基于形态匹配聚类的近邻扩展目标跟踪算法

针对传统扩展目标跟踪（Extended Target Tracking, ETT）算法在处理近邻目标时面临的计算效率低下和跟踪不准确的问题，提出了一种形态匹配聚类量测集划分与高斯逆威沙特概率假设密度（Gaussian Inverse Wishart Probability Hypothesis Density, GIW-PHD）滤波器相结合的跟踪处理方法。该方法首先由GIW-PHD滤波器得到预测的目标状态，其次使用DBSCAN(Density-Based Spatial Clustering of Applications with Noise, DBSCAN)算法完成量测集的初步划分，在此基础上利用较高权重的预测分量实现对多个近邻目标混合量测簇的判断，进而使用椭圆形状约束（Elliptic Shape Constraint, ESC）的FCM(Fuzzy C-Means, FCM)算法（ESC-FCM）对混合簇进行二次划分得到更精确的划分结果，最后将划分结果合并后送入GIW-PHD滤波器完成目标状态的更新。仿真结果表明，本文所提量测集划分方法能够充分利用GIW-PHD滤波器预测步获取...

Extended Target Tracking Algorithm Based on Morphological Matching Clustering in Near Spaced Environment

? ?Aiming at the problems of low computational efficiency and inaccurate tracking of traditional extended target tracking （ETT） algorithm under the scenario where the targets are near spaced， a tracking algorithm combining the partition of morphological matching clustering measurement set and Gaussian inverse Wishart probability hypothesis density （GIW-PHD） filter is proposed in this paper. Firstly， the predicted target state is obtained by the GIW-PHD filter. Density based spatial clustering of applications with noise（DBSCAN） algorithm is used to preliminarily partition the measurement set. On this basis， the prediction component with higher weight is used to judge the mixed measurement cluster of multiple adjacent targets， and then the fuzzy c-means （FCM） algorithm with elliptical shape constraints is used to partition the mixed clusters to obtain more accurate partition results. Finally， the partition results are integrated and sent to the GIW-PHD filter to update the target state. The simulation results show that the proposed method can make full use of the predicted target position and shape information obtaining by GIW-PHD filter to accurately partition the mixed clusters， and thus to fulfill fast and accurate estimation of the kinematic state and extended state for the adjacent extended targets.