CIELUV is a three-dimensional color space standardized in 1976 by the International Commission on Illumination (CIE). It expresses a color through three coordinates: , representing lightness, and and , representing chromatic components relative to a specified reference white. CIELUV was designed so that equal distances between coordinates correspond approximately to equal perceived color differences. It is derived from CIE XYZ tristimulus values and has been used particularly in the specification and analysis of colored lights and electronic displays.
Background and development
CIE XYZ, established in 1931, provides a standardized numerical description of color based on a standard observer. Its coordinates are useful for colorimetry and additive color calculations, but distances in XYZ do not directly represent perceived differences: an identical numerical displacement can appear conspicuous in one region and barely noticeable in another.
The CIE developed more uniform coordinate systems to address this limitation. Its 1960 uniform chromaticity scale used coordinates called and . CIELUV incorporates a modified version of that chromaticity diagram, together with a nonlinear lightness scale.
In 1976, the CIE recommended both CIELUV and CIELAB as approximately uniform color spaces. They share the same lightness coordinate but use different transformations for their chromatic components. CIELUV is closely associated with additive light and display applications, whereas CIELAB became especially widespread in surface-color measurement and industrial color control. This distinction reflects common practice rather than a restriction on what either space can represent.
Coordinates and interpretation
A CIELUV color is written as .
- : lightness. A nonlinear function of luminance relative to the reference white. It equals 0 for ideal black and 100 when the sample luminance equals the reference-white luminance.
- : one chromatic coordinate. Positive values generally point toward reddish colors, and negative values toward greenish colors.
- : the other chromatic coordinate. Positive values generally point toward yellowish colors, and negative values toward bluish colors.
These directional descriptions are approximate. Neither chromatic axis is a direct physical measurement of a particular wavelength or a simple red–green or yellow–blue signal.
Colors with the same chromaticity as the reference white lie on the neutral axis, where . Moving away from that axis increases chroma. The usual interpretation of between 0 and 100 assumes colors no brighter than the reference white; the mathematical transformation can also produce values above 100.
CIELUV coordinates are dimensionless. The asterisks distinguish the perceptual coordinates from related quantities, particularly the chromaticity coordinates and .
Conversion from CIE XYZ
Conversion requires the sample tristimulus values and the reference-white values , expressed on a consistent scale. The component is proportional to luminance; its ratio to determines lightness.
Chromaticity coordinates
The CIE 1976 chromaticity coordinates are
For the reference white,
These coordinates describe chromaticity independently of the overall scaling of the tristimulus values. Multiplying by the same positive factor leaves and unchanged.
The relationship to the CIE 1960 coordinates is
Thus, the 1960 and 1976 diagrams must not be treated as numerically interchangeable.
Lightness
Let
Then
The cube-root branch approximates the nonlinear relationship between relative luminance and perceived lightness. The linear branch near black avoids an infinite slope at zero and joins smoothly to the cube-root branch.
Chromatic components
The remaining coordinates are
Consequently, and depend both on displacement from the reference white in the chromaticity diagram and on lightness.
For ideal black, , the chromaticity expressions become indeterminate. Black is conventionally assigned , and numerical implementations must handle this case explicitly.
Reference white and viewing assumptions
CIELUV is relative to a specified white point. A sample’s coordinates can change when a different reference white is selected, even if the sample’s XYZ values remain unchanged. Common references include CIE standard illuminants D65 and D50.
A complete color specification therefore needs more than three CIELUV numbers. It should also identify the reference white and the standard-observer conditions underlying the XYZ values.
Changing the reference white in the equations is not, by itself, a complete model of chromatic adaptation. When predicting the appearance of colors under different illumination or adaptation conditions, an appropriate adaptation transform or appearance model may be needed.
CIELUV also does not fully account for the effects of surround, background, absolute luminance, or spatial context. Its coordinates should therefore not be interpreted as a complete description of appearance under arbitrary viewing conditions.
Cylindrical representation
CIELUV can be expressed in cylindrical coordinates, commonly written . Chroma is
and the hue angle is
usually converted to degrees and expressed from to .
In this representation:
- controls lightness;
- measures distance from the neutral axis;
- specifies the direction around that axis.
Hue is undefined when chroma is zero and becomes numerically unstable very close to the neutral axis. Cylindrical CIELUV is sometimes called LChuv, distinguishing it from the analogous cylindrical form of CIELAB.
For , a saturation correlate can be defined as
Chroma and saturation are therefore related but not interchangeable quantities.
Color differences
The basic CIELUV color-difference formula is the Euclidean distance between two coordinate triples:
This provides a convenient scalar measure for comparing colors calculated under compatible reference conditions. Smaller distances generally indicate more similar colors.
However, CIELUV is only approximately perceptually uniform. A fixed does not guarantee equal visibility across all hues, chromas, lightness levels, or viewing conditions. In particular, a value of 1 should not be treated as a universal threshold for a just-noticeable difference.
Color-difference formulas associated with other spaces are not automatically interchangeable with this one. For example, CIEDE2000 is formulated using CIELAB coordinates, not CIELUV coordinates.
Applications
CIELUV and its associated diagram have several related uses:
- Display characterization: comparing measured display colors and examining the color gamut of additive color systems.
- Lighting measurement: expressing chromaticity and evaluating chromaticity variation among light sources. Many such tasks use only the two-dimensional diagram rather than the full CIELUV space.
- Color selection and visualization: constructing palettes in which lightness, chroma, and hue can be manipulated more directly than in a device-dependent RGB encoding.
- Color processing: computing approximate perceptual distances or performing interpolation in a space designed to be more uniform than XYZ or encoded RGB.
A distinction is necessary between perceptual interpolation and physical light mixing. Straight-line interpolation in CIELUV is not generally equivalent to mixing lights in corresponding proportions. Physical additive mixing is linear in XYZ, whereas CIELUV is a nonlinear transformation.
Limitations and implementation considerations
CIELUV is device-independent in its definition, but it does not make every coordinate reproducible on every device. Conversion to a display or printer may produce out-of-gamut values, requiring clipping or a gamut-mapping method.
Conversion from encoded RGB also requires care. Encoded channel values generally need to be decoded into linear-light values before transformation to XYZ. Treating nonlinear encoded values as linear changes the resulting colorimetry.
The apparent simplicity of the diagram can also be misleading. It omits lightness, and distances within it are not identical to full CIELUV distances. Two lights with the same chromaticity but different luminances occupy the same point in that diagram while potentially having different CIELUV coordinates.
For tasks requiring more complete predictions across different adaptation and viewing conditions, color appearance models provide additional structure beyond CIELUV’s lightness-and-chromaticity transformation.