Design of an Automatic Traffic Control System with Real Time Vehicle Density Analysis

Abstract

Growth in developing countries forcefully leads their roads becoming increasingly congested thus bringing traffic to a standstill. There are two major measures that can address this issue viz. manual and automatic approaches. Because of its inefficiency, manual approach is not considered under the preview of this study and the automatic approach, which is the focus of this study, can further be classified into two viz. timer based and density based approaches. Timer based approaches are primarily static in nature and hence cannot consider real time requirements. Incorporating Density analysis on to the problem under preview adds dynamism to the solution and in the proposed study, traffic signals are controlled by considering real time density of vehicles on each roads. After capturing real time images and performing appropriate preprocessing, density of vehicles are measured using Canny Edge detection approach; which is used as a parameter for later decision making. Considering density parameters alone may lead to starvation to certain roads and here in the proposed study, authors consider a weight factor to propose a better solution. A prototype was developed using Arduino mega 2560 board, Webcams and Airplane 9g Mini Servo - SG-90 and led lights and it was observed that the efficacy of the proposed system is higher while comparing it to other prevailing methods.