Extração de características de imagens de impressão digital

• Ano de defesa: 2006
• Autor(a) principal: Medeiros, Luciano Xavier
• Orientador(a): Flores, Edna Lúcia
• Banca de defesa: Veiga, Antônio Cláudio Paschoarelli , Carrijo, Gilberto Arantes , Rezende, Marcos Ferreira de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia
• Programa de Pós-Graduação: Programa de Pós-graduação em Engenharia Elétrica
• Departamento: Engenharias
• País: BR
• Palavras-chave em Português: Minúcias; Orientação de campo; Saliências; Afinamento e impressão digital; Processamento de imagens-Técnicas digitais
• Palavras-chave em Inglês: Minutiae; Filed orientation; Ridges; Thinning and fingerprint
• Área do conhecimento CNPq: CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/14544

The fingerprints are lines present on fingers of each human and they are unique, in other words, there are not two people with the same distribution of lines on their fingers. For this reason, it has been used for identification of people for many purposes, such as to check the entrancy and exit of workers in a company, the identify criminals and to restrict the access of people to restrict areas of extreme security. The proposal of this work is to improve the characteristic extraction algorithm of fingerprint developed by Jain and others. The goal of this work is to accelerate the processing time in orientation field estimation, in order to get a better quality of images, resulting of binarization process and to decrease the number of spurious minutiae on the images. The time processing of orientation field estimation improved in these work uses the commutative propriety is smaller than the same estimation wich does not use this propriety in 90%. The method of ridges extration developed in this work uses the DDA algorithm, and results in better quality images. Due to this improviment got in binarization process and the ridges break removing, the detection minutiae algorithm implemented in this work find a small quantity of spurious minutae present in the fingerprint images.