引用本文:[点击复制]
[点击复制]
【打印本页】 【在线阅读全文】【下载PDF全文】 查看/发表评论下载PDF阅读器关闭

←前一篇|后一篇→

过刊浏览    高级检索

本文已被:浏览 141次   下载 213 本文二维码信息
码上扫一扫!
基于谷歌地球引擎平台的海上养殖信息提取方法研究——以福建省平潭县为例
闫锦崴,郑蔚恒,于鹏
0
(厦门理工学院计算机与信息工程学院, 福建 厦门 361024;自然资源部空间海洋遥感与应用重点实验室, 北京 100081)
摘要:
海上养殖业对粮食安全有着至关重要的作用。然而,海上养殖的无序扩张和开发,阻碍了海上交通,同时也造成了海洋环境问题。为及时、准确地获取海上养殖信息,满足海岸带调查以及推进海上养殖规范化、科学化,提出一种基于谷歌地球引擎(Google Earth Engine, GEE)平台实现长时间序列下海上养殖区信息快速提取的方法。本研究构建了一个基于随机森林分类的海上养殖区信息提取模型,该模型综合利用了Sentinel-1卫星SAR影像数据的VV和VH极化波段,以及Sentinel2卫星的多光谱影像数据。此外,模型还融合了4个用于增强养殖区特征的指数,以提高养殖区域信息提取的准确性和效率。这种方法的应用旨在优化海上养殖区的识别过程,通过精确分析和利用不同数据源的互补优势,展现了-遥感技术在海洋养殖监测领域的巨大潜力。本研究对2017—2021年平潭县海上养殖区域进行判定与提取,实验结果表明,以养殖密度较低,养殖特征不明显为特征的海上养殖区,基于GEE平台的海上养殖区信息提取方法精度在90%以上,表明在复杂水体背景下对养殖区快速识别取得较好的效果,可为海上养殖科学规划与规范化管理提供有效的参考依据。
关键词:  海洋物理学  谷歌地球引擎  海上养殖区提取  Sentinel-1/2  随机森林分类  平潭县
DOI:10.3969/J.ISSN.2095-4972.20230612001
基金项目:福建省自然科学基金(2021J05259);厦门理工学院高层次人才项目(YKJ21009R)
Extraction of coastal aquaculture data based on Google Earth Engine with case study in Pingtan County, Fujian Province
YAN Jinwei,ZHENG Weiheng,YU Peng
(College of Computer and Information Engineering, Xiamen University of Technology, Xiamen 361024, China;Key Laboratory of Space Ocean Remote Sensing and Application, MNR, Beijing 100081, China)
Abstract:
Coastal aquaculture is an important part of marine economic development and plays a vital role in food security. However, the disorderly expansion and development of coastal aquaculture has hindered maritime traffic and also caused marine environmental problems. In order to obtain coastal aquaculture data timely and accurately for the coastal zone survey and for the standardization and reasonable coastal aquaculture, this paper proposes a method based on Google Earth Engine (GEE) platform for a rapid information extraction of coastal aquaculture area in long time series. We used Sentinel-1 and Sentinel-2 satellite image data to construct an extraction model based on random forest classification, determined and extracted the coastal culture area of Pingtan county from 2017 to 2021. Experimental results show that in Pingtan County, where the aquaculture density is low and the aquaculture characteristics are not prominent, the extraction method based on GEE platform utilized in this study exhibits an accuracy rate exceeding 90%. This indicates that rapid identification of aquaculture areas amidst complex water bodies has yielded favorable outcomes, thereby offering valuable insights for scientific planning and standardized management of coastal aquaculture.
Key words:  marine physics  Google Earth Engine  coastal aquaculture region extraction  Sentinel-1/2  random forest classification  Pingtan Country

用微信扫一扫

用微信扫一扫