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

HaRGB with duo narrowband filter

  • Please log in to reply
34 replies to this topic

#1 Riaandw

Riaandw

    Vostok 1

  • *****
  • topic starter
  • Posts: 117
  • Joined: 24 May 2021
  • Loc: South Africa

Posted 19 September 2021 - 11:56 AM

So here's a question.

 

Would like to experiment with HaRGB with the 533MC.

 

I realize that OSC would only be 25% efficient using an H alpha filter and all ifs ands or buts I'll invest in a mono camera to capture this channel to combine with the color data. These are big purchases to me and the classifieds has seen my entire budget for the last few months.

 

My question is could I using for instance the L-extreme simply extract the H alpha channel to combine as Luminance using a one shot color camera? Sounds good in theory, not sure if it would work in practice.



#2 Riaandw

Riaandw

    Vostok 1

  • *****
  • topic starter
  • Posts: 117
  • Joined: 24 May 2021
  • Loc: South Africa

Posted 19 September 2021 - 11:59 AM

This would justify the purchase of a duo narrowband since HA filters seem to be just as expensive. I would then be able to with the Hubble palette also.


  • 17.5Dob likes this

#3 DJL

DJL

    Viking 1

  • -----
  • Posts: 948
  • Joined: 15 Aug 2020

Posted 19 September 2021 - 12:05 PM

I did a fair bit of HaRGB with my OSC before getting my L-Extreme. I didn't get to do a whole lot with the L-Extreme but you can think of it as an Ha and OIII, shoot with it and pick the Ha and OIII apart in Astro Pixel Processor, possibly combining with white light exposures for star color. One thing not to do with it is very wide field for example using the Samyang / Rokinon 135, as it gives less throughput. I spent night after night on the Spaghetti Nebula using the L-Extreme and apparently would have had better results with an Ha filter.


  • Riaandw likes this

#4 DJL

DJL

    Viking 1

  • -----
  • Posts: 948
  • Joined: 15 Aug 2020

Posted 19 September 2021 - 12:05 PM

This would justify the purchase of a duo narrowband since HA filters seem to be just as expensive. I would then be able to with the Hubble palette also.

It won't give you the Sulphur channel. You can do HOO.


  • Riaandw likes this

#5 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 25,978
  • Joined: 27 Oct 2014

Posted 19 September 2021 - 12:16 PM

So here's a question.

 

Would like to experiment with HaRGB with the 533MC.

 

I realize that OSC would only be 25% efficient using an H alpha filter and all ifs ands or buts I'll invest in a mono camera to capture this channel to combine with the color data. These are big purchases to me and the classifieds has seen my entire budget for the last few months.

 

My question is could I using for instance the L-extreme simply extract the H alpha channel to combine as Luminance using a one shot color camera? Sounds good in theory, not sure if it would work in practice.

The duoband filters are designed to work well with OSC.  And they do.  <smile>

 

So, your idea is basically good.  Just don't try to use the LExtreme as a narrowband Ha filter.  A mistake _many_ make.

 

With the 533, gather LExtreme data, as the filter is designed to do, and use it as luminance.  Gather OSC data as RGB.  Don't shoot (Ha)RGB (I do those, with a mono camera).  Shoot (LExtreme)RGB or, stated differently (LExtreme)(OSC) images.  That will work better, _and_ be easier to process.  With that equipment, it's the way to go.

 

Bottom line.  Don't try to pretend the 533 is mono or the LExtreme is an Ha filter.  Use both as designed.  As a general principle, using equipment as the engineers designed them to do is better.  People are forever trying to come up with "creative" alternatives.  It's very rare that they work better.


Edited by bobzeq25, 19 September 2021 - 12:22 PM.

  • AhBok, cucubits and Riaandw like this

#6 Riaandw

Riaandw

    Vostok 1

  • *****
  • topic starter
  • Posts: 117
  • Joined: 24 May 2021
  • Loc: South Africa

Posted 19 September 2021 - 12:22 PM

It won't give you the Sulphur channel. You can do HOO.

