Alignment-Free Cross-Sensor Fingerprint Matching based on the Co-Occurrence of Ridge Orientations and Gabor-HoG Descriptor

The existing automatic fingerprint verification methods are designed to work\nunder the assumption that the same sensor is installed for enrollment and\nauthentication (regular matching). There is a remarkable decrease in efficiency\nwhen one type of contact-based sensor is employed for enrolment and another\ntype of contact-based sensor is used for authentication (cross-matching or\nfingerprint sensor interoperability problem,). The ridge orientation patterns\nin a fingerprint are invariant to sensor type. Based on this observation, we\npropose a robust fingerprint descriptor called the co-occurrence of ridge\norientations (Co-Ror), which encodes the spatial distribution of ridge\norientations. Employing this descriptor, we introduce an efficient automatic\nfingerprint verification method for cross-matching problem. Further, to enhance\nthe robustness of the method, we incorporate scale based ridge orientation\ninformation through Gabor-HoG descriptor. The two descriptors are fused with\ncanonical correlation analysis (CCA), and the matching score between two\nfingerprints is calculated using city-block distance. The proposed method is\nalignment-free and can handle the matching process without the need for a\nregistration step. The intensive experiments on two benchmark databases\n(FingerPass and MOLF) show the effectiveness of the method and reveal its\nsignificant enhancement over the state-of-the-art methods such as VeriFinger (a\ncommercial SDK), minutia cylinder-code (MCC), MCC with scale, and the\nthin-plate spline (TPS) model. The proposed research will help security\nagencies, service providers and law-enforcement departments to overcome the\ninteroperability problem of contact sensors of different technology and\ninteraction types.\n