CLASSIFICATION OF NEAR-INFRARED SPECTRA USING WAVELENGTH DISTANCES - COMPARISON TO THE MAHALANOBIS DISTANCE AND RESIDUAL VARIANCE METHODS

A simple and easy to understand method for classification of near-infr ared spectra is reported. The method uses a sample's normalized distan ce from a library of mean spectra, The probability distribution of the test is described, and its ability to discriminate between similar ma terials was tested and is reported, Its ability to detect samples that fail to meet product specifications and samples adulterated with mino r levels of impurities was also tested and is reported. The performanc e of the method is compared to methods based on principal component an alysis, Mahalanobis distances, and SIMCA residual variance distances. Overall, the wavelength distance method gave better classification res ults than the Mahalanobis and SIMCA methods when small training sets w ere used, but poor results were obtained in the detection of samples t hat do not meet product specifications and samples adulterated with lo w levels of contamination.