TRANSFER OF CALIBRATION FUNCTION IN NEAR-INFRARED SPECTROSCOPY

A procedure for the transfer of the regression equation in near-infrar ed spectroscopy (NIRS), from a first instrument to a second instrument , is presented. The procedure uses partial least squares (PLS) regress ion twice: in the first step to compute the relationship between the s pectra of transfer samples of the two instruments, and in the second s tep to compute the regression equation (relationship between chemical variables and spectral variables) of the first instrument. These two P LS steps are combined to predict the regression equation of the second instrument. Sometimes the PLS relationship between the two instrument s is obtained from the principal components of the spectra of the two instruments. The procedure is applied to a set of 60 samples of soy fl our, representative of the Italian soy production. 40 samples were use d both as transfer samples and to compute the regression equation. 20 samples were used as evaluation set. Spectra were recorded with four d ifferent instruments, in four different laboratories. The result of th e transfer procedure were evaluated by means of the standard error of prediction (SEP) with the predicted regression equation. Owing also to the great number of samples in the transfer set, and to the noise fil tering effect of the twin PLS procedure, SEP with the predicted regres sion equation is not greater than that with the regression equation co mputed directly from the second instrument. The effect of some paramet ers, such as the number of PLS latent variables in the two steps, is a lso studied.