基于注意力机制的轻量级SAR船舶检测器

于楠晶; 冯大权; 朱颖; 张恒嘉; 陆平

doi:10.11959/j.issn.2096-3750.2024.00407

您当前的位置：

首页 >

文章列表页 >

基于注意力机制的轻量级SAR船舶检测器

理论与技术 | 更新时间：2025-01-20

- 基于注意力机制的轻量级SAR船舶检测器
- Lightweight attention-based SAR ship detector
- 物联网学报 2024年8卷第4期页码：156-166
- 作者机构：
  
  1.深圳大学电子与信息工程学院，广东深圳 518060
  2.中国信息通信研究院，北京 100191
  3.中兴通讯股份有限公司，广东深圳 518055
  4.移动网络和移动多媒体技术国家重点实验室，广东深圳 518055
- 作者简介：
  
  [ "于楠晶（1998‒ ），女，深圳大学电子与信息工程学院博士生，主要研究方向为目标检测算法、点云编解码算法等。" ]
  [ "冯大权（1986‒ ），男，博士，深圳大学电子与信息工程学院副教授、博士生导师，主要研究方向为沉浸式通信、VA/AR、频谱共享技术。" ]
  [ "朱颖（1987‒ ），女，中国信息通信研究院高级工程师，主要研究方向为移动通信射频测试技术。" ]
  [ "张恒嘉（1998‒ ），男，中兴通讯股份有限公司战略规划师、XRExplore平台工程师，移动网络和移动多媒体技术国家重点实验室成员，主要研究方向为深度学习、计算机视觉、SLAM、实时云渲染、空间计算等。" ]
  [ "陆平（1971‒ ），男，博士，移动网络和移动多媒体技术国家重点实验室副主任、正高级工程师，中兴通讯股份有限公司副总裁，主要研究方向为云计算、AR、媒体大数据。" ]
- 基金信息：
  
  深圳市科技计划;The Shenzhen Science and Technology Program(JCYJ20210324095209025)
- DOI：10.11959/j.issn.2096-3750.2024.00407
  中图分类号：
- 纸质出版日期：2024-12-10，
  
  收稿日期：2024-06-03，
  
  修回日期：2024-09-27，
- 稿件说明：
移动端阅览
于楠晶, 冯大权, 朱颖, 等. 基于注意力机制的轻量级SAR船舶检测器[J]. 物联网学报, 2024,8(4):156-166.

YU NANJING, FENG DAQUAN, ZHU YING, et al. Lightweight attention-based SAR ship detector. [J]. Chinese journal on internet of things, 2024, 8(4): 156-166.
于楠晶, 冯大权, 朱颖, 等. 基于注意力机制的轻量级SAR船舶检测器[J]. 物联网学报, 2024,8(4):156-166. DOI： 10.11959/j.issn.2096-3750.2024.00407.

YU NANJING, FENG DAQUAN, ZHU YING, et al. Lightweight attention-based SAR ship detector. [J]. Chinese journal on internet of things, 2024, 8(4): 156-166. DOI： 10.11959/j.issn.2096-3750.2024.00407.

摘要

合成孔径雷达（SAR

synthetic aperture radar）遥感图像凭其全天候、全时段优势，在军事侦察、交通监管等领域得到了广泛的应用。卷积神经网络因其较强的学习能力，被广泛用于SAR图像船舶检测算法。然而，SAR图像中船舶特征提取难度较大。此外，计算资源和内存空间受限，实际应用对算法推理速度需求较高。为此，提出了一种基于注意力的轻量级船舶检测（LASD

lightweight attention-based ship detector）算法。该算法设计了一种新的线性混合注意力残差模块，先后用全局通道注意力和局部空间注意力在深层特征空间中提取船舶潜在特征；基于跨阶段部分通道连接的空间金字塔池化模块优化多尺度特征融合质量，用串联小核池化组替换并行大核池化组以提升算法推理速度；设计了一种新的基于局部注意力的特征融合策略，在特征融合阶段利用局部注意力进一步扩大船舶和背景噪声的差异。在公开数据集SSDD和LS-SSDD-v1.0上的实验数据表明，LASD算法同时兼顾了检测精度和推理速度，相比其他先进算法更具竞争力。

Abstract

Synthetic aperture radar (SAR) remote sensing images have been widely applied in military reconnaissance and traffic supervision

owing to their all-weather and all-day abilities. With excellent learning performance

convolutional neural networks are employed in the SAR ship detection algorithms. However

it is difficult to extract features. In practical applications

computing resources and memory space are limited

and high inference speed is required. Therefore

a lightweight attention-based ship detector (LASD) was proposed. A novel linear hybrid attention module was designed which extracted potential ship features from deep-level space by using global channel attention and local spatial attention. A spatial pyramid pooling module based on cross-stage partial connections optimized the quality of multi-scale feature fusion

which replaced the parallel max-pooling group with large kernels with the serial max-poolings with small kernels to improve the inference speed. A novel feature fusion scheme via the local channel attention was suggested which widened the gap between the objects and background noise using local attention during the feature fusion. The results on the public datasets SSDD and LS-SSDD-v1.0 show that LASD achieves the balance of detection precision and inference speed

and is more competitive than the other advanced algorithms.

