Authors : S. Sadeghi, M. Pourahmadi and M.R. Shayesteh
Abstract: In recent years, as a result of many applications in daily lives such as the management of stolen cars, management of parking lots and traffic flow control, the necessity of intelligent and automatic recognition of cars’ License Plates (LP) is swiftly increasing. Taking into account the specific application to define the algorithms, hardware and the network structure, the process of recognizing involves localization of the license plate region, exploiting of the plate characters and identification of the characters. This work is based on a designed algorithm to recognize the plates by means of the pictures taken under wide-ranging environmental conditions from bright direct sunlight to poorly illuminated conditions at night as well as diverse angles and distances. The threshold for license plate region is determined by using Otsu’s thresholding method and the plate features. Vertical and horizontal histograms are used for character segmentation. At last, recognition of Iranian plate characters is achieved by a new algorithm based on neural network. Simulations are performed using MATLAB Software.
S. Sadeghi, M. Pourahmadi and M.R. Shayesteh, 2017. An Efficient Method for License Plate Recognition System Based on Neural Networks. International Journal of Signal System Control and Engineering Application, 10: 126-133.