ISO 13655:1996
(Main)Graphic technology — Spectral measurement and colorimetric computation for graphic arts images
Graphic technology — Spectral measurement and colorimetric computation for graphic arts images
Technologie graphique — Mesurage spectral et calcul colorimétrique relatifs aux images dans les arts graphiques
Grafična tehnologija - Spektrometrija in kolorimetrični izračuni za grafične upodobitve
General Information
Relations
Buy Standard
Standards Content (Sample)
INTERNATIONAL
STANDARD
First edition
1996-I O-01
Graphic technology - Spectral
measurement and calorimetric computation
for graphic arts images
Technologie graphique - Mesurage spectral et calcul colorim&rique
relatifs aux images dans /es arts graphiques
Reference number
IS0 13655:1996(E)
---------------------- Page: 1 ----------------------
Page
Contents
................................................................................................ 1
1 Scope
1
.......................................................................
2 Normative references
1
...........................................................
3 Definitions and abbreviations
2
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4 Spectral measurement requirements
3
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .*.
5 Calorimetric computation requirements
Annexes
A Procedures for widening the bandwidth of narrow bandpass
6
......................................................................................
instruments
.......... 7
B Computation of CIELAB, CIELUV and CMC(I:c) parameters.
...................... 9
C Spectral weights for illuminant Da5 and 2” observer.
11
...................................................................
D Sample backing material
12
. . . . .*.
E Measurement geometry
14
F Aperture size in reflectance measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
15
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
G Fluorescence in measurement
16
.............................................
H Improving inter-instrument agreement
18
......................................................................................
J Bibliography
0 IS0 1996
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced
or utilized in any form or by any means, electronic or mechanical, including photocopying and
microfilm, without permission in writing from the publisher.
International Organization for Standardization
Case postale 56 l CH-1211 Geneve 20 l Switzerland
Printed in Switzerland
ii
---------------------- Page: 2 ----------------------
@ IS0 IS0 13655:1996(E)
Foreword
IS0 (the International Organization for Standardization) is a worldwide
federation of national standards bodies (IS0 member bodies
The work of
preparing International Standards is normally carried out
‘through IS0
technical committees. Each member body interested in a sut ject for which
a technical committee has been established has the right to be rep-
resented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. IS0
collaborates closely with the International Electrotechnical Commission
(IEC) on all matters of electrotechnical standardization.
Draft International Standards adopted by the technical committees are
circulated to the member bodies for voting. Publication as an International
Standard requires approval by at least 75 % of the member bodies casting
a vote.
International Standard IS0 13655 was prepared by Technical Committee
lSO/TC 130, Graphic technology.
Annex A forms an integral part of this International Standard. Annexes B
to J are for information only.
---------------------- Page: 3 ----------------------
IS0 13655:1996(E) 0 IS0
Introduction
There are many practices for making spectral measurements and
calorimetric computations allowed in CIE Publication 15.2. The choice of
instrument geometry, illuminant, observer, etc. are all left to the user.
Unfortunately, the selections made will result in different numerical values
for the same parameter for the same material. Furthermore, measure-
ments made under one method usually cannot be converted to correspond
to a different method. Thus, one may not be able to make valid compari-
sons using data from different methodologies. The purpose of this lnter-
national Standard is to specify a methodology for the measurement of
graphic arts images which results in valid and comparable data. While this
International Standard references the standard established for graphic arts
viewing conditions, it is not intended to provide an absolute correlation
with visual colour appearance.
IV
---------------------- Page: 4 ----------------------
INTERNATIONAL STANDARD @ IS0 IS0 13655:1996(E)
Spectral measurement and
Graphic technology -
calorimetric computation for graphic arts images
investigate the possibility of applying the most recent
1 Scope
editions of the standards indicated below. Members
of IEC and IS0 maintain registers of currently valid
This International Standard establishes a methodology
International Standards.
for reflection and transmission spectral measurement
and calorimetric parameter computation for graphic
IS0 5-2: 1991, Photography - Density measurements
arts images. Graphic arts includes, but is not limited
- Part 2: Geometric conditions for transmission
to, the preparation of material for, and volume pro-
density.
duction by, production printing processes which
include offset lithography, letterpress, flexography,
IS0 5-4: 1995, Photography - Density measurements
gravure and screen printing.
- Part 4: Geometric conditions for reflection density.
This International Standard does not apply to three-
IS0 3664: 1975, Photography - Illumination con-
filter (tristimulus) calorimeters although annexes B, D,
ditions for viewing co/our transparencies and their
E, F and G may also be relevant to those instruments.
reproductions.
This International Standard applies to colour
Cl E Publication 15.2: 1986, Calorimetry.
measurement of limited volume reproductions of
coloured images such as those produced with photo-
graphic, ink jet, thermal transfer, diffusion, electro-
photography, mechanical transfer or toner technology
3 Definitions and abbreviations
(e.g. off-press proofs) when used for graphic arts
applications.
For the purposes of this International Standard, the
following definitions and abbreviations apply.
This International Standard does not address the
spectral measurement of light emitted by video
3.1 CIE: Commission lnternationale de I ’Eclairage.
monitors nor does it supersede the specification of
other measurement geometries appropriate to
specific application needs, such as the evaluation of
3.2 CIE illuminants: llluminants A, D50, D65 and
materials (e.g. ink and paper) used in the graphic arts.
other D illuminants, defined by the CIE in terms of
relative spectral power distributions.
NOTE 1 Procedures for colour measurement of spectral
data from video monitors are included in ASTM E 1336-91[4].
3.3 illuminant: Radiation with a relative spectral
The use of integrating sphere geometry for paper evaluation
is covered in IS0 2469[*1. power distribution defined over the wavelength range
that influences object colour perception.
2 Normative references
3.4 measurement illuminant: Characteristic of the
radiant flux (light) incident on the specimen surface.
The following standards contain provisions which,
through reference in this text, constitute provisions of
this International Standard. At the time of publication, 3.5 radiance factor: Ratio of the radiance of the
the editions indicated were valid. All standards are surface element in the given direction to that of a
subject to revision, and parties to agreements based perfect reflecting or transmitting diffuser identically
on this International Standard are encouraged to irradiated.
1
---------------------- Page: 5 ----------------------
@ IS0
IS0 13655:1996(E)
NOTE 4 It is recognized that many instruments presently
3.6 reflectance factor: Ratio of the radiant or lumi-
do not have a measurement source that matches illuminant
nous flux reflected in the directions delimited by the
DsO. Annex G provides further information on fluorescence
given cone to that reflected in the same direction by a
and techniques to test for its presence.
perfect reflecting diffuser identically irradiated or
illuminated.
4.3 Wavelength range and interval for
3.7 sample backing: Surface on which the sample
measured values
is placed for measurement.
The data should be measured from 340 nm to 780 nm
3.8 transmittance factor (for incident radiation of a at 10 nm intervals and shall be measured from
given spectral composition, polarization and geometri- 400 nm to 700 nm, inclusive, at intervals of no more
cal distribution): Ratio of the transmitted radiant or than 20 nm. The reference for spectral data shall be
luminous flux to the incident flux in the given con- based on computed data at 10 nm intervals, where
ditions. the spectral response function is triangular with a
10 nm bandwidth.
