Application of wavelet transform on hyperspectral reflectance for soybean lai estimation in the songnen plain, China

Abstract

In this study, we present spectral measurements of soybean LAI and their estimation from reflectance spectra data in Songnen Plain. Soybean canopy reflectance and its derivative were subsequently used in a linear regression analysis against LAI on one by one spectral reflectance. It was found that determination coefficient for LAI was high in blue, red and near infrared spectral region, and it was low in green spectral region, however LAI obtained its high determination coefficient in blue, green and red edge spectral region, especially in red edge region. Regression models were established based upon spectral vegetation indices and wavelet energy coefficient. It was found that wavelet transforms is an effective method for hyperspectral reflectance variables extraction to retrieve LAI, and the best multivariable regressions R2 up to 0.90 for LAI. Further studies are still needed to refine the methods for determining and estimating corn bio-physical/chemical parameters or other physiological parameters of different vegetation as well in the future.