Color Measurement

The objective of this article is to present the principles of the digitalization of color, the technologies used in its measurement, the spaces of representation and its extrapolation to human logic and perception.

It is common to discuss the color of materials and products in everyday life and, on many occasions, causing numerous controversies as there is no reference source for comparison.

In short, this is because the same source of color information can be interpreted differently by each human being.


Representation of different colours in Pantone scale


The objective of this article is to present the principles of the digitalization of color, the technologies used in its measurement, the spaces of representation and its extrapolation to human logic and perception.

In industrial color measurement applications and processes, it is normal to ask oneself, as in the perception of human color,

Where is the limit of this color detection?

Which technologies to consider for each color measurement process?

What is the target value to consider for our measurement?


  1. Color Definition

Color is NOT a property of light, but a subjective human perception of its physical principles.

To understand this statement we must understand the electromagnetic spectrum, as subjectively translated into human perception of color.


  1. The electromagnetic spectrum

The electromagnetic spectrum ranges from the smallest wavelengths of radiation (gamma rays and X-rays), through ultraviolet light, visible light and infrared rays, to the longest wavelengths of electromagnetic waves, such as radio waves.

Above the frequency of infrared radiation is what is usually called “light”, an electromagnetic radiation that has a wavelength in the range of 380 to 780nm. The light we see with our eyes is therefore a very small part of the electromagnetic spectrum.


Espectro electromagnético y “luz” visible


  1. Human perception

If the source of radiation has a frequency in the visible region of the electromagnetic spectrum and impacts an object, it will be a transmitted part, an absorbed part, a reflected part, and the remaining part emitted by the object.

The radiation that impacts the human eye gives rise to the visual perception of the scene. Our brain processes the multitude of frequencies that are reflected in different tones and nuances, and through this, the perception of the object is formed.


Evolución del estímulo del color



  • Spectral absorption of the human eye

Cones and rods are the two types of photoreceptors present in the human retina and are responsible for “color” vision.

On the retina, the Cones are less sensitive to light than the Bars (which support vision at low light levels). They are also able to perceive fine details and rapid changes in the image as the response to stimuli is faster than the Bars.

There are three types of Cones, each with a different pigment, and each one sensitive to light corresponding to short, medium or long length.

  • L-Cones (Long), Maximum at 560 nm, Greenish Yellow.
  • M-Cones (Medium), Maximum at 530 nm, Yellow-green.
  • S-Cones (Short), Maximum at 420 nm, Blue.


Respuesta espectral normalizada de los Conos del ojo humano


  1. Principles of colorimetry

Colorimetry is the science and technology used in quantification and physically describes human perception. It reduces the spectrum to the physical parameters of color perception. The CIE 1931 XYZ color space is commonly referred to and can be said to have become the standard.


CIE 1931 (Commission Internationale de l’Éclairage)


The CIE 1931 has but one great limitation in the measurement of color….

The chromatic diagram is obtained from the calculation of the ratios of each color.

y = X / (X+Y+Z) for the Red component

y = Y / (X+Y+Y+Z) for the Green component y

z = 1 – x – y for the Blue component.

Problem: The CIE 1931 color space is not equidistant!


Diagrama cromático CIE 1931


4.1. The colour space L*a*b**

Built according to the theory of color opposition. Its great advantage is that it is an equidistant color space and is built from:

Axis a*: opposition of red and green.

Axis b*: opposition of blue and yellow.

L* axis (Brightness): vertical direction of the axes a* and b*.


Diagrama cromático L*a*b*


Color deviation ΔE

It is calculated from:


It indicates the distance between two colors in an equidistant color space L*a*b, as a result interpretable in:

Deviation considered very small, not noticeable to the human eye.

1…2 Color deviation small, noticeable to the “trained eye”.

2…3.5 Medium colour deviation, perceptible to the “non-trained” eye.

3.5…5 Clear colour deviation.

5 Strong colour deviation: different colours.

In short, efficient colour measurement is done in L*a*b colour space!


4.2. Stage lighting

The lighting of the scene is key to the correct measurement of color. Within the light sources we find different technologies and some of them are optimal for the correct reproduction of color.



Light sources with optimum colour rendering (marked with red outline).


Comparison of light sources


  1. Methods for color measurement

For color measurement we can distinguish between: Equivalence Method, Spectral Method and Tristimulus Method. The latter two are the most commonly used in automated color measurement.


5.1. Spectral Method

The Spectral method measures the power of the spectral distribution of a light source. Its operation consists of capturing the light by means of an optical detector that directs a monochromator for its subsequent reading in different wavelength bands.


Método Espectral


  • High precision.
  • Independent of the light source.
  • The colour values of each illuminant are calculable.


  • Measurement cycle time.
  • Expensive hardware.
  • In most industrial processes its use becomes unviable.

5.2. Tri-stimulus method

The Tri-stimulus method acquires a limited number of spectral bandwidths in the visible spectrum using photodetectors and corresponding filters.



Método Triestímulo



  • High speed (e.g. ASTECH CR100 performs measurements at 50μs).
  • Very low cost components.
  • Compact design.
  • Ideal for industrial inline applications.


  • Less accurate than the Spectral method.
  • The measured colour values depend on the light source used.



ASTECH CR100 Colorimeter



Autor: Jaume Fontanella

vision_logo blue + transparent

Subscribe to our Newsletter

Contact Info

Vision Online SL

Tel: +34 93 795 92 70

Skype: Vision-Online

© Vision Online SL - All rights reserved