3.9 bandwidth: Width of the spectral response
NOTE 5 Instrumentation with different intervals and
function at the half-power point.
response functions will produce different results. These
differences can be reduced by proper selection of band-
NOTE 2 For spectral measurement equipment a triangular
pass shape for a given interval and by applying the proper
response function is assumed.
method of calculation for the bandpass characteristic and
interval selected.
4 Spectral measurement requirements
4.4 Reflectance factor measurement
4.1 Instrument calibration
4.4.1 Sample backing material
The measurement instrument shall be calibrated in
accordance with its manufacturer ’s instructions. The
A sample backing material as defined in IS0 5-4:1995,
calibration standard provided by the manufacturer
4.7, shall be placed under or behind the sample during
shall be traceable to a national standardizing insti-
measurement to eliminate variability due to sample
tution.
on the reverse side
backing and any material printed
of the sample. See annex D.
NOTE 3 Where multiple instruments are used for
measurement, there will be differences in the resulting
data due to the individual characteristics of the instru-
ments. Annex H provides a methodology by which such
4.4.2 Measurement geometry
data can be brought into better agreement. The method-
ology is applicable to both reflection and transmission
Measurement geometry shall be 45 ”/O” or 0 ”/45” and
spectrophotometry.
conform with the geometric conditions defined in
IS0 5-4.
4.2 Spectral power distribution of the
measurement source
NOTES
4.2.1 Non-fluorescing materials
6 The use of 45 ”/0” or 0 ”/45” geometry will not adequately
address variations in all surface characteristics. Other
instrumentation can be used to detect specific character-
If the materials do not fluoresce, the spectral power
istics such as “bronzing ”. See annex E.
distribution of the measurement source is not a
concern and so no specification is given for the con-
7 It is recognized that many instruments do not conform
formity of the spectral power distribution of the
to the requirement in IS0 5-4 for a 2 mm boundary beyond
measurement source to the illuminant specified in
the sampling aperture due to the physical size of the press
51 . .
colour bars which are normally measured. Annex F provides
further information on aperture size.
4.2.2 Fluorescing materials
4.4.3 Measurement reporting
To minimize the variations in measurements between
instruments due to fluorescence, the spectral power
distribution of the measurement source shall match
Measured reflectance factors shall be multiplied by
CIE illuminant D 50 specified in 5.1 over the wave- 100 and shall be reported to the nearest 0,Ol %, or
length range of potential energy absorption and decimal equivalent, relative to a perfect reflecting
emission. diffuser having 100 % reflectance at all wavelengths.
2
---------------------- Page: 6 ----------------------
IS0 13655:1996(E)
@ IS0
The general form of these computations is:
4.5 Transmittance factor measurement
Reflection Transmission
4.5.1 Measurement geometry
a=780
a=780
Measurement geometry shall be normal/diffuse (OO/d) x = ~[R(a)-w#l x = ~m4*~~(a)l
a=340 a=340
or diffuse/normal (d/O ”) and conform either to the
geometric conditions defined in IS0 5-2 or those of
a=780 a=780
CIE 15.2
Y = ~w)4vy~a~i Y = x[T(A
a=340 a=340
The measurement geometry and the use of an inte-
a=780 it=780
grating sphere or opal diffuser shall be reported. (See
2 = c [Rta)-wzwi Z= C[T(A #I -w,(a)1
annex E.)
a=340
4.5.2 Measurement reporting
Measured transmittance factor shall be multiplied by
is the reflectance factor at wavelength 2;
R(4
100 and shall be reported to the nearest 0,Ol %, or
decimal equivalent, relative to the perfect transmitting
is the transmittance factor at wavelength a;
diffuser having 100 % transmittance at all wave-
lengths. (See annex E.)
Wx(A) is the weighting factor at wavelength il for
tristimulus value X;
5 Calorimetric computation
W,(A) is the weighting factor at wavelength a for
requirements
tristimulus value Y;
5.1 Calculation of tristimulus values
W,(A) is the weighting factor at wavelength il for
tristimulus value 2.
To provide consistency with graphic arts viewing
conditions, defined in IS0 3664, calculated tristimulus
If measured data is at intervals and bandpass is
values shall be based on CIE illuminant DFJ) and the
smaller than 10 nm, the method described in annex A
CIE 1931 standard calorimetric observer (often re-
shall be used to widen the bandpass of the data.
ferred to as the 2O standard observer) as defined in
CIE Publication 15.2. Computation shall be at 10 nm
NOTE 9 The weighting factors given in table 1 and table 2
or 20 nm intervals. Factors representing the product are based on triangular bandpass characteristics as referred
to in 4.3.
of CIE illuminant D50 and the 2O standard observer
data, to be used for weighting spectral reflectance
and transmittance data shall be those given in table 1 The values of Xn = 96,422, Yn = 100,000 and 2, =
for 10 nm intervals and table 2 for 20 nm intervals, as 82,521 shall be used to do calorimetric calculations.
taken from ASTM E 308i31. The user is strongly en-
couraged to use data at 10 nm intervals to improve
NOTES
the accuracy of the results.
IO Adding the weighting factors from 340 nm to 780 nm
NOTE 8 The 2’ standard observer was selected rather
in table 1 or in table 2 does not give a sum equal to the
than the IO0 standard observer, because it more closely
values for Xn, Yn and Zn. This is because Xn, Yn and Zn
matches the size of image detail found in printed material.
were computed to greater precision in ASTM E 308 than as
given by the summation of the table values. The sums for
If the measured spectral data begin at a wavelength
X, Y and 2 in the tables are useful as a data entry check.
greater than 340 nm, then all the weighting factors in
table 1 or table 2 for wavelengths less than the first
11 As a convenience for those applications which cannot
measured wavelength shall be summed and added to conform to this International Standard but which use CIE
illuminant DG5, weighting factors used to calculate tristimu-
the weighting factor for the first wavelength
lus values for CIE illuminant Ds5 and the CIE 1931 standard
measured.
calorimetric observer (often referred to as the 2’ standard
observer) are included in annex C.
If the last measured spectral data are at a wavelength
less than 780 nm, then all the weighting factors in
12 Tables 1 and 2 and tables C.1 and C.2 have been
table 1 or table 2 for wavelengths greater than the last
reproduced, with permission, from the Annual Book of
measured wavelength shall be summed and added to
ASTM Standards, copyright American Society for Testing
for the last wavelength
the weighting factor
and Materials, 1916 Race St., Philadelphia, PA 19130, USA.
measured.
