基于智能视频处理技术的监狱监控系统设计与实现

发布时间：2018-03-29 15:37

本文选题：嵌入式　切入点：多线程编程　出处：《中山大学》2014年硕士论文

【摘要】：在人工智能技术和嵌入式技术不断发展的今天，，模式识别技术有了长足的发展，也使得人们对视频监控有了更高的要求。而基于智能识别技术的视频监控在嵌入式产品中的应用也越来越广泛。具体到监狱安防领域，其对视频监控产品更是提出了实时监控和预警的需求。在这种需求的激励系啊，也为了能够更好地完善监狱安防工作，及时发现被监控对象的异常行为并实现实时预警，文中重点对基于视频处理技术的监狱监控系统进行设计与实现。在文中系统的设计和实现过程中，所采用的核心芯片为S3C2410，同时以国内某公司所研发的ARM9开发板基础，对一种多层次实时监控和预警的智能监控网络系统进行开发。该系统以嵌入式的Web视频实时监控设备为实现核心，在结合网络通信技术和图像识别技术的基础上，实现以本地视频智能分析预警为中心、以及区、市和省级监控中心为骨干的多层次网络。从而能够更好的满足监狱安防这一特殊行业对安防产品的要求。在本文系统中，微处理器采用S3C2410，而嵌入式的操作系统则主要采用Linux系统，从而有效节约了系统的研发和使用成本。其中，系统中的软件部分还包括有对摄像头的驱动和采集。而采集过程又可以分为单帧/多帧图像的采集和视频流采集。在软硬件环境构建的基础上，系统还对基于C/S模式的网络视频实时监控、网络实时图像智能分析和多级自动报警等功能进行了详细说明。此外，通过对文中内容的深入研究，文中作者对各种嵌入式编程技术、多线程编程、图像分割技术等进行了深入分析。同时，还对嵌入式系统的软件开发流程以及嵌入式Linux的远程调试等进行了分析和挖掘，为系统的开发和调试奠定基础，有效缩短了系统的研发周期。
[Abstract]:With the continuous development of artificial intelligence and embedded technology, pattern recognition technology has made great progress. It also makes people have higher requirements for video surveillance. And video surveillance based on intelligent identification technology is more and more widely used in embedded products. It has put forward the demand for real-time monitoring and early warning for video surveillance products. In the incentive system of this demand, ah, in order to better improve prison security work, to detect the abnormal behavior of the monitored objects in time and to realize real-time early warning, This paper focuses on the design and implementation of the prison monitoring system based on video processing technology. In the design and implementation of the system, the core chip is S3C2410, and the base of the ARM9 development board developed by a domestic company. An intelligent monitoring network system with multi-level real-time monitoring and early warning is developed. The system uses embedded Web video real-time monitoring equipment as the core, and combines network communication technology and image recognition technology. To realize the multi-level network with local video intelligent analysis and warning as the center and the district, city and provincial monitoring center as the backbone, which can better meet the requirements of the security products of the special industry of prison security. In this system, the microprocessor uses S3C2410, while the embedded operating system mainly adopts Linux system, which saves the cost of research and development of the system. The software of the system also includes the drive and acquisition of the camera, and the acquisition process can be divided into single-frame / multi-frame image acquisition and video stream acquisition. Based on the construction of software and hardware environment, the functions of real-time monitoring of network video based on C / S mode, intelligent analysis of network real-time images and multi-level automatic alarm are described in detail. In this paper, various embedded programming techniques, multi-thread programming and image segmentation techniques are deeply analyzed. At the same time, the software development flow of embedded system and remote debugging of embedded Linux are analyzed and mined. It lays a foundation for the development and debugging of the system and effectively shortens the research and development cycle of the system.
【学位授予单位】：中山大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN948.6

【参考文献】