Development of Adaptive Realtime Mobile Based Traffic Management System for Congested Urban Cities

Abstract

congestion in Nigeria, focusing on Port Harcourt. The study analyzed four main routes: Location, Rumuokoro, Rumuokwuta, and GRA Junction. Field surveys collected traffic volume and video recordings of cars traveling the indicated routes between 7 am and 7 pm. The Rivers State Command of the Federal Road Safety Corps provided further data. A mobile-based Traffic Management System was designed and simulated using Proteus. The technology transmits real-time traffic data from congested areas to a local server. This data is easily accessible on mobile devices. Python software was used to classify traffic congestion as heavy (0-14km/hr), medium (15-28km/hr), and free (29km/hr and above). Background subtraction was used for vehicle image analysis. The system can detect road traffic, correctly determine car speeds, count the number of vehicles, and send this traffic information to consumers via mobile devices via online communications. After simulating 1200 traffic data instances, 36 had mistaken, while 1164 were classified. This represents good accuracy, with 97% accuracy and 3% inaccuracy. The traffic data will be maintained in a database to help researchers identify places with recurring traffic congestion. These findings suggest that authorities should build alternative routes. It can also help build traffic-congestion-reducing policy review, organization, and enforcement initiatives. Keywords: Background Subtraction Technique, Mobile devices, Proteus, Traffic Congestion, Python