首页 | 本学科首页   官方微博 | 高级检索  
     

基于轨迹特征分析的行人异常行为识别
引用本文:周维柏,李蓉.基于轨迹特征分析的行人异常行为识别[J].电脑编程技巧与维护,2010(12):107-109.
作者姓名:周维柏  李蓉
作者单位:华南师范大学增城学院,广州,511363
摘    要:提出一个以行人为目标的视频监控系统,利用行人的移动轨迹特征来判断是否发生异常行为。利用背景相减法来检测是否有目标的存在,并利用目标的各种信息来追踪行人;建立目标行为模型,利用模型有效地识别目标的各种行为;利用轨迹对比的方式完成相关事件的轨迹检索,并供后续查询。实验表明,本算法可行、性能稳定。

关 键 词:目标检测  轨迹特征  异常行为  主成份分析

Pedestrian Abnormal Behavior Recognize Based on Trajectory Features
ZHOU Weibai,LI Rong.Pedestrian Abnormal Behavior Recognize Based on Trajectory Features[J].Computer Programming Skills & Maintenance,2010(12):107-109.
Authors:ZHOU Weibai  LI Rong
Affiliation:(Zengcheng college of South China Normal University,Guangzhou 511363)
Abstract:A video surveillance system for pedestrian abnormal behavior is presented.Firstly,background subtraction technique is employed to detect moving objects from video sequences.Then,all features are extracted to track each detected object.After that,the model is introduced to model pedestrians'behaviors.According to the parameters of the models,different behaviors like walking,running and falling can be successfully recognized and analyzed.Finally,two curve-matching algorithms are employed to complete the trajectory retrieval.Experiment results show that the algorithm can solve the test paper composition problem quickly and effectively,and also has more capability and utility.
Keywords:Target Detection  Track Features  Abnormal Behavior  Principle Component Analysis
本文献已被 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号