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

Why do narrowband imaging at all?

  • Please log in to reply
26 replies to this topic

#1 hyperion0001

hyperion0001

    Sputnik

  • -----
  • topic starter
  • Posts: 45
  • Joined: 26 Sep 2007
  • Loc: Concord, MA

Posted 04 December 2020 - 10:23 AM

Hi,

 

First, I image with a 12inch SCT, so my question is in reference to mostly small DSOs.

I can understand that narrowband filters are more selective in the wavelengths they pass through. In that case, images may look a little different because they focus on certain "colors" more than others. However, that data is still present in an RGB image, is it not? If LP and moonlight are not a problem, what real benefit is there to doing narrowband imaging? Today, you can get some really nice tri-band or quadband filters for OSCs.  If I have a sensitive OSC in Bortle 5 or better skies, with no moonlight, and I can use long exposures, what is the benefit ? (Note the mono-vs-color camera argument is not relevant to my question so I'd rather not go into that discussion).

Also, I see filters applied to the same standard objects being imaged all the time: Orion Nebula, Heart Nebula, California Nebula, Soul Nebula, Veil Nebula and Wizard Nebula etc. - all extended objects - I rarely see extensive use of filters for e.g. planetary nebula or other small DSOs ...why is that...?

 

terry

 

 

 



#2 gatsbyiv

gatsbyiv

    Viking 1

  • *****
  • Posts: 976
  • Joined: 29 Mar 2009
  • Loc: Doylestown, PA

Posted 04 December 2020 - 10:52 AM

For common imaging setups, a faint emission nebula may emit on the order of 1-5 photons per minute per pixel, especially in the OIII and SII lines.  Due to the nature of ionized gases, every one of those photons has exactly the same wavelength.  Even for Bortle 5, the skyglow may be on the order of 100 or 1000 photons per minute per pixel, spread across the full visible spectrum.  So yes, you're capturing the nebula's photons with an R, G, or B filter, but it's mixed in with a hundred times more skyglow photons.  The shot noise of those skyglow photons makes it hard to see the additional nebula photons.  A narrowband filter gets rid of 95-97% of the sky glow photons while keeping every one of the nebula photons.  

 

With an R, G, or B filter, imagine 5 nebula photons + 100 skyglow photons.  With a narrowband filter, imagine 5 nebula photons + 5 skyglow photons. 


  • X-Lurker, NMCN, jdupton and 12 others like this

#3 astroian

astroian

    Ranger 4

  • -----
  • Posts: 318
  • Joined: 21 Sep 2016
  • Loc: Swindon, UK

Posted 04 December 2020 - 05:38 PM

If you want examples of small objects imaged with narrow band filters then have a look at my astronomy page:

https://www.astrobin...users/astroian/

I’m sure you will also be able to find many others.

For me the the narrow band filters allow me to image well into the lunar time of the month (depending on faintness of the target and angular distance from the Moon) They also allow me to image far into the summer months. Here in the UK it doesn’t get properly dark during summer, but with NB filters I can image until late June and restart around mid August.

And since planetary nebula shine mostly in the NB wavelengths there seems little point in collecting other light that will simply lower the contrast of what can be very faint structures.

Hope this helps.

Cheers,
Ian

#4 Peregrinatum

Peregrinatum

    Surveyor 1

  • *****
  • Posts: 1,560
  • Joined: 27 Dec 2018
  • Loc: South Central Valley, Ca

Posted 04 December 2020 - 06:25 PM

more efficient use of imaging time, and you are not limited by the moon cycle


  • epdreher likes this

#5 dan_hm

dan_hm

    Viking 1

  • *****
  • Posts: 726
  • Joined: 31 Dec 2012
  • Loc: New Jersey

Posted 04 December 2020 - 11:53 PM

I rarely see extensive use of filters for e.g. planetary nebula or other small DSOs ...why is that...?


Narrowband filters are definitely used extensively for planetary nebulae, probably even more than for emission nebulae and large SNRs. All of the best pictures of M57, for example, are done by deep Ha integrations.

