I have a canon 600d that says it uses sRGB. How is this even possible without having the same spectral sensitivity spectrum as a standard observer? It is possible to define the matching functions of the camera for a set of known display primaries (as the sRGB ones) when the sensitivity spectrum is known. This I believe, and please correct me if I am wrong, means that I know where the locus of the camera lies on the sRGB colorspace and consequently on XYZ.

But I do not really know the transformation between the camera space and sRGB. Nor do I know the type of transformation.

My question is: what are the calculations involved in transforming the outputs of the camera sensors to provide an image in the sRGB space?


The camera sensor does not have a color space that allows going back and forth to XYZ since it does not have the same sensitivity curves as the human eye (Luther - Ives condition). The best that can be done is come up with a transform matrix that minimizes the sum of the errors in LAB for the set of standard colors (Gretab -Macbeth). I believe this process in defined in ISO17321. Camera manufacturers may choose a different matrix to produce colors that they believe are more pleasing.

  • Ok, this makes perfect sense to me. Thank you. Just to complete the answer, the Luther-Ives condition says something like "a sensor can distinguish exactly the same colors as the average human eye, if and only if the spectral responses of the sensor can be obtained by a linear combination of the eye cone responses". – strangelyput Apr 16 '15 at 9:06

You're right; the camera has its own native color space. When a camera is said to use or support sRGB (or Adobe RGB, as many also do), that means that it has native support for transforming its raw sensor data into that color space.

When you use an out-of-camera RAW converter, like Lightroom or Darktable, that program needs to know about your individual camera so it can do the conversion properly. This can either be with a complete camera profile, or through a much simpler but usually perfectly adequate "color matrix" — see this post on creating one for command-line converter dcraw (which is used as the internal image-load engine for many other RAW conversion tools).

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.