A modified BP network using malsburg learning for rotation and location invariant fingerprint recognition and localization with and without occlusion

Abstract

This present paper designs and develops a modified Malsburg Learning and Back Propagation (BP) Network combination for recognizing and localizing clear as well as occluded fingerprints in single and multiple fingerprint image frames. The present method of fingerprint recognition is completely rotation and location invariant of the different fingerprints in an image frame. The technique of using the combination of Malsburg learning and BP Network to perform learning of the different fingerprint images and subsequent identification and location invariant localization of clear and occluded images is efficient, effective and fast. Also the accuracy, precision, recall and F-score of the classifier are substantially moderate and the recognition time of fingerprints are quite low.