#6 epdreher

epdreher

    Apollo

  • *****
  • Posts: 1,034
  • Joined: 12 Jun 2011
  • Loc: Texas Hill Country

Posted 05 December 2020 - 10:00 AM

more efficient use of imaging time, and you are not limited by the moon cycle

Not to speak of the structure/detail in NB that simply isn't there in LRGB for many objects.



#7 WadeH237

WadeH237

    Fly Me to the Moon

  • *****
  • Posts: 6,525
  • Joined: 24 Feb 2007
  • Loc: Ellensburg, WA

Posted 05 December 2020 - 11:19 AM

In that case, images may look a little different because they focus on certain "colors" more than others. However, that data is still present in an RGB image, is it not?

To answer your question in a single word: Contrast.

 

Even if you have pristine skies, there is still a good amount of broad band light in the sky.  When you image with narrow band filters, you get signal from your object, and next to nothing for the rest of the sky.  When you image with broad band filters, you get the same signal from your object (as you have surmised), but you also capture all that broad band light.

 

I know that conventional wisdom here is that you should do narrow band imaging to beat light pollution.  But the reality is, that narrow band images are improved by dark skies just about as much as broad band images.  If my imaging plan includes emission objects, then I absolutely use my narrow band filters, even at the Bortle 1 sites that I visit.  I don't consider it a wasted use of dark skies at all.


  • Andy Brown, LauraMS, Jon Rista and 4 others like this

#8 WadeH237

WadeH237

    Fly Me to the Moon

  • *****
  • Posts: 6,525
  • Joined: 24 Feb 2007
  • Loc: Ellensburg, WA

Posted 05 December 2020 - 11:21 AM

Not to speak of the structure/detail in NB that simply isn't there in LRGB for many objects.

Not really.

 

Filters work by blocking undesirable wavelengths of light.  They cannot produce signal from the object.  The reality is that LRGB filters will likely collect more signal from the target object than narrow band filters (including all the details).  The benefit is, as I mentioned in my above post, that the narrow band filters block more of the undesirable light, giving you greater contrast in your target.


  • Jon Rista likes this

#9 WebFoot

WebFoot

    Surveyor 1

  • *****
  • Posts: 1,827
  • Joined: 02 Jun 2005
  • Loc: Redmond, WA, USA

Posted 05 December 2020 - 04:06 PM

I image for fun, making pictures I enjoy.

 

Imaging with NB filters can, for me, enhance that fun; I like how certain objects (primarily emission nebulae) look in pure Ha, and how planetary nebulae with significant OIII look better with OIII data woven in; and I like how a Hubble palette image shows, in one image, where there are emissions in SII and/or OIII.  I like comparing different ways of processing, using different data-sets.  For instance, I really enjoyed comparing these images (click on each image for a comparison of processing the same object with different data-sets):
http://de-regt.com/A...GC7380.RCOS.htm

http://de-regt.com/A...GC6888.RCOS.htm

It's not that I think one way is better; it's just that it's fun to see how they're different

Adding a component of Ha to a galaxy image can bring out the HII star-forming regions much better:

http://de-regt.com/A...my/M74.RCOS.htm

http://de-regt.com/A...omy/M51RCOS.htm

 

One can perhaps have more success imaging in badly light-polluted skies with NB filters.  And one also can image closer to the moon, and with a little more moon, with NB filters (but my experience has convinced me that it is a fallacy to pretend that NB filters allow you to image with a huge moon in the sky, or with a smaller moon very close to the object, without adverse consequences).

Mark


Edited by WebFoot, 05 December 2020 - 08:40 PM.


#10 Topographic

Topographic

    Viking 1

  • *****
  • Posts: 806
  • Joined: 09 Jan 2019
  • Loc: Tarn et Garonne, France

Posted 06 December 2020 - 08:47 AM

The need for narrowband imaging decreases as your Bortle number decreases. Many people have said sky quaility has the most effect on imaging. I have B3 but I am going to experiment with F2 Ha imaging, L/OSC and Ha/OSC using a mono and  OSC cameras (ASI 183MM/MC). I may then do some traditional narrowband if the mood takes me.


  • elmiko likes this

#11 RogeZ

RogeZ

    Surveyor 1

  • *****
  • Posts: 1,632
  • Joined: 21 Nov 2004
  • Loc: Palm Beach Gardens, FL

Posted 06 December 2020 - 10:55 AM

Not really.

