为预防城市轨道交通车站拥挤踩踏事故,保障乘客人身和财产安全,有必要实行拥挤踩踏预警,但目前相关研究较为匮乏。在综合分析地铁客流预警、拥挤踩踏事故风险评估和客流状态检测(主要是视频行人检测、跟踪和计数)研究现状的基础上,对城市轨道交通车站拥挤踩踏预警体系进行了探讨。建立了基于视频检测的城市轨道交通车站拥挤踩踏预警框架。明确了预警的技术流程和主要方法,包括基于视频检测的客流信息提取、客流信息分析与短时预测、拥挤踩踏风险评估、拥挤踩踏预警决策与措施制订等。最后,提出了城市轨道交通车站拥挤踩踏预警面临的挑战和研究建议。
In order to prevent crowd trample accidents and protect passengers' personal and property safety in urban rail transit stations,it is necessary to implement crowd trample warning. However,we are short of relevant research recently. Based on comprehensively analyzing the research status about early warning of subway passenger flow,crowd trample accident risk assessment and passenger traffic monitoring( mainly video pedestrian detection,tracking and counting),we discussed the crowd trample warning system for urban rail transit stations. We established a crowd trample warning framework for urban rail transit stations based on video detection,and also determined the technical process and major method of the early warning system,including extraction, analysis and short-term prediction of passenger flow information based on video detection,assessment of crowd trample accident risk,decision-making on early warning of crowd trample and measure development. The challenges and research recommendations of crowd trample warning for urban rail transit station are presented at last.