ISO/TR 16066:2003

TECHNICAL ISO/TR
REPORT 16066
First edition
2003-03-15
Graphic technology — Standard object
colour spectra database for colour
reproduction evaluation (SOCS)
Technologie graphique — Base de données de spectres de couleurs
d'objets normalisée pour l'évaluation de la reproduction des couleurs
(SOCS)
Reference number
©
ISO 2003
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.

©  ISO 2003
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 either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland
ii © ISO 2003 — All rights reserved

Contents Page
Foreword. iv
Introduction . v
1 Scope. 1
2 Standard object colour spectra . 1
2.1 Object categories and sample selection . 1
2.2 Typical set samples and difference set samples. 4
3 Use of the colour spectra database . 5
3.1 Use of typical sets. 5
3.2 Use of difference sets. 7
4 Permissions. 8
Annex A (informative) Spectral reflectance and transmittance source data. 9
Annex B (normative) Typical set selection method for artificial colour groups. 27
Annex C (normative) Typical set selection method for non-skin colour, natural colour groups . 29
Annex D (normative) Typical set selection method for skin groups. 31
Annex E (normative) Difference set selection method. 32
Annex F (informative) Correspondence between typical/difference samples and original collected
data samples . 33

Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies
(ISO member bodies). The work of preparing International Standards is normally carried out through ISO
technical committees. Each member body interested in a subject for which a technical committee has been
established has the right to be represented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the
International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. 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.
In exceptional circumstances, when a technical committee has collected data of a different kind from that
which is normally published as an International Standard (“state of the art”, for example), it may decide by a
simple majority vote of its participating members to publish a Technical Report. A Technical Report is entirely
informative in nature and does not have to be reviewed until the data it provides are considered to be no
longer valid or useful.
Attention is drawn to the possibility that some of the elements of this document may be the subject of patent
rights. ISO shall not be held responsible for identifying any or all such patent rights.
ISO/TR 16066 was prepared by Technical Committee ISO/TC 130, Graphic technology.
The TC 130 Japanese National Committee prepared this database, and their efforts have made this Technical
Report possible. The original form of this report was published as a technical report TR X 0012 by the
Japanese Standards Association in Japanese in December 1998.
iv © ISO 2003 — All rights reserved

Introduction
The simplest way to evaluate the colour reproduction of colour image input devices is to input images of objects
whose colours are exactly known and then to compare the pixel values to exact ones. For that purpose, input
colour target standards have been already established in ISO 12641:1997, Graphic technology — Prepress
digital data exchange — Colour targets for input scanner calibration. Evaluation becomes more complicated,
however, when we consider metamers.
The perceived colours of a pair of objects are referred to as metamers if, under a particular illumination, they
have the same tristimulus values even though they are spectrally different. The spectral pairs of such objects
may be used to advantage in the assessment of differences among lighting conditions. If, for instance, a pair
of spectra yield the same tristimulus values X, Y, Z under CIE illuminant D50, the difference between the two
perceived colours of the pair of spectra as measured in the field under a different illumination is referred to as
a metameric index, which can be used as a measure of the non-conformance of that illumination to D50.
While it may be sometimes preferable, e.g. for standards purposes, to use artificial pairs of perceived colours,
i.e. those not coming from natural objects, natural metameric pairs have the advantage that one can assess
the effect of non-standard lighting for a particular lighting condition. One may find, for instance, that a
particular light source leads to unacceptably large deviations in skin tones, whereas the same source is quite
acceptable for furniture colours.
Definitions with respect to metameric indices and the procedures for their evaluation are described in the
following CIE publications:
CIE 15.2, Colorimetry, 2nd ed. (1986) (Contains Special Metamerism Index: Change in Illuminant)
CIE 51.2, A method for assessing the quality of daylight simulators for colorimetry (1999)
CIE 80, Special metamerism index: Change in observer (1989)
It can also be useful to consider metamers in the evaluation of such colour image input devices as colour
scanners and digital cameras, which, though designed to capture images in a way similar to that of the human
visual system (HVS), nonetheless deviate enough from HVS sensitivity so that colour reproduction of sensed
colours in display devices or print outputs are significantly different from that desired, even when the
illumination conditions of the original human observation of an object have been recreated for the observation
of the output image.
To evaluate deviations due to variations in light sources and/or sensor sensitivities under actual conditions, it
is useful to know the range of spectral differences in existing objects. Committee members have created an
exhaustive collection of colours of existing objects, a database containing more than 50 000 items. This report
details the extraction from the data of 365 colour samples and their classification into sets, of which there are
two types, “typical sets” and “difference sets”. “Typical sets” refers to sets of typical spectral reflectances and
transmittances of objects as classified into a number of different categories. “Difference sets” refers to sets of
metamers whose tristimulus values are roughly typical but whose spectral values are significantly non-typical.
The entire original collection of more than 50 000 spectral data items is included, in electronic form, as part of
this Technical Report in the data directory SourceData as described in Annex A.

