HaLRGB

[Nigel Ball]

Morning all

I've already posted an image of the Horsehead nebula in the Gallery section

Last night I collect some Ha data to try and bring out the faint nebulosity.

My question is how do I add this in. Ha is in the red region so do I combine with the red channel? or use the Ha data as Luminance? If so some guide on how to 'add' channel data together would also be appreciated. I haven't used PixelMath yet and I assuming it is this tool I'll need to become familiar with.

I appreciate this is a fairly fundamental question but it is sometimes the basic stuff that holds us up

Thanks in advance

Nigel

[Philip de Louraille]

Hi Nigel,

That topic was already discussed here:
http://pixinsight.com/forum/index.php?topic=2298  and an older discusion is here as well
http://pixinsight.com/forum/index.php?topic=562.0

In the first post, Jack explains how to create two new master channels which are subsets of the Red master channel and the Ha master channel. I'm explaining what I have read because it is not clear (sorry Jack!)
Suffice to say that he shot with 4 filters: Ha, R, G, and B.  (notice no separate L.)

So say your narrow filter bandpass is 5nm
You create a new red channel where you remove the contribution from Ha

new_red = m_red - m_Ha

new_Ha = m_Ha - 0.05 * new_red

Finally a mix
red_channel = 0.6 * new_Ha + 0.4 * new_red, and you can play with the 0.6 and 0.4 coefficients.

Now get you color image with assembling
red_channel, m_green and m_blue

Well! I just found a new tutorial from Vicent Peris here about this
http://pixinsight.com/tutorials/narrowband/theory/en.html


There is also a tutorial about PixelMath but do not read the whole thing, it will make your head explode.
It is a complete tutorial with examples.
Skip to the "Merging Images" part. Near the end.

[Nigel Ball]

Thanks Philip

I always forget to search the forum 

Nigel

[Emanuele]

maybe I am missing something but to do the initial  M_Red - M_ha , do we need the images to be linear ?

[Philip de Louraille]

Yes. We are dealing with the Masters, just out of integrations.

[Emanuele]

Thanks Pde.

I just tried this method on an image I am working on with Ha, R, G, B.

The problem that I have with this method is that I can't really squeeze every bit of information from the Ha channel, since it is blended with the Red one and combined right away with the G, and B channels.

By doing an Ha-RGB combine using the LRGBCombination process, and treating the Ha data as the L, one could squeeze every bit of data there is in the Ha frame.

The drawback is that the result using LRGB (Ha as L) is a pink image.

Any way to avoid that?
E.