Abstract
“SPECTRAL AMPLIFICATION” is the significant name of a new algorithm of wavelength selection developed to improve the precision of the partial least squares (PLS) calibration of near infrared a (NIR) spectrometer for quantitative chemical analyses. This algorithm amplifies selectively some spectral data by mutiplicative coefficients so that they are predominant in the spectra and lower the prediction error of the PLS calibration. The poster presents a demonstration of “spectral amplification” in the determination of moisture on milk powders by NIR diffuse reflectance spectroscopy.
Get full access to this article
View all access options for this article.