You are on page 1of 15

BAHIR DAR UNIVERSITY

BAHIR DAR INSTITUTE OF TECHNOLOGY


FACLUTY OF ELECTRICAL ENGINEERING
PROPOSAL
ON
Traffic Sign Recognition System using deep learning
No Name ID

1 MESFIN BEYENE 1103251


2 BENEYA LUKUDU JOSEPH 1111170

3 YOHANNES DEREJE 1102919

Advisor Name:
Mr. Ahmed N.

Dec, 2022
Bahir Dar, Ethiopia
DECLARATION

This is to certify that the work in our final year Research titled, “TRAFFIC SIGN
RECOGNITION SYSTEM USING DEEP LEARNING” submitted to the Department of
Electrical and Computer Engineering, Bahir Dar University, Ethiopia, in partial fulfillment of the
requirement, we therefore hereby declare that this report represents our own work which has been
done after registration for the degree of Computer Engineering at Bahir Dar University.

We have read the University’s current research ethics guidelines, and accept responsibility for the
conduct of the procedures in accordance with the University’s Committee on the Use of Human &
Animal Subjects in Teaching and Research (HASC). We have attempted to identify all the risks
related to this research that may arise in conducting this research, obtained the relevant ethical
and/or safety approval (where applicable), and acknowledged my obligations and the rights of the
participants.

i
Acknowledgment
This work would not have been possible without the support of our parents. We are especially
indebted to MR Molla, who have been supportive of our career goals individually and
corporate, and who worked actively to provide us with the protected academic time to pursue those
goals.

We are grateful to all of those with whom we have had the pleasure to work during this Amazing
Project, and other related projects. Each of the members of our Research Committee (Mesfin
Beyene, Beneya Joseph, and Yohannes Dereje) has provided extensive personal and professional
guidance and taught a great deal about both scientific research and life in general. We would
especially like to thank MR Ahmad N our Supervisor at Bahir Dar University-FACULTY of
ELECTRICAL AND COMPUTER ENGINEERING. As our teacher and mentor, he has taught
us more than we could ever give him credit for here. He has shown us, by his example, what a good
scientist (and person) should be.

We would like to thank our parents, whose love and guidance are with us in whatever we ought to
pursue. They are the ultimate role models.

ii
Abstract/Summary
Traffic Sign Recognition (TSR) is particularly useful for novice drivers and self-driving cars.
Driver Assistance Systems (DAS) involves automatic traffic sign recognition. Efficient
classification of the traffic signs is required in DAS and unmanned vehicles for safe navigation.
Convolutional Neural Networks (CNN) is known for establishing promising results in the field of
image classification, which inspired us to employ this technique in our thesis. Computer vision is
a process that is used to understand the images and retrieve data from them. OpenCV is a Python
library used to detect traffic sign images in real-time.

iii
List of Figures
Figure 1 specific objectives ............................................................................................................................4
Figure 2 system methodology ........................................................................................................................4

iv
Abbreviations

TSRS-Traffic sign Recognition System

CNN-Convolutional Neural Network

AI-Artificial Intelligence

Ds-Dataset

RS-Relevant Staff

DL-Deep Learning

SP-Scope of Project

SoS-Significance of Study

MU-Methodology Used

LR-Literature Review

v
Contents
DECLARATION ........................................................................................................................................... i
Acknowledgment........................................................................................................................................... ii
Abstract/Summary ....................................................................................................................................... iii
Introduction/ Background .............................................................................................................................1
Literature Review ..........................................................................................................................................1
Statement of the Problem.............................................................................................................................2
Objective/Aim of the study ...........................................................................................................................2
General Objective.....................................................................................................................................3
Specific Objective [2] ...............................................................................................................................3
System Methodology ....................................................................................................................................4
Block Diagram Description .....................................................................................................................4
Input ......................................................................................................................................................4
Data........................................................................................................................................................4
Deep Learning ......................................................................................................................................4
Train and Test ......................................................................................................................................4
Classification .........................................................................................................................................4
Scope of the Project ......................................................................................................................................5
Significance of the Study..............................................................................................................................5
Time Frame for the Project/Work Plan ..........................................................................................................5
References......................................................................................................................................................8

