Today’s Question: What are the causes of color noise and luminance noise? I understand noise in general from high ISO but why are there two different kinds of noise? And why are each corrected differently in Lightroom Classic?
Tim’s Quick Answer: Color and luminance noise are both caused by the same general issues. They are just exhibited differently in an image. Separate corrections for color versus luminance noise are provided in most software for optimizing photos because each should be dealt with a little differently.
More Detail: You might say that “all noise is noise”. Noise in a digital image is represented by inaccurate variations in pixel value at the level of the individual pixel. These variations can be caused by a number of factors, including strong amplification due to a high ISO setting, long exposures, and heat buildup in the camera in general, among other causes.
In most cases a digital camera records information for three individual color channels (red, green, and blue). Each of those channels is really a monochromatic channel consisting of only luminance values. In other words, it would be fair to say that in most cases noise in digital photography is all luminance noise.
However, the three color channels are combined to create a full-color image, and variations in noise between the channels contributes to color noise. Again, noise is simply the variation in values at the pixel level, so the key question is whether the inaccurate variations in pixel values are more of a tonal difference with similar colors versus differences in color values (and possibly tonal values as well).
More to the point, color noise and luminance noise need to be dealt with somewhat differently. Color noise can be averaged out relatively aggressively, blending color values into the area surrounding each pixel without too much of a problem in terms of visual artifacts or loss of color. However, with luminance noise you need to be much more cautious. With just a little too much luminance noise reduction the variations in tonal value in the overall photo will be averaged out too much, resulting in a potentially significant loss of sharpness and overall detail.