Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

Separate luminance in nb


Arc minute

Recommended Posts

Hi all

I'm just starting the plunge to the deep end of mono and narrowband imaging. I have read the available info from here (SGL) and other sources but the findings seem to be rather inconclusive.

So here's the thing.

Some of our fellow imagers seem to be boosting the nb signal by using luminance channel. Methods for LUM vary from

  1. using just the Ha,
  2. using all the nb channels combined or
  3. using lum filter or better yet using ccd grade light pollution filter. 

I'm sure there are other methods as well..?

From what I gather, there are some "drawbacks" to each

  1. does't represent all the signal
  2. just uses the same total signal ratio (does not actually increase the S\N that is already there in each of the channels) --> there's no new "information" gained from here
  3. star size is/might be different than in nb channels?

So what is the "industry standard" method to be used or is there any? According to my logic the only method that actually could add information is the last one. The problems caused might be fixable with post-processing?

I am probably missing something here, but what?

I should also note that ccd users might use different hw binning. Since cold cmos cameras (which yours truly has) do not support hw binning and low read noise negates the benefits somewhat anyhow, I would consider this to be a non-factor.

Regarding the option 3. I would think that this same drawback would affect all those that image in LRGBHa...

 

Link to comment
Share on other sites

3 hours ago, Arc minute said:

Thanks Olly and Dave, interesting points!

Olly, do you also mean that you don't use lum channel technique at all in nb? Even either of the first two methods?

 

Thanks guys!

I don't do pure NB imaging, so I can't comment. Because I work from a dark site I'm wedded to natural colour imaging enhanced (often very considerably) by narrowband. In my situation I'm reluctant to use Ha as a luminance layer simply because it's false. Ha is deep red and luminance is full visible spectrum. However, on Ha-dominated targets I will sometimes allow about 15% of the Ha into the luminance channel.

From what I read NB imagers will sometimes use Ha as Luminance applied to a bi- or tri-colour palette because it contains the strongest signal and the highest structural contrasts.

I can't see any point in combining three NB channels and using them as L because that is exactly what you have embedded in the tricolour NB image to start with. (Extract a synthetic L from that and you have a clone of the proposed combined filter L channel.) What you might do is replicate the procedure often used in one shot colour or RGB-only imaging and that is to extract a synthetic luminance, process that to bring out sharpness, detail and contrast, then apply that to the original colour layer which had been processed for colour intensity and low noise at the cost of resolution. I don't know if NB imagers ever do this, though. 

Planetary and lunar imagers sometimes use red as luminance because it is the least affected by the seeing, whch matters at their very high levels of resolution.

Olly

PS I was called for domestic duties by madame before fully explaining my original answer on L-filter luminance for NB imaging. Sadly, if you compare an L-filter image of an emission nebula with an Ha-filter image you'll see why you would never want to use the L filter data! It has big soft stars, low structural contrasts, skyglow effects and, in short, looks like a load of mush! We broadband imagers dream of combining the NB look with full spectrum colour. It remains, for the moment, at the end of the rainbow!

Link to comment
Share on other sites

In a similar vein, I've now got some reasonable Ha data of M27 (from last night with the moon around) and some poorer RGB data from last weekend that I'd like to use to create a LRGB of this emission nebula.

Would it be best to create a synthetic lum channel from the Ha and then proceed as normal for LRGB? How do I create a synthetic lum?  Any advice on how best to do this would be very much appreciated.

Regards, Andy

Link to comment
Share on other sites

On 10/09/2017 at 10:34, fireballxl5 said:

In a similar vein, I've now got some reasonable Ha data of M27 (from last night with the moon around) and some poorer RGB data from last weekend that I'd like to use to create a LRGB of this emission nebula.

Would it be best to create a synthetic lum channel from the Ha and then proceed as normal for LRGB? How do I create a synthetic lum?  Any advice on how best to do this would be very much appreciated.

Regards, Andy

What I would do is process an RGB image first. Don't try to drag out the faint details, you'll soon be flogging a dead horse. Go for a nice neutral background sky and try to keep the stars as small as possible. (A synthetic lum would be this RGB image with the colour discarded. I'd ignore that option for the moment.) Then...

Process your Ha as a hard stretch. Drag out everything you can and don't panic if there is noise in the background sky. Just be sure that your Ha background is a tad darker than your RGB background and then the Ha background will not appear in the final image using this method. 

Select and copy the Ha image. (CtrlA CtrlC in Ps.) Split the channels in the RGB image and paste (CtrlV) the Ha on top of the red channel.

Go to Layers and change the blend mode to Lighten. Now the Ha layer will only be applied to the red where the Ha is brighter. If you want to get clever you can use curves to tweak the Ha while it is in situ over the red. Just don't let the Ha brighten the background sky in red or you'll wind up with a red background sky. Flatten and recombine the RGB in RGB mode.

Olly

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.