Using Fractal Dimension and Singular Values for Image Forgery Detection and Localization

Verification of the authenticity of images that are circulated through media and public outlets is a critical issue in image forensics. Copy-move forgery is the most common image forgery that conceals a particular feature from the scene by replacing it with another feature of the same image. We propose a hybrid approach based on local fractal dimension (LFD) and singular value decomposition (SVD) to efficiently detect and localize the duplicated regionsin images. First the image is partitioned into fixed size blocks and the local fractal dimension is estimated. In order to reduce the inherent computational complexity of fractal techniques, the image blocks are arranged in a B+ tree based on the LFD values. Pair of blocks within each segment is compared using singular values, to find regions that exhibit maximum resemblance. This reduces the need for comparison to the most suspected image portions alone. The experimental results show how effectively the method identifies the duplicated region; also presents the robustness of the method to detect forgery in the presence of after-copying manipulations such as rotation, blurring of edges and noise addition. The method proves to be effective in detecting multiple forgeries within the image. (C) 2016 The Authors. Published by Elsevier Ltd.