You are on page 1of 3

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882

Volume 5, Issue 4, April 2016

REAL TIME VEHICLE LICENSE PLATE RECOGNITION USING MY


RIO KIT
Sagar Joshi
PG Student, Embedded System
G.H. Patel College of Engineering &Technology
Anand, India

Rohit Parmar
Lecturer, Electronics & Communication Department
G.H. Patel College of Engineering &Technology
Anand, India

Abstract

C. Character recognition

In todays era security and safety becomes the most


important things. Automatic number plate recognition
used in many field like parking area, access control,
traffic control, tolling etc. Here in this paper an efficient
approach to recognize the license plate of vehicle is
presented. LabView is used as a basic software and
National Instruments myRIO board is used as a hardware
module. For image processing vision development
module (VDM) and vision acquisition software (VAS) is
used. Image segmentation and basic morphological
image processing is used to localize the plate region.
Here the color image is convert in binary image. The
grayscale image is converted into binary thorough
Niblack thresholding. After that doing erosion and
applying particle filter the final ROI is visible. Finally
filling the holes of image and masking with original
image gets exact license plate region. Optical character
recognition (OCR) is used to read the characters from
image. OCR is built in tool available in LabView. The
Microsoft LifeCam HD camera is interfaced with
myRIO board. The LCD is interfaced to show the output
license plate number.

It is the third and last phase of the system. The important


role is played by this phase in terms of accuracy and
success rate of system. There are many algorithms like
artificial neural network, support vector machine, OCR.
In our proposed system we have simply used the OCR
toll which is in built available in LabView.

Keywords: LabView, OCR, myRIO, ANPR, Microsoft


Lifecam HD, Niblack Thresholding, LCD

I. INTRODUCTION
Automatic license plate recognition is very essential now
a days.it is used in many places for the security reason.in
this paper we recognize the number plate using LabView
software.
Any automatic number plate recognition system
consisted of three parts:
A. Plate region localization

From whole image, extract the region of our interest that


is license plate region is the first phase of the whole
system.in other words we are only focusing on license
plate instead of whole image. The output of this step is
fed to second step.
B. Character partition

Partition or segmentation is the second phase of whole


system. Character segmentation is done automatically in
Labview.so no need to use precise algorithm for that.

II. RELATED WORK


Much work related to license plate detection is done
earlier, But very few has been done in the LabView. In
[1] the authors have used different image segmentation
and morphological image processing techniques like
thresholding, removing small objects, finding centroid
from image etc. Character is recognized through OCR.
LabView has an inbuilt OCR feature. The image size are
640x480 pixels. There is also assumption by author that
image is taken under clean condition. In [2] the authors
firstly convert the color image into grayscale image and
dilate it to remove noise because of conversion. The
authors have used vertical and horizontal edge
processing to localize the license plate region. Then
applies low pass filtering to smooth the histogram. Then
by filtering they found highest value of histogram region
which is most probable region of license plate. Then
recognition is done using template matching technique.
In [3] the authors have used RFID for reserved parking
system and recognition of license plate by using
OpenCV. They have used PIC controller as a
microcontroller. Recognition of authorized vehicle is
done through OpenCV. RFID reader read the RFID tag
at same instance which is inside the license plate. The
authors have used thresholding, edge detection and
connected component analysis to locate the license plate
and OCR is used for recognition.
In [4] the authors have used Mexican hat filtering to
remove noise from B & W images.it is 9x9 size kernel.
They have used cropping method of MATLAB for
license plate extraction. Then applied inverse technique
on the plate region. Character segmentation is done
through bounding box.it will detect the boundary of
black pixel. The recognition is through OCR.The authors
also created GUI in MATLAB.The whole system is
implemented for MALASIAN license plates. In [5] the
authors have use vision module of the LabView
software. Firstly the image is acquire through sensing

www.ijsret.org

277

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volume 5, Issue 4, April 2016

device after that some preprocessing and filtering is


done. After that some morphological process is done by
author as a part of segmentation and finally character is
read by OCR tool.

