Melola Discovery

← Melola Discovery

Open access Journal SINTA 1

REGRESSION AND CLASSIFICATION MODELS FOR HUMAN AGE PREDICTION

Jurnal Ilmiah Ilmu Terapan Universitas Jambi · Vol. 8 No. 2 (2024) · pp. 424–435

DOI
10.22437/jiituj.v8i2.32505
ISSN
2580-2259, 2580-2240
Published
2024-10-10
Publisher
LPPM Universitas Jambi
Language
en
License
creativecommons.org/licenses/by/4.0

Abstract

This study aims to enhance automated age prediction from facial images, a task with significant potential in security, law enforcement, and Human-Computer Interaction (HCI). While age estimation has seen progress, it remains a challenging problem due to the diverse factors influencing facial aging, such as genetics, environment, lifestyle, and facial expressions. These variations result in individuals of the same chronological age looking markedly different. Most existing age estimation methods rely on computationally intensive pre-trained models, often treated as "black boxes" with predefined input sizes and limited flexibility. To address these limitations, we propose using Convolutional Neural Networks (CNNs) for age prediction. Our approach combines classification and regression techniques to predict age more accurately. We applied our model to publicly available datasets, including FGNET, Adience, APPA-REAL, UTKFace, and All-Age-Face, encompassing images from constrained and unconstrained environments. The proposed CNN model was evaluated against existing pre-trained models, demonstrating comparable performance in age prediction tasks. Both classification and regression results underscored the model's accuracy, offering additional benefits in reduced computational complexity, increased flexibility, and adaptability. This study introduces a CNN-based approach as a viable alternative to pre-trained models for automated age prediction. It offers competitive accuracy while addressing critical limitations of current models, such as computational demands and lack of flexibility, thus contributing a more efficient solution for age estimation tasks in various real-world applications.

Keywords

Collaboration network of the authors

3 authors · 3 co-authorship links

Hover or tap a name to highlight its co-authors. Double-click to open the author's page.

  • Gulbarga University
  • Affiliation unknown
  • Different organisations
  • Same organisation
  • Affiliation unknown
  • Thicker line = more joint articles · bigger circle = more articles
Show as table
Co-authorOrganisationJoint articles

Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.

Metadata source: OAI-PMH (metadata jurnal) https://online-journal.unja.ac.id/index.php/JIITUJ/oai, first harvested 5 Nov 2026, last seen 15 Nov 2026.
Original metadata (oai_dc)
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en">REGRESSION AND CLASSIFICATION MODELS FOR HUMAN AGE PREDICTION</dc:title>
	<dc:creator xml:lang="en">Patil, Bhuvaneshwari</dc:creator>
	<dc:creator xml:lang="en">Rumma, Shivanand S.</dc:creator>
	<dc:creator xml:lang="en">Hangarge, Mallikarjun</dc:creator>
	<dc:subject xml:lang="en">Classification</dc:subject>
	<dc:subject xml:lang="en">Convolutional Neural Network</dc:subject>
	<dc:subject xml:lang="en">Regression</dc:subject>
	<dc:description xml:lang="en">This study aims to enhance automated age prediction from facial images, a task with significant potential in security, law enforcement, and Human-Computer Interaction (HCI). While age estimation has seen progress, it remains a challenging problem due to the diverse factors influencing facial aging, such as genetics, environment, lifestyle, and facial expressions. These variations result in individuals of the same chronological age looking markedly different. Most existing age estimation methods rely on computationally intensive pre-trained models, often treated as "black boxes" with predefined input sizes and limited flexibility. To address these limitations, we propose using Convolutional Neural Networks (CNNs) for age prediction. Our approach combines classification and regression techniques to predict age more accurately. We applied our model to publicly available datasets, including FGNET, Adience, APPA-REAL, UTKFace, and All-Age-Face, encompassing images from constrained and unconstrained environments. The proposed CNN model was evaluated against existing pre-trained models, demonstrating comparable performance in age prediction tasks. Both classification and regression results underscored the model's accuracy, offering additional benefits in reduced computational complexity, increased flexibility, and adaptability. This study introduces a CNN-based approach as a viable alternative to pre-trained models for automated age prediction. It offers competitive accuracy while addressing critical limitations of current models, such as computational demands and lack of flexibility, thus contributing a more efficient solution for age estimation tasks in various real-world applications.</dc:description>
	<dc:publisher xml:lang="en">LPPM Universitas Jambi</dc:publisher>
	<dc:date>2024-10-10</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://online-journal.unja.ac.id/JIITUJ/article/view/32505</dc:identifier>
	<dc:identifier>10.22437/jiituj.v8i2.32505</dc:identifier>
	<dc:source xml:lang="en">Jurnal Ilmiah Ilmu Terapan Universitas Jambi; Vol. 8 No. 2 (2024): Volume 8, Nomor 2, December 2024; 424-435</dc:source>
	<dc:source>2580-2259</dc:source>
	<dc:source>2580-2240</dc:source>
	<dc:language>en</dc:language>
	<dc:relation>https://online-journal.unja.ac.id/JIITUJ/article/view/32505/19139</dc:relation>
	<dc:rights xml:lang="en">Copyright (c) 2024 Bhuvaneshwari Patil, Shivanand S. Rumma, Mallikarjun Hangarge</dc:rights>
	<dc:rights xml:lang="en">https://creativecommons.org/licenses/by/4.0</dc:rights>
</oai_dc:dc>