In this paper we propose a semi-automatic approach for the enhancement of very low quality fingerprints such as latent fingerprints. A specific markup tool is designed to allow fingerprint examiners to simply and quickly provide sparse estimates of local orientations and frequencies. These estimates are then interpolated though Delaunay triangulation and fed to a contextual Gabor-based enhancement algorithm that significantly improves the image quality, thus making the successive automatic feature extraction much more reliable. Experimental results (both qualitative and quantitative) confirm the effectiveness of this method over a fully-automatic state-of-the-art approach.
R. Cappelli, D. Maio , D. Maltoni (2009). Semi-automatic Enhancement of Very Low Quality Fingerprints. SALZBURG : IEEE.
Semi-automatic Enhancement of Very Low Quality Fingerprints
CAPPELLI, RAFFAELE;MAIO, DARIO;MALTONI, DAVIDE
2009
Abstract
In this paper we propose a semi-automatic approach for the enhancement of very low quality fingerprints such as latent fingerprints. A specific markup tool is designed to allow fingerprint examiners to simply and quickly provide sparse estimates of local orientations and frequencies. These estimates are then interpolated though Delaunay triangulation and fed to a contextual Gabor-based enhancement algorithm that significantly improves the image quality, thus making the successive automatic feature extraction much more reliable. Experimental results (both qualitative and quantitative) confirm the effectiveness of this method over a fully-automatic state-of-the-art approach.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.