Automatic license plate recognition in the Mercosur standard based on deep machine learning

• Grant number: 22/05407-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2022
• End date: March 31, 2023
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Murilo Varges da Silva
• Grantee: Lucas Mity Kamado
• Host Institution: Instituto Federal de Educação, Ciência e Tecnologia de São Paulo (IFSP). Campus Birigui. Birigüi , SP, Brazil

Abstract

Optical Character Recognition (OCR) for automatic identification of license plates from images enables a large number of relevant applications. To perform automatic license plate recognition from images, it is necessary to perform three main steps: I. locate the plate in the image, II. segment the characters and III. recognize segmented characters, methods that solve this problem, are generally referred to as Automatic License Plate Recognition (ALPR). Although there are already several techniques proposed in the literature to solve this problem, they cannot achieve good accuracy rates for all types of license plates, for example, each country adopts a different plate standardization (size, shape, color, font, and standard of digits). Therefore, in this project, we propose a study on computer vision and pattern recognition techniques for the automatic recognition of license plates in the new Mercosur license plate standard and in the old 3-letter standard.(AU)