Abstract

Evaluation of color-rendering capability of commercial lamps is considered from the point of view of the lampmaker and the typical user, rather than for precision color-matching applications. Only an approximate magnitude of color-rendering capability of a test lamp has meaning for the average user. Only illuminants of superb color-rendering capability are used here as reference illuminants. Such ideal reference illuminants are defined for a range of colors, both on and off the blackbody line. Chromatic adaptation approximates object-color constancy adequately closely as long as both initial and final illuminants have color-rendering capability approximately as good as that of a daylightlike illuminant. These considerations lead to more universal reference illuminants and allow usefully valid comparison of lamps of different colors with acceptable accuracy.

© 1972 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. The IES-CIE interim method2 of evaluating the color-rendering properties of lights compares the chromaticities of a set of eight objects (spectral reflectance curves) illuminated by a reference illuminant to the chromaticities of the same objects illuminated by a test illuminant. Distance in 1960 CIE u,υcolor space from the chromaticity of one object as rendered by one illuminant to the chromaticity of the same object as rendered by the other illuminant is averaged for the eight objects, and the average is used as a measure of the discrepancy in color rendition between the two illuminants. Color-rendering index CRI = 100 − ka, where ais this average distance and kis a constant setting the CRI of the commercial warm white fluorescent lamp, with respect to a 3000-K blackbody, equal to about 50. When the color-rendering properties of two illuminants are thus compared, one illuminant acts as reference illuminant and the other acts as test illuminant; it is merely a matter of definition which functions as which. For example, the CRI of the warm white fluorescent lamp is about 50 when a 3000-K blackbody acts as reference illuminant, and, since ais the same, the CRI of the 3000-K blackbody is about 50 when a warm white fluorescent lamp acts as reference illuminant. The interim method is thus entirely relative, requiring the important decision as to what is to be used as reference illuminant. The reference illuminant is in effect the ideal, and the color-rendering rating of the test illuminant is penalized in proportion to the discrepancies represented by awhether these discrepancies represent more saturated colors, less saturated colors, or hue differences. The eight test objects (spectral reflectance curves) of the interim method were chosen so that the chromaticity pattern in average daylight (for example, illuminant C) would be approximately round, i.e., the chromaticities would be about equally saturated and about equally spaced around the hue circuit in the 1960 CIE u,υdiagram.
  2. IES Subcommittee on Color Rendering, Illum. Eng. 57, 471 (1962); Comm. Intern. de l’Eclairage (CIE), Publ. CIE 13 (E-1.3.2), Bur. Central de la CIE, 57 rue Cuvier, Paris, France (1965).
  3. D. Nickerson, Illum. Eng. 53, 77 (1958).
  4. D. L. MacAdam, J. Opt. Soc. Am. 27, 294 (1937).
    [CrossRef]
  5. G. Wyszecki, W. S. Stiles, Color Science (Wiley, New York, 1967).
  6. G. T. Winch, W. Harrison, H. R. Ruff, Trans. Illum. Eng. Soc. (London) 16, 3 (1951).
  7. D. B. Judd, G. Wyszecki, Color in Business, Science, and Industry (Wiley, New York, 1963).
  8. Most of the parameters of these color diagrams were obtained from Ref. 7, p. 277.
  9. W. D. Wright, The Measurement of Color (Van Nostrand-Reinhold, New York, 1969).
  10. W. D. Wright, Proc. Phys. Soc. 53, 93 (1941).
    [CrossRef]
  11. P. J. Bouma, Physical Aspects of Color (Philips Industries, Netherlands, 1947).
  12. That is, illuminant E, equal power at all wavelengths, generates a spectral power distribution of light reflected from a test object that is proportional, wavelength by wavelength, to the spectral reflectance of that test object as conventionally presented. This reflected SPD represents in that sense the true or inherent color of an object.
  13. A. C. Barr, C. N. Clark, J. Hessler, Illum. Eng. 47, 649 (1952).
  14. D. B. Judd, D. L. MacAdam, G. Wyszecki, J. Opt. Soc. Am. 54, 1031 (1964).
    [CrossRef]
  15. Ref. 9, p. 165.
  16. H. Helson, D. B. Judd, M. H. Warren, Illum. Eng. 47, 221 (1952).
  17. D. L. MacAdam, J. Opt. Soc. Am. 46, 500 (1956).
    [CrossRef] [PubMed]
  18. R. W. Burnham, R. M. Evans, S. M. Newhall, J. Opt. Soc. Am. 42, 597 (1952).
    [CrossRef] [PubMed]
  19. R. W. Burnham, R. M. Evans, S. M. Newhall, J. Opt. Soc. Am. 47, 35 (1957).
    [CrossRef]
  20. P. J. Bouma, A. A. Kruithoff, Philips Tech. Rev. 9, 257 (1947).
  21. D. Nickerson, C. W. Jerome, Illum. Eng. 60, 262 (1965).
  22. B. H. Crawford, Trans. Illum. Eng. Soc. (London) 28, 50 (1963).
  23. J. L. Ouweltjes, Die Farbe 9, 207 (1960).
  24. Or one can think of it as provision of an infinite number of reference illuminants of geometrically identical pattern and of whatever chromaticity is needed to match that of the test lamp.

