Fingerprint Recognition

Abstract

This paper deals with the development of computational model for fingerprint recognition. A person can be recognized and identified by comparing the characteristics of his fingerprint with those of known individuals. Features of the fingerprint under test are compared with a set of enrolled fingerprints already stored in database. If the features match with the database then fingerprint can be recognized and identified.The most popular technique is Minutia Extraction for Fingerprint. The development of computation model for fingerprint recognition by using slantlet transform instead of FFT due to the ability of the slantlet transform to decompose signal (images at various resolutions allows accurate extraction of features from non stationary signal images). The low frequency coefficients which contain the maximum information about the signal, were selected from slantlet decomposition, these coefficient fed to another step in Minutia Extraction and then to Minutia Matching. This new method of fingerprint recognition tasks with lower number of calculations. Hence it raises the algorithmic efficiency and reduces the execution time. The proposed algorithm is tested upon a database consisted of 10 persons. An overall accuracy of recognition rate of the proposed approaches 100.The performance of this work has been evaluated by using MATLAB (2008 A)software.

Keywords

AFAS, AFIA, AFRS, SLT, FVC2002