color measurement

Evaluating Color: Systems and Instruments – Color Measurement

The human eye can interpret over a million colors. There are three types of photoreceptors in the human eye that are responsible for interpreting a different range of wavelengths: long, medium, or small, or red, green, or blue sensitive wavelengths. Many factors can impact the way color is perceived including the light source and the illuminating and viewing angles. The description of color and color difference can be complex and inconsistencies in evaluating and reporting color can result in confusion. This article describes several systems used to represent the different attributes of color as well as the instruments used to measure color.

Systems for Evaluating and Reporting Color

The Munsell color system was created by Albert H. Munsell in the early 1900’s and represents three properties of color: hue (basic color), chroma  (brilliance), and value (lightness or darkness). This color system has been revised multiple times because there are some issues with the physical application of the theoretical system. This system is still in use and it has served as a foundation for numerous other color scales.

In 1931 the International Commission on Illumination (CIE) introduced the first system developed to describe all visible colors that the human eye can see and was named the CIE 1931 color space or CIEXYZ. This color space can be shown as a cube with X, Y, and Z indicating the primary colors red, blue, and yellow (also known as the tristimulus values) and x, y the chromaticity coordinates. The Y value is often referenced since it corresponds to the relative brightness. Building on this work, Richard S. Hunter developed the Hunter LAB color scale in 1948. The Hunter LAB scale was designed to be easily derived by mathematical formulas from the CIEXYZ color scale. In this model L is the lightness and is derived from the Y-tristimulus value, and a and b are opposite color axes. Red (positive)-green (negative) is represented by ‘a’ and yellow (positive)-blue (negative) is represented by ‘b’. This scale can be used to measure color on any object. The most current formulas were published in 1966.

In 1976 the CIE defined the CIELAB color space and is often referred to as CIE L*a*b*, or L*a*b* for short. The CIE L*a*b* is a 3-dimensional color model in the real number space. It expresses color as three values: L*, a*, and b with L* representing the lightness value, where darkest black is represented by L*=0 and the brightest white when L*=100. The a* axis represents the green-red component, with green in the negative direction and red in the positive direction. The b* axis represents the blue-yellow component, with blue in the negative direction and yellow in the positive direction. True neutral gray is represented when a* and b* equal 0. CIE L*a*b* was derived from the CIE 1931 XYZ color space and is heavily influenced by the Munsell Color System. It was designed to approximate human vision and does not hold any copyright or licensing.

Measuring Color Differences

Measuring the difference between two colors is complicated by the fact that there are three parameters to account for. The delta E (ΔE) value is used to express the difference between two colors as a single parameter and is calculated by:

ΔE = √(L*)2 + (a*)2 +(b*)2

The ΔE ranges in value from 0 to 100 and if ΔE=0 the standard and sample are identical in color. The ΔE considers the differences between the L, a, and b values of the sample and standard and is useful for determining whether a specimen color is within a specified tolerance from a standard. However, this can be problematic because one of the individual values may be out of tolerance while the ΔE value may be within the specification allowance. Also, if the ΔE is determined to be out of tolerance, the L, a, or b value causing the intolerance is not readily known. Even though the individual delta values for L*, a*, and b* may have a negative value, the resultant ΔE will always be positive. This metric is often used for quality control or formula adjustment. A commonly referenced rule of thumb is that the untrained human eye cannot differentiate between colors with a ΔE less than 3.

It is important to note that the color scales described in this article are not interchangeable, and there is no factor that can be used to convert color differences or tolerances from one measuring system to another for all colors of samples. The color differences obtained for the same samples evaluated in different color scale systems are not likely to be identical.

Two more important factors established by the CIE are the illuminant and standard observer. They are usually reported together, with the illuminant followed by the observer (e.g., D65/2° or D65/10°) where D65 indicates the illuminant and 2° or 10° indicates the observer. The illuminant is a standardized way to compare images or colors recorded under different light conditions. Illuminants A represent incandescent light sources of tungsten filament bulbs, and illuminants B and C are used to represent direct sunlight at noon and average day light. The most common illuminant used is the D illuminant representing daylight. The D illuminant is further subdivided according to the color temperature it is trying to replicate. Horizon light with a color temperature of 5000° Kelvin (K) is represented as D50, while D55 represents midmorning light with a color temperature of 5500°K, D65 represents noon daylight at 6500°K, and D75 represents north sky daylight at 7500°K. The paint industry typically uses the D65 illuminant.

