摘要
对拥挤地区人群行为的检测对于多种安全任务都至关重要。这个问题包括2个部分:对拥挤程度的估计和对异常的检测。本文定义了一个运动特征来进行群体建模和实时检测。在此基础上,通过分析大量的运动信息。提出了一种通过自定义的能量函数来评估观测区域内的拥挤程度和异常事件检测的新方法。此方法被用于地铁的视频监控系统中,具有良好的效果。
Surveillance on pedestrian flows in crowded areas is of significance for various security tasks. This problem involves two parts： evaluation of crowdedness and detection of abnormality. This paper defines a motion feature to deal with crowd modeling and processing in the real-time surveillance. Based on it, a new approach is presented to evaluate the level of crowdedness and monitor abnormal events of the scene according to an analysis on the amount of motion information. The result of a metro subway video surveillance system demonstrated the effectiveness of this method.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2007年第4期614-620,共7页
Chinese Journal of Scientific Instrument
基金
Supported by the grant from the Research Grants Council of the Hong Kong Special Administrative Region(CUHK4163 /03E)
Hong Kong Innovation and Technology Fund under grant ITS/140 /01
关键词
群体检测
运动特征
群体能量
群体异常
crowd surveillance
motion feature
crowd energy
crowd abnormality
