TRAFFIC-SIGN DETECTION AND RECOGNITION USING DEEP

LEARNING
Submitted in partial fulfillment of the requirements for the award of Bachelorof
Engineering degree in Computer Science and Engineering

by

G.sai jithin, (Reg. No. 38110164)

Yanamala Umesh (Reg. No.38110662)

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

SCHOOL OF COMPUTING

SATHYABAMA
INSTITUTE OF SCIENCE AND TECHNOLOGY

(DEEMED TO BE UNIVERSITY)
Accredited with Grade “A” by NAAC

JEPPIAAR NAGAR, RAJIV GANDHI SALAI, CHENNAI - 600 119

MAY - 2022
 DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERINGBONAFIDE

CERTIFICATE

This is to certify that this Project Report is the bonafide work of KORAPATI MAHESH (Reg.
No: 38110288), SILUVERU HAKHILESH RAM (Reg. No: 38110189) who

carried out the project entitled “TRAFFIC-SIGN

DETECTION AND RECOGNITION
USING DEEP LEARNING
” under my supervision from November 2021 to April 2022.

Internal Guide External Guide (if Applicable)

Dr. G. Nagarajan M.E., Ph.D.,

ad of the Department
Dr. L. LAKSHMANAN M.E., Ph.D.,

Submitted for Viva voce Examination held on

Internal Examiner External Examiner

 DECLARATION

I G.sai jithin, Yanamala Umesh hereby declare that the Project Report entitled TRAFFIC-

SIGN DETECTION AND RECOGNITION USING DEEP

LEARNING done by me under the guidance of Dr. G. Nagarajan M.E., Ph.D., (Internal)

is submitted in partial fulfillment of the requirements for the award of Bachelor of Engineering

Degree in Computer Science and Engineering.

DATE: 25-03-2022 K. Mahesh.

PLACE: CHENNAI SIGNATURE OF THE CANDIDATE

 ACKNOWLEDGEMENT

I am pleased to acknowledge my sincere thanks to Board of Management of

SATHYABAMA for their kind encouragement in doing this project and for completing it
successfully. I am grateful to them.

I convey my thanks to Dr. T. Sasikala M.E., Ph. D, Dean, School of Computing Dr. L.
Lakshmanan M.E., Ph.D., and Dr. S. Vigneshwari M.E., Ph.D. Head of the Department,
Dept. of Computer Science and Engineering for providing me necessary support and
details at the right time during the progressive reviews.

I would like to express my sincere and deep sense of gratitude to my Project Guide Dr.
G. Nagarajan ME., Ph. D for his valuable guidance, suggestions and constant
encouragement paved way for the successful completion of my project work.
I wish to express my thanks to all Teaching and Non-teaching staff members of the
Department of Computer Science and Engineering who were helpful in many ways for
the completion of the project.
 TABLE OF CONTENTS

CHAPTER TITLE PAGE NO

NO

ABSRACT i

LIST OF FIGURES vi

LIST OF TABLES viii

1 INTRODUCTION 1

1.1 Data Science 1

1.1.1 Data Scientist 1

1.2 Artificial Intelligence 2

1.3 Natural Language Processing 3

1.4 Machine Learning 3

1.5 Objectives 4

1.6 Project Goals 5

2 LITERATURE SURVEY 6

2.1 Literature Survey 6

3 AIM AND SCOPE OF THE 14

INVESTIGATION

3.1 Project Proposal 14

3.2 Scope of the Project 14

3.3 Existing System 14

3.3.1 Disadvantages 15
 3.4 Preparing Dataset 15

3.5 System Study 15

3.5.1 Classification of Attacks 15
3.5.2 Attacks Summary 17

3.6 Feasibility Study 21

3.6.1 Data Wrangling 21
3.6.2 Data Collection 21
3.6.3 Preprocessing 21
3.6.4 Building the Classification model 21
3.6.5 Construction of a Predictive 22
model

