A Matching Algorithm Based on Local Topologic Structure
How to cope with non-linear distortions in the matching algorithm is a real challenge. In this paper, we proposed a novel fingerprint matching algorithm based on the local topologic structure and a novel method to compute the similarity between two fingerprints. The algorithm firstly aligns the template fingerprint and the input fingerprint. Then local topologic structure matching was introduced to improve the robustness of global alignment. Finally a novel method was introduced to compute the similarity between the template fingerprint and the input fingerprint. The proposed algorithm has been participated in Fingerprint verification competition (FVC2004). The performance was ranked 3rd position in open category in FVC2004.
KeywordsGlobal Alignment Fingerprint Image False Acceptance Rate Plastic Distortion Fingerprint Match
Unable to display preview. Download preview PDF.
- 1.Cappelli, R., Maio, D., Maltoni, D.: Modelling plastic distortion in finger print images. In: Proc. ICAPR 2001, Rio de Janeiro (March 2001)Google Scholar
- 2.Chen, H., Tian, J., Yan, X.: Fingerprint Matching with Registration Pattern Inspection. In: Oral Report, AVBPA 2003, pp. 327–334. Springer, Heidelberg (2003)Google Scholar
- 3.Biometric Systems Lab, Pattern Recognition and Image Processing Laboratory, Biometric Test Center, http://bias.csr.unibo.it/fvc2004/
- 5.Luo, X., Tian, J.: Knowledge based fingerprint image enhancement. In: Proc. 15th ICPR, Barcelona, Spain (September 2000)Google Scholar
- 6.Luo, X., Tian, J., Wu, Y.: A Minutia Matching algorithm in Fingerprint Verification. In: 15th ICPR, Barcelona, vol. 4, pp. 833–836 (2000)Google Scholar