National Technical University of Athens (NTUA)
Καινοτόμοι μηχανισμοί βελτίωσης απόδοσης και αποτελεσματικότητας των βιομετρικών συστημάτων
Abstract
dc:descriptionThe current market trend for the provision of personalised services as well as the intensifying need for reinforcing security in various sectors over the past years are gradually leading towards the quest for novel, more effective and robust identification mechanisms, methods and systems. Biometric technologies, being based on human characteristics with special innate properties, such as uniqueness, universality and collectiveness, which comprise important requirements for person identification, are a joint multidisciplinary research effort for covering these high needs. The intense commercial and governmental interest for these technologies and their capabilities has led to worldwide feverish research regarding boosting their accuracy while reducing their cost, so that they are able to cover the continuously increasing range of applications of varying requirements aiming at providing more secure and effective identity management. This thesis presents innovative mechanisms which are either inextricable parts of biometric systems or focus on the infrastructural requirements in order to improve the effectiveness and the efficiency of these systems. As far as effectiveness is concerned, we present an innovative modeling of biometric data at score-level based on Student t mixture models and the application of a likelihood-based decision rule. This mechanism is compared against existing methods for score-level fusion with the simulation results proving the effectiveness and the efficacy of the proposed method. It should be noted that this mechanism is independent of the biometric system in which it is incorporated as well as the biometric feature extraction algorithms used. In the context of increasing the accuracy of the biometric systems, a novel biometric systems architecture is presented which is an extension of the existing generic architecture of biometric systems and which focuses on the more effective and intelligent exploitation of the informational wealth of biometric data through its combination with the respective information regarding the specific conditions under which the incorporated biometric feature extraction algorithms perform best. This study expands towards security and privacy aspects of biometric systems during which we analyse the main points of attack within a biometric systems and map them with the major privacy aspects. Finally, based on the fact that reliability, effectiveness and efficiency issues of a biometric system are tightly linked with the availability of the biometric data and the access latency to the latter, we present novel replica creation and maintenance mechanisms within a distributed environment, with their evaluation being based on the required computational power and their execution time while their effectiveness lies within acceptable levels.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2009
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Andronikou, Vassiliki
- Ανδρόνικου, Βασιλική
Subjects
dc:subject × 19- Βιομετρικά χαρακτηριστικά
- Βιομετρικά συστήματα
- Αρχιτεκτονική βιομετρικών συστημάτων
- Διαχείριση αντιγράφων
- Ασφάλεια δεδομένων
- Πληροφοριακή ιδιωτικότητα
- Συνδυασμός βιομετρικών χαρακτηριστικών
- Ποιότητα υπηρεσιών
- Biometric features
- Biometric systems
- Quality of services (QoS)
- Replication management
- Data security
- Informational privacy
- Biometric fusion
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/26780