Resin Code Classification on Plastic Packaging Using Few-Shot Learning
DOI:
https://doi.org/10.56705/ijodas.v7i2.406Keywords:
Few-Shot Learning, Prototypical Network, RIC Classification, EfficientNet-B2, Plastic Waste SortingAbstract
Poorly managed plastic waste poses a serious environmental threat. Accurate sorting based on the Resin Identification Code (RIC) can improve recycling quality; however, conventional deep learning methods generally require large labeled datasets that are costly and difficult to obtain, particularly for small RIC symbols printed on packaging. This study applies Few-Shot Learning using Prototypical Networks with a 7-way 5-shot episodic training scheme. A self-collected dataset containing 350 images was used, consisting of 50 images for each of seven RIC classes: PETE, HDPE, PVC, LDPE, PP, PS, and OTHER. Three feature-extraction backbones—ConvNet4, ResNet-18, and EfficientNet-B2—were evaluated, while an ablation experiment examined the contribution of light augmentation applied to the support set. Performance was assessed using episodic training accuracy and accuracy and F1-score on an independent test set of 42 images. EfficientNet-B2 achieved the highest episodic training accuracy of 93.36%, outperforming ResNet-18 at 88.71% and ConvNet4 at 54.14%. With light augmentation, EfficientNet-B2 obtained 85.71% test accuracy and correctly classified all PVC, LDPE, and PP samples. Most errors involved visually similar categories, particularly HDPE samples misclassified as PP. Subsequent fine-tuning corrected five of six initial errors, increasing test accuracy to 90.48% and macro F1-score from 0.857 to 0.903. These findings demonstrate that cosine-distance Prototypical Networks are effective for RIC classification under limited-data conditions and provide a promising foundation for automated plastic-waste sorting systems. Training code and experimental configurations are available upon request for research purposes.
Downloads
References
[1] Sustainable Waste Indonesia, “Collection and Recycling Rate Index of Plastic Waste in Indonesia,” 2025.
[2] OECD, Regional Plastics Outlook for Southeast and East Asia. OECD Publishing, 2025.
[3] J. Vaucher, A. Demongeot, V. Michaud, and Y. Leterrier, “Recycling of Bottle Grade PET: Influence of HDPE Contamination on the Microstructure and Mechanical Performance of 3D Printed Parts,” Polymers, vol. 14, no. 24, p. 5507, Dec. 2022, https://doi.org/10.3390/polym14245507.
[4] T. Werner, I. Taha, and D. Aschenbrenner, “An overview of polymer identification techniques in recycling plants with focus on current and future challenges,” Procedia CIRP, vol. 120, pp. 1381-1386, 2023, https://doi.org/10.1016/j.procir.2023.09.180.
[5] J. Bobulski and M. Kubanek, “Deep Learning for Plastic Waste Classification System,” Applied Computational Intelligence and Soft Computing, vol. 2021, pp. 1-7, May 2021, https://doi.org/10.1155/2021/6626948.
[6] A. A. P. Chazhoor, E. S. L. Ho, B. Gao, and W. L. Woo, “Deep transfer learning benchmark for plastic waste classification,” Intelligence & Robotics, 2022, https://doi.org/10.20517/ir.2021.15.
[7] E. Gothai, R. Thamilselvan, P. Natesan, M. Keerthivasan, K. Kabinesh, and D. K. Ruban, “Plastic Waste Classification Using CNN for Supporting 3R’s Principle,” in 2022 International Conference on Computer Communication and Informatics (ICCCI), Jan. 2022, pp. 01-07, https://doi.org/10.1109/ICCCI54379.2022.9740902.
[8] K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition,” in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2016, pp. 770-778, https://doi.org/10.1109/CVPR.2016.90.
[9] M. Tan and Q. V. Le, “EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks.”
[10] L. Listyalina, Y. Yudianingsih, A. W. Soedjono, E. L. Utari, and D. A. Dharmawan, “Deep-RIC: Plastic Waste Classification using Deep Learning and Resin Identification Codes (RIC),” Telematika, vol. 19, no. 2, p. 215, June 2022, https://doi.org/10.31315/telematika.v19i2.7419.
[11] S. Agarwal, R. Gudi, and P. Saxena, “Image Classification Approaches for Segregation of Plastic Waste Based on Resin Identification Code,” Transactions of the Indian National Academy of Engineering, vol. 7, no. 3, pp. 739-751, Sept. 2022, https://doi.org/10.1007/s41403-022-00324-4.
[12] Y. Wang, Q. Yao, J. T. Kwok, and L. M. Ni, “Generalizing from a Few Examples: A Survey on Few-shot Learning,” ACM Computing Surveys, vol. 53, no. 3, pp. 1-34, May 2021, https://doi.org/10.1145/3386252.
[13] Y. Song, T. Wang, P. Cai, S. K. Mondal, and J. P. Sahoo, “A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities,” ACM Computing Surveys, vol. 55, no. 13s, pp. 1-40, Dec. 2023, https://doi.org/10.1145/3582688.
[14] J. Snell, K. Swersky, and R. S. Zemel, “Prototypical Networks for Few-shot Learning.” arXiv, June 19, 2017, Accessed: Feb. 15, 2024, http://arxiv.org/abs/1703.05175.
