摘要
随着公共安全问题的日益突出,公共场所人群异常事件的及时发现将有助于相关部门的及时响应和救援,从而降低群众人身伤亡和财产的损失。近年来,在智能监控和安防领域的发展下,基于视频的人群异常事件检测已成为图像处理、机器视觉、机器学习等相关领域的研究热点。概述了基于视频的人群异常事件检测相关研究的概况、研究现状及未来的发展趋势。人群异常事件检测有两个基本问题,一个是基本事件的表示,一个是异常事件检测模型的建立。重点从这两个方面回顾人群异常事件检测技术的发展和常用的处理方法,并对研究难点及未来的发展趋势作了较为详细的分析。
As the increasingly prominent of public security issues,timely detection of anomaly in public crowds facilitates the in-time response and rescue from related departments,reducing personal casualty and property loss.In recent years,with the development of intelligent surveillance and security monitoring,video anomaly detection in crowds has becoming a hot research topic in related computer vision fields such as image processing,machine vision and machine learning.This paper outlines the overview,research status,and trends of the study on anomaly detection in crowds.There are two elemental problems in the study of anomaly detection:one is the presentation of primitive events,and the other is the construction of anomaly detection model.From the two aspects,this paper mainly reviews the development of anomaly detection in crowds,and the processing methodology commonly used.additionally,difficulties in research and study trend in the future are analyzed in details.
出处
《电子测量与仪器学报》
CSCD
2014年第6期575-584,共10页
Journal of Electronic Measurement and Instrumentation
基金
国家自然科学基金(61005012)资助项目
关键词
异常事件检测综述
人群异常事件
基本事件表示
异常事件检测模型
overview of abnormal events detection
Anomaly in crowds
primitive events presentation
anomaly detection model