Comparing Different Methods for Wheat LAI Inversion Based on Hyperspectral Data

Gaussian process regression (GPR) can effectively solve the problem of high-dimensional modeling with a small sample size. However, there is lack studies comparing GPR other methods for leaf area index (LAI) inversion using hyperspectral data. In this study, winter wheat was used as research material to evaluate performance different LAI inversion, i.e., GPR, an artificial neural network (ANN), partial least squares (PLSR) and spectral (SI). To end, 2-year water nitrogen coupled experiment conducted, canopy data were measured at critical growth stages wheat. Based on these data, calibration validation datasets obtained, prediction model constructed above four validated. The results showed that models SI effective compared methods, R2 RMSE ranging from 0.42–0.76 0.80–1.04 during 0.37–0.55 0.94–1.09 validation. ANN had best results, 0.89 0.85 0.46 0.53 0.74 0.71 both PLSR intermediate values 0.80 0.61 0.67 Thus, recommended