Smart Waste Sorting Through Advanced Computer Vision: Optimizing YOLOv11 for High-Accuracy Waste Classification
Downloads
This research aims to develop a high-accuracy, real-time waste classification system to overcome the inefficiencies and errors associated with manual sorting. The methodology utilizes the advanced YOLO11x-cls architecture, enhanced through transfer learning and optimized using Stochastic Gradient Descent (SGD). Based on the TrashNet dataset containing 2,390 images across five categories (cardboard, glass, metal, paper, and plastic) the study involved rigorous hyperparameter tuning, identifying an optimal initial learning rate of 0.00075. The system was subjected to systematic hyperparameter tuning using a targeted grid search strategy to identify the optimal balance between convergence speed and stability. Key findings demonstrate a superior testing accuracy of 98.16%, an F1-Score of 0.9816, and an inference speed of 35 FPS, proving the system's readiness for real-time applications. Notably, the SGD optimizer provided better stability and generalization than AdamW on this dataset. The novelty of this study lies in being among the first to implement YOLOv11 for waste management, leveraging its new C3K2 and C2PSA blocks to achieve a 1–5% F1-Score improvement over previous YOLO versions while requiring 20% fewer parameters than YOLOv8. This improvement offers a scalable, lightweight solution for edge-computing devices, directly supporting global environmental sustainability goals. The proposed system offers a scalable and hardware-efficient solution for real-time edge deployment in smart waste management infrastructure.
Downloads
[1] Maalouf, A., & Agamuthu, P. (2023). Waste management evolution in the last five decades in developing countries – A review. Waste Management and Research, 41(9), 1420–1434. doi:10.1177/0734242X231160099.
[2] Agamuthu, P., & Babel, S. (2023). Waste management developments in the last five decades: Asian perspective. Waste Management and Research, 41(12), 1699–1716. doi:10.1177/0734242X231199938.
[3] Arumdani, I. S., Puspita, A. S., & Budihardjo, M. A. (2021). MSW handling of top 5 leading waste-producing countries in Southeast Asia. IOP Conference Series: Earth and Environmental Science, 896(1), 12003. doi:10.1088/1755-1315/896/1/012003.
[4] Alharbi, E., Alsulami, G., Aljohani, S., Alharbi, W., & Albaradei, S. (2025). Real-time detection and monitoring of public littering behavior using deep learning for a sustainable environment. Scientific Reports, 15(1), 3000. doi:10.1038/s41598-024-77118-x.
[5] Zamathula Queen Sikhakhane Nwokediegwu, Ejike David Ugwuanyi, Michael Ayorinde Dada, Michael Tega Majemite, & Alexander Obaigbena. (2024). Ai-Driven Waste Management Systems: a Comparative Review of Innovations in the USA and Africa. Engineering Science & Technology Journal, 5(2), 507–516. doi:10.51594/estj.v5i2.828.
[6] Choi, W. H., Pae, K. P., Kim, N. S., Kang, H. Y., & Hwang, Y. W. (2023). Feasibility Study of Closed-Loop Recycling for Plastic Generated from Waste Electrical and Electronic Equipment (WEEE) in South Korea. Energies, 16(17), 10 3390 16176358. doi:10.3390/en16176358.
[7] Fang, B., Yu, J., Chen, Z., Osman, A. I., Farghali, M., Ihara, I., Hamza, E. H., Rooney, D. W., & Yap, P. S. (2023). Artificial intelligence for waste management in smart cities: a review. Environmental Chemistry Letters, 21(4), 1959–1989. doi:10.1007/s10311-023-01604-3.
[8] Gunaseelan, J., Sundaram, S., & Mariyappan, B. (2023). A Design and Implementation Using an Innovative Deep-Learning Algorithm for Garbage Segregation. Sensors, 23(18), 10 3390 23187963. doi:10.3390/s23187963.
[9] Fotovvatikhah, F., Ahmedy, I., Noor, R. M., & Munir, M. U. (2025). A Systematic Review of AI-Based Techniques for Automated Waste Classification. Sensors, 25(10), 3181. doi:10.3390/s25103181.
[10] Cheng, R., He, X., Zheng, Z., & Wang, Z. (2021). Multi-scale safety helmet detection based on sas-yolov3-tiny. Applied Sciences (Switzerland), 11(8), 3652. doi:10.3390/app11083652.
[11] Rajeev, P. A., Dharewa, V., Lakshmi, D., Vishnuvarthanan, G., Giri, J., Sathish, T., & Alrashoud, M. (2025). Advancing e-waste classification with customizable YOLO based deep learning models. Scientific Reports, 15(1), 18151. doi:10.1038/s41598-025-94772-x.
