Finding noise?

I have read the tutorial pdf's trying to find out how to interpret noise with FRV, but I'm not getting it. It looks as though using "D" (Display fine details) helps with noise definition...the more red dots, the more noise???  Thanks.

Dear Sir:

When comparing 2 shots of the same scene having the same lightness, yes, in most cases more red dots means more noise.

flycaster's picture

OK, got that for "most cases."  But, when would the dots not be indicating noise.  Thanks.

Dear Sir:

When very fine details are present (with the size of less than 3 pixels) dots represent those details.

flycaster's picture

"When very fine details are present (with the size of less than 3 pixels) dots represent those details." So, how does one discern between noise and detail...that is, how do can you tell that a dot is less than 3 pixels...other than magifying to about 500% and count pixels?

Unfortunately, there is no way to separate noise (non-existing details) from real details based on raw data only.

If it were possible, then noise reduction would be much easier and would not remove small details.

--

Alex Tutubalin/FastRawViewer team

Add new comment