全球10m分辨率红树林（2020）

利用遥感大数据和Google Earth Engine（GEE）云平台，集成影像最大值合成算法（MSIC）以及面向对象机器学习算法（OBRF），提出了一种高效、高精度、高鲁棒性的红树林提取方法体系，构建了全球首套高空间分辨率（10米）的全球红树林分布数据集并命名为HGMF_2020。该方法体系的基本思路：利用现有的红树林数据集，确定全球红树林具体的分布范围；应用影像最大值合成算法在GEE平台上合成全年最低潮影像；结合面向对象分割和随机森林算法，提取2020年全球红树林。与此前公布的全球红树林数据相比，HGMF_2020提供了最新的、分辨率更高的全球红树林的空间分布以及斑块结构信息，总体制图精度达到95%以上，被精细刻画的红树林斑块包含具有地理学意义的属性信息，可直接用于后续的研究和分析。

Leveraging remote sensing big data and the Google Earth Engine (GEE) cloud platform, and integrating the Image Maximum Value Synthetic Composite (MSIC) algorithm and the object-based machine learning algorithm (OBRF), this study proposes an efficient, high-precision, and highly robust mangrove extraction methodology, and constructs the world's first global mangrove distribution dataset with 10-meter high spatial resolution, named HGMF_2020. The core workflow of this methodology is as follows: first, use existing mangrove datasets to define the precise global distribution range of mangroves; second, apply the MSIC algorithm on the GEE platform to synthesize annual lowest-tide composite images; third, combine object-oriented segmentation and random forest algorithms to extract global mangrove distributions for the year 2020. Compared with previously published global mangrove datasets, HGMF_2020 provides the latest, higher-resolution spatial distribution and patch structure information of global mangroves, with an overall mapping accuracy exceeding 95%, and the finely delineated mangrove patches carry geographically meaningful attribute information that can be directly used for subsequent research and analysis.

搜集汇总

背景与挑战

背景概述

该数据集是全球首套10米高空间分辨率的红树林分布数据，覆盖2020年全球范围，通过遥感大数据和先进算法提取，总体制图精度超过95%。它提供了精细的红树林斑块结构和属性信息，适用于生态保护和地理研究，并以开放获取方式共享。