计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (10): 151-158.doi: 10.11896/jsjkx.250100097

• 计算机图形学&多媒体 • 上一篇    下一篇

基于小目标特征增强RT-DETR的SAR图像舰船目标检测方法

张弘森1,2, 吴蔚2, 徐建2, 吴飞1,2, 季一木3   

  1. 1 南京邮电大学自动化学院、人工智能学院 南京 210003
    2 信息系统工程全国重点实验室 南京 210003
    3 南京邮电大学计算机学院 南京 210003
  • 收稿日期:2025-01-15 修回日期:2025-04-24 出版日期:2025-10-15 发布日期:2025-10-14
  • 通讯作者: 吴飞(wufei_8888@126.com)
  • 作者简介:(zhs9586@163.com)
  • 基金资助:
    信息系统工程全国重点实验室开放基金(05202305);国家自然科学基金(62076139)

Ship Detection Method for SAR Images Based on Small Target Feature Enhanced RT-DETR

ZHANG Hongsen1,2, WU Wei2, XU Jian2, WU Fei1,2, JI Yimu3   

  1. 1 College of Automation,Nanjing University of Posts and Telecommunications,Nanjing 210003,China
    2 National Key Laboratory of Information Systems Engineering,Nanjing 210003,China
    3 School of Computer Science,Nanjing University of Posts and Telecommunications,Nanjing 210003,China
  • Received:2025-01-15 Revised:2025-04-24 Online:2025-10-15 Published:2025-10-14
  • About author:ZHANG Hongsen,born in 2001,postgraduate.His main research interest is pattern recognition.
    WU Fei,born in 1989,Ph.D,professor,is a member of CCF( No.86938S).His main research interests include pattern recognition and machine learning.
  • Supported by:
    Science and Technology on Information System Engineering Laboratory(05202305) and National Natural Science Foundation of China(62076139).

PDF (PC)

1

摘要： 在舰船检测任务中,SAR图像因其优异的成像条件被广泛应用于海洋资源管理、海上救援等场景。然而,舰船目标尺寸较小和海面杂波等问题,导致传统目标检测算法的性能表现不佳。近年来,许多算法通过引入Transformer的注意力机制,实现更好的语义解释;或采用较为复杂的网络结构,以提高特征提取能力。这在一定程度上改善了检测精度,却牺牲了检测速度。对此,提出了一种基于小目标特征增强RT-DETR的SAR图像舰船目标检测方法。该方法由以下3部分组成:1)大模型提示生成网络:借助多模态大模型的零样本学习能力生成提示,以提取图像模态中更具判别性的信息;2)AIFI-EAA模块:以RT-DETR为基线,改进尺度内特征交互模块,引入高效加性注意力机制,降低算法计算复杂度;3)轻量化小目标特征增强融合网络:在多尺度特征融合网络中加入小目标检测层,设计CSP-OmniKernel模块进行多尺度特征融合,提升小目标的检测性能。在SSDD,HRSID和SAR-Ship-Dataset 3个公开数据集上进行实验验证,结果表明所提方法在准确性上具有优势。

关键词: 舰船检测, SAR图像, 轻量化, RT-DETR, 小目标检测

Abstract: In ship detection tasks,SAR images are widely used in maritime resource management,search and rescue,and other scenarios due to their excellent imaging conditions.However,traditional target detection algorithms perform poorly due to issues such as the small size of ships and sea surface clutter.Recently,many algorithms have introduced the attention mechanism of Transformer to achieve better semantic interpretation or adopted more complex network structures to improve feature extraction capabilities.This has improved detection accuracy to some extent but has sacrificed detection speed.This paper proposes a ship detection method for SAR images based on small target feature enhanced RT-DETR.The method consists of three parts:1)Large model prompt generation network:Leveraging the zero-shot learning capability of multimodal large models,prompts are generated to extract more discriminative information from the image modality;2)AIFI-EAA module:Using RT-DETR as the baseline,the scale-invariant feature interaction module is improved by introducing an efficient additive attention mechanism to reduce the computational complexity of the algorithm;3)Lightweight small target feature enhancement fusion network:A small target detection layer is added to the multi-scale feature fusion network,and the CSP-OmniKernel module is designed for multi-scale feature fusion to enhance small target detection performance.Experiments on three public datasets(SSDD,HRSID,and SAR-Ship-Dataset) demonstrate that the proposed method has advantages in terms of accuracy.

Key words: Ship detection,SAR image,Lightweight,RT-DETR,Small target detection

中图分类号: 

  • TP751
[1]CUI Z,QUAN H,CAO Z,et al.Sar target cfar detection via gpu oarallel operation[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2018,11(12):4884-4894.
[2]LIN H,LIU J,LI X,et al.Dcea:Detr with concentrated deformable attention for end-to-end ship detection in sar images[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2024,17:17292-17307.
[3]REN S,HE K,GIRSHICK R,et al.Faster r-cnn:Towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,39(6):1137-1149.
[4]REDMON J,DIVVALA S,GIRSHICK R,et al.You only look once:Unified,real-time object detection[C]//IEEE Conference on Computer Vision and Pattern Recognition.2016:779-788.
[5]CARION N,MASSA F,SYNNAEVE G,et al.End-to-end object detection with transformers[C]//European Conference on Computer Vision.2020:213-229.
[6]ZHAO Y,LV W,XU S,et al.Detrs beat yolos on real-time object detection[C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition.2024:16965-16974.
[7]LIU L,FU L,ZHANG Y,et al.Clfr-det:Cross-level feature refinement detector for tiny-ship detection in sar images[J].Knowledge-Based Systems,2024,284:111284.
[8]CAI X,LAI Q,WANG Y,et al.Poly kernel inception network for remote sensing detection[C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition.2024:27706-27716.
[9]QIN C,WANG X,LIU Y,et al.A novel end-to-end transformer network for small scale ship detection in sar images[C]//International Geoscience and Remote Sensing Symposium.2024:8158-8162.
[10]BEYER L,STEINER A,PINTO A S,et al.Paligemma:A versatile 3b vlm for transfer[J].arXiv:2407.07726,2024.
[11]SHAKER A,MAAZ M,RASHEED H,et al.Swiftformer:Efficient additive attention for transformer-based real-time mobile vision applications[C]//IEEE/CVF International Conference on Computer Vision.2023:17425-17436.
[12]CUI Y,REN W,KNOLL A.Omni-kernel network for image restoration[C]//AAAI Conference on Artificial Intelligence.2024:1426-1434.
[13]LI J,QU C,SHAO J.Ship detection in sar images based on an improved faster r-cnn[C]//SAR in Big Data Era:Models,Me-thods and Applications.2017:1-6.
[14]WEI S,ZENG X,QU Q,et al.HRSID:A high-resolution sar images dataset for ship detection and instance segmentation[J].IEEE Access,2020,8:120234-120254.
[15]WANG Y,WANG C,ZHANG H,et al.A sar dataset of ship detection for deep learning under complex backgrounds[J].Remote Sensing,2019,11(7):765.
[16]HE K,ZHANG X,REN S,et al.Deep residual learning for image recognition[C]//IEEE Conference on Computer Vision and Pattern Recognition.2016:770-778.
[17]CHEN J,LEI B,SONG Q,et al.A hierarchical graph network for 3d object detection on point clouds[C]//IEEE/CVF Confe-rence on Computer Vision and Pattern Recognition.2020:392-401.
[18]DING X,ZHANG X,MA N,et al.Repvgg:Making vgg-styleconvnets great again[C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition.2021:13733-13742.
[19]YAN H,LIU Y L,JIN L W,et al.The development,applica-tion,and future of llm similar to chatgpt[J].Journal of Image and Graphics,2023,28(9):2749-2762.
[20]LI L H,ZHANG P,ZHANG H,et al.Grounded language-image pre-training[C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition.2022:10965-10975.
[21]XU Y,ZHANG M,FU C,et al.Multi-modal queried object detection in the wild[C]//Advances in Neural Information Processing Systems.2024:1-18.
[22]ZHOU K,YANG J,LOY C C,et al.Learning to prompt for vision-language models[J].International Journal of Computer Vision,2022,130(9):2337-2348.
[23]KHATTAK M U,RASHEED H,MAAZ M,et al.Maple:Multi-modal prompt learning[C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition.2023:19113-19122.
[24]ZHANG H,LI F,LIU S,et al.Dino:Detr with improved denoi-sing anchor boxes for end-to-end object detection[J].arXiv:2203.03605,2022.
[25]ZHU X,SU W,LU L,et al.Deformable detr:Deformable transformers for end-to-end object detection[C]//International Conference on Learning Representations.2021:1-16.
[26]ZONG Z,SONG G,LIU Y.Detrs with collaborative hybrid assignments training[C]//IEEE/CVF International Conference on Computer Vision.2023:6748-6758.
[27]JOCHER G,NISHIMURA K,MINEEVA T,et al.Yolov8 byultralytics[EB/OL].https://github.com/ultralytics/ultraly-tics.
[28]SHEN J,BAI L,ZHANG Y,et al.Ellk-net:An efficient light-weight large kernel network for sar ship detection[J].IEEE Transactions on Geoscience and Remote Sensing,2024,62:5221514.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发