• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利:CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验

基于视觉注意力增强网络的光学遥感图像船舶检测

Ship Detection for Optical Remote Sensing Images Based on Visual Attention Enhanced Network.

作者信息

Bi Fukun, Hou Jinyuan, Chen Liang, Yang Zhihua, Wang Yanping

机构信息

School of Information Science and Technology, North China University of Technology, Beijing 100144, China.

School of Information and Electronics, Beijing Institute of Technology, Beijing 100081, China.

出版信息

Sensors (Basel). 2019 May 16;19(10):2271. doi: 10.3390/s19102271.

DOI:10.3390/s19102271
PMID:31100909
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC6567313/
Abstract

Ship detection plays a significant role in military and civil fields. Although some state-of-the-art detection methods, based on convolutional neural networks (CNN) have certain advantages, they still cannot solve the challenge well, including the large size of images, complex scene structure, a large amount of false alarm interference, and inshore ships. This paper proposes a ship detection method from optical remote sensing images, based on visual attention enhanced network. To effectively reduce false alarm in non-ship area and improve the detection efficiency from remote sensing images, we developed a light-weight local candidate scene network( L 2 CSN) to extract the local candidate scenes with ships. Then, for the selected local candidate scenes, we propose a ship detection method, based on the visual attention DSOD(VA-DSOD). Here, to enhance the detection performance and positioning accuracy of inshore ships, we both extract semantic features, based on DSOD and embed a visual attention enhanced network in DSOD to extract the visual features. We test the detection method on a large number of typical remote sensing datasets, which consist of Google Earth images and GaoFen-2 images. We regard the state-of-the-art method [sliding window DSOD (SW+DSOD)] as a baseline, which achieves the average precision (AP) of 82.33%. The AP of the proposed method increases by 7.53%. The detection and location performance of our proposed method outperforms the baseline in complex remote sensing scenes.

摘要

舰船检测在军事和民用领域都发挥着重要作用。尽管一些基于卷积神经网络(CNN)的先进检测方法具有一定优势,但它们仍无法很好地应对挑战,包括图像尺寸大、场景结构复杂、大量虚警干扰以及近岸舰船等问题。本文提出了一种基于视觉注意力增强网络的光学遥感图像舰船检测方法。为有效减少非舰船区域的虚警并提高遥感图像的检测效率,我们开发了一种轻量级局部候选场景网络(L2CSN)来提取含有舰船的局部候选场景。然后,对于选定的局部候选场景,我们提出了一种基于视觉注意力DSOD(VA-DSOD)的舰船检测方法。在此,为提高近岸舰船的检测性能和定位精度,我们既基于DSOD提取语义特征,又在DSOD中嵌入视觉注意力增强网络来提取视觉特征。我们在大量由谷歌地球图像和高分二号图像组成的典型遥感数据集上测试了该检测方法。我们将先进方法[滑动窗口DSOD(SW+DSOD)]作为基线,其平均精度(AP)为82.33%。所提方法的AP提高了7.53%。在复杂遥感场景中,我们所提方法的检测和定位性能优于基线。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/067cf1c76bd6/sensors-19-02271-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/76090c917591/sensors-19-02271-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/f2d4460b14c9/sensors-19-02271-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/7ba1bf9cdd5f/sensors-19-02271-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/03869669e5b8/sensors-19-02271-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/06f3a3cc3881/sensors-19-02271-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/b2340159d50b/sensors-19-02271-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/f530a6a3361e/sensors-19-02271-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/1fa90300df1e/sensors-19-02271-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/c95138fc507c/sensors-19-02271-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/067cf1c76bd6/sensors-19-02271-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/76090c917591/sensors-19-02271-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/f2d4460b14c9/sensors-19-02271-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/7ba1bf9cdd5f/sensors-19-02271-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/03869669e5b8/sensors-19-02271-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/06f3a3cc3881/sensors-19-02271-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/b2340159d50b/sensors-19-02271-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/f530a6a3361e/sensors-19-02271-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/1fa90300df1e/sensors-19-02271-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/c95138fc507c/sensors-19-02271-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/6ee2/6567313/067cf1c76bd6/sensors-19-02271-g010.jpg

