IRJET- Features Extraction OCR Algorithm in Indian License Plates

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 09 | Sep 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 1530
Features Extraction OCR Algorithm in Indian License Plates
Priya Sahu1, Silky Pareyani2
1Department of Electronics and Communication Engineering, Gyan Ganga College of Technology, Jabalpur
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract - Automated Number Plate Recognition (ANPR)
is an Image Processing Technology, used to recognize the
Characters of License Plate (LP) of vehicle using optical
character recognition (OCR). The modern technologies of
OCR achieve good performances on modern License Plate
produced with uniform layout and known fonts. However, for
old License Plate, OCR results are of lower quality. The OCR
quality assessment is a real challenge for the ANPR. The
objective is to design an efficient algorithm using the
features extraction in the OCR. The developed system first
detects the vehicle and then catches the image of the vehicle.
The Vehicle Number Plate area is extracted using image
splitting on the image. Now OCR is used for character
recognition. The resultant data is then used to compare the
records on the database so that specific information such as
the owner of the vehicle, the place of registration, the
address etc. can be revealed. The algorithm is implemented
and simulated in MATLAB, and it performance is tested on
numeric digits from 0-9 and alphabets from A-Z. It was
observed from the simulation that the proposed algorithm
successfully detects and recognize 32 characters among 36.
Therefore in order to recognize remaining two characters
(M & W) one other field has been introduced.
Key Words: OCR, ANPR, MATLAB.
1. INTRODUCTION
The escalating increase of contemporary urban and
national road networks over the last three decades
emerged the need of efficient monitoring and management
of road traffic. Conventional techniques for traffic
measurements, such as inductive loops, sensors or EM
microwave detectors, suffer from serious shortcomings,
expensive to install, they demand traffic disruption during
installation or maintenance, they are bulky and they are
unable to detect slow or temporary stop vehicles. On the
contrary, systems that are based on video are easy to
install, use the existing infrastructure of traffic
surveillance. Furthermore, they can be easily upgraded and
they offer the flexibility to redesign the system and its
functionality by simply changing the system algorithms.
Those systems allow measurement of vehicle's speed,
counting the number of vehicles, classification of vehicles,
and the identification of traffic incidents (such as accidents
or heavy congestion) [13].There is a wide variety of
systems based on video and image processing employing
different methodologies to detect vehicles and objects.
2. LITERATURE SURVEY
A typical ANPR system consists of a traffic camera network,
which processes captured traffic video on-site and transmits
the extracted parameters in real time. Here our focus is on
the study of algorithmic part of such a system. In this thesis,
we present full-featured vehicle detection, tracking and
license plate recognition system framework, particularly
designed to work on video footage [8, 9]. This system
consists of four modules:-
i. Video Acquisition
ii. Vehicle detection and tracking
iii. License plate extraction
iv. Character recognition
Automatic Number Plate Recognition
Automatic number plate recognition (ANPR) is a mass
surveillance and cataloguing the movements of traffic or
individuals and also for traffic law enforcement. ANPR can
be used to store the images captured by the cameras as well
as the text from the license plate. ANPR technology tends to
be region specific, owing to plate variation from place to
place. Concerns about these systems have centered on
privacy fears of government tracking citizens' movements,
misidentification, high error rates and increased
government spending [12, 9]. The system is designed for
real time videos where a camera is used for continuous
recording of videos. The view of camera or the area covered
by camera is fixed between entry zones and exit zone. Each
frame is continuously processed to check the presence of a
vehicle. A defined connected component area is taken as
threshold; if the detected area is above that threshold value
then it will be recognized as a vehicle and will be tracked [5].
Fig -1: ANPR system implementation in practical scenario

