面向城市应急事件管理的监控视频智能解析
发布时间：2019-01-20 09:17
【摘要】：随着城市现代化进程的不断加速,建立高效、完善的城市应急事件管理系统迫在眉睫。遍布城市各处的监控摄像头拍摄的监控视频是城市应急事件管理系统重要的数据来源,但在现阶段主要依赖人工识别,费时费力且效果不佳。提出一种面向城市应急管理的监控视频智能解析方法,利用计算机视觉库获取视频中监控对象的数量和位置,生成其运动轨迹并匹配预设的异常运动模板,从而构建应急事件的规范化情境模型用以分析和预警。以某大学校园开放区域和某高新科技园区的监控视频为例,验证了所提出方法的可行性和合理性。
[Abstract]:With the acceleration of urban modernization, it is urgent to establish an efficient and perfect urban emergency management system. Surveillance video captured by surveillance cameras all over the city is an important data source of urban emergency management system, but at this stage, it mainly relies on manual identification, which is time-consuming and laborious, and the effect is not good. This paper presents an intelligent analysis method of surveillance video for urban emergency management. The computer vision library is used to obtain the number and position of the monitored objects in the video, to generate its motion trajectory and to match the default abnormal motion template. In order to build a standardized emergency situation model for analysis and early warning. The feasibility and rationality of the proposed method are verified by taking the surveillance video of a university campus open area and a hi-tech park as an example.
【作者单位】： 上海交通大学机械与动力工程学院;上海仪电(集团)有限公司中央研究院;
【基金】：国家自然科学基金(No.70971085,No.71271133) 上海市经信委产学研项目(No.沪CXY-2015-006)
【分类号】：TP391.41
[Abstract]:With the acceleration of urban modernization, it is urgent to establish an efficient and perfect urban emergency management system. Surveillance video captured by surveillance cameras all over the city is an important data source of urban emergency management system, but at this stage, it mainly relies on manual identification, which is time-consuming and laborious, and the effect is not good. This paper presents an intelligent analysis method of surveillance video for urban emergency management. The computer vision library is used to obtain the number and position of the monitored objects in the video, to generate its motion trajectory and to match the default abnormal motion template. In order to build a standardized emergency situation model for analysis and early warning. The feasibility and rationality of the proposed method are verified by taking the surveillance video of a university campus open area and a hi-tech park as an example.
【作者单位】： 上海交通大学机械与动力工程学院;上海仪电(集团)有限公司中央研究院;
【基金】：国家自然科学基金(No.70971085,No.71271133) 上海市经信委产学研项目(No.沪CXY-2015-006)
【分类号】：TP391.41
【相似文献】
相关期刊论文 前10条
1 陈姣姣;张晓如;周永梅;;基于本体的监控视频语义事件探测[J];计算机应用研究;2012年01期
2 汪贤锋;;基于内容的二次聚类监控视频摘要生成分析[J];内蒙古财经大学学报;2013年05期
3 朱丽英;梁辰;张鸿洲;;基于运动目标的监控视频检索方法[J];计算机应用与软件;2011年12期
4 周渝斌;;海量监控视频快速回放与检索技术[J];计算机应用;2012年11期
5 孙鹏;周纯冰;杨洪臣;沈U，
本文编号：2411899
本文链接：https://www.wllwen.com/guanlilunwen/gonggongguanlilunwen/2411899.html
