bayer pattern FITS header

Discussions, Bug Reports and Issues related to Beta versions of SharpCap
Post Reply
maclean
Posts: 10
Joined: Sat Dec 19, 2020 7:59 pm

bayer pattern FITS header

#1

Post by maclean »

hi Robin,
I noticed you had changed the FITS header fields BAYERPAT and COLORTYP from GBRG to RGGB in the 3.3 beta. I saw some of your earlier posts about there not really being a universal interpretation of ordering (viewtopic.php?t=3125). Astro Pixel Processor seems to like the old GBRG designation by default - while it's not a problem to force the correct pattern in APP when loading files (there's a combo box for specifying it), I thought perhaps you might be willing to put an option somewhere in the SharpCap Settings dialog to set the header pattern so that a user like me can set it to whatever format matches our workflow and have it remembered?

Other than that, 3.3 is working well for me so far. The scripting sequencer seems quite promising. I had been trying to figure out how to have PHD2 dither during capture sequences, and that's a really nice thing to have available now.

Matthew MacLean
NY (USA)
User avatar
admin
Site Admin
Posts: 13177
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: bayer pattern FITS header

#2

Post by admin »

Hi Matthew,

thanks for the suggestion – I will try to make that update at some point in the near future. Actually I'm a bit surprised that any software actually reads the header – when I last tested this a year or two back I couldn't find any processing software that actually took any notice of those pattern headers!!

There's a new update to 3.3 available this evening with improvements to the sequencer - more to follow when I get a moment to post info about it :)

cheers,

Robin
maclean
Posts: 10
Joined: Sat Dec 19, 2020 7:59 pm

Re: bayer pattern FITS header

#3

Post by maclean »

Thanks so much; it saves a selection when post-processing files. I assume most other users have their own process, but they probably do the same thing every time so just saving whatever setting works for each person seems flexible enough.

Yes, if you are interested, Astro Pixel Processor seems to read and utilize those fields (I'm not clear which one is the trigger). There is a combo box to force it to use a certain pattern for import, but the default is to try to detect it from the header data. Curiously, the active file table in APP lists a CFA pattern field for each images that it has imported which is the reverse of the pattern that you are specifying in your file header, so I imagine APP inverts the data orientation from SharpCap, but it seems to handle that all correctly from what you are exporting. I just did a quick image in my living room and V3.2 definitely imports correct colors and orientation, V3.3 imports with inverted coloring when using the automatic option.
drblevy13
Posts: 4
Joined: Mon May 10, 2021 3:58 pm

Re: bayer pattern FITS header

#4

Post by drblevy13 »

Hi Robin,

Just wanted to follow up on this. I too am still running into problems with FITS header and Astro Pixel Processor when trying to stack raw images. I suspect it is because of the way the Bayer pattern is being reported. When I force Astro Pixel Processor to recognize as RGGB, all is well. If not, my final image is mostly gray with a green tint.

I am running latest SharpCap 4 beta with ZWO ASI 2600 MC Pro one shot color camera.

When I compare the header of the FITS from the beta to a header taken from the latest SharpCap 3 iteration, the beta has a "0" next to the Bayer Y parameter whereas in SharpCap 3, the Bayer Y is "1". Only the beta has the issue with colors being off when stacking in Astro Pixel Processor.

Thanks,
Ben
User avatar
admin
Site Admin
Posts: 13177
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: bayer pattern FITS header

#5

Post by admin »

Hi,

thanks Ben, it looks like I am going to have to add an option before 4.0 releases to allow invert of the bayer settings. It's good to know that APP is currently 'wrong'.

cheers,

Robin
Post Reply