From my youtube degree ;) Pixelmath in Pixinsght creates the Sulphur channel by partial combination of Ha and Oii for a fake false color image... 


  • 17.5Dob and DJL like this

#7 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 25,978
  • Joined: 27 Oct 2014

Posted 19 September 2021 - 12:24 PM

From my youtube degree wink.gif Pixelmath in Pixinsght creates the Sulphur channel by partial combination of Ha and Oii for a fake false color image... 

Why do fake when you can do real?  <smile>  Live with what the equipment is designed to do, don't try to make it something else.

 

Better to do the images that are suited to the equipment.


Edited by bobzeq25, 19 September 2021 - 12:26 PM.

  • AhBok, cucubits, DJL and 1 other like this

#8 Riaandw

Riaandw

    Vostok 1

  • *****
  • topic starter
  • Posts: 117
  • Joined: 24 May 2021
  • Loc: South Africa

Posted 19 September 2021 - 12:25 PM

Thanks for the advise. It would then seem the L-extreme or Idas nebula booster filter would be a better purchase right now than the narrowband H alpha filter.


  • bobzeq25 likes this

#9 DJL

DJL

    Viking 1

  • -----
  • Posts: 948
  • Joined: 15 Aug 2020

Posted 19 September 2021 - 08:39 PM

The duoband filters are designed to work well with OSC.  And they do.  <smile>

 

So, your idea is basically good.  Just don't try to use the LExtreme as a narrowband Ha filter.  A mistake _many_ make.

 

With the 533, gather LExtreme data, as the filter is designed to do, and use it as luminance.  Gather OSC data as RGB.  Don't shoot (Ha)RGB (I do those, with a mono camera).  Shoot (LExtreme)RGB or, stated differently (LExtreme)(OSC) images.  That will work better, _and_ be easier to process.  With that equipment, it's the way to go.

 

Bottom line.  Don't try to pretend the 533 is mono or the LExtreme is an Ha filter.  Use both as designed.  As a general principle, using equipment as the engineers designed them to do is better.  People are forever trying to come up with "creative" alternatives.  It's very rare that they work better.

What's your opinion of using Astro Pixel Processor Tab 1 to separate the Ha and OIII from the L-Extreme?


  • Riaandw likes this

#10 AhBok

AhBok

    Gemini

  • *****
  • Posts: 3,359
  • Joined: 02 Dec 2010
  • Loc: Lakeland, TN

Posted 19 September 2021 - 10:30 PM

I couldn’t agree with Bob more on this. Actually, extracting OIII and Ha is a good strategy, especially when combining with OSC color data. Simply split the color data into separate RGB channels and normalize them with the Ha and OIII extracts and then use the combine tool for a RGBHOO image. Go easy on the Ha and OIII for excellent OSC color images with enhanced Ha and OIII regions.

You can certainly create narrowband-like images that sort of emulate SHO images, but I find them much inferior to the real thing. It is personal taste, but I would rather see an enhanced OSC image than a poor false color image that will never compete with an actual NB image.

But that is just my take. Others MMV.


Edited by AhBok, 20 September 2021 - 09:32 AM.

  • bobzeq25 and Riaandw like this

#11 Riaandw

Riaandw

    Vostok 1

  • *****
  • topic starter
  • Posts: 117
  • Joined: 24 May 2021
  • Loc: South Africa

Posted 19 September 2021 - 10:39 PM

Why do fake when you can do real?  <smile>  Live with what the equipment is designed to do, don't try to make it something else.

 

Better to do the images that are suited to the equipment.

A science guy and AP borders on art... Not sure how to deal with it ;)


  • DJL likes this

#12 imtl

imtl

    Skylab

  • *****
  • Moderators
  • Posts: 4,124
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 19 September 2021 - 11:01 PM

SHO images or narrowband in general are not about the ''Hubble color''. They are about capturing the different gaseous emission structures of Ha, Oiii and Sii (Nii for planetary nebulae) seperately so one can control the processing.

