A Simple and Efficient Roadmap to Process Fingerprint Images in Frequency Domain

In the field of computer vision, image enhancement is one of the most important and critical stage, which eventually or indirectly decides the final results in the boolean form, as far as image recognition or comparison is concern, fingerprint recognition system is also the part of computer vision a...

Full description

Saved in:
Bibliographic Details
Published in:International journal of computer applications 2015-01, Vol.112 (4)
Main Authors: Kumar, Avenash, Jilani, Tahseen A
Format: Article
Language:English
Subjects:
Approximation
Computer vision
Fingerprints
Frequency domains
Image enhancement
Matlab
Object recognition
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In the field of computer vision, image enhancement is one of the most important and critical stage, which eventually or indirectly decides the final results in the boolean form, as far as image recognition or comparison is concern, fingerprint recognition system is also the part of computer vision and considered as one of the most matured and accepted biometric system, which implies the matching of fingerprint impression with template data. The normal issue which arise in the making such systems is the noise in input fingerprint image which actually depends upon the devices i. e. used to capture fingerprint image. In this paper we demonstrate the techniques for fingerprint image enhancement in frequency domain, after getting back in spatial (time) domain, we exact the ROI from the output image of frequency domain using least square approximation method and finally we extract minutiae from fingerprint image using cross number (CN) [5] and compare with template data in post-processing stage. The demonstration has been made under the MATLAB's background and the experiments conducted on FVC 2002 fingerprint dataset of University of Bologna [1]
ISSN:0975-8887
0975-8887
DOI:10.5120/19655-1292