Evaluating different vegetation index for estimating lai of winter wheat using hyperspectral remote sensing data

Leaf area index (LAI) is an important parameter which always be used to estimate vegetation cover and forecast the crop growth and yield. Currently the statistic relationship between LAI and vegetation indices (VI) has been widely applied to predict vegetation LAI. Each vegetation index for inversing LAI has applicable area and conditions, the best vegetation index or spectral parameter of them were not sure for estimating LAI of winter wheat in China. In this paper, PSR-3500 spectrometer and LAI-2200 plant canopy analyser were used to acquire the spectrum and LAI synchronously from April to June in 2013, in xiaotangshan of Beijing. After calculating VIs selected in this study, the correlation relationships between VIs and LAI were established under different spectral widths and center wavelengths. The results show that DVI is the best index with R² of 0.7. 3nm was verified the best bandwidth with center wavelength of NIR and red was 815nm and 746nm, respectively. The method that multiple VIs were used to inverse LAI synergistically, was proposed in this paper, which established the optimal linear regression model. Finally, the R² we got between prediction LAI and the measured value reached to 0.9235, which reaffirmed the feasibility of multiple VIs in the estimation of vegetation LAI.