Modelling and Simulation of Smart Traffic Light System for Emergency Vehicle using Image Processing Techniques

Abstract

Saving time is very essential for humans. Every day people are spending some time at the traffic signal due to the drawbacks of the conventional traffic light system. In the existing traffic light system, a defined timer system is used and it is working based on preset timing. Due to the preset timing, there is no flexibility of ON/OFF in the signal light based on the emergency vehicle and congestion of the vehicle. Sometimes emergency vehicle like an ambulance needs to wait at a traffic signal for a long time and this would lead to a risk to a patient's life. Traffic police must personally identify an ambulance and release the congestion, but this is not possible as there are an enormous number of vehicles present these days. This project aims to providea solution for the issue in the conventional system. The model was designed using an image processing system that reads the image and determines the presence of an emergency vehicle and the density of vehicles in each lane the ON/OFF signal for the particular lane will be given to the traffic light system which helpsto reduce the unnecessary waiting time of vehicles. The system calculates the vehicle's density and to detect the emergency vehicle using image processing to provide the green light signal tothe lane. This project used Open CV and Yolo (you only look once)algorithm in the image processing method to develop the system. The simulation has been done on the proposed smart traffic systemand it identifies that the proposed system is efficient. Multiple times of programming and testing have been done on the proposedsystem to ensure accuracy and for validation.