面向视频监控的时空上下文算法的研究

发布时间：2018-06-27 03:26

本文选题：时空上下文 + 树莓派　；参考：《郑州大学》2017年硕士论文

【摘要】：本文在分析近年目标跟踪的基本理论和最新算法的基础上,根据跟踪算法时空上下文,提出了一种改进算法。时空上下文算法利用目标周围时间和空间的上下文信息进行跟踪,算法实时性好,鲁棒性强,具有较高的应用价值。但当目标受到严重遮挡时,难以利用该算法重新寻回目标。本文根据时空上下文算法的特点,结合加速鲁棒特征转换算法,提出了改进型时空上下文算法,有效地检测时空上下文算法的错误更新,并予以纠正。其次,在目标受到严重遮挡时,利用改进型算法能够重新寻回目标,提高了算法的准确性,实验表明改进型算法在原时空上下文算法的基础上有了较大提高。本文搭建了视频无线监控系统,并将改进型算法移植到该系统中,对可疑目标进行跟踪,验证了改进型算法的有效性。硬件采用树莓派开发板作为硬件的核心模块,罗技C270摄像头作为视频的采集端,实现监控系统图像的采集。并对嵌入式处理器的架构和操作系统进行了分析和配置,通过WIFI无线网络实现了树莓派开发板与计算机之间的Socket通信,利用无线网络实现客户端和服务器端之间数据的实时传输。基于OpenCV视觉库实现无线视频的压缩,在计算机端应用OpenCV视觉库来实现系统视频的接收、播放。通过MFC开发出易操作的人机交互界面,结合上述技术,实现视频的无线监控和可疑目标的跟踪。测试结果表明,系统工作可靠、稳定,具有较高的实用和研究价值。
[Abstract]:Based on the analysis of the basic theory and the latest algorithms of target tracking in recent years, an improved algorithm is proposed according to the temporal and spatial context of the tracking algorithm. The spatio-temporal context algorithm uses the temporal and spatial context information around the target to track. The algorithm is real-time and robust, and has high application value. However, when the target is seriously blocked, it is difficult to find the target again by using the algorithm. According to the characteristics of the spatio-temporal context algorithm and the accelerated robust feature transformation algorithm, an improved spatio-temporal context algorithm is proposed in this paper, which can effectively detect the error update of the spatio-temporal context algorithm and correct it. Secondly, when the target is severely occluded, the improved algorithm can find the target again and improve the accuracy of the algorithm. The experimental results show that the improved algorithm has been greatly improved on the basis of the original spatio-temporal context algorithm. In this paper, a video wireless monitoring system is built, and the improved algorithm is transplanted to the system to track suspicious targets, which verifies the effectiveness of the improved algorithm. The Raspberry pie development board is used as the core module of the hardware and Logitech C270 camera is used as the video acquisition terminal to realize the image acquisition of the monitoring system. The architecture and operating system of embedded processor are analyzed and configured. The socket communication between raspberry pie development board and computer is realized through WiFi wireless network. Real-time data transmission between client and server is realized by wireless network. Wireless video compression is realized based on OpenCV visual library. OpenCV visual library is used in computer to receive and play video. A human-computer interface which is easy to operate is developed by MFC. With the above technology, video wireless monitoring and suspicious target tracking are realized. The test results show that the system works reliably and stably, and has high practical and research value.
【学位授予单位】：郑州大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN948.6

【相似文献】