Abstract-Terahertz Spectroscopy Determination of Benzoic Acid Additive in Wheat Flour by Machine Learning

Xudong Sun, Ke Zhu, Junbin Liu, Jun Hu, Xiaogang Jiang, Yande Liu Zhiyuan Gong

https://link.springer.com/article/10.1007%2Fs10762-019-00579-z

Investigations were carried out to develop terahertz (THz) techniques combined with machine learning methods for rapidly measuring benzoic acid (BA) additive in wheat flour. The absorption coefficient had a significant characteristic peak at 1.94 THz, which usually varies with the increase of BA additive concentrations. Correlation analysis was initiated to choose a frequency pair of 1.94 and 1.86 THz that improved the predictive ability of multiple linear regressions (MLR) model. Then, quantitative determination models were modeled between BA concentrations and THz absorption coefficients using partial least square (PLS) regression and least square support vector machine (LS-SVM). Compared with MLR and PLS models, the LS-SVM model could be regarded as an effective tool for quality control of wheat flour with a correlation coefficient of prediction (Rp) of 0.994 and root mean square error of prediction (RMSEP) of 0.12%. The results suggest that combining THz spectroscopy with LS-SVM has the potential to quantitatively analyze BA additive in wheat flour.

Abstract-Analytical-performance improvement of laser-induced breakdown spectroscopy for the processing degree of wheat flour using a continuous wavelet transform

Ping Yang, Yining Zhu, Shisong Tang, Zhongqi Hao, Lianbo Guo, Xiangyou Li, Yongfeng Lu, and Xiaoyan Zeng

https://www.osapublishing.org/ao/abstract.cfm?uri=ao-57-14-3730

Quality and safety of food are two of the most important matters in our lives. Wheat is one of the most important products in the modern agricultural processing industry. Issues of mislabeling and adulteration are of increasingly serious concern in the grain market. They threaten the credibility of producers and traders and the rights of the consumers. Therefore, it is very significant to guarantee the processing degree of wheat flour. In this work, two different spectral peak recognition methods, i.e., artificial spectral peak recognition and automatic spectral peak recognition, are carried out to study the adulteration problem in the food industry. Three grades of the processing degree of wheat flour from northern China are classified by laser-induced breakdown spectroscopy (LIBS). To search for an automatic classification model, continuous wavelet transform is used for the automatic recognition of the LIBS spectrum peak. Principal component analysis is used to reduce the collinearity of LIBS spectra data. First, 20 principal components were selected to represent the spectral data for the following discrimination analysis by a support vector machine. The results showed that the classification accuracies of automatic spectral peak recognition are better than those of artificial spectral peak recognition. The classification accuracies of artificial spectral peak recognition and automatic spectral peak recognition are 95.33% and 98.67%; the fivefold cross-validation classification accuracies are 94.67% and 96.67%; and the operation times were 240 min and 2 min, respectively. It can be concluded that LIBS can provide simpler and faster classification without the use of any chemical reagent, which represents a decisive advantage for applications dedicated to rapidly detecting the processing degree of wheat flour and other cereals.

© 2018 Optical Society of America