Title :
Orthogonal curved-line Gabor filter for fast fingerprint enhancement
Author :
Yuan Mei ; Bo Zhao ; Yu Zhou ; Songcan Chen
Author_Institution :
Jiangsu Eng. Center of Network Monitoring, Nanjing Univ. of Inf. Sci. & Technol. (NUIST), Nanjing, China
Abstract :
Fingerprint enhancement, which aims to improve the clarity of fingerprint ridge structures, plays a very important role in automatic fingerprint recognition. Recently, a novel enhancement method called curved Gabor was proposed, which achieves a state-of-the-art enhancement effect, but its high computation becomes a bottleneck. An orthogonal curved-line Gabor filter is proposed to overcome this issue. This method decomposes the unrotated two-dimensional (2D) Gabor into a 1D bandpass Gabor and a 1D lowpass Gabor, then implements them with two separable steps: a 1D bandpass Gabor and a 1D lowpass Gabor filtering on curved lines, respectively, perpendicular and parallel to the ridge orientation. The experiments show that the method is about 42 times faster than the curved Gabor while retaining comparable enhancement effect.
Keywords :
Gabor filters; band-pass filters; fingerprint identification; image enhancement; low-pass filters; 1D bandpass Gabor; 1D lowpass Gabor filtering; automatic fingerprint recognition; fingerprint enhancement; fingerprint ridge structures; orthogonal curved-line Gabor filter; ridge orientation; unrotated two-dimensional Gabor;
Journal_Title :
Electronics Letters
DOI :
10.1049/el.2013.2619