Reconhecimento de edifícios utilizando a transformada de Gabor, a transformada wavelet e a rede neural perceptron de múltiplas camadas

• Ano de defesa: 2018
• Autor(a) principal: Silva, Leandro Aureliano da
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Elétrica
• Palavras-chave em Português: Reconhecimento de edifícios; Recognition of buildings; Representação de wavelet Gabor; Gabor wavelet representation; Extração de características; Extraction of features; Rede Perceptron de múltiplas camadas; Multi-layer Perceptron; Engenharia elétrica; Processamento de imagens; Percepção geográfica; CNPQ::CIENCIAS EXATAS E DA TERRA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/21665; http://dx.doi.org/10.14393/ufu.te.2018.776

Recognition of buildings is a very difficult task, because the images used have different angles and lighting conditions, partial obstructions of trees, moving vehicles or other buildings. Due to all these difficulties, this paper proposes a recognition system which uses Gabor wavelet representation for the extraction of features, the wavelet transform for dimensionality reduction and the multi-layer perceptron neural network for recognition. In order to check the efficiency of this system, we used a database of images of new and old buildings. The performance of the proposed method was satisfactory compared to other methods found in the literature.