TECHNICAL REPORT ISO/TR 16066:2003(E)

Graphic technology — Standard object colour spectra database
for colour reproduction evaluation (SOCS)
1 Scope
This Technical Report provides a database of typical and difference sets of existing object colour spectral data
that are suitable for evaluating the colour reproduction of image input devices. It also includes the spectral
reflectance and transmittance source data from which these data sets have been derived.
2 Standard object colour spectra
2.1 Object categories and sample selection
2.1.1 Categories and groups
The following categories and subcategories were first established:
(1) Photographic materials
a. Transparencies
b. Reflection prints
(2) Offset prints
(3) Computer colour prints
a. Dye sublimation printer
b. Electrostatic printer
c. Ink-jet printer
(4) Paint (not for art)
(5) Paints (for art)
a. Oil paints
b. Water colours
(6) Textiles
a. Synthetic dyes
b. Plant dyes
(7) Flowers and leaves
(8) Outdoor scenes (Krinov data except for flowers and leaves)
(9) Human skin
a. Bare North Asian skin
b. Foundation-applied North Asian skin
c. Bare South Asian skin
d. Foundation-applied South Asian skin
e. Bare Caucasian skin
f. Bare Negroid skin
Spectral reflectance/transmittance data were then collected for more than 50 000 items falling into these
categories/subcategories.
Categories (1) to (5) and subcategory (6)a are for artificial colours, while subcategory (6)b and categories (7)
to (9) are for natural colours. Typical sets and difference sets were established from 365 samples in this
database. A typical set is a set of representative spectral data of colour objects, while a difference set is a set
of metamers whose colour under D65 illuminant is similar to typical set samples but differs significantly from
them spectrally. Samples for the typical and difference sets were selected as shown in Table 1. The manner
of selecting typical samples depended on whether colours were artificial or natural. In most artificial colour
groups, all colours are synthesized by mixing three or four colorants, and distributed almost uniformly in their
colour gamut. Colours in the paint (not for art) and paints (for art) categories, however, are synthesized by
mixing more than four colorants, and these categories were dealt with in the same manner as with natural
colours. Selection strategies are described in following subsections.
Table 1 — Numbers of selected typical/difference set colour samples
Group Typical sets Difference sets
Photo (transparency) 15 15
Photo (reflection print) 15 15
Offset prints 15 15
Dye sublimation printer 15 15
Electrostatic printer 15 15
Ink-jet printer 15 15
Textiles (synthetic dyes) 15 15
Flowers/grasses/leaves (includes 25 25
Krinov’s grasses and leaves)
Paint (not for art) 15 —
Oil paints 15 —
Water colours 15 —
Textiles (plant) 15 —
Non-grass/leaf Krinov 15 —
Bare North Asian skin 5 —
FD-applied North Asian skin 5 —
Bare South Asian skin 5 —
FD-applied South Asian skin 5 —
Bare Caucasian skin 5 —
Bare Negroid skin 5 —
Total 365
2 © ISO 2003 — All rights reserved

