2016 1st International Conference on New Research Achievements in Electrical and Computer Engineering

Proposal for Automatic License and Number Plate

Recognition System for Vehicle Identification

Hamed Saghaei
Department of Electrical Engineering, Faculty of Engineering
Shahrekord Branch, Islamic Azad University, Shahrekord - Iran.
Email: [email protected],
Tel Number: +983833361000-399

Abstract— In this paper, we propose an automatic and orientation [11]. Since the orientation and font used for number
mechanized license and number plate recognition (LNPR) system plates differ in different countries/states/provinces, this
which can extract the license plate number of the vehicles passing algorithm is needed to be modified accordingly keeping its
through a given location using image processing algorithms. No structure intact, if we want to apply this system for recognizing
additional devices such as GPS or radio frequency identification the number plates of those places.
(RFID) need to be installed for implementing the proposed
system. Using special cameras, the system takes pictures from The purpose of this paper is to develop and implement a
each passing vehicle and forwards the image to the computer for smart system for optimum use of information and
being processed by the LPR software. Plate recognition software communication technology (ICT) for managing executive
uses different algorithms such as localization, orientation, organizations as well as forming a database for facilitating
normalization, segmentation and finally optical character decision-making and adopting better staff and strategic
recognition (OCR). The resulting data is applied to compare with planning methods. License plate number is one of the most
the records on a database. Experimental results reveal that the appropriate information items for identifying a vehicle and its
presented system successfully detects and recognizes the vehicle owner in Iran. Vehicle traffic can be controlled through
number plate on real images. This system can also be used for different techniques. If such traffic involves a variety of vehicle
security and traffic control. models, then installing special technical equipment for each
model would not be economical and cannot provide the
Keywords— License and number plate recognition (LNPR) expected level of security. License plate identification is a
system, image processing, orientation, normalization, useful and the only practical method used in individual
segmentation, identification, optical character recognition (OCR). countries to ensure security, prevent theft, and manage vehicle
fleet.
I. INTRODUCTION The rest of the paper is organized as follows: section II will
Vehicle’s license and number plate recognition (LNPR) present the system model of the developed LNPR system.
system has been an important area of research interest in image Section III will present the LNPR software with detailed
monitoring and processing systems [1]. With the advent of information on used algorithms. Section IV presents LNPR
high-tech cameras, number plate recognition system has system advantages briefly and finally section V will end the
numerous applications for traffic management applications, paper with the conclusion.
and especially in the parking lot [2, 3]. LNPR system has many
applications such as border crossing control [4], identification II. SYSTEM MODEL
of stolen vehicles [5], automated parking attendant [6], red
light camera [7], petrol station surveillance, speed enforcement, LNPR system is proposed for monitoring and managing
security. For many of these applications, most of the basic traffic in the parking lots of private and public organizations
processing algorithms remain the same. via identifying vehicle license plate numbers at the parking
gate. This system can also be used to identify stolen vehicles
The LNPR system works in three steps, the first step is the on roads. No additional equipment need to be installed on
detection and capturing a vehicle image, the second one is the vehicles for operating this system. Fig. 1 shows that the only
detection and extraction of number plate in an image. The third requirement of this system is installing special cameras for
step uses image segmentation technique to get the individual identifying license numbers at the entrance and exit gates of the
character and optical character recognition (OCR) to recognize parking lots. The images taken by these cameras are
the individual character with the help of database stored for subsequently processed in a computer. All vehicle traffic
each and every alphanumeric character [5, 8, 9]. information (including the driver’s image) is stored in the
system database for a long time. Thus, detailed traffic
In LNPR system, the main detection hardware of the first
information can be retrieved from different parking gates at
step including some cameras will be installed in places of
different times. Moreover, this system can apply intelligent
interest for intersection control, traffic monitoring etc., to
control in the parking lots through automatic opening of the
identify vehicles that violate traffic laws or to find stolen
gate only to authorized vehicles upon recognizing their license
vehicles. The technique used in some papers is based on
numbers according to Fig. 1. Other advantages of the proposed
pattern matching [10], which is fast and accurate enough for
system include online access to information such as the total
real-time applications and is developed for recognition of
number of vehicles currently present in the parking lot, the
license plates with prior knowledge of letters and numbers
number of authorized vehicles, etc. Moreover, too much
2016 1st International Conference on New Research Achievements in Electrical and Computer Engineering

