DVR & Dr.

HS MIC COLLEGE OF TECHNOLOGY,

ECE Department.

RECOGNITION OF VEHICLE NUMBER PLATE

USING MATLAB
Name1,Name2

Abstract:- The vehicle number plate segmentation and finally, the character
recognition. Among this, Segmentation is the most
recognition automatically controls access to a important part in the method because it affect the
secured area for authorized members. We tested system accuracy.
our method for number plate recognition which
includes five main testimonials first is pre- Vehicle number plate recognition mainly
processing of the input image, then cropping the consists of the five main testimonials. All the
steps have their own importance and methods
pre processed image that follows by the image
followed in the number plate recognition. In the
edge extraction operated on the cropped image first stage, pre-processing which reduces the
after this the character segmentation process noise of the image and make it more clearly
algorithm and then finally to the character visible. Next step will crop the image and then,
recognition which gives the improved results the number plate edge extraction is performed.
that shows our method has the average precision This can be done using many simple methods; it
that can be put into practice. extracts the features of the number plate as well
as their shape and symmetry. Then it will
Keywords: ANPR,Number plate,Image edge segment the characters individually to recognise
extraction,Vehicle number plate each character of the number plate using
algorithm which also gives the match degree
recognition,Character segmentation.
results of the each characters of the number
plate.
1.Introduction

The automatic vehicle number plate recognition

portray in real-life applications to spot vehicles
by encapsulating their number plate. Several
works analyze this issue and specify many
unfolding to allocate with this work. This
increasing passion for the improvement of road
safety in various areas like the automatic-toll tax
collection, rush hour law enforcement, parking
transport access control, road traffic supervision
and with the crime prevention. It is executed
with the synthesis of five main procedure flow
one after other in vehicle number plate
recognition method that are the pre-processing
of the query image, cropping the pre-processed
image then the edge extraction applied on the
cropped image that follows the character Fig 1:Flow chart.
 DVR & Dr.HS MIC COLLEGE OF TECHNOLOGY,
ECE Department.

2.Pre-Processing and the numbers are visible to eliminate the

The pre-processing of the number plate unwanted region of the query image.
recognition is the first step to remove the
useless information from the input image I(x, y). 4.The edge extraction using
The noise of the image should be removed and morphological operations
the useful information should be acquired by First, the cropped image C(x, y) should be
enhancing visual appearance of the image. converted to the binary image consists of only
When the useless information is cleared then the 1's and 0's (for black and white). The edge of
greyscale contrast enhancement improves the the image which is directly extracted by the
appearance by brightening the dataset and the morphological basic operations gives the E(x, y)
resultant image is I1(x,y). from the cropped image C(x, y).
This is performed by using two basic
morphological operations that are dilation and
erosion.

(a)

(b)
Fig. 2 Original input vehicle number plate (a)
plate1.jpg, (b) plate2.jpg Fig. 4 Morphological edge detection applied
on plate1.jpg (a) Binarized images, (b)
Morphological edge detection images, and (c)
3.Cropping Dilated images using morphologic operation.
Next, precisely containing the useful
information from the I1(x,y) then by using the
binary coordinates to crop the vehicle number
plate which is only required for further
processes that results in the C(x, y) image. At
this stage, the characters and numbers that were
cropped from the query number plate may also
contain garbage objects as well as useful
objects.

(a)

Fig. 5 Morphological edge detection applied

on the plate2.jpg. (a) Binarized images, (b)
(b) Morphological edge detection images, and (c)
Fig. 3 Cropped images of the vehicle number Dilated images using morphologic operation.
plates (a) plate1.jpg cropped image, (b)
plate2.jpg cropped image These two methods are based on the
This can be obtain by simply cropping the query fundamental morphological operations by
image only to the extend where the characters giving the limits 1.1 and the image terminals by
selecting different size and extensions converted
 DVR & Dr.HS MIC COLLEGE OF TECHNOLOGY,
ECE Department.

