Hybrid DenseNet201-ViT Architecture for Robust Solid Waste Classification

Lailia  Rahmawati; Catur  Supriyanto

doi:10.29207/resti.v10i5.7669

Authors

Lailia Rahmawati Universitas Darul Ulum
Catur Supriyanto Universitas Dian Nuswantoro

DOI:

https://doi.org/10.29207/resti.v10i5.7669

Keywords:

CNN, deep learning, DenseNet201, hybrid architecture, waste classification

Abstract

Automatic sorting of solid waste is difficult because discarded items are often badly deformed and some material classes, such as transparent plastic and glass, look almost identical. Conventional Convolutional Neural Networks (CNNs) mainly learn local patterns; as a result, they tend to overlook global context and to overfit when only a small and unevenly distributed dataset is available. This study develops a hybrid deep learning model that pairs the dense local feature learning of DenseNet201 with the global context modeling of a Vision Transformer (ViT) in order to make waste classification more robust. An ImageNet-pretrained DenseNet201 serves as the backbone, and its high-level feature maps are converted into a sequence of patch tokens that is processed by a custom ViT encoder. Experiments were conducted on a public garbage image dataset with six unevenly represented classes, namely cardboard, glass, metal, paper, plastic, and trash. On-the-fly data augmentation and early stopping were employed during training to limit overfitting. The hybrid model reached an overall accuracy of 87% and a macro F1-score of 0.85, exceeding the CNN baselines trained under identical settings. It was especially accurate on fibrous materials, with precision values of 0.95 for cardboard and 0.89 for paper. These results indicate that coupling self-attention with densely connected convolutional layers alleviates the main weaknesses of pure CNNs and provides an accurate, interpretable, and computationally efficient basis for practical automated waste sorting systems.

Downloads

Download data is not yet available.

References

[1] N. A. Alsadhan, "DeepWasteSort-SI-SSO: A Vision Transformer-Based Waste Image Classification Framework Optimized with Self Improved Sparrow Search Optimizer," Sustainability, vol. 18, no. 4, p. 2080, 2026, doi: 10.3390/su18042080.

[2] F. Y. Manik et al., "Explainable CNN-Based Multiclass Household Waste Classification Using Grad-CAM for Smart Waste Management," Int. J. Adv. Comput. Sci. Appl., vol. 17, no. 1, 2026, doi: 10.14569/IJACSA.2026.0170155.

[3] C. Dewi, R. A. Susanto, H. P. Chernovita, M. V. V. Santoso, and A. P. S. Chen, "Evaluation of Garbage Detection Efficacy Utilising Deep Learning," J. Inf. Commun. Technol., vol. 24, no. 4, pp. 36–57, Oct. 2025, doi: 10.32890/jict2025.24.4.2.

[4] N. S. Kolla, M. Anumula, S. Sujana, and M. Ratnababu, "Road Garbage Classification Using ResNet50," in 2023 Int. Conf. Signal Process. Comput. Electron. Power Telecommun. (IConSCEPT), May 2023, pp. 1–5, doi: 10.1109/IConSCEPT57958.2023.10170073

[5] N. Nnamoko, J. Barrowclough, and J. Procter, "Solid Waste Image Classification Using Deep Convolutional Neural Network," Infrastructures, vol. 7, no. 4, p. 47, 2022, doi: 10.3390/infrastructures7040047.

[6] K. Liu and X. Liu, "Recycling Material Classification using Convolutional Neural Networks," in 2022 21st IEEE Int. Conf. Mach. Learn. Appl. (ICMLA), Nassau, Bahamas, 2022, pp. 83-88, doi: 10.1109/ICMLA55696.2022.00019

[7] U. L. M. Rijah and P. K. W. Abeygunawardhana, "Smart Waste Segregation for Home Environment," in 2023 3rd Int. Conf. Adv. Res. Comput. (ICARC), 2023, pp. 184-189, doi: 10.1109/ICARC57651.2023.10145659

[8] L. Yong, L. Ma, D. Sun, and L. Du, "Application of MobileNetV2 to waste classification," PLoS ONE, vol. 18, no. 3, 2023, doi: 10.1371/journal.pone.0282336

[9] T. Tabata, "Environmental and Social Impacts of Community-Based Household Plastic Waste Collection for High-Value Recycling," Appl. Sci., vol. 16, no. 3, p. 1326, 2026, doi: 10.3390/app16031326.

