International Conference on Embedded Systems - (ICES 2014)

162

Advanced License Plate Recognition System For

Car Parking
Judith Sen E,Deepa Merlin Dixon K,Ansy
Anto,Anumary M V,Daine Micheal,Fincy Jose
Dept. Electronics and Communication
Jyothi Engineering College
Thrissur, India
[email protected] ,[email protected]
Abstract The Advanced License Plate Recognition System For
Car Parking consists of two section: Radio frequency
Identification (RFID) based reserved car parking system and
number plate recognition by using Open Computer Vision(CV).
The system provides high security and there is no need for a
guard. In this system only authorized vehicles are permitted to
the parking area by checking some conditions. Authorized
vehicles have a valid RFID tag number and standardized license
plate. RFID is used to detect certain objects and access the data
using certain devices such as transponders or RFID tags. RFID is
a small wireless device based on an RFID reader and and an
RFID tag. The technology requires some extent of cooperation of
an RFID reader and an RFID tag. The number in license plate is
recognised by using open CV. Open CV contains many
programming functions used in the real time computer vision
systems. The library consists of many optimized algorithms and
C is considered as the backbone of open CV. Reserved car
parking system offer convenience for vehicle users and efficient
usage of space. This system save time, space and make the
parking easier.
KeywordsALPR ; RFID ; OpenCV ; Python ;Embedded Systems

I. INTRODUCTION
A good understanding of car parking system is
necessary for an efficient and secured design of parking
facilities. There are a range of parking facilities available to
people with disabilities in towns and cities. Every regulated
car parking system consists of a barrier which can be
overcome only with the help of a ticket. The system consists
of issual of a token and the clearance of the barrier and finally
the car is allowed to enter the parking area. Payment can be
made anywhere. A flexible payment charge is equipped in
order to increase the revenue. The system eliminates robbery
that can happen in a manual system. The cashier takes the
ticket and puts into the machine and the amount of parking
fees is displayed.
The system mainly consists of two sections: RFID
number plate based automatic car parking system; number
plate identification using image processing. RFID is an
automatic secure and convenient identifier which identifies the
objects automatically. RFID system employs many
frequencies in ultra high frequency region, low frequency
978-1-4799-5026-3/14/$31.00 2014 IEEE

Jinesh K J
Dept. Electronics and Communication
Jyothi Engineering College
Thrissur, India
[email protected]

region, high frequency region and microwave frequency

regions.

Fig 1. Automated Hands free car parking system

Intelligent transportation systems possess an important

role in peoples lives. Automatic License Plate Recognition
(ALPR) is used to obtain the license number of the vehicles
and it possesses various applications but also many challenges.
As they use many technologies in them, they are very
costly.The open source technologies have many applications
in these times. Python is the major contribution of open source
community . Open CV was introduced by Intel. Open CV can
be used for free under BSD license. Applications include
advanced robotics, mine inspection.
II.METHODOLOGY
Advanced License Plate Recognition System For Car
Parking involves mainly two parts: RFID number plate based
automatic car parking system; number plate identification
using image processing.

International Conference on Embedded Systems - (ICES 2014)

163

Fig 2. Block diagram of entire system

The main parts of the system are microcontroller, GSM

module, RFID reader, RFID tag, PC. The PIC16F877A used
as a microcontroller and chose RFID with operating frequency
125KHz. This system check two condition , then only it allow
the vehicle to the parking area. Due to this, it provide high
security in reservation of cars. Here reservation is done by
sending the license plate number of vehicle and, to the parking
entrance through SMS. To obtain this data a GSM module is
placed in the parking system. The fig:2 shows the block
diagram of the overall system. These process are discussed in
the following sub-sections:
A. RFID number plate based automatic car parking system
RFID technique and RFID tags are intelligent bar codes
that can talk to a networked system. Radio frequency signals
are used to send information to identify a particular situation
or item. RFID is a passive and a secure identifier which is
contactless and obtains the data from the RF signals.

Fig. 3 Flow chart

ALPR is divided in two main steps: plate detection and

plate recognition. Plate detection has the purpose of detecting
the location of the plate in the whole camera frame. When a
plate is detected in an image, the plate segment is passed to the
second step - plate recognition - which uses an OCR algorithm
to determine the alphanumeric characters on the plate.
The plate detection step consists of other three steps
such as capture, preprocess and localize. Plate recognition
involves connected component analysis, segment and
character recognition. The fig.4 shows the entire block
diagram of number plate recognition system.

B. Number plate recognition using Open Computer Vision

The figure given below shows the flow chart of the
number plate recognition system.

Fig 4. Block diagram of number plate recognition system

C. Capture
The input image is obtained from the web camera. The
readable characters may be distorted due to the obliqueness of
the camera. Use of a web camera which has got better
resolution will improve the image quality and hence the
perfection of the output. The fig.5. shows the captured image.

