An algorithm based on hierarchical clustering for multi-target tracking of multi-sensor data fusion
Hao Wang, Tangxing Liu, Qing Bu, Bo Yang · 2016
This paper proposes an efficient algorithm to deal with multi-target tracking of multi-sensor data fusion. The radar tracks have complex patterns such as irregular shapes, have no overlapping ranges, track number is uncertain and dense targets problem etc. Different solutions for different requirements may be impractical. To solve this problem, the similar score between track sets is defined by generalized hausdorff distance first. Then, based on the hierarchical clustering model the cluster search tree is presented, which can efficiently determine the optimal classification result. Through experiments, the effectiveness of the hierarchical clustering algorithm proposed in this paper is verified.