 

Filters work by blocking undesirable wavelengths of light.  They cannot produce signal from the object.  The reality is that LRGB filters will likely collect more signal from the target object than narrow band filters (including all the details).  The benefit is, as I mentioned in my above post, that the narrow band filters block more of the undesirable light, giving you greater contrast in your target.

Actually, while you are correct I also agree with Eric.

 

When doing LRGB the emissions lines are "blended" together and therefore the structure of each specific gas is hidden. When using NB you can isolate a specific gas and see its distinct structure. This is incredibly beautiful in PN.



#12 WadeH237

WadeH237

    Fly Me to the Moon

  • *****
  • Posts: 6,525
  • Joined: 24 Feb 2007
  • Loc: Ellensburg, WA

Posted 06 December 2020 - 11:06 AM

When doing LRGB the emissions lines are "blended" together and therefore the structure of each specific gas is hidden. When using NB you can isolate a specific gas and see its distinct structure. This is incredibly beautiful in PN.

Oh, I agree with this, too.

 

My point is that all of the signal with the detail is actually there with LRGB.  The reason that it's blended, as you say, is that there is also undesirable light that's competing with it.  That's my point.  It's subtle, but I find that, especially with troubleshooting or improving results, it helps me to understand what's actually happening.  A good understanding of this is a direct answer - along with the rationale - to the question in the title of this thread.



#13 sn2006gy

sn2006gy

    Apollo

  • ****-
  • Posts: 1,453
  • Joined: 04 May 2020
  • Loc: Austin, TX

Posted 06 December 2020 - 11:55 AM

 

My point is that all of the signal with the detail is actually there with LRGB.  The reason that it's blended, as you say, is that there is also undesirable light that's competing with it.  That's my point.  It's subtle, but I find that, especially with troubleshooting or improving results, it helps me to understand what's actually happening.  A good understanding of this is a direct answer - along with the rationale - to the question in the title of this thread.

 

I don't think anyone has to be convinced that Narrowband affords capabilities that survive moon and skyglow imaging.

 

However, In dark skies, it really comes down to personal preference. 

 

"undesirable light" to me is gradients - gradients can be diminished by NB imaging but nothing beats dark skies.  If the basis of this discussion is dark skies then I think we should speak to the merits of dark skies.

 

In dark skies, to some people "undesirable light"  or "blended light" (reduction of over all color complexity into shades of tri-color) could be the fact that SHO is a tri-color remix and while it creates contrast - it means that it is contrasted in ways that are also limited.  Take horsehead for example. It shines so bright in SHO and RGB  but i'd say it actually has less "undesirable light" in dark sky RGB than if taken with any NB filter because a lot of the natural color texture is lost when you make it glow hyper red in NB (Ha) and to be honest, I don't really think Ha or even an HaRGB does it justice because you lose out on the blue caverns and bright white  clouds and dark nebula dust in contrast with faint wispy glowing red dust.. 

 

As far as contrast is concerned, people shoot m42 and do all sorts of things to it to represent that contrast in many different ways. With and without NB.  So i'm not sure contrast in itself is the reason to say NB vs Broadband is better than the other outside the obvious fact that NB has benefits with moon/skyglow.

 

In the end, it's all personal preference.

 

So do what pleases you smile.gif

 

I look at these cameras, sensors, filters, scopes and everything as tools. There is no one best tool to get the job done. Use all of the tools you have and be creative. If all you like is SHO then shoot SHO, if all you want is RGB then shoot RGB - if you want to experiment, test, learn and be creative then use everything at your disposal.

 

Most importantly. Have fun! 

 

(and i guess tl;dr - the universe really is colorful up there regardless of how you choose to capture and express those colors)


Edited by sn2006gy, 06 December 2020 - 11:59 AM.


#14 WadeH237

WadeH237

    Fly Me to the Moon

  • *****
  • Posts: 6,525
  • Joined: 24 Feb 2007
  • Loc: Ellensburg, WA

Posted 06 December 2020 - 12:43 PM

It shines so bright in SHO and RGB  but i'd say it actually has less "undesirable light" in dark sky RGB than if taken with any NB filter because a lot of the natural color texture is lost when you make it glow hyper red in NB (Ha) and to be honest, I don't really think Ha or even an HaRGB does it justice because you lose out on the blue caverns and bright white  clouds and dark nebula dust in contrast with faint wispy glowing red dust.. 