[12] Li, Z., Jia, D., He, Z., & Wu, N. (2025). Multi-Frame Joint Detection Approach for Foreign Object Detection in Large-Volume Parenterals. Mathematics, 13(8), 1333. doi:10.3390/math13081333.
[13] Feng, W. Y., Zhu, Y. F., Zheng, J. T., & Wang, H. (2021). Embedded YOLO: A Real-Time Object Detector for Small Intelligent Trajectory Cars. Mathematical Problems in Engineering, 6555513. doi:10.1155/2021/6555513.
[14] Dipo, M. H., Farid, F. Al, Mahmud, M. S. Al, Momtaz, M., Rahman, S., Uddin, J., & Karim, H. A. (2025). Real-Time Waste Detection and Classification Using YOLOv12-Based Deep Learning Model. Digital, 5(2), 10 3390 5020019. doi:10.3390/digital5020019.
[15] Wan, D., Lu, R., Fang, Y., Lang, X., Shu, S., Chen, J., ... & Ye, Z. (2025). Yolov11-rgbt: Towards a comprehensive single-stage multispectral object detection framework. arXiv Preprint, arXiv:2506.14696. doi:10.48550/arXiv.2506.14696.
[16] Geetha, A. S. (2025). Yolov4: A breakthrough in real-time object detection. arXiv Preprint, arXiv:2502.04161. doi:10.48550/arXiv.2502.04161.
[17] Arishi, A. (2025). Real-Time Household Waste Detection and Classification for Sustainable Recycling: A Deep Learning Approach. Sustainability (Switzerland), 17(5), 1902. doi:10.3390/su17051902.
[18] Yasiri, J. R., Prathivi, R., & Susanto, S. (2025). Detection of Plastic Bottle Waste Using YOLO Version 5 Algorithm. Sinkron, 9(1), 20–30. doi:10.33395/sinkron.v9i1.14242.
[19] Maharana, K., Mondal, S., & Nemade, B. (2022). A review: Data pre-processing and data augmentation techniques. Global Transitions Proceedings, 3(1), 91–99. doi:10.1016/j.gltp.2022.04.020.
[20] Çetin, V., & Yıldız, O. (2022). A comprehensive review on data preprocessing techniques in data analysis. Pamukkale University Journal of Engineering Sciences, 28(2), 299–312. doi:10.5505/pajes.2021.62687.
[21] Talebi, H., & Milanfar, P. (2021). Learning to Resize Images for Computer Vision Tasks. Proceedings of the IEEE International Conference on Computer Vision, 487–496. doi:10.1109/ICCV48922.2021.00055.
[22] Khanam, R., & Hussain, M. (2024). Yolov11: An overview of the key architectural enhancements. arXiv Preprint, arXiv:2410.17725. doi:10.48550/arXiv.2410.17725.
[23] Iman, M., Arabnia, H. R., & Rasheed, K. (2023). A Review of Deep Transfer Learning and Recent Advancements. Technologies, 11(2), 40. doi:10.3390/technologies11020040.
[24] Farahani, A., Pourshojae, B., Rasheed, K., & Arabnia, H. R. (2020). A Concise Review of Transfer Learning. Proceedings - 2020 International Conference on Computational Science and Computational Intelligence, CSCI 2020, 344–351. doi:10.1109/CSCI51800.2020.00065.
[25] Hassan, E., Shams, M. Y., Hikal, N. A., & Elmougy, S. (2023). The effect of choosing optimizer algorithms to improve computer vision tasks: A comparative study. Multimedia Tools and Applications, 82(11), 16591-16633. doi:10.1007/s11042-022-13820-0.
[26] Li, Y., Chen, S. C., & Huang, S. J. (2023). Implicit Stochastic Gradient Descent for Training Physics-Informed Neural Networks. Proceedings of the 37th AAAI Conference on Artificial Intelligence, 37(7), 8692–8700. doi:10.1609/aaai.v37i7.26046.
[27] Li, D., Wu, J., & Zhang, Q. (2023). Stochastic gradient descent in the viewpoint of graduated optimization. arXiv Preprint, arXiv:2308.06775. doi:10.48550/arXiv.2308.06775.
[28] Tian, Y., Zhang, Y., & Zhang, H. (2023). Recent Advances in Stochastic Gradient Descent in Deep Learning. Mathematics, 11(3), 682. doi:10.3390/math11030682.
- This work (including HTML and PDF Files) is licensed under a Creative Commons Attribution 4.0 International License.
