作者: Farid Benhammadi , Hamid Hentous , Kadda Bey-Beghdad , Mohamed Aissani
DOI: 10.1007/11492542_65
关键词: Coordinate system 、 Minutiae 、 Image processing 、 Matching (statistics) 、 Artificial intelligence 、 Fingerprint 、 Computer science 、 Projection (set theory) 、 Computer vision 、 Fingerprint database 、 Pattern recognition 、 Fingerprint recognition 、 Blossom algorithm 、 Biometrics 、 Dynamic time warping
摘要: In this paper, we propose an original fingerprint matching algorithm using a set of intrinsic coordinate systems which each one is attached to minutia according its orientation estimated from image. Exploiting these systems, minutiae locations can be redefined by means projection coordinates on the relative reference minutia. Thus, our use calculate score between two fingerprints. To avoid directional field estimation errors, variation in order manage errors location minutiae. With technique, approach doesn't embedding alignment into stage design robust algorithms for recognition. The was tested database DB2 used FVC2000 and results are promising.