红茶
发酵
高光谱成像
质量标准
近红外光谱
食品科学
化学成像
人工智能
化学
生物系统
数学
模式识别(心理学)
计算机科学
色谱法
生物
神经科学
作者
Luqing Li,Yujie Wang,Qingqing Cui,Ying Liu,Jingming Ning,Zhengzhu Zhang
标识
DOI:10.1016/j.jfca.2021.104300
摘要
Fermentation is the most critical step in black tea processing and contributes to the taste and appearance of black tea. Conventionally, the degree of fermentation is evaluated through sensory reviews by experts. However, this method has low accuracy and is subject to reviewers’ experience. In this study, the fermentation quality of black tea was qualitatively and quantitatively evaluated through near-infrared hyperspectral imaging (HSI). The spectral information of samples of four fermentation degrees was collected and preprocessed using standard normal variate. The developed qualitative model could discriminate fermentation degrees, with a correct classification rate of 83.33 % for the prediction set. The quantitative prediction model achieved excellent prediction accuracy for the catechin content, with a residual prediction deviation of 2.18. Finally, the spatial distribution of catechins during fermentation was mapped through chemical imaging. Our study confirmed the feasibility of HSI for monitoring the fermentation quality of black tea.
科研通智能强力驱动
Strongly Powered by AbleSci AI