Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

How to fix the weird shadow on Jupiter?

  • Please log in to reply
11 replies to this topic

#1 controlpower29

controlpower29

    Sputnik

  • -----
  • topic starter
  • Posts: 42
  • Joined: 21 Sep 2020
  • Loc: Chicago, IL, USA

Posted 29 October 2020 - 03:05 AM

This is Jupiter with Io's shadow. I did the normal processing with AutoStarkkert stacking and RegiStax sharpening. I don't quite understand how does red color around the shadow form. Is there a way to avoid it in one of the processing steps or any way to fix it after processing?

20 22 20 lapl5 ap35

 



#2 happylimpet

happylimpet

    Fly Me to the Moon

  • *****
  • Posts: 5,210
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 29 October 2020 - 03:54 AM

Your R ,G and B channels need alignment. You can do this in photoshop, gimp, registax, etc. Notice the whole jupiter image suffers from red at the bottom and blue at the top. The Io shadow is just the same thing.



#3 Daniele

Daniele

    Vostok 1

  • -----
  • Posts: 102
  • Joined: 14 May 2020
  • Loc: Tuscany, Italy.

Posted 29 October 2020 - 03:58 AM

Buy the ZWO ADC 


  • happylimpet and dcaponeii like this

#4 kisstek

kisstek

    Viking 1

  • *****
  • Posts: 642
  • Joined: 25 Jul 2018

Posted 29 October 2020 - 10:52 AM

Your R ,G and B channels need alignment. You can do this in photoshop, gimp, registax, etc. Notice the whole jupiter image suffers from red at the bottom and blue at the top. The Io shadow is just the same thing.

As you say, the red fringe around Jupiter is on the bottom. But the red fringe on IO's shadow is on top. If you move the red channel up to correct for the planet, you'll make the red fringe on IO's shadow worse.

 

Controlpower29 doesn't say whether he's using a OSC or a mono camera.



#5 happylimpet

happylimpet

    Fly Me to the Moon

  • *****
  • Posts: 5,210
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 29 October 2020 - 10:53 AM

As you say, the red fringe around Jupiter is on the bottom. But the red fringe on IO's shadow is on top. If you move the red channel up to correct for the planet, you'll make the red fringe on IO's shadow worse.

 

Controlpower29 doesn't say whether he's using a OSC or a mono camera.

Not so. both are red on the bottom of the bright area - a shadow obviously has the reverse colour fringing to a bright object (jupiter). So shifting the red up and the blue down will fix it. I'd do it now if i wasnt working. (yeah, i know)


  • RedLionNJ and KiwiRay like this

#6 controlpower29

controlpower29

    Sputnik

  • -----
  • topic starter
  • Posts: 42
  • Joined: 21 Sep 2020
  • Loc: Chicago, IL, USA

Posted 29 October 2020 - 05:49 PM

Controlpower29 doesn't say whether he's using a OSC or a mono camera.

Sorry that I forgot to mention my camera is ASI290MC. It is a color camera.



#7 controlpower29

controlpower29

    Sputnik

  • -----
  • topic starter
  • Posts: 42
  • Joined: 21 Sep 2020
  • Loc: Chicago, IL, USA

Posted 29 October 2020 - 06:22 PM

Not so. both are red on the bottom of the bright area - a shadow obviously has the reverse colour fringing to a bright object (jupiter). So shifting the red up and the blue down will fix it. I'd do it now if i wasnt working. (yeah, i know)

Well I tried both ways.

 

This is move red up and blue down by 10.

move Red Up And blue down
 

This is move blue up and red down by 10.

move blue Up And Red down

 

Neither of above fixed the shadow fringe issue. It looks the "move blue up and red down" looks slightly better. I tuned the red/blue up/down values in all kinds of ways, but the red fringe never disappeared. I am not sure what does the green box do. I also tried to move the green box to cover the shadow, but the color shifting still works on the whole picture.



#8 controlpower29

controlpower29

    Sputnik

  • -----
  • topic starter
  • Posts: 42
  • Joined: 21 Sep 2020
  • Loc: Chicago, IL, USA

Posted 29 October 2020 - 06:27 PM

Neither of above fixed the shadow fringe issue. It looks the "move blue up and red down" looks slightly better. I tuned the red/blue up/down values in all kinds of ways, but the red fringe never disappeared. I am not sure what does the green box do. I also tried to move the green box to cover the shadow, but the color shifting still works on the whole picture.

Suddenly I realized that the image was flipped. So the "up" and "down" may be opposite. But anyway, I tried both "move red up and blue down" and "move blue up and red down". Neither works.

 

This is possibly similar to the weird Saturn ring issue I earlier posted: https://www.cloudyni...ert/?p=10585893

Someone also suggested to use the ZWO ADC. The image quality itself is too bad, no way to fix in the post-processing.


Edited by controlpower29, 29 October 2020 - 06:34 PM.


