ABSTRACT digital images by using computers. It uses computer

                                    ABSTRACT

 

There
are certain rules that have been prepared for the benefit of people and the
idea of preparing these rules is not that they should be understood by the
drivers, but it should also be understood by the driver and other people. It is
essential to fallow all the rules and regulation and they are clearly listed
here.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

Traffic signs defines a visual language that can be
interpreted by drivers. They represent the traffic signs on the road, give them
warnings, and help them with their careful driving by providing useful
information that makes driving safe and convenient.

The human visual perception abilities depend on
individual’s physical and mental conditions. In certain circumstances, these
abilities can be affected by many factors such as fatigue, and observatory
skills.

Giving this information in a good time to drivers can
prevent accidents, save lives, increase driving performance.

 

                                      CHAPTER 1

                           INTRODUCTION

 

1.1 INTRODUCTION TO IMAGE
PROCESSING

 

Image processing is among rapidly growing research
area today, with in application in various aspects of a business. Image
processing is used to convert an image into digital form and also to receive
some kinds of information from the same. The two types of methods used and for
image processing include analog and digital image processing.

 Digital image
processing techniques helps in manipulation of digital images by using
computers. It uses computer algorithms to perform image processing. It is
better than analog processing and avoids problems during processing such as
noise and signal distortion. It allows of more complex algorithms and
implementation of methods which is impossible in analog. It is the only
practical technology for classification feature extraction, pattern recognition
and projection.

 Most suitable algorithm will be used in
detecting and recognizing the road signs. As the car moves and approaches near
the traffic sign boards (in certain distance) the software will detect and recognize
the traffic signs. Major motivation in this project is to alert and give the
traffic sign information to driver in time which saves life.

 

 1.2
INTRODUCTION TO TRAFFIC SIGNS DETECTION

 

Following
the rules and regulation of traffic is one of the main aspect of every
individual drivers. Identifying the traffic sign board is important to drivers
at right time and at the right place.

 Camera based traffic sign recognition is a
valuable tool for any drivers, particularly with respect to signs indicating
danger and speed limits. It is able to recognize selected traffic signs,
identify their meaning and sustainably assist the driver in adapting his or her
driving style to the situation and regulations.

 Traffic sign detection and recognition (TSDR) is
a system that aims on developing an application of computer vision. This
application will automatically recognize the various traffic signs. TSDR in
general is an application that can be implemented with two main modules:
Detector and Classifier. Detector can be implemented using algorithms such as
color based and shape based. Detected traffic signs may need to be
pre-processed and corrected with their appearance before classification stage.
The classifier will make use of machine learning techniques such as vector
machine to identify the traffic signs.

This
detection can be done through canny edge detection algorithm. Canny Edge Detector is the best and
widely used algorithm for edge detection .This algorithm provides robust
Detection, Localization and Number of response. It is a multi-stage algorithm
which involves following stages:

•       
Gaussian Smoothing

•       
Gradient Filtering

•       
Non-Maximum Suppression

•        
Hysteresis Thresholding

 

Here in your project we can improve the
processing speed by using Principle       Component Analysis (PCA) techniques .This
PCA algorithm is used to analyze the signs.

 

1.4 INTRODUCTION TO
BACKGROUND OF PROBLEM

 

Organizing and managing of huge amount of
image data set of traffic signs have ever been a trivial problem and a
challenging task. The complexity increases when recognition of traffic sign
boards having a discontinuous signs, unconditional and irregularities found in
traffic signs.

 

The major challenging is to compress the captured
image and comparing with the stored images and segmenting of an traffic sign
images.

 

1.5 
OBJECTIVES
OF RESEARCH

 

·        
This system converts RGB images
into Gray scale images.

·        
This system removes the
noise in Gray scale images using Gradient Filter.

·        
This system compress the
captured images using ‘WCompress algorithm’.

·        
This system helps in
segmentation of captured image using                     algorithm.

 

 

1.6 
 APPLICATION AND CONTRIBUTION

 

       1.6.1 APPLICATION

·        
It will help the driver to drive
the vehicle without any accidents.