III. PROPOSED METHOD


The proposed method is explained using flowchart
shown below:

Figure 2 original image & masked grayscale image

Figure 3 Binary image

Figure 1 Flowchart
Firstly, read the image from pc or capture it from the
camera.After that we can observe that license plate
region will be always in the center part of image.so
image masking is done which removes border area and
unnecessary area of image.Then image is transfered into
grayscale image.After that histogram equalization is
done.then binary image is obtained using the niblack
thresholding.
Small objects is removed using erosion process.so
objects which are very small compare to license plate
will be removed. The hall filling process and particle
filtering is done after the erosion process. Particle
filtering will remove too big or too small objects
compare to license plate. After that by using grayscale
value pyramid algorithm pattern matching is done which
find out exact rectangle region from image which may be
our license plate region. Then after image masking is
done with the original image which will give the exact
license plate.

Figure 4 Result after processing

IV. RESULTS
By applying above algorithm the result we got are as
follow:

www.ijsret.org

Figure 5 Final output

278

International Journal of Scientific Research Engineering & Technology (IJSRET), ISSN 2278 0882
Volume 5, Issue 4, April 2016

system which is very useful when pc have not space to


put.it can be also useful in area where human cannot
easily survive. The system is working well under proper
condition like illumination and gets good efficiency.

REFERENCES
[1]

Figure 6 OCR output

V. DISCUSSIONS ON PROPOSED SYSTEM


All the images have 640x480 resolution. Here we have
taken only black and white number plates not yellow or
other plates. We have taken only single line number
plates. The LCD interfacing with myRIO is shown
below:

[2]

[3]

[4]

[5]

[6]

Figure 7 LCD interfacing


The LCD is connected through myRIO with SPI (serial
peripheral interface).camera interfacing with myrio is
shown below:

Y. K. Al-Audah, A. K. Al-Juraifani*, M. A. Deriche


A Real-Time License Plate Recognition System for
Saudi Arabia using Lab VIEW, image processing
theory, tool, application 2012 .
Dipti Jadhav,Manoj Sabnis Open Road Tolling in
India by pattern Recognition 2015 international
Conference on Technologies for Sustainable
Development ICTSD-2015.
Judith Sen E, Deepa Merlin Dixon K,Ansy
Anto,Anumary M V, Daine Micheal,Fincy
Jose,Jinesh K J Advanced license Plate Recognition
System For Car Parking International Conference
on Embedded Systems-ICES 2014.
Norizam Sulaiman, Sri Nor Hafidah Mohammad
Jalani, Mahfuzah Mustafa, Kamarul Hawari
Development of Automatic Vehicle Plate Detection
System 2013IEEE 3rd International Conference on
System Engineering and Technology
Tao Zhang, Yanqiu Cui and Yaning Yang Design
of a Character Recognition System Based on
LabVIEW 4th International Conference on
Mechatronics, Materials, Chemistry and Computer
Engineering ICMMCCE 2015 .
Paramvir Singh, Rashpinder Kaur, Gurjinder Singh
Labview Based Real Time Alphanumeric Character
Recognition System. 2015 second International
Conference on Advances in Computing and
Communication Engineering.

Book
[1] Anil K Jain, Fundamentals of Digital image
Processing. PHI publications.
[2] Rafel C. Gonzalez, Richard E. Woods, Steven L.
Eddins, Digital image Processing. McGraw Hill
publication.
[3] Christopher G. Relf, Image Acquisition and
Processing with LabView. CRC press.
Handbook/Manual
[1] User guide and specifications NI-myRIO-1900.
[2] Digilent reference manual of LCD
[3] Microsoft Lifecam HD-3000 technical data sheet.

Figure 8 Camera interfacing


The Microsoft Lifecam HD-3000 camera is used. It is
USB type webcam which can attach with myrio board.
The still image resolution of camera is 1280 x 800.

World Wide Web


[1] www.ni.com.

VI. CONCLUSION
This system can be useful in many places for example
tolling, traffic control etc. The ANPR can be further
extended by deducting money from the persons account
who breaks the rules. The system can work as standalone
www.ijsret.org

279

You might also like