Implementasi Model Hibrida RT-DETR dan Vision Transformer untuk Deteksi dan Klasifikasi Produk Retail Berdasarkan Citra Kemasan
DOI:
https://doi.org/10.54065/artificial.1982Keywords:
Citra Kemasan, Deteksi dan Klasifikasi Produk, Model Hibrida, RT-DETR, Vision TransformerAbstract
Identifikasi produk retail pada kemasan merupakan salah satu penerapan computer vision yang dapat mendukung proses pengenalan produk secara otomatis. Penelitian ini bertujuan mengimplementasikan model hibrida Real-Time DEtection TRansformer (RT-DETR) dan Vision Transformer (ViT) untuk mendeteksi dan mengklasifikasikan produk retail berdasarkan citra kemasan. Penelitian menggunakan pendekatan eksperimental dengan dataset yang dikumpulkan secara langsung melalui dokumentasi citra kemasan produk retail. Dataset terdiri atas 1.202 citra dari 15 kelas produk, yang dibagi menjadi data pelatihan, validasi, dan pengujian dengan rasio 80:10:10. RT-DETR digunakan sebagai model deteksi untuk menentukan lokasi objek produk melalui bounding box, sedangkan ViT digunakan untuk mengklasifikasikan hasil cropping objek ke dalam kelas produk. Model RT-DETR-L dilatih selama 36 epoch dengan ukuran citra 640 × 640 piksel dan batch size 8, sedangkan ViT-Base Patch16-224 dilatih selama 6 epoch menggunakan citra berukuran 224 × 224 piksel. Evaluasi RT-DETR dilakukan menggunakan metrik mean Average Precision (mAP), precision, dan recall, sedangkan kinerja klasifikasi hibrida dievaluasi menggunakan accuracy, precision, recall, dan F1-score. Hasil pengujian menunjukkan bahwa RT-DETR mampu mendeteksi objek produk dengan nilai precision dan recall sekitar 0,90 serta mAP@50 sekitar 0,93–0,95. Pada tahap klasifikasi, ViT mencapai akurasi validasi sebesar 99,2% pada akhir pelatihan. Berdasarkan evaluasi rinci sistem hibrida, diperoleh nilai accuracy, precision, recall, dan F1-score masing-masing sebesar 100%. Hasil tersebut menunjukkan bahwa kombinasi RT-DETR sebagai pendeteksi objek dan ViT sebagai pengklasifikasi mampu menjalankan proses deteksi dan klasifikasi produk retail pada dataset penelitian dengan kinerja yang sangat baik.
References
Amjoud, A. B., & Amrouch, M. (2023). Object Detection Using Deep Learning, CNNs and Vision Transformers: A Review. IEEE Access, 11, 35479–35516. https://doi.org/10.1109/ACCESS.2023.3266093
Arkin, E., Yadikar, N., Xu, X., Aysa, A., & Ubul, K. (2023). A survey: Object detection methods from CNN to transformer. Multimedia Tools and Applications, 82(14), 21353–21383. https://doi.org/10.1007/s11042-022-13801-3
Chen, S., Liu, D., Pu, Y., & Zhong, Y. (2022). Advances in deep learning-based image recognition of product packaging. Image and Vision Computing, 128, 104571. https://doi.org/10.1016/j.imavis.2022.104571
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., & Houlsby, N. (2020). An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (Version 2). arXiv. https://doi.org/10.48550/ARXIV.2010.11929
Fuchs, K., Grundmann, T., & Fleisch, E. (2019). Towards Identification of Packaged Products via Computer Vision: Convolutional Neural Networks for Object Detection and Image Classification in Retail Environments. Proceedings of the 9th International Conference on the Internet of Things, 1–8. https://doi.org/10.1145/3365871.3365899
Gebru, T., Morgenstern, J., Vecchione, B., Vaughan, J. W., Wallach, H., Iii, H. D., & Crawford, K. (2021). Datasheets for datasets. Communications of the ACM, 64(12), 86–92. https://doi.org/10.1145/3458723
Gothai, E., Bhatia, S., Alabdali, A. M., Sharma, D. K., Kondamudi, B. R., & Dadheech, P. (2022). Design features of grocery product recognition using deep learning. Intelligent Automation & Soft Computing, 34(2), 1231–1246. https://doi.org/10.32604/iasc.2022.026264
Gothai, E., Bhatia, S., M. Alabdali, A., Kumar Sharma, D., Raj Kondamudi, B., & Dadheech, P. (2022). Design Features of Grocery Product Recognition Using Deep Learning. Intelligent Automation & Soft Computing, 34(2), 1231–1246. https://doi.org/10.32604/iasc.2022.026264
