Car Plate Recognition By Neural Networks And Image Processing Using Integration Of Wavelets

Car Plate Recognition by Neural Networks and Image Processing Using Integration of Wavelets Jamal Fathi Abu Hasna Abstract-This paper presents an experimental system for the recognition of North Cyprus-style car license plates. Images are usually taken from a camera at a toll gate and preprocessed by a fast and robust 1-D DFT scheme to find the plate and character positions. We examine a new wavelet video method of processing signals for continuous-wave.

Keywords- Feedforward, Denoising, WaveNet,

and Discrete

Fourier Transform.

C

I. INTRODUCTION

ar license recognition is important in several fields of application: traffic control in restricted areas; automatic payment of tolls on highways or bridges; general security systems wherever there is the need of identifying vehicles. Some approaches exist and have been described in literature. They are mainly based on pattern matching and normalized correlation with a large database of stored templates. In this paper we describe an experimental system for the recognition of Cyprus-style car license plates. The system is based on the use of a feedforward neural network (FNN). This learning approach has been shown to guarantee high rates of convergence and properties of stability and robustness of the solution. The data at hand consist of digitized images of cars, acquired by a high-resolution x4 photo camera and collected in a Photo CD. The processed images (see Fig.1) are 390 by 480 pixels wide. The distance and the angle of view simulate a car passing through a toll gate.

The recognition process starts with the search and the extraction of the portion of the original image containing the car plate, or even any part of the car. The characters contained in the plate are localized by a robust processing using a non-traditional Discrete Fourier Transform (DFT), and subsequently isolated and classified by the neural network. The scores are validated by a post-processor which takes into account the syntax of Cyprus-style plates. The processing of video data in real time is considered to be somewhat impractical given the current state of technology. The utility of such processing in real-world applications would therefore seem to be limited. However, recent developments at Trident Systems, Incorporated have made available real-time wavelet processing of video, in the form of the WaveNet technology [2]. Also, in the future a variety of fast architectures for computing wavelet transforms will surely be developed. With respect to a recently published work, our approach is able to reduce the complexity of the learning phase (no feature extraction and pattern matching are required). The character recognition has been speeded up by the parallel architecture of the FNN. The algorithm has been tested on a workstation and the Matlab software.

Fig. 1 Original Image

Manuscript received June 26, 2006. This work was supported in part by the Near East University, Electrical & Electronics Engineering, North Cyprus, Turkey via Mersin-10, KKTC. Jamal Fathi Abu Hasna was born in Tulkarm on November 19, 1964. He came to the North Cyprus in 1997, and graduated with a Bachelor of Science degree in Electrical & Electronics Engineering from Near East University in 1998-99. He continued to the graduate program at Near East University and completed the Master of Science in 20002001 and continued the Doctor of Philosophy degrees in 2001 and now in process.

II. PREPROCESSING WITH WAVELET DENOISING To improve performance for noisy Doppler signals, we apply Donoho’s O (n) wavelet denoising algorithm [3]. The algorithm first does the discrete wavelet transform with Mallat’s pyramid algorithm [4]. The pyramid algorithm computes the transform for some J dyadic levels of scale, resulting in vectors of detail and smooth wavelet coefficients d1, d2,…,dJ-1,sJ. The algorithm then shrinks the detail coefficients for scales

His current interests include neural computing, adaptive signal processing, cellular communications, control systems, and wavelets, Phone: 00905338658472, mail: [email protected],

j  J  1 to obtain d1 , d2 ,..dJ 1 . Here the d j are d   (d ) , j

 j j

j

  ( x) is a nonlinear threshold shrinkage function

where

given by

if | x |  0   ( x )    sign( x)(| x |  ) if | x |  This threshold shrinkage function is shown in Fig 2. δλσ(x)

-λσ x λσ

Fig. 2 Nonlinear threshold shrinkage function for wavelet denoising

III. ARTIFICIAL NEURAL NETWORKS FOR PATTERN RECOGNITION The continuous wavelet transform correlates a Doppler signal with time-localized wavelets at various scales and shifts. It gives the change in local signal scale over time, which in this case is the Doppler period or inverse frequency. When a moving window is placed on the incoming Doppler signal and the windowed signal is wavelet transformed, the corresponding time-varying transform imagery constitutes video. Samples of this wavelet-generated video over time then form signal features for pattern recognition neural networks. These networks are then trained to extract the Doppler frequency shift over time. This frequency shift is critical information for proximity sensing. The continuous wavelet transform constitutes a frame rather than a basis. Such a redundant representation allows more flexibility in the selection of signal features. In terms of the most efficient signal representation, these features should be orthogonal. However, such a representation in which the features are completely independent is less robust with respect to noise immunity and fault tolerance. The search for the best representation is therefore a tradeoff between redundancy and robustness [5].