[10] N. Umar, B. Asrul, and Y. Yuyun, "Type Deep Learning Model for Multi-Label Waste Classification in Canal Environments: A Comparative Study with CNN Architectures," J. Appl. Data Sci., vol. 7, no. 1, pp. 261-276, 2025, doi: 10.47738/jads.v7i1.1066.

[11] A. Alkılınç, F. Y. Okay, İ. Kök, and S. Özdemir, "Deep Ensemble Learning Model for Waste Classification Systems," Sustainability, vol. 18, no. 1, p. 24, 2026, doi: 10.3390/su18010024.

[12] D. Kumar et al., "Optimizing Litter Detection in Images: An Integration of Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs)," IEEE Access, vol. 14, pp. 3325-3344, 2026, doi: 10.1109/ACCESS.2025.3646075.

[13] Z. Lv, Y. Diao, C. Zeng, W. Wang, and S. An, "Research on Encrypted Transmission and Recognition of Garbage Images in Low-Illumination Environments," Electronics, vol. 15, no. 2, p. 302, 2026, doi: 10.3390/electronics15020302.

[14] M. Tarequzzaman, M. Akash, Z. Nayon, M. S. Reza, and S. Haque, "Enhancing waste management through municipal solid waste classification: a convolutional neural network approach," IAES Int. J. Artif. Intell., vol. 14, no. 6, pp. 4775-4786, 2025, doi: 10.11591/ijai.v14.i6.pp4775-4786.

[15] H. C. Yoong, S. L. Jong, and K. G. Tay, "Waste Classification Using NasNet-Mobile: A Multi-Stage Deep Learning Approach for Environmental Sustainability," Indones. J. Electr. Eng. Informatics, vol. 13, no. 4, pp. 952-967, Dec. 2025, doi: 10.52549/ijeei.v13i4.7198.

[16] H. T. N. Le and H. Q. T. Ngo, "Application of the vision-based deep learning technique for waste classification using the robotic manipulation system," Int. J. Cogn. Comput. Eng., vol. 6, pp. 391-400, 2025, doi: 10.1016/j.ijcce.2025.02.005.

[17] J. Cui et al., "Research on optimization of waste sorting and transportation network in smart cities based on garbage volume prediction," Discov. Comput., vol. 28, p. 54, 2025, doi: 10.1007/s10791-025-09537-x.

[18] Cchang, "Garbage Classification," Kaggle, 2018, doi: 10.34740/Kaggle/DS/81794.

[19] I. Wiguna, I. Desnanjaya, and I. Sandika, "Performance evaluation of pre-trained deep learning model on garbage classification with data augmentation approach," IAES Int. J. Artif. Intell., vol. 14, no. 6, pp. 4971-4981, 2025, doi: 10.11591/ijai.v14.i6.pp4971-4981.

[20] V. Lingamuthu et al., "ARCNet: A Novel Deep Learning Model for Robust Solid Waste Image Classification in Waste Management Systems," Int. Arab J. Inf. Technol., vol. 22, no. 6, Nov. 2025, doi: 10.34028/iajit/22/6/6.

[21] S. Shetty, S. Kallianpur, R. Fernandes, A. P. Rodrigues, and V. Padmanabha, "ECO-HYBRID: Sustainable Waste Classification Using Transfer Learning with Hybrid and Enhanced CNN Models," Sustainability, vol. 17, no. 19, p. 8761, 2025, doi: 10.3390/su17198761.

[22] G. Celik, "Multi-layer feature fusion for high-accuracy solid waste classification using a hybrid deep learning model," Vis. Comput., vol. 41, pp. 10227–10249, 2025, doi: 10.1007/s00371-025-04031-3.

[23] R. G. Tiwari and V. K. Shukla, "Hybrid Squeeze-and-Excitation Convolutional Neural Network with Elastic Weight Consolidation for Longitudinal Learning in High-Accuracy Waste Classification," J. Online Inform., vol. 10, no. 2, pp. 395-406, Dec. 2025, doi: 10.15575/join.v10i2.1628.

[24] V. S. Malagitti, M. C. Elemmi, S. Jagadeesha, and P. Kiran, "Cognitive Approach and Modified Vision Transformer for Improved E-Waste Management System," Int. J. Intell. Eng. Syst., vol. 18, no. 6, 2025, doi: 10.22266/ijies2025.0731.35.

[25] S. Potharaju et al., "Optimizing Waste Management with Squeeze-and-Excitation and Convolutional Block Attention Integration in ResNet-Based Deep Learning Frameworks," J. Artif. Intell. Technol., vol. 5, pp. 211–220, 2025, doi: 10.37965/jait.2025.0709.

