Research On Fingerprint Recognition Based On Spiking Neural P Systems And Its Application

On account of its safety,reliability and accuracy in identification recognition,biometric recognition technology is considered the most convenient and secure recognition technology currently.Meanwhile,as a result of its stability,consistency,uniqueness and dependability,fingerprint has nearly become the synonym of biometric identification.However,in spite of the relatively high accuracy of fingerprint recognition based on categorization and analysis scheme,it is still not able to achieve a completely precise result.Furthermore,the categorization of fingerprints is neither sufficiently precise nor systematicHuman brain has accumulated abundant computational intelligence throughout the millions of years of evolution,which offers ample ideas towards the development of cerebral heuristic computation model.Spiking neural P systems is a computational model inspired from cerebral neural system which possesses supreme characteristics such as high level of computational parallelism,readability and extendibility,and high fault tolerance.In this article,spiking neural P systems has been applied in fingerprint recognition system for the first time and it will mainly demonstrate1.A new spiking neural P systems will be raised in this article.It is double layer spiking neural P systems with self-organization using anti-spikes that achieves Turing universality2.This article will provide a clearer demonstrative strategy and straight-forward pairing process of spiking neural P system built on fingerprint minutiae comparing to the traditional modeling method3.In this article,a fingerprint pairing method based on the minutiae is raised and it will increase the accuracy of recognition from the extension of minutiae pairing and fingerprint pairingIn this article,fingerprint images will be binary conversed and refined at the preprocessing stage then ridge endings and bifurcation minutiae will be paired.The new spiking neural P systems is proposed,based on which,a new fingerprint recognition algorithm is presented and has been proved with great experimental performance.