Skip to content

Commit

Permalink
Add support for bia-bob
Browse files Browse the repository at this point in the history
  • Loading branch information
haesleinhuepf committed Oct 10, 2024
1 parent 92dbd00 commit a993f4b
Show file tree
Hide file tree
Showing 2 changed files with 57 additions and 0 deletions.
55 changes: 55 additions & 0 deletions napari_segment_blobs_and_things_with_membranes/_bia_bob_plugins.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,55 @@
def list_bia_bob_plugins():
"""List of function hints for bia_bob"""
return """ ## napari-segment-blobs-and-things-with-membranes (nsbatwm)
nsbatwm is a Python library that processes images, mostly using the scikit-image library, but with simpler access.
When you use it, you always start by importing the library: `import napari_segment_blobs_and_things_with_membranes as nsbatwm`.
When asked for how to use nsbatwm, you can adapt one of the following code snippets:
* Splits touching objects in a binary image using an algorithm similar to the ImageJ watershed.
nsbatwm.split_touching_objects(binary_image)
* Applies Otsu's method to binarize an intensity image (also works with yen, isodata, li, mean, minimum, triangle instead of otsu).
nsbatwm.threshold_otsu(image)
* Labels connected components in a binary image.
nsbatwm.connected_component_labeling(binary_image)
* Applies seeded watershed segmentation using labeled objects, e.g. nuclei, and an image showing bright borders between objects such as cell membranes.
nsbatwm.seeded_watershed(image, labeled_objects)
* Segments blob-like structures using Voronoi-Otsu labeling.
nsbatwm.voronoi_otsu_labeling(image, spot_sigma=4, outline_sigma=1)
* Applies a Gaussian blur for noise reduction.
nsbatwm.gaussian_blur(image, sigma=5)
* Applies median filter to reduce noise while preserving edges.
nsbatwm.median_filter(image, radius=5)
* Smooth a label image using a local most popular intensity (mode) filter.
nsbatwm.mode_filter(labels)
* Applies a percentile filter.
nsbatwm.percentile_filter(image)
* Removes background in an image using the top-hat filter.
nsbatwm.white_tophat(image)
* Applies local minimum filtering to an image (also works with maximum, and mean).
nsbatwm.minimum_filter(image)
* Subtracts background in an image using the rolling ball algorithm.
nsbatwm.subtract_background(image)
* Removes labeled objects touching image borders.
nsbatwm.remove_labels_on_edges(label_image)
* Expands labels by a specified distance.
nsbatwm.expand_labels(label_image, distance=2)
* Segments using seeded watershed with local minima as seeds.
nsbatwm.local_minima_seeded_watershed(image, spot_sigma=10, outline_sigma=2)
* Skeletonizes labeled objects.
nsbatwm.skeletonize(image)
"""
2 changes: 2 additions & 0 deletions setup.cfg
Original file line number Diff line number Diff line change
Expand Up @@ -45,3 +45,5 @@ install_requires =
[options.entry_points]
napari.plugin =
napari-segment-blobs-and-things-with-membranes = napari_segment_blobs_and_things_with_membranes
bia_bob_plugins =
nsbatwm_plugin = napari_segment_blobs_and_things_with_membranes._bia_bob_plugins:list_bia_bob_plugins

0 comments on commit a993f4b

Please sign in to comment.