Minutiae location and direction adjustment for nonlinear distorted fingerprint image matching

Nonlinear distorted fingerprint image matching is a challenge task in most computer-based Auto Fingerprint Identification Systems. Especially for minutiae-based matching algorithm, which is most widely adopted, both the location and direction of the minutiae are changed because of distortion. We carefully observe many distorted fingerprint images, find out some useful characteristics of minutiae´ deviation on nonlinear distorted image, and try to use them to improve the matching accuracy. In this paper, we propose a method to adjust the minutiae position and direction on the basis of some landmark minutiae pairs got in pre-processing step. As a result, the minutiae deviation caused by nonlinear distortion can be recovered to a certain extent. Experimental results on FVC2004 DB2 show that our method can improve the matching performance clearly. Also this method can be deployed in most existing minutiae-based Auto Fingerprint Identification Systems easily.