J. Qi (PRC), S. Wang (USA), Z. Shi, X. Zhao, and Yangsheng Wang (PRC)
Fingerprints, correspondence, matching, minutiae, orientation attribute
A novel fingerprint representation scheme is introduced us ing both the feature of orientation field of the fingerprint pattern around the minutia and the geometrical configura tion of the neighboring minutiae around the minutia. For each minutia, this representation allows the derivation of a fixed-size feature vector which is invariant to the rota tion and translation of the fingerprint. With these vectors, we can easily define the similarity level between minutiae that is used to find the correspondence of two minutiae sets and increase the reliability of the matching. A finger print matching algorithm,based on the well-defined similar ity between minutiae, is developed and tested with a series of experiments performed on the public domain collection of fingerprint images, DB3 in FVC2002. The results reveal that our method outperforms other alternative approaches implemented for comparison.
Important Links:
Go Back
