Classificação de impressões digitais

• Autor(a) principal: Mizobuchi Júnior, Alvaro Toshiyuki [UNESP]
• Data de Publicação: 2011
• Tipo de documento: Trabalho de conclusão de curso
• Idioma: por
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://hdl.handle.net/11449/120007
• topic: Computação; Biometria; Reconhecimento de padrões; Processamento de imagens; Datiloscopia

The biometric characteristics have been used increasingly as a way to identify an individual, mainly for security reasons. Among them, the fingerprint is the most used biometric characteristic around the world, because it is relatively simple and very efficient. In this scene, there was a significant increase in the size of databases containing information on fingerprints, necessary to perform the recognition of a person. The task of classifying them beforehand has become extremely important as it reduces dramatically the size of the problem during a search, because it is not necessary to go through the whole database. Considering its importance, in the last thirty years, many techniques have been developed to try to increase the efficiency of the classification process. This project followed the rules-based approach and the Software Development Kit (SDK) VeriFinger 6.1 was used to assist in the detection of cores and deltas. Additionally, the classification was also implemented by means of directional map and the Poincar´e index. To make the experiments, the number four database from the National Institute of Standards and Technology (NIST) was used, which is a standard in this area