IoT-Enabled Speed and Accident Detection Platform

The IoT-Enabled Speed and Accident Detection Platform is an intelligent traffic monitoring system designed to enhance road safety through automated vehicle detection, speed estimation, and overspeeding identification using video analytics. With the increasing number of vehicles and rising traffic accidents, especially due to overspeeding, there is a strong need for accurate, real-time, and cost-effective traffic surveillance systems. This platform employs deep learning–based computer vision techniques, utilizing YOLOv8 for real-time vehicle detection and ByteTrack for robust multi-object tracking across consecutive video frames. A calibration-based speed estimation method is applied to convert pixel displacement into real-world distance, enabling precise speed calculation for each tracked vehicle. The system also incorporates line-crossing logic for vehicle counting and generates alerts when predefined speed limits are exceeded. Developed using Python, OpenCV, and the Ultralytics framework, the system processes prerecorded traffic video streams and visualizes detection results, including bounding boxes, vehicle IDs, speed values, and overspeed warnings, in real time. The modular architecture supports future integration with IoT devices, cloud platforms, and automated accident detection and notification systems. Experimental results demonstrate reliable vehicle detection, consistent tracking, and effective speed estimation under varying traffic conditions. The proposed platform offers a scalable and cost-effective solution for intelligent transportation systems, smart city applications, and improved road safety management.