Fig. 1. LNPR system for opening our desired mechanized gate in a parking lot

information of vehicles traffic statistics can be extracted from the number of current vehicles, time of entry and departure,
the system. number of vehicles exiting the area, duration of each vehicle’s
stay in the area, vehicles which have been denied entry,
The cameras used in the system can be deployed under all unidentified vehicles, etc. Implementing the proposed system
weather conditions and are equipped with powerful infrared can provide a higher level of security and safety in the
radiation units for identifying vehicle license plates in absolute
controlled area for the vehicles. The system can be connected
darkness. Moreover, these cameras are equipped with to some gates to provide automatic and intelligent
thermometers which detect ambient temperature and activate opening/closing operation for authorized vehicles. If connected
the cooling system of the camera at very high temperatures so to alarm lights, traffic lights, or smart boards, the system can
as to provide high-quality imaging in the -50 to +70 deg C display predefined messages for specific vehicles (like
temperature range. The camera case is built in accordance with “welcome”, etc.). The user interface of the system is designed
the IP66 standard and is fully water and dust proof. The lenses for speedy access to system events and can facilitate usage by
used in the camera provide high-quality image processing issuing audio alarms.
capabilities. The proposed comprehensive vehicle traffic
control and monitoring system is used for monitoring vehicle
access to a specified area. The system normally comprises a III. AUTOMATIC NPR SOFTWARE
camera for monitoring the vehicle path, an identification In this section, we introduce LNPR software. a picture was
system for recognizing license plate number used for further taken by a camera shown in Fig. 2(a) is sent to LNPR software
identification and control purposes, gates and also traffic lights for image processing and number recognition.
presented as optional accessories.
To maximize the flexibility of the LNPR system, a modular
The software design and programming with high precision structure is chosen. In every module image processing
provide great processing speeds and fully reliable system algorithm(s) will be implemented. The software uses nine
security. The software architecture is so designed that the algorithms means nine modules for identifying a license and
applications selected by the users can be easily developed with number plate follows:
minimum cost in the shortest possible time. The advanced 1- Plate localization: this algorithm is responsible for
artificial intelligence and neural networks techniques finding and isolating the plate on the picture. It is
implemented in the processing engine offer users powerful and shown in Fig. 2(b)
precise processing capabilities.
2- Plate orientation and sizing: this algorithm
In the proposed software, the administrator defines both compensates for the skew of the plate and adjusts the
“white list” for authorized vehicles which can access the site dimensions to the required size that is illustrated in Fig.
and a “black list” for unauthorized vehicles which are denied 2(c).
entry to that site. The system consequently would allow only
the authorized (predefined) vehicles to enter the parking lot, 3- Conversion: Using some conversion image processing
and, meanwhile, prepares reports of the parking lot in terms of techniques, the image can be converted as desired for
2016 1st International Conference on New Research Achievements in Electrical and Computer Engineering

instance to have a simpler processing of the image, we

convert the image from red-green-blue (RGB) layers to
gray scale layer demonstrated in Fig. 2(d)
4- Normalization: this algorithm adjusts the contrast and
brightness of the image that is illustrated in Fig. 2(e).
5- Edge detection: it is applied to increase the picture
difference between the letters and the plate backing. A
median filter may also be used to reduce the visual
noise on the image. As can be observed in Fig. 2(f), a
useless image part similar to the left side of Fig. 2(e)
has been omitted. (a) A taken picture by a camera installed at the gate place

6- Character segmentation: this algorithm finds the

individual characters on the plate and segments them
for extra enhancement and also additional lines are
deleted as shown in Fig. 2(g).
7- Optical character recognition: it is the electronic
conversion of images or printed text into machine-
encoded text demonstrated in Fig. 2(h).
(b) Plate localization
8- Syntactical and geometrical analysis: it checks
characters and positions against country-specific rules.
9- The averaging of the recognized value over multiple
fields/images to produce a more reliable or confident
result. Especially since any single image may contain a (c) Plate orientation and sizing
reflected light flare, be partially obscured or another
temporary effect.

