基于局部特征组合的目标跟踪算法

为了克服目前大多数观测模型在小样本空间中鲁棒性不高的弱点,文中在粒子滤波框架下提出基于局部特征组合的粒子滤波视频跟踪算法。局部特征能更有效描述目标模板细节信息,可降低特征匹配中目标形变、光照变化和部分遮挡的影响。该方法借鉴混合高斯模型思想,采用多模式描述有效局部观测信息,这种融合策略更加准确可靠,能够较好地通过最新观测减轻了粒子退化现象,从而提高目标跟踪效率。小样本空间一定程度上降低了粒子数量和计算代价。实验结果表明该算法相比单一特征或一般多特征融合跟踪算法具有优越性,并能实现复杂场景下的目标跟踪。

Object Tracking Algorithm Based on Partial Feature Combination

In order to avoid the poor robustness based on most of present observation models in small sample space, a particle filter algorithm for visual tracking based on partial feature combination is proposed. Partial features can represent the detail of target template effectively, and can alleviate the affection of object deformation, illumination change and partial occlusion in feature matching. The proposed method employs the idea of mixture of Gaussian and uses multiple modes to represent valid partial observation information. The strategy of fusion is more precise and reliable, thus can overcome the degeneracy problem by new measurement and improve the efficiency of object tracking. The small sample space can reduce quantity of particle and computational load in a certain extent. Experimental results indicate the proposed method is more effective than tracking algorithm with single feature or common multi-features fusion, and it has good performance in complex scene.