SIST ISO 12640-3:2022

INTERNATIONAL ISO
STANDARD 12640-3
Second edition
2022-08
Graphic technology — Prepress digital
data exchange —
Part 3:
CIELAB standard colour image data
(CIELAB/SCID)
Technologie graphique — Échange de données numériques de
préimpression —
Partie 3: Données d'images en couleur normalisées CIELAB (CIELAB/
SCID)
Reference number
© ISO 2022
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on
the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below
or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: +41 22 749 01 11
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
ii
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Requirements . 2
5 Data description and definition . .3
5.1 Data set definition . 3
5.2 Colour encoding used in this document . 3
5.2.1 Image data encoding . 3
5.2.2 Image data arrangement . 3
5.2.3 CIELAB image data (16 bits per channel) . 3
5.2.4 CIELAB image data (8 bits per channel) . 4
5.3 Natural images. 4
5.4 Synthetic images . 7
5.4.1 Colour charts . 7
5.4.2 Vignettes . 9
5.4.3 Synthetic image format . 10
6 Electronic data .11
6.1 Image file characteristics . 11
6.2 Image file formats . 11
Annex A (normative) Guidance for use of digital data .12
Annex B (informative) Definition of the reference colour gamut .14
Annex C (normative) Checksum data .21
Annex D (informative) Typical TIFF file headers used for image data .22
Annex E (informative) Text insertion .24
Annex F (informative) Histogram and gamut plots .25
Annex G (informative) CIELAB values in colour charts .29
Bibliography .34
iii
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.
The procedures used to develop this document and those intended for its further maintenance are
described in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the
different types of ISO documents should be noted. This document was drafted in accordance with the
editorial rules of the ISO/IEC Directives, Part 2 (see www.iso.org/directives).
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. Details of
any patent rights identified during the development of the document will be in the Introduction and/or
on the ISO list of patent declarations received (see www.iso.org/patents).
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and
expressions related to conformity assessment, as well as information about ISO's adherence to the
World Trade Organization (WTO) principles in the Technical Barriers to Trade (TBT), see www.iso.org/
iso/foreword.html.
This document was prepared by Technical Committee ISO/TC 130, Graphic technology.
This second edition cancels and replaces the first edition (ISO 12640-3:2007), of which it constitutes a
minor revision. The changes are as follows:
— CIE Publication 15:2004 has been changed to CIE Publication 15 Colorimetry;
— in 3.4, the definition of colour space has been updated based on revision to CIE Publication 17 in
2020;
— in the Bibliography, CIE S 17:2020 ILV has been updated to International lighting vocabulary, 2nd
edition.
A list of all parts in the ISO 12640 series can be found on the ISO website.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www.iso.org/members.html.
iv
Introduction
0.1  Need for standard digital test images
Standard test images provide a set of data that can be used for any of the following tasks:
— evaluating the colour reproduction of imaging systems;
— evaluating colour image output devices;
— evaluating the effect of image processing algorithms applied to the images;
— evaluating the coding technologies necessary for the storage and transmission of high-definition
image data.
Because they exist as standard, well-defined, high-quality image data sets, typical of the range of image
content commonly encountered, they enable users to be confident that the images will produce good
quality reproductions, if properly rendered, and that they provide a reasonable test of the evaluation
task being undertaken. No limited set of images can fully test any system, but the sets provided give
as reasonable a test as can be expected from a limited image set. Furthermore, the existence of a
standard image data set enables users in different locations to produce comparisons without the need
to exchange images prior to reproduction.
However, different applications require that the standard image data be provided in different image
states using different image encodings (see ISO 22028-1). The user needs to select those appropriate to
the evaluation task being undertaken. Whilst transformation of the image data to another image state
is always possible, there is, in general, no agreement amongst experts as to how this is best done. Thus,
it has been considered preferable to provide data in three different image states in the various parts of
ISO 12640.
ISO 12640-1 provides a set of 8-bits-per-channel data that is defined in terms of CMYK dot percentages.
The colours resulting from reproduction of CMYK data are strictly defined only at the time of printing
and, as such, the data are only applicable to evaluation of CMYK printing applications. Transformations
to other image states and colour encodings are not necessarily well defined. In fact, the data might not
even be useful for CMYK printing processes different from those typically found in traditional graphic
arts applications as the image data are defined to produce “pleasing” images when reproduced on
systems using “typical” inks and producing “typical” tone value rendering. Printing systems that use
inks of a distinctly different colour, or produce a very different tone value rendering, will not reproduce
them as pleasing images without a well-defined colour transformation. Moreover, with a bit depth of
only 8 bits per channel, any colour transformation employed will probably introduce artefacts.
ISO 12640-2 provides a set of test image data encoded both as XYZ tristimulus values with a depth of
16 bits per channel and as sRGB (defined in IEC 61966-2-1) with a bit depth of 8 bits per channel. (The
higher bit depth for the XYZ encoding is necessary because of the perceptual non-uniformity of the
