A Robust Finger Knuckle Print Authentication using Topothesy and fractal dimension

Madasu Hanmandlu; Neha Mittal; Ritu Vijay

doi:10.14429/dsj.1.9003

Madasu Hanmandlu Electrical Engineering Department, Indian Institute of Technology Delhi, New Delhi
Neha Mittal Electronics Engineering Department, JP Institute of Engineering and Technology, Meerut
Ritu Vijay Electronics Department, Banasthali University, Banasthali

DOI: https://doi.org/10.14429/dsj.1.9003

Keywords: Finger-knuckle-print, Gaussian smoothed high pass, Gaussian smoothed oriented directives, Structure function, Fractal parameters, Score level fusion, Refined scores method

Abstract

This paper presents the finger knuckle based biometric authentication system using the approaches like structure entropy, GSHP (Gaussian smoothed High pass), GSOD (Gaussian Smoothed Oriented Directives) and also the well known method for surface roughness measurement called the fractal profiles represented by Topothesy and fractal dimension which describe not only the roughness but also the affine self similarity. We have also implemented Daisy descriptor for the representation of texture. The results of fractal parameters along with the refined scores are comparable to those of the compcode and impcompcode.

Author Biographies

Madasu Hanmandlu, Electrical Engineering Department, Indian Institute of Technology Delhi, New Delhi

Professor

Electrical Engineering Department

Neha Mittal, Electronics Engineering Department, JP Institute of Engineering and Technology, Meerut

Assistant Professor

Electronic and communications department

Ritu Vijay, Electronics Department, Banasthali University, Banasthali

Associate Professor

Department of electronics