| 注册
首页|期刊导航|信号处理|Pol-ISARSpaceTarget-1.0:极化ISAR空间目标精细识别电磁仿真数据集

Pol-ISARSpaceTarget-1.0:极化ISAR空间目标精细识别电磁仿真数据集

陈思伟 李铭典 崔兴超

信号处理2025,Vol.41Issue(8):1443-1454,12.
信号处理2025,Vol.41Issue(8):1443-1454,12.DOI:10.12466/xhcl.2025.08.013

Pol-ISARSpaceTarget-1.0:极化ISAR空间目标精细识别电磁仿真数据集

Pol-ISARSpaceTarget-1.0:Polarimetric ISAR Electromagnetic Simulation Dataset for Space Target Fine Recognition

陈思伟 1李铭典 1崔兴超1

作者信息

  • 1. 国防科技大学电子科学学院,湖南长沙 410073
  • 折叠

摘要

Abstract

Space targets,such as satellites,play a critical role in remote sensing mapping,meteorological monitoring,wireless communication,and reconnaissance,which serve as essential objects in space situational awareness.Polarimet-ric inverse synthetic aperture radar(ISAR)offers unique advantages for space target awareness by providing high-resolution images and polarimetric scattering information,which are sensitive to target structures.Nevertheless,chal-lenges persist due to the scarcity of publicly available real measured data for polarimetric ISAR space targets and the complexity of fine-grained component-level annotation,which remains underexplored.To address these challenges,this study introduced the Pol-ISARSpaceTarget-1.0 dataset,the first publicly available dataset for the fine recognition of po-larimetric ISAR space targets.The dataset consists of polarimetric ISAR images of six space targets,including radar sat-ellites with parabolic antennas,two radar satellites with plane antennas,a communication satellite,and an optical satel-lite.It also contains the semantic annotations of nine typical components.This study details the dataset composition,electromagnetic simulation and imaging,and annotation workflow.To validate its effectiveness,representative deep-learning object-recognition methods were adopted for comparison studies,producing benchmark results that offer valu-able references for researchers.The dataset features various kinds of space targets,comprehensive polarimetric informa-tion,and detailed component labels that provide fundamental data support for the fine classification and recognition of space targets.

关键词

雷达极化/逆合成孔径雷达/空间目标/数据集/目标识别

Key words

radar polarimetry/inverse synthetic aperture radar/space target/dataset/target recognition

分类

信息技术与安全科学

引用本文复制引用

陈思伟,李铭典,崔兴超..Pol-ISARSpaceTarget-1.0:极化ISAR空间目标精细识别电磁仿真数据集[J].信号处理,2025,41(8):1443-1454,12.

基金项目

军队高层次科技创新人才工程项目 ()

国家自然科学基金(U24B20189,62122091)The Military High-level Scientific and Technological Innovation Talent Project and the National Natural Science Foundation of China(U24B20189,62122091) (U24B20189,62122091)

信号处理

OA北大核心

1003-0530

访问量0
|
下载量0
段落导航相关论文