·        
It will provide comprehensive
assistance to the driver for the traffic signs.

·        
It will compress the capture image
which help in storage efficiency and time complexity.

·        
Erosion and Dilation methods is
used to detect uneven condition of a traffic signs.                                         

1.6.2     
CONTRIBUTION

·        
The algorithm we used is
robust and can detect signs even when the traffic sign board is rotated.

·        
The traffic sign template
can easily update to the database.

·        
This method is aimed at
achieving high accuracy in recognition of sign boards.

 

      

 

                            CHAPTER 2

 

            LITERATURE SURVEY

 

 

2.1 PAPERS REVIEWED

 

 ‘1Aakash Darekar, Pushkar Sindekar’ proposed
‘Traffic
Sign Detection and   Alert System’.
In this paper they introduce approach based on Principle Component images for
traffic sign detection and recognition. In the above stated methodology,
firstly, the connectivity of components in the obtained image checked and if it
is less than 1000 then that component is removed so that the noise is
eliminated and after that morphology operations are performed. This paper is
used Euclidean distance in the detection phase and Eigen Vectors in the
recognition stage. In this project they have achieved about 95.66% for
recognition and detection.

 

‘2G.Revathi
,Dr. G.Balakrishnan  proposed ‘ Indian Sign Board Recognition         Using Image Processing Techniques2’.
This paper deals with object in outdoor environments which are useful for
Driver Support Systems and Intelligent Autonomous ?Vehicles to take some
decisions about their speed, trajectory and send a warning signal indicating
over speed .It works in identifying traffic sign boards detection and
recognition. In this paper they introduce both Sobel Edge Detector and Canny
Edge Detector, they used Median Filter to removing of noises in  traffic sign images.

 

‘3Mohamed
Boumediene, Christophe Cudel  proposed ‘Triangular Traffic Signs Detection Based on RSLD
Algorithm3’. This paper has presented a novel
approach to triangular traffic sign detection. This idea is to detect the corners
and the symmetric sides of a triangle. The corners are detected using a
well-known Harris detector followed by a corner coding process. A future aim is
to combine this triangular traffic sign detection with an eye tracking system
developed in MIPS laboratory

 

 

 

 

     
        2.1.1 OBSERVATON IN THE FORM OF TABLE

      Table 1: Literature
Survey

Paper Title/ Author
Name

        Objective

    Methodology/
    Algorithm

         Result

Traffic
Sign Detection and Driver Alert System

This
project is to detect and recognition of traffic signs. PCA algorithm is used,
speed change according to sign.

PCB
Algorithm, ARM Controller,
Canny
Edge Detection Algorithm

95.6%
of accuracy in recognition and detection of traffic signs

Indian
Sign Board Recognition Using Image Processing

In
this paper they use Morphological
Process
for shape detection .The data which is obtained by neural network training is
used to classify the road sign   

Preprocessing
,Mathematical
Morphology Edge detection , Sobel edgy detection , can-edge detection 

It
is useful for driver support system and intelligent  autonomous vehicles to take some detection
about their speed and send alert messages 
about traffic signs.

Triangular
traffic signs detection based on RSLD Algorithm.

In
this paper mainly focuses on triangular shape traffic signs using RSLD Algorithm

Mobile
mapping system is used for data collection .
“Adaboost”
Algorithm is used for pattern recognition

Detecting
of  triangular shape traffic signs are
detected using “Harris Algorithm”

 

 

2.2 MOTIVATION:

During
the literature survey it is found that most of the works are focused on   different technologies and feature
extraction techniques. Since many works do not have much accuracy in
recognition and detection of traffic signs .Therefore more research and
implementation is required to building up an powerful algorithms for detection
of traffic signs.

 

2.3 IMPLEMENTATION OF
BASE PAPER 1

    ‘1Aakash Darekar, Pushkar Sindekar’ proposed
‘Traffic
Sign Detection and Alert System’. In
this paper they use Principle Component Algorithm (PCA) is used for recognition
and segmentation of captured images. They use of Euclidean Distance formula for
morphological segmentation operation on both stored images and captured images.   By implementing this algorithm it get precise
results and an efficiency of 95.66%.

