IJCSN - International Journal of Computer Science and Network, Volume 6, Issue 6, December 2017

ISSN (Online) : 2277-5420

www.IJCSN.org
Impact Factor: 1.5

Automatics Vehicle License Plate Recognition

using MATLAB
Alhamzawi Hussein Ali mezher

Faculty of Informatics/University of Debrecen

Kassai ut 26, 4028 Debrecen, Hungary.

Abstract - The objective of this paper is to be able to detect the area of a car license in a photograph. Using computer vision and
digital image processing techniques. The different processes to which a photograph of a car is subjected will be explained in order to
detect its registration. It will be indicated under what conditions the presented algorithm is effective and correctly detects the
registration and under what conditions it can produce false positives. Indicate that the algorithm will only detect the area where the
license plate is located, adjusting as much as possible the size of this zone to the size of the license plate.

Keywords - Median filter , canny edge detection, image convolution , image binarization.

1. Introduction
the OCR process is performed later. When done in situ,
he ANPR ( Automatic Number Plate Recognition the information captured from the alphanumeric
registration, date and time, identification of the place and

T
or ANPR in English ) is a method of mass
surveillance which uses optical character any other information that is required is completed in
recognition on images to read the license plates of about 250 milliseconds. This information, now converted
vehicles. In 2005, systems can scan license plates into small data packets, can be easily transferred to a
with an approximate frequency of one per second in remote computer for further processing if necessary or
vehicles with speeds of up to 160 km / h. They can use the stored in the place to be retrieved later.
existing closed circuit television or radars, or ones
designed specifically for this task. They are used by the In the other arrangement, there is typically a large number
various police forces and as a method of electronic toll of PCs used in a server farm to handle high workloads,
collection on the payment highways, and to monitor such as those found in the London Congestion Charge
traffic activity as a red light at an intersection. project. Often in such systems, there is a need to send
images to the remote server and this may require high
The ANPR can be used to store the images captured by bandwidth transmission means. The drawbacks of these
the cameras, as well as the license plate text, and some systems are centered on the fear regarding the privacy of
can be configured to store a photograph of the driver. citizens' movements and media reports about
These systems often use infrared illumination to make it misidentification and high error rates. However, as they
possible for the camera to take pictures at any time of the have been developed, these systems have been much more
day. At least one camera version for intersection accurate and reliable.
monitoring includes a powerful flash, which serves to
illuminate the scene and make the offender aware of his 1.1. Methodology
error. The ANPR technology tends to be region-specific,
due to the variation between license plates from one place The proposed algorithm is designed to recognize license
to another. plates of cars currently used in Europe , both those located
at the front and at the rear. This algorithm will take as
The system software runs on a standard PC hardware and input an image of a vehicle, perform a pre-processing
can be linked with other applications or databases. First, it process to adjust the quality of the image, then mark the
uses a series of image manipulation techniques to detect, area of the image where the license plate is located, be
normalize and enhance the image of the license plate making segmentation, and finally proceed to the location
number, and finally, optical character recognition to and extraction of the license plate. In summary, the steps
extract the alphanumeric from the license plate. The of the algorithm are:
ANPR / ALPR systems can be used in two ways; one
allows the process to be performed in its entirety in the 1-Pre-processed image
place of the shot in real time, while the other transmits all 2-Segmentation of the image.
images from many cameras to a remote computer where 3-Location and extraction of the license plate

677

Copyright (c) 2017 International Journal of Computer Science and Network. All Rights Reserved.
IJCSN - International Journal of Computer Science and Network, Volume 6, Issue 6, December 2017
ISSN (Online) : 2277-5420
www.IJCSN.org
Impact Factor: 1.5

Captured image from the camera result of this process is an image with less noise and
greater smoothing.

