Agreed, it is a fact that these image processing operations are not science,
and what works for one image may not work for another. So we do want to
have a bag of them.
Summary of what we have so far:
1. Manual contrast enhancing:
out = (in - average) * contrast + average;
out = pow(in * pow(pivot, 1.0 / contrast - 1.0), contrast);
Maybe there is a "better" formula?
There are probably many other formulas so I suggest we have both. As more
formulas are suggested by people, we can expand. There is no one best way
with these things anyway.
2. Automatic stretching by remapping the old range to a new one :
I already have these methods:
- auto contrast: map current range to min-max range.
- modified auto contrast: explicitly specify which range to map to
min-max range,
and what to do with values
outside of that range.
- histogram specification: adjust contrast so that it matches the
contrast of some
other image given as input.
3. True histogram equalization:
Got this one.
4. Tone mapping - some conversion of HDR to [0,1] in a perceptually
pleasing way:
- Could do some research on this later.
SS
_______________________________________________
Oiio-dev mailing list
[email protected]
http://lists.openimageio.org/listinfo.cgi/oiio-dev-openimageio.org
