Jump to content

Narrowband

Recommended Posts

Hi all

i hope you can help out?

I have a ZWO ASI585MC that I have been using for planetary imaging and more recently turned it to DSO AP until I get a ASI533. The issue I have been having, as the title suggests are that it saves the Fits files with incorrect colours. I have been using SharpCap with great success especially at LiveStacking and saves correctly, however as soon as I open the raw fits files they are all blue. I use Siril to force change the bayer pattern in its settings but that either comes out blue or green depending on the pattern selected. It seems that it saves as BGGR not RGGB. I understand the fits reads the bayer pattern back to front so to speak but can’t work out how to make it see correctly or at least save correctly to start with? 

I have tried different settings in SharpCap and I’m at a loss as to what I am either missing or simply not doing. I turn to you for help and advice. Please. 
 

Thanks. Jon

Edited by JonHigh
Link to comment
Share on other sites

Thank you for responding. Yup it was set to “on”.  Not sure I ever touched it from the default setting. I take it that forcing a Bayer pattern might not be a solution then as I was going to try it out next clear skies. I have even checked the setting in SharpCap Preferences as it can be changed to inverted for PIPP etc but it was configured correctly for DSS etc. 

Link to comment
Share on other sites

Hi Jon.

@alacant needs the raw. unprocessed .FITS image straight from capture save in SharpCap to be able to offer a Debayer solution for Siril.

If you can't attach a .FITS format image directly to your reply use a free file share host such as GoogleDrive, DropBox or OneDrive to host the .FITS file and post a public share link back with your reply.

FWIW, I believe there is a Debayer bug in the current build 1.0.6 of Siril where if you have Siril preferences configured to use both the .FITS header Bayer mosaic pattern manually configured and also the option to use "Bayer information from FITS header if available" then Siril gets confused when the manually configured and auto detect patterns differ.

For SharpCap single frame grabs using an old ZWO ASI120MC for solar Ha imaging I have the following FITS/SER Debayer preferences settings in Siril:

"Bayer information from FITS header if available" = Deselected.

"Bayer mosaic pattern" = GRBG (for an old ZWO ASI120MC).

"Debayer interpolation" = VNG4 (for stars, solar and planetary, or select method RCD for star images)

"Debayer FITS in top-down if no explicit keywords found" = Selected.

Note: the Bayer mosaic pattern selected by Siril uses the specific top-down or bottom-up readout order FITS header keyword value provided by the camera driver (ZWO). If that keyword value in the FITS header is incorrect then the manual choice of Debayer pattern will need to be inverted.

Remember that the preview image open in Siril is not the latest version of the Debayered image, after running the process to Debayer a file with a particular bayer pattern selected you have to go back to the file menu and open that new image that was saved by the process, otherwise you will not see any changes.

HTH.

Edited by Oddsocks
Auto dictation spelling corrections were needed
Link to comment
Share on other sites

Hi Oddsocks. Thank for your detailed help. You me was very much appreciated. 
It was a bit late last night but I followed your procedure regardless. Unfortunately that didn’t work for me. It just came out green! 
The one thing I had noticed in my setting was that I had previously image flipped both axis in the SharpCap settings and wondered if that flips the bayer pattern as well? 
I might check that out next time out? 
Any thoughts?

Link to comment
Share on other sites

Hi Jon.

I’m afraid it’s impossible to guess without having access to the original .FITS image, many forum members here use SiriL and could most likely quickly determine the correct deBayer and post display settings in SiriL for you, or at least have a suggestion as to where the problem originates.

Image flip in SharpCap should not affect the raw data, no astrophotography capture program should ever modify the raw data in a destructive manner, if they do there’re not worth using.

A green cast across the entire image (both background and objects) is not unusual after DeBayer for a colour camera since there are twice as many green pixels as red and blue and the green imbalance is normally fixed in post processing, colour calibration, but whether this is the case with your image is just conjecture at this point.

Link to comment
Share on other sites

45 minutes ago, JonHigh said:

I had previously image flipped both axis in the SharpCap settings and wondered if that flips the bayer pattern as well? 

