利用决策树模型的湿地提取与分类

发布时间：2018-01-14 16:23

本文关键词：利用决策树模型的湿地提取与分类　出处：《辽宁工程技术大学学报(自然科学版)》2016年05期 　论文类型：期刊论文

【摘要】：为解决因湿地环境复杂且类型多样导致光谱混淆而难以对其自动遥感提取的问题,采用决策树模型的湿地信息提取方法,以Landsat OLI影像光谱特征和经缨帽变换后的数据为基础,结合不同类型湿地的环境特征和空间特征信息,提出先分区再分类的思想,构建决策树分类模型.对原始影像进行缨帽变换,利用变换后的湿度分量及地物的光谱特征规律,将研究区划分为水体区域、植被区域和非植被区域,然后分别对各个区域进行再分类,逐层分级,最终实现不同类型湿地的分级提取.研究结果表明:采用分区分类思想构建决策树模型,可以有效提取湿地信息,精度达87.50%.
[Abstract]:In order to solve the problem of complex and diverse types of wetland environment lead to the confusion of spectral remote sensing automatic extraction, extraction methods of wetland information decision tree model, with the characteristics of the image spectrum of Landsat OLI and the data after tasseled cap transformation as the basis, combined with different types of environmental features and spatial features of wetland, put forward first division the classification idea, constructing decision tree classification model. The tasseled cap transformation to the original image using spectral features and terrain humidity component after transformation, the study area is divided into water area, vegetation and non vegetation area, then the area classification, layer classification, finally realize the classification of different types the extraction of wetland. The results show that the classification thought of constructing decision tree model can effectively extract wetland information, accuracy of 87.50%.

【作者单位】：辽宁工程技术大学测绘与地理科学学院;辽宁省基础地理信息中心;
【分类号】：P237
【正文快照】： 0引言湿地作为一种人类赖以生存的特殊生态系统,拥有巨大的环境和经济功能,素有“地球之肾”的美誉[1].近30年来,受社会经济和自然环境不断发展的影响,越来越多的湿地生存环境正在遭受工业污染的威胁,面积持续萎缩,生态功能逐渐退化[2-3],因此,在人类发展经济的同时,湿地保护

【相似文献】