Title :
Mixing fingerprints for template security and privacy
Author :
Ross, Arun ; Othman, Asem
Author_Institution :
Lane Dept. of Comput. Sci. & Electr. Eng., West Virginia Univ., Morgantown, VA, USA
fDate :
Aug. 29 2011-Sept. 2 2011
Abstract :
Securing a stored fingerprint image is of paramount importance because a compromised fingerprint cannot be easily revoked. In this work, an input fingerprint image is mixed with another fingerprint (e.g., from a different finger), in order to produce a new mixed image that obscures the identity of the original fingerprint. Mixing fingerprints creates a new entity that looks like a plausible fingerprint and, thus, (a) it can be processed by conventional fingerprint algorithms and (b) an intruder may not be able to determine if a given print is mixed or not. To mix two fingerprints, each fingerprint is decomposed into two components, viz., the continuous and spiral components. After pre-aligning the two components of each fingerprint, the continuous component of one fingerprint is combined with the spiral component of the other fingerprint image in order to generate a mixed fingerprint. Experiments on the WVU and FVC2000 datasets show that the mixed fingerprint can potentially be used for authentication and that the identity of the original fingerprint cannot be easily deduced from the mixed fingerprint. Further, the mixed fingerprint can facilitate in the generation of cancelable templates.
Keywords :
data privacy; fingerprint identification; FVC2000 dataset; WVU dataset; cancelable template generation; component prealignment; continuous components; data authentication; fingerprint decomposition; fingerprint mixing; input fingerprint image; spiral components; stored fingerprint image security; template privacy; template security; Equations; Fingerprint recognition; Indexes; Probes; Security; Spirals;
Conference_Titel :
Signal Processing Conference, 2011 19th European
Conference_Location :
Barcelona