Doug Kerr
Well-known member
The color of light is wholly defined in terms of human perception. If we see two instances of light side-by-side, and they look as if they are the same color, they are the same color - by definition.
The color of an instance of light is determined by its spectrum - the "plot" of the distribution of power in the light over the different wavelengths in the "visual range". A certain spectrum will always have a certain color. But there can be an infinity of other spectrums that also have exactly that color - a situation called metamerism.
We would like it if our digital camera sensor gave the same report of the color of light upon it - by way of the output voltages of its three types of photodetector - for light of any given color, regardless of its spectrum. But this is not ordinarily quite so, for various reasons - including some intentional design compromises. This situation is said to be metameric error on the part of the sensor. It cannot be corrected during the processing of the sensor output data, or later.
However, by "crafty" choice of some details of sensor data processing, we can minimize the average amount of metameric error for some chosen collection of colors.
And there is an objective "scoring system" that describes the amount of "residual" metameric error, allowing us to compare the performance of different cameras in this regard.
This fascinating story is told in my new technical article, "Metameric Error in Digital Photography", available here:
http://dougkerr.net/Pumpkin#Metameric_Error
Enjoy!
The color of an instance of light is determined by its spectrum - the "plot" of the distribution of power in the light over the different wavelengths in the "visual range". A certain spectrum will always have a certain color. But there can be an infinity of other spectrums that also have exactly that color - a situation called metamerism.
We would like it if our digital camera sensor gave the same report of the color of light upon it - by way of the output voltages of its three types of photodetector - for light of any given color, regardless of its spectrum. But this is not ordinarily quite so, for various reasons - including some intentional design compromises. This situation is said to be metameric error on the part of the sensor. It cannot be corrected during the processing of the sensor output data, or later.
However, by "crafty" choice of some details of sensor data processing, we can minimize the average amount of metameric error for some chosen collection of colors.
And there is an objective "scoring system" that describes the amount of "residual" metameric error, allowing us to compare the performance of different cameras in this regard.
This fascinating story is told in my new technical article, "Metameric Error in Digital Photography", available here:
http://dougkerr.net/Pumpkin#Metameric_Error
Enjoy!