WasteSegNet: A Deep Learning Approach for Smart Waste Segregation in Urban Environments

Efficient waste segregation is paramount for sustainable urban waste management. In this study, we present WasteSegNet, a custom-designed Convolutional Neural Network (CNN) for automated waste segregation into Organic and Recyclable categories. Trained on 22,564 waste images with a test set of 2,513 samples, WasteSegNet achieves an impressive accuracy of 87.5% on the test dataset. Comparative analysis with traditional methods highlights the superior performance of WasteSegNet in accurate waste item classification. The model's strength lies in its ability to discern intricate visual patterns, enabling effective segregation. By harnessing deep learning techniques, WasteSegNet offers a promising solution to optimize waste sorting processes, reduce landfill waste, combat environmental pollution, and foster sustainable waste management practices in smart cities.