基于特征可分性和稳定性度量的多特征融合目标跟踪算法

针对单一特征目标跟踪算法鲁棒性较差的问题, 提出一种基于特征可分性和稳定性度量的多特征融合目标跟踪算法. 在粒子滤波框架下, 通过计算不同特征对目标和背景的可区分性和稳定性, 设置重要性权值并自适应选择区分能力强、稳定性好的特征描述目标, 建立多特征融合目标模型. 在状态转移过程中, 给出一种基于特征稳定性度量的选择性模板更新策略, 并进行遮挡处理. 实验结果表明, 所提出的算法能够在复杂场景下鲁棒地跟踪目标.

Fusing multi-features target tracking algorithm based on discriminability and stability of features

In order to solve the poor robustness problem of using single feature in the target tracking process, an adaptive fusing multi-features tracking algorithm is proposed based on the discriminability and stability of features in the particle filter framework. Several reliable features are adaptively selected by calculating their discriminative ability and stability, which are used to describe the target model, the multi-features fusion target model is established and the importance weights of features are set. In the process of state transition, a selective template updating method is presented based on the measurement of feature stability, and the occlusion problem is handled. Experimental results show that the proposed method can track the target under the complex scene in robust performance.