An aliasing detection algorithm based on suspicious colocalizations of Fourier coefficients

We propose a new algorithm able to detect the presence and the localization of aliasing in a single digital image. Considering the image in Fourier domain, the fact that two frequencies in aliasing relation contribute to similar parts of the image domain is a suspicious coincidence, that we detect with an a-contrario model. This leads to a localization of the aliasing phenomenon in both spatial and spectral domains, with a detection algorithm that keeps control of the number of false alarms. Experiments on several images show that this new method favorably compares to the state of the art, and opens interesting perspectives in terms of image enhancement.

[1]  Dennis Gabor,et al.  Theory of communication , 1946 .

[2]  C. Latry,et al.  Super resolution: quincunx sampling and fusion processing , 2003, IGARSS.

[3]  Jean-Michel Morel,et al.  From Gestalt Theory to Image Analysis , 2008 .

[4]  Gwendoline Blanchet,et al.  Automatic detection of well sampled images via a new ringing measure , 2010, 2010 IEEE International Conference on Acoustics, Speech and Signal Processing.

[5]  Jean-Michel Morel,et al.  From Gestalt Theory to Image Analysis: A Probabilistic Approach , 2007 .

[6]  Amy R. Reibman,et al.  A no-reference Spatial Aliasing Measure for digital image resizing , 2008, 2008 15th IEEE International Conference on Image Processing.

[7]  Lionel Moisan,et al.  Periodic Plus Smooth Image Decomposition , 2011, Journal of Mathematical Imaging and Vision.

[8]  Alan C. Bovik,et al.  The analytic image , 1997, Proceedings of International Conference on Image Processing.

[9]  Amir Said A New Class of Filters for Image Interpolation and Resizing , 2007, 2007 IEEE International Conference on Image Processing.

[10]  Ingrid Daubechies,et al.  Ten Lectures on Wavelets , 1992 .