Multi-target tracking in clutter with histogram probabilistic multi-hypothesis tracker

This study presents a recently developed tracking algorithm, namely histogram probabilistic multi-hypothesis tracker (H-PMHT), a modified version of PMHT, for multi-target tracking. Even though the theory of H-PMHT could be easily extended to multi-dimensional case, its applications have only been realized for one-dimensional cases. In this work the theory of H-PMHT has been extended into two-dimensional case and its performance has been compared to that of interacting multi-model probabilistic data association filter (IMMPDAF) with amplitude information (IMMPDAF-AI). Simulation results reveal that H-PMHT algorithm outperforms the IMMPDAF-AI under various conditions explained in the following sections.