Gregory Lee

Yeah, I don't remember closing this. Thanks for reopening

In terms of the numerical `distance` returned by the `return_distance` option for `medial_axis`, it should just be a matter of adding a `spacing=None` kwarg to the API and then passing...

Thanks @moi90! I haven't gone through the publication yet, but the implementation here looks simple and low-maintenance. In terms of adding a citation, I would just add a `Notes` section...

I am also interested in the result if you do any timings of this vs. binary erosion with footprint decomposition. There is substantial benefit to footprint decomposition relative to not...

When citing, please use this form of the reference rather than the researchgate link: ``` References ---------- .. [1] Cuisenaire, O. and Macq, B., "Fast Euclidean morphological operators using local...

Hi @xactbupt, the image I/O operations will get sent to an underlying package such as `imageio`, `tifffile` or `pillow` so it is likely the difference in one of these that...

Thanks, I was planning to open a PR soon, porting the same changes I did for `PyWavelets` here. Aside from the small .yaml file updates, we have to update the...

> BTW, I don't really like the aliases defined in skimage._shared.testing for numpy.testing and pytest functions. These functions are somehow standard, using these aliases can confuse contributors. But this is...

I am fine with using the `assert_*` from `numpy.testing` instead of `skimage._shared.testing` too, but am reluctant to make that change immediately because it will introduce a ton of conflicts with...

Another motivation for adding this argument is my proposed structuring element decomposition feature (see #5388)