ViT4Waste: Vision Transformer-Based Deep Learning for Efficient E-Waste Management

Electronic waste (E-waste) includes discarded electronic devices such as batteries, televisions, PCBs, and household appliances. The improper disposal of these items releases toxic substances such as lead, mercury, and cadmium, which pose serious environmental and health hazards. Lead exposure causes neurological damage, especially in children, and can harm DNA. People near processing sites face a high risk of respiratory problems. Rapid technological advancement, frequent device upgrades, and inadequate recycling systems have intensified this issue. An effective classification of electronic waste is crucial for safe disposal. Therefore, this paper proposes a Vision Transformer (ViT)-based framework to classify E-waste by extracting meaningful features from images. The proposed framework is compared with several state-of-the-art transfer learning models for image classification, such as VGG-16, ResNet50, and Xception. The proposed framework performed better by achieving accuracies of 99.67%, 97.61%, and 98.88% in three benchmark datasets, highlighting its potential to improve the management of electronic waste.