Noisy facial recognition with convolutional neural network (CNN)

The aim of this research is to evaluate how Gaussian noise affects the accuracy of two Convolutional Neural Network (CNN) architectures—LeNet and AlexNet—in facial recognition applications. By utilizing the Faces95 dataset, the networks were subjected to varying levels of Gaussian noise to determ...

Full description

Saved in:
Bibliographic Details
Main Author: Heng, Yin Qi
Other Authors: Anamitra Makur
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2024
Subjects:
CNN
Online Access:https://hdl.handle.net/10356/176638
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-176638
record_format dspace
spelling sg-ntu-dr.10356-1766382024-05-24T15:50:32Z Noisy facial recognition with convolutional neural network (CNN) Heng, Yin Qi Anamitra Makur School of Electrical and Electronic Engineering EAMakur@ntu.edu.sg Engineering Machine learning CNN Facial recognition The aim of this research is to evaluate how Gaussian noise affects the accuracy of two Convolutional Neural Network (CNN) architectures—LeNet and AlexNet—in facial recognition applications. By utilizing the Faces95 dataset, the networks were subjected to varying levels of Gaussian noise to determine how noise impacts model performance. The study revealed that while LeNet was moderately resilient to noise, AlexNet's accuracy suffered significantly. Specifically, at a noise variance of 0.05, AlexNet experienced a marked accuracy reduction compared to LeNet. This divergence highlights the importance of architectural complexity in noise sensitivity. Importantly, the study shows that incorporating noise during training can significantly improve the noise robustness of both models, indicating that noise-adaptive training is a powerful strategy for enhancing CNN performance in visually noisy environments. These findings deepen our understanding of CNN resilience against visual noise and underscore the potential of revisiting simpler CNN architectures to unlock and leverage innate noise-resistant features. The insights gained here pave the way for developing more reliable facial recognition systems that can function effectively in real-world noise disturbances. Bachelor's degree 2024-05-19T23:27:11Z 2024-05-19T23:27:11Z 2024 Final Year Project (FYP) Heng, Y. Q. (2024). Noisy facial recognition with convolutional neural network (CNN). Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/176638 https://hdl.handle.net/10356/176638 en A3005 – 231 application/pdf Nanyang Technological University
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering
Machine learning
CNN
Facial recognition
spellingShingle Engineering
Machine learning
CNN
Facial recognition
Heng, Yin Qi
Noisy facial recognition with convolutional neural network (CNN)
description The aim of this research is to evaluate how Gaussian noise affects the accuracy of two Convolutional Neural Network (CNN) architectures—LeNet and AlexNet—in facial recognition applications. By utilizing the Faces95 dataset, the networks were subjected to varying levels of Gaussian noise to determine how noise impacts model performance. The study revealed that while LeNet was moderately resilient to noise, AlexNet's accuracy suffered significantly. Specifically, at a noise variance of 0.05, AlexNet experienced a marked accuracy reduction compared to LeNet. This divergence highlights the importance of architectural complexity in noise sensitivity. Importantly, the study shows that incorporating noise during training can significantly improve the noise robustness of both models, indicating that noise-adaptive training is a powerful strategy for enhancing CNN performance in visually noisy environments. These findings deepen our understanding of CNN resilience against visual noise and underscore the potential of revisiting simpler CNN architectures to unlock and leverage innate noise-resistant features. The insights gained here pave the way for developing more reliable facial recognition systems that can function effectively in real-world noise disturbances.
author2 Anamitra Makur
author_facet Anamitra Makur
Heng, Yin Qi
format Final Year Project
author Heng, Yin Qi
author_sort Heng, Yin Qi
title Noisy facial recognition with convolutional neural network (CNN)
title_short Noisy facial recognition with convolutional neural network (CNN)
title_full Noisy facial recognition with convolutional neural network (CNN)
title_fullStr Noisy facial recognition with convolutional neural network (CNN)
title_full_unstemmed Noisy facial recognition with convolutional neural network (CNN)
title_sort noisy facial recognition with convolutional neural network (cnn)
publisher Nanyang Technological University
publishDate 2024
url https://hdl.handle.net/10356/176638
_version_ 1800916424047198208