Online Multi-Object Tracking via maximum entropy Intuitionistic Fuzzy Data Association
Zhan Xi-yang, Xiaoli Wang, Liangqun Li · 2018
Aiming at the uncertainty of target occlusion and background cluster in multi-target tracking of a single camera, an Online Multi-Object Tracking via maximum entropy Intuitionistic Fuzzy Data Association algorithm is proposed. Firstly, the distance function between the target and the observation is obtained by calculating the similarity of multiple features between the target and each observation, and local information of the target and the observation is introduced. Moreover, the optimized intuitionistic fuzzy membership degree of maximum entropy intuitionistic clustering is used to construct the correlation matrix between the target and the observation, and the correlation between the target and the observation is achieved. Finally, using the Kalman filter to filter and predict on the association target, the uncorrelated target, and the uncorrelated observation respectively. The experimental results show that the proposed algorithm is robust and accurate in tracking multiple targets continuously and effectively when they are occluded and interfered with each other. Simulation results verify the effectiveness of the algorithm.