Fingerprint Preprocessing Algorithm Research

With the development of society, the traditional security systems which based mainly on token or password become fragile and can not meet the demand of modern security system. This situation encourages people research for new security system that is safer and more difficult to fabricate. Fingerprint recognition technique that is safer and harbored feasibility appears under this circumstance. Fingerprint recognition technique would increase the reliability of the recognition by means of our bodies'physical characters.The jobs completed in this dissertation are selective analysis:1. In the step of fingerprint segmentation, the segmentation based on gray level and the segmentation based on orientation are compared, and then the layered segmentation based both gray level and orientation applied to improve these two fingerprint segmentation algorithms.2. This dissertation used uses the Gabor filter to filtering the fingerprint and overcome the disadvantage of other filtering algorithm, get a better effect.3. The binaryzation based on gray level and binaryzation based on orientation are compared in this dissertation, and then the binaryzation combining the orientation and the blocks of fingerprint is applied and get a better effect.4. This dissertation tests a series of means of thinning, the emphasis is algorithm of OPTA and modified OPTA algorithm. And then this dissertation modified the two algorithms to get a better effect of thinning.5. In the end of this dissertation, the theorem and the algorithm of extract the minutia of fingerprint are introduced, and the match algorithm based on the minutia is applied to match the fingerprint effectively.