---------------------- Page: 7 ----------------------
IS0 13655:1996(E) @ IS0
Table 1 -Weighting factors (W) for illuminant D50 Table 2 - Weighting factors (W) for illuminant D50
and 2’ observer for calculating tristimulus values and 2” observer for calculating tristimulus values
at 10 nm intervals at 20 nm intervals
Wavelength Wavelength
wx (4 wy (4 wz (4 WJf (4 wy (4 wz (it)
nm nm
340 0,000
340 0,000 0,000 0,000 0,000 0,000
0,001
360 0,000 0,000 360 -0,001 0,000 -0,003
370 0,001 0,000 0,005
380 -0,007 0,000 -0,034
0,000 0,013
380 0,003 400
0,100 0,001 0,459
390 0,012 0,000 0,057
420
1,651 0,044 7,914
400 0,060 0,002 0,285
440
4,787 0,325 24,153
0,006 1,113
410 0,234
460 4,897
1,018 28,125
420 0,775 0,023 3,723
480 1,815 2,413
15,027
1,610 0,066 7,862
430
500 0,044
6,037 4,887
12,309
440 2,453 0,162
520 1,263
13,141 1,507
450 2,777 0,313 14,647
540 5,608
18,442 0,375
0,514 14,346
460 2,500
560
11,361 18,960 0,069
470 1,717 0,798 11,299
580
16,904 16,060 0,026
1,239 7,309
480 0,861
600
19,537 11,646 0,014
4,128
490 0,283 1,839
620 15,917 7,132
0,003
500 0,040 2,948 2,466
640 8,342
3,245 0,000
4,632 1,447
510 0,088
660 3,112
1,143 0,000
520 0,593 6,587 0,736
680 0,857
0,310 0,000
1,590 8,308 0,401
530
700
0,178 0,064 0,000
540 2,799 9,197 0,196
720
550 4,207 9,650 0,085 0,044 0,016 0,000
0,037
560 5,657 9,471 740 0,011 0,004
0,000
570 7,132 8,902 0,020
760 0,002 0,001
0,000
0,015
580 8,540 8,112
780 0,001
0,000 0,000
590 9,255 6,829 0,010
Sums
96,423 100,002 82,522
600 9,835 5,838 0,007
NOTE - Although weighting factors are provided for 20 nm
610 9,469 4,753 0,004
intervals, the user is strongly encouraged to use data at IO nm
620 8,009 3,573 0,002
intervals to improve the accuracy of the results.
2,443 0,001
630 5,926
640 4,171 1,629 0,000
5.2 Calculation of other calorimetric
0,000
650 2,609 0,984
parameters
660 1,541 0,570 0,000
Calorimetric parameters shall be calculated using the
670 0,855 0,313 0,000
equations given in CIE Publication 15.2. The equations
680 0,434 0,158 0,000
for CIELAB L*, a*, b*, C& and hab and their associ-
690 0,194 0,070 0,000
ated colour difference equations are included in
700 0,097 0,035 0,000
annex B, together with the equations for CMC colour
710 0,050 0,018 0,000
difference.
0,008 0,000
720 0,022
730 0,012 0,004 0,000
5.3 Data reporting
740 0,000
0,006 0,002
When data generated in accordance with this lnter-
750 0,002 0,001 0,000
national Standard are reported, they shall be ac-
760 0,001 0,000 0,000
companied by the following information:
770 0,001 0,000 0,000
a) confirmation that measurements and compu-
780 0,000 0,000 0,000
I
tations are in conformance with IS0 13655;
Sums 96,421 99,997 82,524
1 b) originator of the data;
4
---------------------- Page: 8 ----------------------
IS0 13655:1996(E)
@ IS0
c) date of creation of the data; f) measurement source (light source and filter) con-
ditions used;
d) a description of the purpose or contents of the
data being exchanged; g) wavelength interval used.
e) a description of the instrumentation used, includ-
ing, but not limited to, the brand and model num-
.
ber
I
---------------------- Page: 9 ----------------------
IS0 13655:1996(E) @ IS0
Annex A
(normative)
Procedures for widening the bandwidth of narrow bandpass instruments
The body of this International Standard describes The function is defined
in the interval given
bY
procedures for tristimulus integration of spectral
aY?z-axrz I< Aa.
I
measurements taken with either 10 nm or 20 nm
bandwidth instruments. The method used for tristimu-
In those situations where data is not available at the
lus integration assumes that the instrument band-
ends of the measurement range, the data shall be
width and sampling interval are approximately equal (a
assumed to be uniform and the last available
10 nm sampling interval assumes a 10 nm bandwidth
measured value shall be used to define the end values.
and a 20 nm sampling interval implies a 20 nm band-
width). A triangular response function of the measur-
NOTE 13 The following example assumes that data is
ing instrument, with the half-power points defining
available at 3 nm intervals and that data is desired at 10 nm
the bandwidth, is also assumed. This assumption is
intervals. In the region of 420 nm the specific values are at
based on the design of the classic laboratory instru-
wavelengths of 403 nm, 406 nm, 409 nm, . . . 436 nm. The
ment which uses slit apertures and a diffraction
computation for the value at 420 nm is accomplished as
grating or prism.
follows:
Where data is available which has been collected at
1 Since the bandwidth (AA) is 10 nm, only data from 410
intervals that do not correspond to the desired 10 nm to 430 will be used in computation (data values at 412,
415,418,421,424, 427 and 430).
or 20 nm intervals of the available calorimetric
weighting functions weighting, it must be modified
2 The weighting functions will be 412 (0,2),
(resampled) to provide estimated (pseudo) data at the 415 (0,5), 418
(0,8), 421 (0,9), 424 (0,6),
427 (0,3) and 430 (0). The
required interval. This shall be done only if the data
sum of the weights is 3,3.
has been collected at an interval that is less (smaller)
than the desired 10 nm or 20 nm interval and if the
3 The spectral data at each wavelength Xn is multiplied
bandwidth corresponds to the sampling interval.
by the value of the weighting factor Xn and the prod-
ucts are summed and divided by the sum of the
The technique that shall be used to create the desired
weights (3,3 in this example). This is then the value to
data is to successively apply a triangular weighting
be used for a 10 nm bandpass centred at 420 nm.
function to the existing data based on the desired
(new) sampling intervals and bandwidth. This data is
4 This process is repeated at wavelengths within the
then summed over the interval and normalized by the
range of 340 nm to 780 nm at 10 nm intervals.
sum of the weights used. This process is repeated for
each new data point required.
The same procedure is used to modify other available data
intervals to provide input for calorimetric computation with
the available 10 nm and 20 nm weighting functions.
The weighting function is as follows:
Aq~Yn-axnI
w(n,) =
Ail
where
W&J is the weighting function at wavelength Xn;
is the wavelength for which data is to be
computed;
is the wavelength of available data;
Aa is the desired bandwidth.
