Digital Image Forensics Using EM Algorithm

Digital image forensics has become a very important research topic. This paper proposes a method to detect the forgery of digital image by (1) computing the interpolated coefficient for the images using expectation-maximization (EM) algorithm, (2) generating the probability map, (3) obtaining the frequency spectrum of the probability map, (4) determining whether an image has been tampered based on the periodicity characteristics of the spectrum. The experimental results show that our approach is effective to detect three different image forgeries: (a) air-brush or brush strokes, (b) different blurring filters, and (c) composite image taken from different cameras.