IoT based Intelligent Object Sorting System using YOLO

The advent of Internet-of-Things technology has opened newer horizons for industrial automation. This paper addresses the traditional inefficiencies in sorting systems highly dependent on the intervention of humans, proposing an intelligent object sorting system based on IoT. It aims at the integration of IoT sensors with the YOLO Deep Learning Model to accomplish fast, error-free object sorting in parameters like size, colour, shape, and material based on the use of as little human effort as possible. Using a dataset of 25,000 annotated images, the system achieved a Mean Average Precision (MAP) of 0.87 and a classification accuracy of 0.93, outperforming traditional rule-based methods by reducing misclassification rates by up to 32%. The integration of IoT enables real-time monitoring and scalability, making the system suitable for applications in recycling, manufacturing, and logistics. This research highlights the potential of IoT and deep learning in improving productivity and reducing costs in industrial automation.