摘要
针对目前广泛使用的光流法计算耗时严重问题,提出了基于差分图像绝对值和(SAD)与光流法相结合的人体运动检测方法.通过计算SAD检测出运动区域,在已确定的运动区域内进行Horn-Schunck光流场计算,准确地计算出人体的运动信息.在后续处理中,应用形态学的闭运算和连通性分析,较完整地分割出人体运动目标.实验结果表明,该方法有效地提高了系统的计算速度,能够实时准确地对人体运动进行检测.
Considering the problem that the traditional optical flow algorithm is a time-consuming computation. An integrated method of detecting human motion is proposed based on the sum of absolute differences （SAD） and optical flow. The motion sub-region is detected by the computation of SAD. The optical flow vectors of the object can be obtained by computing optical flow on the motion sub-region. In the post processing, closing and connected components analysis are used for the segmentation of human object. Experimental results showed the effectiveness and robustness of the proposed method.
出处
《北京理工大学学报》
EI
CAS
CSCD
北大核心
2008年第9期794-797,共4页
Transactions of Beijing Institute of Technology
基金
国家部委预研项目(06104040)
关键词
人体运动
差分图像绝对值(SAD)
光流法
目标分割
human motion
sum of absolute differences （SAD）
optical flow method
object segmentation
