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

Extracting Ha and OIII from Dual Band Filters

  • Please log in to reply
31 replies to this topic

#1 mralpaca

mralpaca

    Vostok 1

  • -----
  • topic starter
  • Posts: 100
  • Joined: 09 May 2018
  • Loc: New York

Posted 14 December 2019 - 09:59 AM

Hello!

 

I am using a DSLR with an STC Duo-Narrow band filter and am wondering if anyone has a way to separate out the OIII and Ha channels using APP or PI. I am hoping to create a hubble palette image, and realize that I need to get the channels right before I run the color swapping.

 

Any help would be greatly appreciated!

 

Thanks,

 

Alex



#2 kathyastro

kathyastro

    Vanguard

  • *****
  • Posts: 2,147
  • Joined: 23 Dec 2016
  • Loc: Nova Scotia

Posted 14 December 2019 - 10:16 AM

Red = Ha.  Green = Oiii.  If you do a channel extraction on the image, you should be able to distinguish the two bands easily.  Then recombine as desired for the Hubble palette or HOO image.



#3 tkottary

tkottary

    Messenger

  • *****
  • Posts: 490
  • Joined: 06 Dec 2015
  • Loc: SunnyVale ,CA

Posted 14 December 2019 - 11:25 AM

pretty Simple with APP.

1. Load all your files and chose algorithm as Ha-OIII extract Ha in Tab 0 (raw/fits)

2. proceed all the way & integrate. You will now get Ha integrated file.

3. Without changing anything, go back to Tab 0, change algorithm to Ha-OIII extract OIII. It will ask you to confirm, press ok. 

4. Go to Tab 5 Normalize & then integrate to get oiii master.

 

 

ha-oii.jpg


  • Scott Mitchell, psandelle, AhBok and 1 other like this

#4 mewmartigan

mewmartigan

    Apollo

  • *****
  • Posts: 1,280
  • Joined: 02 Jul 2008
  • Loc: outside of Pittsburgh, PA

Posted 14 December 2019 - 11:26 AM

In Pixinsight you can split the channels and then use PixelMath to make Red the Ha and Green+Blue  the OIII channel.

 

Here is a thread on the subject including weights for the various colors/pixels instead of just straight red=Ha:

 

 https://www.cloudyni...d-and-osc-help/


  • 42itous1 and Swordfishy like this

#5 mralpaca

mralpaca

    Vostok 1

  • -----
  • topic starter
  • Posts: 100
  • Joined: 09 May 2018
  • Loc: New York

Posted 16 December 2019 - 01:24 PM

Amazing! Thank you everyone :)



#6 mralpaca

mralpaca

    Vostok 1

  • -----
  • topic starter
  • Posts: 100
  • Joined: 09 May 2018
  • Loc: New York

Posted 17 December 2019 - 03:39 PM

When I perform the steps that tko suggests, I get individual subs that look almost like they need to be debayered. When I debayer, they still get that same weird square artifact. Is this normal?



#7 tkottary

tkottary

    Messenger

  • *****
  • Posts: 490
  • Joined: 06 Dec 2015
  • Loc: SunnyVale ,CA

Posted 17 December 2019 - 03:47 PM

Can you check the force debayer option on raw 0 tab and try .

#8 mralpaca

mralpaca

    Vostok 1

  • -----
  • topic starter
  • Posts: 100
  • Joined: 09 May 2018
  • Loc: New York

Posted 17 December 2019 - 04:00 PM

It was on but I had RGGB as my pattern, and not supported. I am trying without force debayer (mostly through that) then will flip that on!



#9 mralpaca

mralpaca

    Vostok 1

  • -----
  • topic starter
  • Posts: 100
  • Joined: 09 May 2018
  • Loc: New York

Posted 17 December 2019 - 04:23 PM

Still has those debayer-looking boxes, unfortunately. Has anyone else seen this?



#10 AhBok

AhBok

    Mercury-Atlas

  • *****
  • Posts: 2,705
  • Joined: 02 Dec 2010
  • Loc: Lakeland, TN

Posted 18 December 2019 - 11:04 PM

You must check force Bayer CFA in APP if you are using an OSC cam, even when extracting Ha. I would go ahead and select the correct debayer pattern for your camera rather than rely on the default “supported” setting. APP makes extracting individual color channels as well as Ha and OIII channels very easy.
  • psandelle likes this

#11 mralpaca

mralpaca

    Vostok 1

  • -----
  • topic starter
  • Posts: 100
  • Joined: 09 May 2018
  • Loc: New York

Posted 14 January 2020 - 11:28 AM

Well now I am stumped. I used an L-eNhance filter with my ZWO ASI 294MC and tried to separate the channels and... the images come out in full color. Is the L-eNhance not "narrow" enough for this algo?



