航空学报 > 2022, Vol. 43 Issue (S1): 726922-726922   doi: 10.7527/S1000-6893.2022.26922

一种轻量化SAR图像舰船目标斜框检测方法

苏航1, 徐从安1,2, 姚力波1, 李健伟3, 凌青1, 高龙1   

  1. 1. 海军航空大学 信息融合研究所,烟台 264000;
    2. 北京理工大学 前沿技术研究院,济南 250300;
    3. 392877部队,舟山 316000
  • 收稿日期:2022-01-10 修回日期:2022-02-10 发布日期:2022-04-12
  • 通讯作者: 徐从安,E-mail:xcatougao@163.com E-mail:xcatougao@163.com
  • 基金资助:
    国家自然科学基金(61790550,61790554,61971432,62022092);中国科协青年人才托举工程(2020-JCJQ-QT-011)

A lightweight oriented ship detection method in SAR images

SU Hang1, XU Congan1,2, YAO Libo1, LI Jianwei3, LING Qing1, GAO Long1   

  1. 1. Institute of Information Fusion, Naval Aviation University, Yantai 264000, China;
    2. Advanced Technology Research Institute, Beijing Institute of Technology, Jinan 250300, China;
    3. Troops of 92877, Zhoushan 316000, China
  • Received:2022-01-10 Revised:2022-02-10 Published:2022-04-12
  • Supported by:
    National Natural Science Foundation of China (61790550, 61790554, 61971432, 62022092);Young Elite Scientists Sponsorship Program by China Association for Science and Technology (2020-JCJQ-QT-011)

摘要: 针对目前基于深度学习的合成孔径雷达(SAR)图像舰船目标斜框检测方法难以满足实时化检测需求的问题,提出了一种轻量化SAR舰船目标斜框检测方法。该方法基于无锚框框架,设计了轻量化的网络结构,对模型参数量和运行速度进行了优化,可直接从头训练。同时,为解决基于角度回归的斜框参数表示方法存在的角度敏感性问题,提出了基于旋转向量的斜框参数表示方法。在公开的SAR图像舰船目标检测数据集上的实验结果表明,所提方法在无需预训练的情况下取得了与迁移学习方法相近的检测精度,模型参数量和运行速度取得了最优结果,充分验证了所提方法的有效性。

关键词: SAR图像, 舰船目标斜框检测, 轻量化, 从头训练, 旋转向量

Abstract: Current oriented ship detection methods in Synthetic Aperture Radar (SAR) images based on deep learning cannot meet the requirements of real-time detection. This paper proposes a lightweight oriented ship detection method. A lightweight network structure is designed based on the anchor-free framework. The number of model parameters and running speed are optimized, so that the model can be directly trained from scratch. To solve the problem of angle sensitivity existing in the method based on angle regression, an oriented bounding box representation method is proposed based on the rotated vector. Experiments are conducted on public SAR ship detection dataset. The results show that the proposed method can reduce model parameters and improve the detection speed while maintaining the detection accuracy, which fully verifies the effectiveness of the method.

Key words: SAR image, oriented ship detection, lightweight, train from scratch, rotated vector

中图分类号: