Fingerprint recognition system for low quality images

Fingerprints are widely used in biometric techniques for automatic personal identification. In this paper, a fingerprint recognition system is developed to identify a person according to fingerprint image on Myanmar National Registration Cards (NRCs). It can be acquired only the low quality fingerprint image from NRC card. Generally, fingerprint identification approaches are minutiae-based and correlation-based. Although the minutiae-based method is popular and extensively used method for fingerprint identification, it shows poor performance for low quality images. In proposed system, the correlation-based approach for fingerprint recognition is developed by calculating the mean value of correlation factors. The mean value of correlation factors of input image and template image are computed and compared. If it is over a certain threshold the result of the matching process is positive otherwise negative. Before calculating correlation factor, fingerprint enhancement algorithm using gabor filters are applied for good ridge line feature extraction. Therefore, the proposed system can identify not only on the good quality fingerprint but also on the bad or poor quality fingerprint. The effectiveness of the proposed system can be confirmed through the experimental results.