MSWNet: A visual deep machine learning method adopting transfer learning based upon ResNet 50 for municipal solid waste sorting

Abstract

● MSWNet was proposed to classify municipal solid waste.

● Transfer learning could promote the performance of MSWNet.

● Cyclical learning rate was adopted to quickly tune hyperparameters.

An intelligent and efficient methodology is needed owning to the continuous increase of global municipal solid waste (MSW). This is because the common methods of manual and semi-mechanical screenings not only consume large amount of manpower and material resources but also accelerate virus community transmission. As the categories of MSW are diverse considering their compositions, chemical reactions, and processing procedures, etc., resulting in low efficiencies in MSW sorting using the traditional methods. Deep machine learning can help MSW sorting becoming into a smarter and more efficient mode. This study for the first time applied MSWNet in MSW sorting, a ResNet-50 with transfer learning. The method of cyclical learning rate was taken to avoid blind finding, and tests were repeated until accidentally encountering a good value. Measures of visualization were also considered to make the MSWNet model more transparent and accountable. Results showed transfer learning enhanced the efficiency of training time (from 741 s to 598.5 s), and improved the accuracy of recognition performance (from 88.50% to 93.50%); MSWNet showed a better performance in MSW classsification in terms of sensitivity (93.50%), precision (93.40%), F1-score (93.40%), accuracy (93.50%) and AUC (92.00%). The findings of this study can be taken as a reference for building the model MSW classification by deep learning, quantifying a suitable learning rate, and changing the data from high dimensions to two dimensions.

Graphical abstract

Keywords

Municipal solid waste sorting / Deep residual network / Transfer learning / Cyclic learning rate / Visualization

References

[1] Alom M Z, Taha T M, Yakopcic C, Westberg S, Sidike P, Nasrin M S, Hasan M, Van Essen B C. (2019). A state-of-the-art survey on deep learning theory and architectures. Electronics (Basel), 8(3): 292–358

[2] Ding Y, Zhao J, Liu J W, Zhou J, Cheng L, Zhao J, Shao Z, Iris Ç, Pan B, Li X, Hu Z T. (2021). A review of China’s municipal solid waste (MSW) and comparison with international regions: Management and technologies in treatment and resource utilization. Journal of Cleaner Production, 293: 126144

[3] Fu B, Li S, Wei J, Li Q, Wang Q, Tu J. (2021). A novel intelligent garbage classification system based on deep learning and an embedded linux system. IEEE Access: Practical Innovations, Open Solutions, 9: 131134–131146

[4] Gundupalli S P, Hait S, Thakur A. (2017). A review on automated sorting of source-separated municipal solid waste for recycling. Waste Management (New York, N.Y.), 60: 56–74

[5] Guo Y, Zhu Z, Zhao Y, Zhou T, Lan B, Song L. (2021). Simultaneous annihilation of microorganisms and volatile organic compounds from municipal solid waste storage rooms with slightly acidic electrolyzed water. Journal of Environmental Management, 297: 113414

[6] He K, Zhang X, Ren S, Sun J. (2015). Delving deep into rectifiers: Surpassing human-level performance on ImageNet classification. Santiago, Chile: IEEE, 1026–1034

[7] He K, Zhang X, Ren S, Sun J. (2016). Deep residual learning for image recognition. Las Vegas, NV, USA: CVPR, 770–778

[8] Huang Y, Chen J, Duan Q, Feng Y, Luo R, Wang W, Liu F, Bi S, Lee J. (2022). A fast antibiotic detection method for simplified pretreatment through spectra-based machine learning. Frontiers of Environmental Science & Engieering, 16(3): 38

[9] Kaza S, Yao L, Bhada-Tata P, Van Woerden F. (2018). What a waste 2.0: A global snapshot of solid waste management to 2050. Washington DC: World Bank Publications, 1–295

[10] Kwan C, Chou B, Yang J, Rangamani A, Tran T, Zhang J, Etienne-Cummings R. (2019). Deep learning-based target tracking and classification for low quality videos using coded aperture cameras. Sensors (Basel), 19(17): 3702–3734

[11] Leslie N S. (2017). Cyclical learning rates for training neural networks. Santa Rosa, CA, USA: IEEE, 464–472

[12] Li J, Pan L, Suvarna M, Wang X. (2021a). Machine learning aided supercritical water gasification for H₂-rich syngas production with process optimization and catalyst screening. Chemical Engineering Journal, 426: 131285

[13] Li J, Suvarna M, Li L, Pan L, Pérez-Ramírez J, Ok Y S, Wang X. (2022a). A review of computational modeling techniques for wet waste valorization: Research trends and future perspectives. Journal of Cleaner Production, 367: 133025

[14] Li J, Zhang L, Li C, Tian H, Ning J, Zhang J, Tong Y W, Wang X. (2022b). Data-driven based in-depth interpretation and inverse design of anaerobic digestion for CH₄-rich biogas production. ACS ES&T Engineering, 2(4): 642–652

[15] Li Z, Liu F, Yang W, Peng S, Zhou J. (2021b). A survey of convolutional neural networks: analysis, applications, and prospects. IEEE Transactions on Neural Networks and Learning Systems, 2(1): 1–21

[16] Liang Y, Song Q, Wu N, Li J, Zhong Y, Zeng W. (2021). Repercussions of COVID-19 pandemic on solid waste generation and management strategies. Frontiers of Environmental Science & Engieering, 15(6): 115

[17] Lin K, Zhao Y, Gao X, Zhang M, Zhao C, Peng L, Zhang Q, Zhou T. (2022a). Applying a deep residual network coupling with transfer learning for recyclable waste sorting. Environmental Science and Pollution Research International, 10(2): 1–15

