基于局部敏感直方图的稀疏表达跟踪算法

为解决目标跟踪过程中光照变化、姿态变化等问题,提出了一种基于局部敏感直方图特征的稀疏表达跟踪方法。对粒子滤波获取的多个候选目标提取局部敏感直方图特征,并根据模板字典,采用改进的L1范数模型求取每个候选目标的稀疏表示系数;然后计算每个候选目标的权重,选取权重最大的候选目标作为跟踪结果。实验结果表明,本算法能很好实现对目标的跟踪,在解决光照变化、姿态变化等问题方面有较好的效果。

Sparse representation tracking method based on locality sensitive histogram

In order to solve the problems of illumination and pose change during target tracking, a sparse representation tracking method based on local sensitive histogram was proposed. Local sensitive histogram features of multiple candidate targets were extracted, and sparse representation coefficient of each candidate target was calculated based on template dictionary by using modified L1 norm model. Then, the weight of each candidate target was calculated. The candidate target which had the largest weight was selected as tracking result. Experimental results demonstrated that the method can track the target accurately and effectively and has advantage in illumination and pose change.