The angular glint in the near field plays an important role on radar tracking errors. To predict it more efficiently for electrically large targets, a new method based on graphical electromagnetic computing （GRECO） ...The angular glint in the near field plays an important role on radar tracking errors. To predict it more efficiently for electrically large targets, a new method based on graphical electromagnetic computing （GRECO） is proposed. With the benefit of the graphic card, the GRECO prediction method is faster and more accurate than other methods. The proposed method at the first time considers the special case that the targets cannot be completely covered by radar beams, which makes the prediction of radar tracking errors more self-contained in practical circumstances. On the other hand, the process of the scattering center extraction is omitted, resulting in possible angular glint prediction in real time. Comparisons between the simulation results and the theoretical ones validate its correctness and value to academic research and engineering applications.展开更多
极低频电磁台网成功观测到大量的Pc1地磁脉动事件,研究极低频Pc1地磁脉动的自动识别方法对于全面分析地球空间电磁物理环境具有重要意义.本文采用了YOLOv8目标检测网络、ResNet残差网络和定向特征增强技术,提出了一种基于计算机视觉的Pc...极低频电磁台网成功观测到大量的Pc1地磁脉动事件,研究极低频Pc1地磁脉动的自动识别方法对于全面分析地球空间电磁物理环境具有重要意义.本文采用了YOLOv8目标检测网络、ResNet残差网络和定向特征增强技术,提出了一种基于计算机视觉的Pc1地磁脉动自动识别模型(Automatic Detection Model for Pc1 Geomagnetic Pulsation,简称ADM-Pc1).以大连台站和丽江台站的极低频观测数据为例,利用2015—2016年的数据作为训练集进行模型的监督学习,并使用2017—2022年的数据作为测试集对模型性能进行评估.实验结果显示,ADM-Pc1模型的F1-Score值达到了95%,错分率仅为0.9%,虚警率仅为5.8%,漏检率仅为9%,处理1天数据平均耗时是2.72 s,显著优于现有的最优识别模型.这表明,ADM-Pc1模型在识别效果和计算速度方面均能更好地满足实际工程需求.展开更多
基金supported by the National Natural Science Foundation of China (60871069)
文摘The angular glint in the near field plays an important role on radar tracking errors. To predict it more efficiently for electrically large targets, a new method based on graphical electromagnetic computing （GRECO） is proposed. With the benefit of the graphic card, the GRECO prediction method is faster and more accurate than other methods. The proposed method at the first time considers the special case that the targets cannot be completely covered by radar beams, which makes the prediction of radar tracking errors more self-contained in practical circumstances. On the other hand, the process of the scattering center extraction is omitted, resulting in possible angular glint prediction in real time. Comparisons between the simulation results and the theoretical ones validate its correctness and value to academic research and engineering applications.
文摘极低频电磁台网成功观测到大量的Pc1地磁脉动事件,研究极低频Pc1地磁脉动的自动识别方法对于全面分析地球空间电磁物理环境具有重要意义.本文采用了YOLOv8目标检测网络、ResNet残差网络和定向特征增强技术,提出了一种基于计算机视觉的Pc1地磁脉动自动识别模型(Automatic Detection Model for Pc1 Geomagnetic Pulsation,简称ADM-Pc1).以大连台站和丽江台站的极低频观测数据为例,利用2015—2016年的数据作为训练集进行模型的监督学习,并使用2017—2022年的数据作为测试集对模型性能进行评估.实验结果显示,ADM-Pc1模型的F1-Score值达到了95%,错分率仅为0.9%,虚警率仅为5.8%,漏检率仅为9%,处理1天数据平均耗时是2.72 s,显著优于现有的最优识别模型.这表明,ADM-Pc1模型在识别效果和计算速度方面均能更好地满足实际工程需求.
