基于形态滤波和推理矩阵的群目标跟踪  

Group Target Tracking Based on Morphology Filtering and Reasoning Matrix

作　　者：王胜华 邓宇坤 赵晨博 徐家宁 WANG Shenghua;DENG Yukun;ZHAO Chenbo;XU Jianing(School of Communications and Information Engineering,Xi′an University of Posts and Telecommunications,Xi′an Shaanxi 710061,China)

机构地区：[1]西安邮电大学通信与信息工程学院,陕西西安710061

出　　处：《现代雷达》2023年第10期94-99,共6页Modern Radar

摘　　要：针对雷达群目标跟踪时存在交叉、分裂和合并问题,提出一种基于形态滤波的群目标跟踪方法。首先建立笛卡尔坐标系下的群目标量测分布初始图像矩阵,对图像矩阵进行形态滤波和连通域标记处理,获得场景中的群目标个数以及每个群目标的几何区域。然后计算每个群目标的等效质心,并提出推理矩阵来判断群目标之间的交互行为,完成对群目标交叉、分裂和合并行为下的目标关联和稳定跟踪。最后通过仿真验证了该方法对群目标跟踪的有效性。A group target tracking method based on image morphology processing is proposed aiming at the problem of group target tracking under crossing,splitting and merging behavior.First,the initial image matrix is established for the measurement distribution of group targets in the Cartesian coordinate system.Then image morphology filtering and connected domain marking processing are performed on the image matrix to obtain the number of the group targets and the geometric area of each group target.The method calculates the equivalent center of each group target,and judges the interactive behavior between group targets based on the proposed reasoning matrix.And then the target association and robust group target tracking when existing crossing,splitting and merging is realized.Finally,the effectiveness of the proposed method for group target tracking is verified by numerical simulation.

关 键 词：群目标跟踪 形态滤波 连通域标记 推理矩阵 目标关联 

分 类 号：TN957.52[电子电信—信号与信息处理]