Current - Issue

Road accidents involving two-wheelers are increasing rapidly due to unsafe riding practices such as riding without helmets, mobile phone usage while driving, and unsafe passenger seating. To address these challenges, this paper presents a vision-based real-time helmet recognition and automated bike ignition control system using the YOLOv8 deep learning model. The proposed system integrates Artificial Intelligence (AI), Internet of Things (IoT), and embedded technologies to improve rider safety through intelligent automation. A WiFi camera captures real-time video streams and performs helmet detection, mobile phone usage detection, and child sitting-in-front detection using the YOLOv8 object detection model. Additional safety verification is carried out using buckle detection, alcohol sensing, fall detection, and weather monitoring modules. The ignition system is enabled only when all safety conditions are satisfied; otherwise, ignition remains disabled. In case of violations or accidents, alerts and system status notifications are transmitted through Telegram communication. Experimental results demonstrate high detection accuracy with reliable real-time performance suitable for practical deployment. The proposed system provides a low-cost, scalable, and intelligent solution for enhancing two-wheeler road safety and reducing accident risks through AI-powered monitoring and automated ignition control.