Exposing Blurred Image Forgeries through Blind Image Restoration

Abstract

When creating a digital forgery, it is often necessary to combine several images. Image splicing is one such form of tampering. In recent years, researchers have proposed various methods for detecting such splicing. In this paper, we propose a novel method for blurred image splicing localization based on blind image restoration. We use blur parameters estimation through the spectrum characteristics of blurred images in order to restore the spliced region and the rest of the image. We also develop a new measure to assist in inconsistent region segmentation in restored images that contain large amounts of ringing effect. Experimental results show efficacy of the method even if the images to be tested have been noised or compressed with a low quality factor.