针对交互式多模型目标跟踪算法中模型转移概率固定对跟踪精度造成的影响,提出了一种隐马尔科夫模型修正的模型转移概率自适应交互多模跟踪算法。该算法通过对跟踪过程建立隐马尔科夫模型,采用Viterbi 算法求解修正系数,在检测到目标运动发生机动性变化时,将修正系数用于交互式多模型算法以达到实时调整模型转移概率的目的。仿真结果表明,该算法的跟踪结果优于传统的交互式多模型算法,具有很好的稳健性、实时性,有效降低了主观因素对跟踪精度造成的影响。

Aiming at the influence of the fixed pair tracking accuracy of model transition probability in the interactive multi-model target tracking algorithm, a modified Markov model modified model transition probability adaptive interactive multi-mode tracking algorithm was proposed. The algorithm establishes a hidden Markov model for the tracking process, uses the Viterbi algorithm to solve the correction coefficient, and uses the correction coefficient in an interactive multi-model algorithm to achieve real-time adjustment of the model transition probability when detecting the movement of the target motion. The simulation results show that the tracking result of this algorithm is superior to the traditional interactive multi-model algorithm. It has good robustness and real-time performance and effectively reduces the impact of subjective factors on the tracking accuracy.

TP391.9

内蒙古自治区科技计划资助项目(201502013-1)