Background

Tea is one of the most popular beverages around the world for its health benefit. The availability of rapid and reliable techniques for evaluating tea quality and safety is important for the tea industry. However, tea components and pesticide residues detection, sensory analysis, grading, and pricing are carried out by laborious, time-consuming, and destructive methods. Therefore, vibrational spectroscopy has emerged as an increasingly popular technology in tea quality and safety evaluation owing to its simple, nondestructive and rapid detection procedures.

Scope and approach

The current review compares the employment of different vibrational spectroscopic techniques including near-infrared (NIR), mid-infrared (MIR), Raman, Terahertz (THz) spectroscopies and hyperspectral imaging (HSI) in tandem with regression and classification models in the quantitative and qualitative analyses of tea quality and safety. Furthermore, band assignments of tea components based on these spectroscopies are summarized.

Key findings and conclusions

NIR spectroscopy dominates the applications of vibrational spectroscopy in the evaluation of tea quality, while only one study is related to THz spectroscopy. Raman technology is mainly applied to detect pesticide residues. HSI providing both spectral and spatial information has a potential in the optimization of the tea production process. Additionally, regression models find widespread use in the quantitative analysis of tea, such as phenolics, flavonoids, pigments, sensory evaluation, and pesticide residues, while classification models are useful for discriminating geographical origins, varieties, and fermentation degrees. The online detection of tea quality and safety by vibrational spectroscopy on the industrial scale is still an important and challenging task for future studies.

中文翻译：

---

振动光谱技术在茶质量和安全性分析中的最新进展

背景

茶因其有益于健康而成为世界上最受欢迎的饮料之一。快速，可靠的技术来评估茶的质量和安全性对于茶产业而言至关重要。但是，茶叶成分和农药残留的检测，感官分析，定级和定价是通过费力，费时和破坏性的方法进行的。因此，振动光谱由于其简单，无损和快速的检测程序而已成为茶质量和安全性评估中越来越流行的技术。

范围和方法

当前的审查比较了不同振动光谱技术的应用，包括近红外（NIR），中红外（MIR），拉曼，太赫兹（THz）光谱学和高光谱成像（HSI），以及定量和定量分析中的回归和分类模型。对茶叶质量和安全性的定性分析。此外，总结了基于这些光谱学的茶组分的频带分配。

主要发现和结论

近红外光谱法在振动光谱法在茶叶质量评估中的应用占主导地位，而只有一项研究与太赫兹光谱法有关。拉曼技术主要应用于检测农药残留。提供光谱和空间信息的HSI在优化茶叶生产过程中具有潜力。此外，回归模型广泛用于茶的定量分析中，例如酚类，黄酮类，色素，感官评估和农药残留，而分类模型可用于区分地理来源，品种和发酵程度。在工业规模上通过振动光谱在线检测茶的质量和安全性仍然是未来研究的重要且具有挑战性的任务。