Mean Shift算法在目标跟踪领域的应用研究

Application of Mean Shift Algorithm in Target Tracking Field

目标跟踪物在运动过程不断移动以及在运动过程中被遮挡的问题一直是研究的一个焦点和重点问题。本文通过阐述Mean Shift算法原理,研究了Mean Shift算法对目标物的快速跟踪过程。并通过两组实验分别对运动目标物及在遮挡条件下运动目标物进行跟踪。第一组实验中,选取移动的物体篮球为目标物,并通过Mean Shift算法,不断迭代跟踪新的篮球位置,实现对篮球的跟踪。第二个实验中,选取移动的人为目标物,目标物在移动过程中被其他人遮挡,通过将颜色灰度特征应用到Mean Shift算法中,减少了其他物体对目标物跟踪的干扰,从而实现不断跟踪新目标。实验结果证明,Mean Shift算法能够快速、有效地实现目标跟踪。

It is still a challenge to study a real-time and accurate algorithm to solve the problem of target tracker moving constantly and being blocked in the process of movement. Mean Shift algorithm is simple in calculation and can quickly track the target. The Mean Shift algorithm principle is expounded, and two groups of experiments are carried out to track moving targets and moving targets under occlusion condition respectively. The experimental results prove that Mean Shift algorithm can realize fast and effective tracking.