东南亚地区红树林空间分布数据集

红树林是生长在热带、亚热带海岸潮间带或河流入海口，受到海水周期性浸没的木本植物群落。有着防风、防浪、促淤造陆、稳定海岸、保护海堤等的作用。红树林在遥感影像上具有明显的植被光谱特征，尤其在短波红外波段能有效地与其他植被类型进行区分。采用 2013-2015年期间共计428景30m空间分辨率的Landsat-8多光谱遥感数据及ASTER GDEM数据，结合GOOGLE EARTH高空间分辨率影像，采用多分类器组合方法对东南亚地区以及中国广西和雷州半岛的红树林进行了提取。其中选取的单分类器包括：支持向量机分类器、人工神经网络分类器、波谱角分类器、最大似然分类器、最小距离分类器。采用并行方式对各分类器输出信息进行组合，分类器的组合采用贝叶斯算法。经过部分地区的抽样检验。数据表明，研究区红树林提取面积为4.52万平方公里。本数据集包含.shp和.kmz两种格式。

Mangroves are woody plant communities that grow in tropical and subtropical coastal intertidal zones or river estuaries, periodically submerged by seawater. They play multiple roles including windbreak, wave dissipation, sediment accumulation and land formation, coastal stabilization, and seawall protection. Mangroves exhibit distinct vegetation spectral characteristics in remote sensing images, and can be effectively differentiated from other vegetation types, particularly in the short-wave infrared band. A total of 428 scenes of Landsat-8 multispectral remote sensing data with 30 m spatial resolution and ASTER GDEM data from 2013 to 2015, combined with high-spatial-resolution Google Earth imagery, were utilized to extract mangroves in Southeast Asia, as well as Guangxi and Leizhou Peninsula of China, via a multi-classifier combination approach. The selected individual classifiers are: Support Vector Machine (SVM) classifier, Artificial Neural Network (ANN) classifier, Spectral Angle Mapper (SAM) classifier, Maximum Likelihood Classifier (MLC), and Minimum Distance Classifier. The outputs of all classifiers were combined in a parallel manner, with the classifier integration adopting the Bayesian algorithm. Following sampling inspections in partial regions, the data indicates that the extracted mangrove area in the study area totals 45,200 square kilometers. This dataset includes two formats: .shp and .kmz.

搜集汇总

背景与挑战

背景概述

该数据集基于2013-2015年的Landsat-8和ASTER GDEM遥感数据，采用多分类器组合方法提取了东南亚地区及中国广西、雷州半岛的红树林分布，总面积达4.52万平方公里，提供.shp和.kmz两种格式。