Quantitative prediction and visualization of matcha color physicochemical indicators using hyperspectral microscope imaging technology

Authors: LI, DENGSHAN - Jiangsu University; Park, Bosoon; CHEN, QUANSHENG - Jiangsu University; OUYANG, QIN - Jiangsu University; KANG, RUI - Jiangsu Academy Agricultural Sciences

Submitted to: Food Control
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/17/2024
Publication Date: 4/18/2024
Citation: Li, D., Park, B., Chen, Q., Ouyang, Q., Kang, R. 2024. Quantitative prediction and visualization of matcha color physicochemical indicators using hyperspectral microscope imaging technology. Food Control. https://doi.org/10.1016/j.foodcont.2024.110531.
DOI: https://doi.org/10.1016/j.foodcont.2024.110531

Interpretive Summary: Matcha is a powder made by grinding green tea. High-quality matcha usually appears as a bright green color, whereas low-quality matcha may exhibit a yellowish or greenish-gray color with a grayish-brown hue. During the matcha production and storage process, chlorophyll is unstable in oxygen, heat, and light, which leads to chlorophyll oxidative and induces negative impressions on matcha color. Therefore, color is a crucial factor in the sensory evaluation of matcha, and chlorophyll significantly influences the color of matcha. The traditional method for evaluating the matcha color is human sensory evaluation, which relies on professional tea tasters to score the color of matcha considering its appearance. The spectrophotometer method is a conventional technique for detecting chlorophyll contents with high precision. However, this method is complicated, time-consuming, and cannot meet the requirements of rapid detection in matcha production processing. Hyperspectral microscope imaging (HMI) technology combines the merits of spectroscopy and imaging technologies, which can simultaneously provide the spectra and spatial features of the samples. With the help of image processing techniques, HMI can visualize the distribution of the measured indicators. Visible and near-infrared spectra were extracted from the regions of interest, and the calibration models were established based on the characteristic wavelength variables. Furthermore, the distribution maps of color physicochemical indicators and chlorophyll contents were visualized based on the optimal models. HMI technology combined with multivariate calibration methods could achieve non-destructive detection and visualization of color physicochemical indicators and chlorophyll contents in matcha, which provided an alternative approach for intelligent quality assessment of matcha.

Technical Abstract: Color is a crucial factor in the sensory assessment of matcha, and chlorophyll significantly influences the color of matcha. The color physicochemical indicators and chlorophyll contents in matcha were determined and visualized using hyperspectral microscope imaging (HMI) technology. The average spectra were calculated from the regions of interest (ROI). After spectra preprocessing, various variable selection algorithms and the hybridization of interval random frog (iRF) and successive projections algorithm (SPA) were adopted to optimize the partial least squares (PLS) calibration models. The correlation analysis showed that the matcha color was closely related to the chlorophyll contents. The iRF-SPA-PLS models outperformed other calibration models, with coefficients of prediction (Rp) of 0.9262, 0.8826, 0.8583, 0.8243, 0.7518, and 0.8093 for L*, a*, b*, chlorophyll a, chlorophyll b, and chlorophyll total, respectively. Furthermore, the distribution maps of color physicochemical indicators and chlorophyll contents were visualized using these iRF-SPA-PLS models. The results indicated that the HMI technology coupled with chemometrics could achieve non-destructive determination and visualization of matcha color physicochemical indicators and chlorophyll contents.