Effective fingerprint classification by localized models of support vector machines

Finger-print classification is useful as a preliminary step of the matching process and is performed in order to reduce searching time, Various classifiers like support vector machines (SVMs) have been used to Fingerprint classification. Since the SVM which achieves high accuracy in pattern classification is a binary classifier, we propose a classifier-fusion method, multiple decision templates (MuDTs). The proposed method extracts several clusters of different characteristics from each class of fingerprints and constructs localized classification models in order to overcome restrictions to ambiguous Fingerprints. Experimental results show the feasibility and validity of the proposed method.