7rty
7rty
7rty
RECOGNITION
ABSTRACT
Automatic Number Plate Recognition (ANPR) has many applications as (highway electronic toll
collection, Police applications, Average Speed Enforcement, Bus Lane Enforcement, Parking
Management Systems, Journey Time Systems, detection of stolen vehicles, etc.).It is worth mentioning
that there is a scarcity in researches that introduce an automatic number plate recognition which depends
on Arabic language.
In this study, a smart and simple algorithm is presented for Egyptian vehicle’s number plate
recognition system. The proposed algorithm consists of two major phases: plate region extraction and
plate recognition. For extracting the plate region, edge detection algorithms and the alignment algorithm
are used. And finally DCT or LDA is used for plate recognition.
The performance of the proposed algorithm has been tested on real images. Based on the experimental
results, we noted that our algorithm shows superior performance special in number plate recognition
phase.
KEYWORDS
Extraction plate region, Number plate recognition, Edge detection, Prewitt edge detector, dilation, a
label connected components technique, Discrete Cosine Transform, Linear Discriminate Analysis,
Euclidean distance.
1. INTRODUCTION
Nowadays vehicles play a very big role in transportation. Also the use of vehicles has been increasing
because of population growth and human needs in recent years. Therefore, control of vehicles is
becoming a big problem and much more difficult to solve.
Due to the rapidly increase in number of vehicles across the world's big cities and one of them is Cairo,
number plate recognition system has become one of the most important digital image processing systems
to be used.
Real time ANPR plays a major role in automatic monitoring of traffic rules and maintaining law
enforcement on public roads. ANPR is used to identify vehicles by only their number plates. Since every
vehicle carries a unique number plate, no external cards, tags or transmitters need to be recognizable, only
number plate.
The field of ANPR and its application has attracted many researchers to search and develop systems
which can process images and get useful information from them. Most previous researches and
applications have faced some kind of poor performance due to the diversity of plate formats, the non
uniform outdoor illumination conditions during image acquisition, noisy patterns connecting characters
and poor edge enhancement.
Accordingly, these researches and applications have in some way restricted their working conditions,
such as limiting them to indoor stationary backgrounds, fixed illumination, limited vehicle speeds,
designated ranges of the distance between camera and vehicle, the best position of the camera and the
maximum vertical view angle of the camera.
So many researches of car identification have been approached by car license plate extracting and
recognition, some of the related work is as follows: Syed Ali Khayam [1], the results presented in this
document show that the DCT exploits interpixel redundancies to render excellent decorrelation for most
natural images. Thus, all (uncorrelated) transform coefficients can be encoded independently without
compromising coding efficiency.
In addition, the DCT packs energy in the low frequency regions. Therefore, some of the high frequency
content can be discarded without significant quality degradation. Such a (course) quantization scheme
causes further reduction in the entropy (or average number of bits per pixel). Lastly, it is concluded that
successive frames in a video transmission exhibit high temporal correlation (mutual information). This
correlation can be employed to improve coding efficiency.
Serkan Ozbay, and Ergun Ercelebi [2], this paper presented application software designed for the
recognition of car license plate. Firstly it extracted the plate location, and then it separated the plate
characters individually by segmentation and finally applied template matching with the use of correlation
for recognition of plate characters. This system is designed for the identification Turkish license plates.
Ibrahiem M. M. El Emary [3], the medical image of human chromosome is treated by this software. It is
entered in filter algorithm to remove the noise of the image that it can contain with cleaning percentage
reach to 91.7%. Then this filtered image entered to segmentation algorithm, which cuts the original image
into most likely segments which can contain one object (one chromosome) or in some bad cases it can
contain more than one object (overlapping or touching chromosomes) then it can classify the human
chromosomes by entering the images after segmentation into the Neural Network to study each segment
and classify the chromosome that is contained in it.
N. Senthilkumaran and R. Rajesh [4], this paper mainly focuses on the study of soft computing
approach to edge detection for image segmentation. The soft computing approaches namely, fuzzy based
approach, Genetic algorithm based approach and Neural network based approach is applied on a real life
example image of nature scene.
