E-waste Classification Using Pre-trained Deep Learning CNN Model

Electronic waste is one of the fastest-growing waste streams in the world, posing a threat to human health and the environment. E-waste classification is an essential step for adequately managing and recycling E-waste. In this research, we have proposed a method for E-waste classification using YOLOv7 as an E-waste detection model. After that, a pre-trained model is used as a classification model. First, we have used YOLOv7 to detect E-waste objects in images and crop them into separate regions. Then, we have used a pre-trained model to classify each part into one of the different categories of E-waste composition. Finally, we have got 93.5% mAP for the E-waste detection model and 99.7% accuracy for the classification of E-waste.