直方图概率多假设跟踪方法技术综述

直方图概率多假设跟踪(histogram probabilistic multi-hypothesis tracking, H-PMHT)算法是高效的多目标检测前跟踪(track before detect, TBD)方法, 其由概率多假设跟踪(probabilistic multi-hypothesis tracking, PMHT)算法发展起来。首先深入剖析了H-PMHT算法, 并重点分析比较了该算法和PMHT算法间的深刻联系, 分析了该算法的特点和适用条件; 其次, 介绍了近年来H-PMHT算法的改进和推广情况, 梳理了算法应用中的相关研究工作; 最后, 总结了算法的优缺点及需要改进的问题, 展望了其发展前景。

A survey on histogram probabilistic multi-hypothesis tracker technique

Histogram probabilistic multi-hypothesis tracking (H-PMHT) is an efficient multi-target track before detect (TBD) method, which is developed from the probabilistic multi-hypothesis tracking (PMHT). Firstly, a deep analysis of H-PMHT is made. Then the relationship between H-PMHT and PMHT are compared; meanwhile, the characteristics and application conditions of the algorithm are indicated. Next, the improvement and promotion of H-PMHT in recent years are introduced, and the related application researches of the algorithm are sorted out. Finally, the advantages and disadvantages of the algorithm are summarized; meanwhile, the development and application prospects of the algorithm are forecasted.