The Effect of Spectral Pre-Treatments on the Partial Least Squares Modelling of Agricultural Products

Abstract

Spectral pre-treatment, such as scatter correction, smoothing and derivatisation is considered, to the point of being almost folklore, an integral component to the development of near infrared (NIR) partial least squares (PLS) regression equations. This study was undertaken to examine the importance of pre-treatments. Diffuse reflectance NIR (1100–2500 nm) spectra of ground wheat and forages were separately analysed. For ground wheat, the effect of spectral pre-treatment on the PLS equations for protein content and sodium dodecyl sulphate (SDS) sedimentation volume (a protein quality index) was examined. For forages, similar examinations were performed on crude protein content and lignin content. Results indicate that while pre-treatment is indeed important, statistical significance, as determined by the F-test of correlated variances, is often not established. Protein content calibrations tend to be enhanced by scatter correction, as opposed to smoothing or derivatisation, whereas the SDS sedimentation volume and lignin content calibrations favoured these convolution functions. It is recommended that the selection of the best pre-treatment for an analyte be based on the combination of statistical testing and the modeller's judgement.