Enhancement of low quality fingerprints based on anisotropic filtering

  • Authors:

  • Xinjian Chen;Jie Tian;Yangyang Zhang;Xin Yang

  • Affiliations:

  • Center for Biometrics and Security Research, Key Laboratory of Complex Systems and Intelligence Science, Institute of Automation, Chinese Academy of Sciences, Graduate School of the Chinese Academ ...;Center for Biometrics and Security Research, Key Laboratory of Complex Systems and Intelligence Science, Institute of Automation, Chinese Academy of Sciences, Graduate School of the Chinese Academ ...;Center for Biometrics and Security Research, Key Laboratory of Complex Systems and Intelligence Science, Institute of Automation, Chinese Academy of Sciences, Graduate School of the Chinese Academ ...;Center for Biometrics and Security Research, Key Laboratory of Complex Systems and Intelligence Science, Institute of Automation, Chinese Academy of Sciences, Graduate School of the Chinese Academ ...

  • Venue:
  • ICB'06 Proceedings of the 2006 international conference on Advances in Biometrics
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

The enhancement of the low quality fingerprint is a difficult and challenge task. This paper proposes an efficient algorithm based on anisotropic filtering to enhance the low quality fingerprint. In our algorithm, an orientation filed estimation with feedback method was proposed to compute the accurate fingerprint orientation. The gradient-based approach was firstly used to compute the coarse orientation. Then the reliability of orientation was computed from the gradient image. If the reliability of the estimated orientation is less than pre-specified threshold, the orientation will be corrected by the mixed orientation model. And an anisotropic filtering was used to enhance the fingerprint, with the advantages of its efficient ridge enhancement and its robustness against noise in the fingerprint image. The proposed algorithm has been evaluated on the databases of Fingerprint verification competition (FVC2004). Experimental results confirm that the proposed algorithm is effective and robust for the enhancement of the low quality fingerprint.