It is not possible to save/load itk::seedsimage onto the harddisc. Since a seedsimage is pretty much the same as a normal itk::image it should not be too hard to make a filewriter for it.
The bundle "Graphcutsegmentation" is creating seedsimages into the dataset which can be used for testing.
I just made an easy fix ,that allows seedsImages to be saved like and image.
But the loading is a problem. Images are usually loaded as mitk::image. If you want them to be laded as mitk::seedsimage, there has to be information about that in the image file. But there is no possibility for that, unless you make a new image format specially for seedsimage.
So, my proposed workaround: Have the seedsimage as mitk::image in your dataset and cast it in your bundle to an mitk::seedsimage whenever necessary. I just try to establish that in the graphcut bundle.
Another solution: Establish a novel file ending "msi" for mitk seeds image. Its actually saved as PIC, but when it is loaded, its put into a mitk::seedsimage
Resetting all bugs without active assignee flag to "CONFIRMED". Change status to IN_PROGRESS if you are working on it.
Bug was discussed with Bastian. Seed images are a new feature for segmentation algorithm. Hence, we switch bug componente from ExtApp to Segmentation.
Bastian, are you still working on that bug? Is it still relevant or can it be closed?
This bug is nearly 3 years old. It doesn't look like there is realy a need for seedsimage io. Basti can you please check whether we can close this bug?
The Graphcut plugin was excluded due to Patent-Issues
(ITK also included a graphcut function, but excluded it again as well due to patents).
Thus we do not need mitk:Seedsimage anymore, i guess.