A machine vision based autonomous waste managment system using deep learning

Rapid urbanization has intensified the challenges of waste management, as traditional garbage collection methods remain inefficient, labor-intensive, and prone to improper segregation. This project proposes an AI-based Garbage Sorting and Monitoring System that uses the YOLOv5n object detection model to automatically classify waste into categories such as plastic, paper, glass, metal, organic, and e-waste. The lightweight YOLO architecture enables real-time detection with minimal computational resources, supported by a custom annotated waste image dataset. Detected data is visualized through an analytics dashboard, offering real-time insights into waste types, frequency, and recycling potential to assist smart city decision-making. The system is scalable and can be integrated with IoT-based smart bins or robotic collectors, contributing to automated, data-driven, and sustainable waste management solutions.