Mimiking a color palette is not doing that. It's just coloring. The same as using a LP filter on broadband objects and then tweaking the colors to be ''right'' does not recover the details lost due to the filter.

Of course it's an art and everyone can do what they want. It just does not make a tomato into a cucumber when you just paint it green.

What Bob and others are trying to tell you is it's better to use tools for what they were designed for and if one is interested in other outcomes then get the proper tools for them. Just renaming something doesn't change its essence.

But, do as you want of course, be happy with your own hobby.


  • Juno18 and Riaandw like this

#13 Riaandw

Riaandw

    Vostok 1

  • *****
  • topic starter
  • Posts: 117
  • Joined: 24 May 2021
  • Loc: South Africa

Posted 19 September 2021 - 11:37 PM

Agreed. I enjoy true color images. SHO has scientific relevance and I guess the Hubble palette is representative of that.

 

Will experiment with the L-extreme. Basically I want to experiment with HaRGB, if I enjoy that I invest in that direction. If separating the Ha and Oiii gives me the same result as an 7mn Ha filter with a OSC camera then the duo narrow band is a more useful tool.

 

Read that mono binning should be more efficiency than 25% with OSC when using Ha only at the cost of resolution... Not sure I completely grasp this yet, but interesting.
https://www.cloudyni...era/?p=6846949 



#14 Riaandw

Riaandw

    Vostok 1

  • *****
  • topic starter
  • Posts: 117
  • Joined: 24 May 2021
  • Loc: South Africa

Posted 19 September 2021 - 11:40 PM

I like Alex McConahay perspective here. Its innovative thinking that anything can be done with anything.



#15 imtl

imtl

    Skylab

  • *****
  • Moderators
  • Posts: 4,124
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 19 September 2021 - 11:55 PM

Agreed. I enjoy true color images. SHO has scientific relevance and I guess the Hubble palette is representative of that.

 

Will experiment with the L-extreme. Basically I want to experiment with HaRGB, if I enjoy that I invest in that direction. If separating the Ha and Oiii gives me the same result as an 7mn Ha filter with a OSC camera then the duo narrow band is a more useful tool.

 

Read that mono binning should be more efficiency than 25% with OSC when using Ha only at the cost of resolution... Not sure I completely grasp this yet, but interesting.
https://www.cloudyni...era/?p=6846949 

One thing I would say is that L-extreme has a bandpass of 7nm for Ha (and Oiii) which is pretty broad for NB imaging. Especially in LP areas. I think experimenting with what you have is great. However, have a perspective on the the outcome. It's also a lot about the processing. 


  • Juno18 and Riaandw like this

#16 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 25,978
  • Joined: 27 Oct 2014

Posted 19 September 2021 - 11:58 PM

What's your opinion of using Astro Pixel Processor Tab 1 to separate the Ha and OIII from the L-Extreme?

I think it's a bad idea.  You're trying to make a duoband filter and a one shot camera pretend they're single narrowband filters and a mono camera.  That's fraught with potential problems (look at the overlaps between colors with the Bayer matrix filter used on every one shot color camera).   _And_ a lot of extra work.

 

If you want to record Ha and O(III) get a mono camera and two filters.  That's what they're designed to do.  They cleanly separate these signals.  And use all the pixels on all the light that comes through the filter.  They let you adjust exposure for the generally strong Ha and the generally weak O(III).  They are the right tools for that job.

 

If you could do narrowband imaging with a one shot color camera, and processing, why would anyone spend the money for a mono camera and narrowband filters.  Are the people who do it stupid?

 

Yes, you can do anything you want to.  You'll even get an image.  The question is, is it better than doing standard practice?  I don't think so.


Edited by bobzeq25, 20 September 2021 - 12:01 AM.

