Leveraging Deep Learning and Convolutional Neural Network for Digital Waste Image Classification
Waste management is a significant environmental challenge, with 14 million tons of waste uncollected in 2023, accumulating at landfill sites. The mixing of household and industrial waste complicates adequate segregation. Baseline sorting techniques are costly in labor, time, and resources. This pape...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
EDP Sciences
2025-01-01
|
| Series: | E3S Web of Conferences |
| Online Access: | https://www.e3s-conferences.org/articles/e3sconf/pdf/2025/22/e3sconf_interconnects2025_03009.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849738858970742784 |
|---|---|
| author | Fauzan Naufal Anis Sukmasetya Pristi Nuryanto Nuryanto |
| author_facet | Fauzan Naufal Anis Sukmasetya Pristi Nuryanto Nuryanto |
| author_sort | Fauzan Naufal Anis |
| collection | DOAJ |
| description | Waste management is a significant environmental challenge, with 14 million tons of waste uncollected in 2023, accumulating at landfill sites. The mixing of household and industrial waste complicates adequate segregation. Baseline sorting techniques are costly in labor, time, and resources. This paper looks at the application of CNNs for the automated sorting of wastes to enhance speed, capacity, and precision. CNNs can be used to classify digital images of waste to minimize the manual work in waste management, especially on a large scale. The research evaluates two CNN architectures—ResNet101 and a custom-designed CNN model—aiming to enhance waste classification. The dataset includes images of 12 different waste categories: battery, biological waste, cardboard, clothes, green glass, metal, paper, plastic, shoes, trash, and white glass. The performance of models was measured by metrics such as accuracy, precision, recall, and F1-score during the model training and assessment periods. The fine-tuned ResNet101 model offered a validation accuracy of 97.74% and classification accuracy of 74.21%, though with lesser accuracy when compared to the custom model, which had 90.31% validation accuracy and classification accuracy of 52.5%. Such results have shown the ability of deep learning models, especially CNNs, to enhance waste classification accuracy. The study, therefore, provides perceptions of improved efficiency in waste management. Thus, it is recommended that future work improve these models, include more data, and explore more practical applications to encourage appropriate waste management and demarcate the impact on the environment. |
| format | Article |
| id | doaj-art-562ce79c9ecd422980d8b348b8d1998f |
| institution | DOAJ |
| issn | 2267-1242 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | EDP Sciences |
| record_format | Article |
| series | E3S Web of Conferences |
| spelling | doaj-art-562ce79c9ecd422980d8b348b8d1998f2025-08-20T03:06:25ZengEDP SciencesE3S Web of Conferences2267-12422025-01-016220300910.1051/e3sconf/202562203009e3sconf_interconnects2025_03009Leveraging Deep Learning and Convolutional Neural Network for Digital Waste Image ClassificationFauzan Naufal Anis0Sukmasetya Pristi1Nuryanto Nuryanto2Universitas Muhammadiyah MagelangUniversitas Muhammadiyah MagelangUniversitas Muhammadiyah MagelangWaste management is a significant environmental challenge, with 14 million tons of waste uncollected in 2023, accumulating at landfill sites. The mixing of household and industrial waste complicates adequate segregation. Baseline sorting techniques are costly in labor, time, and resources. This paper looks at the application of CNNs for the automated sorting of wastes to enhance speed, capacity, and precision. CNNs can be used to classify digital images of waste to minimize the manual work in waste management, especially on a large scale. The research evaluates two CNN architectures—ResNet101 and a custom-designed CNN model—aiming to enhance waste classification. The dataset includes images of 12 different waste categories: battery, biological waste, cardboard, clothes, green glass, metal, paper, plastic, shoes, trash, and white glass. The performance of models was measured by metrics such as accuracy, precision, recall, and F1-score during the model training and assessment periods. The fine-tuned ResNet101 model offered a validation accuracy of 97.74% and classification accuracy of 74.21%, though with lesser accuracy when compared to the custom model, which had 90.31% validation accuracy and classification accuracy of 52.5%. Such results have shown the ability of deep learning models, especially CNNs, to enhance waste classification accuracy. The study, therefore, provides perceptions of improved efficiency in waste management. Thus, it is recommended that future work improve these models, include more data, and explore more practical applications to encourage appropriate waste management and demarcate the impact on the environment.https://www.e3s-conferences.org/articles/e3sconf/pdf/2025/22/e3sconf_interconnects2025_03009.pdf |
| spellingShingle | Fauzan Naufal Anis Sukmasetya Pristi Nuryanto Nuryanto Leveraging Deep Learning and Convolutional Neural Network for Digital Waste Image Classification E3S Web of Conferences |
| title | Leveraging Deep Learning and Convolutional Neural Network for Digital Waste Image Classification |
| title_full | Leveraging Deep Learning and Convolutional Neural Network for Digital Waste Image Classification |
| title_fullStr | Leveraging Deep Learning and Convolutional Neural Network for Digital Waste Image Classification |
| title_full_unstemmed | Leveraging Deep Learning and Convolutional Neural Network for Digital Waste Image Classification |
| title_short | Leveraging Deep Learning and Convolutional Neural Network for Digital Waste Image Classification |
| title_sort | leveraging deep learning and convolutional neural network for digital waste image classification |
| url | https://www.e3s-conferences.org/articles/e3sconf/pdf/2025/22/e3sconf_interconnects2025_03009.pdf |
| work_keys_str_mv | AT fauzannaufalanis leveragingdeeplearningandconvolutionalneuralnetworkfordigitalwasteimageclassification AT sukmasetyapristi leveragingdeeplearningandconvolutionalneuralnetworkfordigitalwasteimageclassification AT nuryantonuryanto leveragingdeeplearningandconvolutionalneuralnetworkfordigitalwasteimageclassification |