4 EXPERIMENTAL OR MATERIALS AND 23

METHODS, ALGORITHMS USED

23
4.1 Project Requirements
23
4.1.1 Functional Requirements
23
4.1.2 Non-Functional Requirements
24
4.1.3 Environment Requirements

24
4.2 Software Description
24
4.2.1 Anaconda Navigator
26
4.2.2 Jupyter Notebook
28
4.2.3 Python

30
4.3 Python Libraries Needed
30
4.3.1 NUMPY Library
31
4.3.2 PANDAS Library
 31
4.3.3 MATPLOTLIB Library
31
4.3.4 SCIKIT-LEARN Library
32
4.3.5 TKINTER

32
4.4 System Architecture

33
4.5 UML Diagrams
33
4.5.1 Class Diagram
34
4.5.2 Use Case Diagram
35
4.5.3 Workflow Diagram
36
4.5.4 Activity Diagram
37
4.5.5 Sequence Diagram
38
4.5.6 Entity Relationship Diagram

4.6 Algorithms Used 39

4.7 Comparing Algorithm with prediction in 44

the form of best Accuracy result

4.8 Module Description 48

48
4.8.1 Data Validation process and

Visualization
54
4.8.2 Prediction of DOS Attacks
55
4.8.3 Prediction of R2L Attacks
56
4.8.4 Prediction of U2R Attacks
57
4.8.5 Prediction of Probe Attacks
 58
4.8.6 Prediction of Overall Network

Attacks
59
4.8.7 GUI based prediction results
of

Network Attacks

5 61
RESULTS AND DISCUSSION,
PERFORMANCE ANALYSIS

61
5.1 Performance Analysis
64
5.2 Discussion

6 65
SUMMARY AND CONCLUSION

65
6.1 Summary
65
6.2 Conclusion
65
6.3 Future Work

66
REFERENCES

67
APPENDIX

A. SOURCE CODE 67
B. SCREENSHOTS 118
 LIST OF FIGURES

FIGURE NO FIGURE NAME PAGE NO

1.1 Process of Machine Learning 4
3.1 Process of data flow Diagram 22
4.1 Anaconda Navigator Interface 26
4.2 System Architecture 33
4.3 Class Diagram 33
4.4 Use Case Diagram 34
4.5 Workflow Diagram 35
4.6 Activity Diagram 36
4.7 Sequence Diagram 37
4.8 Entity Relationship Diagram 38
4.9 Logistic Regression 40
4.10 Decision Tree Classifier 41
4.11 Random Forest Classifier 42
4.12 Support Vector Classifier 43
4.13 Data Frame 49
4.14 Percentage level of protocol type 50
4.15 Comparison of service type and protocol type 51
4.16 Data Validation 53
4.17 Prediction of DOS Attacks 54
4.18 Prediction of R2L Attacks 55
4.19 Prediction of U2R Attacks 56
4.20 Prediction of Probe Attacks 57
4.21 GUI based prediction 60
5.1 Accuracy comparison of DOS Attack 61
5.2 Accuracy comparison of R2L Attack 61
5.3 Accuracy comparison of U2R Attack 62
5.4 Accuracy comparison of Probe Attack 62
5.5 Accuracy comparison of Overall Attack 63
6.1 GUI before input 118
6.2 GUI after giving input 118
 TRAFFIC-SIGN DETECTION AND RECOGNITION USING
DEEP LEARNING

