Real-time computer vision system for monitoring conveying systems

Conveyors play a crucial role in transporting packages and containers in manufacturing and production facilities. While computer vision has emerged as a promising technology for real-time monitoring of transportation systems, its application in conveyor operations remains in the early stages. This paper introduces an Industrial Internet of Things (IIoT) framework for real-time conveyor monitoring. We first evaluate state-of-the-art object detection algorithms (YOLO10, YOLO12, RTMDet, and RT-DETER) using different model variants for identifying diverse package types including boxes, envelopes, trays, and totes. Our analysis reveals that RTMDet-t achieves a balanced trade-off between accuracy and speed, with an F1 score of 86% at 10 frames per second, while YOLO10-n and YOLO12-n models deliver higher speeds up to 27 frames per second with reduced accuracy. We then introduce novel algorithms for conveyor extraction, applicable to a range of conveyor monitoring applications. To demonstrate the framework's robustness, we implement the proposed system on two use-cases: jam detection and throughput monitoring, achieving an F1 score of 97% in throughput monitoring and 84% in jam detection on real-life test datasets, validating its potential for diverse industrial applications.