A Blind Forensic Algorithm for Detecting Copy-Paste Images Based on Krawtchouk Invariant Moments

For a common class of copy-paste tampering in digital image forensics,a forensics algorithm is proposed based on Discrete Wavelet Transform(DWT) and radial krawtchouk invariance moment.The algorithm uses DWT to extract low frequency components of the image to propose the krawtchouk invariance moment of the blocks of low frequency of components.This characterization method is robust against image rotation.The feature vectors are then sorted according to the dictionary.Combined with mathematical morphology for image temper detection and region localization.Experiments show that the algorithm can locate the tampered region and strong robustness to the rotation of the paste region.