Restoration

Restoration algorithms alter the smoothness or shape of image data without changing its physical interpretation: denoising, resampling, spatial averaging and binary clean-up.

Denoising and regularization

TVD([key])	Total variation denoising interface.
tvd(img[, method, weight, max_num_iter, eps])	Inline application of TVD.
split_bregman_tvd(img[, mu, omega, ell, ...])	Split Bregman algorithm for TV denoising.
H1_regularization(img, mu[, omega, dim, solver])	Inline application of H1 regularization.
Median([key])	Median through skimage.filter.rank.

Resizing

Resize([ref_image, shape, fx, fy, ...])	Object for resizing 2d arrays.
resize(image[, ref_image, shape, fx, fy, ...])	Function wrapper to Resize object.
equalize_voxel_size(image[, voxel_size])	Resize routine which keeps physical dimensions, but unifies the voxel length.
uniform_refinement(image, levels)	Uniform refinement.

Spatial averaging

VolumeAveraging(rev, mask[, labels])	Average a field over representative elementary volumes (REV).
REV(size, img)	Class defining an representative elementary volume.
volume_average(img, mask, size)	Fast-access function for volume averaging.
porosity_based_averaging(labels, porosity, ...)	Custom (FFUM) averaging function that uses porosity in averaging.

Binary clean-up

BinaryFillHoles([area_threshold, key])	Wrapper for filing holes in binary images with skimage.
BinaryRemoveSmallObjects([min_size, key])	Wrapper for removing small objects in binary images with skimage.
BinaryLocalConvexCover([cover_patch_size, key])	Local convex cover using local convex hulls with skimage.