A. Technical Specification of the NPR Processor

NPR Software has some features including input image, (d) RGB to gray conversion of the Plate
output data and operating systems supported by system core as
follows:
Input Image Features:
 BMP or JPEG image formats (e) Plate normalization
 PAL or NTSC video formats
 Average processing time: 60 ms (2 GHz CPU)
Output Data:
 License plate number (f) Plate edge detection
 License plate position
 Characters positions
 Driver’s image
 License plate color
(g) Plate character segmentation
 Character recognition reliability
 Possibility of identifying all kinds of license plates
regardless of the type and place of registration
 Possibility of recognizing license plates for stationary
(h) Optical character recognition
and moving vehicles (up to 180 km/h speed)
 Very high processing speed and precision Fig. 2. LNPR software with nine image processing algorithms
Operating Systems Supported by the System Core:
2016 1st International Conference on New Research Achievements in Electrical and Computer Engineering

 The possibility of exerting smart control on the gates

and traffic lights.
A. System Capabilities
 Defining and issuing admission tickets for authorized
vehicles to enter the area
 Defining unauthorized vehicles for preventing their
entry to the area and also defining a variety of
permissions for entering the area
 Defining the authorized day and time of arrival for
different permits
 Automatic opening the gate to authorized vehicles
Fig. 3. LNPR system for stolen vehicles identification in highways immediately upon recognizing their license plate and
manual opening of the gate by the guard and issuing
 Windows: XP (SP2 and SP3), Vista (SP1), and Server
2003, 2008. permission for specific vehicles as well as registering
 Linux: Red Hat Enterprise (3-5), Debian (4,5), Fedora their license plates
Core (7-10), and Open SUSE (7-11).  Defining guards’ working shifts and working hours
The main core of the system can be executed under  Issuing temporary permissions for specific vehicles
Windows and Linux operating systems. For the high-security
and high-sensitivity projects where Windows cannot provide  Possibility of taking the driver’s picture (optional)
the required security, the Linux-based LPR version can be used  Possibility of defining specific days and hours for
for more improved reliability. The received data can be sent exerting better control and taking images of the driver
using wireless communication system to a defined server. I and the vehicle
studied power control process for 3rd and 4th generations of
 Recording the entry/exit date and time for the vehicles
wireless cellular communication systems [12-15].
 Reporting the authorized and unauthorized vehicles
IV. LNPR SYSTEM ADVANTAGES parked within the area
LNPR system has some advantages as follows:  Providing multiple reports regarding the current and
past status of vehicle traffic to and from the area
 Easier vehicle’s arrivals/departures to/from a parking  Web-based monitoring the authorized and
lot unauthorized entries (the number of entries can be
 Issuing alarms when authorized or unauthorized increased)
vehicles try to enter a site  Storing vehicle traffic information for an unlimited
 Preparing updated and instantaneous reports from the time.
situation within the parking lot as well as individual
vacant and occupied parking spaces B. Reports
 Recording and retaining the driver’s picture during The proposed system presents a number of predefined
default reports. Based on customer demand, new reports can be
arrivals/departures
added to the system. The default reports include license plate
 Archiving vehicle traffic into and out of the area for a number, vehicle model, driver information, arrivals/departures
very long period of time time, entrance and departure gate numbers, parking duration,
 Increasing security in the area and gate shift guard reports.
 Presenting various statistical reports about vehicle
C. Applications
traffic in and out of the area
The proposed system can be widely used in the following
 No need to install additional equipment on vehicles applications:
for recognizing them (like different cards, etc.)
 Facilitating traffic inflow/outflow during rush hours  Stolen vehicles identification in roads and highways
shown in Fig. 3
 Possibility of displaying or broadcasting messages for
certain predefined vehicles  Office areas
2016 1st International Conference on New Research Achievements in Electrical and Computer Engineering

