An Intelligent License Plate Recognition System: Zyad Shaaban
An Intelligent License Plate Recognition System: Zyad Shaaban
An Intelligent License Plate Recognition System: Zyad Shaaban
7, July 2011
55
Summary
Automatic license plate recognition plays an essential role in
many applications and a number of methods have been proposed.
These applications range from complex security systems to
common areas and from parking admission to urban traffic
control. License plate recognition has hard properties due to
varied effects as fog, rain, shadows, irregular illumination
conditions, variable distances, cars velocity, scene's angle on
frame, plate rotation and conservation, number of vehicles in the
scene and other. These effects make plate recognition much more
complex and difficult than the traditional pattern recognition
systems. The proposed system is composed of the following
three stages: 1) detection and extraction of a license plate area by
video camera 2) segmentation of the plate characters and digits;
and 3) character and digit recognition. The plate image area
extracted using the proposed system is segmented into disjoint
characters for the character recognition phase, where the neural
network classifier is used to recognize the characters based on
edge moment invariants and principal component analysis
features of wavelet coefficient matrix. The main goal of this
research is to develop a new plate recognition system with
intelligent issues surpass the systems introduced in literature and
to reduce many of the restrictions in the working environment.
here the part of summary.
Key words:
License plate detection; character recognition; moment
invariants; principal component analysis; neural networks
1. Introduction
All Computer vision and character recognition
algorithms for license plate recognition are used as core
modules for intelligent infrastructure systems like
electronic payment systems (toll payment and parking fee
payment) and freeway and arterial management systems
for traffic surveillance [1,8-12]. A system for automatic
car license plate recognition consists of a camera, a frame
grabber, a computer, and custom designed software for
image processing, analysis and recognition [2].
In [1], a comprehensive and critical survey of up-todate License plate recognition (LPR) methods is presented.
This survey focused on LP detection, character
segmentation and character recognition. The extraction of
the plate region methods are categorized into the following
Manuscript received July 5, 2011
Manuscript revised July 20, 2011
IJCSNS International Journal of Computer Science and Network Security, VOL.11 No.7, July 2011
56
2. Preprocessing
2.1 Image Acquisition and Filtering
The video is captured using a digital camera by the
program Auto Movie Creator 3.26 [14] . The video file is
saved in wmv format. Then the file is divided into
sequence of frames(isolated images) by mmread matlab
program [15]. The input isolated image is filtered to
improve the quality of the image by median and Wiener
filters [22].
2.4 Normalization
There are several methods to compute the skew angle
for the vehicle license plate: horizontal skew, vertical
skew and combination of both horizontal and vertical
skew [17]. PCA is used to analyze data in several
application fields such as pattern recognition and image
processing. This method is a statistical technique that is
used to find the Eigen vectors. These vectors represent the
relation of these data sets along that line [16]. In this paper,
the image of the car plate is normalized based on PCA
technique to determine the skew angle[16-17]. The
extracted plate image is rotated using negative of the
calculated skew angle.
2.5 Segmentation
3. Feature Extraction
In the following subsections, different feature
extraction approaches will be used to extract some useful
features to use them in the recognition stage.
IJCSNS International Journal of Computer Science and Network Security, VOL.11 No.7, July 2011
57
4. Neural Networks
In this paper, back propagation neural network
(BPNN) is used. BPNN uses a supervised learning
technique, and is built from simple computational units
referred to as neurons. Neurons are connected by weighted
links [26]. The architecture for a BPNN is given in Fig. 1.
For more details, refer to [26].
and
Figure 1. The architecture for a BPNN
and
58
IJCSNS International Journal of Computer Science and Network Security, VOL.11 No.7, July 2011
IJCSNS International Journal of Computer Science and Network Security, VOL.11 No.7, July 2011
59
IJCSNS International Journal of Computer Science and Network Security, VOL.11 No.7, July 2011
60
References
[1]
Christos-Nikolaos
E.
Anagnostopoulos,
Ioannis
E.