  • Riaandw likes this

#17 imtl

imtl

    Skylab

  • *****
  • Moderators
  • Posts: 4,124
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 20 September 2021 - 12:02 AM

I like Alex McConahay perspective here. Its innovative thinking that anything can be done with anything.

What Alex wrote (which was about 12nm Ha filter vs. a 3nm. And not about OSC with NB) was:

 

 

 

If you can get a deal on a twelve or five (nm bandpass), give it a shot......Nearly anything can be done with anything.

 

It is easier to get results you want with ideal equipment, but you can get something with almost anything.

 

Alex

Which is what we said over here. Can you do it? yes. To a point. Will it be ideal? No.


  • Riaandw likes this

#18 DJL

DJL

    Viking 1

  • -----
  • Posts: 948
  • Joined: 15 Aug 2020

Posted 20 September 2021 - 12:18 AM

I think it's a bad idea.  You're trying to make a duoband filter and a one shot camera pretend they're single narrowband filters and a mono camera.  That's fraught with potential problems (look at the overlaps between colors with the Bayer matrix filter used on every one shot color camera).   _And_ a lot of extra work.

 

If you want to record Ha and O(III) get a mono camera and two filters.  That's what they're designed to do.  They cleanly separate these signals.  And use all the pixels on all the light that comes through the filter.  They let you adjust exposure for the generally strong Ha and the generally weak O(III).  They are the right tools for that job.

 

If you could do narrowband imaging with a one shot color camera, and processing, why would anyone spend the money for a mono camera and narrowband filters.  Are the people who do it stupid?

 

Yes, you can do anything you want to.  You'll even get an image.  The question is, is it better than doing standard practice?  I don't think so.

Just taking the opportunity to ask you as one of the resident APP experts, because there seem to be alternatives with APP and with duoband filters. I started APP for OSC only, learned to add Ha and then really wasn't sure what to do for the best when processing L-Extreme images.

 

I haven't used the L-Extreme since the spring as I had a clouded over summer followed by planetary boot camp.  A mono camera and SHO filters is definitely in my future.


  • Riaandw likes this

#19 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 25,978
  • Joined: 27 Oct 2014

Posted 20 September 2021 - 12:42 AM

Just taking the opportunity to ask you as one of the resident APP experts, because there seem to be alternatives with APP and with duoband filters. I started APP for OSC only, learned to add Ha and then really wasn't sure what to do for the best when processing L-Extreme images.

 

I haven't used the L-Extreme since the spring as I had a clouded over summer followed by planetary boot camp.  A mono camera and SHO filters is definitely in my future.

I can tell you what I do with a 2600MC and my NBZ filter, the equivalent of the LExtreme for fast scopes, like my C8 RASA.  I process the data as I would RGB from the OSC camera.  I split nothing.  I color balance as I would with RGB.

 

Here's the result.  Bortle 7 skies, dim target.  Acquisition data here, click on it to enlarge, for better detail.

 

https://www.astrobin.com/kis712/

 

Yes there are alternatives.  When I can do images like that, it's not worth my time to mess around with them.  That image used all the photons.  My default is always standard practice, it's hard enough to make that work.

 

The one thing I could have done was split the luminance data from the color data, and process the color differently, then recombine.  You can be more aggressive with noise reduction on the color, if it's a bit blurry the luminance will sharpen it up.  EXCEPT...

 

I use PixInsight.  Some of its noise reduction tools let you apply them separately to luminance and color within one overall application.  So, no need to split them manually, PI does it automatically.


Edited by bobzeq25, 20 September 2021 - 12:59 AM.

  • imtl and Riaandw like this

#20 Riaandw

Riaandw

    Vostok 1

  • *****
  • topic starter
  • Posts: 117
  • Joined: 24 May 2021
  • Loc: South Africa

Posted 20 September 2021 - 07:44 AM

These results though, makes me want to consider the Ha nb OSC again.