Abstract
Automatic detection and recognition of traffic signs plays a crucial role in
management of the traffic-sign inventory. It provides an accurate and timely way to
manage traffic-sign inventory with a minimal human effort. In the computer vision
community, the recognition and detection of traffic signs are a well-researched
problem. A vast majority of existing approaches perform well on traffic signs needed
for advanced driver-assistance and autonomous systems. However, this represents a
relatively small number of all traffic signs (around 50 categories out of several
hundred) and performance on the remaining set of traffic signs, which are required
to eliminate the manual labor in traffic-sign inventory management, remains an open
question. In this paper, we address the issue of detecting and recognizing a large
number of traffic-sign categories suitable for automating traffic-sign inventory
management. We adopt a convolutional neural network (CNN) approach, the mask
R-CNN, to address the full pipeline of detection and recognition with automatic end-
to-end learning. We propose several improvements that are evaluated on the
detection of traffic signs and result in an improved overall performance. This
approach is applied to detection of 200 traffic-sign categories represented in our
novel dataset. The results are reported on highly challenging traffic-sign categories
that have not yet been considered in previous works. We provide comprehensive
analysis of the deep learning method for the detection of traffic signs with a large
intra-category appearance variation and show below 3% error rates with the
proposed approach, which is sufficient for deployment in practical applications of
the traffic-sign inventory management.

CHAPTER 1

INTRODUCTION

Proper management of traffic-sign inventory is an important task in ensuring safety

and efficiency of the traffic flow. Most often this task is performed manually. Traffic
signs are captured using a vehicle-mounted camera and manual localization and
recognition is performed off-line by a human operator to check for consistency with
the existing database. However, such manual work can be extremely time consuming
when applied to thousands of kilometers of roads. Automating this task would
significantly reduce the amount of manual work and improve safety through quicker
detection of damaged or missing traffic signs. A crucial step towards the automation
of this task is replacing manual localization and recognition of traffic signs with an
automatic detection. In the computer-vision community the problem of traffic-sign
recognition has already received a considerable attention, and excellent detection
and recognition algorithms have already been proposed. But these solutions have
been designed only for a small number of categories, mostly for traffic signs
associated with advanced driver-assistance systems (ADAS) and autonomous
vehicles. Detection and recognition of a large number of traffic-sign categories
remains an open question. Various previous benchmarks have addressed the traffic-
sign recognition and detection task. However, several of them focused only on
traffic-sign recognition (TSR) and ignored the much more complex problem of
traffic-sign detection (TSD) where finding accurate location of traffic sign is needed.
Other benchmarks that do address TSD mostly cover only a subset of traffic-sign
categories, most often ones important for ADAS and autonomous vehicles
applications. Most categories appearing in such benchmarks have a distinct
appearance with low inter-category variance and can be detected using handcrafted
detectors and classifiers. Such examples include round mandatory signs or triangular
prohibitory signs.

1.1 PROBLEM DEFINITION

Automatic detection and recognition of traffic signs plays a crucial role in

management of the traffic-sign inventory. The issue of detecting and recognizing a
large number of traffic-sign categories suitable for automating traffic-sign inventory
management. We adopt a convolutional neural network (CNN) approach, the mask
R-CNN, to address the full pipeline of detection and recognition with automatic end-
to-end learning. We propose several improvements that are evaluated on the
detection of traffic signs and result in an improved overall performance.
 CHAPTER 2

LITERATURE SURVEY

Literature survey is the most important step in software development process. Before
developing the tool it is necessary to determine the time factor, economy and
company strength. Once these things are satisfied, then the next step is to determine
which operating system and language can be used for developing the tool. Once the
programmers start building the tool the programmers need lot of external support.
This support can be obtained from senior programmers, from book or from websites.
Before building the system the above consideration are taken into account for
developing the proposed system. The major part of the project development sector
considers and fully survey all the required needs for developing the project. For
every project Literature survey is the most important sector in software development
process. Before developing the tools and the associated designing it is necessary to
determine and survey the time factor, resource requirement, man power, economy,
and company strength. Once these things are satisfied and fully surveyed, then the
next step is to determine about the software specifications in the respective system
such as what type of operating system the project would require, and what are all the
necessary software are needed to proceed with the next step such as developing the
tools, and the associated operations.

