Computer vision-based multi-bean coffee grading using deep learning networks
Procedia Computer Science · 2026 · 284 · 388-395
DOI: 10.1016/j.procs.2026.07.016 ↗Kurz gefasst
Fachartikel in der Zeitschrift Procedia Computer Science, 2026. Verfasst von Jinky G. Marcelo und 14 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Manual coffee bean grading is time-consuming and inconsistent, while the small size of beans poses challenges for automated computer vision systems. To address this, we present a comparative benchmark of deep learning models (YOLOv8, SSD, Faster R-CNN, and RetinaNet) for real-time coffee bean grading using a new public dataset of 6,468 images with 35,413 annotations. Our comparative analysis shows YOLOv8 achieved the highest performance with 98.86% mAP@50 and the fastest inference time. The result of this study validates the potential of Yolov8 for multi-bean processing in coffee grading towards deployment in real-world settings.
Zitieren
APA
Marcelo, J. G., Bonifacio, K. L. A., Verula, D. R. P., Casildo, N. J. G., Verula, V. T., Ayunar, G. S., Tajones, J. D., Jomaya, N. J. J., Buntod, R. M., Jr., F. D. B., Aguirre, M. R. C., Jr., A. V. M., Jr., D. R. A., Borres, K. M. B., & Ybañez, J. M. (2026). Computer vision-based multi-bean coffee grading using deep learning networks. Procedia Computer Science, 284, 388-395. https://doi.org/10.1016/j.procs.2026.07.016
Vancouver
Marcelo JG, Bonifacio KLA, Verula DRP, Casildo NJG, Verula VT, Ayunar GS, et al. Computer vision-based multi-bean coffee grading using deep learning networks. Procedia Computer Science. 2026;284:388-395. doi:10.1016/j.procs.2026.07.016
BibTeX
@article{Marcelo2026,
author = {Marcelo, Jinky G. and Bonifacio, Kent Levi A. and Verula, Dona rae P. and Casildo, Nathalie Joy G. and Verula, Viceliogems T. and Ayunar, Gladys S. and Tajones, John D. and Jomaya, Neil John J. and Buntod, Rinante M. and Jr., Ferdinand D. Bascones and Aguirre, Mindell Rey C. and Jr., Almerando V. Mendez and Jr., Danilo R. Antonio and Borres, Kathelyn Meg B. and Ybañez, Johnrey Mark},
title = {Computer vision-based multi-bean coffee grading using deep learning networks},
journal = {Procedia Computer Science},
year = {2026},
volume = {284},
pages = {388--395},
doi = {10.1016/j.procs.2026.07.016}
}
EUVIMED