---------------------- Page: 10 ----------------------
@ IS0
Annex B
(informative)
Computation of CIELAB, CIELUV and CMC(Z:c) parameters
where
B.1 CIELAB calorimetric parameters
(see CIE Publication 15.2)
U’ =4X/(X+ 15Y+ 32)
L* =
116[f(Y/Y,)I - 16
v’ =
9Y/(X+ 15Y+32)
a* =
5OO[f(X/X~) -fcylqJl
and u ’,, v ’, are the values of u ’, v’ for the reference
white.
b * = 2OO[f(Y/Y,) - jIz/zJ
The two spaces defined above are examples of Uni-
for: Xix, > 0,008 856, f(xIxn) = (XIX,)113
form Colour Spaces. They are called this because the
uniformity of them, in terms of numerical difference,
Y/Y, > 0,008 856,flY/Y,) = (Y/Y,)1/3
between colours which are perceived as having equal
Z/Z, > 0,008 856,f(zIz,) = (2./ ‘. ‘.~) “~~
differences, is far better than for XYZ. Two such
spaces were approved by the CIE in 1976 because
for . Xix, G 0 I 008 856 I f(xIxn) = 7,786 7(X/X,) + 16/l 16
there were somewhat conflicting requirements. One
of these was that the colour space should have an
Y/r, s 0 I 008 856 I fly/Y,) = 7,786 7(Y/Y,) + 16/116
associated chromaticity diagram whose coordinates
must be linearly related to x and y.
Z/Z, s 0 I 008856 I fiZ&)= 7,7867(2/&J + 16/116
where For users who are concerned with the mixing of
coloured lights (which includes the television industry)
xn = 96,422, Yn = 100,000 and Zn = 82,521 for
the linearity of the XYZ system is an important prop-
the conditions described in 5.1.
erty since it means that the colour obtained by mixing
coloured lights is easily predicted because of addi-
ib = (a** + b**) “*
C
tivity. It follows from this that the colour gamut ob-
tained by mixing three additive stimuli can be defined
h
ab = tan-l(b*/a*) simply by constructing linear boundaries in colour
space between the primaries and the white and black.
When specified for a chromaticity diagram this sim-
where
plifies to a triangle joining the chromaticity values of
0
O
0
the primaries. Hence the requirement that a Uniform
*c
b 0 Colour Space must have an associated chromaticity
diagram as achieved by plotting u’ against v ’.
’ sh,b<180” ifa*GO
90
Pigments do not exhibit additive behaviour. Non-turbid
b” > 0
media, such as dyes, approximate it well when color-
180" Sh,b<270” ifa*
use to graphic technology where pigments, which
b "GO
exhibit turbid behaviour, are the normal reproduction
270" G hab < 360’ if a* a 0 colorants. It is frequently stated, though rarely proved,
that CIELAB provides a more uniform space in the
b* < 0
region of interest to graphic technology. In this con-
text it has become the preferred colour space for this
industry and is widely quoted. However, since it does
B.2 CIELUV calorimetric parameters
not have a linear relationship to XYZ (because of the
(see CIE Publication 15.2)
cube-roots in the calculation of a* and b*) there is no
chromaticity diagram associated with it. Thus the
L” =
116[f(Y/Y,)] - 16 colour gamut of a set of additive primaries cannot be
easily calculated. Whilst it is not strictly accurate to do
u” =
13L* (u’ - u ’,) so, because of the non-additive behaviour exhibited by
pigments, the colour gamut of a set of pigments for
v* = colour reproduction is sometimes approximated by a
13L” (v’ - v ’,)
7
---------------------- Page: 11 ----------------------
@ IS0
IS0 13655:1996(E)
and where
hexagon in the u ’v’ diagram joining the primaries and
secondaries. This can be directly compared with the
l/2
gamuts of other pigment sets plotted similarly or,
F= and
(c;b j4/ [(c;(, I4 + 1 9001
more importantly, that obtained from a colour monitor
display (or any other additive system). Obviously such
a comparison should be treated with some caution
T=0,36 + ( o,4COS(ha@5)(; unless 164” c hab
because of the non-additive nature of the primaries
< 345O, then
(and also because such a diagram does not show
luminance or lightness). Nevertheless, it is for such
T= 0 ’56 + 1 O,ZCOS(hab +I 68)).
applications that CIELUV proves of some value in
graphic technology.
NOTE 14 The CMC (Colour Measurement Committee, a
British organization) colour difference is not presently CIE
approved or recommended but a modified form is being
B.3 CIELAB colour differences
considered by CIE along with other colour difference
equations.
(see CIE Publication 15.2)
The values of the parameters in the CMC equation are
AL* =r;;-L;
derived from visual judgements based on accept-
ability, not perceptibility, differences for textiles. The
Aa* =a;-a$
value of AEcmc correlates well with visual assessment
of textiles when I = 2. The value of c is always 1 as
Ab” = by - b;
presently used and is explicitly given here to show
agreement with British Standard BS 6923L5]. (See also
AATCC Test Method 173-l 990.) However, other types
AC& = c&j - c&,2
of surface colours or acceptability differences might
require other values of I and c, and even different
Ah h h
ab = abl - ab2
values in different relations for SL, SC, SH, F and T.
The CMC colour difference model can be useful for
and b* for sample 1 and
For AEb from L*, a* establishing empirical tolerances.
sample 2,
For the colour differences below 3 the formula AEg4
may be of advantage [see CIE Publication 116-l 995
= [(AL*)* + (Aa*)* + (Ab*)*] “* .
AEib
(formula 2.1 I)].
The CIE presently defines a metric hue difference,
AH&, as
= [(~;b)*-(AL*)*-(Ac;b)*ll ’*
AHib
B.4 CMC(l:c) colour difference AE,,,
(see BS 692315])
where
AL*, AC& and AHib are as defined in B.3;
SL = 0,040 975L*/(l + 0,017 65L*) unless
L* < 16, then SL = Of51 1;
SC = 0,063 8c;b /(I + 0,013 1 c;b ) + 0,638;
&=Sc(FT+ 1 -F)
---------------------- Page: 12 ----------------------
IS0 13655:1996(E)
Annex C
(informative)
Spectral weights for illuminant D65 and 2” observer
NOTE 15 Adding the values of the weighting factors from
As a convenience for those applications which cannot
340 nm to 780 nm in table C.l or in table C.2 does not give
conform to this International Standard but which use
a sum equal to the values for Xnl Yn and Zn. This is be-
CIE illuminant D 65, weighting factors used to calculate
cause Xnl Yn and Zn were computed to greater precision in
tristimulus values for CIE illuminant D65 and the CIE
ASTM E 308 than as given by the summation of the table
1931 standard calorimetric observer (often referred to
values. The sums for X, Y and 2 in the tables are of value
as the 2” standard observer) are included for infor-
as a data entry check of the tables.
mation.
The values of Xn = 95,047, Y, = 100,000 and Zn =
108,883 may be used to do calorimetric calculations.
9
---------------------- Page: 13 ----------------------
@ IS0
IS0 13655:1996(E)
-Weighting factors (W) for illuminant
Table C.l -Weighting factors (W) for illuminant Table C.2
De5 and 2” observer for calculating tristimulus
D65 and 2” observer for calculating tristimulus
values at IO nm intervals values at 20 nm intervals
Wavelength Wavelength
wx (4 wy (Al wz (4 wx (4 wy (4 wz (4
nm nm
0,000
340 0,000 0,000 340 0,000 0,000 0,000
350 0,000 0,000 0,000
360 -0,001 0,000 -0,005
360 0,000 0,000 0,001
380 -0,008 0,000 -0,039
370 0,002 0,000 0'010
40
...