NOTE ‘Foundation’ is a cosmetic used as a base for facial make-up. However, in this Technical Report, ‘foundation
applied skin’ means skin that is not bare, but covered with foundation and/or face powder.
2.1.2 Typical set selection for artificial colour groups
There are seven artificial colour groups in Table 1: photographic transparency, photographic reflection prints,
offset prints, dye sublimation printer, electrostatic printer, ink-jet printer, and textiles (synthetic dyes). Colour
samples can be obtained for every hue in these groups.
Colorants can be expected to vary within any one group, as, for example, among the many products of
different photographic prints material manufacturers. Spectral reflectance measurements were carried out for
several representative products among them. Statistical analysis was carried out on the measurement data,
and the product whose characteristics most closely approximated the statistical average for the products as a
whole was determined to be a typical colour product. A mathematical explanation for this is found in Annex B.
From a large number of colour samples for a typical product, fifteen samples were selected whose colours are
nearest to pre-determined basic colours, which consist of three achromatic colours and twelve chromatic
colours that are homogeneously distributed in twelve hues. Lightness and chromaticity of each chromatic
colour was such that the colour gamut of each artificial colour group contained all the chromatic colours.
Table 2 shows the achromatic colours (1-3) and the chromatic colours (4-15). The colour for each spectral
data is calculated under D65 illuminant.
Table 2 — Basic colours for artificial colour groups
L* H* C* a* b*
1 20 — 0 0,0 0,0
2 50 — 0 0,0 0,0
3 80 — 0 0,0 0,0
4 40 0 30 30,0 0,0
5 45 30 35 30,3 17,5
6 50 60 37 18,5 32,0
7 60 90 45 0,0 45,0
8 60 120 30 −15,0 26,0
9 45 150 30 −26,0 15,0
10 45 180 23 −23,0 0,0
11 45 210 22 −19,1 −11,0
12 45 240 20 −10,0 −17,3
13 40 270 20 0,0 −20,0
14 35 300 27 13,5 −23,4
15 40 330 30 26,0 −15,0
2.1.3 Typical set selection for non-skin colour, natural colour groups
There are six natural colour groups in Table 1 for non-skin colours: flowers/grasses/leaves (including Krinov’s
grasses/leaves), paint (not for art), oil paints, watercolours, textiles (plant), and non-grass/leaf Krinov.
Samples in natural colour groups are not distributed in whole hues, and typical samples cannot be selected on
the basis of their colours. To select typical set samples for natural colour groups, an algorithm based on
spectral distribution was developed. When principal component analysis was applied to all samples in a group,
it became possible to express the data distribution in a low dimensional subspace. An equi-distanced lattice
was set in the subspace, and a representative sample was selected from each lattice point. A mathematical
explanation for this is found in Annex C. Twenty-five typical samples were selected from the
‘flowers/grasses/leaves’ group, while 15 typical samples were selected from each of the other groups.
2.1.4 Typical set selection for skin groups
There are six skin colour groups in Table 1: Bare North Asian skin, FD (foundation)-applied North Asian skin,
Bare South Asian skin, FD-applied South Asian skin, Bare Caucasian skin, and Bare Negroid skin. Skin
colours are not distributed over a wide range, and only five samples were selected from each group. These
colours correspond to average and extreme colours over their distribution in CIELAB space. The mathematical
explanation for this is found in Annex D.
2.1.5 Difference set selection
Difference set samples were selected for typical samples in the artificial colour groups and in the
flowers/grasses/leaves group. Colour samples in difference sets have colours that, while similar to typical set
sample colours under D65 illuminant, are significantly different spectrally. Samples whose colour was least
different from a typical set were selected first. The most spectrally different colours among these were then
selected to create the difference set. The mathematical explanation for this is found in Annex E.
2.2 Typical set samples and difference set samples
Typical set spectral data and difference set spectral data are stored in electronic form as files in the
subdirectories designated ‘typical’ and ‘difference’ described in the directory TRDatabase that is part of this
Technical Report. All spectral data are 31 dimensional, from 400 nm to 700 nm at 10 nm interval. The spectral
data are presented as percent reflectance factor or percent transmittance factor. File names corresponding to
the above groups are summarized in Table 3. The originally collected data can also be accessed from the
data directory SourceData which is described in Annex A. Correspondence with original sample identification
numbers (IDs) is summarized in Annex F.
The files ‘photo_t’, ‘photo_r’, ‘offset’, ‘print_ds’, ‘print_es’, ‘print_ij’, and ‘textiles_s’ in ‘typical’ directory include
typical set samples from artificial colour groups, each sample with a number from 1 to 15 has a colour similar
to its corresponding basic colour. The same holds for the difference set colours in the files ‘photo_t-d’,
‘photo_r-d’, ‘offset-d’, ‘print_ds-d’, ‘print_es-d’, ‘print_ij-d’, and ‘textiles_s-d’ in the ‘difference’ directory, since
they are samples in artificial colour groups. For the other typical set samples and for the other difference set
samples, which belong to natural colour groups, there is no such relation to basic colours. Nevertheless, the
order of colour samples in ‘flowers_leaves’ file in the ‘typical’ directory corresponds to the order of colour
samples in the ‘flowers_leaves-d’ file in the ‘difference’ directory. Such correspondences can be confirmed by
comparing the L*, a* and b* values of the samples in Annex F.
4 © ISO 2003 — All rights reserved

Table 3 — File names that contain typical set samples and difference set samples for each group
Kind of set Group Directory\File name
Typical set Photo (transparency) typical\photo_t.txt
Typical set Photo (reflection print) typical\photo_r.txt
Typical set Offset prints typical\offset.txt
Typical set Dye sublimation printer typical\print_ds.txt
Typical set Electrostatic printer typical\print_es.txt
Typical set Ink-jet printer typical\print_ij.txt
Typical set Textiles (synthetic dyes) typical\textiles_s.txt
Typical set Flowers/grasses/leaves (incl. Krinov’s grasses and leaves) typical\flowers_leaves.txt
Typical set Paint (not for art) typical\paint.txt
Typical set Oil paints typical\oil.txt
Typical set Water colours typical\water.txt
Typical set Textiles (plant) typical\textiles_p.txt
Typical set Non-grass/leaf Krinov typical\n_krinov.txt
Typical set Bare North Asian skin typical\n_asian_b.txt
Typical set FD-applied North Asian skin typical\n_asian_f.txt
Typical set Bare South Asian skin typical\s_asian_b.txt
Typical set FD-applied South Asian skin typical\s_asian_f.txt
Typical set Bare Caucasian skin typical\caucasian_b.txt
Typical set Bare Negroid skin typical\negroid_b.txt
Difference set Photo (transparency) difference\photo_t-d.txt
Difference set Photo (reflection print) difference\photo_r-d.txt
Difference set Offset prints difference\offset-d.txt
Difference set Dye sublimation printer difference\print_ds-d.txt
Difference set Electrostatic printer difference\print_es-d.txt
Difference set Ink-jet printer difference\print_ij-d.txt
Difference set Textiles (synthetic dyes) difference\textiles_s-d.txt
Difference set Flowers/grasses/leaves (incl. Krinov’s grasses and leaves) difference\flowers_leaves-d.txt
3 Use of the colour spectra database
3.1 Use of typical sets
In this section examples of how the database may be used to evaluate the colour reproduction accuracy of a
device are given. The usual colour reproduction evaluation scheme is as follows:
Object colour is calculated using colour matching functions where xx= 400 ,x 410 , .,
(xy, ,z) () ( )