 

https://www.cloudyni...mera/?p=6850242



#21 jonnybravo0311

jonnybravo0311

    Mercury-Atlas

  • *****
  • Posts: 2,664
  • Joined: 05 Nov 2020
  • Loc: NJ, US

Posted 20 September 2021 - 09:08 AM

I did a comparison a while back between using the L-eXtreme data "as is" - i.e. processing it as RGB - vs separating and constructing an HOO image using pixel math. Here's the result:

 

gallery_347158_15661_5818492.png

 

The "RGB" processed data is on the left, the "HOO" is on the right. You can see some subtle differences... for example, the image on the left is more "salmon and teal". Are these differences profound enough for you to do the channel splitting? Only you can answer that. I image with a mono setup, so this is completely irrelevant for me. If I want HOO, I image with the Ha and O3 filters :).


  • bobzeq25 and Riaandw like this

#22 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 25,978
  • Joined: 27 Oct 2014

Posted 20 September 2021 - 01:51 PM

I did a comparison a while back between using the L-eXtreme data "as is" - i.e. processing it as RGB - vs separating and constructing an HOO image using pixel math. Here's the result:

 

gallery_347158_15661_5818492.png

 

The "RGB" processed data is on the left, the "HOO" is on the right. You can see some subtle differences... for example, the image on the left is more "salmon and teal". Are these differences profound enough for you to do the channel splitting? Only you can answer that. I image with a mono setup, so this is completely irrelevant for me. If I want HOO, I image with the Ha and O3 filters smile.gif.

I'll bet I can take the one on the left and make it look every bit as good (maybe better) as the one on the right - with a lot less trouble than splitting, processing two things, recombining.

 

The same is true with the other OSC/duoband/splitting images the OP is impressed by.

 

People think Ha is "magic", the term causes them to swoon.   With a mono camera that uses all the pixels on what comes through the filter - it is.  With OSC which only uses 25% of the pixels on the Ha data - not so much.  OSC is where duoband shines.  Without any splitting.


  • jonnybravo0311 and Riaandw like this

#23 imtl

imtl

    Skylab

  • *****
  • Moderators
  • Posts: 4,124
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 20 September 2021 - 02:03 PM

These results though, makes me want to consider the Ha nb OSC again.

 

https://www.cloudyni...mera/?p=6850242

One word of caution. Don't be blind by one of the brightest emission nebula in the sky together with 10 hours integration and in the hands of a very capable imager (and image processor). This should not lead you to a generalized conclusion. That would be my advice.


  • bobzeq25 and Riaandw like this

#24 Riaandw

Riaandw

    Vostok 1

  • *****
  • topic starter
  • Posts: 117
  • Joined: 24 May 2021
  • Loc: South Africa

Posted 21 September 2021 - 01:56 AM

We are talking getting my foot in the door and not an instant perfect solution. Thing is theat when a mono CMOS is later acquired the duo narrow band will not be useful. Where as a NB Ha filter would be useful now with my color sensors and then with a mono later right?

 

Mostly image at f/4 to f/6. At that focal length I am capable of 10 minute exposures even on my fork mounted 14" sct. The Ha is also useful for galaxies. Can be used when the moon's out.

 

Processing skills at this stage are basic monkey see monkey do level, but this will always be a place to grow into skill wise. An investment of time rather than money. The equipment is the expensive part and processing skill is also limited by that unless we process someone else data. 

 

I will probably have osc around always and that is the only consideration towards spending the 300usd on a duo NB filter. 


  • imtl and Juno18 like this

#25 imtl

imtl

    Skylab

  • *****
  • Moderators
  • Posts: 4,124
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 21 September 2021 - 02:40 AM

Well, when you move to mono setup you should also move to a 3nm or 5nm bandpass Ha.

Some galaxies have Ha spots that give an extra dimension to it. I personally don't bother but sure a lot of people like that.

Processing is 80% of AP.
  • Riaandw 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