A training method for an adaptive evaluation algorithm, a hyperspectral measuring device, as well as a device for dispensing of an operating medium

摘要

The present invention relates to a training method for an adaptive evaluation algorithm, in particular for an artificial neural network, for the determination of concentrations (20) in at least one plant (25) on the basis of said hyperspectral image data (17) of the at least one plant (25), wherein parameters of the adaptive evaluation algorithm on the basis of a plurality of training cases (23), to be calibrated, wherein each of these training cases (23) is given by an input vector (11), of the image data (17) of a said hyperspectral image (1) a to the training case (23) associated with the test plant (2), and an output vector (21), which by means of a chemical analysis (19) certain concentrations (20) within these to the training case (23) associated with the test plant (2) is given, whereby each individual one of the training cases (23) is produced by first the material to the training case is associated with the test plant (23) into a predetermined physiological state is guided.The present invention also relates to a hyperspectral measuring device (24) with an evaluation unit (18) and a device (27, 31, 36) for dispensing of an operating medium comprising such a measuring device.