#12 erictheastrojunkie

erictheastrojunkie

    Viking 1

  • *****
  • Posts: 787
  • Joined: 15 Jul 2016
  • Loc: Salt Lake City

Posted 14 January 2020 - 11:46 AM

My process for this in PixInsight was:

1) Perform all frame calibration (flats, bias, darks) on light frames

2) Debayer all light frames

3) Use batchchannelextractor to extract red files into one folder and green/blue files into another

4) Use subframe selector to analyse the red files and the green/blue files, if blue files contained enough SNRweight and overall signal I'd include them, if not I'd discard them. Output frames with weighting/FITS keyword

5) Use staralign to align red files to the best subframe from the red analysis and then use green reference frame from the same red frame source (so if the red frame came from Image #1, use the green frame from Image #1 to align the green/blue to) and align the green/blue frames.

6) Stack red frames (name the red stack H-alpha) and green/blue frames (name the green/blue stack Oiii)

7) Do all my cropping, background extraction, and stretching/curves on the H-alpha and Oiii, be sure to match the crop.

8) Use PixelMath to combine the H-alpha and Oiii. 

 

For step 5, if you use difference reference frames for the red and green/blue then you'll have to align the stacked H-alpha and Oiii frames using staralign during step 7. 



#13 SilverLitz

SilverLitz

    Apollo

  • -----
  • Posts: 1,024
  • Joined: 17 Feb 2018
  • Loc: Louisville, KY

Posted 14 January 2020 - 12:15 PM

Using a multiple band pass filter with a OSC, with result in somewhat true color image, and it will NOT look like what you see with mono cameras and NB filters processed in the typical Hubble Pallet.

 

These will put Ha and Sii into the Red channel and Oiii into the both the Green and Blue.  This will be somewhat like HOO NB processing, with more red as it adds some Sii into red as well.  All of the NB images you see posted have had various other processing steps, such shifting the particular hue curves (especially for Hubble, or SHO).

 

You will need to separate you combined RGB image from you OSC into the separate R,G,B channels.  This can be done in Photoshop by going into the Channels tab.  PixInsight has specific functions that can separate RGB into Luminescence and the the separate R,G,B channels as well. 



#14 OhmEye

OhmEye

    Viking 1

  • *****
  • Posts: 781
  • Joined: 15 Sep 2019
  • Loc: Western NY Southern Tier

Posted 14 January 2020 - 12:56 PM

Astro Pixel Processor (APP) has a convenient option to extract these during debayering to process as separate channels, it's what I used with my OSC+L_Enhance filter.



#15 mralpaca

mralpaca

    Vostok 1

  • -----
  • topic starter
  • Posts: 100
  • Joined: 09 May 2018
  • Loc: New York

Posted 14 January 2020 - 01:48 PM

Astro Pixel Processor (APP) has a convenient option to extract these during debayering to process as separate channels, it's what I used with my OSC+L_Enhance filter.

When I do this, the image comes out in full color, is this what happens with you too? I expect the OIII to be blue/green and not red. 



#16 mralpaca

mralpaca

    Vostok 1

  • -----
  • topic starter
  • Posts: 100
  • Joined: 09 May 2018
  • Loc: New York

Posted 14 January 2020 - 01:48 PM

Using a multiple band pass filter with a OSC, with result in somewhat true color image, and it will NOT look like what you see with mono cameras and NB filters processed in the typical Hubble Pallet.

 

These will put Ha and Sii into the Red channel and Oiii into the both the Green and Blue.  This will be somewhat like HOO NB processing, with more red as it adds some Sii into red as well.  All of the NB images you see posted have had various other processing steps, such shifting the particular hue curves (especially for Hubble, or SHO).

 

You will need to separate you combined RGB image from you OSC into the separate R,G,B channels.  This can be done in Photoshop by going into the Channels tab.  PixInsight has specific functions that can separate RGB into Luminescence and the the separate R,G,B channels as well. 

I will need to dig into this in more detail!



#17 DubbelDerp

DubbelDerp

    Viking 1

  • *****
  • Posts: 994
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 14 January 2020 - 02:25 PM

When I do this, the image comes out in full color, is this what happens with you too? I expect the OIII to be blue/green and not red. 