[15] C. Finn, P. Abbeel, and S. Levine, “Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks.” arXiv, July 18, 2017, https://doi.org/10.48550/arXiv.1703.03400.
[16] J. Y. Lim, K. M. Lim, C. P. Lee, and Y. X. Tan, “A review of few-shot image classification: Approaches, datasets and research trends,” Neurocomputing, vol. 649, p. 130774, Oct. 2025, https://doi.org/10.1016/j.neucom.2025.130774.
[17] E. Triantafillou et al., “Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples.” arXiv, Apr. 08, 2020, https://doi.org/10.48550/arXiv.1903.03096.
[18] Y. Tian, Y. Wang, D. Krishnan, J. B. Tenenbaum, and P. Isola, “Rethinking Few-Shot Image Classification: A Good Embedding is All You Need?,” in Computer Vision - ECCV 2020, vol. 12359, A. Vedaldi, H. Bischof, T. Brox, and J.-M. Frahm, Eds. Cham: Springer International Publishing, 2020, pp. 266-282.
[19] S. Hou, W. Liu, A. Karim, Z. Jia, W. Jia, and Y. Zheng, “Few‐shot logo detection,” IET Computer Vision, vol. 17, no. 5, pp. 586-598, Aug. 2023, https://doi.org/10.1049/cvi2.12205.
[20] L. Jamieson, E. Elyan, and C. F. Moreno-García, “Few-Shot Symbol Detection in Engineering Drawings,” Applied Artificial Intelligence, vol. 38, no. 1, p. 2406712, Dec. 2024, https://doi.org/10.1080/08839514.2024.2406712.
[21] M. Alfaro-Contreras, A. Ríos-Vila, J. J. Valero-Mas, and J. Calvo-Zaragoza, “Few-shot symbol classification via self-supervised learning and nearest neighbor,” Pattern Recognition Letters, vol. 167, pp. 1-8, Mar. 2023, https://doi.org/10.1016/j.patrec.2023.01.014.
[22] I. Ranjbar, Y. Ventikos, and M. Arashpour, “Zero-shot and few-shot multimodal plastic waste classification with vision-language models,” Waste Management, vol. 202, p. 114815, July 2025, https://doi.org/10.1016/j.wasman.2025.114815.
[23] O. Russakovsky et al., “ImageNet Large Scale Visual Recognition Challenge,” International Journal of Computer Vision, vol. 115, no. 3, pp. 211-252, Dec. 2015, https://doi.org/10.1007/s11263-015-0816-y.
[24] H. Wang, S. Tian, Y. Fu, J. Zhou, J. Liu, and D. Chen, “Feature augmentation based on information fusion rectification for few-shot image classification,” Scientific Reports, vol. 13, no. 1, p. 3607, Mar. 2023, https://doi.org/10.1038/s41598-023-30398-1.
[25] P. Zhu, Z. Zhu, Y. Wang, J. Zhang, and S. Zhao, “Multi-granularity episodic contrastive learning for few-shot learning,” Pattern Recognition, vol. 131, p. 108820, Nov. 2022, https://doi.org/10.1016/j.patcog.2022.108820.
[26] S. Agustiani, A. Junaidi, R. Aryanti, and A. A. B. Kamil, “Waste Classification using EfficientNetB3-Based Deep Learning for Supporting Sustainable Waste Management,” vol. 4, no. 1, 2025.
[27] A. Tharwat, “Classification assessment methods,” Applied Computing and Informatics, vol. 17, no. 1, pp. 168-192, Jan. 2021, https://doi.org/10.1016/j.aci.2018.08.003.
[28] X. Li, X. Yang, Z. Ma, and J.-H. Xue, “Deep metric learning for few-shot image classification: A Review of recent developments,” Pattern Recognition, vol. 138, p. 109381, June 2023, https://doi.org/10.1016/j.patcog.2023.109381.
[29] V. O. Kuzevanov and D. V. Tikhomirova, “Deep Learning for Effective Visualization and Classification of Recyclable Material Labels,” Scientific Visualization, vol. 16, no. 5, pp. 179-196, Dec. 2024, https://doi.org/10.26583/sv.16.5.12.
[30] W. Li, L. Wang, J. Xu, J. Huo, Y. Gao, and J. Luo, “Revisiting Local Descriptor Based Image-To-Class Measure for Few-Shot Learning,” in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2019, pp. 7253-7260, https://doi.org/10.1109/CVPR.2019.00743.
Published
Issue
Section
License
Copyright (c) 2026 Anindita Septiarini, Alyani Noor Septalia, Masna Wati
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Authors retain copyright and full publishing rights to their articles. Upon acceptance, authors grant Indonesian Journal of Data and Science a non-exclusive license to publish the work and to identify itself as the original publisher.
Self-archiving. Authors may deposit the submitted version, accepted manuscript, and version of record in institutional or subject repositories, with citation to the published article and a link to the version of record on the journal website.
Commercial permissions. Uses intended for commercial advantage or monetary compensation are not permitted under CC BY-NC 4.0. For permissions, contact the editorial office at ijodas.journal@gmail.com.
Legacy notice. Some earlier PDFs may display “Copyright © [Journal Name]” or only a CC BY-NC logo without the full license text. To ensure clarity, the authors maintain copyright, and all articles are distributed under CC BY-NC 4.0. Where any discrepancy exists, this policy and the article landing-page license statement prevail.
