基于状态转换图的公共场所行人视频监控方法研究

Pedestrian Video Surveillance Approach Based on the State-transition Graph

自动视频监控对于公共场所的安全控制等方面具有重要意义。本文实现了一个实用的行人监控系统,采用了基于改进Kalman滤波的背景更新方法进行背景的有效更新,同时开创性地提出了基于状态转换图的行人计数系统,用于处理场景中前景遮挡、粘连等复杂情况,弥补了复杂条件下前景运动物体检测存在的准确度不高的问题,实验结果证明了此方法的有效性。

The automatic visual surveillance plays an important role in the security control of public places.In this paper,a practical pedestrian surveillance system is proposed,the system is based on the improved Kalman filter background update approach.We also implement a new state-transition graph system to solve some complicate situations,like foreground occlusion and conjunction.Experimental results show the effciency of our approaches.