Welcome to BioLab 23/11/2024 10:50
 
 


Presentation


Brief presentation of the main laboratory activities: PDF version - PowerPoint version

Documento di presentazione del laboratorio (in Italian): PDF version - Word version

Main achievements

2019 BioLab participation in the SOTAMD european project.
2017 Biolab introduced one of the first technique (Face Demorphing) to detect face morphing attacks.
2016 BioLab demonstrated the feasibility of enrolling double-identity fingerprint biometrics in electronic documents.
2015 A novel fingerprint identification algorithm running on GPUs has been designed. The new algorithm allows a throughput of more than 40 million comparisons per second on a single PC. Read more...
2014 BioLab demonstrated the feasibility of enrolling double-identity face biometrics in electronic documents.
2014 A two-factor biometric template protection technique based on MCC was developed: Two-Factor Protected Minutia Cylinder-Code (2P-MCC). Read more...
2013 BioLab participation in the European Commission VII Framework Programme within INGRESS (IP).
2012 A novel biometric template protection technique based on MCC was developed: Protected Minutia Cylinder-Code (P-MCC). Read more...
2012 BioLab participation in the European Commission VII Framework Programme within FIDELITY (IP).
2010 FVC-onGoing, an innovative web-based automated evaluation system for biometric recognition algorithms was developed and made available to the biometric community. FVC-onGoing is the latest evolution of FVC: tests are carried out on a set of sequestered datasets and results are reported on-line by using well known performance indicators and metrics. Read more...
2009 The second edition of the Handbook of Fingerprint Recognition was published. Read more...
2009 A new representation and matching technique for fingerprint recognition was developed and patented: Minutia Cylinder-Code (MCC). The new method is extremely fast and very accurate: it can be implemented even on very light architectures and can perform 200K comparisons per second on a low-cost PC. Read more...
2008 BioLab introduced a new operational definition of fingerprint scanner quality and carried out a large experimentation to understand the effects of the various quality parameters on fingerprint recognition accuracy. Read more...
2007 BioLab published the first effective approach to reconstruct fingerprint images from standard minutiae-based templates and showed how much the reconstructed images are similar to the original ones. Read more...
2005 Two novel approaches for fake finger detection (patent pending) were developed, based on odor and skin distortion. Read more...
2004 BioLab participation in the European Commission VI Framework Programme within BioSec (IP) and Biosecure (NoE).
2003 The first monographic book on automated approaches to fingerprint recognition was coauthored by two BioLab members (published by Springer). The book received the prestigious 2003 PSP award for the "Computer Science" category given by the Association of American Publishers. Read more...
2002 BioLab published the first fingerprint classification algorithm able to meet the FBI fingerprint classification requirements. Read more...
2000 BioLab co-organized the first international competition for fingerprint verification algorithms (FVC2000). The great success of the event prompted the organizers to set up similar competitions in the years 2002, 2004 and 2006. FVC databases are nowadays the most widely adopted benchmark for fingerprint recognition. Read more...
2000 BioLab developed SFinGe: the first approach able to generate realistic synthetic fingerprint images. Read more...
1997 A novel fingerprint indexing strategy was introduced: Continuous Classification. The new approach overcomes the typical problems of exclusive classification techniques: small number of classes, non-uniform distribution among classes, ambiguous fingerprints. Read more...
1997 The first direct gray-scale minutiae detection approach was proposed: the new approach is able to extract minutiae from fingerprint patterns without the need of binarizing and thinning the ridge lines. Read more...

Copyright © 2024 Biometric System Laboratory