Fingerprint Partial Matching using a Minutiae-based Combined Feature

Authors
Monireh Abdoos
Nasser Mozayani
Abstract
Fingerprint is a biometric characteristic that is widely used for authentication. Many studies have been done on fingerprint recognition. Most methods use the total image of the fingerprint for matching. Fingerprint matching using a part of the image is an important challenge especially when a partial image doesn’t include singular points of fingerprint such as core and delta. In this work, we present a partial matching method that is independent of the singular points. The proposed method is based on a new feature named “Combined Feature”. The combined feature is defined for every two minutiae by using the skeletonized fingerprint. The feature is invariant with respect to the global transformations such as rotation and transition. Experimental results on FVC2004 show efficiency and accuracy of the proposed method.

Keywords