论文部分内容阅读
根据确认的众多量测和众多目标跟踪窗之间的几何关系,引入确认矩阵并计算所有联合事件及其对应的参数,不论量测是否落入跟踪窗相交区域,根据JPDA算法计算每一个量测与其可能的各个源目标之间互联的概率。将互联的概率与Kalman滤波器相结合从而完成对每一个目标的预测和更新。理论及实验结果表明,该算法适用于序列图像密集杂波环境下的全程跟踪,并取得了一定的理论和仿真结果。
According to the confirmation of the many measurements and many target tracking window between the geometric relationship, the introduction of the confirmation matrix and calculate all joint events and their corresponding parameters, whether the measurement falls into the intersection of the tracking window region, calculated according to JPDA algorithm for each measurement Probability of interconnection with its possible source targets. The probability of interconnection is combined with the Kalman filter to complete the prediction and updating of each target. Theoretical and experimental results show that the proposed algorithm is suitable for the full tracking of dense clutter in a sequence of images and has obtained some theoretical and simulation results.