International Conference on Embedded Systems - (ICES 2014)

164

thersholding technique is applied to form these contours. This

technique makes use of two thresholds:lower threshold and
upper threshold. These thresholds are compared with the pixel
gradients. Pixel gradient is calculated as

Fig 5. Captured Image

D. Preprocess
Conversion to grayscale and binarization are two
important preprocessing steps. In this proposed system ,
preprocessing involves two processes; Resize, gray scale
conversion. In resizing, the image in hand is resized to a
feasible aspect ratio. Normally the images will be multi
coloured and it is to be converted to gray scale under
preprocessing stage. large and can drive the system slow. It is
to be resized to a feasible aspect ratio.
E. Localize
The input image can be caught from rear or front part of
the vehicle. The image may also contain other parts of the
vehicle and the environment, which are not under
consideration. The area under the interest is the license plate
and it should be localized .Localization includes the
binarization of the image.
There are two motivations for this operation,
Highlighting characters and Suppressing background.
Localization is done by a technique called Thresholding and
Edge detection. In thresholding, the pixels are separated to two
values depending upon the value of threshold. Appropriate
selection of the threshold value is to be done for accurate
outputs. There are different algorithms are used to
thresholding. In this paper, adaptive thresholding is used for
the binarizing process.

where Gx and Gy represent horizontal and vertical

gradients respectively. If the gradient is higher than the upper
threshold, it is taken as an edge pixel. If the gradient is lower
than lower threshold, it is not an edge pixel. If the gradient
comes in between, check whether the gradient of the pixel
under consideration is in the neighborhood of any edge pixel.
If yes, it is taken as an edge pixel. Otherwise, it is taken as a
non edge pixel.
E. Connected component analysis
Binarization of plate is done as a critical step and it
undergoes connected component algorithm so as to remove
the unwanted portions of the image. Connected components
are finally labeled and extracted.
F. Segmentation
Major step involved in the automatic number plate
recognition is the segmentation procedure. It is the initial
step and many other steps follows it. If the segmentation
fails, a character can be improperly divided into two
pieces, or two characters can be improperly merged
together. Basically the process of partitioning a digital
image into multiple segments is called segmentation.

Fig 7. Image after segmentation

The objects and boundaries in images are located by this

process.
H. Character recognition using OCR algorithm
Fig 6. Image after localization

Canny edge detection algorithm is used to detect the

edge pixels and assembles these pixels into contours. Hystersis

The optical character recognition is used for extracting

the features intended for training. It is used for data entry from
some kind of papers such as passport documents business

International Conference on Embedded Systems - (ICES 2014)

card,mail etcFor editing electronically, printed texts are

digitized and stored appropriately.OCR is extensively used in
pattern recognition and artificial intelligence. The Fig.8 shows
the image after character recognition.

Fig 8. Image after character recognition

III. TOOLS
A . Open CV
It was designed for enhancing the computational
efficiency and is focused on real time application.It uses
multicore processors.OpenCV provides a simple and
convenient computer vision infrastructure which helps people
to build vision applications. It is used by many computer
scientists and practical programmes.Although intel has
started OpenCV, as its name indicates, it is open and free for
all users. Open CV is simple, free and an open source. Open
CV also promotes ease of execution and possess easy run
time operations.
B. Python
It is used for Linux and microcontroller to
communicate serially. The system cant be communicated
directly to the motor instead it is interfaced through
microcontroller with the help of python. Performance can be
upgraded by using the python also coding in simple in python.
Python is very user friendly in the sense it can be understood
easily.
IV. RESULTS
This approach provides the output under two
conditions, first it check the entered vehicle license plate
number to the reserved number then it check the RFID tag
number of the vehicle. When an authorized vehicle reach at
the parking area, the number plate is recognised by Open CV.
RFID tag in the license plate is read by RFID reader
simultaneously. These two parameters are compared with the
database in the system. Whether the conditions are satisfied,
vehicle is permitted to the parking area. Else the permission is
denied. The algorithm is tried for approximately 20 vehicles

165

and success rate depends on the font styles and formats of

number plate.
V. CONCLUSION
Reserved car parking system offer convenience for
vehicle users and efficient usage of space. This system save
time, space and make the parking more easy. Open source
technologies are matured enough for scientific computing
domains. It is a better alternative to the existing proprietary
systems, due to the high speed and accuracy of the system.
This system is a real time embedded system which
automatically recognizes the license plate of vehicles. There
are many applications ranging from complex security systems
to common areas and from parking admission to urban traffic
control. Performance of the system can be improved by
training the neural networks.
REFERENCES
[1] K.M. Sajjad Automatic License Plate Recognition Using
Python And Open CV. Department of Computer Science and
Engineering M.E.S. College of Engineering, Kuttippuram,
Kerala ,[email protected].
[2] Abbas M. Al Ghaili, Syamsiah Mashohor, Abdul Rahman
Ramli and Alyani Ismail " Vertical Edge Based Car License
Plate Detection Method" IEEE transactions on vechicular
technology,Vol.62,No .1 ,Januvary .
[3] V. Koval I, V. Turchenko , V. Kochan , A. Sachenko , G.
Markowsky "Smart License Plate Recognition System Based
on Image Processing Using Neural Network" IEEE Workshop
on Intelligent Dam Acquisition and Advanced Comuting
Systems: retechnology and Apptication September 2003. Lviv.
Ukraine.
[4] Zhen Xue Chen, Cheng Yun Liu, Fa Liang Chang and
Guo You Wang "Automatic License-Plate Location and
Recognition Based on Feature Salience" IEEE transactions on
vehicular technology ,Vol. 58, No. 7,September 2009.
[5] Shan Du- Member IEEE Mahmoud Ibrahim, Mohamed
Shehata- Senior Member IEEE and Wael Badawy" 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.