Copy-Move Image Forgery Detection Using Deep Learning Methods: A Review

Abstract

In recent years, the manipulation of digital images can be done with relative ease. This can be attributed to the technological advancement in the field of computing specifically with advanced, sophisticated image editing tool software. A majority of these software are user-friendly which results in its widespread use. However, this also presents a new problem where anyone with access to the software can easily manipulate an image and can use it for nefarious purposes such as spreading fake news. Due to this development of sophistication of tools and software like Adobe Photoshop, Pixir, and Affinity, digital images content is often simply manipulated and thus forged images are produced. Therefore, the process authenticating a digital image becomes difficult such as to distinguish between manipulated images and actual images through the naked eyes. Therefore, the importance of digital image forensics has attracted many researchers who are deeply involved in this area and has established many techniques for forgery detection in image forensics. Recently, deep learning approach has a high interest among researchers across the field and has shown good result in its application. Thus, forensic researchers attempt to apply deep learning approach as a method for detecting forgery image. This paper presents the understanding and extensive literature review of state-of-the-art techniques of deep learning in the detection of copy-move image forgery.