EUVIMEDEUVIMED entdecken
Evaluationsstudie

Tea grading, blending, and matching based on computer vision and deep learning

Jilong Guo, Kexin Zhang, Selorm Yao-Say Solomon Adade, Jinsu Lin, Hao Lin, Quansheng Chen iD

Journal of the Science of Food and Agriculture · 2025-04 · 105 (6) · 3239-3251

DOI: 10.1002/jsfa.14088 ↗

Kurz gefasst

Evaluationsstudie in der Zeitschrift Journal of the Science of Food and Agriculture, online erschienen am 22.12.2024. Verfasst von Jilong Guo und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Food and Biological Engineering, Jiangsu University, Jiangsu, People's Republic of China. Hauptthemen laut MEDLINE: Deep Learning, Camellia sinensis, Tea. Die Arbeit wurde bisher 32-mal zitiert (Crossref) und nennt 32 Literaturangaben. Gefördert von: National Natural Science Foundation of China, China Spark Program in Fujian Province (2023S0035), Fujian Province Science and Technology Support Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND Accurate tea blending assessment and sample matching are critical in the tea production process. Traditional methods face efficiency and accuracy challenges, which can be addressed by advances in computer vision and deep learning. This study developed an efficient and non‐destructive method for fast tea grading classification, blending ratio evaluation, and sample matching. The method trained a Residual Network (ResNet) model on an enhanced dataset of tea images and used Convolutional Block Attention Module (CBAM) to improve the model's feature‐extraction ability. RESULTS The enhanced grade classification model achieved 95.05% accuracy for oolong tea and 99.13% accuracy for black tea, outperforming other deep‐learning models such as EfficientNet, MobileNet, and VGG16. For oolong tea blends, the model demonstrated greater efficiency than manual evaluation with an average absolute error of 2.26%. In black tea sample matching, the model achieved an average error of 3.34%. CONCLUSION These results highlight the importance of attention mechanisms in improving the analysis of images with intricate textures. The integration of deep learning and attention modules enhanced the accuracy and efficiency of tea quality evaluation processes effectively. This study underscores the transformative potential of intelligent classification and analysis methods in modernizing tea production, ensuring higher standards of consistency and quality. © 2024 Society of Chemical Industry.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningCamellia sinensisTeaPlant LeavesQuality ControlImage Processing, Computer-Assisted

Stichwörter der Autor:innen

attention mechanismcomputer visiondeep learningsample matchingtea blendingtea grade classification

Substanzen

Tea

Autor:innen und Institutionen

  • School of Food and Biological Engineering, Jiangsu University, Jiangsu, People's Republic of China.
  • College of Ocean Food and Biological Engineering, Jimei University, Xiamen, People's Republic of China.
  • Bama Tea Co., Ltd, Shenzhen, People's Republic of China.

ORCID-Profile: Quansheng Chen

Förderung

  • National Natural Science Foundation of China
  • China Spark Program in Fujian Province (2023S0035)
  • Fujian Province Science and Technology Support Project

Zitieren

APA

Guo, J., Zhang, K., Adade, S. Y. S., Lin, J., Lin, H., & Chen, Q. (2025). Tea grading, blending, and matching based on computer vision and deep learning. Journal of the Science of Food and Agriculture, 105(6), 3239-3251. https://doi.org/10.1002/jsfa.14088

Vancouver

Guo J, Zhang K, Adade SYS, Lin J, Lin H, Chen Q. Tea grading, blending, and matching based on computer vision and deep learning. Journal of the Science of Food and Agriculture. 2025;105(6):3239-3251. doi:10.1002/jsfa.14088

BibTeX
@article{Guo2025,
  author = {Guo, Jilong and Zhang, Kexin and Adade, Selorm Yao-Say Solomon and Lin, Jinsu and Lin, Hao and Chen, Quansheng},
  title = {Tea grading, blending, and matching based on computer vision and deep learning},
  journal = {Journal of the Science of Food and Agriculture},
  year = {2025},
  volume = {105},
  number = {6},
  pages = {3239--3251},
  doi = {10.1002/jsfa.14088}
}
Kostenlos starten →