Fingerprint identification – a support vector machine approach

Abstract

In this work a hybrid technique for classification of fingerprint identification has been developed to decrease the matching time. For classification a Support Vector Machine is described and used. Automatic Fingerprint Identification Systems are widely used today, and it is therefore necessary to find a classification system that is less time-consuming. The given fingerprint database is decomposed into four different subclasses and a SVM algorithm is used to train the system to do correct classification. The classification rate has been estimated to about 87.0 % of unseen fingerprints. The average matching time is decreased with a factor of about 3.5 compared to brute force search applied.