 Parking lots of different places such as commercial [7] S. Du, M. Ibrahim, M. Shehata, and W. Badawy, "Automatic license
plate recognition (ALPR): A state-of-the-art review," Circuits and
complex, public parking and etc. Systems for Video Technology, IEEE Transactions on, vol. 23, pp. 311-
 Passenger terminals 325, 2013.
[8] G. Balamurugan, S. Punniakodi, K. Rajeswari, and V. Arulalan,
 Airports "Automatic number plate recognition system using super-resolution
 Railway stations technique," in Computing and Communications Technologies (ICCCT),
2015 International Conference on, 2015, pp. 273-277.
 Shopping centers, auto repair shops, and carwash [9] M. V. Srinu and B. S. Shankar, "Real Time Car Parking System and
areas Parking Fee Display Using Raspberry Pi," International Journal of
Research, vol. 3, pp. 421-426, 2016.
 Other places where traffic control is required. [10] D. E. Knuth, J. Morris, James H, and V. R. Pratt, "Fast pattern matching
in strings," SIAM journal on computing, vol. 6, pp. 323-350, 1977.
[11] A. Broumandnia and M. Fathy, "Application of pattern recognition for
Farsi license plate recognition," ICGST International Journal on
V. CONCLUSION Graphics, Vision and Image Processing, vol. 5, pp. 25-31, 2005.
In summary, this paper presented the automatic vehicle [12] H. Saghaei, B. Seyfe, H. Bakhshi, and R. Bayat, "Novel approach to
identification system using vehicle license and number plate adjust the step size for closed-loop power control in wireless cellular
code division multiple access systems under flat fading," IET
recognition. The LNPR software of the system uses series of communications, vol. 5, pp. 1469-1483, 2011.
image processing algorithms for number plate recognition and [13] H. Saghaei and A. A. L. Neyestanak, "Variable step closed-loop power
finally identifying the vehicle from the database stored on the control in cellular wireless CDMA systems under multipath fading," in
PC. This software has been written in .Net C# based on the Communications, Computers and Signal Processing, 2007. PacRim
studied and simulated algorithms in Matlab. The SQL database 2007. IEEE Pacific Rim Conference on, 2007, pp. 157-160.
has been used to store different achieved records of vehicles. [14] H. Saghaei and B. Seyfe, "New Approach to Closed-Loop Power
Control in Cellular CDMA Systems under Multipath Fading," in
We have evaluated the system performance on real images. Wireless Communications, Networking and Mobile Computing, 2008.
Both the simulation and practical results revealed that the WiCOM'08. 4th International Conference on, 2008, pp. 1-4.
LNPR system can robustly detect and recognize the vehicle [15] H. Saghaei and B. Seyfe, "Power control with diversity antenna arrays in
using license plate in different lightening and weather CDMA wireless cellular systems under multipath fading," in Wireless
conditions and can be implemented on the entrance of highly Communications, Networking and Mobile Computing, 2008.
WiCOM'08. 4th International Conference on, 2008, pp. 1-4.
restricted areas. The prototype system can be integrated to the
intersection surveillance video system for traffic surveying,
stolen vehicles, or for some application specific purposes
discussed in the paper.

REFERENCES
[1] M. T. Qadri and M. Asif, "Automatic number plate recognition system
for vehicle identification using optical character recognition," in
Education Technology and Computer, 2009. ICETC'09. International
Conference on, 2009, pp. 335-338.
[2] M. Ondrej, V. Zboril Frantisek, and D. Martin, "Algorithmic and
mathematical principles of automatic number plate recognition systems,"
BRNO University of technology, p. 10, 2007.
[3] S.-L. Chang, L.-S. Chen, Y.-C. Chung, and S.-W. Chen, "Automatic
license plate recognition," Intelligent Transportation Systems, IEEE
Transactions on, vol. 5, pp. 42-53, 2004.
[4] Y. Wen, Y. Lu, J. Yan, Z. Zhou, K. M. Von Deneen, and P. Shi, "An
algorithm for license plate recognition applied to intelligent
transportation system," Intelligent Transportation Systems, IEEE
Transactions on, vol. 12, pp. 830-845, 2011.
[5] C. A. Rahman, W. Badawy, and A. Radmanesh, "A Real Time Vehicle?
s License Plate Recognition System," in null, 2003, p. 163.
[6] T. Sirithinaphong and K. Chamnongthai, "The recognition of car license
plate for automatic parking system," in Signal Processing and Its
Applications, 1999. ISSPA'99. Proceedings of the Fifth International
Symposium on, 1999, pp. 455-457.