Two-Stage Enhancement Scheme for Low-Quality Fingerprint Images by Learning From the Images

摘要: Fingerprint authentication for content protection in the human-machine systems, cybernetics, and computational intelligence is very popular. Because of complex input contexts, low-quality fingerprint images always exist with cracks scars, dry skin, or poor ridges valley contrast ridges. Usually, are enhanced by one stage either spatial frequency domain. However, performances not satisfactory because complicated ridge structures that affected unusual contexts. In this paper, we propose a novel effective two-stage enhancement scheme both domain learning from underlying images. To remedy areas enhance local ridges, first image ridge-compensation filter With help step, second-stage filter, i.e., bandpass separable radial- angular-frequency domains, employed. It noted parameters filters learnt original first-stage instead acquiring solely. enhances significantly fast sharp attenuation radial domains. Experimental results show our proposed algorithm able to handle various contexts achieves better compared some state-of-the-art algorithms over public databases, improve fingerprint-authentication systems.