The angle at which an object is viewed effects how its color is perceived. To remove this variability two standard observers were introduced. The 2° observer was defined in the 1931 CIE color model and is named to reflect the 2° angle to which color receptors in the human eye are activated. Visually the 2° observer is illustrated by the example of looking at your thumbnail at arm’s length. In 1964 the standard observer was modified to allow for a 10° angle on the human retina. The 10° observer is like looking at the palm of your hand at arm’s length.

Color Measuring Instrumentation

Today color can be measured using either a colorimeter or a spectrophotometer. A colorimeter uses red, green, and blue filters to mimic the way humans perceive color. Colorimeters can be used to calibrate computer monitors but cannot handle metamerism (a phenomenon when a color appears different when viewed under different lighting conditions). Spectrophotometers, on the other hand, use 16, 32, or more filters to evaluate color under different lighting sources. Because of these different lighting sources, they are unaffected by metamerism and can measure a variety of different materials, such as liquids, plastics, paper, metal, and fabrics.

There are three types of spectrophotometers that differ in the geometry of the light source: the object, and the observer (or detector). They are generally referenced by two numbers with the first number representing the starting point of the light and the second representing the destination of the light after being reflected off the surface of the sample such as 45°/0° (or 0°/45°), spherical (d/8°), and multi-angle. The two most common types are the 45°/0° and the d/8°. With the 45°/0° geometry the object is illuminated from a ring of lights fixed at 45°and the detector is set directly overhead at 0°and reversed for the 0°/45°. The 45°/0° geometry is most frequently used because it captures the reflection from an object the way the human eye would see it. It is ideal for measuring color on smooth or matte surfaces and excludes the influence of the texture of the surface; however, the glossy surface can appear darker or more saturated due to the difference in light reflection.

The spherical, or d/8° spectrophotometer illuminates the sample diffusely by reflecting the light of the mirror-like inside of the sphere. Measurements are taken at an 8° viewing angle. The spherical spectrophotometers also have a port that can be closed or opened to exclude or include specular light, which is the light that reflects from a smooth surface in a specific direction. Diffuse reflection occurs when light is scattered in many directions. Specular light included eliminates the influence of gloss and surface texture, so it only measures the color of the sample. Specular light excluded includes the influence of the surface (i.e., the gloss and any texturing) when measuring color.  They are most often used for measuring color on textured surfaces such as textiles, carpets, plastics, and samples with high gloss.

The multi-angle spectrophotometers present light as a single source, usually at 45°, but have several detectors to capture the light reflected in multiple directions. They are commonly used for pigments and coatings containing light reflective additives and can compensate for optical brighteners. This geometry is being used more frequently as color effect pigments are used in a wide range of applications including automotive paints and fingernail polish.

Application of Color Difference Criterion for Industrial/Marine Finish Coats

SSPC: The Society for Protective Coatings publishes performance-based standards for various industrial/marine coatings; some include color difference criterion. One example is SSPC Coating Specification No. 36, “Two-Component Weatherable Aliphatic Polyurethane Topcoat, Performance-Based.” Table 1 in this Specification contains weathering performance testing criteria based on either 1,000, 2000, or 3,000 hours accelerated weathering exposure (per ASTM D4587) or 12, 24, or 36 months natural outdoor weathering (per ASTM D1014). Independent of whether the exposure is accelerated or natural, and independent of the exposure time, the criteria for color change (when measured according to ASTM D2244) is < 2.0 ΔE* C.I.E 1976 L*a*b*.


The perception of color is influenced by a range of factors including the light source, the object’s surface, and the lighting and viewing geometry. Color systems have been developed for evaluating and reporting color. Spectrophotometers with different geometries are appropriate based on the specific characteristics of the object being evaluated. Performance-based industry standards establish criterion for maximum color difference after accelerated or natural weathering that are quantified using the color systems and measuring devices described in this article.

About the Author

Melissa Swogger is a Chemical Technician with KTA-Tator, Inc., where she has performed compositional and physical testing of coatings and coating systems for over 14 years. Melissa holds a BS in Chemistry from the University of Pittsburgh.   Melissa can be reached at