Related Experiment Video
Updated: Jul 7, 2025
04:36
Author Spotlight: Exploring Tea Aroma Using Solvent-Assisted Flavor Evaporation Technique
Published on: May 26, 2023
3.2K
Rapid and comprehensive grade evaluation of Keemun black tea using efficient multidimensional data fusion
Luqing Li1, Yurong Chen1, Shuai Dong1
1State Key Laboratory of Tea Plant Biology and Utilization, Anhui Agricultural University, 130 Changjiang West Road, Hefei 230036, China.
Food Chemistry: X
|December 25, 2023
Summary
This study introduces a new method for evaluating Keemun black tea quality using spectroscopy, computer vision, and sensors. The combined approach achieved 98.57% accuracy in grading tea and accurately predicted key flavor compounds.
Area of Science:
- Agricultural Science
- Analytical Chemistry
- Food Science
Background:
- Assessing the quality of Keemun black tea is crucial for market value and consumer satisfaction.
- Traditional methods for tea quality evaluation can be time-consuming and subjective.
- Objective and rapid analytical techniques are needed to complement existing methods.
Purpose of the Study:
- To develop a comprehensive and rapid evaluation method for Keemun black tea quality.
- To assess the feasibility of combining spectral, visual, and colorimetric data for tea analysis.
- To identify key chemical compounds related to black tea flavor and quality.
Main Methods:
- Micro-near-infrared spectroscopy, computer vision, and colorimetric sensor arrays were employed for data acquisition.
- Machine learning algorithms including Support Vector Machine (SVM), Least-Squares Support Vector Machine (LS-SVM), Extreme Learning Machine (ELM), and Partial Least Squares Discriminant Analysis (PLS-DA) were utilized for classification.
- Support Vector Regression (SVR) was applied for quantitative analysis of flavor substances.
Main Results:
- The LS-SVM model, incorporating mid-level data fusion, achieved a high accuracy of 98.57% in distinguishing between different grades of Keemun black tea.
- Quantitative analysis using SVR demonstrated strong correlation coefficients (0.81111–0.94249) for predicting key flavor compounds like caffeine and catechins.
- All predicted compounds showed residual predictive deviation values greater than 2, indicating robust model performance.
Conclusions:
- A multi-modal data fusion approach combining spectral, visual, and colorimetric information offers a rapid and comprehensive method for Keemun black tea quality assessment.
- The developed methodology can accurately classify tea grades and quantify important flavor constituents.
- This integrated approach provides a valuable tool for the tea industry, enhancing quality control and consistency.
