This project provides an intelligent traffic surveillance and management system based on artificial intelligence that takes advantage of MATLAB in conjunction with a microcontroller to facilitate road safety and efficient traffic flow. It utilizes image processing to identify traffic violations, such as not wearing a helmet or seatbelt, not carrying loads beyond the weight limit, or that do not have appropriate safety equipment. The captured pictures are converted to greyscale, and then features are derived via the Hough Transform. The derived features are passed to the trained model of the machine learning algorithm which determines if it matches any known violations. If a violation is found, a challan is automatically sent to the mobile number registered in the vehicle's record. If no violations are detected, it sends an ‘A’ to the microcontroller and displays this on a $20 \times 4$ LCD which can indicate that the violator is in compliance. The intelligent traffic surveillance and management system also improves traffic congestion. The microcontroller uses IR sensors to measure vehicle density on the roads approaching intersections. When it senses a high volume of traffic, it can tell the 8051 microcontroller to pass a command to extend the green signal time for that lane, and therefore improve traffic volume during peak hours. As an improvement factor, the system makes use of an RF module for emergency vehicles. The RF transmitter that interrupts the normal function of a traffic signal is located in the emergency vehicle, where it transmits an RF signal to a receiver located at the traffic signal junction. The receiver sends a signal to the microcontroller to turn the light on in the emergency lane and displays an alert on the LCD. The system contributes towards Smart City motivation goals by reducing human involvement while also improving overall safety across the traffic context and environment as a whole.
AI-Powered Traffic Management System
25.06.2025
793685 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
IEEE | 2023
|