Jupiter colour balance help please

[runoffshed]

Looking for some help please.  I am using the ASI120mc with Firecapture and, as seems to be recommended, getting Autostakkert to do the Bayering.  Whether I set the Bayering option to automatically detect or use the Force GRBG all my images/stacks come out with a strong purple cast.  Using RGB correct in Registax doesn't produce a great result and neither do my attempts at manually trying to balance the colours.  I have looked through the SL posts and can't find anything about this specific problem although there are some posts about setting the camera white balance.  I haven't changed this in Firecapture so wonder if this is the issue or whether I have accidentally changed a setting in Autostakkert?   If anyone has any ideas for me to try I would be grateful.  (Getting fed up with capturing the Great Purple Spot)  Many thanks

 

[Stargazer33]

Are you shooting in RAW mode or RGB24? You could also just try all the other Bayer patterns and see if one comes out looking like other peoples images.

[runoffshed]

Thanks for reply Bryan.  I think I'm shooting in Raw in that I simply set it to record as an AVI and turn Debayering off and it says 'Recording in raw and preview in colour' or something like that.  Using the other Bayer patterns in Autostakkert doesn't improve the output either.  I'm sure I've changed a setting or other in some bit of software as the colour preview image on the screen when recording shows a much better colour balance.  Thanks for your input though - very much appreciated.

[Stub Mandrel]

Try RGGB.

[runoffshed]

Thanks Neil but unfortunately that makes my purple cast turn green.

This is probably a really dim question so apologies in advance but if recording in raw do I need to change any other Firecapture setting with regard to the Debayering/Bayering or is this simply recorded in the raw output and not user changeable?  Thanks again.

 

[Ben the Ignorant]

Are you using your 102mm Vixen lens or your 300mm dob? Given the resolution I'd say the 102, but is it an achro or an apo? Because if it's an achro the purple cast is unavoidable unless you filter it out with a Baader semi-apo or a regular pale yellow filter.

[Cosmic Geoff]

I have the same camera, the ASI120MC and I had to debayer recorded .avi files in PIPP using a setting of GBRG.

I now record in .ser format which avoids the bother of debayering.

I use Sharpcap, the program recommended by ZWO, for capturing image video.

If you are still stuck, you could put the fisheye lens on the camera, record a short colour video in daylight, and see which debayer option debayers it correctly.

[runoffshed]

22 minutes ago, Ben the Ignorant said:

Are you using your 102mm Vixen lens or your 300mm dob? Given the resolution I'd say the 102, but is it an achro or an apo? Because if it's an achro the purple cast is unavoidable unless you filter it out with a Baader semi-apo or a regular pale yellow filter.

Thanks Ben, yes the 102mm and it's an achro.

Funnily enough I wondered if that was an issue but had it in my head that I'd captured better images of Jupiter in the past in 'glorious colour' -  but I was wrong.  On checking images from last year, it's true that I had captured the red spot as red but you are quite right that as even that had a slight purple colour cast in the bands.  So I think you are right and I will investigate the filters you mention.  ( I am guessing the pale yellow will be cheaper than the Badder semi apo which sounds expensive - update, just checked £53 from FLO!  Sounds like the way to go)  Anyway, many thanks Ben and I will try that first anyway.

[runoffshed]

4 minutes ago, Cosmic Geoff said:

I have the same camera, the ASI120MC and I had to debayer recorded .avi files in PIPP using a setting of GBRG.

I now record in .ser format which avoids the bother of debayering.

I use Sharpcap, the program recommended by ZWO, for capturing image video.

If you are still stuck, you could put the fisheye lens on the camera, record a short colour video in daylight, and see which debayer option debayers it correctly.

Thanks Geoff.  I used to use Sharpcap but have got used to Firecapture now.  However, I use Autostakkert to Bayer (if that's the phrase).  Great idea about the daylight colour video though.  Never thought of that!  Many thanks.

[Ben the Ignorant]

Same scene through my 80mm f/7.5 achro…

...and my 80mm f/7 apo:

[runoffshed]

Many thanks - pretty conclusive.  Interestingly I think my camera enhances this colour cast as, to my eye, it appears less pronounced with visual observation...or is that wishful thinking?

Well I think the semi apo or yellow filter is the way to go and thanks to you and everyone for their input...really interesting and helpful. Many thanks all.

[Stub Mandrel]

I think you have the channels mapped incorrectly. I changed them over in photo paint, which made the image very magenta, then did an auto-equalise. Clearly this looks a bit naff but tits close than the magenta version :

 

[runoffshed]

11 minutes ago, Stub Mandrel said:

I think you have the channels mapped incorrectly. I changed them over in photo paint, which made the image very magenta, then did an auto-equalise. Clearly this looks a bit naff but tits close than the magenta version :

 

You may well be right Neil - and thank you for your help  - but all I can say is that I think I have the force bayer setting in AS2 correct at GRBG and that even if I let AS2 do an Automatic Detection instead the result is the same.  Your result is definitely better though but I'm wondering if you're just much better at processing than I am.  If I knew how do process properly I have often thought that I should be able to sort this out at the processing stage.

