对变形目标的跟踪研究

Study on Tracking of Deformational Targets

在目标跟踪过程中,目标在图像中的形状和大小常发生不同程度的变化,为了准确、有效地实现目标跟踪,采用多次迭代的连续自适应平均值移动算法。该算法是一种基于颜色跟踪的算法,根据多次迭代的计算结果,自适应调整图像序列中搜索窗口的大小和位置,从而得到当前图像中的目标中心,实现对运动目标的跟踪。最后分别在不同变形情况下和多个运动目标下进行实验。实验结果表明,当目标发生旋转、大小变化及存在多个目标的情况下,该算法仍能有效地对运动目标进行准确跟踪。

In the process of the target tracking,the shape and size of a target in the images often change to some degree.In order to accurately and effectively achieve the target tracking,the multiple iterative continuous adaptive mean shift（camshift） algorithm is used.The algorithm is a color-based tracking algorithm,which can make adaptive adjustments for the size and location of the search window in the image sequence to locate the center of the object in the current image according to multiple iterative calculation results,and to track the object.A experiment was performed in the case of different target deformation and multiple moving targets.The experimental results show that the algorithm can still effectively track the moving targets accurately even if the target rotates,changes its size and exists with multiple targets.