tgv_phantom

odl.phantom.geometric.tgv_phantom(space, edge_smoothing=0.2)

Piecewise affine phantom.

This phantom is taken from [Bre+2010] and includes both linearly varying regions and sharp discontinuities. It is designed to work well with Total Generalized Variation (TGV) type regularization.

Parameters:

spaceDiscretizedSpace, 2 dimensional

Discretized space in which the phantom is supposed to be created. Needs to be two-dimensional.

edge_smoothingnonnegative float, optional

Smoothing of the edges of the phantom, given as smoothing width in units of minimum pixel size.

Returns:

phantomspace-element

The generated phantom in space. Values have range [0, 1].

Notes

The original phantom is given by a specific image. In this implementation, we extracted the underlying parameters and the phantom thus works with spaces of any shape. Due to this, small variations may occur when compared to the original phantom.

References

[Bre+2010] K. Bredies, K. Kunisch, and T. Pock. Total Generalized Variation. SIAM Journal on Imaging Sciences, 3(3):492-526, Jan. 2010