Recuperación de información de placas vehiculares en base al método optical character recognition (OCR)

• Autor: Castañeda Haro, Carlos Enrique
• Formato: tesis de grado
• Fecha de Publicación: 2020
• Institución: Universidad de Lima
• Repositorio: ULIMA-Institucional
• Lenguaje: español
• OAI Identifier: oai:repositorio.ulima.edu.pe:20.500.12724/12343
• Enlace del recurso: https://hdl.handle.net/20.500.12724/12343
• Nivel de acceso: acceso abierto
• Materia: Optical character recognition; Automobile license plates; Image processing; Reconocimiento óptico de caracteres; Placas de automóviles; Proceso de imágenes; https://purl.org/pe-repo/ocde/ford#2.02.04

Vehicle impersonation is one of the main problems based on license plates adulteration. Furthermore, license plate recognition has been field of study of computer vision, due to the high impact it has on intelligent transport, road safety, etc. This work presents an OCR process based on license plate recognition. To achieve this, we have established a rectified license plate database to experiment. The process consists of an initial noise filtering to the images, to later segment the region of the plate that contains the characters, then, perform a pre-processing on the images and, finally, isolate each of the characters to apply the OCR individually to each one of them. The results show an 89.33% accuracy in the license plate recognition in accordance with similar researches, and a 96.11% accuracy in the recognition of each character individually.