针对实时视频中的运动物体跟踪问题,提出了一种基于自适应Kalman滤波的运动物体跟踪新算法.首先利用基于∑-△背景估计算法检测运动物体,并提取主要颜色特征.然后构建物体运动模型,并生成自适应Kalman滤波的系统状态模型.最后利用主要颜色特征进行物体跟踪,其结果反馈给自适应Kalman滤波器,并通过遮挡率自动调整参数达到正确跟踪.实验结果表明,所提出的自适应Kalman滤波算法在运动物体被遮挡等复杂条件下的鲁棒性好,还具有跟踪准确性高和数据计算量小等优点,可用于实时运动物体的检测与跟踪.%For real-time video moving objects tracking, a new tracking method using adaptive Kalman filter is proposed. Firstly, the moving objects are detected using S-A background estimation algorithm and their dominant color is extracted. Then, a motion model is constructed to set the system model of adaptive Kalman filter. At last, the dominant color is used to track moving objects. The tracking result is fed back to adaptive Kalman filter. The parameters of adaptive Kalman filter are adjusted by occlusion ratio adaptively. The experimental results show that the proposed algorithm has the robust ability on some complex situations such as occlusion and the advantages of high accuracy and low calculation, and it can be used for real-time moving object detection and tracking.
展开▼
