Gregory Lee

cuTENSOR 1.6: failure in math operations on some arrays with singleton dimensions

6

### Description I observed one test failure in cuCIM today when running the test suite using CuPy 11, but only when `CUPY_ACCELERATORS` contains "cutensor" . I have not yet gone...

scipy.ndimage.distance_transform_edt (Euclidean distance transform)

2

### Description We have recently proposed an implementation for `scipy.ndimage.distance_transform_edt` in: https://github.com/rapidsai/cucim/pull/318 I have started this there as we would like to use this to implement a few other cuCIM...

## Description closes #5337 I see no warnings from the IO module on my system, but it is possible some other versions of dependencies may still warn. Let's see how...

handling newly introduced deprecations (SKIP 4 clarification needed)

11

## Description We have some existing kwarg deprections that were introduced in 0.19 where the "removed_version" is listed as 1.0, but I think that was only correct assuming that were...

Add homomorphic filtering (with demo for removing variations in illumination)

1

## Description Please review #6251 first as the first commit here is overlapping with that PR. This PR adds a simple function to help perform homomorphic filtering and shows a...

add warning on non-integer image inputs to local_binary_pattern

4

## Description closes #2522 by adding the warning suggested by @stefanv (https://github.com/scikit-image/scikit-image/issues/2522#issuecomment-281206863) This function does end up converting the input to double precision internally, though so this seems slightly annoying...

Avoid redundant computations during central moments calculation

4

## Description The current calculation of central moments is inefficient in two ways: 1.) Raw first order moments are calculated using `moments_central` with ``center = (0, 0)`` and then then...

## Description **NOTE:** the first commit here is #6318, so this will be easier to review after that one is merged A subset of functions have a `preserve_range` keyword argument...

Make boundary mode names consistent and improve their documentation

4

## Description This PR makes edge modes consistent as discussed previously in https://github.com/scikit-image/scikit-image/issues/5439#issuecomment-903344391. closes #6345: cross-links to documentation illustrating the edge modes I went with the `scipy.ndimage` mode names because...

rename img_as_* to rescale_to_*

5

## Description This PR implements a prior suggestion to rename `img_as_float` to `rescale_to_float`, etc. This is done primarily to make it clearer that these function rescale the range of the...