A distance is defined between the vehicle and the camera
and when the vehicle comes within that range i.e. vehicle’s
connected component area is maximum, these frames of
video are passed to license plate recognition algorithm.
After that recognition of character takes place and data is
stored and compare with data base.
Fig -2: Flow chart of the system
ANPR System has mainly four modules:
• Video Acquisition– In this module videos are taken by
the static camera situated at traffic scenario. A camera
network that has the ability to transmit images in real time
to a central operational centre. The processing of the
images can be carried out on-site saving valuable network
bandwidth as it transmits only the outcome of the
calculations. The whole process can also be performed
either in real time video streaming from an operational
centre or in already stored video material.
• Vehicle Detection and Tracking– In vehicle detection
we have simulated various background subtraction
techniques available in the literature. The background
subtraction technique should overcome the problems of
varying illumination condition, background clutter,
camouflage and shadow. Motion segmentation of
foreground object has been done in real time. It’s hard to
get this entire problem solved in one background
subtraction technique. So the idea was to simulate and
evaluate their performance on various video data taken in
different situations [1, 3].
• License Plate Extraction- License plates are first located
in current frame then they are extracted using various
available techniques in the literature based on Hough
Transform method, Template matching technique, Region
growing algorithm, Histogram Approach and Edge
Detection Approach [5, 12].
• Character Extraction– Images of the extracted plates are
the input to this module. Here first license plate image is
cropped in lines, and then characters are segmented and
recognized.
In the article presented by Kartikeya Jain, Tanupriya
Choudhury and Nirbhay Kashyap et al they describe the
application of OCR technology in the identification of the
number plate of the vehicle and allotting parking space for
the vehicle [1]. The vehicle is identified by the image of
number plate and the image is processed. The mechanism of
the OCR system consists of the Number Plate Localization,
Processing of Image, Character Segmentation and
Recognition. The final output is stored in the database [1].
3. PROPOSED METHODOLOGY
In this paper he overviewed the problem of Vehicle Number
Plate recognition with various types of Number Plates. The
system has been tested on MATLAB environment with
satisfactory results. Most of the time the input image is
taken from low-resolution mobile camera which does not
have very good quality image output [5, 10]. Given a better
device the result should increase in accuracy significantly.
Edges in an image are calculated by either horizontal or
vertical edge detection methods [3]. It is necessary for the
Localization algorithm to yield high accuracy as the
Character Recognition Technique would be ineffective if the
Number Plate is not localized correctly. Then by using
deskewing, the LP is rotate and fit along longitudinal axis
[4].
Optical Character Recognition using Feature
Extraction
This algorithm uses the fact that every character has unique
set of features such as corners, ending and bifurcations.
Inheriting these features makes the algorithm fast and less
complicated. The input character is converted to edge image
and the features are extracted from it in iterative process.
Every character has unique lines or slopes when observed
from the boundary of the segmented character image. This
methodology uses this feature to recognize the characters.
After the character is segmented out, it is resized in such a
way that the character touches from all sides to the
boundary. Features are extracted here by Optical Character
Recognition (OCR) method. There are three features in each
of the characters.
X1=Numbers of Triangle
X2=Numbers of Square
X3=Numbers of Corner
X4=Numbers of Pore
X5=Numbers of End
X6=Positions of Ends

Table -1: Recognition of alphabets or numeric digits from
image
Conversion of alphabets or numeric digits into
binary
Table -2: Conversion of Characters into binary formatALP. X1 X2 X3 X4 X5 X6
0 0 0 0 1 0 0
1 1 0 1 0 3 113
2 0 0 1 0 2 13
3 1 0 0 0 3 111
4 0 1 2 1 2 34
5 0 0 2 0 2 13
6 0 0 0 1 1 3
7 0 0 1 0 2 14
8 0 1 0 2 0 0
9 0 0 0 1 1 1
ALP. X1 X2 X3 X4 X5 X6
A 2 0 1 1 2 44
B 2 0 2 2 0 0
C 0 0 0 0 2 33
D 0 0 2 1 0 0
E 1 0 2 0 3 333
F 1 0 1 0 3 334
G 0 0 2 0 2 134
H 2 0 0 0 4 2244
I 2 0 0 0 4 1133
J 0 0 1 0 2 11
K 0 1 0 0 4 2244
L 0 0 1 0 2 23
M 0 0 3 0 2 44
N 0 0 2 0 2 24
O 0 0 0 1 0 0
P 1 0 1 1 1 4
Q 1 0 0 1 1 4
R 2 0 1 1 2 34
S 0 0 0 0 2 13
T 1 0 0 0 3 134
U 0 0 0 0 2 22
V 0 0 1 0 2 22
W 0 0 3 0 2 22
X 0 1 0 0 4 2244
Y 1 0 0 0 3 224
Z 0 0 2 0 2 13
ALP. S. No. Binary of S. No.
0 1 1
1 2 10
2 3 11
3 4 100
4 5 101
5 6 010
6 7 111
7 8 1000
8 9 1001
9 10 1010
A 11 1011
B 12 1100
C 13 1101
D 14 1110
E 15 1111
F 16 10000
G 17 10001
H 18 10010
I 19 10011
J 20 10100
K 21 10101
L 22 10110
M 23 10111
N 24 11000
O 25 11001
P 26 11010
Q 27 110110
R 28 11100
S 29 11101
ALP. S. No. Binary of S. No.
T 30 11110
U 31 11111
V 32 100000
W 33 100001
X 34 100010
Y 35 100011
Z 36 100100