vi
Introduction/ Background
Lately autonomous vehicles have been a center of attraction for research study and development.
And when we speak about autonomous vehicles Traffic Signs Recognition system is the first and
foremost concept to include in any autonomous vehicles. Sometimes drivers may tend to miss the
traffic signs along the route this can be dangerous and very unsafe with concern to road safety. In
such cases automatic process of classification of traffic signs and reduce the number of road
accidents on very large scale and can ensure complete safety. Many big names in automation used
this system in their cars using computer vision and machine learning approach but this was soon
replaced by deep learning approach based on classifiers. Recently deep convolutional techniques
have been proved to be the most effective for object detection. It proves to be advantageous to look
at the traffic signs recognition/classification with the deep learning the deep learning perspective.
Classification of traffic signs is not a simple task it requires a huge dataset to go through various
processes in deep learning. Initially the dataset gets divided into some ratio where those many
images will go through one process and other number of images will go through other process and
then will show the accuracy of the classes and then further it will get recognized. This concept has
notable research work history and existing work going on still there are few areas/drawbacks that
are yet to overcome.

Literature Review
There is a notable amount of research work done on traffic signs recognition/classification in
history and still is going on. Different types of datasets have been used to solve different types of
problems which includes detection, classification, tracking etc. After trying out and testing every
object detection approach researchers got to the implementation with deep networks. In recent
years with increasing technology and availability of standard datasets deep learning method is
being more preferable. First CNN architecture ever used for traffic signs recognition was LENET
architecture. In the paper ‘Novel Deep Learning Model for Traffic Sign Detection Using Capsule
Networks’ by Amara Dinesh Kumar they had stated the various approaches been tried fpr traffic
signs recognition in early stages. Also they have stated how CNN method is effective and
preferable over other methods. They have also briefly stated about the capsule techniques. In the

1
paper ‘Traffic Sign Classification Using Deep Inception Based Convolutional Networks’ by Mrinal
Haloi IIT Guwahati1 again they have explained the noval deep learning for traffic signs detection.
In paper’A Novel Neural Network Model for Traffic Sign Detection and Recognition under
Extreme Conditions’ by Haifeng Wan they have mainly focused on the traffic signs detection by
autonomous vehicles under extreme weather conditions. In the paper’ Traffic Sign Detection for
Intelligent Transportation Systems’ by Ayoub Ellahyani1 they have survey regarding TSR system
being installed in transportation vehicles and this paper proposes that it is a step towards intelligent
transportation system. In the paper ‘Traffic Sign Detection and Recognition using Image
Processing’ by 1Karthikeyan D they have proposed the image processing approach towards TSR.
In the paper ‘Two-stage traffic sign detection and recognition based on SVM and convolutional
neural networks’ by Ahmed Hechri1 In this study, a novel two-stage approach for realtime traffic
sign detection. [1]

Statement of the Problem


Traffic sign detection and recognition have gained importance with advances in image processing
due to the benefits that such a system may provide. The recent developments and interest in self-
driving cars have also increased the interest in this field.

Automatic detection and recognition of traffic signs is very important and could potentially be used
for driver assistance to reduce accidents and eventually in driverless automobiles. There are many
sign boards on roads that people are unaware of. They might have seen traffic or road signs all the
time but don’t understand what those signs are indicating.

Our Traffic Sign Recognition System detects and recognizes road signs or traffic signs from an
image and video. It will also recognize signs in real-time using CNN. Here, Deep Convolutional
Neural Network (CNN) is used to develop Autonomous Traffic Sign detection.

Objective/Aim of the study


The overall aim is to develop a system that can be used for traffic sign inventory. Traffic sign
detection and recognition have gained importance with advances in image processing due to the
benefits that such a system may provide. In this system, the user can upload an image for the system
to detect and recognize traffic signs. The system can recognize uploaded images/videos. For real-

2
time traffic sign recognition, the user will need to turn on their device camera and click on the
camera button to start detecting.

General Objective
The most important objective of this project is to help the driving population residing in the country
to be able to recognize the different Ethiopia traffic signs using image processing techniques. This
will make their life simpler and better improve their quality and standard of living by reducing their
dependency on other individuals and thereby taking care of safety during driving

- The system automatically detects traffic signs.


