CASIA OpenIR  > 毕业生  > 硕士学位论文
Alternative TitleThe Design and Implementation of the Multi-camera Video Objects Trajectory Mining System
Thesis Advisor李子青
Degree Grantor中国科学院大学
Place of Conferral中国科学院自动化研究所
Degree Discipline计算机技术
Keyword视频监控 多摄像机 目标再辨识 轨迹挖掘 Video Surveillance Multi-camera Target Re-identification Trajectory Mining
Abstract随着硬件成本的降低和网络的普及,视频监控系统往往会采用多个摄像机来监控环境,一个多摄像机构成的监控网络能够有更大的覆盖面积,也为后续的大范围视频内容分析和理解打下基础。为了实现这一过程的智能化,需要利用图像处理和计算机视觉的方法来实现多摄像机的协作处理和分析,其中的一个重要研究课题就是在多个视频场景中实现对目标的轨迹描述和挖掘,这对于协助公安人员办案分析具有重大意义。 本课题实现一个多摄像机视频目标轨迹挖掘系统。在已有对海量监控数据进行目标提取功能的基础上,在多个摄像机视频中检索挖掘疑似目标,然后根据摄像机的空间信息及目标的时间信息等,挖掘目标可能的运动轨迹,然后将该运动轨迹呈现在地图中。 在本文中,作者首先对相关技术做了简要介绍,并详细介绍了目标特征提取和特征匹配等轨迹挖掘算法的基础,然后具体阐明了轨迹挖掘算法的设计思路。在系统开发方面,首先对系统做了全面的功能分析,以此为基础进行了架构设计、模块设计、界面设计和数据存储设计,最后,作者对本系统的实现从多个层次及角度进行了详细说明,并对测试情况进行了详细报告,结果表明本系统符合课题对功能和性能的要求,能够解决公安民警目前迫切的业务需求,提高办案效率。
Other AbstractWith the reductions of hardware cost and the popularity of networks, video surveillance systems usually use multiple video cameras to monitor the environment. Monitoring network consisting of more than one camera can cover a greater area, which is good for the subsequent large-scale video content analysis. In order to achieve this process intelligence, image processing and computer vision methods are needed to achieve multi-camera collaborative processing and analysis. One of the most important research topics is to describe and mine the track of target from different videos, which has great significance in helping police officers handling the case analysis. This thesis implements a multi-camera video object trajectory mining system. On the basis of the extraction of targets from multiple camera video, the system search suspected targets. According to the time information and spatial information of targets, the system will mine the movement trajectory of the targets and present it in the map. In this thesis, the author first gave a brief introduction of related technologies. Then he told the details of the target feature extraction and matching algorithm, which was the basis of track mining algorithm. After that, the author clarified the design of trajectory mining algorithm design. As for the system development, the author first made a comprehensive functional analysis. Then he made the architecture design, module design, interface design and data storage design. And finally, the author described the realization of the system from multiple levels and angles in detail. He also carried out a detailed testing report, which showed that the system meets the functional and performance requirement and it can help improving the work efficiency of police.
Other Identifier2011E8009061080
Document Type学位论文
Recommended Citation
GB/T 7714
张鑫龙. 多路视频目标轨迹挖掘系统的设计与实现[D]. 中国科学院自动化研究所. 中国科学院大学,2014.
Files in This Item:
File Name/Size DocType Version Access License
CASIA_2011E800906108(2556KB) 暂不开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[张鑫龙]'s Articles
Baidu academic
Similar articles in Baidu academic
[张鑫龙]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[张鑫龙]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.