The aim of this study was to develop a rapid method for determining geographical origin of black pepper based on combination of near-infrared spectroscopy and chemometric algorithms. Near-infrared spectroscopy was used to determine the geographical origin of 111 black pepper samples from three regions in Vietnam: North Central Region, Central Highlands and Southeast Region. A portable NIR spectrometer was used to collect the spectra of intact black pepper. Support vector machine (SVM), linear discriminant analysis (LDA) and K-nearest neighbors (KNN) were established and the identification effect of these models with different pre-processing methods were optimized and compared. By using support vector machine (SVM) classification and standard normal variate (SNV) spectral pre-processing, classification accuracy was 98.7% on the calibration and 100% on the validation sets for the determination of origin of black pepper. This study has demonstrated that portable spectrometers can be effective tools to use in the classification of black pepper samples according to their geographical origin.
Keyword
NIR, Black pepper, Geographical origin, Chemometrics, SVM, KNN, LDA
