A Hybrid Framework Based on YOLOv8 and Vision Transformer for Multi-Class Detection and Classification of Coffee Fruit Maturity Levels
DOI:
https://doi.org/10.30871/jaic.v9i5.10590Keywords:
YOLOv8, Computer Vision, Object detection, multi-class classification of coffee fruitsAbstract
Detection and classification of coffee cherries based on maturity levels present a significant challenge in agricultural product processing systems, primarily due to the high visual similarity among classes within a single bunch. This study aims to develop a multi-class detection and classification system for coffee cherries by integrating YOLOv8 and Vision Transformer (ViT) as a classification enhancer. The initial detection process is conducted using YOLOv8 to identify and automatically crop coffee cherry objects from bunch images. These cropped images are then re-classified using the Vision Transformer to improve prediction accuracy. The training process was carried out with a learning rate of 0.0001, a batch size of 16, and epoch variations of 50, 100, and 150. Evaluation results demonstrate that the integration of YOLOv8 and ViT significantly improves classification accuracy compared to using YOLOv8 alone. At 100 epochs, the YOLOv8+ViT model achieved an accuracy of 89.52%, a precision of 90.43%, and a recall of 89.52%, outperforming the standalone YOLOv8 model, which only reached an accuracy of 75.44%. These results indicate that the Vision Transformer effectively enhances classification performance, particularly for visually similar coffee cherry classes. The integration of these two methods offers a promising alternative solution for improving image-based multi-class classification in agriculture and other domains involving complex visual objects.
Downloads
References
[1] A. Subki and B. Imran, “Implementasi Deep Learning Menggunakan CNN dengan Arsitektur Alexnet Untuk Klasifikasi dan Identifikasi Jenis Kopi Khas Lombok Ahmad,” Explore, vol. 14, no. 2, pp. 135–140, 2024.
[2] N. Pradita, Hayati, Suwardji, Muktasam, and Mulyati, “Analisis Keberlanjutan Dimensi Ekologi Kopi Arabika di Lahan Kering Desa Sajang Kecamatan Sembalun Kabupaten Lombok Timur,” Agroteksos, vol. 34, no. 2, pp. 383–391, 2024.
[3] L. Y. K. Chandra, B. I. Linggarweni, and S. Novida, “Analisis Pendapatan Usaha Kopi Bubuk Arabika di Desa Sajang Kecamatan Sembalun Kabupaten Lombok Timur,” J. Ekon. dan Bisnis, vol. 3, no. 2, pp. 148–155, 2023, doi: 10.56145/jurnalekonomidanbisnis.v3i2.71.
[4] T. C. Pham, V. D. Nguyen, C. H. Le, M. Packianather, and V. D. Hoang, “Artificial intelligence-based solutions for coffee leaf disease classification,” IOP Conf. Ser. Earth Environ. Sci., vol. 1278, no. 1, 2023, doi: 10.1088/1755-1315/1278/1/012004.
[5] E. Elbasi et al., “Artificial Intelligence Technology in the Agricultural Sector: A Systematic Literature Review,” IEEE Access, vol. 11, pp. 171–202, 2022.
[6] B. Ye, R. Xue, and H. Xu, “ASD-YOLO: a lightweight network for coffee fruit ripening detection in complex scenarios,” Front. Plant Sci., vol. 16, no. February, pp. 1–13, 2025, doi: 10.3389/fpls.2025.1484784.
[7] H. C. Bazame, J. P. Molin, D. Althoff, and M. Martello, “Detection of coffee fruits on tree branches using computer vision,” Sci. Agric., vol. 80, no. October, 2022, doi: 10.1590/1678-992X-2022-0064.
[8] S. Velásquez, A. P. Franco, N. Peña, J. C. Bohórquez, and N. Gutiérrez, “Classification of the maturity stage of coffee cherries using comparative feature and machine learning,” Coffee Sci., vol. 16, no. March, p. 1, 2021, doi: 10.25186/.v16i.1710.
[9] M. N. Izza and G. P. Kusuma, “Image Classification of Green Arabica Coffee Using Transformer-Based Architecture,” Int. J. Eng. Trends Technol., vol. 72, no. 6, pp. 304–314, 2024, doi: 10.14445/22315381/IJETT-V72I6P128.
[10] M. García, J. E. Candelo-Becerra, and F. E. Hoyos, “Quality and defect inspection of green coffee beans using a computer vision system,” Appl. Sci., vol. 9, no. 19, 2019, doi: 10.3390/app9194195.
[11] H. L. Gope, H. Fukai, F. M. Ruhad, and S. Barman, “Comparative analysis of YOLO models for green coffee bean detection and defect classification,” Sci. Rep., vol. 14, no. 1, pp. 1–16, 2024, doi: 10.1038/s41598-024-78598-7.
[12] A. Rincon-Jimenez et al., “Ripeness stage characterization of coffee fruits (coffea arabica L. var. Castillo) applying chromaticity maps obtained from digital images,” in Materials Today: Proceedings, Elsevier Ltd., 2021, pp. 1271–1278. doi: 10.1016/j.matpr.2020.11.264.
[13] A. Michael and M. Garonga, “Classification model of ‘Toraja’ arabica coffee fruit ripeness levels using convolution neural network approach,” Ilk. J. Ilm., vol. 13, no. 3, pp. 226–234, 2021, doi: 10.33096/ilkom.v13i3.861.226-234.
[14] A. G. Costa, D. A. G. De Sousa, J. L. Paes, J. P. B. Cunha, and M. V. M. De Oliveira, “Classification of robusta coffee fruits at different maturation stages using colorimetric characteristics” Eng. Agrícola, vol. 4430, no. 4, pp. 518–525, 2020, [Online]. Available: https://doi.org/10.1590/1809-4430-Eng.Agric.v40n4p518-525/2020
[15] B. Xiao, M. Nguyen, and W. Q. Yan, “Fruit ripeness identification using YOLOv8 model,” Multimed. Tools Appl., vol. 83, no. 9, pp. 28039–28056, 2024, doi: 10.1007/s11042-023-16570-9.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Ahmad Subki, M. Zulpahmi, Bahtiar Imran
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) ) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
