Intelligent Traffic Management using YOLOv8 Object Detection & Pedestrian Safety Integration with IoT

Abstract

With the rapid growth of cities and the increasing number of vehicles on the road, managing traffic efficiently has become a major challenge. Most traditional traffic signal systems still operate using fixed time intervals, regardless of the actual traffic conditions. As a result, vehicles often experience unnecessary delays and congestion, especially during peak hours. These systems are also not effective at responding to changing road situations or reducing traffic violations. This not only affects traffic flow but also increases safety risks for all road users, particularly pedestrians, who are among the most vulnerable on busy roads. To address these systemic shortcomings, this research introduces an integrated, multi-functional traffic management framework that converges sophisticated computer vision with real-time embedded hardware actuation. Instead of treating traffic optimization, safety enforcement, and pedestrian assistance as disparate challenges, this system employs a custom-trained YOLOv8 object detection model as a centralized intelligence engine. The system focuses on four main functions to improve traffic management. It gives priority to emergency vehicles such as ambulances and fire trucks, automatically checks whether motorcycle riders are wearing helmets, adjusts traffic signal timings based on the traffic density in each lane, and provides voice-guided crossing assistance for visually impaired pedestrians. By using live camera feeds and an IoT-based control system, the project is able to monitor traffic conditions in real time and respond accordingly. Unlike traditional traffic systems that operate on fixed signal timings, this approach adapts to changing road conditions, helping to reduce congestion and improve overall traffic flow. Testing showed that the system reduced average waiting times at intersections by around 20–30% while also making road crossings safer and more accessible for everyone. Overall, the project demonstrates that AI and IoT can be combined to create a practical, scalable, and cost-effective traffic management solution for modern smart cities.