Vegetation extraction in Taishan region based on high-resolution satellite remote sensing images

Abstract

When using satellite remote sensing images to identify vegetation areas, only using spectral information to extract information will cause the phenomenon of "same spectrum foreign matter". Aiming at the limitations of the above methods, this paper proposes a vegetation area identification method based on vegetation index and texture features. Firstly, the normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) are used to classify the vegetation areas in satellite images. Then, the Gray-level Co-occurrence Matrix (GLCM) of the image is calculated and the image texture feature information parameters are superimposed. Based on the initial division result, the boundary of the vegetation area is accurately identified. Compared with the method of using only spectral information, the extraction result of vegetation area in this paper obtains the research result with higher accuracy.