[Digikam-devel] [Bug 142055] Which whitebalance is used

[Guillaume Castagnino]

------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
         
http://bugs.kde.org/show_bug.cgi?id=142055         




------- Additional Comments From casta xwing info  2007-08-25 19:33 -------
Hi

Peter, I think you are confusing RAW white balance and RGB white balance.
Currently, the digiKam "White Balance" acts on the RGB space, so AFTER raw conversion.
So in fact, temperature provided within the WB tool is relative to the original WB of the picture (which is not necessary neutral). So there is a biais in temperature interpretation in this tool.

Currently, white balance during RAW conversion is only handeled for automatic and camera embeded white balance (in dcraw setup in digiKam configuration box).

By the way, I agree that the values are quite strange. Having a "Neutral" WB (that provides RGB multipliers = (1, 1, 1)) at 4750K and 1.2 green multiplier sounds quite strange to me.
IMHO it would be far less strange to have the neutral point at standard CIE D65 with 1.0 green multiplier...

To change that, here[1] is a proposal to change the way of calculating the RGB multipliers from the pair (temperature, green multiplier). I have ported the algorithm from ufraw, that set the neutral white point at D65 illumination. That sounds to me far much logical since it's a widely spread standard.
Moreover, it does not need anymore the enormous blackbody.h matrix.
At your appreciation, but I think it could be a good thing to apply to digiKam ;)

[1] http://casta.nerim.net/digiKam/digikam-kde3-WB-ufraw-multiplier.patch

Regards