Classification of brands of instant noodles using Vis/NIR spectroscopy and chemometrics

Visible and near infrared (Vis/NIR) spectroscopy combined with chemometric methods was applied for the classification of brands of instant noodles. Six brands of instant noodles and a total of 360 samples were prepared for the discrimination analysis. Partial least squares (PLS) analysis was implemented for the extraction of principal components (PCs). The first nine PCs were regarded as the inputs to develop the back propagation neural network (BPNN) model and least squares-support vector machine (LS-SVM) model. The performance of the model was validated by the 90 unknown samples and an excellent precision and recognition ratio of 98.9% and 100% were achieved by BPNN and LS-SVM, respectively. Simultaneously, certain sensitive wavelengths for the identification of brands were proposed by x-loading weights and regression coefficients. The results indicated that Vis/NIR spectroscopy could be used as a rapid and non-destructive method for the classification of different brands of instant noodles.