You will likely find it has, it's very likely down to the different ways different software read the Bayer array sequence. I had that issue when I'd used SCPro on the Moon and inverted the image forgetting to return it back the next time I did DSO work. I was surprised many others hadn't fallen foul of that little problem. Best to rotate as a final image after processing. Good luck.

Cheers,

Steve

Edited by SteveNickolls
Link to comment
Share on other sites

Okay. I have linked to one of the original files on Google drive for you to take a look. Again, very much appreciated! 
I will definitely check the image flip back to off so I can at least rule that out. Hope that it is as simple as that. Thanks for all your help so far. Jon
 

Link to comment
Share on other sites

Hi Jon.

Ok, the image you linked has very little signal but there was just enough to tease out the Rosette structure.

The .FITS header has these keys and values for the bayer pattern, derived from the camera manufacturers driver:

image.png.52f31ec0135b352ba6f4ba894e89a810.png

(Note: I read this .FITS header in PixInsight as I could not remember where the .FITS header viewer was in SiriL).

The default pattern is BGGR (BAYERPAT) and the manufacturer adds the note "Try GRBG if image upside down or R/B swapped".

The Bayer pattern that was used in SiriL Preferences to open the image was the default BGGR, as shown below, you will see that as expected the image is green because as explained earlier this is raw data from the camera and it has twice as many green pixels as red and blue:

image.thumb.png.3b3ac0f6cfb7d8fc7112337aae2a76b6.png

After running Colour Calibration, Background Neutralisation and White Balance adjustment followed by a Histogram Transformation you can just begin to see the Rosette is beginning to emerge from the background noise with the correct hue:

image.thumb.png.8a436b63cc1cc3f74239e10d00553c77.png

Above the preview image are four tabs, Red, Green, Blue and RGB, if you click on the Red tab you will see only the information in the Red channel and in this image the only nebula structure that appears is in the red channel, the blue and green channels have only noise and a few stars, this is a double check that the Bayer pattern used, BGGR was the correct choice. 

When you import a stacked image the noise will be much reduced, the signal stronger and the colours easier to separate.

For future reference, when opening a raw .FITS image from this camera in SiriL the image should appear green when "AutoStretch" is applied and before you begin with colour calibration, if the image appears magenta then the image read sequence is reversed and the Bayer pattern selected in SiriL Preferences should be GRBG.

HTH

William.

 

  • Thanks 1
Link to comment
Share on other sites

The FITS header is showing as BGGR in Siril. You can access it under Image info which is available when you click the 3 bars

image.png.8d7632a10698be0de6f48e95441fb115.png

Once you do a background extraction, the green cast is removed. Doing a Colour calibration should help get rest of colours right too (already demonstrated by Oddsocks 🙂  )

 

Edited by AstroMuni
  • Thanks 1
Link to comment
Share on other sites

2 hours ago, Oddsocks said:

The default pattern is BGGR (BAYERPAT) and the manufacturer adds the note "Try GRBG if image upside down or R/B swapped".

Indeed...

Strange, since ASI list the Bayer on a 585 as RGGB.

Maybe something happened after the image left the camera but before being written to storage🤨

Edited by alacant
Link to comment
Share on other sites

Good evening. Wow, thank you so much for all your help. I am so overhemingly grateful. Nice to know there are kind and helpful people out there!

I have processed as you said and although noisy, at least the colours are now correct out of Siril.

Many thanks again, Jon

 

 

 

 

New NGC2244.jpg

Edited by JonHigh
  • Like 2
Link to comment
Share on other sites

15 hours ago, alacant said:

Strange, since ASI list the Bayer on a 585 as RGGB.

Maybe something happened after the image left the camera but before being written to storage

I guess a simple way to check is capture an image using a different software (say) ASICap and see how it records the FITS header

Link to comment
Share on other sites

  • 2 weeks later...

Thank you the suggestion. I have noticed the bayer pattern is fine in APT so it must be SharpCap. 
I’ll have another look at the settings again. There must be something I’ve missed. I prefer and like SharpCap and would like to keep using it but it should be reading the pattern correctly?!. However, now I have sorted my stacking woes with all your help, I’m a happy customer again!

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.