论文部分内容阅读
本文讨论了利用多种异质特征数据诸如空间状态、信号幅度、多普勒频移等信息来实现多目标关联跟踪的方法。文中由品质函数的概念定义了一种关联测度，通过关联测度定量地将可能获得的多特征数据引入关联跟踪算法中。文中阐明这一关联测度是常规关联决策的推广。为验证这一方法的性能，本文还进行了计算机仿真实验，并给出实验结果。
In this paper, we discuss the method to realize multi-target correlation tracking by using a variety of heterogeneous data such as spatial state, signal amplitude and Doppler shift. In this paper, an association measure is defined by the concept of quality function, and the multi-characteristic data that can be obtained are quantitatively introduced into the correlation tracking algorithm through the correlation measure. The paper states that this correlation measure is a generalization of the conventional decision-making. In order to verify the performance of this method, we also carry out computer simulation experiments and give the experimental results.