Stuart Weir
Enhancing AI-Generated Image Detection with a Novel Approach and Comparative Analysis
Weir, Stuart; Khan, Muhammad Shahbaz; Moradpoor, Naghmeh; Ahmad, Jawad
Authors
Muhammad Shahbaz Khan M.Khan2@napier.ac.uk
Student Experience
Dr Naghmeh Moradpoor N.Moradpoor@napier.ac.uk
Associate Professor
Dr Jawad Ahmad J.Ahmad@napier.ac.uk
Visiting Lecturer
Abstract
This study explores advancements in AI-generated image detection, emphasizing the increasing realism of images, including deepfakes, and the need for effective detection methods. Traditional Convolutional Neural Networks (CNNs) have shown success but face limitations in generalization and accuracy , particularly with newer technologies like Diffusion Models. With the evolution of AI image generation models, from CNNs to Generative Adversarial Networks (GANs) and Diffusion Models, detecting synthetic images has become more challenging. Issues include dataset diversity, adversarial attacks, and inconsistencies in pre-processing methods. While state-of-the-art models like CNNs, Vision Transformers (ViTs), and hybrid approaches exist, their accuracy in detecting increasingly sophisticated fake images remains suboptimal. This research proposes a novel hybrid detection model combining CNNs and ViTs with an additional attention mechanism layer. This structure aims to improve the interaction between local and global features, enhancing detection accuracy. The model was trained using the CIFAKE dataset, which contains 120,000 real and AI-generated images. The added attention mechanism enhances feature extraction, addressing limitations in existing models when faced with next-generation synthetic images. The hybrid CNN/ViT+Attention model demonstrated improved detection accuracy, achieving 99.77%, surpassing previous methods. This research lays a foundation for stronger AI-generated image detection, helping to mitigate the risks of synthetic image fraud.
Citation
Weir, S., Khan, M. S., Moradpoor, N., & Ahmad, J. (2024, December). Enhancing AI-Generated Image Detection with a Novel Approach and Comparative Analysis. Presented at 2024 17th International Conference on Security of Information and Networks (SIN), Sydney, Australia
Presentation Conference Type | Conference Paper (published) |
---|---|
Conference Name | 2024 17th International Conference on Security of Information and Networks (SIN) |
Start Date | Dec 2, 2024 |
End Date | Dec 4, 2024 |
Acceptance Date | Oct 15, 2024 |
Online Publication Date | Feb 13, 2025 |
Publication Date | 2025 |
Deposit Date | Oct 16, 2024 |
Publicly Available Date | Feb 13, 2025 |
Publisher | Institute of Electrical and Electronics Engineers |
Peer Reviewed | Peer Reviewed |
Book Title | 2024 17th International Conference on Security of Information and Networks (SIN) |
DOI | https://doi.org/10.1109/SIN63213.2024.10871900 |
Keywords | Vision Transformer, Convolutional Neural Networks, Hybrid models, attention mechanism, CIFAKE dataset |
Public URL | http://researchrepository.napier.ac.uk/Output/3889224 |
Related Public URLs | https://www.sinconf.org/sin2024/ |
Files
Enhancing AI-Generated Image Detection With a Novel Approach and Comparative Analysis (accepted version)
(360 Kb)
PDF
You might also like
A Novel Cosine-Modulated-Polynomial Chaotic Map to Strengthen Image Encryption Algorithms in IoT Environments
(2024)
Presentation / Conference Contribution
Image Encryption Using A Novel Orbital-Extraction Permutation Technique and Chaotic Key Generation
(2024)
Presentation / Conference Contribution
Chaotic Quantum Encryption to Secure Image Data in Post Quantum Consumer Technology
(2024)
Journal Article
PermutEx: Feature-Extraction-Based Permutation — A New Diffusion Scheme for Image Encryption Algorithms
(2024)
Presentation / Conference Contribution
Downloadable Citations
About Edinburgh Napier Research Repository
Administrator e-mail: repository@napier.ac.uk
This application uses the following open-source libraries:
SheetJS Community Edition
Apache License Version 2.0 (http://www.apache.org/licenses/)
PDF.js
Apache License Version 2.0 (http://www.apache.org/licenses/)
Font Awesome
SIL OFL 1.1 (http://scripts.sil.org/OFL)
MIT License (http://opensource.org/licenses/mit-license.html)
CC BY 3.0 ( http://creativecommons.org/licenses/by/3.0/)
Powered by Worktribe © 2025
Advanced Search