• 首页
  • 期刊简介
  • 编委会
  • 版权声明
  • 投稿指南
  • 期刊订阅
  • 相关下载
    雷达数据
    下载专区
  • 过刊浏览
  • 联系我们
引用本文:汪 翔, 王彦斌, 汪育苗, 崔国龙. 基于图神经网络的多尺度特征融合雷达目标检测方法[J]. 雷达科学与技术, 2025, 23(1): 39-47.[点击复制]
WANG Xiang, WANG Yanbin, WANG Yumiao, CUI Guolong. Graph Neural Network Based Radar Target Detection Method with Multi⁃Scale Feature Fusion[J]. Radar Science and Technology, 2025, 23(1): 39-47.[点击复制]
【打印本页】   【下载PDF全文】   【查看/发表评论】  【下载PDF阅读器】  【关闭】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 264次   下载 97次 本文二维码信息
码上扫一扫!
分享到: 微信 更多
字体:加大+|默认|缩小-
基于图神经网络的多尺度特征融合雷达目标检测方法
汪 翔, 王彦斌, 汪育苗, 崔国龙
电子科技大学信息与通信工程学院, 四川成都 611731
摘要:
本文针对复杂杂波环境下的雷达目标检测问题,提出了一种基于图神经网络的多尺度特征融合雷达目标检测方法,该方法利用多个脉冲回波之间的特征关联性检测目标。具体而言,其首先利用多个级联的特征提取模块从回波中提取多尺度特征。然后,该方法利用多尺度特征构造多个有向完全图,图中每个节点对应一个脉冲。之后,每个节点利用图神经网络的消息传播机制聚合其邻居节点的信息,以此学习脉间回波高阶相关性。进一步地,该方法融合多尺度特征以丰富对目标和杂波的特征表示。最后,对融合后特征进行非线性映射,以二分类的形式得到检测结果。基于公开雷达数据集的试验验证了所提方法的有效性。
关键词:  雷达目标检测  杂波环境  图神经网络  多尺度特征融合
DOI:DOI:10.3969/j.issn.1672-2337.2025.01.004
分类号:TN959.1+1
基金项目:国家自然科学基金(No.62271126)
Graph Neural Network Based Radar Target Detection Method with Multi⁃Scale Feature Fusion
WANG Xiang, WANG Yanbin, WANG Yumiao, CUI Guolong
School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 61731, China
Abstract:
Aiming at the problem of radar target detection in complex clutter environment,a graph neural network?based radar target detection method with multi?scale feature fusion is proposed in this paper, which uses the feature correlations among multiple pulse echoes for the detection. Firstly, the proposed method utilizes multiple cascaded feature extraction modules to extract multi?scale features. Subsequently, it constructs multiple directed complete graphs using muti?scale features, where a node in a graph corresponds to a pulse, and the node features of each graph are the features of the corresponding scale. Then, each node can aggregate the information of its neighbors using the graph neural network, and the proposed method can therefore learn the higher?order correlation among the pulse echoes. Further, the proposed method fuses multi?scale features to enrich the feature representation of the target and the clutter. Finally, the fused features are mapped nonlinearly, and the detection results are obtained in the form of binary classification. The effectiveness of the proposed method is verified using the public radar database.
Key words:  radar target detection  clutter environment  graph neural network  multi⁃scale feature fusion

版权所有:《雷达科学与技术》编辑部 备案:XXXXXXX
主办:中国电子科技集团公司第三十八研究所 地址:安徽省合肥市高新区香樟大道199号 邮政编码:230088
电话:0551-65391270 电子邮箱:radarst@163.com
技术支持:北京勤云科技发展有限公司