视频和GIS协同的人群状态感知模型  

Crowd Status Analysis Based on Surveillance Videos and GIS

作　　者：邓仕虎[1,2] 张兴国 王小勇 朱俊丰 王秀 

机构地区：[1]重庆市地理信息中心,重庆401121 [2]重庆知行地理信息咨询服务有限公司,重庆401121 [3]信阳师范学院地理科学学院,河南信阳464000

出　　处：《信阳师范学院学报（自然科学版）》2018年第1期59-63,共5页Journal of Xinyang Normal University(Natural Science Edition)

基　　金：国家自然科学基金项目(41401436);对地观测技术国家测绘地理信息局重点实验室开放课题(K201510);信阳师范学院博士科研启动基金项目(0201402)

摘　　要：针对监控视频在图像空间难以统一、宏观及真实量化人群状态的问题,提出了一种基于地理信息技术进行人群状态地图展示、量化分析及预警的方法.首先,通过摄像机厂商或标准视频流接口接入监控视频;然后,将视频图像调整为较小尺寸,采用光流法进行计算,获取图像空间下的光流场;最后,将各个监控视频相应的光流场映射至地理空间,即可在地图中观察人群运动状态,并可通过散点内插、等值线等进行分析,通过相关阈值的设置实现人群异常检测与预警.本文基于MATLAB、ArcGIS Engine、C#等技术,采集多个实验视频,并研发了基于GIS的人群状态感知原型系统,对相关算法进行了验证.结果表明,该方法相对于传统图像空间下的人群状态检测方法,具有人群运动可定位、可量测、可宏观观察和预警等优势.In order to solve the problcmi that surveillance videos were clifficuispace and the crowd statuswas dificuh to be perceived macroscopically and quantitatively,a new method was proposed based on GIS,which can be used to display,analyze and early warn crowd status through 2D miap. Firstly, the surveillance video was accessed through the API of camicra mianufacturcrs or the standard video stream interface. Secondly, the video images were adjusted to a smiallcr size and the optical flow field in imagespacewas calculated through some optical calculation algorithm. Finally,the optical flow field of each surveil-lance video was miappcd to the corresponding geographic space. T＇hcn through GIS, thcrowd can be observed in the map. T＇wo analysis methods,including scatter point interpolation and contour map,were put forward. The crowd alarmcanbe realized through relevant thresholds setting. BaLAB,ArcGISE：ngine and MicrosotVisual Studio C #,a prototype system was developed to validate the algo- rithmi using several crowd videos. T＇he results showed that the new micthod had the advantages of geographic lo-cation ,micasurcmicnt, miacroscopic observation and early warning,etc. when comiparcd with the previous micthods.

关 键 词：GIS 视频 人群 光流 异常行为 

分 类 号：P208[天文地球—地图制图学与地理信息工程]