1965 (1)

D. Nickerson, C. W. Jerome, Illum. Eng. 60, 262 (1965).

1964 (1)

1963 (1)

B. H. Crawford, Trans. Illum. Eng. Soc. (London) 28, 50 (1963).

1962 (1)

IES Subcommittee on Color Rendering, Illum. Eng. 57, 471 (1962); Comm. Intern. de l’Eclairage (CIE), Publ. CIE 13 (E-1.3.2), Bur. Central de la CIE, 57 rue Cuvier, Paris, France (1965).

1960 (1)

J. L. Ouweltjes, Die Farbe 9, 207 (1960).

1958 (1)

D. Nickerson, Illum. Eng. 53, 77 (1958).

1957 (1)

1956 (1)

1952 (3)

R. W. Burnham, R. M. Evans, S. M. Newhall, J. Opt. Soc. Am. 42, 597 (1952).
[CrossRef] [PubMed]

H. Helson, D. B. Judd, M. H. Warren, Illum. Eng. 47, 221 (1952).

A. C. Barr, C. N. Clark, J. Hessler, Illum. Eng. 47, 649 (1952).

1951 (1)

G. T. Winch, W. Harrison, H. R. Ruff, Trans. Illum. Eng. Soc. (London) 16, 3 (1951).

1947 (1)

P. J. Bouma, A. A. Kruithoff, Philips Tech. Rev. 9, 257 (1947).

1941 (1)

W. D. Wright, Proc. Phys. Soc. 53, 93 (1941).
[CrossRef]

1937 (1)

Barr, A. C.

A. C. Barr, C. N. Clark, J. Hessler, Illum. Eng. 47, 649 (1952).

Bouma, P. J.

P. J. Bouma, A. A. Kruithoff, Philips Tech. Rev. 9, 257 (1947).

P. J. Bouma, Physical Aspects of Color (Philips Industries, Netherlands, 1947).

Burnham, R. W.

Clark, C. N.

A. C. Barr, C. N. Clark, J. Hessler, Illum. Eng. 47, 649 (1952).

Crawford, B. H.

B. H. Crawford, Trans. Illum. Eng. Soc. (London) 28, 50 (1963).

Evans, R. M.

Harrison, W.

G. T. Winch, W. Harrison, H. R. Ruff, Trans. Illum. Eng. Soc. (London) 16, 3 (1951).

Helson, H.

H. Helson, D. B. Judd, M. H. Warren, Illum. Eng. 47, 221 (1952).

Hessler, J.

A. C. Barr, C. N. Clark, J. Hessler, Illum. Eng. 47, 649 (1952).

Jerome, C. W.

D. Nickerson, C. W. Jerome, Illum. Eng. 60, 262 (1965).

Judd, D. B.

D. B. Judd, D. L. MacAdam, G. Wyszecki, J. Opt. Soc. Am. 54, 1031 (1964).
[CrossRef]

H. Helson, D. B. Judd, M. H. Warren, Illum. Eng. 47, 221 (1952).

D. B. Judd, G. Wyszecki, Color in Business, Science, and Industry (Wiley, New York, 1963).

Kruithoff, A. A.

P. J. Bouma, A. A. Kruithoff, Philips Tech. Rev. 9, 257 (1947).

MacAdam, D. L.

Newhall, S. M.

Nickerson, D.

D. Nickerson, C. W. Jerome, Illum. Eng. 60, 262 (1965).

D. Nickerson, Illum. Eng. 53, 77 (1958).

Ouweltjes, J. L.