tt t
   
xy700 , =y 400 ,y 410 , .,y 700 , zz= 400 ,z 410 , .,z 700 recommended by CIE 15.2.
() ()() () ()() ()
   
t
 
If the i-th object’s spectral reflectance is ββ= 400 ,β 410 , . ,β 700 and illumination intensity is
() () ( )
ii i i
 
t

SS= 400 ,S 410 , . ,S 700 , the CIE-1931 XYZ values for the i-th object may be expressed as in
()( ) ( )

equation (1):


Xk= S()λβλ()x()λ
0ii
∑

λ= 400



Yk= S λβλyλ
 () ()()
0ii
∑

λ= 400

700 (1)

Zk= S λβλzλ
() ()()
 0ii
∑
 λ= 400

where k=
Syλλ
()()
∑
λ= 400
NOTE CIE 15.2 recommends that calculation shall be carried out from 380 nm to 780 nm at an interval of 5 nm.
However, in the practical cases for which this data is used (with a data range of 400 nm to 700 nm at an interval of 10 nm)
it is recommended that weighting functions such as those defined in ASTM E-308 be used.
Colour image input devices usually have sensors with three different spectral sensitivities. If actual sensor
t
t
   
sensitivities are s,,ss ss= ()400 ,s()410 , . ,s()700 , ss= ()400 ,s()410 , . ,s()700 ,
()
rg b rr r r gg g g
   
t

ss= ()400 ,s()410 , . ,s()700 output signals (R , G , B ) for the i-th object may be expressed as in
bb b b
i i i

equation (2):
 700

RS= λβλs λ
() () ()
ii∑ r

λ= 400



GS= λβλs λ (2)
() () ()

ii∑ g

λ= 400


BS= λβλs λ
() () ()
ii∑ b
 λ= 400

The (R , G , B ) is converted to colour values in the CIE-1931 XYZ colour space. Equation (3) is often used for
i i i
the conversion — though a higher order function may sometimes be desirable.
Xa a a R
  
ii11 12 13
  
Ya= a a G (3)
ii21 22 23
  
  
Z aa a B
ii 31 22 23 
Matrix elements a can be determined using the least square method, where the sum of squared colour
{}
ij
differences for typical set colours are minimized in some colour space. The sum cannot be zero in practical
cases. This residual colour difference is evaluated in a uniform colour space (e.g. CIELAB space). The
conversion from the XYZ space to the CIELAB space is also described in CIE 15.2. Letting the converted
** * ** *
values from (XY,,Z ) and (XY,,Z ) be La00ii,,b0i and Laii,,bi, respectively, the
00ii 0i ii i ( ) ( )
average colour difference for n typical set object colours may be expressed as in equation (4):
n
22 2
** * * **
LL−+a−a +b−b
()ii00(i i)(i0i)
∑
i= 1
*
∆E = (4)
ab
n
6 © ISO 2003 — All rights reserved

*
The smaller the ∆E , the better the colour reproduction. That is, in the event that sensor spectral sensitivities
ab
of colour image input devices can somehow be measured, it is possible to evaluate their colour reproducibility
using actual object colours. It is also possible to estimate colour reproducibility using this database when
designing the spectral sensitivities of image input devices.
Usually, a single image input apparatus does not need to input all the object colours in the world. Film
scanners, for example, need only to scan colour films; it is not necessary for scanners to reproduce the
colours of natural objects or scenery, and the colour reproduction of colour scanners only needs to be
evaluated for colour films. That is why data here have been classified into object groups. For a scanner,
therefore, only the typical set samples identified as “photo” in Table 3 (i.e. photo_t.txt and photo_r.txt) would
be used for evaluation.
On the other hand, for evaluating digital cameras, human faces, flowers, leaves and outdoor objects always
have high importance, and typical sets for these object groups would be used.
3.2 Use of difference sets
Difference sets contain samples whose appearance is similar to the appearance of typical sets samples under
the standard illuminant D65, but whose spectral reflectance/transmittance is significantly different from that of
typical set samples. That is, difference sets contain metamers. Let us first assume that β represents the
j
spectral reflectance of a given difference set sample, a reflectance that corresponds to β , the spectral
i
reflectance of a typical set sample. In this case, two sensor outputs will be calculated using two light sources
S and S . Four sets of R, G, and B values can then be obtained, as expressed in equations (5) through (8).
1 2
 700