Pre-processing image

Segmentation of the Location and

image extraction of the
license plate
Binarization of the
pre-processed image Morphological Fig. 3 using median filter to smooth the image
transformations

2. Segmentation of the image

Edge detection
Calculation of
intensities by rows The next step is to segment the image to locate the
Image convulsion
possible area where the license plate is located. This step
Calculation and does not pretend to detect the registration, but only to
Crop the image extraction of regions segment the image to stay with the best area of the image
where the registration is possible.
Removal of the license The steps to be taken in this process are:
plate
1- Binarization of the pre-processed image.
2- Calculation of intensities by rows.
3- Crop the image.

Output Image 2.1 Binarization of the pre-processed image

Fig.1 License Plate Recognition diagram
A process of binarization is applied to the resulting image
of the pre-processing in order to detect its edges. In this
1.2-Pre-processed image case, the image is processed using the Canny algorithm.
This algorithm is based on the theory of first derivative
The input image will consist, as a general rule, of a color operators and is especially useful since, in addition to
RGB image. The first step will be to scale the image to a extracting edges, it closes the contours avoiding possible
suitable size for further treatment. In this case, the image ruptures. The threshold used for this algorithm was 50.
is scaled to a size of 480 × 640. Then, we proceed to the
conversion of the image to a grayscale, eliminating the The result of this process is the following:
information of tone and saturation but maintaining the
luminance. The result of these two processes is the
following:

Fig.4 binarization the plate region

Fig. 2 scale the image to a suitable size
2.2 Calculation of intensities by rows
Once the image is converted, it is subjected to a
smoothing process in order to eliminate the noise. For this
Once binarized the image proceeds to perform a
purpose, an order filter is applied, in this case, the median
calculation of intensities by rows. This calculation is
filtering has been used. Order filters are based on methods
based on several assumptions:
that use image statistics, known as order statistics. In the
case of filtering the median for each pixel, the intensities
1- The area where the registration is will have a
are ordered by their neighboring pixels, the median is
very intense row (1's) zones very close to each
determined and this value is assigned to the pixel. The
size of the neighborhood that has been used is 5 × 5. The

678

Copyright (c) 2017 International Journal of Computer Science and Network. All Rights Reserved.
IJCSN - International Journal of Computer Science and Network, Volume 6, Issue 6, December 2017
ISSN (Online) : 2277-5420
www.IJCSN.org
Impact Factor: 1.5

other. Keep in mind that the registration will 2- Edge detection

occupy several rows. 3- Image convulsion
2- A row containing a piece of the license plate will 4- Calculation and extraction of regions
follow a certain pattern. That is, the numbers will 5- Removal of the license plate
have blank space between them and in their
region. 3.1 Morphological transformations
The algorithm consists of traversing all the rows of the This first step is to perform a series of morphological
binarized image one by one. For each row, its value will transformations in the image. These transformations will
be increased by 1 (starting from 0) as long as we meet allow us to soften the edges of the different regions of the
some of the following patterns in their pixels: (0, 0, 1) or image (specifically the region of the license plate),
(1, 0, 1). The idea is to penalize those areas with separate the different regions and facilitate their
continuous borders or without borders and favors those computation.
rows with discontinuous edges, a pattern that will follow
the numbers of a license plate. Three types of transformations will be used:

The result of this process is the following graphic: 1- First of all, an erosion of the image is applied.
This morphological transformation combines
two sets using vector subtraction. One of these
sets will be the image resulting from the previous
processes and the other will be the following: [0
0 1 0 0; 0 1 1 1 0; 1 1 1 1 1; 0 1 1 1 0; 0 0 1 0 0].
2- Then the transformation called opening is
calculated. To calculate this transformation to the
result of applying erosion to the image, a dilation
process is applied. This transformation, dilation,
Fig.5 intensities of the plate region combines two sets using the addition of vectors.
3- Finally, the image is closed. For this the result of
If we look at the graph we can see that in the area of the the erosion is subtracted from the result of the
registration is where the rows are concentrated with opening.
greater intensity.

