以沙漠化问题较突出的辽宁省北部地区为例,选取2007年Landsat 5TM 遥感影像作为基本数据源,通过对影像中耕地、林地、草地、水域等常见地物及典型沙漠化土地进行光谱特征分析和波段间的相互运算,将修改型土壤调整植被指数(MSAVI)、归一化差异水体指数(NDWI)和遥感图像缨帽变换后的土壤亮度指数(SBI)、绿度植被指数(GVI)及湿度指数(WVI)等特征变量融入决策树分类模型后进行分层分离,从而实现对沙漠化信息的高精度提取.结果显示,决策树分类法可排除提取地物时的干扰信息,是保证沙漠化土地信息快速自动提取的方法之一.%To a more prominent problem of desertification in northern Liaoning Province for example, selecting the Landsat 5 TM remote sensing image data in 2007, it carries on the analysis of spectral characteristics and the mutual operation of the band through the images of the cultivated lands,the forest land, the lawn, the waters and other common surface features as well as the typical desertified land. After merging the MSAVI, NDWI as well as the SBI、 GVI、 WVI and other characteristic variables of the remote sensing image by K-T transform into the decision tree classification model,it separated hierarchically in order to achieve the high-precision extraction of desertification information. The results show that decision tree classification which is the means of the fast automatic extraction of the desertified land can eliminate the interference information when extracting surface features.
展开▼