It's all about the processing.

 

Your description of "hyper red" and "lose out on the blue..." suggests to me a processing choice (intentional or accidental).  You can certainly process HaRGB to maintain the qualities that you like, while enhancing the faint Ha emissions in a way that does not overwhelm the RGB.

 

Also, in terms of "undesirable light" that is definitely not sky gradients, Adam Block just released a new Horizons video that demonstrates attenuating star light to dramatically bring out the emission nebulae near Gamma Cassiopeia.  His end result was very nice, and the emission nebulae were not overwhelming at all.  Another example would be using Ha to bring out star forming regions in galaxies.



#15 sn2006gy

sn2006gy

    Apollo

  • ****-
  • Posts: 1,453
  • Joined: 04 May 2020
  • Loc: Austin, TX

Posted 06 December 2020 - 12:56 PM

It's all about the processing.

 

In that case the filters matter not.

 

Also, if the case for "undesirable light" is entirely circumstantial then it could be said with broadband you have much broader ranges of light to attenuate or enhance.

 

Not here to argue one is better than the other, but to show that in dark skies it really is personal preference.

 

There are "tricks of the trade" no matter which route you go. It's not uncommon to do HaRGB or any combination of any combination of SHO (Nii, Hb) LRGB to whatever effect you want depending on which filters you use.  You can shoot Ha and use it as a lume. I totally get that.



#16 sn2006gy

sn2006gy

    Apollo

  • ****-
  • Posts: 1,453
  • Joined: 04 May 2020
  • Loc: Austin, TX

Posted 06 December 2020 - 01:01 PM

The simple answer of "why do narrowband at all" is because of what it can do for LP and Moonglow.

 

The complex answer is because even without LP and Moonglow you can be creative with NB and the quality of NB data you get in dark skies is superior to that with moonglow or LP.

 

Dark skies improve your signal and reduce integration times for both NB and Broadband.



#17 hyperion0001

hyperion0001

    Sputnik

  • -----
  • topic starter
  • Posts: 45
  • Joined: 26 Sep 2007
  • Loc: Concord, MA

Posted 06 December 2020 - 03:31 PM

 

I know that conventional wisdom here is that you should do narrow band imaging to beat light pollution.  But the reality is, that narrow band images are improved by dark skies just about as much as broad band images.  If my imaging plan includes emission objects, then I absolutely use my narrow band filters, even at the Bortle 1 sites that I visit.  I don't consider it a wasted use of dark skies at all.

First I didn't look at everything on your site, but your Crescent Nebula and Melotte 15 images are outstanding.

 

I would agree with Byron though, that if you're not using narrow band imaging to beat moonlight or light pollution, it comes down to personal preference ... even in very dark skies, you seem to *like* the more targeted light from emission objects vs. broadband (which, given a very dark site, would be broadband emitted almost entirely by the object of interest and so completely appropriate and good for imaging). So in the end, it sounds to me like that is really the overriding factor - what do you want/like? 

 

The fact that the narrowband filter *also* improves contrast in light polluted or moonlit skies for the particular bands it lets through is secondary. Of course, it's a nice bonus, but the bottom line still seems to be one of preference - do you want to image the broadband signal, or do you want to image the narrowband signal? Either way, the end result is a unique representation of the object of interest based on the data you collect and how you want to process/display it.

 

My conclusion then is that "need" has little to do with it when it comes to narrow band imaging. However, I can clearly understand the rationale though: "since my sky is light polluted so badly or I can only image when the Moon is out, I might as well get a decent narrowband image rather than a poor broadband image in a comparable amount of imaging time."

 

terry



#18 WadeH237

WadeH237

    Fly Me to the Moon

  • *****
  • Posts: 6,525
  • Joined: 24 Feb 2007
  • Loc: Ellensburg, WA

Posted 06 December 2020 - 05:20 PM

Also, if the case for "undesirable light" is entirely circumstantial then it could be said with broadband you have much broader ranges of light to attenuate or enhance.

