AN INNOVATIVE AND EFFICIENT DEEP LEARNING ALGORITHM FOR COPY MOVE FORGERY DETECTION IN DIGITAL IMAGES

Abstract

Digital images have become more easier to tamper with the rapid advancement in image processing tools and software. Digital image manipulation takes part to deform the content of a picture in order to accomplish some deceit purposes. Such deceits are acknowledged as forgeries.The forensic people  need an effective means of observing such maliciously tampered data. Some important and efficient forgeries are; image retouchingand,splicing and Copy-move forgery(CMF). Copy-move forgery is the most significant type of image forgery due to its effective nes and simplicity. CMF is nothing but the image copying from one location and pasting it in another location. In this process, we conceal the existing data in the picture or to generate a simulated image. This region duplication process modifies the meaning of the picture totally. The precise forgery detection plays a key role in digital images. Image forgery detection approaches may be active or passive. Copy-move forgery detection (CMFD) is a passive-Blind image forgery detection method. CMFD mainly impresses on the speed and rigorous of the detection method. The proposed CMFD method presents a new approach to solve the issues in existing methods that the tampered area is resized or rotated after attachment. The proposed method is an innovative and efficient algorithm called Generalized Approximate Reasoning-Based Intelligence Control (GARIC) algorithm. Hence, GARIC deep learning approcch is used to detect the presence of falcification in images.