Skip to Main Content
A fingerprint identification technique is presented, which mainly consists of three modules: enrolment module, identification module and feedback module. In the identification module, a clustering algorithm is used to detect similar minutiae groups from multiple template images generated from the same finger and create the cluster core set. An algorithm compares the similarity level between the minutiae of the test fingerprint and the cluster core set and returns a likely list of candidates. In feedback module, we propose a path to learn and train the cluster core vector based on the assessment of cluster solution. The experiment results demonstrate that this similarity-searching approach proves suitable for one-too-many matching of fingerprints on large-scale databases. With the feedback module the proposed fingerprint identification scheme has inspiring identification performance.