That is kinda my point.  It is totally circumstantial!

 

If I am going to image a globular cluster, then there is very little reason to use narrow band filters, regardless of the sky.  But if I am going to image M57 with its outer shell, then there is every reason to use narrow band filters - again, regardless of the sky.

 

As for my personal preference, I actually like to image galaxies, and I usually don't collect narrow band data on them.  When I image emission nebulae, I always collect narrow band data.  The less obvious point is that just because I have narrow band data, doesn't mean that I let the narrow band data swamp the RGB.  In my experience, narrow band data always improves emission nebulae, even if it's just enhancing the dim stuff while preserving the overall color balance.

 

Generically speaking, in most cases, narrow band data will get you the best contrast in the details, regardless of the sky.  Personally, my opinion is that, unless you are doing Hubble palette (or some other interesting mapping), the best results come from a blend of RGB and narrow band.  That is certainly not to say that it's wrong to do something else.  Unless you are doing pure science, this is a creative activity, after all.

 

If you get any takeaway from this, I would want it to be that all of the negative aspects of narrow band imaging that you've mentioned are processing choices and not something inherent in the use of specific filters.



#19 WadeH237

WadeH237

    Fly Me to the Moon

  • *****
  • Posts: 6,525
  • Joined: 24 Feb 2007
  • Loc: Ellensburg, WA

Posted 06 December 2020 - 05:29 PM

First I didn't look at everything on your site, but your Crescent Nebula and Melotte 15 images are outstanding.

Thanks!
 

I would agree with Byron though, that if you're not using narrow band imaging to beat moonlight or light pollution, it comes down to personal preference ... even in very dark skies, you seem to *like* the more targeted light from emission objects vs. broadband (which, given a very dark site, would be broadband emitted almost entirely by the object of interest and so completely appropriate and good for imaging). So in the end, it sounds to me like that is really the overriding factor - what do you want/like?

My assertion is that narrow band data always provides superior contrast to broad band on emission nebulae, and that this is true regardless of the sky conditions.  Yes, light polluted skies make the difference obvious, but narrow band images taken in dark skies still have superior contrast to RGB (or LRGB) images.  Where we might disagree is that I would challenge the statement that "given a very dark site, would be broadband emitted almost entirely by the object of interest".  When looking for faint details, the sky always glows, even at the darkest sky sites.

 

The what you want/like aspect comes in with how you use the narrow band data.  You don't have to use it as a sledge hammer.


  • rockstarbill likes this

#20 hyperion0001

hyperion0001

    Sputnik

  • -----
  • topic starter
  • Posts: 45
  • Joined: 26 Sep 2007
  • Loc: Concord, MA

Posted 07 December 2020 - 01:26 PM

 

My assertion is that narrow band data always provides superior contrast to broad band on emission nebulae, and that this is true regardless of the sky conditions.  Yes, light polluted skies make the difference obvious, but narrow band images taken in dark skies still have superior contrast to RGB (or LRGB) images.  

I would still rephrase your comment though, just to bring out a subtle point (we might as well talk subtleties because that's the fun of communication:) ... when you talk of superior contrast, you're still talking about preference, because the superior contrast is of features that you find interesting in bringing out - specifically - the features associated with those narrow band wavelengths. I think it would be inaccurate to say (nor are you saying) that the filter improves contrast for wavelengths outside of the pass-through wavelengths.

 

terry



#21 sn2006gy

sn2006gy

    Apollo

  • ****-
  • Posts: 1,453
  • Joined: 04 May 2020
  • Loc: Austin, TX

Posted 07 December 2020 - 01:37 PM

I would still rephrase your comment though, just to bring out a subtle point (we might as well talk subtleties because that's the fun of communication:) ... when you talk of superior contrast, you're still talking about preference, because the superior contrast is of features that you find interesting in bringing out - specifically - the features associated with those narrow band wavelengths. I think it would be inaccurate to say (nor are you saying) that the filter improves contrast for wavelengths outside of the pass-through wavelengths.

 

terry

That's my thought process.

 

One can favor what they get out of either methods, but that doesn't mean you can't process both to bring out features/contrast/color/texture.



