Chlorophyll content inversion with hyperspectral technology for wheat canopy based on support vector regression algorithm

In order to provide scientific basis for wheat growth monitoring and agronomic decision-making, the wheat canopy chlorophyll content was estimated by using hyperspectral technology in this paper. Eighteen kinds of hyperspectral indices were comparative analyzed. The index REP, which could respond wheat canopy chlorophyll content sensitively, was selected. The inversion model of wheat canopy chlorophyll content was then built by using the field spectra as the training samples and the least squares support vector regression (LS-SVR) algorithm as the modeling method, with the calibration R-square and prediction R-square 0.751 and 0.722, respectively, indicating the accuracy of estimation predicted by REP was highest in all indices. Further more, the prediction accuracy of REP was least sensitive to the change of chlorophyll content and LAI values among 18 indices and therefore least affected by the range of sample values and canopy density when used to estimate the chlorophyll content of wheat canopy. Using the inversion model, the remote sensing mapping for OMIS image was accomplished. The inversion and measured values were then compared by the method of regression fitting. The R-square and RMSE of the fitting model was 0.676 and 1.715, respectively, indicating the similarity between the inversion value and measured value was high. The result showed that it was feasible to estimate chlorophyll content accurately by using hyperspectral index REP to build a LS-SVR inversion mode. Therefore, this method proposed can be used as a rapid and non-destructive method for getting wheat chlorophyll content.