Forgery Detection in Digital Images Via Discrete Wavelet and Discrete Cosine Transforms

When an image is forged through some sophisticated software editing tool, with imperceptibility being the goal, the idea is not to leave any observable trace that may help to distinguish the forged image from the original one at least with a naked eye. We believe that no matter which type of forgery is employed, there ought to be imperfections in the tampered image that may eventually prove its fakery. Even, forging an image with noble intentions, needs a reasonable amount of care. With this in perspective, we present a forgery detection method that depends on the discrete wavelet transform (DWT) as well as the discrete cosine transform (DCT) for feature reduction. The DCT is applied to the individual blocks obtained after dividing the DWTed image. The blocks are then compared on the basis of correlation coefficients. A mask-based tampering method is also developed as part of the experiments in order to test the detection method. The method shows interesting results when compared to two methods from the literature. (C) 2017 Elsevier Ltd. All rights reserved.