2.3.1 Results and Discussion

In
the above stated methodology, firstly, the connectivity of the components in
the obtained image is checked and if it is less than 1000 (our experimental
value) then that component is removed so that the noise is eliminated and after
that morphology operations are performed; and then the letters are segmented
and recognized by Multiclass Support Vector Machine (SVMs) classifiers . We
make the use of Euclidean distance in the detection phase. Use Eigen vectors in
the recognition stage.

 2.3.2
Conclusion

We
have verified our method on some traffic speed sign board database and the
accuracy rate has been reached to 95.66% for detection and recognition stage.
Also sometimes accuracy decreased when the light intensity is very low. This
algorithm has higher speed and efficiency. Future scope is to transmit these
detected and recognized traffic signs to nearby vehicles to alert them, this
can be done by using a GSM module.

 

2.4 IMPLEMENTATION OF
BASE PAPER 2

 ‘2G.Revathi ,
Dr. G.Balakrishnan  proposed ‘ Indian Sign Board Recognition Using Image Processing
Techniques2’. In this paper they  use of two methods for detecting of and edges i.e.
Canny Edge Detector and Sobel Edge Detector to find the edges in the images.
They uses several methods /algorithm in this project like mathematical
morphology for to argues the shape and appearance of an images and they use
median filter to removal of noises in the images and they perform the image
classification techniques.

 2.4.1 Result
and Discussion

The software first uses color processing techniques
to isolate relevant color data from the image. A variety of Image Processing
Techniques are used to threshold, filter, detect edges, and further process the
image. Morphological processing algorithms are applied in order to remove non pertinent
data and isolate the sign boards. Shape detection is used to determine if a
sign is present in the current image. If present the sign will be resized and
classified. The data which obtained by neural network training is used to
classify the road sign. This proposed work not only recognizes the traffic sign
but also provides information about its condition or state. Finally Recognized
Traffic sign Boards are delivered as Text Message and Voice.

 

 2.4.2
Conclusion

This
paper deals with object detection in outdoor environments which are usefulness
for Driver Support systems and Intelligent Autonomous Vehicles to take some
decisions about their speed, trajectory and send a warning signal indicating
over speed, warn or limit illegal ma-oeuvres. It works in identifying traffic
sign boards detection and recognition.

 

 

 

 

 

  

 

 

 

 

                                          CHAPTER 4                                          

                                       CONCLUSION:

 4.1 SUMMARY

There
are certain rules that have been prepared for the benefits of people and the
idea of preparing these rule is not that they should be understood by drivers.
It is essential to follow all the rules and regulations. People are recommended
that they should be carefully observing all the rules and regulation and it is
effectual to be careful, considerate and patience. This signs have got very
prominent role to play in the traffic system and they are made for the safety
of people.

 The human visual perception abilities depend
on the individuals physical and mental conditions. In certain circumstances
these abilities can be affected by many factors such as fatigue and observatory
skills

Giving
this information in a good time to drivers can prevent accident, save lives and
increase driving performance.

                                   

                                               

 

 

 

 

 

 

 

                                     

                                      

                                BIBLIOGRAPHY

 

1Aakash Darekar, Pushkar Sidekar,”Traffic Sign
Detection and Driver Alert System”, International journal of Advancement
Research in Electronic and Communication Engineering Science and
Technologyvolume 5,1296-1298,5th may 2016.

2G.Revathi, Dr. Balakrishnan,”Indian Sign Board
Recognition Using Image Processing Techniques”, International Journal of
Advanced Research in Biology Engineering Science and
Technology,Volume2,326-330,15th  March 2016.

3Mohamed Boumediene, Christophe Cudel,”Triangular
Traffic Sign Detection Based on RSLD Algorithm”, Machine Vision and
Application, Springer Verlag Volume 3, 1721-1732, 24th Oct 2013.

4 Petia Ivanova Radeva,” Fast Traffic Sign Model
Matching and Recognition on Gray-Scale Images”, Escola Technical Superior
Engineering Department of Science and Computer Volume4, 1332-1336, 10th
July 2005.