Detecting Forgery in Duplicated Region using Keypoint Matching

Region duplication is a common form of image manipulation where part of an image is pasted to another location to conceal undesirable contents. Mostly used region duplication detection methods are directly matching the block of image pixels. But it is not effective when the duplicated regions have geometrical and illumination distortions. The proposed method uses SR-PE (Scale rotation invariant pattern Entropy) for obtaining the matching pattern which is much faster compared to other algorithms. It generally aims to measure the spatial regularity of matching patterns formed by local keypoints. This method consist of various transformations from which the duplicated region can be identified, by estimating the transform between matched SIFT(Scale Invariant Feature Transform) keypoints. The SIFT algorithm along with the SR-PE and RANSAC algorithm helps to find the duplicated regions more effectively. Index Terms Digital image forensics; Region Duplication Detection, Image Feature Matching, Keypoints.