Copy Move Forgery Detection with Quadtree Decomposition Segmentation

The increasing usage of digital images and the wide availability of easy-to-use image editors have made the authenticity of images questionable. Copy-move is one of the most applied forgery types. The keypoint-based copy move forgery detection methods in the literature cannot detect forgeries with smooth regions. In this paper to overcome this problem, a new copy-move forgery detection technique is proposed. In the proposed scheme, the input image is segmented into two sub-image via Quadtree Decomposition. The sub-images are labeled as smooth and textured. The textural form of the smooth labeled segment is obtained based on LBPROT approach. Then the keypoints are extracted from both textured segment and textural form of the smooth segment. After that, they are matched with each other to obtain forgery clues. By matching the extracted keypoints, it is revealed whether the image is forged or not. The experimental results show that the proposed method is robust under scaling and rotation attacks.