A method based on continuous spectrum analysis and artificial immune network optimization algorithm for fingerprint image ridge distance estimation

Abstract

It is important for improving the performance of automatic fingerprint identification system to estimate the ridge distance accurately. The traditional Fourier transform spectral analysis method had the worse redundancy degree in estimating the ridge distance because it was based on the two-dimension discrete Fourier spectrum. The statistical window method cannot obtain the accurate ridge distance because of the noises and the warp of the statistical value. The paper introduces the sampling theorem and artificial immune network into the fingerprint image ridge distance estimation method, transforms the discrete spectrum into the continuous spectrum, acquires the local peak points adopting the artificial immune network optimization algorithm and then obtains the ridge distance in the frequency field. The experimental results indicate that the ridge distance is more accurate and has improved the accuracy rate of automatic fingerprint identification system to a certain extent.