Classification of Urban Waste Materials with Deep Learning Architectures

Six new and most preferred models from deep learning methods to separate waste materials, and training was carried out using fivefold cross-validation using the one-vs-rest method in the MobileNetV2 model to comment on the classes.