Abstract: The brightness of tea infusion is one of the important factors reflecting the quality of black tea. At present, the evaluation of the brightness of tea infusion is dependent on the human sensory evaluation, which lacks a quantitative assessment. In this study, based on the brightness of the infusions obtained by the sensory evaluation, 43 congou black tea infusions with different brightness were measured using color difference analysis in combination with multivariate analysis and linear regression, and an equation for brightness prediction was established. The results show that bright and dark black tea infusions could be discriminated based on their color difference profiles in combination with principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA). Based on linear regression, the key variables L99 (lightness) and C* (chroma) were selected and the established prediction equation was L_pre=-46.028+0.519×L99+0.117×C^*. It was revealed that, the predicted brightness showed high correlation with that obtained by human sensory evaluation with the Pearson correlation of 0.92 (P<0.01). Besides, the effectiveness of established prediction equation for quantitative evaluation of tea infusion brightness had been further proved using validation set. In summary, in this study, a rapid and effective method based on color difference analysis and multi statistical approaches was developed which enabled quantitative evaluation of the brightness of black tea infusions. This study could provide new insights into the intelligent evaluation of black tea quality, as well as oriented manufacture of highly-bright black tea.

Key words: Congou black tea, tea infusion brightness, color difference analysis, multivariate analysis