SLOVENSKI STANDARD
SIST ISO 13655:1997
01-junij-1997
*UDILþQDWHKQRORJLMD6SHNWURPHWULMDLQNRORULPHWULþQLL]UDþXQL]DJUDILþQH
XSRGRELWYH
Graphic technology -- Spectral measurement and colorimetric computation for graphic
arts images
Technologie graphique -- Mesurage spectral et calcul colorimétrique relatifs aux images
dans les arts graphiques
Ta slovenski standard je istoveten z: ISO 13655:1996
ICS:
17.180.20 Barve in merjenje svetlobe Colours and measurement of
light
37.100.01 *UDILþQDWHKQRORJLMDQD Graphic technology in
VSORãQR general
SIST ISO 13655:1997 en
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.
---------------------- Page: 1 ----------------------
SIST ISO 13655:1997
---------------------- Page: 2 ----------------------
SIST ISO 13655:1997
INTERNATIONAL
STANDARD
First edition
1996-I O-01
Graphic technology - Spectral
measurement and calorimetric computation
for graphic arts images
Technologie graphique - Mesurage spectral et calcul colorim&rique
relatifs aux images dans /es arts graphiques
Reference number
IS0 13655:1996(E)
---------------------- Page: 3 ----------------------
SIST ISO 13655:1997
Page
Contents
................................................................................................ 1
1 Scope
1
.......................................................................
2 Normative references
1
...........................................................
3 Definitions and abbreviations
2
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4 Spectral measurement requirements
3
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .*.
5 Calorimetric computation requirements
Annexes
A Procedures for widening the bandwidth of narrow bandpass
6
......................................................................................
instruments
.......... 7
B Computation of CIELAB, CIELUV and CMC(I:c) parameters.
...................... 9
C Spectral weights for illuminant Da5 and 2” observer.
11
...................................................................
D Sample backing material
12
. . . . .*.
E Measurement geometry
14
F Aperture size in reflectance measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
15
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
G Fluorescence in measurement
16
.............................................
H Improving inter-instrument agreement
18
......................................................................................
J Bibliography
0 IS0 1996
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced
or utilized in any form or by any means, electronic or mechanical, including photocopying and
microfilm, without permission in writing from the publisher.
International Organization for Standardization
Case postale 56 l CH-1211 Geneve 20 l Switzerland
Printed in Switzerland
ii
---------------------- Page: 4 ----------------------
SIST ISO 13655:1997
@ IS0 IS0 13655:1996(E)
Foreword
IS0 (the International Organization for Standardization) is a worldwide
federation of national standards bodies (IS0 member bodies
The work of
preparing International Standards is normally carried out
‘through IS0
technical committees. Each member body interested in a sut ject for which
a technical committee has been established has the right to be rep-
resented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. IS0
collaborates closely with the International Electrotechnical Commission
(IEC) on all matters of electrotechnical standardization.
Draft International Standards adopted by the technical committees are
circulated to the member bodies for voting. Publication as an International
Standard requires approval by at least 75 % of the member bodies casting
a vote.
International Standard IS0 13655 was prepared by Technical Committee
lSO/TC 130, Graphic technology.
Annex A forms an integral part of this International Standard. Annexes B
to J are for information only.
---------------------- Page: 5 ----------------------
SIST ISO 13655:1997
IS0 13655:1996(E) 0 IS0
Introduction
There are many practices for making spectral measurements and
calorimetric computations allowed in CIE Publication 15.2. The choice of
instrument geometry, illuminant, observer, etc. are all left to the user.
Unfortunately, the selections made will result in different numerical values
for the same parameter for the same material. Furthermore, measure-
ments made under one method usually cannot be converted to correspond
to a different method. Thus, one may not be able to make valid compari-
sons using data from different methodologies. The purpose of this lnter-
national Standard is to specify a methodology for the measurement of
graphic arts images which results in valid and comparable data. While this
International Standard references the standard established for graphic arts
viewing conditions, it is not intended to provide an absolute correlation
with visual colour appearance.
IV
---------------------- Page: 6 ----------------------
SIST ISO 13655:1997
INTERNATIONAL STANDARD @ IS0 IS0 13655:1996(E)
Spectral measurement and
Graphic technology -
calorimetric computation for graphic arts images
investigate the possibility of applying the most recent
1 Scope
editions of the standards indicated below. Members
of IEC and IS0 maintain registers of currently valid
This International Standard establishes a methodology
International Standards.
for reflection and transmission spectral measurement
and calorimetric parameter computation for graphic
IS0 5-2: 1991, Photography - Density measurements
arts images. Graphic arts includes, but is not limited
- Part 2: Geometric conditions for transmission
to, the preparation of material for, and volume pro-
density.
duction by, production printing processes which
include offset lithography, letterpress, flexography,
IS0 5-4: 1995, Photography - Density measurements
gravure and screen printing.
- Part 4: Geometric conditions for reflection density.
This International Standard does not apply to three-
IS0 3664: 1975, Photography - Illumination con-
filter (tristimulus) calorimeters although annexes B, D,
ditions for viewing co/our transparencies and their
E, F and G may also be relevant to those instruments.
reproductions.
This International Standard applies to colour
Cl E Publication 15.2: 1986, Calorimetry.
measurement of limited volume reproductions of
coloured images such as those produced with photo-
graphic, ink jet, thermal transfer, diffusion, electro-
photography, mechanical transfer or toner technology
3 Definitions and abbreviations
(e.g. off-press proofs) when used for graphic arts
applications.
For the purposes of this International Standard, the
following definitions and abbreviations apply.
This International Standard does not address the
spectral measurement of light emitted by video
3.1 CIE: Commission lnternationale de I ’Eclairage.
monitors nor does it supersede the specification of
other measurement geometries appropriate to
specific application needs, such as the evaluation of
3.2 CIE illuminants: llluminants A, D50, D65 and
materials (e.g. ink and paper) used in the graphic arts.
other D illuminants, defined by the CIE in terms of
relative spectral power distributions.
NOTE 1 Procedures for colour measurement of spectral
data from video monitors are included in ASTM E 1336-91[4].
3.3 illuminant: Radiation with a relative spectral
The use of integrating sphere geometry for paper evaluation
is covered in IS0 2469[*1. power distribution defined over the wavelength range
that influences object colour perception.
2 Normative references
3.4 measurement illuminant: Characteristic of the
radiant flux (light) incident on the specimen surface.
The following standards contain provisions which,
through reference in this text, constitute provisions of
this International Standard. At the time of publication, 3.5 radiance factor: Ratio of the radiance of the
the editions indicated were valid. All standards are surface element in the given direction to that of a
subject to revision, and parties to agreements based perfect reflecting or transmitting diffuser identically
on this International Standard are encouraged to irradiated.
