Fingerprint Identification Algorithms Based On PCNN And Genetic Algorithms

Fingerprint is widely used as an important means of identity recognition based on its exclusive and stable characteristic. Automatic Fingerprint Identification System (AFIS) is a way of identity identification based on digital image processing, database technology and so on. Although there are many mature AFIS product but most of which are not public because of commercial reasons. The research of fingerprint identification is still a popular subject.This paper is based on the achievements o four predecessors, and get some achievements as followed:(1) Have realized a complete processing of fingerprint minutia extracting, including calculating orientation, segmentation, and reduce noising, thinning, searching minutia and erasing fault minutia.(2) Put forward a PCNN based on orientation algorithm for fingerprint segmentation. PCNN neural network has property of pulse transmit, so we use the Orientation of fingerprint to enable pulse transmit along the ridge of fingerprint, which can make good result of fingerprint segmentation.(3) in fingerprint minutia matching process , genetic algorithm was used. This article first proposed one kind of probability algorithm to determine two fingerprints the direction angles of rotation θ, so only displace dx and dy to the fingerprint code, and the genetic algorithms was used to determine two fingerprints match. In the search process, opposite in the tradition algorithms, which coded dx, dy and 9 mixed, this algorithm reduced search space greatly, so the searching speed was improved greatly.(4) Consider of the wide application of automatic fingerprint recognition system, a simple but complete system of attendance recoding was developed, which can be used by some medium and small sized enterprises.