#22 WadeH237

WadeH237

    Fly Me to the Moon

  • *****
  • Posts: 6,525
  • Joined: 24 Feb 2007
  • Loc: Ellensburg, WA

Posted 07 December 2020 - 04:00 PM

I would still rephrase your comment though, just to bring out a subtle point (we might as well talk subtleties because that's the fun of communication:) ... when you talk of superior contrast, you're still talking about preference, because the superior contrast is of features that you find interesting in bringing out - specifically - the features associated with those narrow band wavelengths.

Well, sure.  But for emission nebulae, the details are in the narrow band frequencies.

 

If your point is that narrow band filters don't help the contrast of objects that don't emit light in those specific wavelengths, then I agree.  I don't really consider that a preference as much as I consider it a property of the target object.  The selection of a target object is, of course, a preference of the astrophotographer.

 

Where I don't agree is that with objects that emit light in narrow band frequencies, you will always get superior data to work with, with (properly exposed) narrow band filters.  It's fine to collect broad band data as well, if there are multiple objects of interest, or if you want to get the best stars.  You can blend that data with the narrow band data to taste.  If you choose to not emphasize those narrow band details in your processing, that is certainly a valid choice.  But you don't really get a technical benefit in an emission nebula by skipping narrow band (assuming that you are set up to collect it).  That would just be leaving S/N from the object on the table.


  • hyperion0001 and epdreher like this

#23 sn2006gy

sn2006gy

    Apollo

  • ****-
  • Posts: 1,453
  • Joined: 04 May 2020
  • Loc: Austin, TX

Posted 07 December 2020 - 05:49 PM

Well, sure.  But for emission nebulae, the details are in the narrow band frequencies.

 

If your point is that narrow band filters don't help the contrast of objects that don't emit light in those specific wavelengths, then I agree.  I don't really consider that a preference as much as I consider it a property of the target object.  The selection of a target object is, of course, a preference of the astrophotographer.

 

Where I don't agree is that with objects that emit light in narrow band frequencies, you will always get superior data to work with, with (properly exposed) narrow band filters.  It's fine to collect broad band data as well, if there are multiple objects of interest, or if you want to get the best stars.  You can blend that data with the narrow band data to taste.  If you choose to not emphasize those narrow band details in your processing, that is certainly a valid choice.  But you don't really get a technical benefit in an emission nebula by skipping narrow band (assuming that you are set up to collect it).  That would just be leaving S/N from the object on the table.

I guess i'm not happy with the language. I just can't agree anything is superior to one another, but both can be enhanced in their own unique qualities.

 

If you like NB better, that's awesome.

 

I'm actually getting a lot of requests for broadband images because there is a lot of images we see all the time in SHO but actually look more vibrant, rich and detailed in broadband.  On astrobin you'll see 400 versions of the NB target, but only a handful of the broadband and when you see the broadband you're often taken back "i didn't know it was that rich!"

Doesn't distract from one image or another... so i guess  i refuse to accept one is superior or even has to be superior to another. The notion and the language thereof is not needed.  It's like arguing about beaches or mountains and which one is best or which one is more beautiful. I mean, its obviously mountains but... laugh.gif


Edited by sn2006gy, 07 December 2020 - 05:50 PM.


#24 rockstarbill

rockstarbill

    Cosmos

  • *****
  • Posts: 7,973
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 07 December 2020 - 09:11 PM

Because it looks cool. :p


  • epdreher likes this

#25 Ron (Lubbock)

Ron (Lubbock)

    Viking 1

  • -----
  • Posts: 978
  • Joined: 17 Aug 2012
  • Loc: Pittstown, NY

Posted 07 December 2020 - 09:14 PM

For the owner of a 12 inch SCT, here's a little secret: narrowband = small star size!  With my C11 Edge, I have found a great difference in star size with LRGB vs. narrowband, and I am still debating the reason(s).  It is just the lower intensity of the light, or is it also the narrow spread in wavelengths? 

 

get.jpg?insecure

 

With my 11" SCT, I have found that objects that have both H-alpha and O-iii emissions (M27, M76, veil nebula) turn out great in narrowband.  Extended objects that are mostly red are less impressive, and I have gotten better results by just shooting H-alpha and making a black & white image out of it.

 

get.jpg?insecure




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