[2.1] A novel lightweight CNN architecture for traffic sign recognition without
GPU requirements
For a safe and automated vehicle driving application, it is a prerequisite to have a
robust and highly accurate traffic sign detection system. In this paper, we proposed
a novel energy-efficient Thin yet Deep convolutional neural network architecture for
traffic sign recognition. Within the proposed architecture, each convolutional layer
contains less than 50 features enabling our convolutional neural network to be
trained quickly even without the aid of a graphics processing unit. The performance
of the proposed architecture is measured using two publicly available traffic sign
datasets, namely the German Traffic Sign Recognition Benchmark and the Belgian
Traffic Sign Classification dataset. First, we train and test the performance of the
proposed architecture using the large German Traffic Sign Recognition Benchmark
dataset. Then, we retrain the network models using transfer learning on the more
challenging Belgian Traffic Sign Classification dataset to evaluate test performance.
The proposed architecture outperforms the performance of the state-of-the-art traffic
sign methods with at least five times less parameter in the individual end-to-end
network for training.

[2.2]An efficient convolutional neural network for small traffic sign detection

Deep learning has become a ubiquitous method in object detection among multiple
domains recently. However, in the era of edge computing, deploying deep neural
networks on mobile edge platforms are challenging due to long latency and huge
computational cost. As previous research efforts were usually focused on accuracy,
achieving the balance between computational consumption and accuracy is a more
significant problem to be tackled in mobile edge computing domain. To this end, we
proposed an efficient convolutional neural network (CNN), which can remarkably
minimize the redundancy, reduce the parameters and speed up the networks. The
effectiveness of the network is further proved with experiments on a Tsinghua-
Tencent 100K traffic sign dataset. Results show that under the same-level model
size, our network outperforms the state-of-the-art Fast R-CNN and Faster R-CNN
with 10% improvement in accuracy. Compared to similar work, the computational
consumption on running time and memory of our network has been also reduced in
the premise of little loss in accuracy.

[2.3]Traffic Sign Detection and Recognition using a CNN Ensemble

In today's world, almost everything we do has been simplified by automated tasks.

In an attempt to focus on the road while driving, drivers often miss out on signs on
the side of the road, which could be dangerous for them and for the people around
them. This problem can be avoided if there was an efficient way to notify the driver
without having them to shift their focus. Traffic Sign Detection and Recognition
(TSDR) plays an important role here by detecting and recognizing a sign, thus
notifying the driver of any upcoming signs. This not only ensures road safety, but
also allows the driver to be at little more ease while driving on tricky or new roads.
Another commonly faced problem is not being able to understand the meaning of
the sign. With the help of this Advanced Driver Assistance Systems (ADAS)
application, drivers will no longer face the problem of understanding what the sign
says. In this paper, we propose a method for Traffic Sign Detection and Recognition
using image processing for the detection of a sign and an ensemble of Convolutional
Neural Networks (CNN) for the recognition of the sign. CNNs have a high
recognition rate, thus making it desirable to use for implementing various computer
vision tasks. TensorFlow is used for the implementation of the CNN. We have
achieved higher than 99% recognition accuracies for circular signs on the Belgium
and German data sets.

[2.4]The Speed Limit Road Signs Recognition Using Hough Transformation

and Multi-Class Svm
In this paper, a method for the speed limit traffic sign recognition is proposed. The
method is based on Support Vector Machines, which is one of the most efficient
algorithms used for traffic sign recognition. It comprises three phases. In the
preprocessing phase, RGB images are converted into HSL images in order to
increase the contrast. In the detection phase, Hough Transformation is used for
detecting the speed limit signs along with Gauss and Median filters for removing the
noise from the detected images. The detection phase achieves accuracy of 95.3%. In
the classification phase, a Histogram of Oriented Gradients descriptor for feature
extraction is used together with Support Vector Machines for image classification
and speed limit sign recognition. The proposed method was used on the two
databases - GTSRB, German Traffic Sign Recognition Benchmark and rMASTIF,
Croatian traffic sign database. The recognition accuracy of 93.75% is achieved. The
presented method proves to be applicable in advancing driving assistance systems
due to its detection and recognition accuracy as well as its performance, thus making
it appropriate for real-time applications.
 CHAPTER 3

SYSTEM REQUIREMENTS

3.1 HARDWARE REQUIREMENTS

System : Pentium i3 Processor

Hard Disk : 500 GB.

