Efficient Source-Level Waste Segregation Using CNN-Based Waste Classifier with IoT-Enabled Smart Bins and Mobile Application

Image processing and deep learning techniques have demonstrated their efficacy as valuable tools for classifying municipal solid waste. This study presents a comparative review of the recent advancement in waste classification techniques and their significance. Further, an attempt has been made to prepare a reliable dataset by integrating datasets from the public domain and web images. Architectures consisting of pre-trained models, namely Inception V3, ResNet, VGG19, DenseNet, and Xception, based on Convolution Neural Networks, were considered. In that process, it was observed that models Inception V3 and Xception were performing exceptionally well. Further, both the models were fine-tuned through hyperparameter tuning, and Inception V3 and Xception yielded 93% and 94.4% accuracy, respectively. The identified best model was further deployed on the mobile app for waste classification, and the advanced bin monitoring unit (ABMU) was developed for smart bins to facilitate source-level segregation of waste. Both these systems will guide the user in the scientific disposal of waste and improve people's awareness of efficient disposal practices according to prevailing guidelines.