Security Issues in Biometric Systems

EasyChair Preprint 9150

Abstract

Biometrics is a recent discussion in computer science that is defined as the study

of creating computer models. It’s significant in forensics, secure access, and

government and commercial applications (such as international border crossing,

making government identity proofs, etc). It is a branch of computer science that

investigates a person’s correct identification. It’s a pattern recognition system

that uses machine learning algorithms to process data from fingerprinting, iris

scanning, retina scanning, hand geometry, face recognition, voice recognition,

and odor biometrics, among other methods. Uni modal and

multi-modal biometrics are the two methodologies used in biometrics. Noise,

spoofing, lower accuracy, and other issues might occur when uni-modal is used.

Another option is multi-modal biometrics, which solves the problem of uni-modal

biometrics by providing anti-spoofing techniques that make it harder for an at

tacker to infiltrate the security system. They have a look at’s predominant

aim is to realize the position of deep learning in the authentication system in

addition to its use within the enhancement of biometric device protection. We

describe those methods and look at the constraints that hold to restrict biometrically

technology’s full ability. The most critical are: developing robust authentication

methods, assuring the security of enrolled templates, and protecting structures

from various assaults.

Keyphrases: Authentication, Biometric, identification

@booklet{EasyChair:9150,
  author    = {Anupriya Jain and Poonam Mishra and Mohit Sachdeva},
  title     = {Security Issues in Biometric Systems},
  howpublished = {EasyChair Preprint 9150},
  year      = {EasyChair, 2022}}