Study on Brand Identification of Monosodium Glutamate Using Sensitive Wavelengths of Short-Wave Near Infrared Spectroscopy

Xiaojing Chen,Di Wu,Sijie Feng,Yong He,Liu Shou
DOI: https://doi.org/10.1109/icmmt.2008.4540824
2008-01-01
Abstract:Monosodium glutamate (MSG) is a worldwide flavor enhancer. It is well known that to some extant, the brand stands for the quality of the products. Short-wave near-infrared spectroscopy (SWNIR) technique was applied in this study. The objectives of this study were to investigate the feasibility of using SWNIR spectroscopy to discriminate different brands of MSG, and to validate the performance of selected sensitive ebands. 66 MSG samples were randomly selected for the calibration set, while the remaining 134 samples for the prediction set. Smoothing way of moving average and standard normal variate (SNV) were used as pretreatments. Based on principal components analysis, 2-D principal components plot was clustered well. Least-squares support vector machines (LS-SVM) was applied to brand identification based on absorbance spectra from 800nm to 1050nm. Finally, recognition ratio of 99% was obtained. Sensitive bands, 900 and 968 nm, were obtained by loading weights and regression coefficients in partial least squares. The prediction results of 99% indicated that the selected wavelengths reflected the main characteristics of MSG of different brands based on SWNIR spectroscopy and LS-SVM model.
What problem does this paper attempt to address?