fill blind pores and trim floating solids do the same thing

[pmeal]

fill blind pores and trim floating solids do the same thing

[mdrkok]

Not sure if that;s intentional but as far as I can tell they perform identical functions

[Zohaib-Atiq]

trim_floating_solids was made to remove floating solids from image. It is also used in trim_non_percolating_paths function.
I agree that they will currently output the same thing and it is because of a bug inside trim_floating_solids. So line im[holes] = True should be im[holes] = False. Which means also a small change in trim_non_percolating_paths. I will do it and push the changes if all agree.

[Zohaib-Atiq]

[Zohaib-Atiq]

Ignore my above comments. @mdrkok these two functions are behaving differently with the current implementation. They look the same but have a different meaning. The purple is the solid phase with zero's label.

[Zohaib-Atiq]

@jgostick While doing this test I also observed one thing. When we run these function the input image also changes.
For example when we do
im_fill_blind_pores = ps.filters.fill_blind_pores(im)
Check
plt.imshow(im)
plt.imshow(im_im_fill_blind_pores)
gives same image. So I think a copy of image should be used to mask it with holes and then we should return masked copy as output. like

def fill_blind_pores(im):
   holes = find_disconnected_voxels(im)
   new_im = im.copy()
   new_im[holes] = False
   return new_im

Not sure if I am the only one facing this issue.

[jgostick]

It seems the functions work ok, but the issue about not copying the input is a problem. I will open a separate issue about that.