Abstract
The fingerprint verification has come to attract increased attention as a possible solution to e-security applications. This work describes a fast fingerprint matching algorithm based on a multiple reference minutiae. The fingerprint codes for the matching algorithm are extracted from neighbourhoods’ minutiae which replace the classical core point. The main advantage of our approach is that it avoids the implicit alignment fingerprint images during the matching process and improves the fingerprint verification accuracy through the minutiae binary codes and orientation features combination. Testing our approach on the public fingerprint databases DB1-a and DB2-a FVC2002.