基于机器学习的内孤立波遥感探测研究进展综述
DOI:
作者:
作者单位:

自然资源部第一海洋研究所

作者简介:

通讯作者:

中图分类号:

基金项目:

国家自然科学“低空远距离条件下海态与目标一体化雷达探测理论与方法”(U2006207);“基于实验和立体观测的内孤立波光学遥感图像不可视条件参数阈值确定”(42006164)。


A review of research progress on remote sensing detection of internal solitary waves based on machine learning.
Author:
Affiliation:

FirstInstitute of Oceanography,Ministry of Natural Resources

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    海洋内孤立波是一种发生在稳定层化海水内部的波动,广泛分布于全球的各个海洋,在物质能量传输、海洋环流中发挥重要作用,对海洋工程建设、舰船航行安全等也有重要的影响。对海洋中内孤立波进行准确的检测和识别,对于增进我们对物理海洋的了解,以及维护海洋环境的安全具有重要的研究价值。传统内孤立波的检测识别方法主要应用小波分析、边缘检测等图像处理技术,需要人工干预,很容易受到人为判别的干扰。随着人工智能的发展,机器学习方法在检测和识别内孤立波方面显示出了相当的优越性,尤其在准确性和自动化方面。除此之外,机器学习方法在内孤立波的参数反演和传播预测等方向的研究也有一定的发展。

    Abstract:

    Oceanic internal solitary waves are a type of wave motion that occurs within the stable stratified layers of seawater, widely distributed in various oceans worldwide. They play a crucial role in the transfer of material energy and ocean circulation, and have significant impacts on marine engineering construction and the safety of ship navigation. Accurate detection and identification of internal solitary waves in the ocean are of great research value for enhancing our understanding of physical oceanography and maintaining the safety of the marine environment. Traditional methods for detecting and identifying internal solitary waves mainly rely on image processing techniques such as wavelet analysis and edge detection, which require manual intervention and are easily influenced by human judgment. With the development of artificial intelligence, machine learning methods have shown significant superiority in the detection and identification of internal solitary waves, especially in terms of accuracy and automation. In addition, machine learning methods have also made certain progress in research on parameter inversion and propagation prediction of internal solitary waves.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2023-11-20
  • 最后修改日期:2023-12-12
  • 录用日期:2023-12-15
  • 在线发布日期:
  • 出版日期: