ISSN (Online) : 2456 - 0774

Email : ijasret@gmail.com

ISSN (Online) 2456 - 0774

A HYBRID APPROACH FOR DEEP LEARNING BASED FINGERVEIN BIOMETRICS TEMPLATE SECURITY

Abstract

Abstract: We are living in the today’ssociety, where we have fairly-enough storage capacity and processing power, theonly issue is with security. As, the technologies are evolving  with faster rate, we are tend to grow the useof electronic devices rapidly in todays’ society, it started to flow or leakageof personal information around/across, which then leads to breach of thisinformation. Now, personal or identical verification is key problem is beingcrucial. So whatever traditional methods we have for providing authenticationor security those have proven inadequate to be unreliable and do not providestrong security. Biometric template protection is one of the most importantissues in securing today’s biometric system. We have many algorithms whichdon’t give adequate solution for the same. So we tried to give a method whichwill reach to the expectations more satisfactorily and certainly to the extentrequired. In this paper we have discussed a hybrid method for finger veinbiometric recognition based on deep learning approach using BDD and fuzzycommitment schemes. The proposed hybrid method consists of four parts, namelyFinger vein feature extraction, BDD-based secure template generation, Fuzzycommitment scheme and ML based finger vein recognition and decision making.Thus it has four module and each module works efficiently and gives accurateresults on all databases.

 

Keywords:  Biometric,template security, hybrid, binary decision diagram (BDD), fuzzy commitmentscheme, deep learning and machine learning.


Full Text PDF

IMPORTANT 

Submit paper at ijasret@gmail.com

Paper Submission Open For July 2020
UGC indexed  2017-2019
Last date for paper submission 10th July, 2020
Deadline Submit Paper any time
Publication of Paper Within 01-02 Days after completing all the formalities
Paper Submission Open For online Conference 
Publication Fees(6 Authors) Rs.1000    (Up to 06 Authors)