Target Brightness

John Binney

When stretching you recommend a maximum of about 0.8 to allow the image to be nicely colourised. However, at the other end of the scale it appears there should also be a target for maximum sky brightness. It is my experience that if the sky brightness is too high you seem to struggle to effectively handle noise reduction. Even using NXT if the sky is too bright you reduce small scale noise but you appear to be left with a larger scale mottled effect that I am finding impossible to remove. Maybe just avoiding black clipping is a better target criteria than a brightness limit?

BlockHead

The background brightness is an independent adjustment. Typically you set the black point to a value that is just to the left of the histogram "mountain" of the faintest recorded values in an image. There isn't really a well-defined value for this since it really depends on the data and the object. 

This doesn't have anything to do with the colorization at the high end though- which is a different effect. 

Large scale mottling can be a symptom or other possible issues that deal with both acquisition of the data, sensors, and processing (sometimes). Large scale features are NOT noise. This is a key insight. The small scale fluctuations are... but larger scale mottling indicates correlations between pixels... and you need to determine the source. 

-the Blockhead