An Automatic Vehicle License Plate Recognition Based on Edge Detection and Artificial Neural Network

by Open Science Repository Computer and Information Sciences
(December 2013)

Abstract


This paper presents a method to use the 2-D Haar Discrete Wavelet Transform and Artificial Neural Network for the vehicle license plate recognition. This thesis consists of three main sections. The first part is to license plate location and extraction the license plate image. The second part is the training of the license plate. The third part is to use window scan make license plate recognition. Traditional license plate recognition system is complex. The present paper is to use a 2-D Haar Discrete Wavelet Transform license plate twice, to select only the second conversion coefficients of the low-frequency part of the image pixels, This paper use cut image of the pre-treatment, thereby reducing the pixels of the scanned image, so which can be improved recognition rapid implementation and reduce computer memory data. This paper is a scan license plate recognition, we do not use individual character recognition. This new method of identification, the experimental results of license plate recognition rate can be as high as 95.33%.

Keywords: vehicle license plate recognition, 2-D Haar Discrete Wavelet Transform, artificial neural network.


Full text


An Automatic Vehicle License Plate Recognition Based on Edge Detection and Artificial Neural Network

Click here to post comments

Return to journal abstracts.