Deep Learning-Based Approaches for Electronic Waste Detection and Classification: A Step Toward Sustainable Recycling

Abstract

The proper management of electronic waste (e-waste) is of critical importance for environmental and economic sustainability. Classifying waste according to its type is considered one of the effective solutions to mitigate the impacts of environmental pollution and ensure a sustainable standard of living. Incorporating sorted waste into recycling processes enables the conservation of natural resources and their reintegration into the production cycle. Traditional waste management methods are generally based on manual classification and physical detection technologies. In the manual classification of e-waste, the type of waste is determined by its physical properties or chemical analysis. These methods are time-consuming, costly, and prone to human error. The limitations of traditional waste management methods necessitate the use of deep learning-based approaches. In this study, a deep learning-based method is proposed to detect the types of e-waste. The study employs three different feature extraction architectures (EfficientNetB0, Inceptionv3, and AlexNet) and five different classification algorithms to identify e-waste types. System performance was evaluated using metrics such as Accuracy, Precision, Recall, and F1-Score. Experimental results indicate that the highest performance was achieved with the EfficientNetB0+SVM, Inceptionv3+SVM, and Inceptionv3+LR combinations, with the highest accuracy recorded at 97%. The results are presented in tabular form, demonstrating that deep learning-based approaches offer an effective solution for e-waste management. This study highlights the innovative and practical potential of deep learning in the detection and classification of e-waste.