A Hybrid Copy-Move Forgery Detection Technique Using Regional Similarity Indices

Different methods have been experimented for processing and detecting forgery in digital images. Image forgery involves various activities like copy-move forgery, image slicing, retouching, morphing etc. In copy-move forgery a portion within the image is copied and pasted on another part of the same image, generally to conceal or enhance certain portions of the image. This paper proposes a copy-move forgery detection using local fractal dimension and structural similarity indices. The image is classified into different texture regions based on the local fractal dimension. Forgery checking is thus confined to be among the portions within a region. Structural similarity index measure is applied to each block pair in each region to localize the forged portion. Experimental results prove that this hybrid method can effectively detect such kind of image tampering with minimum false positives.