EFFICIENT FINGERPRINT IMAGE ENHANCEMENT ALGORITHM BASED ON GABOR FILTER

Shaikh Mohammedsayeemuddin, Sima K Gonsai, Dharmesh Vandra

Abstract: Fingerprint recognition system is most commonly used biometrics for person identification. The performance of feature extraction algorithm heavily depends on the quality of the fingerprint image. An improved enhancement method based on Gabor filter is presented in this paper. An overlapping block-wise filtering is done in place of conventional pixel wise Gabor filtering. An average frequency of entire image is used instead of each pixel frequency. This reduces the filter complexity and blocking artifacts in an enhanced image. Algorithm is tested on FVC2004 DB1 public fingerprint database. The results show that not only the computation time but accuracy is also improved with our method.

Keywords: gabor filter; fingerprint; image enhancement

DOI: https://doi.org/10.15623/ijret.2014.0304142