Abstract: Digital images can be forged easily with today's widely available image processing software. In this paper, we describe a passive approach to detect digital forgeries by checking inconsistencies of blocking artifact. Given a digital image, we find that the blocking artifacts introduced during JPEG compression could be used as a &numl;atural authentication code&.uml; A blocking artifact measure is then proposed based on the estimated quantization table using the power spectrum of the DCT coefficient histogram. Experimental results also demonstrate the validity of the proposed approach.

  author       = {Shuiming Ye and Qibin Sun and Ee-Chien Chang},
  url          = {http://www.ee.columbia.edu/~qibin/papers/qibin2007_icme_3.pdf},
  journal      = {Multimedia and Expo, 2007 IEEE International Conference on},
  year         = {2007},
  title        = {Detecting digital image forgeries by measuring inconsistencies of blocking artifact},
  pages        = {12--15},