#9 happylimpet

happylimpet

    Fly Me to the Moon

  • *****
  • Posts: 5,210
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 29 October 2020 - 06:28 PM

Those are the wrong way round - red is clearly lower in your image (red at the bottom) and so needs to be moved up. Im surprised it hasnt helped more though.

 

I must say i dont trust this function so much in registax (after i suggested it i know!), maybe it will work better in photoshop?

 

Anyway, I do thnk the bottom one looks better in terms of fringing.

 

But the bottom line is as someone said, get an ADC!


Edited by happylimpet, 29 October 2020 - 06:29 PM.


#10 John Boudreau

John Boudreau

    Surveyor 1

  • *****
  • Posts: 1,871
  • Joined: 06 Apr 2008
  • Loc: Boston Area, MA

Posted 29 October 2020 - 11:34 PM

I first looked at the image in RegiStax 6, simply trying to see why a manual channel realignment wasn't clearing thing up, but I became puzzled too. I then decided to do a channel split in Photoshop--- things became more interesting than I had expected so I had to put the attached image together. grin.gif

 

In the R channel Io's shadow is slightly elongated along the line of dispersion--- no surprise.

 

The G channel shows a double shadow that is barely separated, also along the dispersion line.

 

The B shows a really clean and wider split of Io's shadow along the same line.

 

It appears that you didn't use an IR blocking filter, and so have also recorded IR leak ghost images of Io's shadow in each channel. The QE chart shows how much the IR leak is separated from each color channel, and actually some atmospheric dispersion will add a bit more too, as dispersion is greater in the shorter wavelengths. It's largely these IR ghost shadows that caused your issues when you tried to manually align the channels. A properly adjusted ADC would have merged things nicely even without an IR blocker, but for best color balance an IR block is still required.

 

The intersection of the tick marks on the right side indicate the location of Io, which was in transit at the time. Io has virtually no blue, and so the B image's Io is essentially completely from the IR leak and so is as sharp as it's IR shadow in that channel. In the R and G channels Io is recorded in both visible light and IR and these overlap a bit, causing a bit of blurring--- less so in the R which would be expected.

 

The ASI290MC is one of the ZWO color cameras that can record strong IR signal, so ZWO makes them with an A/R coated 'clear' non-blocking window so that users can choose either an IR-blocking or various IR-pass filters, giving the camera more flexibility. The good news here is that the image quality will noticeably improve just with use use of an IR blocking filter. An ADC will also help but usually takes a bit of experience to get it working just right--- you'll find yourself getting better with it with practice, and probably find yourself enjoying it for visual use too. smile.gif

Attached Thumbnails

  • controlpower29_jupiterRGB_channels.jpg

  • RedLionNJ, happylimpet, AstroEthan and 1 other like this

#11 controlpower29

controlpower29

    Sputnik

  • -----
  • topic starter
  • Posts: 42
  • Joined: 21 Sep 2020
  • Loc: Chicago, IL, USA

Posted 30 October 2020 - 12:39 AM

I first looked at the image in RegiStax 6, simply trying to see why a manual channel realignment wasn't clearing thing up, but I became puzzled too. I then decided to do a channel split in Photoshop--- things became more interesting than I had expected so I had to put the attached image together. grin.gif

 

In the R channel Io's shadow is slightly elongated along the line of dispersion--- no surprise.

 

The G channel shows a double shadow that is barely separated, also along the dispersion line.

 

The B shows a really clean and wider split of Io's shadow along the same line.

 

It appears that you didn't use an IR blocking filter, and so have also recorded IR leak ghost images of Io's shadow in each channel. The QE chart shows how much the IR leak is separated from each color channel, and actually some atmospheric dispersion will add a bit more too, as dispersion is greater in the shorter wavelengths. It's largely these IR ghost shadows that caused your issues when you tried to manually align the channels. A properly adjusted ADC would have merged things nicely even without an IR blocker, but for best color balance an IR block is still required.

 

The intersection of the tick marks on the right side indicate the location of Io, which was in transit at the time. Io has virtually no blue, and so the B image's Io is essentially completely from the IR leak and so is as sharp as it's IR shadow in that channel. In the R and G channels Io is recorded in both visible light and IR and these overlap a bit, causing a bit of blurring--- less so in the R which would be expected.

 

The ASI290MC is one of the ZWO color cameras that can record strong IR signal, so ZWO makes them with an A/R coated 'clear' non-blocking window so that users can choose either an IR-blocking or various IR-pass filters, giving the camera more flexibility. The good news here is that the image quality will noticeably improve just with use use of an IR blocking filter. An ADC will also help but usually takes a bit of experience to get it working just right--- you'll find yourself getting better with it with practice, and probably find yourself enjoying it for visual use too. smile.gif

Your answer is amazing. Really appreciate your help. I will buy an IR blocking filter and give a try later.


  • John Boudreau likes this

#12 happylimpet

happylimpet

    Fly Me to the Moon

  • *****
  • Posts: 5,210
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 30 October 2020 - 05:40 AM

Nice one John!

 

I also did a channel split and wondered about the double image but it didnt occur to me that there was no IR block filter.


  • John Boudreau likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics