摘要
视频图像分析联动预警是对地铁施.工现场既有的实时监控视频进行分析,识别施工人员的不安全行为,通过现场的智能广播及手机设备,对相应的施工人员和现场管理人员发出即时联动预警。视频图像分析联动预警结合施工现场既有的实名制门禁设备和视频监控设备、必要的智能广播设备,以及基于GIS(地理信息系统)的空间分析技术,改变了传统的视频监控模式。即把现有视频监控实时数据接入视频图像分析软件,对施工现场不戴安全帽、进入危险区域等不安全行为进行实时分析识别,通过摄像头的坐标位置和GIS技术找到就近的智能广播,并对违规人员进行语音预警;同时,通过实名制门禁设备分析当前在场的管理人员,并向其手机和APP推送预警信息。通过视频图像分析联动预警,实现了无人值守、及时发现、及时预警、就近处置,有效提高了地铁施工安全管控的效率。
The video image analysis linkage warning is to analyze the existing real-time monitoring video of the subway construction site, identify the unsafe behavior of the constructor, and issue real-time linkage warning to the constructor and site management personnel through the on-site intelligent broadcast and mobile phone. The video image analysis linkage warning combined with the existing real-name access control equipment and video surveillance equipment on the construction site, the necessary intelligent broadcasting equipment, and the spatial analysis technology based on GIS(Geographic Information System) have changed the traditional video surveillance mode. That is, connect the existing video monitoring real-time data to the video image analysis software, identify the unsafe behaviors such as not wearing a helmet or entering a dangerous area on the construction site, and give voice warning to the violators by the nearby intelligent broadcast, which found according to the coordinate position of the camera based on the GIS technology;at the same time, find the site management personnel through the real name access control equipment and push alert information to their mobile phone and app. Through video image analysis and linkage early warning, unmanned duty, timely detection, timely warning, and nearby disposal are realized, which effectively improves the efficiency of subway construction safety management and control.
作者
孙有恒
赵明明
王传宝
SUN Youheng;ZHAO Mingming;WANG Chuanbao(Guangzhou Metro Group,Guangzhou 510335,China;China Railway Construction South China Construction Co.,Ltd,Guangzhou 511458,China;Bejing Jiubim Information Technology Co.,Ltd,Beijing 100000 China)
