基于混合高斯模型和SIR粒子滤波的运动目标检测与跟踪研究

Research on Moving Object Detection andTracking based on Mixture Gauss Model and SIR Particle Filter

针对视频监控系统中运动目标的检测与跟踪技术进行深入研究，采用高斯背景模型的背景差法实现对运动目标的检测，基于SIR粒子滤波算法，利用颜色直方图信息与目标的运动状态信息实现对运动目标的跟踪。采用DaVinci技术对目标检测和目标跟踪算法进行封装和优化，实测结果表明，该算法能够对监控场景中的运动目标进行检测和跟踪，实现对视频图像的实时处理。

A thorough research is conducted to the detection and tracking technique oI moving targets m vlaeo mom- toring system. The background difference method of Gaussian background model is adopted to realize the moving tar- get detection. Based on SIR particle filter algorithm, the moving target tracking can be realized by using color infor- mation and target motion state information. The DaVinci technology is used to encapsulate and optimize the target detection and target tracking algorithm. The experimental results show that the algorithm can detect and track the moving target in the monitoring scene, and realize the real - time processing of the video image.