An automatic enhanced filters with frequency-based copy-move forgery detection for social media images

Using images in various file types has become common in the modern digital world, such as social media posts, research reports, and legal documents. The availability of low-cost image manipulation tools has made it easier to change images, potentially leading to undetected image fraud. One such form...

Full description

Saved in:
Bibliographic Details
Published in:Multimedia tools and applications 2024, Vol.83 (1), p.1513-1538
Main Authors: Abir, Noor Atikah Mat, Warif, Nor Bakiah Abd, Zainal, Nurezayana
Format: Article
Language:English
Subjects:
Computer Communication Networks
Computer Science
Data Structures and Information Theory
Digital imaging
Digital media
Forgery
Image enhancement
Image filters
Image manipulation
Image quality
Multimedia Information Systems
Social networks
Special Purpose and Application-Based Systems
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Using images in various file types has become common in the modern digital world, such as social media posts, research reports, and legal documents. The availability of low-cost image manipulation tools has made it easier to change images, potentially leading to undetected image fraud. One such form of image manipulation is copy-move forgery (CMF), which is difficult to detect due to the similarities in image features. There have been efforts to detect CMF using copy-move forgery detection (CMFD) methods. However, most research has focused on CMF images with attacks rather than social media. Social media has contributed significantly to the image manipulation phenomenon, and additional post-processing techniques on social media platforms have affected the efficiency of the CMFD methods. Therefore, this research proposes a two-stage pre-processing phase combined with frequency-based CMFD to detect CMF images in different social media platforms. The first stage includes automatic image selection, followed by image enhancement with filters to improve image quality. The experimental results show that the proposed method achieves the highest detection score compared to existing CMFD methods, with an average score of 90% for CMF-Facebook, 91% for CMF-WhatsApp, and 85% for CMF-Twitter. This research highlights the importance of developing solutions to detect image forgery in social media and the potential of combining pre-processing with frequency-based methods to improve results.
ISSN:1380-7501
1573-7721
DOI:10.1007/s11042-023-15506-7