Cluster analysis method and Near-infrared spectroscopy applied to the identification of food

Cluster analysis method and Near-infrared (NIR) diffuse reflectance spectroscopy are applied to develop a fast identification method of food. The samples are collected from different manufactures and they are peanut oil, milling balm, and Jinhua ham. NIR spectra are pretreated with first derivative calculation and vector normalization. The NIR data are evaluated by cluster analysis, which uses the components of each spectrum to construct an informative classification of an unclassified data set. The distances between clusters are evaluated by Ward´s method of analysis of variance. The geometric distances in the multidimensional space are measured. The method can both distinguish peanut oil, milling balm, and Jinhua ham successfully. Overall, NIR diffuse reflectance spectroscopy using cluster analysis method is shown to have significant potential as a rapid and accurate method for identification of food.