Something isn't set up right - when you run it with the "extract OIII" debayer option, the resulting image should be greyscale. Under 0) RAW/FITS, under "bayer CFA" change the algorithm to "Ha-OIII extract OIII", run through your calibration, analyze stars, register, normalize.. then when you integrate the image, the integration will be a greyscale extraction of the OIII channel. Go back to 0) RAW/FITS and change it to "Ha-OIII extract Ha" and integrate your data set again to get the Ha channel registered to the OIII channel. 

 

Note that only the integration will be greyscale - if you look at any of the subframes, they will be in color.



#18 mralpaca

mralpaca

    Vostok 1

  • -----
  • topic starter
  • Posts: 100
  • Joined: 09 May 2018
  • Loc: New York

Posted 14 January 2020 - 03:13 PM

I agree that something must not be right, but I re-did the analysis and stacking and did not get grayscale. Bayer CFA was on and I had exactly those settings. 

 

Something isn't set up right - when you run it with the "extract OIII" debayer option, the resulting image should be greyscale. Under 0) RAW/FITS, under "bayer CFA" change the algorithm to "Ha-OIII extract OIII", run through your calibration, analyze stars, register, normalize.. then when you integrate the image, the integration will be a greyscale extraction of the OIII channel. Go back to 0) RAW/FITS and change it to "Ha-OIII extract Ha" and integrate your data set again to get the Ha channel registered to the OIII channel. 

 

Note that only the integration will be greyscale - if you look at any of the subframes, they will be in color.



#19 DubbelDerp

DubbelDerp

    Viking 1

  • *****
  • Posts: 994
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 14 January 2020 - 03:27 PM

What's the filename for the integration? It should start with something like St-avg-#### with somewhere in the file name "HaOIII-OIII". I don't check "force Bayer CFA" and leave the pattern set to "supported." 

 

Which version of APP do you have? Would you mind uploading a few subs somewhere and I can try on my workstation?



#20 OhmEye

OhmEye

    Viking 1

  • *****
  • Posts: 781
  • Joined: 15 Sep 2019
  • Loc: Western NY Southern Tier

Posted 14 January 2020 - 03:29 PM

Debayering in APP when set to extract Ha or OII processes into a mono image. I would do Ha, then go back to the 0 tab and change to OIII and integrate again, say yes when it asks to Normalize again. Then I would use RGB Combine to create a HOO bicolor image and use a fairly typical set of process steps from there. I had to make sure CFA was forced for it to extract, but otherwise it seemed straightforward to me.



#21 DubbelDerp

DubbelDerp

    Viking 1

  • *****
  • Posts: 994
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 14 January 2020 - 03:38 PM

Just for reference, here's a snip from the OIII data from when I shot the Veil Nebula this last summer. This was with the "extract OIII" debayer algorithm...

veil-snip.jpg



#22 mralpaca

mralpaca

    Vostok 1

  • -----
  • topic starter
  • Posts: 100
  • Joined: 09 May 2018
  • Loc: New York

Posted 14 January 2020 - 04:01 PM

Here is my file name: St-avg-5620.0s-SC_1_2.7_none-x_1.0_LZ3-NS-full-qua-add-sc_BWMV_nor-HaOIII-Ha-RE-MBB5_1stLNC_it3 

 

 

Here are some subs: https://drive.google...WMf?usp=sharing



#23 mralpaca

mralpaca

    Vostok 1

  • -----
  • topic starter
  • Posts: 100
  • Joined: 09 May 2018
  • Loc: New York

Posted 14 January 2020 - 04:02 PM

I am using APP version 1.075


Edited by mralpaca, 14 January 2020 - 04:02 PM.


#24 DubbelDerp

DubbelDerp

    Viking 1

  • *****
  • Posts: 994
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 14 January 2020 - 05:17 PM

Those look like stacks - are you trying to extract Ha and OIII from them? That won’t work, since the stacks are already debayered. Try reloading your raw light frames and calibration files, switch the debayer algorithm to extract Ha or OIII, and redo the integration. If you can upload a handful of the raw light frames, I can give it a try too.

You could also take the integrated RGB file and under the calibration tab, select split channels and save the calibrated files at the bottom of the tab. This will give you a rough approximation since the Ha will be in the red channel and OIII in the green and blue, but you’ll get a much better result if you go back to the raw files and change the debayer algorithm, especially for Ha.

#25 mralpaca

mralpaca

    Vostok 1

  • -----
  • topic starter
  • Posts: 100
  • Joined: 09 May 2018
  • Loc: New York

Posted 14 January 2020 - 05:25 PM

Ahhh got it. I used sharpcap to do short exposures (I am in NYC, so short exposures stacked into 5 minute "subs" work best). I will give it a shot, but I appreciate you taking a look!




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