Analysis of Supervised and Unsupervised Deep Learning Approaches for Identifying and Localizing Image Forgeries
DOI:
https://doi.org/10.47392/irjash.2023.S003Keywords:
Deep Learning, Image Processing, CNN, Unsupervised SelfConsistency Learning, Forgery Detection, Forgery Localization, Splicing, Copy-MoveAbstract
The field of image forensics has become important in recent years as the use of digital images continues to grow. With the rise of sophisticated image editing software, it has become increasingly difficult to detect whether an image has been tampered with or not. Moreover, social media platforms have made the distribution of forged images to the general public a simple task. It is hence very important to develop automated methods that can detect such forgeries. In this study, we detect and localize splicing and copy-move image forgeries in images by using two different deep-learning techniques - Convolutional Neural Networks (CNN), which is a supervised approach and Self-Consistency Learning, an unsupervised approach. By comparing and contrasting the performance of these approaches, the research aims to gain a better understanding of how to effectively detect and locate image forgeries using deep learning. Ultimately, this research will contribute to the development of more reliable and accurate image forensic techniques, which will be of great benefit in various fields such as criminal investigations, digital media, and photojournalism.
Downloads
Published
Issue
Section
License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.