Concentrations of total catechins and theanine are important indices for monitoring the degree of fermentation, a key factor in determining the types, qualities, and taste characteristics of tea, during manufacturing process. Until now, there has not been any near-infrared (NIR) study applying theanine as an index for monitoring the degree of fermentation of tea. It is necessary to understand and monitor the degree of fermentation according to the concentrations of total catechins and theanine in tea by using NIR spectroscopy, especially for manufacturers of tea products. NIR spectroscopy integrated with modified partial least-squares regression (MPLSR) method was used for spectral analysis of total catechins and theanine in 161 tea samples in the full visible NIR wavelength range of 400−2498 nm (25000−4003 cm⁻¹). Both the optimal calibration models of total catechins and theanine built by MPLSR showed satisfactory predictability, with coefficient of determination of calibration (R_c²) and residual predictive deviation (RPD) up to 0.94 and 3.38, respectively. High total accuracy rates of 100 % were also achieved for the discrimination of green tea, partially fermented tea, and black tea in the validation sets by the optimal MPLSR calibration model of theanine. The results indicated that NIR spectroscopy has the potential to be adopted as an effective method of rapid and accurate inspection of the degree of fermentation of tea. This technique could contribute substantially to quality assurance during fermentation process of tea.

总儿茶素和茶氨酸的浓度是监测发酵程度的重要指标，发酵程度是制造过程中决定茶的类型、质量和口味特征的关键因素。到目前为止，还没有任何近红外（NIR）研究将茶氨酸作为监测茶叶发酵程度的指标。使用近红外光谱法根据茶叶中总儿茶素和茶氨酸的浓度来了解和监测发酵程度是非常有必要的，特别是对于茶产品的生产厂家。近红外光谱结合改进的偏最小二乘回归 (MPLSR) 方法用于在 400-2498 nm (25000-4003 cm ^-1）。由MPLSR构建的总儿茶素和茶氨酸的最佳校准模型均表现出令人满意的可预测性，校准决定系数(R _c ² )和残余预测偏差(RPD)分别达到0.94和3.38。通过茶氨酸的最佳 MPLSR 校准模型，还实现了 100% 的高总准确率，用于在验证集中区分绿茶、部分发酵茶和红茶。结果表明，近红外光谱有可能成为一种快速、准确检测茶叶发酵程度的有效方法。该技术可以大大有助于茶叶发酵过程中的质量保证。