Detalles del recurso
Pertenece a:
CiteSeerX Scientific Literature Digital Library and Search Engine
Descripción: Recognition systems based on biometrics (faces, hand shapes and fingerprints etc.) are finally taking off although it has taken a long way to come. Fingerprints have been a precious tool for law enforcement, forensics and recently in commercial use for over a century. Evaluate the performance of these emerging technologies is tricky problem. Most fingerprint verification algorithms rely on minutiae features, and these algorithms can only be as robust as the underlying minutiae features. Therefore, reliable minutiae extraction is vital to a system’s performance. Most of the feature extraction techniques extract features from thinned images but while dealing with binarization and skeletization of image it introduces noise or superfluous information, which creates troubles for genuine feature extraction. In this paper we have used the mathematical morphology to remove the superfluous information for genuine feature extraction and measure the feature extraction performance through sensitivity and specificity.
Autor(es): Vikas Humbe - S. S. Gornale - Ramesh Manza - K. V. Kale -
Id.: 45682839
Idioma:
inglés
-
Versión: 1.0
Estado: Final
Tipo: application/pdf -
Palabras clave: Feature Extraction -
Tipo de recurso:
Texto Narrativo
-
Tipo de Interactividad: Expositivo
Nivel de Interactividad: muy bajo
Audiencia:
Estudiante
- Profesor
- Autor
-
Estructura: Atomic
Coste: no
Copyright: sí
: Metadata may be used without restrictions as long as the oai identifier remains attached to it.
Formatos: application/pdf -
Requerimientos técnicos: Browser: Any -
Relación:
[IsBasedOn] http://ijcss.org/Volume1/Issue2/V1_I2_053.pdf
[References] 10.1.1.119.7009
[References] 10.1.1.6.7497
[References] 10.1.1.19.6417
[References] 10.1.1.104.690
[References] 10.1.1.19.6687
Fecha de contribución: 28-jul-2011
Contacto:
Localización:
