When extracting the Lum component from an RGB image, should I use CIE Lab mode?
The concept of "color mode" does not exist in PixInsight, mainly because it is an incorrect/unrigorous way to toy with color spaces. To extract the L* component (lightness) of the CIE L*a*b* color space as a grayscale image, you can use the ChannelExtraction tool. Most of the time you don't need to do explicit component separations, since many tools already work in the appropriate color spaces behind the scenes, or allow you to apply a process separately to the lightness, luminance or chroma components. The CurvesTransformation tool allows you to apply custom transformations to the individual components of the RGB, CIE L*a*b*, CIE L*c*h* and HSV spaces. You can also visualize a variety of components of any color image in different ways. In all cases the components are computed on the fly automatically.
I'm reading that LAB mode non-linear
The
CIE L*a*b* color space is nonlinear by definition. It is a device-independent reference model of human color perception. Its three coordinates separate color values into lightness (L*) and chroma (a*, b*) components. The lightness component attempts to model the visual perception of brightness. See the referenced
Wikipedia entry for more detailed information.
