基于Kalman预测和Mean-shift算法的视频目标跟踪

Video object tracking based on Mean-shift with self-adaptability search window and Kalman prediction

提出一种用Kalman滤波理论和Mean-shift算法结合的视频目标跟踪方法,解决了目标变形、部分遮挡和运动速度过快问题。在初始帧中,确定跟踪目标并计算H分量直方图,将每帧图像转化为该直方图的概率投影图;在当前帧中,用Kalman滤波预测搜索窗口,并在搜索窗口中用自适应的Mean-shift算法精确匹配跟踪目标。实验结果表明,本文方法对刚体、非刚体和多目标的跟踪都具有良好的自适应性。

To solve the tracking problems of transformation,partial occlusion and over-fast motion identified with objects in video sequence,an alternative approach is proposed using mean-shift algorithm and Kalman filtering techniques for efficient tracking of the targets.Firstly,in the initial frame the target is determined and followed by computing its H component in the histogram,then the target within the window of the current frame is transformed to the image of probability.The Kalman filter is adopted to predic...