Two Different Regimes of Fingerprint Identification – a Comparison

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 (SVM) and a Multi-Layered Perceptron (MLP) network are described and used. Automatic Fingerprint Identification Systems (AFIS) are widely used today, and it is therefore necessary to find a classification system that is less time-consuming. The fingerprint patterns generated are based on minutiae extraction from a thinned fingerprint image. The given fingerprint database is decomposed into four different subclasses. Two different classification regimes are used to train the systems to do correct classification. The classification rate has been estimated to about 87.0 % and 88.8% of unseen fingerprints for SVM and MLP classification respectively. The classification rate of both systems is only differing marginally.A benchmark test has been done for both systems. The matching time is estimated to decrease with a factor of about 3.7 compared to a brute force approach.