New Approach to the Combination of Broadband/Narrowband Data by Vincent Paris

[Dimitris Platis]

I have been experimenting with this http://pixinsight.com/tutorials/narrowband/theory/en.html and I am having a huge problem with the final stage of exchanging the a/b components of CIE.
The channel extraction and channel combination module appears not to work properly and result in a final image with greenish hue even if I just extract and recombine the same image.
It appears to work best when both images are stretched.

I have tried to match Luminances as best as possible.

Any ideas???

[Ioannis Ioannou]

Dimitri,

I never managed to use the original article as it was (which of course may be just my fault), but mind that there is an implementation of the theory in the HaRGB script. Look http://pixinsight.com/forum/index.php?topic=3446 and download the latest version, around middle of page 6 so far. (Harry's correction for 1.8 http://pixinsight.com/forum/index.php?action=dlattach;topic=3446.0;attach=5968 )

[Dimitris Platis]

I do have the HaRGB/LRGB/SHO scripts...just thought I give that method a try.....I like the control you have over the process...

[troypiggo]

I assume those 3 scripts you mention are the ones that come by default with PI?  The ones based on the "AIP" method?

The HaRGB Combination script Ioannis is talking about is a different method and gives different results in my limited experience.

I agree with you about control etc.  I tried to follow that article and posted here several times trying to get some clarification on the methodology so I could repeat it.  I haven't had what I consider a definitive response, and people just point to the HaRGB Combination script.  Based on what I understood as "Vicent's Method", that script doesn't follow it to the letter.

See my previous question: http://pixinsight.com/forum/index.php?topic=4120.msg28692#msg28692

[Ioannis Ioannou]

Yap, the HaRGBCombination is a different script, just give it a try.

As to if it is following the method to the letter.....hmmm, I'm not sure if the method is so standarized in the first place anyway In my opinion, what the script is missing is a way to show and control the intermediate Continuum image ( anyone with free time ?  ), otherwise it is pretty much complete.

[Dimitris Platis]

How does the script manage the final stages of exchanging the a/b components?
Is there any way to improve the manual process by borrowing insights from the script?