Efficient fingerprint search based on database clustering
Pattern Recognition
Efficient Fingercode Classification
IEICE - Transactions on Information and Systems
Reference Point Detection for Arch Type Fingerprints
ICB '09 Proceedings of the Third International Conference on Advances in Biometrics
Applications of the discrete hodge helmholtz decomposition to image and video processing
PReMI'05 Proceedings of the First international conference on Pattern Recognition and Machine Intelligence
An efficient algorithm for fingercode-based biometric identification
OTM'06 Proceedings of the 2006 international conference on On the Move to Meaningful Internet Systems: AWeSOMe, CAMS, COMINF, IS, KSinBIT, MIOS-CIAO, MONET - Volume Part I
Hi-index | 0.00 |
A fingerprint recognition algorithm needs to recover the pose transformation between the input fingerprint and the template. One solution is to determine a unique reference point for fingerprint alignment. This work develops a new algorithm to detect a unique reference point consistently for all types of fingerprints. Our detection algorithm works on the orientation field smoothed with an adaptively varying neighborhood. The adaptive window is used to attenuate noise of orientation field effectively while maintaining the detailed orientation information in the high curvature area. A new approach of reference point localization is proposed that is based on hierarchical analysis of the orientation coherence. Experiments demonstrate that our developed algorithm consistently locates a unique reference point with high accuracy for all types of fingerprints.