A new classification algorithm with GLCCM for the altered fingerprints

Abstract

Fingerprints have always been the most practical and positive means of Identification. Offenders, being well aware of this, have been coming up with ways to escape identification by that means. Erasing left over fingerprints, using gloves, fingerprint forgery; are certain examples of methods tried by them, over the years. Failing to prevent themselves, they moved to an extent of mutilating their finger skin pattern, to remain unidentified. This article is based upon obliteration of finger ridge patterns .In this article, we propose a new classification algorithm GLCCM (Gray Level Correlation Coefficient Co-occurrence Matrix) algorithm for altered fingerprints classification. It is based on the fact that altered fingerprint image is composed of regular texture regions that can be successfully represents by co-occurrence matrix. So, we first extract the features based on certain characteristics and then we use these features for classifying altered fingerprints. © 2011 Springer-Verlag.