Forged region detection for scanned images

Abstract

As analog media has been replaced with digital media, legal issues in digital media have been in the spot light. In this paper, we propose a detection method of forged regions scanned images. We analyze a specific noise in the scanned images from many scanners. A suspicious image is sliced into small non-overlapping blocks. We extract the spectral noise from each block and identify the scanner model. In experiments, three synthesis images are examined to prove the performance of our scheme. © 2012 Springer Science+Business Media B.V.