基于图像识别的智能视频监控模块研究
发布时间：2018-08-05 16:31
【摘要】：近年来,随着社会的发展,科学技术的进步,监控技术的应用在保证社会稳定安全发展中的作用也越来越被突显出来。在目前的相关监控技术中,大多数传统的视频监控系统仍然需要人对视频内容进行分析辨别,这造成了极大的资源浪费,而且随着监控技术被应用的领域越来越广,人们对于监控系统的智能化的需求也日益迫切。 本文在深入研究了监控系统的智能化相关实现技术以及图像处理技术后,以智能视频监控系统在监控区域对人脸进行识别,以及在识别出人脸后对采集的人脸图像进行缩放,由此解决监控人员的对视频内容分析辨别困难的问题作为本文研究的重点。围绕TI公司所开发的新型数字信号处理开发板TMS320DM6446设计了一套基于图像识别技术的智能视频模块,并依据此视频模块的架构构建一套智能视频监控系统方案。 本系统在硬件方面以TMS320DM6446为核心,针对智能视频监控系统的需求,分别对,视频采集模块、数据存储模块、视频后端处理模块、视频传输模块、开关量控制模块、以及电源供电模块进行了相关硬件设计。使本系统达到性能与成本的最佳平衡。 本系统在软件方面基于人脸识别原理与图像缩放技术,采用识别正确率较高的AdaBoost算法和图像缩放效果较好的缝隙剪裁技术(Seam Carving)设计了系统的主要软件框架,并分析了实现软件模块的关键代码。并在通过软件仿真验证本系统所设计的相关软件模块能够取得较为良好的人脸识别与图像缩放效果。
[Abstract]:In recent years, with the development of society and the progress of science and technology, the application of monitoring technology in ensuring social stability and security development has become more and more prominent. At present, most of the traditional video surveillance systems still need people to analyze and distinguish the video content, which has caused a great waste of resources, and with the application of monitoring technology has become more and more extensive. The demand for intelligent monitoring system is becoming more and more urgent. In this paper, the intelligent correlation technology and image processing technology of the monitoring system are deeply studied, then the face is recognized by the intelligent video surveillance system in the monitoring area, and the collected face image is scaled after the recognition of the face. It is the focus of this paper to solve the problem of video content analysis and discrimination. A set of intelligent video module based on image recognition technology is designed around TMS320DM6446, a new digital signal processing board developed by TI, and a set of intelligent video surveillance system scheme is constructed according to the structure of this video module. This system takes TMS320DM6446 as the core in hardware, aiming at the demand of intelligent video surveillance system, respectively, video acquisition module, data storage module, video back-end processing module, video transmission module, switch control module, etc. And the related hardware design of power supply module. Make the system achieve the best balance between performance and cost. Based on the principle of face recognition and image scaling technology, this system designed the main software framework of the system by using the AdaBoost algorithm with high recognition accuracy and the slot cutting technology (Seam Carving), which has good image scaling effect. The key code to realize the software module is analyzed. The software simulation shows that the software module designed in this system can achieve better face recognition and image scaling effect.
【学位授予单位】：中北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN948.6;TP391.41
本文编号：2166353
[Abstract]:In recent years, with the development of society and the progress of science and technology, the application of monitoring technology in ensuring social stability and security development has become more and more prominent. At present, most of the traditional video surveillance systems still need people to analyze and distinguish the video content, which has caused a great waste of resources, and with the application of monitoring technology has become more and more extensive. The demand for intelligent monitoring system is becoming more and more urgent. In this paper, the intelligent correlation technology and image processing technology of the monitoring system are deeply studied, then the face is recognized by the intelligent video surveillance system in the monitoring area, and the collected face image is scaled after the recognition of the face. It is the focus of this paper to solve the problem of video content analysis and discrimination. A set of intelligent video module based on image recognition technology is designed around TMS320DM6446, a new digital signal processing board developed by TI, and a set of intelligent video surveillance system scheme is constructed according to the structure of this video module. This system takes TMS320DM6446 as the core in hardware, aiming at the demand of intelligent video surveillance system, respectively, video acquisition module, data storage module, video back-end processing module, video transmission module, switch control module, etc. And the related hardware design of power supply module. Make the system achieve the best balance between performance and cost. Based on the principle of face recognition and image scaling technology, this system designed the main software framework of the system by using the AdaBoost algorithm with high recognition accuracy and the slot cutting technology (Seam Carving), which has good image scaling effect. The key code to realize the software module is analyzed. The software simulation shows that the software module designed in this system can achieve better face recognition and image scaling effect.
【学位授予单位】：中北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN948.6;TP391.41
【参考文献】
相关期刊论文 前4条
1 徐伯庆,孙国强,陈离;数字图像缩放的原理与硬件实现[J];上海理工大学学报;2002年03期
2 王文宁;李慧娟;师磊;;一种基于颜色和形状特征的人脸检测方法[J];计算机系统应用;2008年07期
3 白艳茹;刘贺平;朱珍民;罗海勇;先明春;高玉之;;基于TMS320DM6446的多媒体视讯终端[J];仪表技术与传感器;2008年08期
4 王华斌;;C64+对C64内核的性能提升[J];电子设计应用;2006年06期
，本文编号：2166353
本文链接：https://www.wllwen.com/kejilunwen/wltx/2166353.html
