嵌入式监控系统的运动目标检测与跟踪

Moving Object Detection and Tracking by Embedded Monitoring and Control System

采用高斯背景模型的背景差法对运动目标进行检测,并基于采样粒子滤波算法(SIR),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。最后,采用Da Vinci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。

The background difference method of Gaussian background model is adopted to realize the moving target detection. Based on color histogram information and target movement information, the target tracking is carded out with sampling importance resampling （SIR）. The DaVinci technology is used in the encapsulation and optimization of the algorithm in target detection and tracking. The results show that the algorithm can monitor and track the moving objects in the controlled scene to realize the real-time processing of video images.