Morphological segmentation - save processing and merging regions

Hi all,

I am trying to use the Morphological segmentation plugin for my CT analysis. It looks to segment fine, but I need to merge some parts of the samples (it is beam structures with hollow parts). Since I am working with several images (I am testing with around 25 images, but my whole CT scan corresponds to 1000 images), I am not able to do the whole merging in the morphological segmentation at once. So, I have some questions I could not find a solution:

  1. Is it possible to save the morphological segmentation processing (or post-processing) to continue with the merging in another moment? I saved as images, but then I was not able to continue the merging in the plugin…

  2. I also tried to use the Label Edition plugin. After I do the segmentation, I am not able to open the saved segmented file, because it is in RGB. So I opened the Label Edition direct from the morphological segmentation (I understood I was using the image that was segmented), and the label edition opens an 8-bits image, I did the merging, however it didn´t save the merging in the morphological segmentation…

Then I tried to do the merging in the input sample, and after the merging, running the morphological segmentation, also not successful… Am I using the Label Edition wrongly?

Thanks for any help,



I believe @iarganda is the best person to help you with this question…

1 Like

Hello @rafadebas and @etadobson,

I would say the Label Edition plugin is the way to go, but you have to save the result of the Morphological Segmentation plugin by 1) using the catchment basins display option, 2) clicking on Create Image to generate a separate image, and 3) finally save that image into a file.

The result image should be 32-bit, not RGB, in order to be an actual “label image”.

I hope this helps!


Hi @iarganda and @etadobson,

Thanks, this helped!

Is there any way to go back to overlaid basins after using the label edition? Because with label edition I was not able to use the 3d viewer… when I open it, it asks to convert the image to 8-bits or RGB, and doing this, I am not able to actually remove the background and see only the sample. Is there any other way to remove the background and see the sample in 3d? (Using threshold segmentation I was able to do it, but the segmentation was not very good).


Why would you do that? Once you have your labels correct, you can use them to analyze your original image without mixing both up. Overlaying is just for visualization.

If you want to visualize your label image in the 3D viewer, you need indeed to convert it to 8-bit or RGB. For that, I would suggest you use the Plugins > MorphoLibJ > Label Images > Labels To RGB command.


Thanks! Labeling to RGB worked! :grinning:
I wanted to go back to overlaid basins just to visualize in the 3D viewer…