A study on fingerprint repair algorithm

Fingerprint repair is the post-processing stage of fingerprint recognition process, in which false characteristics left in the thinned pictures will be eliminated. A new fingerprint repair algorithm is proposed in this paper. In order to clear off false characteristics and repair the thinned image, a technology called ridge tracking is used to find out all the ridges on the image and then deal with them. Not only the lengths of the ridges but also their initial points and terminations are adopted to clear off false characteristics such as eyelet, short ridge, bridge, burr, ridge break and so on. The algorithm is realized through C++ programme design, mainly using the standard template library of it. The simulation results at the end have shown almost all false characteristics in the experiments are eliminated, which indicates the proposed algorithm is very effective.