Near-infrared Spectroscopy in Food Quality Control

Abstract

Food quality control is an important issue in wholo production chain, from soil to consumer. We have examined the usage of near-infrared spectroscopy in analysis of Finnish agriculture products, such as tomatoes, where we aimed to classify them based on the quality and anthocyanin content on blueberries. Our measurements of aqueous solutions, such as sodium bimetasulfide and morphine were used in study of applying near-infrared spectroscopy in concentration measurement of chemical.

We have shown that with linear multivariate methods, specially PCA and PLS, give good basis of the spectroscopy data analysis. On comparison of traditional regression models the linear multivariate methods were easier to use. There is no need to find important absorption peaks in spectral data, they are by-product of multivariate model. And we have illustrated how the PLS model predicts blueberry anthocyanin concentrations.