InsightSoftwareConsortium / ITK

Insight Toolkit (ITK) -- Official Repository. ITK builds on a proven, spatially-oriented architecture for processing, segmentation, and registration of scientific images in two, three, or more dimensions.
https://itk.org
Apache License 2.0
1.43k stars 664 forks source link

OME-TIFF metadata IO support in the TIFF ImageIO #3427

Open thewtex opened 2 years ago

thewtex commented 2 years ago

Description

Add OME-TIFF metadata IO to the ITK TIFF ImageIO. OME-TIFF is the most standard TIFF metadata format in adopted by microscopy community.

Expected behavior

OME-TIFF metadata is read and written by ITK's TIFF ImageIO. This in includes spatial metadata and additional metadata in the Image MetaDataDictionary.

Actual behavior

This metadata is not used.

Additional Information

Spatial metadata can be stored in PhysicalSizeX, PhysicalSizeY, and the per-Plane: PositionX, PositionY, PositionZ.

blowekamp commented 2 years ago

As I recall the OME-TIFF meta data is most frequently XML encoded in a TIFF TAG. Is this proposal to place that XML into the ITK's MetaData dictionary or further parse the meta-data information into ITK's intrinsic metadata (spacing, origin etc). Both would be useful, but looking for clarification for the goals of this issue.

thewtex commented 2 years ago

Yes, we should read / write the OME-XML tag. The spatial metadata (items noted in Additional Information) should go into itk::Image Origin and Spacing. The rest into MetaDataDictionary.

zhusihan-python commented 2 years ago

hello, is there any updates about the ome tiff support in itk 5.3.0

dzenanz commented 2 years ago

I don't think there has been any progress on this front.

zhusihan-python commented 2 years ago

I don't think there has been any progress on this front.

hi sir dzenanz, when is the ZarrNGFF will be ready, is it better than the ome tiff?

dzenanz commented 2 years ago

OME-TIFF support is not even started. ZarrNGFF is in early stages. Which one will be better? Probably the one which receives more attention. That seems to be ZarrNGFF.