关键词

SAR船舶检测卷积神经网络注意力机制多尺度特征融合

Keywords

SARship detectionconvolutional neural networkattention mechanismmulti-scale feature fusion

references

ZHANG C J, LIU P, WANG H P, et al. NPA2Net: a nested path aggregation attention network for oriented SAR ship detection[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2024, 17: 9772-9789.

ZHANG J D, SHENG W X, ZHU H R, et al. MLBR-YOLOX: an efficient SAR ship detection network with multilevel background removing modules[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023, 16: 5331-5343.

LV Z S, LU J, WANG Q, et al. ESP-LRSMD: a two-step detector for ship detection using SLC SAR imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 5233516.

CHEN B J, YU C R, ZHAO S, et al. An anchor-free method based on transformers and adaptive features for arbitrarily oriented ship detection in SAR images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2012, 17: 2012-2028.

GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE Press, 2014: 580-587.

LI D, LIANG Q H, LIU H Q, et al. A novel multidimensional domain deep learning network for SAR ship detection[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 5203213.

ZHOU Y C, FU K, HAN B, et al. D-MFPN: a Doppler feature matrix fused with a multilayer feature pyramid network for SAR ship detection[J]. Remote Sensing, 2023, 15(3): 626.

REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE Press, 2016: 779-788.

林鑫伟, 徐志京, 黄海. 复杂背景下的SAR图像多尺度舰船检测[J]. 中国航海, 2023, 46(2): 17-24, 32.

LIN X W, XU Z J, HUANG H. Multi-scale detection of ship target against complex background out of SAR image[J]. Navigation of China, 2023, 46(2): 17-24, 32.

胥小我, 张晓玲, 张天文, 等. 基于自适应锚框分配与IOU监督的复杂场景SAR舰船检测[J]. 雷达学报, 2023, 12(5): 1097-1111.

XU X W, ZHANG X L, ZHANG T W, et al. SAR ship detection in complex scenes based on adaptive anchor assignment and IOU supervise[J]. Journal of Radars, 2023, 12(5): 1097-1111.

谢洪途, 姜新桥, 王国倩, 等. 基于改进CenterNet的轻量级无锚框SAR图像多尺度舰船检测算法[J]. 哈尔滨工程大学学报, 2024, 45(3): 504-516.

XIE H T, JIANG X Q, WANG G Q, et al. Lightweight and anchor-free frame method of multiscale ship detection based on improved CenterNet in SAR images[J]. Journal of Harbin Engineering University, 2024, 45(3): 504-516.

龙昊, 张思佳, 周晶, 等. 基于改进YOLOv5-ResNet的海上舰船SAR图像快速检测[J]. 宇航计测技术, 2024, 44(2): 52-59.

LONG H, ZHANG S J, ZHOU J, et al. Rapid detection of SAR images of naval vessels based on improved YOLOv5-ResNet[J]. Journal of Astronautic Metrology and Measurement, 2024, 44(2): 52-59.

邹珺淏,任酉贵,冷芳玲,等.LW-YOLOv7SAR:轻量SAR图像目标检测方法[J].小型微型计算机系统, 2024, 1-9.

ZHOU J H, REN Y G, LENG F L, et al. LW-YOLOv7SAR: lightweight SAR image object detection method[J]. Journal of Chinese Computer Systems, 2024, 1-9.

何旭鑫, 吴建平, 余咏, 等. 复杂背景下SAR舰船目标检测算法的改进[J]. 计算机技术与发展, 2023, 33(11): 41-49.

HE X X, WU J P, YU Y, et al. Improvement of SAR ship target detection algorithm in complex background[J]. Computer Technology and Development, 2023, 33(11): 41-49.

ZHOU Y S, LIU H C, MA F, et al. A sidelobe-aware small ship detection network for synthetic aperture radar imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61, 1-16.

BAI L, YAO C, YE Z, et al. Feature enhancement pyramid and shallow feature reconstruction network for SAR ship detection[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2023, 16: 1042-1056.

YANG Y G, JU Y W, ZHOU Z Y. A super lightweight and efficient SAR image ship detector[J]. IEEE Geoscience and Remote Sensing Letters, 2023, 20: 1-5.

WANG C Y, MARK LIAO H Y, WU Y H, et al. CSPNet: a new backbone that can enhance learning capability of CNN[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). Piscataway: IEEE Press, 2020: 1571-1580.

LIN T Y, DOLLÁR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE Press, 2017: 936-944.

LIU Y, SHAO Z, HOFFMANN N. Global attention mechanism: retain information to enhance channel-spatial interactions[J].arXiv preprint, 2021, arXiv:2112.05561.

HE K M, ZHANG X Y, REN S Q, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9): 1904-1916.

WANG Q L, WU B G, ZHU P F, et al. ECA-net: efficient channel attention for deep convolutional neural networks[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE Press, 2020: 11531-11539.

ZHANG T W, ZHANG X L, LI J W, et al. SAR ship detection dataset (SSDD): official release and comprehensive data analysis[J]. Remote Sensing, 2021, 13(18): 3690.

ZHANG T W, ZHANG X L, KE X, et al. LS-SSDD-v1.0: a deep learning dataset dedicated to small ship detection from large-scale sentinel-1 SAR images[J]. Remote Sensing, 2020, 12(18): 2997.

FU J M, SUN X, WANG Z R, et al. An anchor-free method based on feature balancing and refinement network for multiscale ship detection in SAR images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(2): 1331-1344.

ZHANG T W, ZHANG X L, KE X. Quad-FPN: a novel quad feature pyramid network for SAR ship detection[J]. Remote Sensing, 2021, 13(14): 2771.

JOCHER G, NISHIMURA K, MINEEVA T, et al. YOLOv5 (minor version 1.0)[EB].2022..

WANG C, BOCHKOVSKIY A, LIAO H. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[J].arXiv preprint, 2022, arXiv: 2207.02696.

DENG Y W, GUAN D H, CHEN Y Y, et al. Sar-shipnet: sar-ship detection neural network via bidirectional coordinate attention and multi-resolution feature fusion[C]//Proceedings of the ICASSP 2022—2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Piscataway: IEEE Press, 2022: 3973-3977.

YU J M, WU T, ZHOU S B, et al. An SAR ship object detection algorithm based on feature information efficient representation network[J]. Remote Sensing, 2022, 14(14): 3489.

GUO H Y, YANG X, WANG N N, et al. A CenterNet++ model for ship detection in SAR images[J]. Pattern Recognition, 2021, 112: 107787.

YU Y, YANG X, LI J, et al. A cascade rotated anchor-aided detector for ship detection in remote sensing images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 60: 5600514.

CUI Z Y, LI Q, CAO Z J, et al. Dense attention pyramid networks for multi-scale ship detection in SAR images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(11): 8983-8997.

YAO C, BAI L, XUE D L, et al. GFB-net: a global context-guided feature balance network for arbitrary-oriented SAR ship detection[C]//Proceedings of the 2022 7th International Conference on Image, Vision and Computing (ICIVC). Piscataway: IEEE Press, 2022: 166-171.

PANG J M, CHEN K, SHI J P, et al. Libra R-CNN: towards balanced learning for object detection[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE Press, 2019: 821-830.

TAN M X, PANG R M, LE Q V. EfficientDet: scalable and efficient object detection[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE Press, 2020: 10778-10787.

ZHANG X S, WAN F, LIU C, et al. Learning to match anchors for visual object detection[C]//Proceedings of the IEEE Transactions on Pattern Analysis and Machine Intelligence. Piscataway: IEEE Press, 2022: 3096-3109.

KONG T, SUN F C, LIU H P, et al. FoveaBox: beyound anchor-based object detection[J]. IEEE Transactions on Image Processing, 2020, 29: 7389-7398.

HU Q, HU S H, LIU S Q. BANet: a balance attention network for anchor-free ship detection in SAR images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1-12.

XU X W, ZHANG X L, ZHANG T W, et al. On-board ship detection in SAR images based on L-YOLO[C]//Proceedings of the 2022 IEEE Radar Conference (RadarConf22). Piscataway: IEEE Press, 2022: 1-5.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

融合多尺度和上下文的无人机救援小目标检测算法

基于CNN-BiGRU的复杂连续人体活动Wi-Fi感知方法

一种基于深度可分离卷积和注意力机制的入侵检测方法

基于AR虚实图像注意力机制的电缆装配质量检测方法

基于SqueezeNet卷积神经网络的车辆检测