Fractal-dimension Based Algorithm to Detect Computer Graphics

Shenghong LiCommunications 2008

Abstract

The fast development of computer graphics generating techniques makes computer graphics highly photorealistic.It is increasingly harder to distinguish photorealistic computer graphics and photographic images by human eyes.In this paper,a fractal-dimension based algorithm is proposed to cope with the above-mentioned problem.Being different from conditional algorithm,the proposed algorithm takes advantage of the surface texture feature and self-similarity differences between these two kinds of images,and extracts the local fractal dimensions for image blocks to distinguish the computer graphics and photographic images.The experiment indicates that this fractal-dimension based algorithm shows a new direction for detecting photorealistic computer graphics,and is of a satisfying detection rate and a bright future for research as well.
BibTeX
@inproceedings{li2008fractal,
  author = {Shenghong Li},
  title = {{Fractal-dimension Based Algorithm to Detect Computer Graphics}},
  booktitle = {Communications},
  year = {2008},
}

Loading related content...

Figure