1
---------------------- Page: 7 ----------------------
SIST ISO 13655:1997
@ IS0
IS0 13655:1996(E)
NOTE 4 It is recognized that many instruments presently
3.6 reflectance factor: Ratio of the radiant or lumi-
do not have a measurement source that matches illuminant
nous flux reflected in the directions delimited by the
DsO. Annex G provides further information on fluorescence
given cone to that reflected in the same direction by a
and techniques to test for its presence.
perfect reflecting diffuser identically irradiated or
illuminated.
4.3 Wavelength range and interval for
3.7 sample backing: Surface on which the sample
measured values
is placed for measurement.
The data should be measured from 340 nm to 780 nm
3.8 transmittance factor (for incident radiation of a at 10 nm intervals and shall be measured from
given spectral composition, polarization and geometri- 400 nm to 700 nm, inclusive, at intervals of no more
cal distribution): Ratio of the transmitted radiant or than 20 nm. The reference for spectral data shall be
luminous flux to the incident flux in the given con- based on computed data at 10 nm intervals, where
ditions. the spectral response function is triangular with a
10 nm bandwidth.
3.9 bandwidth: Width of the spectral response
NOTE 5 Instrumentation with different intervals and
function at the half-power point.
response functions will produce different results. These
differences can be reduced by proper selection of band-
NOTE 2 For spectral measurement equipment a triangular
pass shape for a given interval and by applying the proper
response function is assumed.
method of calculation for the bandpass characteristic and
interval selected.
4 Spectral measurement requirements
4.4 Reflectance factor measurement
4.1 Instrument calibration
4.4.1 Sample backing material
The measurement instrument shall be calibrated in
accordance with its manufacturer ’s instructions. The
A sample backing material as defined in IS0 5-4:1995,
calibration standard provided by the manufacturer
4.7, shall be placed under or behind the sample during
shall be traceable to a national standardizing insti-
measurement to eliminate variability due to sample
tution.
on the reverse side
backing and any material printed
of the sample. See annex D.
NOTE 3 Where multiple instruments are used for
measurement, there will be differences in the resulting
data due to the individual characteristics of the instru-
ments. Annex H provides a methodology by which such
4.4.2 Measurement geometry
data can be brought into better agreement. The method-
ology is applicable to both reflection and transmission
Measurement geometry shall be 45 ”/O” or 0 ”/45” and
spectrophotometry.
conform with the geometric conditions defined in
IS0 5-4.
4.2 Spectral power distribution of the
measurement source
NOTES
4.2.1 Non-fluorescing materials
6 The use of 45 ”/0” or 0 ”/45” geometry will not adequately
address variations in all surface characteristics. Other
instrumentation can be used to detect specific character-
If the materials do not fluoresce, the spectral power
istics such as “bronzing ”. See annex E.
distribution of the measurement source is not a
concern and so no specification is given for the con-
7 It is recognized that many instruments do not conform
formity of the spectral power distribution of the
to the requirement in IS0 5-4 for a 2 mm boundary beyond
measurement source to the illuminant specified in
the sampling aperture due to the physical size of the press
51 . .
colour bars which are normally measured. Annex F provides
further information on aperture size.
4.2.2 Fluorescing materials
4.4.3 Measurement reporting
To minimize the variations in measurements between
instruments due to fluorescence, the spectral power
distribution of the measurement source shall match
Measured reflectance factors shall be multiplied by
CIE illuminant D 50 specified in 5.1 over the wave- 100 and shall be reported to the nearest 0,Ol %, or
length range of potential energy absorption and decimal equivalent, relative to a perfect reflecting
emission. diffuser having 100 % reflectance at all wavelengths.
2
---------------------- Page: 8 ----------------------
SIST ISO 13655:1997
IS0 13655:1996(E)
@ IS0
The general form of these computations is:
4.5 Transmittance factor measurement
Reflection Transmission
4.5.1 Measurement geometry
a=780
a=780
Measurement geometry shall be normal/diffuse (OO/d) x = ~[R(a)-w#l x = ~m4*~~(a)l
a=340 a=340
or diffuse/normal (d/O ”) and conform either to the
geometric conditions defined in IS0 5-2 or those of
a=780 a=780
CIE 15.2
Y = ~w)4vy~a~i Y = x[T(A
a=340 a=340
The measurement geometry and the use of an inte-
a=780 it=780
grating sphere or opal diffuser shall be reported. (See
2 = c [Rta)-wzwi Z= C[T(A #I -w,(a)1
annex E.)
a=340
4.5.2 Measurement reporting
Measured transmittance factor shall be multiplied by
is the reflectance factor at wavelength 2;
R(4
100 and shall be reported to the nearest 0,Ol %, or
decimal equivalent, relative to the perfect transmitting
is the transmittance factor at wavelength a;
diffuser having 100 % transmittance at all wave-
lengths. (See annex E.)
Wx(A) is the weighting factor at wavelength il for
tristimulus value X;
5 Calorimetric computation
W,(A) is the weighting factor at wavelength a for
requirements
tristimulus value Y;
5.1 Calculation of tristimulus values
W,(A) is the weighting factor at wavelength il for
tristimulus value 2.
To provide consistency with graphic arts viewing
conditions, defined in IS0 3664, calculated tristimulus
If measured data is at intervals and bandpass is
values shall be based on CIE illuminant DFJ) and the
smaller than 10 nm, the method described in annex A
CIE 1931 standard calorimetric observer (often re-
shall be used to widen the bandpass of the data.
ferred to as the 2O standard observer) as defined in
CIE Publication 15.2. Computation shall be at 10 nm
NOTE 9 The weighting factors given in table 1 and table 2
or 20 nm intervals. Factors representing the product are based on triangular bandpass characteristics as referred
to in 4.3.
of CIE illuminant D50 and the 2O standard observer
data, to be used for weighting spectral reflectance
and transmittance data shall be those given in table 1 The values of Xn = 96,422, Yn = 100,000 and 2, =
for 10 nm intervals and table 2 for 20 nm intervals, as 82,521 shall be used to do calorimetric calculations.
taken from ASTM E 308i31. The user is strongly en-
couraged to use data at 10 nm intervals to improve
NOTES
the accuracy of the results.
IO Adding the weighting factors from 340 nm to 780 nm
NOTE 8 The 2’ standard observer was selected rather
in table 1 or in table 2 does not give a sum equal to the
than the IO0 standard observer, because it more closely
values for Xn, Yn and Zn. This is because Xn, Yn and Zn
matches the size of image detail found in printed material.
were computed to greater precision in ASTM E 308 than as
given by the summation of the table values. The sums for
If the measured spectral data begin at a wavelength
X, Y and 2 in the tables are useful as a data entry check.
greater than 340 nm, then all the weighting factors in
table 1 or table 2 for wavelengths less than the first
11 As a convenience for those applications which cannot
measured wavelength shall be summed and added to conform to this International Standard but which use CIE
illuminant DG5, weighting factors used to calculate tristimu-
the weighting factor for the first wavelength
lus values for CIE illuminant Ds5 and the CIE 1931 standard
measured.
calorimetric observer (often referred to as the 2’ standard
observer) are included in annex C.
