瞿中, 张宁. 运动目标轨迹网格化分析与徘徊行为检测研究[J]. 微电子学与计算机, 2014, 31(4): 60-63,67.
引用本文: 瞿中, 张宁. 运动目标轨迹网格化分析与徘徊行为检测研究[J]. 微电子学与计算机, 2014, 31(4): 60-63,67.
JU Zhong, ZHANG Ning. Analysis of Moving Object Trajectory Gridded and Hovering Behavior Detection Research[J]. Microelectronics & Computer, 2014, 31(4): 60-63,67.
Citation: JU Zhong, ZHANG Ning. Analysis of Moving Object Trajectory Gridded and Hovering Behavior Detection Research[J]. Microelectronics & Computer, 2014, 31(4): 60-63,67.

运动目标轨迹网格化分析与徘徊行为检测研究

Analysis of Moving Object Trajectory Gridded and Hovering Behavior Detection Research

  • 摘要: 针对传统的目标徘徊检测方法不能准确判断目标行为的局限性,通过对运动目标轨迹进行网格化分析,提出一种新的目标徘徊检测方法.首先采用背景差分法检测出运动目标,利用Camshift算法进行目标跟踪.然后对防区内运动目标的轨迹做网格化分析,统计轨迹在网格中的分布情况,判断目标是否有徘徊行为并进行跟踪报警.最后实验结果表明,该方法能够对几种典型的徘徊行为进行实时、精确判断,同时可以检测出其他复杂的徘徊行为,满足了系统的准确性要求.

     

    Abstract: Traditional target hovering detection methods have many limits and shortages which cannot accurately determine the target behavior.In this paper,object trajectories are gridded,a new target detection method for moving object trajectories was proposed.Firstly,the moving targets were detected with the background difference method.Secondly,the moving target in image sequence was tracked according to the Camshift algorithm.Then the moving target trajectory in the defense area was gridded and some statistics were analyzed.At last,the hovering behavior was detected and the alarm was triggered.The experiment results show that this method can judge several typical hovering behavior accurately and in real-time.

     

/

返回文章
返回