If the Character is not found it means the number plate is not following the rules of Traffic or there is no number in the
plate or there is no plate.
Fig -3: Flow chart of the Proposed Method
4. RESULTS
Firstly we define proposed OCR method which is given in fig. 4. Now run the Code for ANPR. For this we upload the
captured crop image. After achieving 10 segments characters are displayed on the output screen. Finally the Owner
information may be fetched from the database of RTO.

Fig -4: Operation of ANPR using MATLAB

The result of the localization process in the article given in
IEEE paper by Kartikeya Jain, Tanupriya Choudhury and
Nirbhay Kashyap [1] gave Accuracy of 88% whereas the
character recognition technique is 96% Accuracy. The
complete Automatic Number Plate Recognition System
Gives the Accuracy given by-
= 92 %.
The result of the localization process in the article given in
IEEE paper by Bhavin V Kakani [3] gave Accuracy of 96.7%
whereas the character recognition technique gave 92.2%
Accuracy. The complete Automatic Number Plate
Recognition System Gives the Accuracy given by-
= 94.45 %.
In the proposed technique the localization process is kept
same as given by Bhavin V Kakani [3]. Here the problem is
that the 6 fields of Digit ‘0 (Zero)’ and alphabet ‘o’ are
identical. Hence this method gives 97.2% Accuracy, but it is
not a severe problem because according to Indian License
Plate System 1st two characters are Alphabets, 2nd two are
numeric digits 3rd two are Alphabets again and last 4
digits are Numeric digits [15].
So that Zero and Alphabet o may be clearly separated out.
The localization process given by Bhavin V. Kakani et al
[3] is 96.7% which is actually better. On the other hand
the Proposed Method gives 97.2% accuracy observed
from simulation in MATLAB which is more than that of in
previous methods. There is now the complete Automatic
Number Plate Recognition System Gives the Accuracy
given by-
= 94.45 %.
Table -3: Comparison of the Methods
The comparative results are comparison in between
proposed method with the available methods. The observe
Results are given in figure 6.
Fig -4: Comparison of the Methods
5. CONCLUSION
In the Proposed work, the Feature Extraction Algorithm is
employed in Optical Character Recognition (OCR) for
Recognition. Different algorithms that are given in literature
survey are studied. For raising the performance of OCR the
features of each of the characters is studied. Therefore the
system would be able to extract characters in real time.
Furthermore, the algorithm will be utilized in the existing
Traffic Police ANPR while not any modification or
standardization is required. Overall, in traditional traffic
conditions the algorithm will respond well. In future we will
try to increase the accuracy of Character Recognition for all
the font style.
REFERENCES
[1] Kartikeya Jain, Tanupriya Choudhury and Nirbhay
Kashyap, “Smart Vehicle Identification System Using
Ocr”, 3rd IEEE International Conference on
"Computational Intelligence and Communication
Technology", IEEE-CICT 2017.
[2] R Shreyas, Pradeep Kumar B V, Adithya H B,
Padmaja B and Sunil M P, “Dynamic Traffic Rule
Violation Monitoring System Using Automatic
Number Plate Recognition with SMS Feedback”,
2017 2nd International Conference on
Telecommunication and Networks , TEL-NET 2017.
[3] Mr. Bhavin V Kakani, Divyang Gandhi and Sagar Jani,
“Improved OCR based Automatic Vehicle Number
Plate Recognition using Features Trained Neural
Network”, 8th ICCCNT 2017, IIT Delhi, India, July 3-
5, 2017.
[4] Anisha goyal and Rekha Bhatia, “Automated Car
Number Plate Detection System to detect far
number plates”, IOSR Journal of Computer
Engineering (IOSR-JCE) e-ISSN: 2278-0661, p-ISSN:
2278-8727, Volume 18, Issue 4, Ver. III (Jul.-Aug.
2016), PP 34-40.
[5] Manoj D. Chaudhary and Jitendra B. Chinchore,
“Towards Multiple License Plate Localization in
S Authors Year
Localizati
on
Process
Character
Recogniti
on
Complete
Accuracy
1
Kartikeya Jain, Tanupriya
Choudhury and Nirbhay
Kashyap (1)
2017 88 % 96 % 92 %
2
R Shreyas, Pradeep
Kumar B V, Adithya H B,
Padmaja B and Sunil M P
(2)
2017 …………… …………… 95 %
3 Bhavin V Kakani (3) 2017 96.7 % 92.2 % 94.45 %
4 Proposed Technique 2019 96.7 % (3) 97.2 % 96.95 %

