基于ARM的危化品仓库出入人员智能监控系统的研究与设计

发布时间：2018-04-13 02:10

  本文选题：危化品仓库 + 智能监控　； 参考：《北方民族大学》2016年硕士论文

【摘要】：近年来,石油化工产业发展迅速,危化品[1]种类不断增多,存放这些危化品的仓库因此也越来越密集,甚至有些已经深入一些居民区,它们成为了威胁附近居民生命财产安全的定时炸弹。因此,如何对危化品仓库进行管理和监控,避免仓库起火爆炸等事故的发生,成为安全部门和相关企业亟待解决的问题。据统计,人为因素依然是事故发生的主要因素,因此加强对进出危化品仓库人员的监控是解决问题的有效途径。本课题利用计算机视觉技术、嵌入式技术以及传感器技术等,在对国内外现有相关系统进行深入分析的基础上,研究与设计了一套基于ARM的危化品仓库出入人员智能监控软件系统。本文首先介绍了智能监控系统的相关理论及技术,包括运动目标检测、人脸识别以及目标跟踪等算法,并提出了基于混合高斯背景建模与CamShift算法结合的运动目标跟踪算法;接着,综合分析了本系统的功能需求,并结合实际情况,对系统进行了总体分析与设计,将系统划分为前端子系统和后端子系统两大部分,前端子系统采用了以S3C6410 CPU芯片为核心的硬件平台,以嵌入式Linux系统为核心的软件平台及Qt Creator等开发工具完成对监控数据的实时采集、处理、发送等,后端则对前端子系统发送的数据进行显示、处理并提供及时的预报警功能;然后,本文对系统的各个功能模块进行了详细的设计,并且介绍了软硬件开发环境的配置与搭建,本文视频分析模块采用了OpenCV2编程,OpenCV是一个开源的跨平台的计算机视觉库,利用该视觉库极大简化了视频分析模块的开发难度;最后,本文对系统实现方式给出了简要的介绍,并对系统进行了测试,测试结果表明初步实现了系统的需求。
[Abstract]:In recent years, the rapid development of petroleum and chemical industry, hazardous chemicals, [1] growing, storage of these hazardous chemicals warehouse and therefore more and more intensive, and even some have been deep in some neighborhoods, they become a bomb threat to nearby residents life and property safety. Therefore, how the hazardous goods warehouse management and monitoring. Avoid warehouse fire explosion and other accidents, become the urgent security departments and related enterprises. According to statistics, human factors are the main factors of the accident, therefore to strengthen monitoring of the import of hazardous chemicals warehouse personnel is an effective way to solve the problem. The subject of the use of computer vision technology, embedded technology and sensor technology and in-depth analysis based on the existing domestic and foreign system, research and design of a ARM based warehouse staff access intelligent monitoring software System. This paper firstly introduces the related theory and technology of the intelligent monitoring system, including the moving target detection, face recognition and target tracking algorithm, and put forward the moving target tracking algorithm combined with the mixed Gauss background modeling and algorithm based on CamShift; then, the comprehensive analysis of the functional requirements of the system, combined with the actual situation of system analysis and overall design, the system is divided into front-end system and terminal system of two parts, the front terminal system uses S3C6410 CPU chip as the core of hardware platform, the embedded Linux system as the core of the software platform and Qt Creator development tools to monitor real-time acquisition, data processing, transmission etc. the back-end of sending terminal, system data display, provide timely warning and treatment; then, the system can work on the various modules of the The detailed design, and introduces the development environment of software and hardware configuration and set up in this paper, the video analysis module uses the OpenCV2 programming, OpenCV is a cross platform computer vision library, an open source, which greatly simplifies the difficulty of the development of video analysis module using the visual library; most after the system implementation gives a brief is introduced, and the system is tested, test results show that the initial implementation of the system requirements.

【学位授予单位】：北方民族大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP277
，

本文编号：1742453