to images of grey levels and then remove the ii. after this, it calculates the correlation plane
entire unwanted region. This process, first gives for each multiplication,
the number of pixels connected together in a iii. it identifies the maximum peak value as
sequence to form a group of connected objects. max(out) to the correlation for each segmented
Then, it counts the cropped characters and character S(x, y),
number of the connected region from number iv. it sorts the maximum peak correlation values
plate and matches this characters and number from each correlation plane to found the
from the dataset of the template library which is reference image brings about the highest
created to call the objects. correlation peak value, i.e., with the reference
image S(x, y) have the best match with unknown
input image I(x, y).
5.The Character Segmentation
After the edges of the number plates are 6.Results
determined as E(x, y), the character
segmentation is the next process in the The correlation matching algorithm
identification module. Each character of the gives the absolute values of the match degree of
cropped number plate need to be split and the the characters for the two query images namely
acquired image is the resultant image S(x, y). plate1.jpg and plate2.jpg taken from the matlab
The character recognition algorithms are many library of 512*512 sizes. The results were tested
but the recent algorithm proposed is Otsu's and validated using correlation coefficient (CC)
algorithm. This algorithm identifies separate metrics [12]. CC is used to match the characters
objects within the image. It finds, that the region from the template library and the vehicle
of the connected pixels with similar properties number plate.
and also finds the boundaries between the
regions.

The value of CC lies between 0 and 1

where as the value increases towards 1 will be
the best correlation value. The CC value
(a)
calculated for the proposed algorithm for the
recognized image gives the results as Table 1
shown below. This gives the approximation, as
(b) the true result according to all the causes of the
image kept in mind as through the illumination
Fig. 5 Segmented number plates characters factor as well as the environmental factors of
(a) plate1.jpg, (b) plate2.jpg the plate.

6.Character Recognition 7.Conclusion

This paper presents a combination of
The final recognition stage is performed both theoretic practise and the visual results on
by the correlation matching algorithm the vehicle number plate recognition through
implementation and the accuracy rate is the the hybridization of the morphological edge
approximation of value 1 in the F(x, y) image. detection and proposed Otsu's algorithm for the
This final resultant recognized objects of the segmentation. The approximate results show
vehicle implements correlation for the pattern that this approach gives the potential of the
recognition by the following steps: morphological edge detection and segmentation
i. it sequentially multiplies every segmented for vehicle number plate classification that has
objects S(x, y) over the entire reference image been assessed. This method have better
set from the template library, theoretical properties than the traditional filters
 DVR & Dr.HS MIC COLLEGE OF TECHNOLOGY,
ECE Department.

that represent images but still not used for [4] C. Chenyu, C. Baozhi, C. Xin, W. Fucheng,
vehicle verification. The five main modules that Z. Chen, "Application of image processing to
worked together one after one for the improved vehicle license plate recognition," Proceeding to
results even lead to the satisfaction to the 2nd International Conference on Computer
natural moving vehicle number plates. The Science and Electronics Engineering (ICCSEE
purpose to apply this five testimonial approach 2013).
is that to eliminate the unwanted region from [5] A. Conci, J. E. R. de. Carvalho and T. W.
the vehicle number plate and detect the Rauber, "Graphical models for joint
prominent edges which are required for the segmentation and recognition of license plate
recognition of the characters and numbers from characters, "IEEE, Latin America Transactions,
the vehicle number plates. Hence, the Vol.7, No. 5, September 2009.
recognition accuracy is high. The method is [6] M. R. Bai, V.V. Krishna, and J. Sreedevi, "A
quite simple that can be put into practise and new morphological approach for noise removal
also has a good application prospect. cum edge detection," IJCSI International
Journal of Computer Science Issues, Vol. 7,
8.References Issue 6, November 2010 .
[7] D. Ghimrie and Joonwhoan Lee, "Nonlinear
[1] M. Ibrahim and M. Shehata, Automatic transfer function-based local approach for color
license plate recognition (ALPR): A state-of-art image enhancement," IEEE Transaction on
review, IEEE Transaction on Circuits and Consumer Electronics, Vol. 57, No. 2, May
Systems for Video Technology, Vol. 23, NO. 2, 2011.
February 2013. [8] D. F. Martin, J. T. G. Partida, P. A.
[2] J. Arrospide and L. Salgado, "Log-gabor Gonzalez and M. B. Garcis, "Vehicular traffic
filters for image-based vehicle verification," serveillance and road lane detection using radar
IEEE Transaction on Image Processing, Vol. 22, interferometry," IEEE Transaction on Vehicular
No. 6, June 2013. Technology, Vol. 61, No. 3, March 2012.
[3] A. M. Al-Ghaili, S. Mashohor, A. R. Ramli
and A. Ismail, Vertical edge-based car license
plate detection method, IEEE, Manuscript
Received January 25,2012.