If the last measured spectral data are at a wavelength
less than 780 nm, then all the weighting factors in
12 Tables 1 and 2 and tables C.1 and C.2 have been
table 1 or table 2 for wavelengths greater than the last
reproduced, with permission, from the Annual Book of
measured wavelength shall be summed and added to
ASTM Standards, copyright American Society for Testing
for the last wavelength
the weighting factor
and Materials, 1916 Race St., Philadelphia, PA 19130, USA.
measured.
---------------------- Page: 9 ----------------------
SIST ISO 13655:1997
IS0 13655:1996(E) @ IS0
Table 1 -Weighting factors (W) for illuminant D50 Table 2 - Weighting factors (W) for illuminant D50
and 2’ observer for calculating tristimulus values and 2” observer for calculating tristimulus values
at 10 nm intervals at 20 nm intervals
Wavelength Wavelength
wx (4 wy (4 wz (4 WJf (4 wy (4 wz (it)
nm nm
340 0,000
340 0,000 0,000 0,000 0,000 0,000
0,001
360 0,000 0,000 360 -0,001 0,000 -0,003
370 0,001 0,000 0,005
380 -0,007 0,000 -0,034
0,000 0,013
380 0,003 400
0,100 0,001 0,459
390 0,012 0,000 0,057
420
1,651 0,044 7,914
400 0,060 0,002 0,285
440
4,787 0,325 24,153
0,006 1,113
410 0,234
460 4,897
1,018 28,125
420 0,775 0,023 3,723
480 1,815 2,413
15,027
1,610 0,066 7,862
430
500 0,044
6,037 4,887
12,309
440 2,453 0,162
520 1,263
13,141 1,507
450 2,777 0,313 14,647
540 5,608
18,442 0,375
0,514 14,346
460 2,500
560
11,361 18,960 0,069
470 1,717 0,798 11,299
580
16,904 16,060 0,026
1,239 7,309
480 0,861
600
19,537 11,646 0,014
4,128
490 0,283 1,839
620 15,917 7,132
0,003
500 0,040 2,948 2,466
640 8,342
3,245 0,000
4,632 1,447
510 0,088
660 3,112
1,143 0,000
520 0,593 6,587 0,736
680 0,857
0,310 0,000
1,590 8,308 0,401
530
700
0,178 0,064 0,000
540 2,799 9,197 0,196
720
550 4,207 9,650 0,085 0,044 0,016 0,000
0,037
560 5,657 9,471 740 0,011 0,004
0,000
570 7,132 8,902 0,020
760 0,002 0,001
0,000
0,015
580 8,540 8,112
780 0,001
0,000 0,000
590 9,255 6,829 0,010
Sums
96,423 100,002 82,522
600 9,835 5,838 0,007
NOTE - Although weighting factors are provided for 20 nm
610 9,469 4,753 0,004
intervals, the user is strongly encouraged to use data at IO nm
620 8,009 3,573 0,002
intervals to improve the accuracy of the results.
2,443 0,001
630 5,926
640 4,171 1,629 0,000
5.2 Calculation of other calorimetric
0,000
650 2,609 0,984
parameters
660 1,541 0,570 0,000
Calorimetric parameters shall be calculated using the
670 0,855 0,313 0,000
equations given in CIE Publication 15.2. The equations
680 0,434 0,158 0,000
for CIELAB L*, a*, b*, C& and hab and their associ-
690 0,194 0,070 0,000
ated colour difference equations are included in
700 0,097 0,035 0,000
annex B, together with the equations for CMC colour
710 0,050 0,018 0,000
difference.
0,008 0,000
720 0,022
730 0,012 0,004 0,000
5.3 Data reporting
740 0,000
0,006 0,002
When data generated in accordance with this lnter-
750 0,002 0,001 0,000
national Standard are reported, they shall be ac-
760 0,001 0,000 0,000
companied by the following information:
770 0,001 0,000 0,000
a) confirmation that measurements and compu-
780 0,000 0,000 0,000
I
tations are in conformance with IS0 13655;
Sums 96,421 99,997 82,524
1 b) originator of the data;
4
---------------------- Page: 10 ----------------------
SIST ISO 13655:1997
IS0 13655:1996(E)
@ IS0
c) date of creation of the data; f) measurement source (light source and filter) con-
ditions used;
d) a description of the purpose or contents of the
data being exchanged; g) wavelength interval used.
e) a description of the instrumentation used, includ-
ing, but not limited to, the brand and model num-
.
ber
I
---------------------- Page: 11 ----------------------
SIST ISO 13655:1997
IS0 13655:1996(E) @ IS0
Annex A
(normative)
Procedures for widening the bandwidth of narrow bandpass instruments
The body of this International Standard describes The function is defined
in the interval given
bY
procedures for tristimulus integration of spectral
aY?z-axrz I< Aa.
I
measurements taken with either 10 nm or 20 nm
bandwidth instruments. The method used for tristimu-
In those situations where data is not available at the
lus integration assumes that the instrument band-
ends of the measurement range, the data shall be
width and sampling interval are approximately equal (a
assumed to be uniform and the last available
10 nm sampling interval assumes a 10 nm bandwidth
measured value shall be used to define the end values.
and a 20 nm sampling interval implies a 20 nm band-
width). A triangular response function of the measur-
NOTE 13 The following example assumes that data is
ing instrument, with the half-power points defining
available at 3 nm intervals and that data is desired at 10 nm
the bandwidth, is also assumed. This assumption is
intervals. In the region of 420 nm the specific values are at
based on the design of the classic laboratory instru-
wavelengths of 403 nm, 406 nm, 409 nm, . . . 436 nm. The
ment which uses slit apertures and a diffraction
computation for the value at 420 nm is accomplished as
grating or prism.
follows:
Where data is available which has been collected at
1 Since the bandwidth (AA) is 10 nm, only data from 410
intervals that do not correspond to the desired 10 nm to 430 will be used in computation (data values at 412,
415,418,421,424, 427 and 430).
or 20 nm intervals of the available calorimetric
weighting functions weighting, it must be modified
2 The weighting functions will be 412 (0,2),
(resampled) to provide estimated (pseudo) data at the 415 (0,5), 418
(0,8), 421 (0,9), 424 (0,6),
427 (0,3) and 430 (0). The
required interval. This shall be done only if the data
sum of the weights is 3,3.
has been collected at an interval that is less (smaller)
than the desired 10 nm or 20 nm interval and if the
3 The spectral data at each wavelength Xn is multiplied
bandwidth corresponds to the sampling interval.
by the value of the weighting factor Xn and the prod-
ucts are summed and divided by the sum of the
The technique that shall be used to create the desired
weights (3,3 in this example). This is then the value to
data is to successively apply a triangular weighting
be used for a 10 nm bandpass centred at 420 nm.
function to the existing data based on the desired
(new) sampling intervals and bandwidth. This data is
4 This process is repeated at wavelengths within the
then summed over the interval and normalized by the
range of 340 nm to 780 nm at 10 nm intervals.
sum of the weights used. This process is repeated for
each new data point required.
