RT Conference Proceedings
T1 A first step to accelerating fingerprint matching based on deformable minutiae clustering
A1 Romero, Luis F.
A1 Tabik, Siham
A1 Sánchez, Andrés Jesús
A1 Medina Pérez, Miguel Angel
A1 Herrera, Francisco
K1 Identificación biométrica - Congresos
K1 Dactiloscopia - Congresos
AB Fingerprint recognition is one of the most used biometricmethods for authentication. The identification of a query fingerprint requiresmatching its minutiae against every minutiae of all the fingerprintsof the database. The state-of-the-art matching algorithms are costly, froma computational point of view, and inefficient on large datasets. In thiswork, we include faster methods to accelerating DMC (the most accuratefingerprint matching algorithm based only on minutiae). In particular,we translate into C++ the functions of the algorithm which represent themost costly tasks of the code; we create a library with the new code andwe link the library to the original C# code using a CLR Class Libraryproject by means of a C++/CLI Wrapper. Our solution re-implementscritical functions, e.g., the bit population count including a fast C++PopCount library and the use of the squared Euclidean distance for calculatingthe minutiae neighborhood. The experimental results show asignificant reduction of the execution time in the optimized functions ofthe matching algorithm. Finally, a novel approach to improve the matchingalgorithm, considering cache memory blocking and parallel data processing,is presented as future work.
YR 2018
FD 2018-07-12
LK https://hdl.handle.net/10630/16189
UL https://hdl.handle.net/10630/16189
LA eng
NO Universidad de Málaga. Campus de Excelencia Internacional Andalucía Tech.
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 21 ene 2026