2.3 Crop the image The result is an image with perfectly defined edges and
closed regions, which will allow detecting the position of
From the calculation of intensities, the row with greater the license plate.
intensity is obtained. Once this index is obtained, the
image 81 is trimmed up and down, size enough to contain
the license plate.

Fig.7 showing the edge and position of the license plate.

3.2 Edge detection

the image is then subjected to a binarization process in

order to detect its edges. The Canny algorithm is reused
Fig.6 crop the plate region
with the same parameters as in the previous steps.

3. Location and extraction of the license plate

Once the image is trimmed, the last step is to locate and

extract the license plate. To do this, the image is subjected
to several processes: Fig.8 canny algorithm for the license plate.

1- Morphological transformations

679

Copyright (c) 2017 International Journal of Computer Science and Network. All Rights Reserved.
IJCSN - International Journal of Computer Science and Network, Volume 6, Issue 6, December 2017
ISSN (Online) : 2277-5420
www.IJCSN.org
Impact Factor: 1.5

3.3 Image convulsion

In order to facilitate the detection of the regions, a

convulsion process is performed in the binarized image.
This convulsion will serve to soften the image and reduce
the number of connected components, in order to Fig.12 final step from Calculation and extraction of regions to detect the
number
minimize false positives. The following filter will be
used:
From the image with the filled regions each one is
msk = [0 0 0 0 0; identified.
0 1 1 1 0;
3.5 Removal of the license plate
0 1 1 1 0;
As we saw in the previous image, the region of the license
0 1 1 1 0; plate is the one with the largest area. To obtain the
registration, the regions are sorted by area in descending
0 0 0 0 0;]; order. The first region should correspond with the
registration.
The result is as follows:
From the coordinates of the region of the position
registration is calculated and extracted from the image.

Fig.10 convulsion the license plate

3.4 Calculation and extraction of regions Fig.13 region of license plate is detect

The next step is to calculate the existing regions. This will Acknowledgment
allow us to later identify the enrollment since it will be a
The work of H.AlHAMZAWI was supported by the
region itself. The process consists of several steps:
Stipendium Hungaricum Scholarship.
1- We label the connected regions. In our case, we
also indicate that the connected regions must 4. Conclusion
have at least 8 connected objects.
2- The second step is to fill in the connected The method described in this communication has shown a
regions. relatively high success rate when used in sequences of
3- Finally, the regions are identified. images. The phase of locating license plates handles
complex scenes quite well, demonstrating a high
Below are the images resulting from steps 2 and 3. discriminatory power. However, it would be desirable to
improve the character recognition in the future as well as
the location of the license plate sides.

Reference
[1] Sunghoon Kim, Daechul Kim, Younbok Ryu,
Gyeonghwan Kim (2002). “A Robust LicensePlate
Fig.11 the result for 1 and 2 steps in Calculation and extraction of
regions Extraction Method under Complex Image Conditions”,
ICPR’02, p216-219 vol.3.
[2] Shyang-Lih C., Li-Shien C., Yun-Chung C., and Sei-
In the previous image, we can see the filling of connected Wan C. (2004) “Automatic License Plate Recognition”.
regions from the image obtained in the previous steps. IEEE Trans. On Intelligent Transportation Systems, Vol.
5, No. 1, pp.42-53.
[3] Transport for London, http://www.tfl.gov.uk

680

Copyright (c) 2017 International Journal of Computer Science and Network. All Rights Reserved.
IJCSN - International Journal of Computer Science and Network, Volume 6, Issue 6, December 2017
ISSN (Online) : 2277-5420
www.IJCSN.org
Impact Factor: 1.5

[4] Borba, B., Vasconcelos, C., Albuquerque, M. et

al.(2002), “Localization of Brazilian Vehicles Plates
Using Frequency Analysis”. SIBGRAPI’02, p 408-.
[5] Martín, F., Borges, D., (2003). “Automatic car plate
recognition using partial segmentation algorithm”,
SPPRA 2003, 404-045.

681

Copyright (c) 2017 International Journal of Computer Science and Network. All Rights Reserved.