Elliposid transform axis names

[tischi]

Can we have axis names for the elliposid transformed image such that we know what is what?
Then we could use those to also name them like this in the exported image.

[NicoKiaru]

Sure, but where and what would you modify ?

[NicoKiaru]

Can we put different names for different axis ?

[NicoKiaru]

We can only put one unit in ImagePlus's calibration, so not sure I can do anything here

[tischi]

Yes, I had the same thought.... sucks a bit.

But there is: new ImagePlus().getProperty(key)
What about adding keys: AxisNameX, AxisNameY, ... ?

Given that OME.ZARR will support also both: unit and name for each axis (@constantinpape). I think it would be important to also add support for this to our SourceAndConverter eco-system.

I actually think the elliposid-transformed images are a very good use case to develop this, because we now have the axes (correct me if I am wrong):

• relative_radius, no unit
• latitude, radian
• longitude, radian
• time, time point

@constantinpape, would we currently have something in OME.Zarr for "no unit" and just for "time point"?

[constantinpape]

@constantinpape, would we currently have something in OME.Zarr for "no unit" and just for "time point"?

Not sure what you mean by this, but you can currently give arbitrary values for the type, name and unit of the axes.
(There are SHOULD requirements in order to strongly suggest using the defined values in the spec, but for experimental things you can use what you need.)

[tischi]

There are SHOULD requirements in order to strongly suggest using the defined values in the spec, but for experimental things you can use what you need.

I think what I mean is whether something like "count" for time point would be available in the SHOULD requirements.