[cuivenion]

Hi, If you're capturing in RAW then your captures should have a green bias once you've debayered them, and this can be easily colour corrected with the RGB Balance tool in Registax. The green colour bias is there because there are more green pixels than red or blue in the bayer matrix.

The bayer pattern for your camera is GRBG from a quick look on google, but in a post above you mentioned RGGB was giving you a green bias. I'd use that and then try the correction in Registax.

Some software might not detect the bayer matrix correctly and some capture software might output files that have the pattern reversed. I know this happens with FITS files and I've seen autostakkert detect the wrong pattern before too.

[runoffshed]

2 hours ago, cuivenion said:

Hi, If you're capturing in RAW then your captures should have a green bias once you've debayered them, and this can be easily colour corrected with the RGB Balance tool in Registax. The green colour bias is there because there are more green pixels than red or blue in the bayer matrix.

The bayer pattern for your camera is GRBG from a quick look on google, but in a post above you mentioned RGGB was giving you a green bias. I'd use that and then try the correction in Registax.

Some software might not detect the bayer matrix correctly and some capture software might output files that have the pattern reversed. I know this happens with FITS files and I've seen autostakkert detect the wrong pattern before too.

Many thanks for that.  This is the result of the RGGB but I can't seem to correct/remove the green cast in Registax either using autobalance or manual.  When I try to reduce the green the image just goes back to a purple hue.  No idea what I'm doing wrong with the balance but I appreciate all the help.

 

[cuivenion]

Weird, could you upload the original capture file?

[runoffshed]

28 minutes ago, cuivenion said:

Weird, could you upload the original capture file?

Thank you but do you mean the Firecapture log file or the stacked but unprocessed AS2 file? ( Can't upload the AVI as it's 1gb +). Thanks again.

[cuivenion]

I did mean the avi, but yeah that'd take a while. If you could upload an unprocessed as2 file from RGGB and GRBG patterns I'll have a look.

[runoffshed]

15 minutes ago, cuivenion said:

I did mean the avi, but yeah that'd take a while. If you could upload an unprocessed as2 file from RGGB and GRBG patterns I'll have a look.

Many thanks indeed...here they are:

Jup_231728_g4_ap32.tif

Jup_231728_g4_ap30.tif

I presume as they are Tif files they don't show but hopefully they will download etc...I'm just wondering if, as suggested, it is simply a case of my scope being an achromat and also the fact that Jupiter is so low this year?  (I have never found the colour fringing objectionable visually but maybe it is enhanced by the camera)  Thanks again for your help.

[cuivenion]

I've had a quick look at the files. Are you using a barlow lens? I remember now a few years back I was using a very bad quality barlow lens that gave gave me a similer problem to what you're having. You also need an IR cut filter when imaging with any type of lens.

 

[runoffshed]

8 minutes ago, cuivenion said:

I've had a quick look at the files. Are you using a barlow lens? I remember now a few years back I was using a very bad quality barlow lens that gave gave me a similer problem to what you're having. You also need an IR cut filter when imaging with any type of lens.

 

Yes to barlow, but a pretty good one - Celestron Ultima 2x.  With a flip mirror between this and the camera/eyepiece there is also a spacing factor which increases the apparent focal length..straight through for the camera, of course.  I'm pretty sure the asi 120 has an ir filter...the window over the chip.

Thanks.

[cuivenion]

This is the result I was getting with the cheap barlow:

[cuivenion]

I'm a bit baffled I'm afraid. At the end of the day if you try each of the debayer options one of them should be right and give you a workable image, but that doesn't seem to be working. I'd suggest trying to image with your dob just to compare, maybe it's an issue with your scope. The similarities between the result I got with the downloaded file and my image makes me think its a lens issue.

[runoffshed]

10 minutes ago, cuivenion said:

I'm a bit baffled I'm afraid. At the end of the day if you try each of the debayer options one of them should be right and give you a workable image, but that doesn't seem to be working. I'd suggest trying to image with your dob just to compare, maybe it's an issue with your scope. The similarities between the result I got with the downloaded file and my image makes me think its a lens issue.

No, that's great and I really appreciate your input. I will certainly try the same set up with the Newtonian and have also just purchased a #8 yellow to try as well. Thanks again for all your help.

[cuivenion]

No probs, I'd try it without the yellow filter first. Just so you get a proper comparison. You shouldn't need a yellow filter with a newtonian anyway, no chromatic aberation.