By Roy Osbourne 

Photometry concerns itself primarily with measuring the quantity of light emitted by a light source, or reflected or transmitted by a surface. The quantity of light conveyed to the eye from unit area of a source or surface per second flowing through a cone of 1 unit of solid angle is known as the luminance of that source or surface.

Photometry, however, is employed to measure only one parameter of the three needed to determine fully an 'objective' description of a coloured light source or surface. The other two are its so-called dominant wavelength and its purity. These two colorimetric dimensions together constitute the physical colour quality or chromaticity of a source or surface.

In 'subjective' terms, a colour can look red, yellow, green or blue, and one colour more or lees colourful than another. Objectively, the hue can be described reasonably well by its corresponding spectral wavelength, and its colourfulness (or saturation) by its purity.

The concept of 'dominant wavelength' is applied to coloured stimuli other than actual spectral lights, and is an estimate of the spectral wavelength to which the stimulus primarily corresponds.

Wavelengths within the visible range extend from red (approximately 760-645 nanometres) through orange, yellow, green and blue, to violet (approximtely 425-380 nanometres). The 'purity' of a light is a measurement of the degree to which 'coloured' light departs from truly monochromatic light (light of a single wavelength only, of purity value 1) and approaches white light, which consists of a psycho-physical mixture of all spectral wavelengths (exhibiting zero purity).

Three-colour (trichromatic) colorimetry is based on the assumption that any colour can be imitated in appearance by the additive mixture of three amounts of primary lights (red, green and blue). If the spectral composition of the chosen primary lights were known then it would be possible to encode the sample colour by simply measuring and noting the intensity of each component primary. Instruments designed to enable such a match to be made are called colorimetrers; and the three primary lights needed to make the match are called the matching stimuli.

By proposing that the matching stimuli be standardised, the Commission Internationale de l'Eclairage (CIE) opened the way to establishing an internationally agreeed system of objective colour notation; the monochromatic primary wavelengths selected by the CIE (1931) were 700.0 nanometres ('red'), 546.1 nanometres ('green') and 435.8 nanometres ('blue').

A simple, experimental method of observing the full range of additive colour mixtures obtained by combining any three primary lights was demonstrated by the Scots physicist James Clerk Maxwell in 1857. In his arrangement, red, green and blue primary lights are fixed one at each corner of a triangular board. Inside the triangle, all the colours that can be matched by mixing the three lights are not only seen in their appropriate position but their location can be plotted by drawing a grid of linear co-ordinates on the surface of the board (thereby obtaining a rudimentary 'chromaticity diagram').  

However, no matter which three primary lights are chosen, there will always be colours which cannot be included inside the area of the colour triangle; their location cannot therefore be given wholly in terms of positive-value co-ordinates. These colours include, most importantly, the monochromatic spectral wavelengths other than the three selected primary lights which, in relation to the sides of the triangle, are found to lie on a curve known as the 'locus of spectral colours' or spectrum locus.

To overcome this problem, and avoid the use of negative co-ordinates, a system of colour notation was approved by the 1931 conference of the CIE based on a set of 'imaginary stimuli', possessing in theory far greater purity than Maxwell's 'real' primary lights. The CIE imaginary primaries (which are not obtainable physically) are located at the corners of a figure large enough to enclose within it both the Maxwell triangle and the spectrum locus; they are encoded in the CIE system by the letters X, Y and Z, analogous respectively to the red, green and blue primaries of the original colour triangle.

By redrawing the equal-sided triangle as a right-angled triangle it was able to be transferred into a rectilinear graph. The co-ordinates of such a graph, the CIE chromaticity diagram [see illustration] can then be used to fix the position of any point within the area of the graph using standard x and y co-ordinates. The numbers along the curved spectrum locus indicate the wavelengths of the colour spectrum. The straight-line base corresponds to the side of Maxwell's triangle linking the red and blue primary lights, along which are to be found the non-spectral purples (including magenta); the other two sides of the triangle are included inside the area bounded by the curved locus.

In an example of a CIE (1931) colour notation, the chromaticity of an emerald green filter is located on the CIE diagram by the co-ordinates x equals 0.210 and y equals 0.710. From this notation it is possible to determine both the dominant wavelength and the purity of the stimulus.

For mathematical convenience, the three CIE imaginary primaries (the tristimulus values) were defined so that in theory the photometric values of X and Z were zero. In practical colorimetry, therefore, the luminance of a given stimulus is indicated entirely by the magnitude of the Y value, which can be adjusted also to express the so-called luminance factor of the stimulus as a percentage value.

No one system of colour specification has yet succeeded in accounting for all the factors which determine a colour response. However, of the many systems proposed, the CIE system (with subsequent modifications) has gained international recognition as being probably the most precise and flexible in notating what is generally accepted to be an unambiguous and reliably communicable definition of any given visual stimulus.

Bibliography
George A. Agoston (1979), Color Theory and Its Application in Art and Design. Berlin & New York: Springer-Verlag. Revised edition 1987.
G.J. & D.G. Chamberlin (1980), Colour, Its Measurement, Computation and Application. London & Philadelphia: Heyden.
Robert W.G. Hunt (1957), The Reproduction of Colour. King's Langley: Fountain Press. Fifth edition 1997.
W. David Wright (1944), The Measurement of Colour. Bristol: Adam Hilger. Fourth edition 1969.

Copyright © 1999 Roy Osborne, All Rights Reserved

For more information about the wonders of colour visit http://www.coloracademy.co.uk/ColorAcademy%202006/subjects/cie/cie.htm

 
 

 
 
Options
Printer Friendly Version 
 
Help
 
Related Subjects 
 B.S.I. (Standards) Organisations
Areas of Knowledge 
 Classification
Tip:
What colours are in this season? Visit coloracademy.co.uk to find out!
 
 
Return to Homepage