Open haesleinhuepf opened 3 years ago
Hey @pr4deepr ,
have you tried https://clij.github.io/clij2-docs/reference_seededWatershed
?
Thanks!
I just tried it.
The labelling wasn't optimised as much, but this demonstrates the differences.
Awesome! I can see the issue now. Does it also fail in 2D? Can you share the dataset and the script which generates both results? That would make it easy for me to trace it down in more detail.
Suggested by @pr4deepr
''' Marker controlled watershed from MorpholibJ If I am using CLIJx to generate seeds (labelling) and binary images (mask), then if I have edges image I can use these images for segmenting cells using Marker Controlled Watershed. '''
See also: https://github.com/clij/assistant/issues/14