School of Food and Biological Engineering, Jiangsu University, Zhenjiang 212013, China.
School of Food and Biological Engineering, Jiangsu University, Zhenjiang 212013, China.
Food Chem. 2021 Aug 15;353:129372. doi: 10.1016/j.foodchem.2021.129372. Epub 2021 Mar 8.
Matcha tea is rich in taste and bioactive constituents, quality evaluation of matcha tea is important to ensure flavor and efficacy. Near-infrared spectroscopy (NIR) in combination with variable selection algorithms was proposed as a fast and non-destructive method for the quality evaluation of matcha tea. Total polyphenols (TP), free amino acids (FAA), and polyphenols-to-amino acids ratio (TP/FAA) were assessed as the taste quality indicators. Successive projections algorithm (SPA), genetic algorithm (GA), and simulated annealing (SA) were subsequently developed from the synergy interval partial least squares (SiPLS). The overall results revealed that SiPLS-SPA and SiPLS-SA models combined with NIR exhibited higher predictive capabilities for the effective determination of TP, FAA and TP/FAA with correlation coefficient in the prediction set (R) of R > 0.97, R > 0.98 and R > 0.98 respectively. Therefore, this simple and efficient technique could be practically exploited for tea quality control assessment.
抹茶富含口感和生物活性成分，因此对抹茶的质量评估非常重要，以确保其口感和功效。近红外光谱（NIR）结合变量选择算法被提出作为一种快速无损的抹茶质量评估方法。总多酚（TP）、游离氨基酸（FAA）和多酚与氨基酸比（TP/FAA）被评估为口感质量指标。协同区间偏最小二乘（SiPLS）随后分别发展出连续投影算法（SPA）、遗传算法（GA）和模拟退火（SA）。总体结果表明，SiPLS-SPA 和 SiPLS-SA 模型与 NIR 结合，对有效测定 TP、FAA 和 TP/FAA 具有更高的预测能力，预测集的相关系数（R）分别大于 0.97、0.98 和 0.98。因此，这种简单有效的技术可以实际用于茶叶质量控制评估。
