A Research Of Embedded Video Monitoring System Based On OpenCV

Recently, with the development of society, the video surveillance system in people’s daily lives become more important, particularly in the computer vision processing core of the intelligent video surveillance system got a great deal of attention and a lot of research. As the key technology of intelligent monitoring system, computer vision processing technology has also been rapid development, in which OpenCV is a typical representative. OpenCV provides moving target detection, moving target tracking and target identification and other large vision algorithms, so that the video surveillance system to achieve the tracking and recognition functions. In the abnormal situation timely to send a warning message, store video data and do other related operations, which not only reduces the personnel involved and system cost, but also improve the efficiency of video surveillance. Due to computer vision algorithms make full use of video surveillance systems, so that it continues towards the digital, intelligent direction.Along with the rapid development of electronic technology, communication technology and computer technology, embedded technology has also been rapid development, become one of the hottest IT technologies, its low power consumption, high performance and low cost, is widely used in many fields,for example industrial control, network communications, aerospace, instrumentation, military equipment and consumer electronics products. The main content of the embedded technology has two aspects, one is the embedded hardware technology that provides the hardware platform for embedded systems; the other is the embedded software technology, which includes not only embedded operating system, but also include embedded applications. On the hardware side, the use of the ARM RISC embedded processors typically low-power, high-performance application has great advantages, is widely used in embedded systems; On the software side, due to the embedded operating system with resource allocation and task scheduling capabilities, and to facilitate the development of embedded systems applications, etc., so embedded operating systems often are used. Embedded Linux operating system is an open source code, supports a number of processors, and has a wealth of hardware and software resources embedded operating system used widely in embedded systems.In this paper, based on the combination of video surveillance technology with an embedded video monitoring system technology trend, put forward "A Research of Embedded Video Monitoring System Based on OpenCV" subject, and by writing papers and realize system function completed the research topics.At first, this paper introduces the background and significance of the subject, and analyzes the development trends and directions of video surveillance technology and embedded technology. The second, it describes the main contents of the embedded system, starting with the definition of the three aspects, features, and structure a detailed overview of the embedded system, and then introduces embedded microprocessors and embedded operating systems:embedded processors based on different core architecture can be divided into ARM, MIPS, PowerPC, and several other common series, this article selected ARM processor with low power consumption, high performance features as the main object of study; we use open source and good real-time embedded Linux as the operating system. At last, details of the last classic computer vision library OpenCV, which covers the origins and applications from OpenCV to basic data structures and operations functions, in addition, also several common moving object detection algorithm is analyzed theoretically, and selected surendra automatic background updating algorithm to achieve the video surveillance system applications.In practice the system, based on pre-selected embedded processor and embedded operating system to build the system’s hardware and software platform. In the hardware, the use of embedded microprocessor Samsung S3C2440 as the main system controller, combined with the power, reset, memory, and system application functions required for USB and the Ethernet, which form together hardware platform of the system. This article not only describes in detail the hardware resources S3C2440 chip and circuit implementation of each module, but also introduces an effective interface circuit design, to ensure the system hardware can be stable and reliable work. In the software design, as the first choice of U-Boot bootloader, and transplanted to the development board, ready for later kernel porting, then began porting Linux kernel, and a detailed description of the migration process. In addition, in order to realize the moving target detection system, you need to transplant OpenCV library development board.Finally, the paper describes the image acquisition module, moving object detection module and data storage compression and network communication module, and the effect of moving target detection and network communications capabilities are analyzed, the results show that the application to meet the functional needs of the video surveillance system, to achieve the desired effect, and has little practical value.