论文部分内容阅读
本文针对红外自动寻的末制导系统提出了一种适合于检测星空背景下的远距离点目标的快速算法,它可用于检测静止不动、匀速运动和机动运动的点目标。该算法通过构造一个关于3D空间中数据集的轨迹表达的总代价函数将点目标轨迹提取问题转化成为总代价函数对目标参量的最优化问题,通过迭代算法实现这个最优化,文章证明了算法的收敛性,并估算了该检测算法的检测性能,实验表明该方法适合于小SNR时的点目标检测。
In this paper, we propose a fast algorithm which is suitable for detecting long-distance point targets in the sky background. It can be used to detect the point targets of stationary, uniform and maneuvering motions. This algorithm transforms the point target trajectory extraction problem into the optimization problem of the target parameters by constructing a total cost function of the trajectory representation in the dataset in 3D space. The iterative algorithm is used to realize this optimization. The paper proves that the algorithm The convergence of the algorithm and the performance of the proposed algorithm are evaluated. Experiments show that this method is suitable for the detection of point targets at small SNRs.