The paper preprocesses Quickbird high-resolution sensing images based on ENVI and uses 4 kinds of vegetation indices,Normalized Difference Vegetation Index(NDVI),Normalized Differential Green Index(NDGI),Soil Adjusted Vegetation Index(SAVI),and Modified Soil Adjusted Vegetation Index(MSAVI)to derive green land from Gulou district in Nanjing.Tacking the images with the programs of Matlab software,the paper analyses their gray image,histogram,binary image and variance and gives the result through C language.The experiments show that in this study area,SAVI and NDVI could derive green land well;mistakes and omissions appeared in NDGI and MSAVI deriving;MSAVI shows superiority in edge extraction of buildings and vegetation.