Monitor : 15’’ LED

Input Devices : Keyboard, Mouse

Ram : 2 GB

3.2 SOFTWARE REQUIREMENTS

Operating system : Windows 10

Coding Language : Python

 CHAPTER 4

SYSTEM ANALYSIS

4.1 PURPOSE

The purpose of this document is detection of traffic sign using machine learning
algorithms. In detail, this document will provide a general description of our project,
including user requirements, product perspective, and overview of requirements,
general constraints. In addition, it will also provide the specific requirements and
functionality needed for this project - such as interface, functional requirements and
performance requirements.

4.2 SCOPE

The scope of this SRSdocument persists for the entire life cycle of the project.
This document defines the final state of the software requirements agreed upon by the
customers and designers. Finally at the end of the project execution all the
functionalities may be traceable from the SRSto the product. The document describes
the functionality, performance, constraints, interface and reliability for the entire life
cycle of the project.

4.3 EXISTING SYSTEM

Yuan et al. in which though the accuracy rate was high, the processing time was also
very high. Another method used fusion network formation to obtain features of the
signs and background statistics around the observed image, but the complexity was
high.
4.4 Disadvantages of exsisting system
➢ Increased algorithms complexity
 ➢ Heavy system hardware requirement
➢ Different pre-processing for training data are necessary

4.5 PROPOSED SYSTEM

Traffic sign recognition and detection is an important part of any autonomous

vehicle. However, the real challenge lies in the detection and recognition of these
traffic sign from the natural image in real time and with accuracy. This paper gives
an overview of the traffic road sign detection and recognition system, we developed
and implemented using an artificial neural network which is trained using real-life
datasets. This paper presents the usage of convolution neural network along with
dataset as an implementation of our project to attain real-time result with accuracy.
The system developed based on this methodology can be implemented in public
transports, personal cars, and other vehicles in order to keep drivers alert and reduce
human errors that lead to accidents. The project has a wide implementation of self-
driving vehicles.

4.6 ADVANTAGES OF PROPOSED SYSTEM

➢ Reducing the number of accidents caused by driver distraction and to reduce

the seriousness of such accidents.

➢ Improve the driver's safety on the road.

DATA FLOW DIAGRAM

1. The DFD is also called as bubble chart. It is a simple graphical formalism that
can be used to represent a system in terms of input data to the system, various
processing carried out on this data, and the output data is generated by this
system.
2. The data flow diagram (DFD) is one of the most important modeling tools. It
is used to model the system components. These components are the system
process, the data used by the process, an external entity that interacts with the
system and the information flows in the system.
3. DFD shows how the information moves through the system and how it is
modified by a series of transformations. It is a graphical technique that depicts
information flow and the transformations that are applied as data moves from
input to output.
4. DFD is also known as bubble chart. A DFD may be used to represent a system
at any level of abstraction. DFD may be partitioned into levels that represent
increasing information flow and functional detail.
 ER DIAGRAM

UML DIAGRAMS

UML stands for Unified Modeling Language. UML is a standardized general-

purpose modeling language in the field of object-oriented software engineering. The
standard is managed, and was created by, the Object Management Group.
The goal is for UML to become a common language for creating models of
object oriented computer software. In its current form UML is comprised of two
major components: a Meta-model and a notation. In the future, some form of method
or process may also be added to; or associated with, UML.
The Unified Modeling Language is a standard language for specifying,
Visualization, Constructing and documenting the artifacts of software system, as
well as for business modeling and other non-software systems.
The UML represents a collection of best engineering practices that have
proven successful in the modeling of large and complex systems.
The UML is a very important part of developing objects oriented software
and the software development process. The UML uses mostly graphical notations to
express the design of software projects.