Anagnostopoulos, Ioannis D. Psoroulas, Vassili Loumos and
Eleftherios Kayafas, License Plate Recognition From Still Images
and Video Sequences: A Survey, IEEE TRANSACTIONS ON
INTELLIGENT TRANSPORTATION SYSTEMS, VOL. 9, NO. 3,
SEPTEMBER 2008 377.
[2]
[3]
[4]
[5]
[6]
[7]
TABLE I.
The Approach
PCA
MI
Wavelet
The
Proposed
Approach
The
Recognition
Rate
95%
96.3%
93.3%
97.6%
[8]
[9]
7. Conclusion
In this paper, an experimental comparison has been
performed between PCA, MI, Wavelet approaches which
are based on the original binary images of the characters
and the proposed system which is based on EMI and PCA
of wavelet matrix of the edge image. The proposed system
achieved high recognition rate. The proposed system will
be developed in the future to tackle the problem of
recognizing the color of plates and the old types of plates.
Acknowledgment
This work has been supported by a research grant
(1432/26/2) from the Deanship of Scientific Research,
University of Tabuk. The author would like to thank the
university for supporting this research.
[10]
[11]
[12]
IJCSNS International Journal of Computer Science and Network Security, VOL.11 No.7, July 2011
[13] Xian-Sheng
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
HUA ,
Xiang-rong Chen ,
LIU
Wenyin , Hong-Jiang ZHANG, Automatic Location of
Text in Video Frames, Proceeding of ACM Multimedia
2001 Workshops: Multimedia Information Retrieval (2001)J.
Clerk Maxwell, A Treatise on Electricity and Magnetism,
3rd ed., vol. 2. Oxford: Clarendon, 1892, pp.6873.
http://www.deskshare.com/amc.aspx
http://www.mathworks.com/matlabcentral/fileexchange/802
8-mmread
Chirag N. Paunwala, Suprava Patnaik and Manoj
Chaudhary, An Efficient Skew Detection of License Plate
Images Based on Wavelet Transform and Principal
Component Analysis, 2010 International Conference on
Signal and Image Processing,page.17-22.
Nipa D. Modi, Chirag N. Paunwala,Chintan K. Modi and
Suprava Patnaik, Skew Correction for Vehicle License
Plates using Principal Component of Harris Corner Feature,
2011 International Conference on Communication Systems
and Network Technologies, page.339-343.
Ehsan Nadernejad, Sara Sharifzadeh and Hamid Hassanpour,
Edge Detection Techniques: Evaluations and Comparisons,
Applied Mathematical Sciences, Vol. 2, 2008, no. 31, 1507
1520.
Mohsen Sharifi, Mahmoud Fathy, Maryam Tayefeh
Mahmoudi, A Classified and Comparative Study of Edge
Detection Algorithms, Proceedings of the International
Conference on Information Technology: Coding and
Computing (ITCC.02),page.
Marius Bulacu and Lambert Schomaker, Text-Independent
Writer Identification and Verification Using Textural and
Allographic Features, IEEE TRANSACTIONS ON
PATTERN
ANALYSIS
AND
MACHINE
INTELLIGENCE, VOL. 29, NO. 4, APRIL 2007, page.
701-717.
Faisal Shafait, Daniel Keysers, and Thomas M. Breuel,
Performance Evaluation and Benchmarking of Six Page
Segmentation Algorithms, IEEE Transactions on Pattern
Analysis and Machine Intelligence, June 2008 (vol. 30 no.
6), pp. 941-954.
A. Khireddine, K. Benmahammed, and W. Puech, Digital
image restoration by Wiener filter in 2D case, Advances in
Engineering Software 38 (2007) 513516.
Mia Hubert, Peter J Rousseeuw, Karlien Vanden Branden.
Technometrics. February 1, 2005, 47(1): 64-79.
Caixia Deng, Yuling Qu And Lijuan Gu, Characterization
Of Image Space Of A Wavelet Transform, International
Journal of Wavelets, Multiresolution and Information
Processing Vol. 4, No. 3 (2006) 547557.
http://www.mathworks.com/help/toolbox/wavelet/ref/wpfun
.html
R. Gonzalez and R. Woods, Digital Image Processing, 2nd
ed., Prentice-Hall, New Jersey, 2001.
61