Person Verification Based on Multimodal Biometric Recognition
Nowadays, person recognition has received significant attention due to broad applications in the security system. However, most person recognition systems are implemented based on unimodal biometrics such as face recognition or voice recognition. Biometric systems that adopted unimodal have li...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | en |
| Published: |
Pertanika
2022
|
| Subjects: | |
| Online Access: | http://ir.unimas.my/id/eprint/37876/1/Biometric1.pdf http://ir.unimas.my/id/eprint/37876/ http://www.pertanika.upm.edu.my/pjst/browse/regular-issue https://doi.org/10.47836/pjst.30.1.09 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Nowadays, person recognition has received significant attention due to broad applications
in the security system. However, most person recognition systems are implemented
based on unimodal biometrics such as face recognition or voice recognition. Biometric
systems that adopted unimodal have limitations, mainly when the data contains outliers
and corrupted datasets. Multimodal biometric systems grab researchers’ consideration
due to their superiority, such as better security than the unimodal biometric system and
outstanding recognition efficiency. Therefore, the multimodal biometric system based on
face and fingerprint recognition is developed in this paper. First, the multimodal biometric
person recognition system is developed based on Convolutional Neural Network (CNN)
and ORB (Oriented FAST and Rotated BRIEF) algorithm. Next, two features are fused
by using match score level fusion based
on Weighted Sum-Rule. The verification
process is matched if the fusion score is
greater than the pre-set threshold t. The
algorithm is extensively evaluated on UCI
Machine Learning Repository Database
datasets, including one real dataset with
state-of-the-art approaches. The proposed
method achieves a promising result in the
person recognition system. |
|---|