基于平行线坐标变换的检测前跟踪算法  

Track before detect algorithm based on parallel-line-coordinate transformation

作　　者：薄钧天 王国宏 于洪波 彭志刚 BO Juntian;WANG Guohong;YU Hongbo;PENG Zhigang(Institute of Information Fusion,Naval Aeronautical University,Yantai 264001,China;Qingdao Branch,Naval Aeronautical University,Qingdao 266041,China)

机构地区：[1]海军航空大学信息融合研究所,烟台264001 [2]海军航空大学青岛校区,青岛266041

出　　处：《航空学报》2022年第8期564-577,共14页Acta Aeronautica et Astronautica Sinica

基　　金：国家自然科学基金(61731023,61701519,61671462);山东省自然科学基金(ZR2020MF015)。

摘　　要：针对微弱目标检测对提高检测概率和优化实时性的需求,提出一种平行线坐标变换检测前跟踪算法。首先将量测点坐标映射为径向距离-时间坐标并规格化,建立以y轴为对称轴的平行线,按照变换规则实现点到线的转化,网格化平面后进行票数积累和能量积累,再进行峰值提取与航迹修正,后输出航迹。仿真结果显示:基于平行线坐标变换检测前跟踪(PT-TBD)算法的检测概率要优于基于Hough变换的检测前跟踪(HT-TBD)算法,二者虚假航迹率相近,运算速度要快于HT-TBD算法,是一种新的效果更好的基于投影变换的检测前跟踪算法。To optimize the real-time performance of target detection,a track-before-detect algorithm is proposed based on parallel-line-coordinate transformation.First,the measurement point coordinates are mapped to the radial distance-time coordinate and normalized and a parallel line with the y axis as the symmetrical axis is established,and then point-to-line conversion is realized according to transformation rules.After the plane is gridded,the number of votes and energy are accumulated,and then peak extraction and track correction are performed.Finally,the track of the target is output.The simulation results show that the detection probability of the Parallel-line-coordinate Transformation based Track-Before-Detect(PT-TBD)algorithm is better than that of the modified Hough Transformation based Track-Before-Detect(HT-TBD)algorithm,the false track rate of the two is similar,and the calculation speed of the PT-TBD is faster than that of the HT-TBD algorithm.

关 键 词：微弱目标 平行线坐标变换 非相参积累 检测前跟踪 积累损失 

分 类 号：V243.2[航空宇航科学与技术—飞行器设计] TN957[电子电信—信号与信息处理]