IIUM Repository

An improvised CNN model for fake image detection

Hamid, Yasir and Elyassami, Sanaa and Gulzar, Yonis and Balasaraswathi, Veeran Ranganathan and Habuza, Tetiana and Wani, Sharyar (2022) An improvised CNN model for fake image detection. International Journal of Information Technology. ISSN 2511-2104 E-ISSN 2511-2112 (In Press)

[img]
Preview
PDF
Download (560kB) | Preview

Abstract

The last decade has witnessed a multifold growth of image data courtesy of the emergence of social networking services like Facebook, Instagram, LinkedIn etc. The major menace faced by today’s world is the issue of doctored images, where-in the photographs are altered using a rich set of ways like splicing, copy-move, removal to change their meaning and hence demands serious mitigation mechanisms to be thought of. The problem when seen from the prism of Artificial intelligence is a binary classification one, where-in the characterization must be drawn between the original and the manipulated images. This research work proposes a computer vision model based on Convolution Neural Networks for fake image detection. A comparative analysis of 6 popular traditional machine learning models and 6 different CNN architectures to select the best possible model for further experimentation. The proposed model based on ResNet50 employed with powerful preprocessing techniques results in a perfect fake image detector having a total accuracy of 0.99 having an improvement of around 18% performance with other models.

Item Type: Article (Journal)
Uncontrolled Keywords: Artificial Intelligence; Classification; Convolution neural networks; Doctored images; Splicing
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7800 Electronics. Computer engineering. Computer hardware. Photoelectronic devices > TK7885 Computer engineering
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button): Kulliyyah of Information and Communication Technology
Kulliyyah of Information and Communication Technology
Depositing User: Dr. Sharyar Wani
Date Deposited: 07 Dec 2022 13:49
Last Modified: 07 Dec 2022 13:49
URI: http://irep.iium.edu.my/id/eprint/101720

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year