RS= λβλs λ
() () ()
ii11∑ r

λ= 400



GS= λβλs λ (5)
() () ()

ii11∑ g

λ= 400


BS λβλs λ
= () () ()
ii11∑ b
 λ= 400

 700

RS= λβλs λ
() () ()
ii22∑ r

λ= 400



GS= λβλs λ (6)
() () ()

ii22∑ g

λ= 400


BS= λβλs λ
() () ()
ii22∑ b
 λ= 400

 700

RS= λβλsλ
() () ()
jj11∑ r

λ= 400



GS= λβλs λ (7)
() () ()

jj11∑ g

λ= 400


BS= λβλs λ
() () ()
jj11∑ b
 λ= 400



RS= λβλs λ
() () ()
jj22
∑
r

λ= 400



GS= λβλs λ (8)
() () ()
jj22 g
∑

λ= 400


BS= λβλs λ
() () ()
jj22 b
 ∑
 λ= 400

To compare colour differences, CIELAB values for these four colour values can be calculated as
** * ** * ** * ** *
La,,b , La,,b , La,,b and La,,b through CIE-1931 XYZ values
ii11 i1 ii22 i2 jj11j jj22 j2
()()() ( )
*
as described in 3.1. The colour difference between β and β under light source 1 will then be equal to ∆E
i j ab1
*
in equation (9), and the colour difference between β and β under light source 2 will be equal to ∆E in
i j ab2
equation (10). By comparing the differences with those for the human visual system, it becomes possible to
evaluate a sensor’s resemblance to human eyes.
22 2
** * * * * *
∆EL=−L +a−a +b−b (9)
ab11()i j1( i1 j1)( i1 j1)
22 2
** * * * * *
∆EL=−L +a−a +b−b (10)
ab22()i j2( i2 j2)( i2 j2)
4 Permissions
The primary intent of this Technical Report is to provide data in digital form for the use of individuals and
organizations evaluating imaging systems. Therefore, the following is permitted:
a) The data files included as part of the Technical Report may be freely copied and used within the
organization purchasing a copy of this Technical Report from ISO, or an authorized reseller of ISO
documents. The data files (in either their original or any reformatted version) may not be distributed to any
other individual or organization.
b) Where this database is used for research, development, or evaluation, any publication reporting such
work shall identify this Technical Report as the source.

8 © ISO 2003 — All rights reserved

Annex A
(informative)
Spectral reflectance and transmittance source data
A.1 General
The spectral reflectance and transmittance source data collected as part of this project, and used to create the
typical and difference set samples, are included as electronic attachments to this Technical Report. These
data are classified into two groups, original data and interpolated data. The directory structure that contains
these files is shown in Figure A.1.

Figure A.1 — Directory structure of source data
A.1.1 \ORG files
ORG is an abbreviation of ‘ORIGINAL DATA’. The \ORG directory includes original spectral reflectance and
transmittance data measured by a spectral measurement apparatus. The data are classified in ten categories
so that users can select data appropriate for their use, depending on the types of colour image input devices
or the situations in which the data are used. The ten categories are shown below with names of directories
where the data are stored.
(1) Photographic materials (\PHOTO)
(2) Graphic prints (\GRAPHIC)
(3) Computer colour prints (\PRINTER)
(4) Paint (not for art) (\PAINT)
(5) Paints (for art) (\PAINTS)
(6) Textiles (\TEXTILES)
(7) Flowers (\FLOWERS)
(8) Leaves (\LEAVES)
(9) Human skin (\FACE)
(10) Krinov data (\KRINOV)
Sixteen spectral measurement devices were used to collect data in categories from (1) to (9). [Collection of
data in category (10) is explained later.] These devices are reliable but do not output the same values, even if
they measure the same object colours. To calibrate measured data, eight colour patches were produced and
measured by the sixteen measurement devices under the same conditions as were in effect when they
measured the object colours in categories from (1) to (9).
A detailed explanation file "README" for data in each category is included in each directory.
A.1.1.1 Photographic materials (\PHOTO)
Eight files are stored in the \PHOTO directory. They are shown in Table A.1. Transparent and reflection input
targets standardized by ISO 12641:1997 were produced by four major vendors. Spectral transmittance or
spectral reflectance for all colour patches in each target (288 colours) is recorded in each file.
Table A.1 — Files stored in \PHOTO directory
Target File name No. of Colours

Transparency Target-1 ph01_t.org 288
Target-2 ph02_t.org 288
Target-3 ph03_t.org 288
Target-4 ph04_t.org 288
Reflection print Target-1 ph01_r.org 288
Target-2 ph02_r.org 288
Target-3 ph03_r.org 288
Target-4 ph04_r.org 288
10 © ISO 2003 — All rights reserved

