基于跟踪信息的多帧联合检测方法

传统多帧联合检测方法针对搜索阶段进行设计, 由于无任何目标信息, 需要采用较高的检测门限以降低虚警数量。然而, 一旦成功建航, 目标存在性已确定, 可利用目标跟踪信息调整多帧联合检测策略, 以提升微弱目标检测能力。基于上述考虑, 本文提出了一种基于跟踪信息的多帧联合检测方法。首先利用当前时刻跟踪信息建立预测波门, 然后设定较低的单帧检测门限保障微弱目标信号的高概率检测, 最后在航迹恒虚警约束下调整多帧联合检测门限, 从而完成航迹维持和撤销。仿真实验表明所提方法可有效提高目标检测概率和跟踪维持概率。

Multi-frame joint detection method based on tracking information

Traditional multi-frame joint detection methods are mainly designed for searching stage and higher detection threshold is required to decrease false alarms due to no target information. However, once track initiation is successful, target existence has already been confirmed, and the strategy of multi-frame joint detection can be adjusted ultilizing the target tracking information for improved detection performance of weak targets. Based on above considerations, a multi-frame joint detection method based on tracking information is presented in this paper. Firstly, the predicted gate is established based on current tracking information. Secondly, lower threshold of single frame detection is set to ensure high detection probability of weak target signals. Finally, the threshold of multi-frame joint detection under the constraint of constant track false alarm rate is adjusted for track maintenance and track deletion. Simulation experiments show that the presented method can significantly improve target detection probability and track maintenance probability.