Classification of rice based on storage time by using near infrared spectroscopy and chemometric methods

Rice is among the most important food crops, and feeds more than half of the world's population. The freshness of rice decreases with storage time, therefore a fast and easy system for determining quality is greatly needed. We have investigated the potential of near-infrared spectroscopy (NIRS), combined with chemometric methods, for distinguishing rice samples of one, two, and three years of storage. A total of 240 rice samples were analyzed in this study. Principal components analysis (PCA) was initially conducted to look for possible clustering. Next, two pattern recognition methods (partial least-squares discriminant analysis (PLS-DA), and nearest neighbor (KNN)) were compared for their usefulness in building classification models. All two scored high for sensitivity and specificity, but a difference was seen for predictive accuracy. PLS-DA achieved a 97% predictive accuracy, whereas the KNN model, built after first derivative spectral pretreatment, scored 100%. Therefore, NIRS coupled with chemometric methods can be considered as an effective method to classify rice from different years of storage. This article provides a new technology for the evaluation of rice freshness in the market.