|
|
Security of Biometric Systems
Lodrová, Dana ; Busch, Christoph (oponent) ; Provazník, Ivo (oponent) ; Drahanský, Martin (vedoucí práce)
The main contributions of this thesis are two novel approaches for the increase of securing of biometric systems based on fingerprint recognition. The first approach is within the liveness detection and prevents the use of various fake fingers and other spoofing techniques during the capturing processes. This patented approach is based on a combination of change of papillary line color and width caused by pressing of a finger against glass plate. The resultant liveness detection unit can be integrated into an optical fingerprint sensor. The second approach is within standardization and it increases the security and interoperability of minutiae extraction and comparison process. For this purposes, I have created the methodology to determine semantic conformance rates of minutiae extractors. The minutiae extracted by the tested extractors are compared against Ground-Truth-Minutiae obtained by clustering of data provided by dactyloscopic/forensic experts. This proposed methodology is included in the ISO/IEC 29109-2 Amd. 2 WD4.
|
|
|
Fingerprint Image Quality Assessment
Šmida, Vladimír ; Busch, Christoph (oponent) ; Drahanský, Martin (vedoucí práce)
The crucial factor of fingerprint recognition is image acquisition. Quality of the captured fingerprint image influences all other components of the recognition system, from image processing, through feature extraction to decision making. Therefore, to retain the security of the overall biometric system, it is necessary to determine the quality and validity of the captured fingerprint image. Although several quality estimation techniques have been proposed, a missing general definition of fingerprint quality makes it difficult to verify them. This thesis deals with the problem of evaluation of the fingerprint signal quality assessment methods. Several factors that influence the quality of fingerprints are described. The state-of-the-art approaches for assessing fingerprint quality are reviewed and implemented. An evaluation method based on the biometric performance is suggested in order to determine the capability to predict the performance of the reviewed methods.
|
|
|
Security of Biometric Systems
Lodrová, Dana ; Busch, Christoph (oponent) ; Provazník, Ivo (oponent) ; Drahanský, Martin (vedoucí práce)
The main contributions of this thesis are two novel approaches for the increase of securing of biometric systems based on fingerprint recognition. The first approach is within the liveness detection and prevents the use of various fake fingers and other spoofing techniques during the capturing processes. This patented approach is based on a combination of change of papillary line color and width caused by pressing of a finger against glass plate. The resultant liveness detection unit can be integrated into an optical fingerprint sensor. The second approach is within standardization and it increases the security and interoperability of minutiae extraction and comparison process. For this purposes, I have created the methodology to determine semantic conformance rates of minutiae extractors. The minutiae extracted by the tested extractors are compared against Ground-Truth-Minutiae obtained by clustering of data provided by dactyloscopic/forensic experts. This proposed methodology is included in the ISO/IEC 29109-2 Amd. 2 WD4.
|
|
|
Fingerprint Image Quality Assessment
Šmida, Vladimír ; Busch, Christoph (oponent) ; Drahanský, Martin (vedoucí práce)
The crucial factor of fingerprint recognition is image acquisition. Quality of the captured fingerprint image influences all other components of the recognition system, from image processing, through feature extraction to decision making. Therefore, to retain the security of the overall biometric system, it is necessary to determine the quality and validity of the captured fingerprint image. Although several quality estimation techniques have been proposed, a missing general definition of fingerprint quality makes it difficult to verify them. This thesis deals with the problem of evaluation of the fingerprint signal quality assessment methods. Several factors that influence the quality of fingerprints are described. The state-of-the-art approaches for assessing fingerprint quality are reviewed and implemented. An evaluation method based on the biometric performance is suggested in order to determine the capability to predict the performance of the reviewed methods.
|
host ::
RSS.