Competency Demonstration Report: Traffic Light Control Project

This report details a student's competency demonstration project on a real-time traffic light control system. The project utilizes image processing techniques, specifically edge detection and Matlab programming, to analyze traffic flow and dynamically adjust traffic light timings. The student's responsibilities included developing the software and hardware interfacing modules, using Matlab for image processing, and connecting the system to a parallel port. The project involved capturing images from a web camera, converting them to grayscale, performing gamma correction and edge detection, and comparing live feed images to reference images to determine traffic density. The results demonstrated the effectiveness of image processing in optimizing traffic light control, potentially reducing congestion. The student also discusses challenges faced, solutions implemented, collaborative efforts, and future innovations, highlighting the potential for intelligent traffic management systems. The report covers the theoretical knowledge of digital image processing and practical knowledge of Matlab to produce the implementation of the software and the hardware interfacing.

Competency Demonstration Report
Career Episode 3
BILAL ABDULLAH

CE 3.1 Project Information
Name of the project: Real Time Traffic Light Control Using Image Processing
Location of the project:[PLEASE FILL]
Project Duration:[PLEASE FILL]
Organization:[PLEASE FILL]
Role and Designation during the time: Team Member
CE 3.2 Project Background
CE 3.2.1 Characteristics of the project
Surveillance of vehicular traffic is essential for the smooth flow of vehicles in and around
the city. Parameters which are considered to be of the traffic control mechanism has always been
as topic of great research to produce effective traffic controller system. Using of sensors to
capture how fast a vehicle is moving or how much traffic has passed in a single day can be sued
to analyse the traffic flow of the road. Cameras with the ability of capturing videos and images
can be helpful in this situation. With the help of this and image study and measurement
calculation it will be easy to control the flow of traffic in a congested area. This is the idea which
our project has been based on. I have decided to produce a system which will be able to identify
the vehicles on the road and signal automatically to change from green to red and vice versa.
Moreover the data collected from the image processing can be stored and a quantitative analysis
of the data can provide a description of the traffic flow in the city.

CE 3.2.2 Objectives developed for the project
The objectives which I had set forward to follow to provide the best project were:
 To distinguish between the absence and presence of vehicles in the road images.
 To signal a traffic light to change to red when the road is empty.
 To change the traffic light to red if the maximum amount of time permitted has elapsed and
there are still vehicles on the road.
CE 3.2.3 My area of work
Out of the three modules of the project, I had taken up the responsibility of producing the
software and the interfacing module. I had selected these two modules because of the experience
in working with Matlab and knowing how to connect the computer module to the interfacing and
the image processing module would complete the project.
CE 3.2.4 Project Group
Figure 1: The hierarchical project group
Head of Electrical
Department Project Supervisor
Project Team
Member 1
(ME)
Project Team
Member 2

CE 3.2.5 My responsibilities throughout the project
I was responsible for producing the image processing module and connecting it to the
hardware interface to produce the complete module of the project. The connection was
completed using parallel port of the computer.
CE 3.3 Distinctive Activity
CE 3.3.1 Comprehending the theory of the project
The procedure to be followed for the implementation of the project has been divided into
three phases. The first phase consists of the acquisition of the image from the web camera of an
empty road and the image is changed to grey scale. The image is then saved as a reference image
to be used for the empty road. The gamma correction is then done on the reference image to
achieve the enhanced image. The edge detection is done on the image to have the best format of
the reference image. The second phase of the procedure is to collect an image of the road with
traffic and do the grey scale enhancement and the gamma enhancement to produce a secondary
reference image of the road. In the third phase the two reference image are then compared with
the live feed images and the percentage of the matching determines the traffic flow in the street.
The matching percentage of the images is found and the relative action is taken on the traffic
lights:
Result Number Matching Light Time
1 0 to 10% Gree
n
90 seconds
2 10 to 50% Gree
n
60 seconds

3 50 to 70% Gree
n
30 seconds
4 70 to 90% Gree
n
20 seconds
5 90 to 100% Red 60 seconds
Table1: experimental result of the matching system
CE 3.3.2 Engineering knowledge and skills applied in the project
I have applied the theoretical knowledge of digital image processing and the practical
knowledge of Matlab to produce the implementation of the software and the hardware
interfacing which would be used to process the images captured from the source video camera
and then had to be transmitted to the remote system server for image processing.
CE 3.3.3 Accomplishment and task performed
I had used Matlab for the process of edge detection of the image feed received from the
web cameras. The edges lines and points of the vehicles in the cars needs to be analysed to find
the amount of traffic in the road. The method requires the detection of the pixels corresponds to
the edge of a vehicle. The analysis returns a binary value of the detected edges of the pixels. The
process used is gradient based edge detection.

Figure 2: Reference image and Captured image
The intensity of the image is calculated as a vector gradient as follows:

The magnitude of the gradient is found by using: G[f (x,y)] = √ Gx2 +Gy2
The directional gradient of the image is found using: B(z, y) = tan-' (Gy/G)
For the flexibility, in a two dimensional figure the gradient is found by using:
Gx = f (i+1,j) – f (i,j) and Gy = f (I,j+1) – f (i,j)
The edge detection has been done using the Prewitt operator.
CE 3.3.4 Identified issues and their solutions
The major issue which I had to overcome during the implementation of this project was
to send the images received to the computer system to analyse them and send back the correct
command for the signalling system. The images which I was retrieving from the video camera
had to be compressed before sending else the overhead would cause a lot of bandwidth to be
consumed. Just by compressing the image is not the problem. To make the image clear for the
computer to analyse the cars the image had to be enhanced. This process required a lot of coding
on the Matlab. I cleared out the problem from the experience which I had gathered from the
Matlab workshop.
CE 3.3.5 Plan to produce creative and innovative work
Image processing is a tough concept to grasp. I had done a lot of research and study to
finally understand the effective way to understand what the image meant and how it can be
understood by the computer to know if there is a car or not. The major step in doing this was to

identify the edge of the vehicle. Not all vehicle had the similar edge. This was confusing at first,
but with trial and error I finally could teach a computer to identify vehicle.
CE 3.3.6 Collaborative work
I had collaborated with my team mates and my university professor to produce the system
which was to make a better time adjustment of the traffic lights to maintain a smooth flow of
vehicles during the time of congestion. The use of the image processing technology was learnt
from the special workshop which me and my teammates had attended at the transport control
office of the state. They taught us all about the system which they were currently using in the
city. This helped us to know how the current system could be made better for the city.
CE 3.4 Project Review
CE 3.4.1 Project Overview
After the completion and successfully implementation of the project the results showed
that the use of image processing is the best technique to use for the traffic lighting system, of any
country. The intelligent usage of the timer and the image processing can be used to eliminate the
large congestions which are common during the peak times. The project module is able to detect
the presence and absence of any kind of vehicle on the road and take decision according to the
movement of the vehicles. The project being designed based on the reality of presence of the
vehicle at the particular point is able to take better decision than the systems which use the
traditional way of detecting the amount of metal at the point to take the decision.

CE 3.4.2 My contribution to work
My contribution to the project was the production of the image processing system using
Matlab and to connect the image processing module to the main computer with the use of
parallel ports. I was also responsible for recording the test results t6 combine them and put them
into the final documentation of the project system. I had also helped my team mates to complete
the tasks assigned for the project and produce a complete working module of the project .