Fine-Tuning Model YOLOv8 untuk Meningkatkan Robustness pada Implementasi Real-World Deteksi Produk di Kasir Koperasi

Authors

  • Rafi Achmad Fachrudi Universitas Nusantara PGRI Kediri Indonesia
  • Daniel Swanjaya, M.Kom Universitas Nusantara PGRI Kediri Indonesia
  • Danar Putra Pamungkas, M.Kom Indonesia

DOI:

https://doi.org/10.29407/n560a953

Abstract

Penelitian ini bertujuan untuk mengembangkan sistem deteksi objek menggunakan model YOLOv8 guna mendukung sistem kasir cerdas tanpa barcode di koperasi. Model dilatih menggunakan dataset produk minuman kemasan yang dikumpulkan secara mandiri, kemudian dilakukan fine-tuning dengan data tambahan yang lebih bervariasi dari segi sudut, pencahayaan, dan kondisi objek. Hasil pengujian menunjukkan bahwa model fine-tuned lebih andal dalam mengenali objek di kondisi nyata meskipun terdapat penurunan pada beberapa metrik seperti precision dan mAP-50. Pendekatan augmentasi data dan pengaturan ulang pelatihan (freeze layer) terbukti meningkatkan kemampuan generalisasi model. Penelitian ini menunjukkan bahwa YOLOv8 dapat menjadi solusi efektif untuk otomatisasi kasir di lingkungan koperasi dan usaha kecil, serta mendukung upaya digitalisasi di sektor tersebut.

Keywords:

YOLOv8, fine-tuning, koperasi, sistem kasir cerdas

##plugins.themes.default.displayStats.downloads##

##plugins.themes.default.displayStats.noStats##

References

A. Vats and D. C. Anastasiu, “Enhancing Retail Checkout through Video Inpainting, YOLOv8 Detection, and DeepSort Tracking,” IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. Work., vol. 2023-June, pp. 5530–5537, 2023, doi: 10.1109/CVPRW59228.2023.00585.

W. Yulita, U. A. Ramadhani, Z. Mufidah, G. K. M. Atmajaya, and R. Bagaskara, “Improved human image density detection with comparison of YOLOv8 depth level architecture and drop-out implementation,” pp. 33–39, 2025.

V. Gandhi and S. Gandhi, “Fine-Tuning Without Forgetting: Adaptation of YOLOv8 Preserves COCO Performance,” 2025, [Online]. Available: http://arxiv.org/abs/2505.01016

Z. R. Utomo, P. W. Adi, P. S. Sasongko, and G. Rahman, “Development and Optimization of a Construction Personal Protective Equipment ( PPE ) Detection Model on YOLOv8 Architecture,” vol. 16, no. March, pp. 1–14, 2025, doi: 10.14710/jmasif.16.1.71622.

R. Ye, Q. Gao, Y. Qian, J. Sun, and T. Li, “Improved YOLOv8 and SAHI Model for the Collaborative Detection of Small Targets at the Micro Scale: A Case Study of Pest Detection in Tea,” Agronomy, vol. 14, no. 5, 2024, doi: 10.3390/agronomy14051034.

O. Selsa, I. Anggraeni, and P. T. Informatika, “Studi Komparatif Performa Framework Javascript Modern dalam Pengembangan Aplikasi Web,” vol. 2, no. 4, pp. 162–177, 2024.

E. U. Armin, A. Purnama Edra, F. I. Alifin, I. Sadidan, I. P. Sary, and U. Latifa, “Performa Model YOLOv8 untuk Deteksi Kondisi Mengantuk pada pengendara mobil,” BRAHMANA J. Penerapan Kecerdasan Buatan, vol. 5, no. 1, pp. 67–76, 2023, [Online]. Available: https://doi.org/10.30645/brahmana.v5i1.279

D. Reis, J. Hong, J. Kupec, and A. Daoudi, “Real-Time Flying Object Detection with YOLOv8”.

J. Lee, R. Tang, and J. Lin, “What Would Elsa Do? Freezing Layers During Transformer Fine-Tuning,” 2018.

Z. F. Khan, M. Ramzan, M. Raza, S. Member, and M. A. Khan, “Real-Time Polyp Detection From Endoscopic Images Using YOLOv8 With YOLO-Score Metrics for Enhanced Suitability Assessment,” IEEE Access, vol. 12, no. December, pp. 176346–176362, 2024, doi: 10.1109/ACCESS.2024.3505619.

Downloads

Published

2025-07-10

Issue

Vol. 9 No. 2 (2025): Prosiding Seminar Nasional Inovasi Teknologi Tahun 2025

Section

Articles

License

Copyright (c) 2025 Rafi Achmad Fachrudi, Daniel Swanjaya, M.Kom, Danar Putra Pamungkas, M.Kom

Creative Commons License

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:

  1. Copyright on any article is retained by the author(s).
  2. The author grants the journal, right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work’s authorship and initial publication in this journal.
  3. 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 acknowledgment of its initial publication in this journal.
  4. 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.
  5. The article and any associated published material is distributed under the Creative Commons Attribution-ShareAlike 4.0 International License

How to Cite

Fine-Tuning Model YOLOv8 untuk Meningkatkan Robustness pada Implementasi Real-World Deteksi Produk di Kasir Koperasi. (2025). Prosiding SEMNAS INOTEK (Seminar Nasional Inovasi Teknologi), 9(2), 1091-1100. https://doi.org/10.29407/n560a953