Smart Traffic Management System Based on Object Detection Using YOLO

Urban transportation systems are increasingly affected by traffic congestion, long waiting times, and inefficient fixed signal control, resulting in fuel wastage and higher emissions.This study investigates the design and development of an adaptive traffic signal control system using machine learning and computer vision to optimize traffic flow dynamically.Real-time or simulated traffic footage is processed using the YOLO algorithm for vehicle detection and counting, and the extracted data are used to train predictive models that estimate optimal green light durations.The system is implemented and visualized through a Python/Flask/React interface, with key performance indicators such as average waiting time, throughput, and queue length analyzed for evaluation.Experimental results demonstrate that the proposed machine learning-based adaptive traffic control significantly improves traffic efficiency and reduces congestion compared to traditional fixedtime systems.The findings highlight the potential of artificial intelligence to enhance urban mobility and support the development of intelligent transportation frameworks.