Total variation and high order total variation methods for image reconstruction rely on sparsity throughout some domain. The spatial variation of the sparsity is not typically accounted for, however. We propose a new method for image reconstruction from non-uniform Fourier data that uses edge detection to indicate regions of sparsity and targets regularization appropriately.

This presentation is part of Contributed Presentation “CP3 - Contributed session 3”

Authors:

Victor Churchill (Dartmouth College)

Anne Gelb (Dartmouth College )

Richard Archibald (Oak Ridge National Laboratory)

Keywords:

edge detection, image reconstruction, non-uniform fourier data, total variation regularization