Smart Roads, Smarter Cities: Machine Learning Integration for Dynamic Traffic Management

Abstract

As the world's cities become more urbanised, traffic congestion becomes a major problem. Conventional approaches are unable to deliver timely insights, which impedes the application of efficient congestion control strategies. This study presents a novel machine learning-based traffic congestion control system that combines a Euclidean distance tracker with the YOLO (You Only Look Once) object recognition framework. As cities struggle with the intricacies of increasing traffic, the need for intelligent technologies capable of real-time vehicle surveillance and congestion analytics is highlighted. To address this, the suggested solution goes beyond traditional constraints by using machine learning to accurately detect and track automobiles in urban environments. Utilizing the YOLO object detection framework, which is renowned for its speed and accuracy, the study builds on prior research in computer vision and transportation engineering. By connecting object detections between frames, the Euclidean Distance Tracker improves performance and allows a continuous comprehension of vehicle motions. The system's effectiveness in real-world circumstances is demonstrated by the results, which offer high accuracy across a range of vehicle classes. A major advancement in the development of urban mobility has been made with the integration of YOLO and the Euclidean Distance Tracker, which offers a viable solution for intelligent traffic management.