Home // COMPUTATION TOOLS 2018, The Ninth International Conference on Computational Logics, Algebras, Programming, Tools, and Benchmarking // View article

License Plates Recognition of Mexican Private Vehicles

Authors:
Carlos Hiram Moreno-Montiel
Benjamin Moreno-Montiel
Nicolás Trejo
Martha Soto

Keywords: license plate recognition; bayesian classifier; artificial neural network; correlation factor; principal component analysis; Hu’s moments.

Abstract:
In most of the investigations about the recognition of license plates from different countries it is assumed that they have a white background, without texture patterns and with black characters. Vehicle registration plates in Mexico are different because they have different texture patterns and colors in the background depending on the State of the Republic; that is why the algorithms of recognition of these plates are not always successful. This article proposes an algorithm for the recognition of vehicle registration plates of Mexico considering three phases: A) Normalization and Binarization of the plate, which is achieved by using a threshold factor, which separates dark colors that form letters from the clearings that are at the bottom. B) Characterization C) Modeling of symbols by technics such as Hu’s moment, Fourier descriptors and correlation cross factors and D) Classification, where we have used comparisons between different techniques of template matching, Bayesian classifier and Artificial Neural Networks to process images of plates from different states. The results obtained are discussed at the end of the present paper.

Pages: 13 to 19

Copyright: Copyright (c) IARIA, 2018

Publication date: February 18, 2018

Published in: conference

ISSN: 2308-4170

ISBN: 978-1-61208-613-2

Location: Barcelona, Spain

Dates: from February 18, 2018 to February 22, 2018