On the design of 2D Gabor filtering of fingerprint images

The directional bandpass Gabor filter bank approach is one of the most effective and mathematically elegant techniques to date for fingerprint image enhancement. The filter output, however, is very sensitive to ridge orientation and the frequency that the filter is tuned to, and also to the spatial parameters of the Gaussian envelope. Unfortunately, filtering of a fingerprint image with an adaptive 2D Gabor filter bank is computationally expensive because ridge orientation and frequency vary significantly throughout the fingerprint. We propose to use an array of 8/spl times/4 two-dimensional Gabor filters tuned to eight directions and four ridge frequencies. Filtered fingerprint images at any combination of local ridge orientation and frequency can be calculated using a 2D interpolation algorithm The proposed technique produces a better image quality than current Gabor-based techniques. The results are compared using a goodness index measure of the reliability of the automatic minutiae detection.