Mohamed El-Adawi, Hesham Abd el Moneim Keshk, Mona Mahmoud Haragi [5], this system can
identify license plate with a good accuracy. It deals with the characteristics of the Arabic letters and Hindi
numbers, and has a good ratio (80%) success rate.
Kuo-Ming Hung and Ching-Tang Hsieh [6] the purpose of this paper is to establish a low-cost, highly
efficient dynamic vehicle license plate recognition system. Instead of fixed point installation, the video
camera is installed in a car to capture the license plate of a random vehicle in front.
M. M. Shidore, S. P. Narote [7], an algorithm for vehicle number plate extraction, character
segmentation and recognition is presented. Database of the image consists of images with different size,
background, illumination, camera angle, distance etc. The experimental results show that, number plates
are extracted faithfully based on vertical edge detection and connected component algorithm, with the
success rate of 85%. Character segmentation phase using connected component analysis and vertical
projection analysis works well with the success rate of 80%. The success rate achieved for character
recognition is 79.84%.
Saqib Rasheed, Asad Naeem and Omer Ishaq [8], this paper presents a novel method for detection and
recognition of Islamabad vehicle number plates. The proposed technique of automated number plate
recognition is divided into two modules, one license plate localization module using canny detector and
Hough lines, second license number recognition module using template matching.
Raseena A, Asst. Prof. Muhammad sajeer [9], Application software designed for the recognition of car
license plate. It extracted the plate location with the input image and also with the skew angle, and then it
separated the plate characters individually by segmentation and finally applied template matching with
the use of correlation for recognition of plate characters.
Najeem Owamoyo, A. Alaba Fadele, Abimbola Abudu [10], an algorithm for vehicle number plate
extraction, character segmentation and recognition is presented.
The experimental results show that, number plates are extracted faithfully based on vertical edge
detection and connected component algorithm, with the success rate of 85%. Character segmentation
phase using connected component analysis and vertical projection analysis works well with the success
rate of 80%. The success rate achieved for character recognition is 79.84%.
The paper is organized as follows: Section 2 provides an overview of the overall system. Extracting
the plate region is explained in Section 3. Section 4 deals with recognition of number plates. Section 5
provides experimental results and the paper conclude with Section 6.
The input of the system is the original image of the vehicle number plate captured by 14 MP high
resolution digital cameras. This image is processed through the number plate extractor and its output will
be provided to the recognition part. The proposed system considers the following assumption about the
number plates:
1- The plates have a rectangular shape.
2- The plate has dark characters on a bright white background.
3- The width-height relationship of the number plate is approximately known.
4- The Plates consist of two groups of characters and digits have the same size.
5- The number of characters doesn't affect the recognition process.
6- A metallic bar separates two groups of characters, located in the plate centre and is useless in number
plate recognition operation and may be ignored.
7- A coloured area above characters area is useless in the number plate recognition
operation and may be ignored.
A simple edge detection algorithm is first applied on the image to obtain the rectangular shape that has
the same characteristics of the number plate; once the plate is located the alignment algorithm is
processed on it to make it ready to the next stage.
Figure (2) Process of plate extraction
Prewitt edge detector with a hhigh threshold is applied, dilation for all lines in the horizontal
h directions
is then performed to avoid the brbroken line problem, The small lines that are very unli nlikely to be a part of
the required horizontal line are discarded using a label connected components te technique, Then the
algorithm checks for any horizonontal lines connecting the two vertical right and left edges
edg detected earlier,
which represent the vertical boundaries
bou of the number plate, discarding the top an and bottom lines, If
nothing is detected, the algorithm
thm attempts the same process again using a different threshold
th levels until
a certain edge represent the requi
quired horizontal line, The input image is rotated around
und its centre from an
angle of 7” to -7” in incrementents of 0.5”, with every rotation, the locations of the up upper left and upper
right corners of the character area
ar are calculated at each stage, and then the differe erence between those
values is calculated, The minimu
imum difference recorded represents the required angle gle; this angle is then
applied to the image.
