓棠,康运锋,等.智能视频监控技术综述[J].计算机学报,2015,38(6):1093-1118.HUANG Kaiqi,CHEN Xiaotang,KANG Yunfeng,et al. Intelligent visual surveillance:a review[J]. Chinese journal of computers,2015,38(6):1093-1118.
[3]吴群,王田,王汉武,等.现代智能视频监控研究综述[J].计算机应用研究,2016,33(6):1601-1606.WU Qun,WANG Tian,WANG Hanwu,et al. Survey on modern intelligent video surveillance[J]. Application research of computers,2016,33(6):1601-1606.
[4] Markets and Markets. Video surveillance market-global forecast&analysis[EB/OL].[2014-11-01]. http://www.marketsandmarkets.com/Market-Reports/video-surveillance-market-645.html.
[5]王海峰.传统报警与视频智能结合及报警联网发展方向探讨[J].中国安防,2017(3):87-91.WANG Haifeng. Discussion on intelligent combination of traditional alarm and video and development direction of alarm networking[J]. China security&protection,2017(3):87-91.
[6]叶剑锋,王化明.AdaBoost检测结合SOM的自动人脸识别方法[J].哈尔滨工程大学学报,2018,39(1):129-134.YE Jianfeng,WANG Huaming. An automatic face recognition method using AdaBoost detection and SOM[J]. Journal of Harbin Engineering University,2018,39(1):129-134.
[7]朱黎,胡涛,罗锋,等.基于自适应LBP人脸识别的身份验证[J].计算机工程与科学,2014,36(11):2217-2222.ZHU Li,HU Tao,LUO Feng,et al. Identity authentication based on adaptive LBP face recognition[J]. Computer engineering&science,2014,36(11):2217-2222.
[8]曹洁,朱晶晶,李伟,等.融合LSH和LoG特征的人脸识别[J].计算机工程与应用,2018,54(17):158-163.CAO Jie,ZHU Jingjing,LI Wei,et al. Face recognition based on fusing LSH and LoG feature[J]. Computer engineering and applications,2018,54(17):158-163.
[9] LI H. Design and implementation of mobile video surveillance system based on V4L2[J]. International journal of signal processing,image processing and pattern recognition,2016,9(1):409-416.
[10]陈金超,谢东亮.无线网络TCP拥塞控制算法研究综述[J].软件,2015,36(1):82-87.CHEN Jinchao,XIE Dongliang. Survey on TCP optimization in wireless network[J]. Computer engineering&software,2015,36(1):82-87.