Abstract:
Fingerprint recognition is one of the most
well-known and publicized biometrics for
personal identification. Fingerprints exhibit
oriented texture-like patterns. The texture
information of the fingerprint can be used for
fingerprint matching. Gabor filters can optimally
capture global and local texture information
even from poor-quality or incomplete images.
But Gabor filterbank-based approach use only
texture information for fingerprint recognition
and it is not robust to image distortion and
rotation. In this paper, a hybrid fingerprint
matching algorithm is developed for identifying
the low quality fingerprint images by combining
orientation features and the local texture pattern
obtained using a bank of Gabor filters. The
proposed matching approach is compared with
the filterbank-based approach, and the proposed
system produces a much improved matching
performance by combining the orientation
features to the filterbank-based features.
