Detecting copy-paste forgeries using transform-invariant features

Image manipulation has become commonplace with growing easy access to powerful computing abilities. One of the most common types of image forgeries is the copy-paste forgery, wherein a region from an image is replaced with another region from the same image (with possible transformations). A brute-force approach to finding identical regions suffers from many problems, including inability to detect transformed regions. In this paper, we propose a novel technique based on transform-invariant features. These are obtained by adapting the MPEG-7 image signature tools to specifically deal with copy-paste forgeries. Results are provided to justify the application of these tools in copy-paste forgery detection.