GOALS:
The Primary goals in the design of the UML are as follows:
1. Provide users a ready-to-use, expressive visual modeling Language so that
they can develop and exchange meaningful models.
2. Provide extendibility and specialization mechanisms to extend the core
concepts.
3. Be independent of particular programming languages and development
process.
4. Provide a formal basis for understanding the modeling language.
5. Encourage the growth of OO tools market.
6. Support higher level development concepts such as collaborations,
frameworks, patterns and components.
7. Integrate best practices.

USE CASE DIAGRAM:

A use case diagram in the Unified Modeling Language (UML) is a type of
behavioral diagram defined by and created from a Use-case analysis. Its purpose is
to present a graphical overview of the functionality provided by a system in terms
of actors, their goals (represented as use cases), and any dependencies between those
use cases. The main purpose of a use case diagram is to show what system functions
are performed for which actor. Roles of the actors in the system can be depicted.

SEQUENCE DIAGRAM:
A sequence diagram in Unified Modeling Language (UML) is a kind of interaction
diagram that shows how processes operate with one another and in what order. It is
a construct of a Message Sequence Chart. Sequence diagrams are sometimes called
event diagrams, event scenarios, and timing diagrams.

ACTIVITY DIAGRAM:
Activity diagrams are graphical representations of workflows of stepwise activities
and actions with support for choice, iteration and concurrency. In the Unified
Modeling Language, activity diagrams can be used to describe the business and
operational step-by-step workflows of components in a system. An activity diagram
shows the overall flow of control.

CHAPTER 6

MODULES DESCRIPTION
6.1 MODULES
➢ Segmentation Module
➢ Data Pre-processing
➢ Traffic Sign-Recognition
➢ Traffic Sign-Detection

MODULE DESCRIPTION
6.1.1 Segmentation Module
Segmentation is the method of partitioning a visual image into different subgroups
(of pixels) called Image Objects, which reduces the image's complexity and makes
image analysis easier. Thresholding is the method of using an optimal threshold to
transform a grayscale input image to a bi-level image.

6.1.2 Data Pre-Processing Module

To save space or reduce computing complexity, we can find it helpful to remove
redundant details from images in some situations. .Converting colorful images to
grayscale images, for example. This is because color isn't always used to identify
and perceive an image in several objects. Grayscale may be sufficient for identifying
such artefacts. Color images can add needless complexity and take up more memory
space because they hold more detail than black and white images color images are
represented in three channels, which means that converting it to grayscale reduces
the number of pixels that need to be processed. For traffic signs gray values are
sufficient for recognition.
6.1.3 Traffic Sign-recognition Module
Deep Learning is a subdomain of Machine Learning that includes Convolutional
Neural Networks. Deep Learning algorithms store information in the same manner
as the human brain does, but on a much smaller scale .Image classification entails
extracting features from an image in order to identify trends in a dataset. We are
using CNN for traffic sign recognition as it is very good at feature extraction. In
CNN, we use filters. Filters come in a variety of shapes and sizes, depending on their
intended use. Filters allow us to take advantage of a specific image's spatial
localization by imposing a local communication pattern between neurons.
Convolution is the process of multiplying two variables pointwise to create a new
feature. Our image pixels matrix is one function and our filter is another. The dot
product of the two matrices is obtained by sliding the filter over the image. Matrix
called "Activation Map" or "Feature Map". The output layer is made up of several
convolutional layers that extract features from the image. CNN can be optimized
with the help of hyper parameter optimization. It finds hyper parameters of a given
machine learning algorithm that deliver the best performance as measured on a
validation set. Hyper parameters must be set before the learning process can begin.
The learning rate and the number of units in a dense layer are provided by it. In our
system will consider dropout rate, learning rate, kernel size and optimizer hyper
parameter.

6.1.4 Traffic Sign Detection Module

In this Module, we have addressed the problem of detecting and recognizing a large
number of traffic-sign categories for the main purpose of automating traffic-sign
inventory management. Due to a large number of categories with small interclass
but high intra-class variability, we proposed detection and recognition utilizing an
approach based on the Mask RCNN detector. The system provides an efficient deep
network for learning a large number of categories with an efficient and fast detection.

