基于CART决策树方法的遥感影像分类

Remote Sensing Image Classification Based on CART Decision Tree Method

以云南省香格里拉县为研究区域,构建一种基于CART遥感影像的决策树分类方法.对遥感影像采用主成分提取、植被信息提取、纹理信息提取等方法,并结合试验区主要地物类型训练样本,采用Landsat 5 TM影像数据、DEM数据以及遥感处理软件ENVI为平台进行影像分类,并将结果与最大似然分类结果作比较.结果表明,基于CART遥感影像决策树分类精度优于最大似然分类,有较好的分类效果.

Taking Shangri-La County,Yunnan Province as the study area,this paper built a decision tree classification method based on remote sensing images.And Regression Tree.Using the methods of principal component extraction,vegetation information extraction,texture information extraction,combined with the main feature type test area of training samples,and taking Landsat 5 TM image date,DEM date,software ENVI as platform,the remote sensing image classification has been done.The comparison results which with the maximum likelihood classification show that CART-based remote sensing image classification accuracy of decision tree is better than maximum likelihood classification,has a better effect of classification.