Inversion and validation of leaf area index based on CBERDS02B image data in GuangXi province of China

Abstract

CBERDS02B satellite has been successfully launched in September 2007, the target of this paper is to get the vegetation index from visible red-band, near-infrared band and the blue-band surface reflectance data of CBERDS02B satellite, through the empirical model of the relations between the vegetation index and LAI, and combined with the classification data to integrate the appropriate model, in order to get the regional leaf area index image in Binyang County of Nanning City in Guangxi Province of China. To make the operation more rapid and feasible, I decided to use an empirical model to obtain LAI, This method is simple and easy to calculate, more realizable, and suitable for remote sensing application. In this paper I use part of the measured data to validate a wide range of VI-LAI models. In order to identify the advantages and disadvantages of the various models, different plants use different types of vegetation model, I finally choose four VIs, such as SR, NDVI, SAVI, EVI, then combine these with the classification data to get the best mixed model so as to attain the leaf area index image of the research region. Then I use the other part of the measured data to get the validation of the mixed model. Ultimately I improve the overall accuracy of the model, and gain more accurate LAI images in the region.