CHAPTER 7

SYSTEM IMPLEMENTATION

7.1 SYSTEM ARCHITECTURE

Describing the overall features of the software is concerned with defining the
requirements and establishing the high level of the system. During architectural
design, the various web pages and their interconnections are identified and designed.
The major software components are identified and decomposed into processing
modules and conceptual data structures and the interconnections among the modules
are identified. The following modules are identified in the proposed system.

Fig: System Architecture

The above architecture describes the work structure of the system.
➢ The data is collected and image processing is done after pre-processing.

➢ After completing the pre-processing with the collected data and then by
applying machine learning algorithms it detects and recognizes the traffic
signal.
7.2 Problem Statement:

Automatic detection and recognition of traffic signs plays a crucial role in

management of the traffic-sign inventory. The issue of detecting and recognizing a
large number of traffic-sign categories suitable for automating traffic-sign inventory
management. We adopt a convolutional neural network (CNN) approach, the mask
R-CNN, to address the full pipeline of detection and recognition with automatic end-
to-end learning. We propose several improvements that are evaluated on the
detection of traffic signs and result in an improved overall performance.

CONCLUSION
In this paper, we have discussed that how our proposed system detects the
traffic signal and recognizes using machine learning algorithms. The proposed
system is also scalable for detecting and recognizing the traffic sign by image
processing. The system is not having complex process to detect and recognize that
the data like the existing system. Proposed system gives genuine and fast result than
existing system. Here in this system we use cnn algorithm to detect and recognize
the traffic sign.
 REFERENCES
[1] Aziz S, Mohamed E, Youssef F (2018) Traffic sign recognition based on multi-
feature fusion and ELM classifier. Proc Comput Sci 127:146–153.
[2] Jang, C., Kim, H., Park, E., Kim, H. (2016). Data debiased traffic sign
recognition using MSERs and CNN. In 2016 International Conference on
Electronics, Information, and Communications (ICEIC), Da Nang, Vietnam, pp. 1-
4. https://doi.org/10.1109/ELINFOCOM.2016.7562938.
[3] Lai, Y., Wang, N., Yang, Y., & Lin, L. (2018). Traffic signs recognition and
classification based on deep feature learning. In 7th International Conference on
Pattern Recognition Applications and Methods (ICPRAM), Madeira, Portugal (pp.
622-629).
[4] Rosario G, Sonderman T, Zhu X. (2018) Deep Transfer Learning for Traffic Sign
Recognition[C]//2018IEEE International Conference on Information Reuse and
Integration (IRI). IEEE: 178–185. MLA.
[5] Hatolkar, Y., Agarwal, P., & Patil, S. (2018). A Survey on Road Traffic Sign
Recognition System using Convolution Neural Network.
[6] Huang, Z., Yu, Y., Gu, J., & Liu, H. (2017). An efficient method for traffic sign
recognition based on extreme learning machine. IEEE transactions on cybernetics,
47(4), 920-933.
[7] Li, C., Hu, Y., Xiao, L., Tian, L. (2012). Salient traffic sign recognition based on
sparse representation of visual perception. In 2012 International Conference on
Computer Vision in Remote Sensing, Xiamen, China, pp. 273-278.
https://doi.org/10.1109/CVRS.2012.6421274.
[8] Shi W, Xin L, Yu Z et al (2017) An FPGA-based hardware accelerator for traffic
sign detection. IEEETransactions on Very Large Scale Integration Systems 4:1362–
1372.
[9] Rosario G, Sonderman T, Zhu X. (2018) Deep Transfer Learning for Traffic
signal Recognition[C] 2018 IEEE International Conference on Information Reuse
and Integration.
[10] Li, H., Gong, M. (2017). Self-paced Convolutional Neural Networks. In IJCAI,
pp. 2110-2116.