Why don’t camera sensors and displays use CIE XYZ directly instead of RGB?
Asked 9/12/2015
6 views
2 answers
0
If CIE XYZ is meant to model human color vision better than RGB, why don’t camera sensors capture directly in XYZ rather than using RGB-style color filters? And if XYZ is preferable, why don’t displays use XYZ primaries too? Since RGB can only represent a subset of visible colors, I’m trying to understand the practical reason cameras and screens are built around RGB instead of XYZ.
Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0
Photography Stack Exchange contributor
10y ago
2 Answers
3
Currently XYZ filters are produced using thin film technology. It is not very cheap, and not very suitable for multi-megapixel sensors. It also results in somewhat spiky spectral response curves, especially problematic when the light source has spiky spectrum, like fluorescent tubes and some flashes. Yet another reason would be higher noise levels, as XYZ filters allow for very little transmittance in certain areas of the spectrum.
For displays, the problem is also the light source. To apply XYZ filters in an efficient manner, the light source needs to have very smooth spectral power distribution, and close to that of a black body at about 5500K, which is not an option currently.
Originally by user28381. Source · Licensed CC BY-SA 4.0
user28381
10y ago
0
Generated from our catalog & community — verify before relying on it.
CIE XYZ is mainly a mathematical color space for describing human color perception, not a set of physically practical sensor or display primaries.
For capture, camera sensors record light through real filter dyes whose spectral responses are determined by manufacturable materials. Those filters are then transformed by profiling/raw conversion into standard color spaces. Direct XYZ-style filters are not practical because ideal XYZ responses are not based on real, clean primaries; XYZ includes “imaginary” components used to avoid negative values in color matching math. In practice, making sensor filters that behave like XYZ is difficult, can have poor transmission, higher noise, and problematic spectral responses under spiky light sources.
For displays, the same issue applies: displays need real light-emitting primaries. XYZ is not a practical set of display primaries, and efficient XYZ-style filtering would also require a very smooth light source spectrum.
So cameras and displays use real-world RGB-like primaries because they are physically achievable and efficient. Color management then converts between a device’s native responses and spaces like XYZ, sRGB, or Adobe RGB. In other words, cameras don’t need to capture XYZ directly; they capture with practical filters and map the result into standard color spaces later.
Recommended products
UniqueBot
AI10y ago
Your Answer
Related Questions
Why do displays use RGB subpixels instead of a pixel that can emit any wavelength?
Why is the CIE 1931 chromaticity diagram non-linear, why doesn’t it reach y = 1, and how was it derived?
Why would image processing use LUV instead of RGB?
Why are red, green, and blue used as the primary colors of light?
Why are digital images represented in RGB instead of wavelength values?