LRGB combination with LRGBcombination or Channel combination?

M

magnusl

Well-known member
Hi!

I'm processing my first LRGB image, and have found some different advice that makes me wonder - maybe it does not matter, but if it matters, it would be good to know. Basically, some sources, like Inside Pixinsight , the book, version 1, says to combine RGB into one chrominance image and the use the LRGBcombiation to combine with Lum.

However, LightVortexastronomy (https://www.lightvortexastronomy.co...-combination-and-further-post-processing.html) says to use ChannelCombination, if I read it right.

My question comes from having L with bin 1x1 and RGB with bin 2x2. I understand that StarAlignment adjusts my RGB to fit the Lum, if I align them with a Lum as master. However, the Pixinsight book (ver 1) says nothing about StarAlignment in relation to LRGBcombination.

So: does not LRGBcombination rely on the images being registred to each other?

And: is there a reason to use one of these rather than the other? LRGBcombination rather than ChannelCombination? Or are the the same?

(I hope this makes sense)

Best,

Magnus
 
LRGBCombination needs images of the same dimensions, as does ChannelCombination. so if your RGB is bin2, you need to register it so that it's upscaled to the Lum's size.

channelcombination is used to assemble an RGB image from 3 mono masters, usually when those images are in the linear stage. it has other modes as well corresponding to the different color models but it probably doesn't get used that much in practice.

LRGBCombination is able to assemble LRGB images from 4 images (L, R, G, B) or is able to replace the L* (perceptual L) component of an RGB image with a given L image. this is generally how people use it - load an L image, configure the sliders and apply the process to an RGB image. one big difference is that LRGBCombination expects the RGB and L images to be stretched, and also that their brightness is roughly comparable.

anyway what you describe in your first paragraph (i think you are talking about Keller's book) is what I do.

i might use ChannelCombination to create the RGB image from my individual masters, but PixelMath also works.

rob
 
If you combine images using any process/script they have to be registered first to a reference image
(rob beat me to it LOL)
 
Magnus - another technique to consider is to use the Channel Combination Tool and tick CieL*a*b mode with the registered luminance image selected and apply this to the non-linear RGB master, with the RGB master either masked (to control the degree of luminance addition in an iterative manner) or unmasked.
 
You must log in or register to reply here.
Back
Top