A cyclostationarity analysis applied to image forensics

The processing history of images plays an important role in many fields of digital image processing and computer vision. In this paper, we focus on geometrical transformations and show that images that have undergone such transformations contain hidden cyclostationary features. This makes possible employing the well-developed theory and efficient methods of cyclostationarity for blind analyzing of history of images in respect to geometrical transformations. To verify this, we also propose a cyclostationarity detection method and show how the traces of geometrical transformations in an image can be detected and the specific parameters of the transformation estimated. The method is based on the fact that a cyclostationary signal has a frequency spectrum correlated with a shifted version of itself.