Hosseini, A., Hooshanfar, K., Omrani, P., Toosi, R., Toosi, R., Ebrahimian, Z., & Akhaee, M. A. (2025). Brand visibility in packaging: A deep learning approach for logo detection, saliency-map prediction, and logo placement analysis. Discover Applied Sciences, 7(6), 537. https://doi.org/10.1007/s42452-025-07043-9
Hou, S., Li, J., Min, W., Hou, Q., Zhao, Y., Zheng, Y., & Jiang, S. (2024). Deep Learning for Logo Detection: A Survey. ACM Transactions on Multimedia Computing, Communications, and Applications, 20(3), 1–23. https://doi.org/10.1145/3611309
Kuznetsova, A., Rom, H., Alldrin, N., Uijlings, J., Krasin, I., Pont-Tuset, J., Kamali, S., Popov, S., Malloci, M., Kolesnikov, A., Duerig, T., & Ferrari, V. (2020). The Open Images Dataset V4: Unified Image Classification, Object Detection, and Visual Relationship Detection at Scale. International Journal of Computer Vision, 128(7), 1956–1981. https://doi.org/10.1007/s11263-020-01316-z
Maulana, A. W., Wibowo, S. A., Sunarya, U., Rahmania, R., & Insani, A. (2024). Performance Improvement of YOLOv8-RTDETR Method Based Retail Product Detection. Journal of Image and Graphics, 12(3), 292–301. https://doi.org/10.18178/joig.12.3.292-301
Melek, C. G., Battini Sönmez, E., & Varlı, S. (2024). Datasets and methods of product recognition on grocery shelf images using computer vision and machine learning approaches: An exhaustive literature review. Engineering Applications of Artificial Intelligence, 133, Article 108452. https://doi.org/10.1016/j.engappai.2024.108452
Pietrini, R., Paolanti, M., Mancini, A., Frontoni, E., & Zingaretti, P. (2024). Shelf management: A deep learning-based system for shelf visual monitoring. Expert Systems with Applications, 255(Part B), Article 124635. https://doi.org/10.1016/j.eswa.2024.124635
Selvam, P., & Koilraj, J. A. S. (2022). A Deep Learning Framework for Grocery Product Detection and Recognition. Food Analytical Methods, 15(12), 3498–3522. https://doi.org/10.1007/s12161-022-02384-2
Shah, S., & Tembhurne, J. (2023). Object detection using convolutional neural networks and transformer-based models: A review. Journal of Electrical Systems and Information Technology, 10, Article 54. https://doi.org/10.1186/s43067-023-00123-z
Wei, Y., Tran, S., Xu, S., Kang, B., & Springer, M. (2020). Deep Learning for Retail Product Recognition: Challenges and Techniques. Computational Intelligence and Neuroscience, 2020, 1–23. https://doi.org/10.1155/2020/8875910
Yanti, R., Susilawati, F. E., & Akramunnisa, A. (2024). Rancang Bangun Aplikasi Penjualan Berbasis Website pada Toko MD Komputer Palopo. Journal Artificial: Informatika Dan Sistem Informasi, 2(2), 147–160. https://doi.org/10.54065/artificial.552
Zaidi, S. S. A., Ansari, M. S., Aslam, A., Kanwal, N., Asghar, M., & Lee, B. (2022). A survey of modern deep learning based object detection models. Digital Signal Processing, 126, Article 103514. https://doi.org/10.1016/j.dsp.2022.103514
Zhao, Y., Lv, W., Xu, S., Wei, J., Wang, G., Dang, Q., Liu, Y., & Chen, J. (2024). DETRs beat YOLOs on real-time object detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 16965–16974). https://doi.org/10.1109/CVPR52733.2024.01605
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Putri Munawaroh, Harma Oktafia Lingga Wijaya, Davit Irawan
This work is licensed under a Creative Commons Attribution 4.0 International License.
License and Copyright Agreement
- Authors retain copyright and other proprietary rights related to the article.
- Authors retain the right and are permitted to use the substance of the article in their own future works, including lectures and books.
- Authors grant the journal the right of first publication with the work simultaneously licensed under Creative Commons Attribution License (CC BY 4.0) that allows others to share the work with an acknowledgment 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 acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post or self-archive 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.