4. PLATE RECOGNITION
The proposed algorithm does es not need a process for segmentation of input imag age of number plate
(character segmentation) but find
inds features in the whole number plate image. LDA (l (linear Discriminant
Analysis) and DCT (Discrete Cos
Cosine Transform) feature extraction methods are usedd to
t classify the given
number plate candidate. This algo
lgorithm may lead the recognition process to have highh accuracy.
a
Training Phase algorism training on data base image by reading images then using one of the feature
Extraction methods as DCT or LDA then a classifier method is used see the following DCT and LDA
flowchart.
2) Test Phase:
This algorithm try to recognize a test image in the training data base images and the minimum distance
classifier method (Euclidean distance) will be used.
Figure (8) Test Phase by DCT and Euclidean method flowchart
6. CONCLUSION
The purpose of this study has been to create an automatic number plate recognition system. It describes
complete intelligent digital image processing system to recognize the Arabic letters and Hindi digits
within the Egyptian vehicle number plate using a sequential multistage approach performed over gray-
scale images, consists of two parts extraction of vehicle number plate from the image, and recognition of
number plate. For extracting the vehicle number plate edge detection algorithm used.
Unlike the traditional methods for number plate recognition, the innovation of the proposed algorithm
is that it does not need a process for segmentation of input image of number plate (character
segmentation) but finds features in the whole number plate image. LDA and DCT feature extraction
methods are used to classify the given number plate candidate. The experimental results show that our
new method is of higher recognition accuracy than using traditional methods. This new approach provides
a good direction for automatic number plate recognition. The system was designed using MATLAB
2013a.
REFERENCES
[1] Syed Ali Khayam,"The Discrete Cosine Transform", Department of Electrical & Computer
Engineering, Michigan State University, March 10th 2003.
[2] Serkan Ozbay, and Ergun Ercelebi, "Automatic Vehicle Identification by Plate Recognition", World
Academy of Science, Engineering and Technology 9 2005.
[3] Ibrahiem M. M. El Emary, " On the Application of Artificial Neural Networks in Analyzing and
Classifying the Human Chromosomes ", Journal of Computer Science, ISSN 1549-3636, 2006.
[4] N. Senthilkumaran and R. Rajesh, "Edge Detection Techniques for Image Segmentation ".
International Journal of Recent Trends in Engineering, Vol. 1, No. 2, May 2009.
[5] Mohamed El-Adawi, Hesham Abd el Moneim Keshk, Mona Mahmoud Haragi, "Automatic License
Plate Recognition", IEEE Transactions on Intelligent Transport Systems, vol. 5, pp. 42- 53, March 2004.
[6] Kuo-Ming Hung and Ching-Tang Hsieh, "A Real-Time Mobile Vehicle License Plate Detection and
Recognition ", Tamkang Journal of Science and Engineering, Vol. 13, No. 4, pp. 433_442 ,2010.
[7] M. M. Shidore, S. P. Narote, "Number Plate Recognition for Indian Vehicles ", International Journal
of Computer Science and Network Security, VOL.11 No.2, Feb. 2011.
[8] Saqib Rasheed, Asad Naeem and Omer Ishaq, " Automated Number Plate Recognition using hough
lines and template matching", Proceedings of the World Congress on Engineering and Computer Science
2012 Vol I WCECS 2012, October 24-26, 2012, San Francisco, USA.
[9] Raseena A, Asst. Prof. Muhammad sajeer, "Automatic Skew Detection and Localisation of Vehicle
License Plate Using Hough Transform ", International Journal of Scientific & Engineering Research,
Volume 4, Issue 8, August 2013.
[10] Najeem Owamoyo, A. Alaba Fadele, Abimbola Abudu ,"Number Plate Recognition for Nigerian
Vehicles", Academic Research International,
ISSN: 2223 – 9944, VOL.4, No. 4, MAY 2013.
AUTHORS
First Author – Captain.Eng. Belal Rashed Aboushanab, Bachelor of Communication Engineering ,
belalaboushanab@yahoo.com
Second Author – Prof.Dr. Hala Abd EL Kader, Ph.D. in Communications and Electronics Engineering,
Hala.mansour@gmail.com
Third Author – Dr. Hanaa Rafaat , Ph.D. in Communications and Electronics Engineering.
Fourth Author – Dr. Mohamed Sharaf , Ph.D. in Communications and Electronics Engineering,
Director of the Interior Ministry Wireless Networks,msharaf5@hotmail.com