J. L. Ouweltjes, Die Farbe 9, 207 (1960).

Ruff, H. R.

G. T. Winch, W. Harrison, H. R. Ruff, Trans. Illum. Eng. Soc. (London) 16, 3 (1951).

Stiles, W. S.

G. Wyszecki, W. S. Stiles, Color Science (Wiley, New York, 1967).

Warren, M. H.

H. Helson, D. B. Judd, M. H. Warren, Illum. Eng. 47, 221 (1952).

Winch, G. T.

G. T. Winch, W. Harrison, H. R. Ruff, Trans. Illum. Eng. Soc. (London) 16, 3 (1951).

Wright, W. D.

W. D. Wright, Proc. Phys. Soc. 53, 93 (1941).
[CrossRef]

W. D. Wright, The Measurement of Color (Van Nostrand-Reinhold, New York, 1969).

Wyszecki, G.

D. B. Judd, D. L. MacAdam, G. Wyszecki, J. Opt. Soc. Am. 54, 1031 (1964).
[CrossRef]

G. Wyszecki, W. S. Stiles, Color Science (Wiley, New York, 1967).

D. B. Judd, G. Wyszecki, Color in Business, Science, and Industry (Wiley, New York, 1963).

Die Farbe (1)

J. L. Ouweltjes, Die Farbe 9, 207 (1960).

Illum. Eng. (5)

D. Nickerson, C. W. Jerome, Illum. Eng. 60, 262 (1965).

IES Subcommittee on Color Rendering, Illum. Eng. 57, 471 (1962); Comm. Intern. de l’Eclairage (CIE), Publ. CIE 13 (E-1.3.2), Bur. Central de la CIE, 57 rue Cuvier, Paris, France (1965).

D. Nickerson, Illum. Eng. 53, 77 (1958).

A. C. Barr, C. N. Clark, J. Hessler, Illum. Eng. 47, 649 (1952).

H. Helson, D. B. Judd, M. H. Warren, Illum. Eng. 47, 221 (1952).

J. Opt. Soc. Am. (5)

Philips Tech. Rev. (1)

P. J. Bouma, A. A. Kruithoff, Philips Tech. Rev. 9, 257 (1947).

Proc. Phys. Soc. (1)

W. D. Wright, Proc. Phys. Soc. 53, 93 (1941).
[CrossRef]

Trans. Illum. Eng. Soc. (London) (2)

G. T. Winch, W. Harrison, H. R. Ruff, Trans. Illum. Eng. Soc. (London) 16, 3 (1951).

B. H. Crawford, Trans. Illum. Eng. Soc. (London) 28, 50 (1963).

Other (9)

Or one can think of it as provision of an infinite number of reference illuminants of geometrically identical pattern and of whatever chromaticity is needed to match that of the test lamp.

The IES-CIE interim method2 of evaluating the color-rendering properties of lights compares the chromaticities of a set of eight objects (spectral reflectance curves) illuminated by a reference illuminant to the chromaticities of the same objects illuminated by a test illuminant. Distance in 1960 CIE u,υcolor space from the chromaticity of one object as rendered by one illuminant to the chromaticity of the same object as rendered by the other illuminant is averaged for the eight objects, and the average is used as a measure of the discrepancy in color rendition between the two illuminants. Color-rendering index CRI = 100 − ka, where ais this average distance and kis a constant setting the CRI of the commercial warm white fluorescent lamp, with respect to a 3000-K blackbody, equal to about 50. When the color-rendering properties of two illuminants are thus compared, one illuminant acts as reference illuminant and the other acts as test illuminant; it is merely a matter of definition which functions as which. For example, the CRI of the warm white fluorescent lamp is about 50 when a 3000-K blackbody acts as reference illuminant, and, since ais the same, the CRI of the 3000-K blackbody is about 50 when a warm white fluorescent lamp acts as reference illuminant. The interim method is thus entirely relative, requiring the important decision as to what is to be used as reference illuminant. The reference illuminant is in effect the ideal, and the color-rendering rating of the test illuminant is penalized in proportion to the discrepancies represented by awhether these discrepancies represent more saturated colors, less saturated colors, or hue differences. The eight test objects (spectral reflectance curves) of the interim method were chosen so that the chromaticity pattern in average daylight (for example, illuminant C) would be approximately round, i.e., the chromaticities would be about equally saturated and about equally spaced around the hue circuit in the 1960 CIE u,υdiagram.