XYZ colour space.) Both sets of data are optimized for viewing on a reference sRGB CRT display in the
reference sRGB viewing environment, and relative to CIE standard illuminant D65 for which the XYZ
values were computed. The images are mainly designed to be used on systems utilizing sRGB as the
reference encoding, and as such are mainly applicable to the consumer market and those systems for
which the colour monitor is the “hub” device. Although such systems are used for some applications
in the graphic arts industry, sRGB is by no means the most common image encoding. Furthermore, a
particular drawback is the fact that the sRGB colour gamut is quite different in shape than the colour
gamut of typical offset printing. This difference can necessitate fairly aggressive colour re-rendering to
produce optimal prints from sRGB image data.
In order to be useful for applications where large, print-referred output gamuts are encountered,
common in graphic technology and photography, it was felt that it would be desirable to produce an
image set in which some colours are permitted to be encoded close to the boundary of the full colour
gamut attainable with surface colours. Furthermore, from the perspective of colour management it
is advantageous if the images are referenced to illuminant D50, which is the predominant reference
v
illuminant used in graphic arts and photography, both for viewing and measurement. For this reason it
has also become the predominant reference illuminant for most colour management applications.
The purpose of this document is, therefore, to provide a test image data set with a large colour gamut
related to illuminant D50. The bit depth of the natural images is 16 bits per channel, while the colour
charts and vignettes are 8 bits per channel.
0.2 Definition of the reference colour gamut
The reference colour gamut defined for this document originated from three quite separate sources.
However, it was noted that there was considerable similarity between the three. One definition came
from work within ISO/TC 130 itself, and this arose by consideration of various sets of published data,
which together were taken to define the colour gamut of surface colours. The other definitions arose
from work within Hewlett-Packard, which was focused on the colour gamut obtainable by printing, and
that of a group of German photographic printing experts. The similarity of these led to the conclusion
that it is desirable to reconcile them into a single gamut that would be taken as the reference colour
gamut for this document. Full details of the reference colour gamut and its derivation are given in
Annex B.
0.3  Characteristics of the test images
The performance of any colour reproduction system is normally evaluated both subjectively (by
viewing the final output image) and objectively (by measurement of control elements). This requirement
dictated that the test images include both natural scenes (pictures) and synthetic images (colour charts
and colour vignettes). Because the results of subjective image evaluation are strongly affected by the
image content, it was important to ensure that the natural images were of high quality and contained
diverse subject matter. However, by requiring images to look natural, it is difficult within a single,
relatively small sample set to produce elements in the scene that contain the subtle colour differences
required in such test images and that cover the full reference colour gamut defined. It is also important
to have some images that contain subtle differences in near-neutral colours. Thus, while most images
contain colours that extend to the gamut boundary, this is often only for a limited range of hues in each
image. The full reference colour gamut can only be explored by utilizing the synthetic colour chart.
A survey was conducted of all ISO/TC 130 member countries to identify desirable image content and
to solicit submission of suitable images for consideration. The image set that resulted consists of eight
natural images, eight colour charts and two colour vignettes. The natural images include flesh tones,
images with detail in the extreme highlights or shadows, neutral colours, brown and wood-tone colours
that are often difficult to reproduce, memory colours, complicated geometric shapes, fine detail, and
highlight and shadow vignettes. The colour charts and colour vignettes show the reference colour
gamut (in CIE Lab colour space) in cross-sections for 16 and 8 hue angles, respectively.
0.4  File format of the digital test images
* * *
All of the images consist of pixel interleaved data (L then a then b ) with the data origin at the upper
left of the image, as viewed naturally, and organized by rows. These data are available as individual
files, which are a normative part of this part of this document. The image file format is as specified
in ISO 12639:2004, Annex H, with BitsPerSample set to 16, 16, 16. The images can be imported and
manipulated as necessary by a wide variety of imaging software tools and platforms commonly in
general use in the industry. See Annex D for details of the TIFF header.
1)
All colour charts and vignettes consist of files in Adobe® PDF format.
1) This information is given for the convenience of users of this document and does not constitute an endorsement
by ISO of the product named.
vi
INTERNATIONAL STANDARD ISO 12640-3:2022(E)
Graphic technology — Prepress digital data exchange —
Part 3:
CIELAB standard colour image data (CIELAB/SCID)
1 Scope
This document specifies a set of standard large gamut colour images (encoded as 16-bit CIELAB digital
data) that can be used for the evaluation of changes in image quality during coding, image processing
(including transformation, compression and decompression), displaying on a colour monitor and
printing. These images can be used for research, testing and assessing of output systems such as
printers, colour management systems and colour profiles.
2 Normative references
The following documents are referred to in the text in such a way that some or all of their content
