Uma contribuição para um sistema computacional de leitura automática de valores em cédulas monetárias

• Ano de defesa: 2013
• Autor(a) principal: Paula, Camilo de Lelis Tosta
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; BR; Programa de Pós-graduação em Engenharia Elétrica; Engenharias; UFU
• Palavras-chave em Português: Processamento de imagens; Técnicas digitais; Redes neurais (Computação); Software; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/14542; https://doi.org/10.14393/ufu.di.2013.227

Image Digital Processing techniques, standards recognizing, and neural artificial nets are not often used in the technology market, because they are hard to work with and expensive for who uses them. When we make research on technology tools we find them fragmented as they solve few problems and come in software with no license for copywriting. Thus, we propose a Computer System for Automatic Reading of Money Banknotes, who will join programming techniques widespread in academia to develop free software that can be used both in the market and for studies in academia, a software with applications on image capture, digital image processing, pattern recognition that uses artificial neural network techniques, coupled to a voice synthesizer, which will assist people with visual needs and even people with difficulties in achieving the necessary focus such as supermarkets. The system can be adapted to any sales place, or even where people work counting high money values. It brings the reliability needed without errors that can cause losses to the company and people as well.