Banca de QUALIFICAÇÃO: VICTOR HUGO MACEDO AMORIM

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : VICTOR HUGO MACEDO AMORIM
DATE: 04/03/2022
TIME: 15:00
LOCAL: Online
TITLE:

A Hybrid Approach to Automatized Vehicle License Plate Detection, Segmentation and Classification

KEY WORDS:

MERCOSUR plates, detection, segmentation, recognition

PAGES: 143
BIG AREA: Ciências Exatas e da Terra
AREA: Ciência da Computação
SUBÁREA: Metodologia e Técnicas da Computação
SPECIALTY: Processamento Gráfico (Graphics)
SUMMARY:

Over the last decade there has been a great increase on the Brazilian vehicle fleet, which implies in a major volume to many previously manually controlled problems. Traffic control, stolen car detection and restricted locations access control are examples of those situations, usually handled by automatized license plate recognition (ALPR) based systems. Those systems are responsible for identifying vehicles on images or video sources, mostly by locating its licence plate number. ALPR based systems are usually built on a three step approach: license plate detection, character segmentation and character recognition. A large number of methods have been studied over the years for each of those, from image processing to machine learning. A few years past Brazilian vehicle licensing system changed, by switching the old license plate design to a new MERCOSUR unified design. Since the transition has been very slow, both designs are now present on the vehicle fleet. The proposed approach is based on well known image processing and machine learning algorithms, as well as the rising convolutional neural networks. From those will be built an ALPR based system that is able to deal with both plate designs, focusing on low cost computational platforms.

BANKING MEMBERS:
Presidente - 2177445 - BRUNO MOTTA DE CARVALHO
Interna - 1350250 - ANNE MAGALY DE PAULA CANUTO
Externo ao Programa - 1172485 - ANTONIO CARLOS GAY THOME