This is a preprint, a preliminary version of a manuscript that has not completed peer review at a journal. Research Square does not conduct peer review prior to posting preprints. The posting of a preprint on this server should not be interpreted as an endorsement of its validity or suitability for dissemination as established information or for guiding clinical practice.
Research
Study on the Determination Model of Four Kinds of Tea Polyphenols in Fresh Tea Based on Visible and Near-infrared Spectroscopy
Xuemei Liu
East China JiaoTong University
Corresponding Author
ORCiD: https://orcid.org/0000-0001-9542-8656
License:
This work is licensed under a CC BY 4.0 License. Read Full License
Background: For tea, tea polyphenols is an essential indicator to measure the quality of tea. In this paper, the content of four tea polyphenols in fresh tea was determined by visible and near-infrared spectroscopy combined with chemometrics.
Results: First, the spectrum data of three kinds of tea, Juhuachun (J), Zhenong25 (Z) and Yingshuang (Y) were collected. A total of 159 samples were collected, 106 of which were used for calibration and 53 for prediction. Then the content of tea polyphenols was determined by HPLC and the physicochemical value samples were established. Subsequently, the spectral data was preprocessed to eliminate noise interference, and a partial least squares (PLS) model was established to select the optimal preprocessing method. In order to improve the efficiency and accuracy of detection, Competitive adaptive reweighted sampling (CARS), Successive projections algorithm (SPA) and Random frog algorithm (RF) were used to extract characteristic wavelengths from the pretreatment spectrum. Based on characteristic wavelengths, PLS, multiple linear regression (MLR) linear models and least squares support vector machine (LS-SVM) nonlinear models were established to predict the content of four tea polyphenols. The performance of LS-SVM models is superior to that of PLS and MLR models. The RP2 values of the four tea polyphenols LS-SVM models based on SPA and CARS were increased to 0.996, 0.991, 0.997, 0.988 and 0.997, 0.991, 0.997, 0.984, respectively. The RP2 values of the four tea polyphenols LS-SVM models based on RF were also increased to 0.996, 0.986, 0.994 and 0.977.
Conclusions: It can be found that the LS-SVM model based on SPA is the most suitable prediction model for the content of tea polyphenols. It has the least input variables and better performance. Therefore, visible and near-infrared spectroscopy can be used as an effective method to measure the content of tea polyphenols in fresh tea.
Keywords
Visible and near-infrared spectroscopy, Tea polyphenols, Partial least squares, Multiple linear regression, Least squares support vector machine, Characteristic wavelengths
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Due to technical limitations, full-text HTML conversion of this manuscript could not be completed. However, the manuscript can be downloaded and accessed as a PDF.
Badges
Prescreen
This manuscript passed our Prescreen assessment
Complete author information
Appropriate declaration statements
Potential risks to human health
Prescreen
History
CURRENT STATUS: Posted
Version 1
Posted 15 Dec, 2020
No community comments so far
Metrics
Comments: 0
PDF Downloads: ...
HTML Views: ...
Learn more about our company.
This is a preprint. It has not completed peer review.
Research
Study on the Determination Model of Four Kinds of Tea Polyphenols in Fresh Tea Based on Visible and Near-infrared Spectroscopy
Xuemei Liu
East China JiaoTong University
Corresponding Author
ORCiD: https://orcid.org/0000-0001-9542-8656
Badges
Prescreen
This manuscript passed our Prescreen assessment
Complete author information
Appropriate declaration statements
Potential risks to human health
Prescreen
History
CURRENT STATUS: Posted
Version 1
Posted 15 Dec, 2020
No community comments so far
Metrics
Comments: 0
PDF Downloads: ...
HTML Views: ...
License:
This work is licensed under a CC BY 4.0 License. Read Full License
Background: For tea, tea polyphenols is an essential indicator to measure the quality of tea. In this paper, the content of four tea polyphenols in fresh tea was determined by visible and near-infrared spectroscopy combined with chemometrics.
Results: First, the spectrum data of three kinds of tea, Juhuachun (J), Zhenong25 (Z) and Yingshuang (Y) were collected. A total of 159 samples were collected, 106 of which were used for calibration and 53 for prediction. Then the content of tea polyphenols was determined by HPLC and the physicochemical value samples were established. Subsequently, the spectral data was preprocessed to eliminate noise interference, and a partial least squares (PLS) model was established to select the optimal preprocessing method. In order to improve the efficiency and accuracy of detection, Competitive adaptive reweighted sampling (CARS), Successive projections algorithm (SPA) and Random frog algorithm (RF) were used to extract characteristic wavelengths from the pretreatment spectrum. Based on characteristic wavelengths, PLS, multiple linear regression (MLR) linear models and least squares support vector machine (LS-SVM) nonlinear models were established to predict the content of four tea polyphenols. The performance of LS-SVM models is superior to that of PLS and MLR models. The RP2 values of the four tea polyphenols LS-SVM models based on SPA and CARS were increased to 0.996, 0.991, 0.997, 0.988 and 0.997, 0.991, 0.997, 0.984, respectively. The RP2 values of the four tea polyphenols LS-SVM models based on RF were also increased to 0.996, 0.986, 0.994 and 0.977.
Conclusions: It can be found that the LS-SVM model based on SPA is the most suitable prediction model for the content of tea polyphenols. It has the least input variables and better performance. Therefore, visible and near-infrared spectroscopy can be used as an effective method to measure the content of tea polyphenols in fresh tea.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Due to technical limitations, full-text HTML conversion of this manuscript could not be completed. However, the manuscript can be downloaded and accessed as a PDF.