Indian Conditions: An Edge Density Based
Approach”, 2014, International Conference on
Control, Instrumentation, Communication and
Computational Technologies (ICCICCT),
Kanyakumari, 2014.
[6] Shan Du, Member, IEEE, Mahmoud Ibrahim,
Mohamed Shehata, Senior Member, IEEE, and
Wael Badawy, Senior Member, IEEE, “Automatic
License Plate Recognition (ALPR): A State-of-the-
Art Review”, IEEE transactions on circuits and
systems for video technology, vol. 23, no. 2,
february 2013.
[7] Jin Chong, Chen Tianhua and Ji Linhao, ” License
Plate Recognition Based on Edge Detection
Algorithm”, Ninth International Conference on
Intelligent Information Hiding and Multimedia
Signal Processing, 2013.
[8] Chirag Patel, Dipti Shah and Atul Patel, “Automatic
Number Plate Recognition System (ANPR): A
Survey”, International Journal of Computer
Applications (0975 – 8887) Volume 69– No.9, May
2013.
[9] Sukhpreet Singh Optical Character Recognition
Techniques: A Survey, Journal of Emerging Trends
in Computing and Information Sciences, Vol. 4, No.
6 June 2013.
[10] Ms. Sushama H. Bailmare, Prof. A.B.Gadicha, “A
Review paper on Vehicle Number Plate
Recognition (VNPR) Using Improved Character
Segmentation Method”, International Journal of
Scientific and Research Publications, Volume 3,
Issue 12, December 2013 ISSN 2250-3153.
[11] Sushruth Shastry, Gunasheela G, Thejus Dutt,
Vinay D S and Sudhir Rao Rupanagudi, i - A novel
algorithm for Optical Character Recognition (OCR),
IEEE 2013.
[12] W. Badawy, “Automatic License Plate Recognition
(ALPR): A State of the Art Review”, 2012.
[13] Amr Badr, Mohamed M. Abdelwahab, Ahmed M.
Thabet, and Ahmed M. Abdelsadek, “Automatic
Number Plate Recognition System”, Annals of the
University of Craiova, Mathematics and Computer
Science Series Volume 38(1), 2011, ISSN: 1223-
6934.
[14] Mr. Dipankar Bhattacharya and Anjan Bikash
Maity, “A Study on Vehicle Number Plate
Identification by Morphological Edge Detection
and Template Matching”, International Journal of
Information and Education Technology, Vol. 1, No.
3, August 2011.
[15] https://en.wikipedia.org/wiki/Vehicle_registratio
n_plates_of_India.

IRJET- Features Extraction OCR Algorithm in Indian License Plates

More Related Content

What's hot (20)

Similar to IRJET- Features Extraction OCR Algorithm in Indian License Plates (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Features Extraction OCR Algorithm in Indian License Plates