[18] Lin K, Zhao Y, Kuo J H. (2022c). Deep learning hybrid predictions for the amount of municipal solid waste: a case study in Shanghai. Chemosphere, 307(4): 136119

[19] Lin K, Zhao Y, Kuo J H, Deng H, Cui F, Zhang Z, Zhang M, Zhao C, Gao X, Zhou T, Wang T. (2022b). Toward smarter management and recovery of municipal solid waste: a critical review on deep learning approaches. Journal of Cleaner Production, 346: 130943

[20] Lin K, Zhao Y, Tian L, Zhao C, Zhang M, Zhou T. (2021). Estimation of municipal solid waste amount based on one-dimension convolutional neural network and long short-term memory with attention mechanism model: a case study of Shanghai. Science of the Total Environment, 791: 148088

[21] Liu J, Yue P, Zang N, Lu C, Chen X. (2021). Removal of odors and VOCs in municipal solid waste comprehensive treatment plants using a novel three-stage integrated biofilter: Performance and bioaerosol emissions. Frontiers of Environmental Science & Engieering, 15(3): 48

[22] Long H, Liao Y, Cui C, Liu M, Liu Z, Li L, Hu W, Yan D. (2022). Assessment of popular techniques for co-processing municipal solid waste in Chinese cement kilns. Frontiers of Environmental Science & Engieering, 16(4): 51

[23] Lu J W, Zhang S, Hai J, Lei M. (2017). Status and perspectives of municipal solid waste incineration in China: a comparison with developed regions. Waste Management (New York, N.Y.), 69: 170–186

[24] Lu W, Huo W, Gulina H, Pan C. (2022). Development of machine learning multi-city model for municipal solid waste generation prediction. Frontiers of Environmental Science & Engineering, 16(9): 119

[25] Miraei Ashtiani S H, Javanmardi S, Jahanbanifard M, Martynenko A, Verbeek F J. (2021). Detection of mulberry ripeness stages using deep learning models. IEEE Access: Practical Innovations, Open Solutions, 9: 100380–100394

[26] Nie Y, Wu Y, Zhao J, Zhao J, Chen X, Maraseni T, Qian G. (2018). Is the finer the better for municipal solid waste (MSW) classification in view of recyclable constituents? A comprehensive social, economic and environmental analysis. Waste Management (New York, N.Y.), 79: 472–480

[27] Özkan K, Ergin S, Isik S, Isikli I. (2015). A new classification scheme of plastic wastes based upon recycling labels. Waste Management (New York, N.Y.), 35: 29–35

[28] Serranti S, Gargiulo A, Bonifazi G. (2012). Classification of polyolefins from building and construction waste using NIR hyperspectral imaging system. Resources, Conservation and Recycling, 61: 52–58

[29] Shin H C, Roth H R, Gao M, Lu L, Xu Z, Nogues I, Yao J, Mollura D, Summers R M. (2016). Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning. IEEE Transactions on Medical Imaging, 35(5): 1285–1298

[30] Tong Q, Liang G, Bi J. (2022). Calibrating the adaptive learning rate to improve convergence of ADAM. Neurocomputing, 481: 333–356

[31] Vaverková M D, Paleologos E K, Dominijanni A, Koda E, Tang C S, Małgorzata W, Li Q, Guarena N, Mohamed A M O, Vieira C S. . (2021). Municipal solid waste management under Covid-19: challenges and recommendations. Environmental Geotechnics, 8(3): 217–232

[32] Wang C, Chu Z, Gu W. (2021a). Participate or not: impact of information intervention on residents’ willingness of sorting municipal solid waste. Journal of Cleaner Production, 318: 128591

[33] Wang Y, Shi Y, Zhou J, Zhao J, Maraseni T, Qian G. (2021b). Implementation effect of municipal solid waste mandatory sorting policy in Shanghai. Journal of Environmental Management, 298: 113512

[34] Wei J, Li H, Liu J. (2022). Curbing dioxin emissions from municipal solid waste incineration: China’s action and global share. Journal of Hazardous Materials, 435: 129076

[35] Wen X, Luo Q, Hu H, Wang N, Chen Y, Jin J, Hao Y, Xu G, Li F, Fang W. (2014). Comparison research on waste classification between China and the EU, Japan, and the USA. Journal of Material Cycles and Waste Management, 16(2): 321–334

[36] Wu J, Zhou X, Yan X, Wang F, Bai X, Li Y, Wang Y, Zhou J. (2016). Effects and improvement suggestions of green account system for waste classification and reduction in Shanghai. Journal of Shanghai University (Natural Science Edition), 22(2): 197–202

[37] Yan B, Liang R, Li B, Tao J, Chen G, Cheng Z, Zhu Z, Li X. (2021). Fast identification and characterization of residual wastes via laser-induced breakdown spectroscopy and machine learning. Resources, Conservation and Recycling, 174: 105851

[38] Zeiler M D, Fergus R. (2014). Visual and understanding convolutional networks. In: Proceedings of the 13th European Conference on Computer Vision (ECCV 2014), Zurich, Switzerland. Heidelberg: Springer, 8689: 818–883

[39] Zeiler M D, Taylor G W, Fergus R. (2011). Adaptive deconvolutional networks for mid and high level feature learning. Barcelona, Spain: IEEE, 2018–2025

[40] Zhang J, Zhang Z, Zhang J, Fan G, Wu D. (2021). A quantitative study on the benefit of various waste classifications. Advances in Civil Engineering, 2021: 1–15

RIGHTS & PERMISSIONS

Higher Education Press