Adding Luminance to SHO Narrowband

This is what I initially thought but isn't it the same thing when shooting RGB?

but if the luminance data is just layered on top to add sharpness & reduce noise, aren't the pixels of the luminance layer ignored when there is no signal in the RGB channels below? Maybe I'm misunderstanding something here.

Luminance is essentially just the sum of the RGB broad bands and so can faithfully represent the overall brightness in visible light of  objects like galaxies for example.    Narrow band on the other hand  - after subtracting the underlying broad band component - just comes  from very particular transitions of specific excited gas ions  -  usually just HII and SII at very narrow wavelengths in the far red and OIII in the blue/ green.   Luminance would not faithfully reflect the sum of these specific signals because it is broadband and 99.99% of luminance comprises all the wavelengths other than those few narrow lines of interest that you are trying to image.  So it would just bury the NB signal.  

Sometimes –because for many NB objects the hydrogen alpha is by far the strongest narrow band signal, shows more interesting  structural detail and is often spatially coincident with the other signals – hydrogen alpha is used as a sort of 'luminance'  in images combining H alpha with the other NB signals.

Tim

At least in PixInsight, LRGB combination is not a simple multiplication of RGB by L. This is described by Juan here:

"The "intrinsic" L in your RGB image will basically be replaced with the L image. You should stretch L first, then stretch RGB to achieve similar levels in its luminance (watch the L display channel, or select the RGB+L pixel readout mode and compare readouts, or extract L from the stretched RGB and compare statistics). The LRGBCombination tool also has a luminance midtones transfer function to fine tune the adaptation. The goal is to achieve an optimal adaptation between luminance and chrominance. Too much luminance means more chrominance noise to achieve the required color saturation. Too much chrominance means more noise in the luminance."

The pixels on our screens are basically RGB. The RGB image is converted to L*a*b*, and L* in the CIELAB space basically approximates human perception of lightness. L* is simply a nonlinear mathematical combination of RGB  (through an intermediate linear conversion to XYZ space).

Replacing the L* in the converted image with a version of L* calculated from the luminance data and then converting back to RGB space simply means that high SNR data collected using the luminance filter is selectively distributed between the RGB channels in a way that maximizes our perception of noise and detail. Note that this "replacement" means that the intrinsic L from the RGB image and the L from the luminance data must be matched and correlated. This is at least possible in broadband imaging because the luminance filter is effectively spanning the bandwidth of the RGB filters, but will not be the case, usually, with narrow band data. You could do an LRGB combine in PixInsight with any L mathematically, but the results are only meaningful if the RGB and L are correlated.