Tea Research Institute, Guangdong Academy of Agricultural Science/Guangdong Provincial Key Laboratory of Tea Plant Resources Innovation & Utilization, Guangzhou 510640, China; College of Engineering, South China Agricultural University, Guangzhou 510642, China.
College of Engineering, South China Agricultural University, Guangzhou 510642, China.
Food Chem. 2025 Feb 1;464(Pt 1):141567. doi: 10.1016/j.foodchem.2024.141567. Epub 2024 Oct 9.
To rigorously assess black tea quality in large-scale production, this study introduces a multi-modal fusion approach integrating computer vision (CV) with Near-Infrared Spectroscopy (NIRS). CV technology is first applied to evaluate the tea's appearance quality, while NIRS quantifies key chemical components, including tea polyphenols (TP), free amino acids (FAA), and caffeine (CAF). Additionally, different methods are employed to extract potential quality features from NIR spectra. The information are then fused, and a classifier is utilized to accurately identify tea quality. Results show that the Temporal Convolutional Network (TCN) fused model achieves a 98.2 % accuracy rate, surpassing both the Convolutional Neural Network (CNN) fused model and traditional methods. This study demonstrates that TCNs effectively extract spectral features and that data fusion significantly enhances tea quality testing, offering valuable insights for production optimization.
为了在大规模生产中严格评估红茶质量，本研究引入了一种多模态融合方法，将计算机视觉（CV）与近红外光谱（NIRS）相结合。CV 技术首先用于评估茶叶的外观质量，而 NIRS 则量化关键化学成分，包括茶多酚（TP）、游离氨基酸（FAA）和咖啡因（CAF）。此外，还采用不同的方法从 NIR 光谱中提取潜在的质量特征。然后融合这些信息，并利用分类器来准确识别茶叶质量。结果表明，时频卷积网络（TCN）融合模型的准确率达到 98.2%，优于卷积神经网络（CNN）融合模型和传统方法。本研究表明 TCN 能够有效地提取光谱特征，并且数据融合显著增强了茶叶质量测试，为生产优化提供了有价值的见解。
