基于运动特征的人体异常行为识别
首发时间:2014-08-22
摘要:为了提高监控视频中人体异常行为识别的实时性和准确率,提出了基于运动特征的人体异常行为识别方法。利用分块更新的背景差法从图像中提取出完整的人体轮廓,通过区域关联结合颜色直方图实现人体目标跟踪,解决了非线性运动时漏跟和误跟的问题。通过人体运动轨迹、运动姿态及运动时间3种参数,对人的5种异常行为进行分析判断。实验结果表明,所提算法不仅能实时地对人体进行检测和跟踪,还能快速、准确地识别出异常行为,具有简单实用的特点。
For information in English, please click here
Human abnormal behavior recognition based on motion characteristics
Abstract:Acharacteristics-based human abnormal behavior detection method is proposed to improve accuracy and real-time performance of human abnormal behavior detection in surveillance video. Chunked update-based background subtraction technique isused to extract complete human silhouette from the image.The regional association is combinedwith color histogram to track human target to solve the miss tracking and target-missing problems in nonlinear movement. Three parameters of human movement trajectory, moving posture and time of movement areselected to identify five human abnormal behaviors. In the experiments,the human body is detected and tracked in real time, and the abnormal behaviors could be identified quickly and accurately using proposed algorithm.
Keywords: chunked update motion characteristics target tracking abnormal behavior identification
论文图表:
引用
No.4606670222691408****
同行评议
勘误表
基于运动特征的人体异常行为识别
评论
全部评论0/1000