In India, with the growing number of vehicles, traffic congestion at junctions has become a serious issue. The density of vehicles is increasing day by day and there is an urgent need of adaptive traffic signals which can do real time monitoring of traffic density.
This paper describes a system which uses image processing for regulating the traffic in an effective manner by taking images of traffic at a junction. A step by step approach of image acquisition, image processing and implementation of algorithm to change the traffic light duration as per the density of vehicles on different roads at a traffic signal is followed. The number of objects in a given image is counted and priority is given to the densest road.
EXISTING TRAFFIC LIGHT SYSTEMS
The traffic lights used in India are basically pre-timed wherein the time of each lane to have a green signal is fixed. In a four lane traffic signal one lane is given a green signal at a time. Thus, the traffic light allows the vehicles of all lanes to pass in a sequence. So, the traffic can advance in either straight direction or turn by 90 degrees as shown in Fig.1.
A. Image acquisition:
The image is captured by a webcam. It is then transferred to the computer via a USB cable. The image acquisition and further processing is done by using MATLAB.
B. Image Processing:
The image is captured by using a webcam placed at the road junction. It has the capability of taking images of all the roads meeting at the junction. The webcam is mounted on the DC motor. The motor is responsible for capturing images from all directions in steps of fixed time interval. The speed of rotation of the camera is designed to be such that it is greater than the click-to-capture time of the camera. The acquired image is converted to grey scale image for further processing. The greyscale image is then converted to a binary image that contains only two colours, black and white.
The ARDUINO board is connected to the DC motor and the traffic lights. The DC motor is used for rotating the camera and controlling the timing of the image acquisition.
COMMUNICATION BETWEEN ARDUINO AND MATLAB INTERFACE
The procedure for connecting the Arduino Microcontroller board to MATLAB via the USB port for serial communication is as follows: The first step is to establish serial connection between the Arduino board and MATLAB via the USB port. The code for the Arduino board is written in the software Arduino and loaded on the board. In MATLAB the COM port is first configured. The configuration settings of serial port should match to that of Arduino. Then the data is sent serially from MATLAB to Arduino.
This technique can be effective to combat the growing pressure of traffic on Indian roads. It uses image processing to estimate the density of vehicles on roads and regulates the trafficat fixed intervals of time. It is cost efficient and does not require the installation of complex machinery to monitor the traffic. Deploying this system will not only save the time consumed in waiting at traffic junctions, but will also conserve a lot of resources that are otherwise wasted.
Source: Fr. C Rodrigues Institute of Technology
Authors: Rahishet | Aparajita Sahoo | Aparna Indore | Vaibhavdeshmukh | Pushpa U S