Automated Recyclable Object Classification using VGG-16 with Channel Attention Enhancement

This research paper looks primarily into the deep learning models, too: Convolutional Neural Networks, MobileNetV2, and VGG-16, helping fully automate the classification of recyclable materials with a view to enhance the efficiency and accuracy of waste sorting processes. A Channel Attention Layer has been added to the VGG-16 architecture in order to enhance the performance of classification of different materials like plastic, metal, and paper. The proposed models are validated on accuracy, precision, recall, and F1-score. Results show that VGG-16 with Channel Attention is the best model with a great accuracy score of 99.38%. This depicts the volumes of their capability in detecting and classifying recyclable materials, making it a potent machine in automated waste management. Findings show how deep learning will now help in dealing with the inefficiencies of conventional waste sorting and encourage a lot more sustainable recycling practices, giving massive contributions to environmental conservancy and resources conservation.