The purpose of this study was to develop a detection technique for adulterated powder products using Raman spectroscopy. While conventional methods for food quality analysis require time and pretreatment procedures, Raman spectroscopy can be used to detect a food component rapidly and nondestructively without pretreatment. In this study, Raman spectra were measured using a 785 nm laser source in the range 3,200-170 cm-1 and multivariate analytical method using partial least square regression (PLSR) was applied to develop optimal models for predicting the mixed cornstarch concentration in adulterated onion powders. A range of preprocessing methods were applied to correct Raman spectra by reducing the systematic noise and variations produced by the light source and light scattering of the sample surface. Raman data were predicted with a high R2 value and low standard error for the entire preprocessing data set. The method proved its effectiveness to rapidly detect adulteration in powdered food.
|ジャーナル||Journal of the Faculty of Agriculture, Kyushu University|
|出版ステータス||出版済み - 2 1 2015|
All Science Journal Classification (ASJC) codes