The threshold shrinkage function δλσ(x) is parameterized by a threshold λand an estimate of the standard deviation of the noise σ. We use a universal threshold

 j  2log( N ) ,

where N is the number of data samples [56]. For  we use the median absolute deviation, which is a robust estimation of standard deviation. Finally, the wavelet denoising algorithm computes the inverse discrete wavelet transform using the new coefficients

Fig. 4 Noisy Image

d1 , d2 ,..dJ 1 , d J , s J This results in a non-parametric estimate of the signal without the noise. The entire wavelet denoising algorithm is shown in Fig. 3. Coefficient

Fig. 5 Denoised Image

Shrinkage d1

d2

Noisy Image DWT

d3

Denoised Image Inverse DWT

d4

s4 Wavelet Coefficients

Fig. 3 Wavelet denoising algorithm, where noisy Image and the denoised one are shown in fig 4, and fig 5.

We extract the Doppler shift with feedforward multilayer neural networks, known as multilayer perceptrons [6]. After computing the continuous wavelet transform of the denoised Doppler signal, we sample the transform coefficients to provide inputs for the multilayer perceptrons. The networks are trained with the Levenberg-Marquardt [7][8] rule to provide the Doppler shift at a given time. This rule is a powerful generalization of gradient descent that employs an approximation of Newton’s method. It is much faster than standard gradient descent algorithms such as backpropagation, although it does require more memory. IV. PROPOSED MODEL In our scheme we place a window of fixed width over the incoming signal, so as to localize the processing near the present time. We then perform a continuous wavelet transform on the signal within the window, resulting in an image of the transform. As the signal window then moves forward in time, the corresponding sequence of transform

images forms a video. From this wavelet transform video, we then extract features as input to pattern recognition algorithms such as artificial neural networks. This is shown in fig 6. Our model has the generation of wavelet transform video from time varying camera proximity fuze signal. Temporary expansion of dimensionality allows us to extract salient features, leading to reduced computational complexity OneDimensional window Sensor signal Time-Varying signal in Continuous wavelet window transform Shift

Time Moving In Time Two-Dimensional window

Scale Wavelet Video features

Pattern Moving In Recognition Time Video From timeNeural Net varying Image in window Fig. 6 Proposed Model

Proximity sensing is widely applied in manufacturing automation and robotics. More recently, there has also been a strong interest in proximity sensing for automobile collision avoidance. Our proposed method is also applicable to processing signals in sonar sensors. V. ADVANTAGES OF OUR APPROACH Besides, in our proposed scheme, wavelet video processing is not a particular computational hindrance, but rather allows salient features to be extracted via the wavelet coefficients. Because of the quality of the wavelet video features, it is likely that fewer numbers of inputs will be needed for pattern recognition. In this sense our scheme could be considered to be a form a data compression. In particular, it seems to be a form of data compression that is ideal for pattern recognition. The multiresolution nature of wavelets also allows us to explore the tolerance of imprecision in the processing of signals. This provides the freedom to tailor the design of the sensor to the resolution requirements of the signals being processed. This tolerance of imprecision is in the spirit of fuzzy logic, but in this case the imprecision is in the scale of the signal structures rather than in the membership of sets. The important idea is that useful information in signals is generally found at the larger scales (lower frequencies). The

less useful, smaller scale signal structures can therefore be disregarded. Neglecting unnecessary details allows a reduction in the amount of data to be processed. This in turn reduces the complexity of the processing, leading to improvements in processing time, system size, and system cost. This reduction of data through the explicit use of scale is a powerful form of data compression. While there are several other strategies for data compression, this one has the advantage of being based on the extraction of signal features. Through wavelet transform time integration, a single coefficient provides the correlation between the signal and a wavelet at a particular scale and time shift. Wavelets are known to provide good signal features for pattern recognition algorithms such as artificial neural networks. Indeed, natural sensors such as eyes and ears carry out wavelet-type processing. The continuous wavelet transform effectively increases the dimensionality of the signal representation from one to two. While this might cause some concern at first glance, it is really not a problem. The reason is that the wavelet representation will be used to extract signal features only. Thus the pattern recognition neural networks need not suffer from the “curse of dimensionality.” After all, the extracted features are of a single dimension only, so that the increase in dimensionality is only temporary. Indeed, because of the high quality of wavelet features, it is quite possible that fewer features will be needed, and that recognition performance will be improved. Mallat’s multiresolution analysis [9] leads to discrete orthogonal wavelets at dyadic scales and shifts, implemented via the efficient pyramid algorithm. These discrete wavelets have been successful in many applications, particularly data compression. However, discrete wavelets have limited utility for pattern recognition problems. This is because interesting signal structures are not constrained to follow such power-oftwo patterns. In particular, discrete wavelet transform coefficients are shift-variant, which in general causes problems for pattern recognition. In contrast, the continuous wavelet transform has coefficients at all scales and shifts, not just dyadic ones. The continuous transform therefore has the desirable property of shift invariance. Another advantage of continuous wavelets is that they have less stringent requirements for admissibility, which allows a wider choice of basis functions. They also have the possibility of being basis functions for adaptive wavelet networks. Through the inclusion of all scales and shifts, the continuous wavelet transform effectively increases the dimensionality of the signal representation. That is, the representation is made to be a function of two variables rather than one. We note that the discrete wavelet transform introduces no such increase in dimensionality, since the number of transform coefficients is the same as the number of signal sample points. This is because the discrete wavelet transform employs an orthonormal basis rather than an over complete frame.

