Comparative Analysis of Techniques for Detecting Copy-Move Image Forgery

Abstract: Various image editing tools available today let alter image content without any expert guidance. Hence, image based crime is increasing at a faster rate than before. Various algorithms exist to detect copy-move forgery in images. In this paper, we compare two popular methods based on DWT and DCT to study and analyze their performance based on execution time and visual results. In both the methods, phase correlation has been applied after respective transformation, as a similarity criterion to estimate the translate offset between two similar regions of an image.

Keywords: DCT (Discrete Cosine Transform), DWT (Discrete wavelet Transform), lexicographical sorting, phase correlation.