Enhancing CNN for Forensics Age Estimation Using CGAN and Pseudo-Labelling
Age estimation using forensics odontology is an important process in identifying victims in criminal or mass disaster cases. Traditionally, this process is done manually by human expert. However, the speed and accuracy may vary depending on the expertise level of the human expert and other human fac...
Saved in:
Main Authors: | , , |
---|---|
Other Authors: | |
Format: | Article |
Published: |
Tech Science Press
2024
|
Subjects: | |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.uniten.dspace-34700 |
---|---|
record_format |
dspace |
spelling |
my.uniten.dspace-347002024-10-14T11:21:51Z Enhancing CNN for Forensics Age Estimation Using CGAN and Pseudo-Labelling Alkaabi S. Yussof S. Al-Mulla S. 59070935100 16023225600 36473139200 age estimation convolutional neural network Dental forensics generative adversarial network pseudo-labelling Convolution Convolutional neural networks Deep learning Digital forensics Image enhancement Neural network models Age estimation Convolutional neural network Dental forensic Dental images Forensic odontology Human expert Labelings Neural network model Pseudo-labeling Synthetic images Generative adversarial networks Age estimation using forensics odontology is an important process in identifying victims in criminal or mass disaster cases. Traditionally, this process is done manually by human expert. However, the speed and accuracy may vary depending on the expertise level of the human expert and other human factors such as level of fatigue and attentiveness. To improve the recognition speed and consistency, researchers have proposed automated age estimation using deep learning techniques such as Convolutional Neural Network (CNN). CNN requires many training images to obtain high percentage of recognition accuracy. Unfortunately, it is very difficult to get large number of samples of dental images for training the CNN due to the need to comply to privacy acts. A promising solution to this problem is a technique called Generative Adversarial Network (GAN). GAN is a technique that can generate synthetic images that has similar statistics as the training set. A variation of GAN called Conditional GAN (CGAN) enables the generation of the synthetic images to be controlled more precisely such that only the specified type of images will be generated. This paper proposes a CGAN for generating new dental images to increase the number of images available for training a CNN model to perform age estimation. We also propose a pseudo-labelling technique to label the generated images with proper age and gender. We used the combination of real and generated images to train Dental Age and Sex Net (DASNET), which is a CNN model for dental age estimation. Based on the experiment conducted, the accuracy, coefficient of determination (R2) and Absolute Error (AE) of DASNET have improved to 87%, 0.85 and 1.18 years respectively as opposed to 74%, 0.72 and 3.45 years when DASNET is trained using real, but smaller number of images. � 2023 Tech Science Press. All rights reserved. Final 2024-10-14T03:21:51Z 2024-10-14T03:21:51Z 2023 Article 10.32604/cmc.2023.029914 2-s2.0-85141892582 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85141892582&doi=10.32604%2fcmc.2023.029914&partnerID=40&md5=3d6e4208eda7c0816b49d401c1b39264 https://irepository.uniten.edu.my/handle/123456789/34700 74 2 2499 2516 All Open Access Gold Open Access Tech Science Press Scopus |
institution |
Universiti Tenaga Nasional |
building |
UNITEN Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Tenaga Nasional |
content_source |
UNITEN Institutional Repository |
url_provider |
http://dspace.uniten.edu.my/ |
topic |
age estimation convolutional neural network Dental forensics generative adversarial network pseudo-labelling Convolution Convolutional neural networks Deep learning Digital forensics Image enhancement Neural network models Age estimation Convolutional neural network Dental forensic Dental images Forensic odontology Human expert Labelings Neural network model Pseudo-labeling Synthetic images Generative adversarial networks |
spellingShingle |
age estimation convolutional neural network Dental forensics generative adversarial network pseudo-labelling Convolution Convolutional neural networks Deep learning Digital forensics Image enhancement Neural network models Age estimation Convolutional neural network Dental forensic Dental images Forensic odontology Human expert Labelings Neural network model Pseudo-labeling Synthetic images Generative adversarial networks Alkaabi S. Yussof S. Al-Mulla S. Enhancing CNN for Forensics Age Estimation Using CGAN and Pseudo-Labelling |
description |
Age estimation using forensics odontology is an important process in identifying victims in criminal or mass disaster cases. Traditionally, this process is done manually by human expert. However, the speed and accuracy may vary depending on the expertise level of the human expert and other human factors such as level of fatigue and attentiveness. To improve the recognition speed and consistency, researchers have proposed automated age estimation using deep learning techniques such as Convolutional Neural Network (CNN). CNN requires many training images to obtain high percentage of recognition accuracy. Unfortunately, it is very difficult to get large number of samples of dental images for training the CNN due to the need to comply to privacy acts. A promising solution to this problem is a technique called Generative Adversarial Network (GAN). GAN is a technique that can generate synthetic images that has similar statistics as the training set. A variation of GAN called Conditional GAN (CGAN) enables the generation of the synthetic images to be controlled more precisely such that only the specified type of images will be generated. This paper proposes a CGAN for generating new dental images to increase the number of images available for training a CNN model to perform age estimation. We also propose a pseudo-labelling technique to label the generated images with proper age and gender. We used the combination of real and generated images to train Dental Age and Sex Net (DASNET), which is a CNN model for dental age estimation. Based on the experiment conducted, the accuracy, coefficient of determination (R2) and Absolute Error (AE) of DASNET have improved to 87%, 0.85 and 1.18 years respectively as opposed to 74%, 0.72 and 3.45 years when DASNET is trained using real, but smaller number of images. � 2023 Tech Science Press. All rights reserved. |
author2 |
59070935100 |
author_facet |
59070935100 Alkaabi S. Yussof S. Al-Mulla S. |
format |
Article |
author |
Alkaabi S. Yussof S. Al-Mulla S. |
author_sort |
Alkaabi S. |
title |
Enhancing CNN for Forensics Age Estimation Using CGAN and Pseudo-Labelling |
title_short |
Enhancing CNN for Forensics Age Estimation Using CGAN and Pseudo-Labelling |
title_full |
Enhancing CNN for Forensics Age Estimation Using CGAN and Pseudo-Labelling |
title_fullStr |
Enhancing CNN for Forensics Age Estimation Using CGAN and Pseudo-Labelling |
title_full_unstemmed |
Enhancing CNN for Forensics Age Estimation Using CGAN and Pseudo-Labelling |
title_sort |
enhancing cnn for forensics age estimation using cgan and pseudo-labelling |
publisher |
Tech Science Press |
publishDate |
2024 |
_version_ |
1814061191417298944 |
score |
13.214268 |