However, the fact that we are using the continuous wavelet transform coefficients merely for feature extraction means that we need not be plagued by the curse of increased dimensionality. In particular, the goal is to use only the relatively few coefficients that provide the best features. In fact, the use of such high quality features may well mean that fewer numbers of inputs will be needed for the pattern recognition neural networks. Of course, these high quality features are also likely to improve the performance of the neural networks. Our scheme could therefore be considered a form a data compression. The temporary increase in dimensionality could then improve compression quality, at least when measured with respect to pattern recognition performance. If we disregard the issue of dimensionality, it might still be argued that computation of the discrete wavelet transform is faster, which has complexity O (n) . However, a continuous wavelet transform implemented via the fast Fourier transform has complexity O (n log n) , which is still quite acceptable for many applications. Also, a continuous wavelet transform has the potential for massive parallelism, and allows the possibility of adaptive wavelet bases [10]

VI. SOME EXPERIMENTS These practical experiments taken using the proposed model for not only the plates, also for the body of the car, and these experiments are as shown in fig 7 below.

(e)

(f)

Fig. 7 (a), (b), and (c) Samples of wavelet-generated video for training and testing inputs to pattern recognition neural networks, while (d), (e) and (f) are the outputs of the proposed model.

VII. CONCLUSION We have just demonstrated the effectiveness of features extracted from wavelet-generated video, and We tested the pattern recognition performance of such features in the estimation of time varying Doppler shift from noisy sensor signals. In particular, we sampled the wavelet video; we saw that the frequency estimation performance of the neural networks is overall good.

(a)

(b)

ACKNOWLEDGMENT My deepest thanks are to Prof. Dr. Senol Bektas, and to Prof. Dr. Fakhraddin Mamedov. I would like to express my gratitude to my collegues in the department. Also I would like to express my gratitude to my family. (c) REFERENCES [1]. R.Parisi, E.D. Di Claudio, G. Orlandi and B.D. Rao, “A generalized learning paradigm exploiting the structure of feedforward neural networks,” IEEE Trans. on Neural Networks, vol.7, no.6, November 1996. [2]. P. Comelli, P. Ferragina, M. Notturno Granieri, and F. Stabile, “Optical recognition of motor vehicle license (d)

plates,” IEEE Trans. On Vehicular Technology, Vol. 44, No. 4, November 1995, pp. 790-799. [3]. S.Haykin, Neural Networks-A Comprehensive Foundation, IEEE Press, 1994. [4] A. Papoulis, Probability Random Variables and Stochastic Processes, Mc Graw-Hill, New York, 3rd Edition, 1991. [5] W.A. Gardner, Statistical Spectral Analysis: A NonProbablistic Theory, Prentice Hall, 1988. [6] P.J. Huber, Robust Statistics, John Wiley, New York, 1981. 0-7803-4455-3/98/$10.00 (c) 1998 IEEE [7]. Antoniadis, A.; G. Oppenheim, Eds.(1995), Wavelets and statistics, Lecture Notes in Statistics 103, Springer Verlag. [8]. Burke Hubbard, B. (1996), The world according to wavelets, AK Peters, Wellesley. The French original version is titled Ondes et Ondelettes. La saga d'un outil mathématique, Pour la Science, (1995). [9]. Chui, C.K. (1992a), Wavelets: a tutorial in theory and applications, Academic Press. [10]. Cohen, A. (1995), Wavelets and multiscale signal processing, Chapman and Hall. [11]. Cohen, A.; I. Daubechies, B. Jawerth, P. Vial (1993), "Multiresolution analysis, wavelets and fast wavelet transform on an interval," CRAS Paris, Ser. A, t. 316, pp. 417-421. [12]. Coifman, R.R.; Y. Meyer, M.V. Wickerhauser (1992), "Wavelet analysis and signal processing," in Wavelets and their applications, M.B. Ruskai et al. (Eds.), pp. 153178, Jones and Bartlett. [13]. Teolis, A. (1998), Computational signal processing with wavelets, Birkhauser. [14]. Vetterli, M.; J. Kovacevic (1995), Wavelets and subband coding, Prentice Hall.