A.1.1.2 Graphic prints (\GRAPHIC)
The \GRAPHIC directory is divided into two subdirectories, \OFFSET and \GRAVURE. The \OFFSET directory
includes thirty files for colour patches printed by offset printing. They are summarised in Table A.2.
The colour patches are those standardized by C, M, Y and K ink quantities shown in ISO 12642:1996, Graphic
technology — Prepress digital data exchange — Input data for characterization of 4-colour process printing,
and as colour patch images in ISO 12640:1997, Graphic technology — Prepress digital data exchange —
CMYK standard colour image data (CMYK/SCID). The images were printed using ten kinds of ink and three
kinds of paper. Spectral reflectances of all 928 colour patches in the images are stored in each file.
Table A.2 — Files stored in \OFFSET directory
Printing method Ink Paper File name No. of colours
Offset printing Ink-1 Uncoated paper of01_u.org 928
Dull-coated paper of01_d.org 928
Gloss-coated paper of01_g.org 928
Ink-2 Uncoated paper of02_u.org 928
Dull-coated paper of02_d.org 928
Gloss-coated paper of02_g.org 928
Ink-3 Uncoated paper of03_u.org 928
Dull-coated paper of03_d.org 928
Gloss-coated paper of03_g.org 928
Ink-4 Uncoated paper of04_u.org 928
Dull-coated paper of04_d.org 928
Gloss-coated paper of04_g.org 928
Ink-5 Uncoated paper of05_u.org 928
Dull-coated paper of05_d.org 928
Gloss-coated paper of05_g.org 928
Ink-6 Uncoated paper of06_u.org 928
Dull-coated paper of06_d.org 928
Gloss-coated paper of06_g.org 928
Ink-7 Uncoated paper of07_u.org 928
Dull-coated paper of07_d.org 928
Gloss-coated paper of07_g.org 928
Ink-8 Uncoated paper of08_u.org 928
Dull-coated paper of08_d.org 928
Gloss-coated paper of08_g.org 928
Ink-9 Uncoated paper of09_u.org 928
Dull-coated paper of09_d.org 928
Gloss-coated paper of09_g.org 928
Ink-10 Uncoated paper of10_u.org 928
Dull-coated paper of10_d.org 928
Gloss-coated paper of10_g.org 928
The \GRAVURE directory includes three files of colour patches printed by gravure printing. The files are
shown in Table A.3. The colour patch images were printed using one type of paper under three different sets
of conditions. The colour patch images were the same as those for the offset printing.
Table A.3 — Files stored in \GRAVURE directory
Printing method Ink / Paper Printing condition File name No. of colours
Gravure printing Ink-1 / Gloss- Standard gr_s.org 928
coated paper
Minus gr_m.org 928
Plus gr_p.org 928
A.1.1.3 Computer colour prints (\PRINTER)
The \PRINTER directory contains the 21 files shown in Table A.4. Three kinds of colour patch images were
measured: 928-colour images, 512-colour images and 216-colour images. The 928-colour images, which are
the same as those used for graphic prints colour patch measurement, were printed by printers whose C, M, Y
and K components can be controlled from a computer program. The 512- and 216-colour images were
employed for printers whose only computer-controllable components are R, G and B. For printers each of
whose colour components was controlled in eight steps, 512 colours were printed and measured, while for
printers each of whose colour components was controlled in six steps, 216 colours were printed and
measured.
Table A.4 — Files stored in \PRINTER directory
Printing method Printer ID Files name No. of colours
Dye sublimation DS-1 pr_ds_1.org 512
DS-2 pr_ds_2.org 928
DS-3 pr_ds_3.org 512
DS-4 pr_ds_4.org 512
DS-5 pr_ds_5.org 216
Electrostatic ES-1 pr_es_1.org 216
ES-2 pr_es_2.org 216
ES-3 pr_es_3.org 928
ES-4 pr_es_4.org 216
Inkjet IJ-1 pr_ij_1.org 216
IJ-2 pr_ij_2.org 216
IJ-3 pr_ij_3.org 216
IJ-4 pr_ij_4.org 216
IJ-5 pr_ij_5.org 216
IJ-6 pr_ij_6.org 216
IJ-7 pr_ij_7.org 216
Silver halide SH-1 pr_sh_1.org 512
SH-2 pr_sh_2.org 216
Other OT-1 pr_ot_1.org 216
OT-2 pr_ot_2.org 216
OT-3 pr_ot_3.org 928
12 © ISO 2003 — All rights reserved

