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

Compensating for weak Oiii when processing bicolor NB

  • Please log in to reply
8 replies to this topic

#1 raguramm

raguramm

    Explorer 1

  • *****
  • topic starter
  • Posts: 50
  • Joined: 15 Jan 2019

Posted 13 October 2019 - 03:20 PM

Currently working on Fish Head, but ran into the same issue with Melotte 15.

I get fantastic Ha SNR within 4 hours, but I've hit 7 hours on Oiii and my signal is still weak. I plan on collecting Sii tonight, which hopefully helps, but putting a bicolor image together, my Ha totally overwhelms the Oiii and yields a red image. Tonemapping also hasn't worked really well, since the amount of stretch required to boost the oxygen also brings out a lot of noise in that channel and turns my image background purple/noisy.

Any suggestions in processing? Will doing SHO for weak O images help? I did collect my oxygen during an 80% moon, so I know that could be an issue, but I like to save my new moon times for broadband, which I get to shoot much less often.
  • 2ghouls likes this

#2 HunterofPhotons

HunterofPhotons

    Apollo

  • *----
  • Posts: 1435
  • Joined: 26 Apr 2008
  • Loc: Rhode Island, USA

Posted 14 October 2019 - 10:10 AM

Take more OIII.

Try to make your narrowband channels equally 'clean'.

It's been years since I did IC 1795 but when I did I ended up collecting 3.5 hours Ha, 10 hours SII, and 9 hours OIII.

Processing is much easier then.

 

dan k.


  • Miguelo and 2ghouls like this

#3 ChrisWhite

ChrisWhite

    Aurora

  • *****
  • Posts: 4760
  • Joined: 28 Feb 2015
  • Loc: Colchester, VT

Posted 14 October 2019 - 10:25 AM

Do you use pixinsight?

#4 terry59

terry59

    Cosmos

  • *****
  • Posts: 9185
  • Joined: 18 Jul 2011
  • Loc: Colorado, USA

Posted 14 October 2019 - 10:57 AM

You could blend a bit of Ha into your OIII



#5 miwitte

miwitte

    Apollo

  • *****
  • Posts: 1132
  • Joined: 28 Dec 2016
  • Loc: Charlotte NC

Posted 14 October 2019 - 11:46 AM

What ive seen and done is use J-P METSAVAINIO techniques of creating starless images which then allows you to really stretch the particular channel. But yes more integration time is the ultimate answer but you can cheat using this. Ive used stratton and then Photoshop to remove artifacts, StarNet++ seems to be the new way to do it. I just love his work and have tried my best to emulate some of his shots.

 

J-Ps site

https://astroanarchy.blogspot.com/

 

tutorial

http://www.cedic.at/...Metsavainio.pdf



#6 BenKolt

BenKolt

    Surveyor 1

  • *****
  • Posts: 1992
  • Joined: 13 Mar 2013

Posted 14 October 2019 - 11:47 AM

raguramm:

 

I have to agree with Dan (HunterofPhotons) that your long term strategy needs to be to capture more data on the weak channels.  Can't beat Mother Nature on this one - SNR goes up roughly as the square root of the number of frames.  In the meantime, you can play around with blending Ha into OIII. as Terry suggests.  But think long-term goals of returning to IC1795 and other targets time and time again for the purpose of accumulating sufficient data.  No matter what processing methods I've attempted to compensate for noisy data, in the final analysis the best solution was simply more integration time!

 

Good luck!

 

Ben



#7 2ghouls

2ghouls

    Viking 1

  • *****
  • Posts: 828
  • Joined: 19 Sep 2016

Posted 14 October 2019 - 02:02 PM

Tonemapping also hasn't worked really well, since the amount of stretch required to boost the oxygen also brings out a lot of noise in that channel and turns my image background purple/noisy.

 

I did collect my oxygen during an 80% moon, so I know that could be an issue, but I like to save my new moon times for broadband, which I get to shoot much less often.

This depends a bit on the object and how strong the OIII signal is. For example, in NGC7000, it is pretty strong, and you could probably shoot full moon, and do near 1:1 with Ha and still get a nice photo. But most objects I've shot, the OIII and SII signal are much, much weaker than Ha. OIII is the most affected by moonlight and terrestrial light pollution sources (white LEDs), so I do try to shoot it when the moon is not high, or already set. Like you, I also try to save new moon for broadband, but sometimes will shoot some OIII too. SII and Ha, I mostly shoot from home, and any clear night. For noise introduced from the highly stretched OIII, here are 3 ideas:

1. As others have said, simply shoot more OIII

2. Use heavier noise reduction on the OIII

3. If tonemapping, increase the contrast- make everything but the OIII signal completely black (0,0,0). I see a lot of people who follow JP's tutorial miss this. If you look at his tone maps, the background is black. It will lighten naturally once you add back the stars with the artificial lum.


  • ArkabPriorSol likes this

#8 miwitte

miwitte

    Apollo

  • *****
  • Posts: 1132
  • Joined: 28 Dec 2016
  • Loc: Charlotte NC

Posted 15 October 2019 - 11:47 AM

This is a great point as i missed that "If tonemapping, increase the contrast- make everything but the OIII signal completely black (0,0,0). I see a lot of people who follow JP's tutorial miss this. If you look at his tone maps, the background is black. It will lighten naturally once you add back the stars with the artificial lum."



#9 ChrisWhite

ChrisWhite

    Aurora

  • *****
  • Posts: 4760
  • Joined: 28 Feb 2015
  • Loc: Colchester, VT

Posted 15 October 2019 - 01:21 PM

If you use pixinsight, simply linear fit HA to OIII. Nothing beats more integration time, but if you do what I suggested you will not have your oiii overwhelmed by ha.
  • 2ghouls 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