相似文献

1
Ship Detection for Optical Remote Sensing Images Based on Visual Attention Enhanced Network.基于视觉注意力增强网络的光学遥感图像船舶检测
Sensors (Basel). 2019 May 16;19(10):2271. doi: 10.3390/s19102271.
2
Locally Oriented Scene Complexity Analysis Real-Time Ocean Ship Detection from Optical Remote Sensing Images.基于局部特征的场景复杂度分析实时光学遥感图像海面舰船检测
Sensors (Basel). 2018 Nov 6;18(11):3799. doi: 10.3390/s18113799.
3
A Decision Mixture Model-Based Method for Inshore Ship Detection Using High-Resolution Remote Sensing Images.一种基于决策混合模型的高分辨率遥感影像近岸船舶检测方法。
Sensors (Basel). 2017 Jun 22;17(7):1470. doi: 10.3390/s17071470.
4
R-CenterNet+: Anchor-Free Detector for Ship Detection in SAR Images.R-CenterNet+:一种用于 SAR 图像中船舶检测的无锚探测器。
Sensors (Basel). 2021 Aug 24;21(17):5693. doi: 10.3390/s21175693.
5
Research on Airplane and Ship Detection of Aerial Remote Sensing Images Based on Convolutional Neural Network.基于卷积神经网络的航空遥感图像飞机和船舶检测研究。
Sensors (Basel). 2020 Aug 20;20(17):4696. doi: 10.3390/s20174696.
6
Incorporating Negative Sample Training for Ship Detection Based on Deep Learning.基于深度学习的船舶检测中的负样本训练。
Sensors (Basel). 2019 Feb 7;19(3):684. doi: 10.3390/s19030684.
7
A Moving Ship Detection and Tracking Method Based on Optical Remote Sensing Images from the Geostationary Satellite.一种基于地球静止卫星光学遥感图像的运动船舶检测与跟踪方法
Sensors (Basel). 2021 Nov 13;21(22):7547. doi: 10.3390/s21227547.
8
An Efficient and Lightweight Convolutional Neural Network for Remote Sensing Image Scene Classification.一种用于遥感图像场景分类的高效轻量级卷积神经网络。
Sensors (Basel). 2020 Apr 2;20(7):1999. doi: 10.3390/s20071999.
9
Ship Target Automatic Detection Based on Hypercomplex Flourier Transform Saliency Model in High Spatial Resolution Remote-Sensing Images.基于高空间分辨率遥感图像超复数傅里叶变换显著模型的船舶目标自动检测。
Sensors (Basel). 2020 Apr 29;20(9):2536. doi: 10.3390/s20092536.
10
A Novel Detector Based on Convolution Neural Networks for Multiscale SAR Ship Detection in Complex Background.基于卷积神经网络的复杂背景下多尺度 SAR 舰船检测新方法
Sensors (Basel). 2020 Apr 30;20(9):2547. doi: 10.3390/s20092547.

引用本文的文献

1
Recognition and Classification of Ship Images Based on SMS-PCNN Model.基于SMS-PCNN模型的船舶图像识别与分类
Front Neurorobot. 2022 Jun 13;16:889308. doi: 10.3389/fnbot.2022.889308. eCollection 2022.
2
CAA-YOLO: Combined-Attention-Augmented YOLO for Infrared Ocean Ships Detection.CAA-YOLO:用于红外海洋船舶检测的联合注意增强 YOLO。
Sensors (Basel). 2022 May 16;22(10):3782. doi: 10.3390/s22103782.
3
A Moving Ship Detection and Tracking Method Based on Optical Remote Sensing Images from the Geostationary Satellite.

本文引用的文献

1
TextBoxes++: A Single-Shot Oriented Scene Text Detector.TextBoxes++:一种单阶段的面向场景的文本检测器。
IEEE Trans Image Process. 2018 Aug;27(8):3676-3690. doi: 10.1109/TIP.2018.2825107. Epub 2018 Apr 9.
2
A Decision Mixture Model-Based Method for Inshore Ship Detection Using High-Resolution Remote Sensing Images.一种基于决策混合模型的高分辨率遥感影像近岸船舶检测方法。
Sensors (Basel). 2017 Jun 22;17(7):1470. doi: 10.3390/s17071470.
3
Automatic Subspace Learning via Principal Coefficients Embedding.基于主系数嵌入的自动子空间学习。
一种基于地球静止卫星光学遥感图像的运动船舶检测与跟踪方法
Sensors (Basel). 2021 Nov 13;21(22):7547. doi: 10.3390/s21227547.
IEEE Trans Cybern. 2017 Nov;47(11):3583-3596. doi: 10.1109/TCYB.2016.2572306. Epub 2016 Jun 9.
4
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks.更快的 R-CNN:基于区域建议网络的实时目标检测。
IEEE Trans Pattern Anal Mach Intell. 2017 Jun;39(6):1137-1149. doi: 10.1109/TPAMI.2016.2577031. Epub 2016 Jun 6.
5
Deep cognitive imaging systems enable estimation of continental-scale fire incidence from climate data.深度认知成像系统能够根据气候数据估算大陆尺度的火灾发生率。
Sci Rep. 2013 Nov 13;3:3188. doi: 10.1038/srep03188.