A new method for fingerprint matching using phase-only auto- and cross-bispectrum

• Published in: Signal, image and video processing Vol. 10; no. 7; pp. 1327 - 1333
• Main Authors: Ismaili Alaoui, E. M., Ibn-Elhaj, E.
• Format: Journal Article
• Language: English
• Published: London Springer London 01.10.2016; Springer Nature B.V
• Subjects: Computer Imaging; Computer Science; Image Processing and Computer Vision; Multimedia Information Systems; Original Paper; Pattern Recognition and Graphics; Phase matching; Signal,Image and Speech Processing; Vision; Biometric; Shift fingerprint; Amplitude spectrum; Fingerprint recognition; Auto-bispectrum; Cross-bispectrum; Rotation fingerprint; Fingerprint matching
• ISSN: 1863-1703; 1863-1711
• DOI: 10.1007/s11760-016-0930-5

Fingerprints are widely used in biometric techniques for automatic personal identification. In this work, we present a new correlation-based approach which uses the phase-only auto- and cross-bispectrum. This paper deals with the problem of aligning two fingerprints images under translation and rotation. The method described utilizes the shift invariance property of the auto-bispectrum to eliminate the effect of the translation component. Only the phase information is preserved from the auto-bispectrum in order to achieve better resilience against non-uniform illumination changes. The input fingerprint image that gives the highest correlation peak is selected as the rotation-normalized image. The theoretical basis for fingerprint matching by auto- and cross-bispectrum analysis is presented. To prove the feasibility of the proposed method, we compared it to a the method proposed by Ito et al. (IEICE Trans Fundam E87-A(3):682–691, 2004 ) implemented in a similar manner as our approach. The experiments executed on FVC 2004 have shown that the suggested algorithm is successful on the bad or poor quality fingerprint images. The algorithm is robust providing accurate results in most of the time.