Abstract
A system was developed to map the yield of grass using hyperspectral imaging. First, the yield was estimated by fitting the spectrum of each pixel in the hyperspectral image to an estimation model of the yield. And then, the estimation yield map was created at a suitable spatial resolution calculated by semivariance analysis. The estimation models were developed using four analysis methods, and the PLSR (Partial Least Squares Regression analysis) model indicated good potential for the estimation of the yield. The model validation result showed that the coefficient of determination was 0.418, the standard error of cross validation was 0.077t/10a and the relative error was 0.218. The estimation map of the yield reflected spatial distribution of the actual yield relatively.
