Copy-move is a simple and effective operation to create image forgeries where a continuous part of pixels are copied and pasted to a different location in the same image, after possible geometrical adjustment and illumination adjustment. Most existing copy-move forgery detection methods are based on directly matching blocks of image pixels or transform coefficient, but not are effective when the copy-move regions have geometrical or illumination distortions. There are various types of forgeries possible on the digital image tampering, copy-move forgery and image compositing. In this paper, we describe a new copy-move forgery detection method that is robust to distortions of the duplicated regions. Our methods starts from estimation the transform between matched scale invariant feature transform (SIFT) keypoint, which are effective to geometrical and illumination and then find all pixels within the duplicated regions. The propose method shows robust detection on an automatically synthesized forgery image database with duplicated and distorted regions.
Loading....