A.1.1.4 Paint (not for art) (\PAINT)
There is only one file in the \PAINT directory as shown in Table A.5. The file contains spectral reflectances of
painted objects. The colours are used for exterior/interior objects, and are not for artistic painting.
Table A.5 — File stored in \PAINT directory
File name Number of colours
paint.org 336
A.1.1.5 Paints (for art) (\PAINTS)
The \PAINTS directory contains the four files shown in Table A.6. Each file contains colour patch spectral
reflectances of oil paints, old type oil paints, watercolours, or new type watercolours whose properties lie
between those of water colours and oil paints. The old type oil paints are paints that were formerly used but
are not any longer because they contain poisonous heavy metal ingredients. Watercolour patches were
provided by Turner Colour Works Ltd., and oil paints patches were provided by Kusakabe Co. Ltd.
Table A.6 — Files stored in \PAINTS directory
Kind of paint Files name No. of colours
Water colour pa_a.org 60
New type paints pa_g.org 60
Oil paints pa_o.org 91
Old type oil paints pa_s.org 20

A.1.1.6 Textiles (\TEXTILES)
The \TEXTILES directory contains the six files shown in Table A.7. Colours on cotton, polyester, wool and silk
clothes are produced by synthetic dyes, while others are produced by plant dyes. The data for silk cloth were
provided by Consiglio Nazionale delle Ricerche, Italy, and the data for plant dyes were provided by Fuji Photo
Film Co., Ltd.
Table A.7 — Data files for measured textiles
Type of objects File names No. of colours
Cotton cloth cotton.org 714
Polyester cloth poly.org 714
Wool cloth wool.org 150
Silk cloth silk.org 1000
Cotton cloth dyed with plant-dyes plant_c.org 240
Silk yarn dyed with plant-dyes plant_y.org 14
A.1.1.7 Flowers (\FLOWERS)
There is only one file in the \FLOWERS directory as shown in Table A.8. The data were provided by Japan
Color Research Institute.
Table A.8 — File stored in \FLOWERS directory
File name Number of colours
flower.org 148
A.1.1.8 Leaves (\LEAVES)
There is only one file in the \LEAVES directory, as shown in Table A.9. The data were provided by Japan
Color Research Institute.
Table A.9 — File stored in \LEAVES directory
File name Number of colours
leaf.org 92
A.1.1.9 Human skin (\FACE)
The \FACE directory is divided into six subdirectories: \SHISEIDO, \KAO, \OOKA, \KAWASAKI, \OULU and
\SUN. Data file names were assigned according to the following rule: File names take the form of ‘ttgcppss’,
where
tt represents the name of the data set,
g represents gender (f: female, m: male),
c represents skin condition (b: bare skin, f: foundation-applied skin),
pp represents position (ch: cheek, fh: forehead, nk: neck, zy: zygomatic region, am: arm, ha: hand), and
ss represents the measurement device used.
The \SHISEIDO directory stores spectral reflectance data collected by Shiseido Co., Ltd. and contains the
27 files shown in Table A.10. All data are for females and represent measured spectral reflectances for
various skin positions and conditions. For any given locale, the order of individual subjects within each file is
the same.
14 © ISO 2003 — All rights reserved

Table A.10 — Files stored in \SHISEIDO directory
Subject set (locale, nationality) Skin condition Position File name No. of colours
Bangkok Foundation-applied skin Cheek baffchc1.org 62
Bare skin Forehead bafbfhc1.org 62
Bare skin Cheek bafbchc1.org 62
Bare skin Neck bafbnkc1.org 62
Kualalumpur (Chinese family) Foundation-applied skin Cheek kcffchc1.org 30
Bare skin Forehead kcfbfhc1.org 30
Bare skin Cheek kcfbchc1.org 30
Bare skin Neck kcfbnkc1.org 30
Kualalumpur (India family) Foundation-applied skin Cheek kiffchc1.org 30
Bare skin Forehead kifbfhc1.org 30
Bare skin Cheek kifbchc1.org 30
Bare skin Neck kifbnkc1.org 30
Kualalumpur (Malay family) Foundation-applied skin Cheek kmffchc1.org 30
Bare skin Forehead kmfbfhc1.org 30
Bare skin Cheek kmfbchc1.org 30
Bare skin Neck kmfbnkc1.org 30
Taipei Foundation-applied skin Cheek taffchc1.org 60
Bare skin Forehead tafbfhc1.org 60
Bare skin Cheek tafbchc1.org 60
Bare skin Neck tafbnkc1.org 60
Tokyo-1 Bare skin Neck t1fbnkc1.org 976
Bare skin Cheek t1fbchc1.org 976
Bare skin Zygomatic region t1fbzyc1.org 976
Tokyo-2 Bare skin Forehead t2fbfhc1.org 123
Bare skin Cheek t2fbchc1.org 123
Foundation-applied skin Forehead t2fffhc1.org 123
Foundation-applied skin Cheek t2ffchc1.org 123
The \KAO directory contains spectral reflectance data collected by the Kao Corporation. All data in this
directory are for Japanese females and were collected in Tokyo. It contains the eight files shown in Table A.11.
The order of individual subjects within each file is not the same.
Table A.11 — Files stored in \KAO directory
Subject set (place, nationality) Skin condition Position File name No. of colours
Tokyo (Japanese) Bare skin Forehead tofbfhc2.org 271
Foundation-applied skin Forehead tofffhc2.org 266
Bare skin Zygomatic region tofbzyc2.org 461
Foundation-applied skin Zygomatic region toffzyc2.org 325
Bare skin Cheek tofbchc2.org 460
Foundation-applied skin Cheek toffchc2.org 325
Bare skin Arm tofbamc2.org 37
Bare skin Neck tofbnkc2.org 313