The same procedure is used to modify other available data
intervals to provide input for calorimetric computation with
the available 10 nm and 20 nm weighting functions.
The weighting function is as follows:
Aq~Yn-axnI
w(n,) =
Ail
where
W&J is the weighting function at wavelength Xn;
is the wavelength for which data is to be
computed;
is the wavelength of available data;
Aa is the desired bandwidth.
---------------------- Page: 12 ----------------------
SIST ISO 13655:1997
@ IS0
Annex B
(informative)
Computation of CIELAB, CIELUV and CMC(Z:c) parameters
where
B.1 CIELAB calorimetric parameters
(see CIE Publication 15.2)
U’ =4X/(X+ 15Y+ 32)
L* =
116[f(Y/Y,)I - 16
v’ =
9Y/(X+ 15Y+32)
a* =
5OO[f(X/X~) -fcylqJl
and u ’,, v ’, are the values of u ’, v’ for the reference
white.
b * = 2OO[f(Y/Y,) - jIz/zJ
The two spaces defined above are examples of Uni-
for: Xix, > 0,008 856, f(xIxn) = (XIX,)113
form Colour Spaces. They are called this because the
uniformity of them, in terms of numerical difference,
Y/Y, > 0,008 856,flY/Y,) = (Y/Y,)1/3
between colours which are perceived as having equal
Z/Z, > 0,008 856,f(zIz,) = (2./ ‘. ‘.~) “~~
differences, is far better than for XYZ. Two such
spaces were approved by the CIE in 1976 because
for . Xix, G 0 I 008 856 I f(xIxn) = 7,786 7(X/X,) + 16/l 16
there were somewhat conflicting requirements. One
of these was that the colour space should have an
Y/r, s 0 I 008 856 I fly/Y,) = 7,786 7(Y/Y,) + 16/116
associated chromaticity diagram whose coordinates
must be linearly related to x and y.
Z/Z, s 0 I 008856 I fiZ&)= 7,7867(2/&J + 16/116
where For users who are concerned with the mixing of
coloured lights (which includes the television industry)
xn = 96,422, Yn = 100,000 and Zn = 82,521 for
the linearity of the XYZ system is an important prop-
the conditions described in 5.1.
erty since it means that the colour obtained by mixing
coloured lights is easily predicted because of addi-
ib = (a** + b**) “*
C
tivity. It follows from this that the colour gamut ob-
tained by mixing three additive stimuli can be defined
h
ab = tan-l(b*/a*) simply by constructing linear boundaries in colour
space between the primaries and the white and black.
When specified for a chromaticity diagram this sim-
where
plifies to a triangle joining the chromaticity values of
0
O
0
the primaries. Hence the requirement that a Uniform
*c
b 0 Colour Space must have an associated chromaticity
diagram as achieved by plotting u’ against v ’.
’ sh,b<180” ifa*GO
90
Pigments do not exhibit additive behaviour. Non-turbid
b” > 0
media, such as dyes, approximate it well when color-
180" Sh,b<270” ifa*
use to graphic technology where pigments, which
b "GO
exhibit turbid behaviour, are the normal reproduction
270" G hab < 360’ if a* a 0 colorants. It is frequently stated, though rarely proved,
that CIELAB provides a more uniform space in the
b* < 0
region of interest to graphic technology. In this con-
text it has become the preferred colour space for this
industry and is widely quoted. However, since it does
B.2 CIELUV calorimetric parameters
not have a linear relationship to XYZ (because of the
(see CIE Publication 15.2)
cube-roots in the calculation of a* and b*) there is no
chromaticity diagram associated with it. Thus the
L” =
116[f(Y/Y,)] - 16 colour gamut of a set of additive primaries cannot be
easily calculated. Whilst it is not strictly accurate to do
u” =
13L* (u’ - u ’,) so, because of the non-additive behaviour exhibited by
pigments, the colour gamut of a set of pigments for
v* = colour reproduction is sometimes approximated by a
13L” (v’ - v ’,)
7
---------------------- Page: 13 ----------------------
SIST ISO 13655:1997
@ IS0
IS0 13655:1996(E)
and where
hexagon in the u ’v’ diagram joining the primaries and
secondaries. This can be directly compared with the
l/2
gamuts of other pigment sets plotted similarly or,
F= and
(c;b j4/ [(c;(, I4 + 1 9001
more importantly, that obtained from a colour monitor
display (or any other additive system). Obviously such
a comparison should be treated with some caution
T=0,36 + ( o,4COS(ha@5)(; unless 164” c hab
because of the non-additive nature of the primaries
< 345O, then
(and also because such a diagram does not show
luminance or lightness). Nevertheless, it is for such
T= 0 ’56 + 1 O,ZCOS(hab +I 68)).
applications that CIELUV proves of some value in
graphic technology.
NOTE 14 The CMC (Colour Measurement Committee, a
British organization) colour difference is not presently CIE
approved or recommended but a modified form is being
B.3 CIELAB colour differences
considered by CIE along with other colour difference
equations.
(see CIE Publication 15.2)
The values of the parameters in the CMC equation are
AL* =r;;-L;
derived from visual judgements based on accept-
ability, not perceptibility, differences for textiles. The
Aa* =a;-a$
value of AEcmc correlates well with visual assessment
of textiles when I = 2. The value of c is always 1 as
Ab” = by - b;
presently used and is explicitly given here to show
agreement with British Standard BS 6923L5]. (See also
AATCC Test Method 173-l 990.) However, other types
AC& = c&j - c&,2
of surface colours or acceptability differences might
require other values of I and c, and even different
Ah h h
ab = abl - ab2
values in different relations for SL, SC, SH, F and T.
The CMC colour difference model can be useful for
and b* for sample 1 and
For AEb from L*, a* establishing empirical tolerances.
sample 2,
For the colour differences below 3 the formula AEg4
may be of advantage [see CIE Publication 116-l 995
= [(AL*)* + (Aa*)* + (Ab*)*] “* .
AEib
(formula 2.1 I)].
The CIE presently defines a metric hue difference,
AH&, as
= [(~;b)*-(AL*)*-(Ac;b)*ll ’*
AHib
B.4 CMC(l:c) colour difference AE,,,
(see BS 692315])
where
AL*, AC& and AHib are as defined in B.3;
SL = 0,040 975L*/(l + 0,017 65L*) unless
L* < 16, then SL = Of51 1;
SC = 0,063 8c;b /(I + 0,013 1 c;b ) + 0,638;
&=Sc(FT+ 1 -F)
---------------------- Page: 14 ----------------------
SIST ISO 13655:1997
IS0 13655:1996(E)
Annex C
(informative)
Spectral weights for illuminant D65 and 2” observer
NOTE 15 Adding the values of the weighting factors from
As a convenience for those applications which cannot
340 nm to 780 nm in table C.l or in table C.2 does not give
conform to this International Standard but which use
a sum equal to the values for Xnl Yn and Zn. This is be-
CIE illum
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.