基于Kalman滤波和改进的MeanShift算法的目标跟踪

MeanShift算法因为简单性和稳定性在目标跟踪中得到广泛应用，但是当目标和背景的颜色模型比较接近时，传统的MeanShift算法由于缺少空间信息，且经典的相似性度量函数不易区别，导致跟踪失败。为了克服上述缺点，采用基于空间颜色特征和新的相似性度量的MeanShift算法，并提出一种融合Kalman滤波器和改进的MeanShift算法的目标跟踪方法。首先，利用改进的MeanShift算法计算出当前帧中目标的准确位置，然后使用Kalman滤波器去预测下一个初始搜索位置，用于下一帧中MeanShift迭代，最后实现对目标的跟踪。实验结果表明，该算法可以准确地跟踪目标，并且跟踪的准确率优于传统的MeanShift算法或者Kalman和传统Meanshift的融合算法。

Target Tracking Based on the Improved Mean Shift Algorithm and Kalman Filter

The Mean Shift algorithm is widely used because of its simplicity and stability in the target tracking, but when the color model of the target and the background is similar, the traditional Mean Shift algorithm lead to tracking failure because of its lack of spatial information, and the classic similarity measure function is not easy to find. To overcome these shortcomings,the Mean Shift algorithm based on the space color features and new similarity measure is used. The author proposes a fusion Kalman filter and improved Mean Shift algorithm target tracking. The improved Mean Shift algorithm is used to calculate the exact location of the target in the current frame,and the Kalman filter is used to predict an initial search position for the next frame Mean Shift iteration and to achieve the target tracking. The experimental results show that the proposed algorithm can accurately track the target, and the tracking accuracy is better than traditional Mean Shift algorithm or Kalman and traditional Mean Shift fusion algorithm.