Applied analysis for canopy nitrogen retrieval of winter wheat using hyperspectral vegetation index

摘要

Nitrogen is one of the most important nutrients in crop growth and development. To study the problem of spectral index setting of intelligent remote sensor for crop N prime inversion, and effectiveness of quantitative evaluation and other issues in different application requirements, with winter wheat for example to study the impact of quantitative model inversion of center wavelength, SNR and band width for different intelligent observation mode, analysis the sensitivity and effectiveness of spectral index for N inversion with the quantitative model. The results showed that: (1) when the center wavelength are 420,508 and 405nm, band width is 1nm, SNR>70DB, the MTCI_B is the best vegetation index; (2) using RVIinf_r and MTCI to build joint inversion model, the inversion result R2 = 0.9252, RMSE = 0.3678, better than the best single index of the inversion results; (3 ) the result of simulating HJ1A-HIS and Hyperion showed that joint inversion model has a certain degree of universality in different hyperspectral sensors.