Evaluation of matcha tea quality index using portable NIR spectroscopy coupled with chemometric algorithms

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Chemometrics

Background:

• Matcha tea quality is crucial for consumer satisfaction and market value.
• Traditional methods for assessing tea quality are often time-consuming and destructive.
• Objective and rapid quality assessment methods are needed for the tea industry.

Purpose of the Study:

• To develop a portable near-infrared (NIR) spectroscopy system for predicting matcha tea quality.
• To utilize chemometric algorithms for the accurate quantification of tea polyphenols and amino acids.
• To establish a non-destructive method for classifying matcha tea based on its chemical composition.

Main Methods:

• Utilized a portable near-infrared (NIR) spectroscopy system.
• Preprocessed spectral data using standard normal variate (SNV), mean center (MC), and first-order derivative (1st D) methods.
• Applied partial least squares (PLS) regression combined with variable selection algorithms (RF-PLS, Si-PLS, GA-PLS, CARS-PLS) for model development.

Main Results:

• The random frog partial least square (RF-PLS) model demonstrated optimal performance for predicting tea polyphenols and amino acids.
• Achieved high prediction accuracy with correlation coefficients (R_P) of 0.8625 for polyphenols and 0.9662 for amino acids.
• Successfully classified matcha tea quality into qualified, unqualified, and excellent grades with an 83.33% accuracy rate.

Conclusions:

• The developed NIR spectroscopy system offers a rapid, accurate, and non-destructive platform for matcha tea quality assessment.
• The method effectively classifies tea samples based on the ratio of tea polyphenols to amino acids.
• This technology has the potential to revolutionize quality control in the tea industry.