Hyperspectral models for estimating vegetation chlorophyll content based on red edge parameter

Hyperspectral reflectance and green degree of Platanus orientalis L. and Populus tomentosa Carr. leaves were measured by the ASD portable spectrometer and the portable chlorophyll meter SPAD-502, respectively. The correlations between spectral reflectance, first derivative spectral reflectance and leaf green degree were analyzed. The hyperspectral models for estimating vegetation chlorophyll content based on red edge position, kurtosis and skewness were established, and the red edge position, kurtosis and skewness were used as the input variables of ANN-BP to estimate the content of chlorophyll. The results showed that red edge position of Platanus orientalis L. and Populus tomentosa Carr. had close relations with chlorophyll content, the regression determination coefficients were 0.7366 and 0.7289, respectively. The regression models established with kurtosis and skewness were obtained which could improve the estimating precision effectively, and the least determination coefficients were above 0.8341. The determination coefficients of ANN-BP models of Platanus orientalis L. and Populus tomentosa Carr. were 0.9574 and 0.9523, respectively. Compared with the models of single variable, ANN-BP model was a good hyperspectrum inversion model for estimating vegetation chlorophyll content, which could greatly improve estimation accuracy of vegetation chlorophyll content.