D. B. Judd, G. Wyszecki, Color in Business, Science, and Industry (Wiley, New York, 1963).

Most of the parameters of these color diagrams were obtained from Ref. 7, p. 277.

W. D. Wright, The Measurement of Color (Van Nostrand-Reinhold, New York, 1969).

G. Wyszecki, W. S. Stiles, Color Science (Wiley, New York, 1967).

P. J. Bouma, Physical Aspects of Color (Philips Industries, Netherlands, 1947).

That is, illuminant E, equal power at all wavelengths, generates a spectral power distribution of light reflected from a test object that is proportional, wavelength by wavelength, to the spectral reflectance of that test object as conventionally presented. This reflected SPD represents in that sense the true or inherent color of an object.

Ref. 9, p. 165.

Cited By

OSA participates in CrossRef's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (10)

Fig. 1
Fig. 1

Color-rendering index of new commercial lamps, using reference illuminant specified in interim procedure (lower curve) or average daylight illuminant C as reference illuminant (upper curve). Arrows indicate the results of subjective evaluation of color-rendering capability of the new lamp.

Fig. 2
Fig. 2

Color-rendering index of new commercial lamp, using reference illuminant specified in interim method (SPEC, about 4000 K) or a reference illuminant of another color temperature. Only the reference illuminant at 6700 K (illuminant C) yields values of CRI correlative to subjective evaluation.

Fig. 3
Fig. 3

Pattern of test-color chromaticities in the 1960 CIE u,υ diagram as rendered by an experimental vapor lamp (solid line) and by a blackbody at 3000 K or average daylight illuminant C (dashed lines).

Fig. 4
Fig. 4

Pattern of test-color chromaticities in the 1960 CIE u,υ diagram as rendered by an experimental vapor lamp (solid line), a reference illuminant of reconstituted daylight at ∞K (blue sky, dashed line) and average daylight illuminant C (dotted line).

Fig. 5
Fig. 5

The range of observed daylight chromaticity, with lines of constant correlated color temperature. E = illuminant E, equal power at all wavelengths (Ref. 5, p. 12).

Fig. 6
Fig. 6

A hypothetical color diagram a, b absolutely uniform in perceptible color difference throughout. Octagonal patterns of eight test-color chromaticities as rendered by illuminants O, P, Q. Identical patterns indicate the equivalence of the three illuminants as regards color rendition.

Fig. 7
Fig. 7

The restricted region of color space (heavy outlines) of particular interest in the evaluation of commercial lamps of the familiar daylight (D), cool white (CW), and incandescence (illuminant A) chromaticities. The patterns represent the chromaticities of the eight test colors presently prescribed.

Fig. 8
Fig. 8

Wright’s vectors of equal perceived color difference, in the limited region of color space of Fig. 7, plotted in the 1960 CIE u,υ color diagram. The chromaticities of daylight (illuminant C) and of illuminant A, together with the blackbody–daylight line, are shown.

Fig. 9
Fig. 9

Six of Stiles’s ellipses in the 1931 CIE x,y diagram, covering the same restricted region of color space depicted in Figs. 7 and 8.5

Fig. 10
Fig. 10

Vectors representing changes in perceived color ascribable to change in adaptation illumination from daylight to tungsten light. Test colors (lights) at equivalent Munsell values of 5/ and 8/. Plotted in the new r,s diagram; dashed areas correspond to the restricted color regions of Figs. 7, 8, and 9. Data are from Ref. 18.

Tables (2)

Tables Icon

Table I Color-Rendering Index of Mixtures of Tungsten Light (Illuminant A) and Daylight (Illuminant C) Using (a) Illuminant A as Reference Illuminant and (b) the Reference Illuminant Prescribed by the Interim Method

Tables Icon

Table II Color-Rendering Index of a Daylight Fluorescent Lamp (Color Temperature 6800 K) Referred to Various Phases of Daylight and to Illuminant A

Equations (1)

Equations on this page are rendered with MathJax. Learn more.

r = 11.6986 x 7.94605 y 1.29263 48.5961 x 94.1789 y 21.1724 , s = 7.75392 x 50.0712 y + 3 48.5961 x 94.1789 y 21.1724 .

Metrics