Analysis of Vegetation and Atmospheric Correction Indices for Landsat Images

Vegetation and Atmospheric Indices are mathematical combinations of remote sensing bands which are useful in distinguishing the various values of the spectral reflectance. In this paper we study how the applications of various atmospherically corrected indices and vegetation indices can aide in retrieving the amount of surface reflectance from a remotely sensed image. Specifically, this paper studies and compares three vegetation indices and one atmospherically resistant index. These indices include the Normalized Difference Vegetation Index (NDVI), the Soil Adjusted Vegetation Index (SAVI), the Green Vegetation Index (GVI), and the Atmospherically Resistant Vegetation Index (ARVI), respectively. The algorithms attempt to estimate the optical characteristics of Thematic Mapper (TM) imagery. It will be shown that the NDVI algorithm followed by the ARVI correcting algorithm provided significant improvements in the tonal qualities of the retrieved images. The results are presented on 1987 TM images over the Kennedy Space Center (KSC) and are compared with a set of United States Geological Survey (U.S.G.S) maps.