Figure 12From: Counter-forensics of SIFT-based copy-move detection by means of keypoint classification Number and classes of keypoints for each iteration of the attack on test image face2.jpg . VLFeat only (top) and VLFeat+RobHess (down). In the latter case, the total amount of keypoints is the union of the keypoints detected by the two techniques. The values in the upper portion of each plot correspond to the PSNR (full image) of even iterations.Back to article page