结合特征点匹配的在线目标跟踪算法

提出一种结合特征点匹配的目标跟踪算法.首先,通过显著区域跟踪方法,解决算法对初始化目标框大小敏感的问题,提高样本选取质量,并降低背景杂波对跟踪器的影响.其次,采用中值流法跟踪和特征点匹配相结合的方法估计目标的尺度变化,并通过层级聚类方法剔除干扰点,解决跟踪器漂移及目标平面旋转跟踪失败等问题.最后,提出一种简单的检测器自适应尺度快速搜索目标方法加快检测速度.结果表明:所提方法有效地提高了TLD目标跟踪算法的跟踪鲁棒性,并在标准数据集上得到了很好的效果.

Online Target Tracking Algorithm Based on Feature Point Matching

A target tracking algorithm based on feature points matching has been proposed in this paper. Firstly, the method of tracking target in the saliency region can solve target size sensitivity while improving the quality of choosing samples and reducing the effect on the tracker caused by background clutter. Secondly, the target’s scale can be estimated through median flow tracker and feature points matching and the interference point can be rejected by hierarchical clustering, based on these methods, tracker drifting and out-of-plane tracking failing can be resolved effectively. Finally, a simple fast search objectives method based on adaptive scale detector was proposed to accelerate detection speed. Experimental results demonstrate that the proposed algorithm can enhance the tracking robustness of TLD target tracking method effectively and obtain good results on standard data sets.