TIFF files do not contain all the metadata that FITS contain

[Shlomi Vaknin]

I wanted to save my files in tiffs instead of fits, and it seems that the tiff files do not store the metadata that is stored with fits files (picture parameters, camera, weather, etc).

Any plans of this to be added?

[Shlomi Vaknin]

Seems like we need to add all the goodness found here https://bitbucket.org/Isbeorn/nina/src/61d9b3f48c29ac541996627ed1b2c1100c33ac6d/NINA/Utility/FITS.cs?at=master#FITS.cs-76:79,84:85,88:90,94:95,97:98,101:102,105:106,109:110,113:115,118:119,122:123,126:127,131:132

to the tiff encoder here https://bitbucket.org/Isbeorn/nina/src/61d9b3f48c29ac541996627ed1b2c1100c33ac6d/NINA/Model/ImageData/ImageData.cs?at=master#lines-234:250

using the metadata setters, as described in https://docs.microsoft.com/en-us/dotnet/api/system.windows.media.imaging.bitmapencoder.metadata?view=netframework-4.8#System_Windows_Media_Imaging_BitmapEncoder_Metadata

Unfortunately I don’t have windows dev machine, so I can’t patch this myself..

[Dale Ghent]

Can you point to any other astronomy software that recognizes and utilizes FITS-like keyword metadata in TIFF files?

[Shlomi Vaknin]

Hi Dale, thanks for responding! I hope you are not getting discouraged by me opening all these tickets. I am really enjoying your software and only thinking about ways to make it even better!

To your question, it seems that AstroPixelProcessor is attempting to utilize these fields as one example (that is how I found out).

Honestly though, I am not entirely getting the purpose of the questions: All the information is already there, the mechanism to put it in the image is already there, it seems like one of the lowest hanging fruits in this project that will make the different formats act in a standard way.

If tiff is a choosable option, it should be expected that some users will choose to use it if it fits their workflow (no pun intended!). Why should these users, not only get a degraded experience by using a perfectly valid choice, but actually lose information that may be really useful to them in an unrecoverable way, for no real technical reason? (at least as far as I understand)

I think all formats, as long as it is technically possible and does not introduce unneeded complexity to the code, should provide the same standard experience and be interchangeable.

Does this makes sense?

[Dale Ghent]

TIFF is provided as an option in NINA to serve as a sort of “format of last resort.” It’s for those who do not utilize a traditional astrophoto processing toolchain where FITS is already the lingua franca, such as someone who goes directly from camera to Photoshop and/or Lightroom. This metadata isn’t understood there unless the user also makes their own tooling to make use sense of it… and if they are doing something that requires the ability to grok this information, then their purposes would be probably be better served using a FITS-oriented toolchain anyway.

Currently there is no discernible or published standard for embedding FITS-like keywords into TIFF images, where there are entities who maintain such standards for FITS and XISF. I’m hesitant to put effort into this in the absence of such standards and if there is nothing that will actually utilize the information in a meaningful way. Is there a reason why you cannot use FITS files in APP?

[Shlomi Vaknin]

No, the reasons that I’d like to use TIFF format is not related to an astro program, I was just giving it as an example of an astro program that looks for these metadata items in the tiff. I see your point about the lack of standard, but in my opinion there is nothing to lose by including this metadata in the tiff file, and I there is some to loose for those who may need it by not having the metadata present. I can’t see how having this information in the metadata can cause problems, on the flip side, if it is retained it will be valuable to whoever does choose to have a non-standard workflow.

[Stefan B]

When headers are required, the FITS and XISF standard provide a much better way to store them and astronomy related image readers can process these keywords in a standardized pattern. No need to invent something for TIFF.