基于互相关算法的人员密集场所人群运动速度特征研究  

Empirical study of the characteristics of the pedestrians' velocity in crowded places based on cross-correlation algorithm

作　　者：王嘉悦[1] 翁文国[1] 张小乐[1] 

机构地区：[1]清华大学工程物理系安全科学与技术研究所,北京100084

出　　处：《中国安全生产科学技术》2014年第6期5-11,共7页Journal of Safety Science and Technology

基　　金：国家自然科学基金项目(91024018);国家"973"重大基础研究项目(2012CB719705)

摘　　要：研究人群密集场所的人群运动速度特征可以预测人群的运动趋势,在大型活动组织过程中可以对异常人群运动做出预警,避免过度的拥挤及踩踏事件的发生,保证大型群体性活动的安全顺利开展。利用国内某重要城市核心区公共场所人群运动的视频图像,通过互相关算法提取该场所人群的运动速度,并进一步比较通往景区的四条不同路径上人群运动速度的差异性,分析其人群运动特征。分析结果表明单向通道的人群运动速度较大且运动方向基本与通道的两侧边界平行,而双向通道中由于人群中阻尼效应的影响,人群运动速度的大小和方向都发生了不同程度的改变。分析结果可为核心区管理者进行大型群体性活动的组织、人群疏散与引导提供建议,进而为人群拥挤踩踏事故风险防控、拥挤踩踏事故专项应急预案制定及人群聚集活动安全方案编制提供理论支持。Study on the velocity characteristics of pedestrian behaviors can predict pedestrian movement trend during real mass events in crowded places. It is very important to warn the abnormal pedestrian behaviors and prevent the occurrence of excessive crowd or even stampede,thus ensure the large-scale mass activities successfully and safely. In this paper,empirical study was conducted on the large scale and high density of people＇behaviors in the core area of a domestic major city,by a cross-correlation algorithm. The velocity of pedestrian movement was extracted at different speeds. In addition,the average velocities of pedestrians were compared with four passages to the scenic spot,to investigate the motion characteristics of pedestrians. The results showed that the velocities on unidirectional path were large and the movement directions were nearly parallel to the boundaries. However,the velocities on two-way passage were changed due to the damping effect of the crowd. Some effective suggestions are thus proposed for the organization and the crowd evacuation of large-scale mass activities. The results can also provide some theoretical information to risk prevention and control for stampede,make the emergency plan for stampede and the safety scheme for crowd activities.

关 键 词：人群运动特征 人员密集场所 人群运动速度 

分 类 号：X936[环境科学与工程—安全科学]