特征不连续下的目标跟踪问题优化仿真

Characteristics of Target Tracking Problem under Discrete Optimization Simulation

研究复杂背景下的运动目标准确跟踪问题。运动目标在未知的时间段内可能做未知的随机运动,跟踪特征众多,不可避免地出现跟踪特征丢失问题。由于特征丢失,会使得传统的粒子群跟踪算法中的目标内部出现较多的误判为背景的运动信息点,较难形成一个完整的目标轮廓,很难保证运动目标的连通性,跟踪准确性不高。为此提出一种概率寻优的未知目标视觉跟踪算法。通过计算跟踪目标中的多个相关概率分布,对多个相关概率进行寻优,保证在特征丢失的情况下,计算出最优跟踪解,克服传统方法的弊端。仿真结果表明,概率寻优方法能够在特征不连通的情况下,保证跟踪的准确性。

Study complex background of moving target tracking problem accurately. The paper put forward a visual tracking algorithm for unknown target based on probability searching. First of all, through the calculation of multiple correlation probability distribution in the tracked target , the probability of one or more related optimization, to ensure that in the condition of the missing features, we calculated the optimal tracking solution to overcome the disad- vantages of traditional methods. The simulation results show that this method can guarantee the accuracy of the tracking.