Performance Improvement for Fingerprint Recognition System using Shape and Orientation Descriptors

Main Article Content

A. E. Amin
A. F. Elgamal Elgamal

Abstract

One of the oldest methods to human identification is fingerprints, there are many and varied systems are used to recognition and verification. The fingerprint recognition systems are similar in all implementation steps which represented by pre and post processing of fingerprint images and extract features. In the preprocessing stage, the image has been enhanced by using Histogram equalization and Fourier transform. The gray fingerprint image is converted to 0-value for ridge and 1-value for bifurcation that is called image binarization. The binarized fingerprint image is used to separate the region of actual fingerprint from the background to produce Region of Interest (ROI). Minutiae extraction stage is consist of two steps namely Fingerprint Ridge Thinning and Minutia Marking. The post-processing stage eliminates spurious feature points based on the structural and spatial relationships of the minutiae, and then validates the minutiae points. Speed and high accuracy in the performance are the distinguishing parameters of fingerprint recognition systems. So, the continuing developments of fingerprints matching algorithms have the greatest impact on the development of biometric identification systems. In spite of the different techniques in the matching methods but were similar in the two basic steps are the alignment andscoring. In this paper, fingerprint matching algorithm based on shape and orientation descriptors is used. The proposed algorithm is designed to address the spurious and missing minutiae and the inability to detect minutiae to guide pre-alignment.

 

Keywords: Fingerprint identification, Minutiae extraction, Fingerprint image processing, Scoring matching

Downloads

Download data is not yet available.

Article Details

Section
Articles