基于深度学习的星载SAR海洋内波自动检测研究
作者:
作者单位:

1.中国人民解放军32021部队;2.国防科技大学气象海洋学院;3.航天宏图(上海)空间遥感技术有限公司;4.中国人民解放军61741部队

基金项目:

国家自然科学基金(No. 41876201)与上海市产业协同创新(科技)项目(No.XTCX-KJ-2022-20)联合资助


Oceanic Internal Waves Automatic Detection from SAR Images based on Deep Learning
Author:
Affiliation:

1.Army 32021, PLA;2.College of Meteorology and Oceanography, National University of Defense Technology;3.Piesat Space Remote Sensing Technology(Shanghai)Co., Ltd, Shanghai;4.Army 61741, PLA

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [19]
  • | | | |
  • 文章评论
    摘要:

    海洋内波是由海水密度陡变和外力扰动所引起的一种海洋内部的波动,在合成孔径雷达(Synthetic Aperture Radar,SAR)遥感图像上通常呈现出亮暗相间的条纹特征。本文基于2014年至2021年的390景Sentinel-1 SAR海洋内波遥感图像建立训练与验证数据集,结合旋转目标检测算法,使用迁移学习方法对模型进行训练,得到基于旋转框的海洋内波自动检测模型,并将检测结果与YOLOv8的检测结果进行对比。研究结果表明,旋转目标检测模型较YOLOv8能够取得更为优异的海洋内波自动检测结果,其精确率达到93.06%,召回率为90.24%,在精确率较高同时还能保证较低虚警。旋转目标检测模型为海量星载SAR海洋内波自动快速检测提供了一种新手段,该方法在检测海洋内波的同时还可提取内波传播方向信息,为针对性开展海洋内波动力学参数反演和过程研究提供了技术基础。

    Abstract:

    Oceanic internal waves are a kind of seawater fluctuation caused by the steep change of seawater density and external disturbance, which are usually shown as bright and dark stripes on Synthetic Aperture Radar (SAR) remote sensing images. Based on a training and validation dataset constructed by 390 Sentinel-1 SAR ocean internal wave remote sensing images from 2014 to 2021, combining the algorithm of Rotation Equivariant Detector (ReDet), this paper trains the model with the transfer learning method and obtained an automatic detection model for ocean internal waves based on rotating box, whose detection results are compared with those from YOLOv8 model. The results show that the rotating target detection model performs better than YOLOv8 in automatic detection of internal waves, which yields an accuracy of 93.06% with a recall rate of 90.24% and achieves a high accuracy and a low false alarm at the same time. The rotating target detection model provides an innovative technical solution for the automatic and rapid detection of ocean internal waves among space borne SAR images of massive quantity, and the method can extract the propagation direction information of internal waves while detecting the internal waves, which provides a solid technical basis for the parameter inversion of internal wave dynamics and further researches on process of internal waves.

    参考文献
    [1] Da Silva J C B, Ermakov S A, Robinson I S, et al. Role of surface films in ERS SAR signatures of internal waves on the shelf. 1. Short-period internal waves [J]. Journal of Geophysical Research, 1998, 103(C4): 8009-8031.
    [2] 林珲, 范开国, 申珲, 等.星载SAR海洋内波遥感研究进展[J]. 地球物理学进展, 2010, 25(3): 1081-1091.
    [3] 范开国, 陈鹏, 顾艳镇, 等.星载合成孔径雷达海洋遥感与图像解译[M]. 北京:海洋出版社, 2017, 60-80.
    [4] 范开国, 周晓中, 徐青, 等. 合成孔径雷达海洋内波遥感探测技术与应用[M]. 北京:海洋出版社, 2017, 75-101.
    [5] Rodenas A and Garello R. Internal wave detection and location in SAR images using wavelet transform [J]. IEEE Transactions on Geoscience & Remote Sensing, 1998, 36(5): 1494-1507.
    [6] Rodenas J A, Cabarrocas D, Garello R. Internal waves detection from ERS-1 SAR images using the 2-D wavelet transform [C]. Geoscience & Remote Sensing, IGARSS 97 Remote Sensing-a Scientific Vision for Sustainable Development, 1997.
    [7] Kang J, Zhang J, Song P J, et al. The Application of Two-Dimensional EMD to Extracting Internal Waves in SAR Images [C]. Computer Science and Software Engineering. 2008, CSSE 2008 International Conference.
    [8] 范开国, 梁建军, 韦道明, 等. EMD端点延拓新方法SAR遥感图像实例应用[J]. 地球物理学进展, 2021, 36(2): 458-463.
    [9] 丁灿, 张本涛, 高国兴, 等.合成孔径雷达图像中海洋内波的特征检测[J]. 海洋测绘, 2012, 32(5):15-18.
    [10] 陈捷, 陈标, 陶荣华, 等. SAR图像海洋内波参数自动提取方法[J]. 海洋技术学报, 2014, 33(06): 20-27.
    [11] Surampudi S and Sasanka S. Internal Wave Detection and Characterization with SAR data [C]. 2019 IEEE Recent Advances in Geoscience and Remote Sensing: Technologies, Standards and Applications. 2019.
    [12] 郑应刚, 张洪生, 李晓恋, 等.基于遥感影像边界特性对南海海洋内波检测算法研究[J].热带海洋学报, 2020, 6: 41-56.
    [13] 孟俊敏, 龙瑞, 孙丽娜, 等.基于机器学习的内孤立波遥感探测研究综述[J].数字海洋与水下攻防, 2024, 7(1): 2-8.
    [14] Bao S D, Meng J M, Sun L N, et al. Detection of ocean internal waves based on Faster R-CNN in SAR images [J]. Journal of Oceanology and Limnology, 2020, 38(01): 57-65.
    [15] Li X F,Liu B,Zheng G,et al. Deep-learning-based information mining from ocean remote-sensing imagery[J]. National Science Review,2020, 7(10):1584-1605.
    [16] Celona S, Merrifield S T, Paolo T, et al. Automated detection,classification,and tracking of internal wave signatures using X-band radar in the inner shelf[J]. Journal of Atmospheric and Oceanic Technology, 2021, 38(4):789-803.
    [17] 孙宏亮, 王怡然, 贾童, 等.基于Faster R-CNN的卫星SAR图像南海海洋内波自动检测[J]. 遥感学报, 2023, 27(4): 905-918.
    [18] Tao M K, Xu C J, Guo L X, et al. An Internal Waves Data Set From Sentinel-1 Synthetic Aperture Radar Imagery and Preliminary Detection [J]. Earth and Space Science, 2022, 9: 1-14.
    [19] Han J M, Ding J, Xue N, et al. ReDet: A Rotation-equivariant Detector for Aerial Object Detection [J]. 2021, DOI:10.48550/arXiv.2103.07733.
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文
分享
文章指标
  • 点击次数:45
  • 下载次数: 0
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 收稿日期:2024-07-16
  • 最后修改日期:2024-08-31
  • 录用日期:2024-09-04
文章二维码