11,96
12/1/2021
·
 |
|---|
I'm new on my narrowband journey and am looking for guidance on how best to combine the narrowband captures with LRGB to maintain star colour and keep a natural look to the images. I've only taken a handful of narrowband images so far as I only started using a mono camera and filters at the beginning of November and clouds have been everywhere this month. My only real attempt is:  HaOIIILRGB Crescent Nebula Any thoughts/tutorials that you think would help would be greatly appreciated. N.B. I'm comfortable with using PixelMath in Pixinsight to do some of this stuff, but I'm not convinced by some of the formulas people are suggesting on YouTube etc..For example: the usual ((Ha*Redbandwidth)-(Red*HaBandwidth))/(RedBandwidth-HaBandwidth) doesn't seem to take into account the fact that Red subs are typically much shorter than Ha subs. I may be missing something here. |
8,84
12/2/2021
·
·
3
vind-ik-leuk
|
|---|
The easiest way is to use the NBRGBCombination script in PI. Use the maximum bandwidth (200 nm) for best results in the RGB section of the dialog window. Fill-in the other sections and you're done. Sometimes the script might cut off the low end of the histogram (rare event but did happen) in which case you need to use the LinearFit process between R and Ha, G and OIII and finally between B and OIII. Use PixelMath to combine the fitted Ha and OIII frames to R, G and B. Note that the fitted OIII frames are different (potentially) when G or B are used as reference frames. As for SII, add it to Ha and you're done. The modified R+Ha, G+OIII and B+OIII can then be used as normal in RGB composition.
|
3,97
12/2/2021
·
·
1
vind ik leuk
|
|---|
PixInsight does have a script called NBRGBCombination (script > utilities). It worked reasonably well for me for Ha overlay on M33. For the galaxy, I then also applied PhotometricColorCalibration to help insure a somewhat natural result. If you are looking specifically to apply RGB stars to an otherwise narrowband emission nebula, a common approach is to use the Starnet++ process to remove the narrowband stars and blend the RGB star layer back in as an overlay.
|
|
11,96
12/2/2021
·
|
|---|
andrea tasselli: Andrea, thanks for all the detailed comments. I followed all of your comments until I got to the bit above ... what exactly are you hinting at here? I have been varying the amount I apply the OIII to the blue and green channels, but I've been making up the factors myself; I'm not sure if there is a more empirical way of doing it? |
|
8,84
12/2/2021
·
·
1
vind ik leuk
|
|---|
Andy, When I said the above I meant that the process of applying LinearFit is to renormalize the OIII (or Ha) frame so that it has the same mean background and background distribution of the reference frame, say G (or B or R for the Ha frame). Once this is done the fitted OIII frame is numerically different from the original OIII , not only bacause of the LinerFit process has been applied to it but also because the reference frames (either G or B) are different, even more so if such frames are coming as part of a RGB/LRGB acquisition technique. Once this is done then you add them up together and rescale the result in PixelMath, e.g., G+(LinearFitted)OIII. As simple as that. |
