The effect of orientation map accuracy on fingerprint classification using PCA-MSOM

Fingerprint classification forms an important problem to be solved in identification system and there are various approaches for the solution. In any pattern classification system the most important part is the feature extraction process. For fingerprints orientation map forms the most common and promising feature to be used for classification. In this paper two different orientation map extraction algorithm are compared to find the one that gives the most reliable features for the PCA-MSOM classifier.