- It can recognize signs from an image and videos.
- It can also detect and recognize in real-time.

Specific Objective [2]


1. To understand the properties of traffic signs and their implications for image processing
for the recognition task.
2. To understand colour, colour spaces and colour space conversion/optional.
3. To identify the most appropriate approach for feature extraction from road signs.
4. To evaluate the performance of the aforementioned methods for robustness under
different conditions of weather, lighting geometry, and sign.
5. Thesis can be divided into three stages: detection, classification and recognition.
6. Thesis can be divided into three stages: detection, classification and recognition

DETECTION

CLASSIFICATION

RECOGNITION

3
Figure 1 specific objectives

System Methodology
In this part we will understand how the methodology we have adopted to improve the detecting of
Indian currency here is a block diagram to illustrate the proposed system

INPUT DATASET DEEP LEARNING

CLASSIFICATION TRAIN AND TEST


RESULT

Figure 2 system methodology

Block Diagram Description


This project mainly contains 4 parts:
Input
The input is given in the form of image where the user takes an image and upload in the system.
Data
The dataset is a collection of different types of notes in the form of image taken at different to
improve the training sets and obtain better training results
Deep Learning
It helps to develop the knowledge of what human naturally learns which helps in processing and
the predicting an output which helps to find greater result
Train and Test
We have trained the dataset which help to get better accuracy with better sets of data and with the
help of testing we how accurate our system we achieve and understand we failed to train our data
Classification
We used to classify the kinds of traffic signs we have in the dataset like.

4
Scope of the Project
By the end of these project our system will be able to execute the follow:
 Traffic sign recognition system in image
 Traffic sign recognition system in video
 Traffic sign recognition system in real time
 Develop the system to classify traffic signs accurately
 Develop the graphical user interface to accept user input/another feature

Significance of the Study

With the increasing high road accidents witnessed everyday in and day out in our roads, some due
to drivers not paying attention to the traffic signs have caused unprecedented and yet a drawback
to our road users. With the raising technology we are determined to providing the most legitimate
solution to aiding road users and drivers to safety caution through a digital system that will help in
notifying them of their safety.

The benefits are generally focused around driver convenience, which for many is great news. As
mentioned above, it should take the stress off attempting to keep on top of speed signs when driving
somewhere new.

Also, for those who spend a lot of time on the motorway where variable speed limits are becoming
all the more frequent, system, such as ours traffic sign recognition software, will identify speed
limits displayed on motorway gantries and other signs of significance.

Time Frame for the Project/Work Plan

Week One:

Gathering all the requirement for the implementation of the project and these include the following:

1. Deploying a python_3 environment.

5
2. Acquiring a camera (Since we a re working to creating a handy systems, we will employ
a smart phone camera

Infinix hot 6 Specifications


Single 13 MP, f/2.0, AF

Features Dual-LED flash, panorama, HDR

Video 1080p@30fps

3. Hp Computer running on a Window 10 with 512GB and 4GB RAM and a 2.8GHZ processing
speed which is convenient for the project.

Week Two:

Specifications of the data domain and its algorithms and these involves the full utility of the python
development environment and the deep learning demo using a try and error mechanism of
implementation.

Week Three:

Incorporating the hardware part to the software and testing its functionality, and accessing the raw
data which is traffic sign motorway gantries.

Week Four:

By the end of week four we intend to have the system deployed and running with all its scope
realized and defined.

Week Five:

Site trial of the product (Traffic sign Recognition System) and assessment from our Advisor and
all the relevant staffs involved in the project implementation.

6
Cost and Materials Required

I. Hardware Requirement
Laptop or PC
 Windows 7 or higher
 I3 processor system or higher
 4 GB RAM or higher
 100 GB ROM or higher
II. Software Requirement
Laptop or PC
 Python
 Visual studio
 Server

7
References

[1] P. T. Bhosale, "International Research Journal of Engineering and Technology," Certified Journal, vol.
Volume: 09 Issue: 01 | Jan 2022 , p. Page 1656, 2008 .

[2] A. J. C. B. Rodrigues, "Detection and Recognition of Traffic Signs," Detection and Recognition of
Traffic Sign, p. 3, 2007.

You might also like