[26] I. Dawar et al., "A systematic literature review on municipal solid waste management using machine learning and deep learning," Artif. Intell. Rev., vol. 58, p. 183, 2025, doi: 10.1007/s10462-025-11196-9.

[27] O. Chacón-Albero et al., "AI for Sustainable Recycling: Efficient Model Optimization for Waste Classification Systems," Sensors, vol. 25, no. 12, p. 3807, 2025, doi: 10.3390/s25123807.

[28] H. Zhou et al., "Optimizing Backbone Networks Through Hybrid–Modal Fusion: A New Strategy for Waste Classification," Sensors, vol. 25, no. 10, p. 3241, 2025, doi: 10.3390/s25103241.

[29] M. Castro-Bello et al., "Convolutional Neural Network Models in Municipal Solid Waste Classification: Towards Sustainable Management," Sustainability, vol. 17, no. 8, p. 3523, 2025, doi: 10.3390/su17083523.

[30] M. Castro-Bello et al., "R3sNet: Optimized Residual Neural Network Architecture for the Classification of Urban Solid Waste via Images," Sustainability, vol. 17, no. 8, p. 3502, 2025, doi: 10.3390/su17083502.

[31] R. Risfendra, G. F. Ananda, and H. Setyawan, "Deep Learning-Based Waste Classification with Transfer Learning Using EfficientNet-B0 Model," Jurnal RESTI (Rekayasa Sistem Dan Teknologi Informasi), vol. 8, no. 4, pp. 535-541, 2024, doi: 10.29207/resti.v8i4.5875.

[32] A. Afifah, E. R. Arumi, M. Maimunah, and S. Nugroho, "Comparative Evaluation of Preprocessing Methods for MobileNetV1 and V2 in Waste Classification," Jurnal RESTI (Rekayasa Sistem Dan Teknologi Informasi), vol. 9, no. 3, pp. 444-452, 2025, doi: 10.29207/resti.v9i3.6211.

[33] M. Nahiduzzaman et al., "An automated waste classification system using deep learning techniques: Toward efficient waste recycling and environmental sustainability," Knowledge-Based Syst., vol. 310, p. 113028, 2025, doi: 10.1016/j.knosys.2025.113028.

[34] F. R. Sayem et al., "Enhancing waste sorting and recycling efficiency: robust deep learning-based approach for classification and detection," Neural Comput. Appl., vol. 37, pp. 4567–4583, 2025, doi: 10.1007/s00521-024-10855-2.

[35] A. Pandey et al., "Deep Learning Enabled Garbage Classification and Detection by Visual Context for Aerial Images," Appl. Comput. Intell. Soft Comput., vol. 2025, p. 9106130, 2025, doi: 10.1155/acis/9106130.

[36] M. Chhabra et al., "Intelligent waste classification approach based on improved multi-layered convolutional neural network," Multimedia Tools Appl., vol. 83, pp. 84095–84120, 2024, doi: 10.1007/s11042-024-18939-w.

[37] V. K. Yogamadhavan and G. Mannayee, "Evaluation of Performance of Different Optimizers of Convolutional Neural Network in the Classification of Images of Urban Domestic Solid Waste," J. Internet Serv. Inf. Secur., vol. 14, no. 3, pp. 42-77, Aug. 2024, doi: 10.58346/JISIS.2024.I3.004.

[38] C. J. Yi and C. F. Kim, "AI-Powered Waste Classification Using Convolutional Neural Networks (CNNs)," Int. J. Adv. Comput. Sci. Appl., vol. 15, no. 10, 2024, doi: 10.14569/IJACSA.2024.0151009.

[39] H. Santoso, I. Hanif, H. Magdalena, and A. Afiyati, "A Hybrid Model for Dry Waste Classification using Transfer Learning and Dimensionality Reduction," Int. J. Informatics Vis., vol. 8, no. 2, pp. 623-634, May 2024, doi: 10.62527/joiv.8.2.1943.

[40] R. Singh et al., "Performance Analysis of Garbage Classification Using Balanced and Unbalanced Dataset with EfficientNetV2B1 Architecture," J. Comput. Sci., vol. 20, no. 10, pp. 1310-1321, 2024, doi: 10.3844/jcssp.2024.1310.1321.

[41] Sunardi, A. Yudhana, and M. Fahmi, "Improving waste classification using convolutional neural networks: An application of machine learning for effective environmental management," Rev. d'Intelligence Artificielle, vol. 37, no. 4, pp. 845-855, 2023, doi: 10.18280/ria.370404.