The data in the \SHISEIDO and \KAO directories were all obtained with contact-type spectral measuring
devices. In contrast, the data in the \OOKA and \KAWASAKI directories were obtained with both contact-type
and remote-type measuring devices. Most subjects were males and a small portion of them were non-
Japanese. Table A.12 shows the composition by sex and nationality. Measured positions were forehead,
zygomatic region, and cheek. The data in \OOKA are stored in the 27 files shown in Table A.13 and organized
with respect to sex, skin condition, position, and measuring device, as in \SHISEIDO and \KAO. The same is
true for the data in \KAWASAKI (27 files; Table A.14).
Table A.12 — Composition of subjects in \OOKA and \KAWASAKI
OOKA KAWASAKI
Male FemaleMaleFemaleTotal
Japan 43 7 66 14 130
China 3 3 1 0 7
Korea 1 0 0 0 1
Taiwan 0 0 1 0 1
Singapore 1 0 0 0 1
Egypt 2 0 0 0 2
Bangladeshi 1 0 0 0 1
Turkey 1 0 0 0 1
France 1 0 0 0 1
Canada 0 0 1 0 1
Finland 0 0 1 0 1
Total 53 10 70 14 147
16 © ISO 2003 — All rights reserved

Table A.13 — Files stored in \OOKA directory
Subject set Type of measurement
Position Files name No. of colours
(sex / skin condition) device
Male / Bare skin Forehead Contact-1 oombfhc1.org 53
Zygomatic region Contact-1 oombzyc1.org 53
Cheek Contact-1 oombchc1.org 53
Forehead Contact-2 oombfhc2.org 53
Zygomatic region Contact-2 oombzyc2.org 53
Cheek Contact-2 oombchc2.org 53
Forehead Remote oombfhr.org 53
Zygomatic region Remote oombzyr.org 53
Cheek Remote oombchr.org 53
Female / Bare skin Forehead Contact-1 oofbfhc1.org 5
Zygomatic region Contact-1 oofbzyc1.org 5
Cheek Contact-1 oofbchc1.org 5
Forehead Contact-2 oofbfhc2.org 5
Zygomatic region Contact-2 oofbzyc2.org 5
Cheek Contact-2 oofbchc2.org 5
Forehead Remote oofbfhr.org 5
Zygomatic region Remote oofbzyr.org 5
Cheek Remote oofbchr.org 5
Female / Foundation-
Forehead Contact-1 oofffhc1.org 5
applied skin
Zygomatic region Contact-1 ooffzyc1.org 5
Cheek Contact-1 ooffchc1.org 5
Forehead Contact-2 oofffhc2.org 5
Zygomatic region Contact-2 ooffzyc2.org 5
Cheek Contact-2 ooffchc2.org 5
Forehead Remote oofffhr.org 5
Zygomatic region Remote ooffzyr.org 5
Cheek Remote ooffchr.org 5
Table A.14 — Files stored in \KAWASAKI directory
Subject set Type of measurement
Position Files name No. of colours
(sex / skin condition) device
Male / Bare skin Forehead Contact-1 kambfhc1.org 70
Zygomatic region Contact-1 kambzyc1.org 70
Cheek Contact-1 kambchc1.org 70
Forehead Contact-2 kambfhc2.org 70
Zygomatic region Contact-2 kambzyc2.org 70
Cheek Contact-2 kambchc2.org 70
Forehead Remote kambfhr.org 70
Zygomatic region Remote kambzyr.org 70
Cheek Remote kambchr.org 70
Female / Bare skin Forehead Contact-1 kafbfhc1.org 2
Zygomatic region Contact-1 kafbzyc1.org 2
Cheek Contact-1 kafbchc1.org 2
Forehead Contact-2 kafbfhc2.org 2
Zygomatic region Contact-2 kafbzyc2.org 2
Cheek Contact-2 kafbchc2.org 2
Forehead Remote kafbfhr.org 2
Zygomatic region Remote kafbzyr.org 2
Cheek Remote kafbchr.org 2
Female / Foundation-
Forehead Contact-1 kafffhc1.org 12
applied skin
Zygomatic region Contact-1 kaffzyc1.org 12
Cheek Contact-1 kaffchc1.org 12
Forehead Contact-2 kafffhc2.org 12
Zygomatic region Contact-2 kaffzyc2.org 12
Cheek Contact-2 kaffchc2.org 12
Forehead Remot
...