Scribd
Upload
14 views5 pages

Mailar SEMINAR

The document discusses finger vein recognition using deep learning for biometric authentication. It describes how finger vein patterns are unique identifiers that can be used for identification and verification. The objectives are to improve accuracy, robustness to variations, and enable real-time performance of finger vein recognition systems using deep learning techniques.

Uploaded by

abzalbagawan48
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
14 views5 pages

Mailar SEMINAR

The document discusses finger vein recognition using deep learning for biometric authentication. It describes how finger vein patterns are unique identifiers that can be used for identification and verification. The objectives are to improve accuracy, robustness to variations, and enable real-time performance of finger vein recognition systems using deep learning techniques.

Uploaded by

abzalbagawan48
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

VISVESVARAYATECHNOLOGICALUNIVERSITY

JnanaSangama,Belgavi-590018

Technical Seminar On

“FINGER VEIN RECOGNITION USING DEEP


LEARNING”
Submitted By:

Mailarlingesh

Barigidad 2TG21EC403

Under the guidance of


Dr. Madhusudan K

Dept.of Electronics and Communication Engg. TCE,GADAG

DEPARTMENT OFELECTRONICS AND COMMUNICATION

TONTADARYA COLLEGE OF ENGINEERING


GADAG-582101
2023-24
Abstract

Due to their combination of security and economic viability, finger vein biometrics
have gained considerable traction in recent years. They have the advantage of being the least
vulnerable to identity theft because veins are present beneath the skin, as well as being
unaffected by the ageing process of the user. To address the ever-increasing need for security,
all of these variables necessitate working models. Using face recognition and AI-based
biometrics has become a hot subject in law enforcement because of the accidental demographic
bias it introduces into the process. Biometric prejudice, on the other hand, has far-reaching
implications that transcend into everyday situations. When an ATM transaction or an online
banking transaction is compromised by a fake positive or negative verification, it makes it
simpler for fraudsters to carry out their criminal activities.

Introduction

To authenticate a person's identification, a biometric authentication system measures


certain physical traits or behaviors of the person's body in real-time. Iris scanners, for example,
use biometric data to create digital representations of a per-son's identity. Accordingly,
biometric authentication systems may identify or validate the individual by comparing the data
to other biometric records in the database using algorithms. Biometric authentication systems
have two primary modes of operation: identification and verification. The input data is
compared to all known patterns in the database in the identification mode. It is possible to
determine if this individual is in the database using the system. Biometric input data is
compared to a single person's unique pattern when in verification mode. If they're not the same
person, it's a way to stop many individuals from using the same identity

• Biometric identification is a challenging task in information security. Extrinsic biometrics


such as the fingerprint, iris, and face are all vulnerable to faked input, which causes security
issues. But intrinsic modalities are difficult to be faked as it is hidden under the skin, hence
more secure. Finger veins are intrinsic biometric, which are tiny blood vessels within our finger,
and is unique to every individual. When infrared images are passed, the blood vessels with
hemoglobin are detectable as a sequence of dark lines.

• Retinal fingerprints are unique to each individual, yet hackers can still find a method to get
around them. The amount of protection provided by various biometric features also varies.
Finger Vien Recognition

Biometric recognition approaches can be divided into two types: (i) extrinsic
biometric features such as palm print, iris, fingerprint, face and (ii) intrinsic biometric features
such as palm vein, hand vein, and finger vein. Comparatively extrinsic features are more visible
than the intrinsic features and the security levels also vary among different biometric features.
The biometric characteristics are unique for each person, but hackers find ways to crack the
authentication system.

But the finger-vein based biometric authentication system is harder to be fooled compared to
other methods because it only identifies unique finger-vein patterns beneath the skin of the
living person.

Finger-vein data is gathered with the use of specialized capturing equipment. Near infrared
light, a lens, a light filter, and picture capturing technology make up the bulk of this capture
apparatus. Finger veins are invisible to the naked eye because they are hidden beneath the
surface of the skin. Near-infrared light, which may penetrate through human tissue, is used in
this gadget . Near infrared light is also blocked by pigments like hemoglobin and melanin.

Fig 1: Finger vein biometrics

Finger vein biometrics uses a person's individual vein patterns to determine their identity.
Biometrics derived from the blood vessels beneath the skin is also known as vascular
biometrics. Using near-infrared light or visible light causes hemoglobin —the iron-containing
protein all have in our blood — to change color. This means that the reader is able to scan the
vein patterns of the individual user. In the cloud, the vein pattern is kept in an encrypted digital
format.

Objectives

• Enhanced Accuracy: Improve the accuracy of finger vein recognition systems using
deep learning techniques to achieve high levels of precision in identifying individuals based on
their unique vein patterns.
• Robustness to Variations: Develop models capable of robustly recognizing finger vein
patterns across diverse populations and environmental conditions, minimizing the impact of
factors such as age, gender, health conditions, and lighting variations.
• Real-time Performance: Optimize the recognition process to ensure real-time performance,
enabling quick authentication responses suitable for applications requiring fast and seamless
user verification.

Advantages

▪ Reliability: This method of biometric authentication is far more reliable as it is difficult to


forge. Unlike other biometric scans which scan things that are present on the surface.
▪ Hygienic: Vein recognition is a contact-free method. That means a person is not required to
make contact with the device for the scan to go through with.
▪ Low Maintenance: As this technology requires less or no contact with the device. In other
authentication methods, a person is required to touch the device.

Disadvantages

• As this method is relatively new, so the technology is expensive.


• Due to this, it is only accessible to a few companies and organizations.
• People suffering from arthritis cant use it as they can’t balance their limbs properly.
• Also, it would be difficult to place their hand on the scanner.
• This technology is still not that reachable to the world, and some cultures refuse to use it.
Future scope

• Improved Accuracy and Robustness: Continued research can focus on developing more
advanced deep learning architectures and algorithms to further improve the accuracy and
robustness of finger vein recognition systems.

• Multi-Modal Fusion: Integration of multiple biometric modalities, such as finger vein


recognition with fingerprint, iris, or facial recognition, can enhance the overall security and
reliability of biometric authentication systems.

• Privacy-Preserving Techniques: As biometric data is highly sensitive, there is a growing


need for privacy-preserving techniques in biometric recognition systems.

• Real-Time and Resource-Efficient Implementation: Efficient deployment of finger vein


recognition systems on resource-constrained devices, such as smartphones or embedded
systems, remains a challenge. Future work can focus on developing lightweight deep learning
models and optimization techniques to enable real-time inference and operation on low-power
devices without compromising accuracy.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy