A SIFT and local features based integrated method for copy-move attack detection in digital image

To determine whether a particular image is authentic or not is a principal problem in digital image forensics. Many techniques have been proposed to detect clues of intended manipulation in the literature. In this paper, an integrated method based on SIFT and local features including color and texture features has been developed to deal with copy-move attack. Initially analogous feature points are paired by comparing SIFT features. Then a distance examination is applied to eliminate false matched points. Normalized RGB color feature of matched points neighborhoods is extracted and comparison between pairs produced by previous step is done. At last revised Gabor texture feature is utilized to dispose of the rest false matched pairs by applying statistic tools and final detection result is presented. Experimental results demonstrate that proposed method is able to detect altered area even though these regions are changed in geometry and shading.