Research Article
Automated Recyclable Waste Detection using Deep Neural Networks
@INPROCEEDINGS{10.4108/eai.28-4-2025.2357782, author={Nagaleela Gangula and J S. Sukanya and Nagesh Chilamakuru and Kummara Lokeshnath and Shabana G and Ganesh G}, title={Automated Recyclable Waste Detection using Deep Neural Networks}, proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part I}, publisher={EAI}, proceedings_a={ICITSM PART I}, year={2025}, month={10}, keywords={deep learning recyclable waste detection mobilenetv2 efficientnet-b3 vision transformer (vit) image type cnn waste sorting}, doi={10.4108/eai.28-4-2025.2357782} }- Nagaleela Gangula
J S. Sukanya
Nagesh Chilamakuru
Kummara Lokeshnath
Shabana G
Ganesh G
Year: 2025
Automated Recyclable Waste Detection using Deep Neural Networks
ICITSM PART I
EAI
DOI: 10.4108/eai.28-4-2025.2357782
Abstract
Automatically classifying waste into recyclable and disposable categories is crucial for improving waste management and reducing environmental impact. This study introduces a deep learning system to classify household waste using publicly available datasets across various categories. We evaluate the performance of MobileNetV2, EfficientNet-B3, and Vision Transformer (ViT) in terms of accuracy and efficiency for image classification. To improve the model's reliability and prevent overfitting, standard preprocessing and data augmentation techniques are applied. Assessment of the reliability and performance of the system in handling waste from different categories. The findings show that these advanced neural networks can effectively identify recyclable materials and provide real-time solutions for waste sorting. This method can be integrated into smart bins, recycling centers, and mobile systems, supporting more sustainable waste management practices.
