Novel Method to Detect Multiple Cloning in Targeted Image Invariant to Rotation

Abstract

Digital Media plays a vital role in our society. Digital image is one of the most important parts of digital media. There are various cases filed due to an image tampering on social networking websites. To detect tampering in an image, various techniques are available, but still these techniques have drawbacks. These techniques are not able to detect certain type of tampering in an image, like multiple region duplication with rotation. In this paper, the proposed system uses the approach, which is block based. In block based approach an image is divided into overlapping blocks, for the betterment of results the block is then divided diagonally into four subblocks. Feature vectors are then calculated using Zernike Moments as this is invariant to Rotation and it is insensitive to image noise. Thus the proposed approach is going to be crucial in digital image tamper detection in the upcoming era of digital media.