Identification of Image Forgeries using Machine Learning - A Review

Abstract

Forgery in images is the manipulation of digital images using techniques like copy-move, splicing, removal of parts of image. Image forgery detection is a crucial task in digital image processing field. The growth and use of digital images in various industries such as forensics, journalism and scientific research has increased the number of manipulated and forged images. New and advanced editing tools and techniques are capable of easily manipulating images without leaving traces, which can lead to negative impact for individuals and society. Therefore, the need for reliable and efficient forgery detection techniques has become more important than ever. They are required to protect the authenticity of images and avoid the spread of fabricated and fake news. In this study the overview of the existing methods for identifying forgeries in images, and the summary of the issues found in these methods are discussed.