188宝金博页面版

 Edge-Cloud Collaborative Video Analytics System for CrowdGathering Detection in Metro StationsLi Sun, Jing Sun*, Jun Zhang, Xianbin Peng, Fan Zhang, Desheng Zhang,Kejiang Ye, and Jianping FanAbstract: The safe operation of metro systems, particularly in densely populated cities, relies on the effectivemanagement of crowd gatherings within stations. Existing research primarily focuses on camera-based crowdcounting but fails to consider crowd movement across multiple interconnected spaces, limiting its effectivenessin complex metro environments. This paper proposes a real-time Edge-Cloud collaborative video analyticssystem for Crowd Gathering Event Detection (EC-CGED), integrating an edge-cloud collaboration mechanism,spatial knowledge model, and an event-driven adaptive dynamic scheduling strategy. The system enables fine-grained, real-time monitoring of crowd dynamics and provides early warnings of potential crowd gatherings. Atthe edge nodes, real-time video streams are decoded and analyzed to extract crowd counting indicators forvarious station areas and transit channels. These indicators, along with spatial knowledge of the metro stationlayout, are aggregated at the central cloud server to enhance the accuracy of crowd gathering event detection.Additionally, we introduce an event-driven adaptive dynamic task scheduling strategy to optimize computationalresource allocation, improving system efficiency. By enabling timely detection and proactive alerts for crowdgatherings, the EC-CGED system enhances metro station safety and operational efficiency, addressing criticalchallenges in urban public transportation management.Key words:  edge-cloud collaborative; video analytics; spatial knowledge; Event-Driven Adaptive Dynamic (EDAD)scheduling1　IntroductionThe metro system is a critical component of urbanpublic transportation, accommodating more than halfof daily commutes in certain metropolitan areas. Forinstance, the Shenzhen Metro system recorded a peakdaily passenger flow of over 10.17 million,underscoring the immense pressure on metroinfrastructure. Uncontrolled crowd gatherings in metrostations can lead to severe operational disruptions andpassenger safety risks. Therefore, the timely detectionand early warning of crowd gathering events are    Li Sun, Jing Sun, and Kejiang Ye are with the Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen518055, China. E-mail: li.sun@siat.ac.cn; jing.sun1@siat.ac.cn; kj.ye@siat.ac.cn.   Jun Zhang, Xianbin Peng, and Fan Zhang are with the Shenzhen Institute of Beidou Applied Technology, Shenzhen 518038, China.E-mail: zhangjun@szbit.cn; pengxianbin@szbit.cn; zhangfan@szbit.cn.   Desheng Zhang is with Department of Computer Science, Rutgers University, Piscataway, NJ 08854, USA. E-mail:desheng@cs.rutgers.edu.   Jianping Fan is with the Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China, andalso with University of Chinese Academy of Sciences, Beijing 101408, China. E-mail: jp.fan@siat.ac.cn.* To whom correspondence should be addressed.    Manuscript received: 2024-12-05; revised: 2025-03-30; accepted: 2025-04-22 TSINGHUA  SCIENCE  AND  TECHNOLOGYResearch ArticleISSN  1007-0214    28/38   pp1764−1777DOI:  10.26599/TST.2025.9010082Volume 31, Number 3, June  2026 ©   The author(s) 2026. The articles published in this open access journal are distributed under the terms of theCreative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/).