Author Topic: Plate solving: Attaching WCS data to FITS Image?  (Read 70 times)

34South

  • Newbie
  • *
  • Posts: 5
Plate solving: Attaching WCS data to FITS Image?
« on: May 18, 2020, 07:25:35 AM »
Firstly, I am totally new to Observatory, but love it! I had developed something myself using FileMaker for saving observing sessions and images but having something which is a one stop shop and includes stacking and overlays is absolutely fantastic! Regarding my question, I did check the forum and discovered that one way of overcoming plate solving errors (it's mostly on those images where a nebula or globular cluster covers most of the image) was to upload the image to astronomy.net and get the solved FITS image and to obtain the WCS data. I did this successfully and the image now has a pin icon, confirming its position has been registered within Observatory. However, the overlays, including the grid, do not appear if I turn them on. I imported the WCS file as an attachment to the FITS image, but perhaps this is not how it's done.

Sander Berents

  • Global Moderator
  • Jr. Member
  • *****
  • Posts: 84
    • Code Obsession
Re: Plate solving: Attaching WCS data to FITS Image?
« Reply #1 on: May 18, 2020, 09:28:16 PM »
Glad to hear you like it! Did you also try to use "Blind Match" with the "Upload whole image" option enabled? There shouldn't really be a need to manually upload the image to astronomy.net and then manually download the plate solved one.

Observatory displays the pin icon when it has all information it needs for displaying the overlays. If it then doesn't display any, that means that when it does the actual transformations it rejects it. Maybe there's something wrong with the FITS header, or it contains a transformation Observatory doesn't support. If you see this with more of your images, then please share one with me using dropbox so I can investigate.

Attaching a file to an image does not add its WCS info to it. It's mainly intended for quickly accessing external notes or articles related to an image or album. Observatory doesn't have a way to manually merge WCS info with an image. The only way to do that is automatic, through the "Blind Match" feature.
Sander Berents
Code Obsession, LLC
http://codeobsession.com

34South

  • Newbie
  • *
  • Posts: 5
Re: Plate solving: Attaching WCS data to FITS Image?
« Reply #2 on: May 19, 2020, 08:19:50 AM »
Dear Sander

Yes, I did try the blind match with the upload option. I set the timeout to 10 minutes but it reports an error. When I upload the same image directly to Astrometry.net, it has success in under a minute. I attach the link to the solved FITS image and my original PNG image.
https://www.dropbox.com/sh/v6x8cha3wyw6906/AABexFlVQlsbYt3MhO1ItALva?dl=0

Regards, Paul

Sander Berents

  • Global Moderator
  • Jr. Member
  • *****
  • Posts: 84
    • Code Obsession
Re: Plate solving: Attaching WCS data to FITS Image?
« Reply #3 on: May 19, 2020, 10:19:54 AM »
Hi Paul,

This is strange. The FITS WCS information at first glance looks good, but indeed overlays don't get displayed. But, I have no trouble solving your PNG using "Blind Match" at all, with Observatory only sending star positions to astrometry.net. Its overlays then display just fine. I'll investigate what in the FITS file may be the cause of its trouble, but I would suggest to try "Blind Match" again. The API of astrometry.net that is used by Observatory sometimes does not behave normally, and waiting a few hours before trying again seems to help.
Sander Berents
Code Obsession, LLC
http://codeobsession.com

34South

  • Newbie
  • *
  • Posts: 5
Re: Plate solving: Attaching WCS data to FITS Image?
« Reply #4 on: May 19, 2020, 02:57:52 PM »
Dear Sander

Some of my images that previously failed to solve have now done so using the blind option (with positions), so it strangely appears to be related to the timing of when one attempts the upload. I have quite good bandwidth, so that can't be it but, as you say, it could be the API playing up.

I have shared my enjoyment of Observatory with our local astronomy club, although I am one of the few Mac users for reasons that escape me! :D

Thanks for your help!