Loading…
A robust face recognition system using convolutional neural networks
Biometric technologies are now widely employed to assess and confirm individuals’ distinct features to achieve recognition or authentication. Face recognition-based systems are the most of these, as these are more automated, simple to operate, and do not require constant user assistance. The current...
Saved in:
Main Authors: | , , |
---|---|
Format: | Conference Proceeding |
Language: | English |
Subjects: | |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Biometric technologies are now widely employed to assess and confirm individuals’ distinct features to achieve recognition or authentication. Face recognition-based systems are the most of these, as these are more automated, simple to operate, and do not require constant user assistance. The current work used a three-layer Convolutional Neural Network (CNN) to identify individuals based on their facial aspects, with the convolutional and pooling layers extracting features, to allow the fully connected layer to transforms the extracted characteristics into final output. Four CNN models were examined in this process: ResNet-50, GoogleNet, AlexNet, and DenseNet-201, and these were trained and tested on the Faces95, Yale, and ORL datasets. These experiments showed that the DenseNet-201 model achieved the best accuracy and training time results, while comparison with previous approaches confirmed the superiority of the proposed method. |
---|---|
ISSN: | 0094-243X 1551-7616 |
DOI: | 10.1063/5.0204692 |