constitutes requirements of this document. For dated references, only the edition cited applies. For
undated references, the latest edition of the referenced document (including any amendments) applies.
ISO 12639:2004, Graphic technology — Prepress digital data exchange — Tag image file format for image
technology (TIFF/IT)
ISO 13655, Graphic technology — Spectral measurement and colorimetric computation for graphic arts
images
ISO 22028-1, Photography and graphic technology — Extended colour encodings for digital image storage,
manipulation and interchange — Part 1: Architecture and requirements
PDF Reference: Adobe Portable Document Format, Version 1.4 3rd edition., Adobe Systems Incorporated,
(ISBN 0-201-75839-3)
3 Terms and definitions
For the purposes of this document, the following terms and definitions apply.
ISO and IEC maintain terminology databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at https:// www .electropedia .org/
3.1
checksum
sum of the digits in a file that can be used to check if a file has been transferred properly
Note 1 to entry: Often, only the least significant bits are summed.
3.2
colour gamut
solid in a colour space, consisting of all those colours that are present in a specific scene, artwork,
photograph, photomechanical or other reproduction; or are capable of being created using a particular
output device and/or medium
[SOURCE: ISO 12640-4:2011]
3.3
colour sequence
order in which the colours are stored in a data file
3.4
colour space
geometric representation of colour in space
Note 1 to entry: A colour space is usually of three dimensions
Note 2 to entry: This entry was numbered 845-03-25 in IEC 60050-45:1987
[SOURCE: CIE S 17:2020 ILV: International Lighting Vocabulary, 2nd edition, modified]
3.5
colour value
numeric values associated with each of the pixels
3.6
data range
range of integers for a given variable in between a minimal and maximal value
3.7
global colour change
change to the colours in an image (often selectively by colour region) applied consistently to all parts of
the image
Note 1 to entry: This is in contrast to a local colour change where selected spatial areas of an image are changed
separately from the rest of the image area.
3.8
orientation
specifies the origin and direction of the first line of data with respect to the image content as viewed by
the end user
Note 1 to entry: The codes used to specify orientation are contained in ISO 12639.
3.9
pixel
smallest discrete picture element in a digital image file
3.10
pixel interleaving
* * *
colour data organized such that the L , a and b colour space values for one pixel are followed by the
same sequence of colour values for the next pixel; the specific order of colour components is determined
by the ColorSequence tag as defined in ISO 12639
Note 1 to entry: Other forms of colour data interleaving are line and plane.
4 Requirements
This document consists of the images contained in the 18 image data files which are part of this
document. Their file names are listed in Table 4. The image characteristics of these data are described
in Clause 5 and the electronic data structure in Clause 6.
5 Data description and definition
5.1 Data set definition
The set of standard colour image data consists of eight natural (photographed) images and ten synthetic
images created digitally on a computer. The synthetic images consist of eight colour charts consisting
of various patches, each 10 mm square, and two colour vignettes. The natural images are identified as
N1 to N8, and each of them also has a descriptive name derived from the picture content (e.g. bride and
groom). The synthetic images are identified as CC1 to CC8, CV1 and CV2.
The images are identified by the designation CIELAB/SCID. The co-ordinates of the text insertion are
provided in Annex E.
NOTE The image set defined in this document is based on the large gamut defined in Annex B. Image sets
contained in other document are based on different gamuts and can be more suitable for use in evaluation of
other applications.
5.2 Colour encoding used in this document
5.2.1 Image data encoding
The image data encoding shall be in accordance with the requirements of ISO 22028-1.
[10]
The image data are the desired CIELAB colorimetry defined by CIE Publication 15 , and measured in
accordance with ISO 13655, of reproductions of the images on the reference medium, with the reference
medium white point selected as the colour space white point. The image data are output-referred,
[11]
having been rendered to the reference medium of the ICC.1:2004-04 perceptual rendering intent,
which is defined as a hypothetical print on a substrate specified to have a neutral reflectance of 89 %
(the reference medium white point) and the darkest printable colour on this medium is assumed to have
a neutral reflectance of 0,347 31 % of the substrate reflectance (the reference medium black point). The
rendering target colour gamut for the reference medium is specified in Annex B. The reference viewing
environment is based on standard viewing condition P2, as specified for graphic arts and photography
in ISO 3664, but extended in the following way: the surfaces immediately surrounding the image are
assumed to be a uniform matt grey with a reflectance of 20 %. The reference viewing environment is
also assumed to have a viewing flare of 0,75 % of the luminance of the reference white. The CIELAB
image data are encoded as specified in 5.2.3 and 5.2.4.
5.2.2 Image data arrangement
* * *
The image data are pixel-interleaved in the colour sequence of L then a then b (16 bits) for the natural
* * *
images and L then a then b (8 bits) for the colour charts and the vignettes. The arrangement of data
follows the scanning of each image from the upper left corner to the upper right, then moving to the
next lower horizontal line. The resolution is 12 pixels/mm for every natural image.
5.2.3 CIELAB image data (16 bits per channel)
The CIELAB data for the natural images are encoded as 16-bit integers per channel, derived by
* * *
multiplying the L , a and b values for each pixel with the corresponding value for the data range.
*
 
L
*
L =×round 65535
 
16bit
 
 
* *
aa=×round()256 (1)
16bit
* *
b =×round()256 b
16biit
* * * * * *
where L , a and b represent normalised 16-bit values of L , a and b .
16bit 16bit 16bit
The data range of the values is:
* * *
L ∈{0.100}, a ∈{−128.+127} and b ∈{−128.+127},
* * * * *
L ∈{0.65 535}, a ∈{−32 768.+32 512} and b ∈{−32 768.+32 512}, a and b are
16bit 16bit 16bit 16bit 16bit
signed integers.
NOTE −32 678 = −128 × 256, and 32 512 = 127 × 256.
5.2.4 CIELAB image data (8 bits per channel)
The CIELAB data for the colour charts are encoded as 8-bit integers per channel, derived by multiplying
* * *
the L , a and b values for each pixel with the corresponding value for the data range.
*
 
L
*
L =×round 255
 
8bit
 
 
* *
aa=round() (2)
8bit
* *
b =round(bb )
8bit
* * * * * *
where L , a and b represent normalised 8 bit values of L , a and b .
8bit 8bit 8bit
The data range of the values is:
* * *
L ∈{0.100}, a ∈{−128.+127} and b ∈{−128.+127},
* * * * *
L ∈{0.255}, a ∈{−128.+127} and b ∈{−128.+127}, the a and b are signed integers.
8bit 8bit 8bit 8bit 8bit
5.3 Natural images
The characteristics of the eight natural images, shown in Figure 1, are given in Table 1.
Table 1 — Characteristics for natural images
Parameter Characteristics
Resolution 12 pixels/mm
* * *
Colour values 16 bits/channel L , a and b , with respect to illuminant D50 (defined as
* * *
media-relative, i.e. such that a white in the image has the L , a and b
values of 100, 0, 0)
File format ISO 12639:2004, Annex H, with BitsPerSample set to 16, 16, 16
This format also readable with TIFF 6.0 with extension, photometric
interpretation tag 8, CIELAB, signed encoding.
Label on image CIELAB/SCID
Image data orientation Horizontal scanning starting from top left and ending at bottom right
NOTE The natural images have been colour-rendered to produce the desired image colorimetry on the
reference print medium as described in Annex B. For the most part, the image colours will be within the reference
colour gamut. However, it is possible for some image colours to be slightly outside (this is somewhat dependent
on how the convex hull of the gamut is constructed). It is sometimes necessary to gamut map the results of colour
rendering and re-rendering processing to exactly fit the destination device colour gamut.
The description and typical usage of the natural images are given in Table 2. The descriptive names
of these images are given following the identification code. Figure 1 shows reduced size sRGB
reproductions of the natural images. Statistical and gamut data for each of the natural images are
* * *
shown in Annex F, as histograms of the L values and a versus b plots for each image, respectively.
Table 2 — Natural images
Name Aspect, image Description and typical usage
size
N1 Bride and groom Horizontal, Image of a bride wearing white clothes and groom wearing black
2 560 × 2 048 clothes. Used to evaluate the rendering of human skin tones and neu-
pixels tral colours, especially highlights and shadows.
N2 People Horizontal, Image consisting of five people wearing colourful clothes, sitting on a
2 560 × 2 048 dark leather couch. Used to evaluate the colour rendering of extreme-
pixels ly colourful objects in the presence of skin tones and neutrals.
N3 Cashew nuts Vertical, Image of dried fruits and filled containers used to evaluate tonal
2 048 × 2 560 and colour rendering, in particular adjustments for grey component
pixels replacement.
N4 Meal Horizontal, Image with widely recognizable cooked food and pastel colours. Used
2 560 × 2 048 to evaluate high-key tonal rendering and food memory colours.
pixels
N5 Mandolin Vertical, Image of goods, including metallic objects, used to evaluate the
2 048 × 2 560 reproduction of colours, as well as the reproduction of the lustrous
pixels appearances of metallic objects.
N6 Tailor scene Horizontal, Still-life image of textile used to evaluate the tone reproduction in a
2 560 × 2 048 range of neutrals and textile structures (object moiré).
pixels
N7 Wool Horizontal, Image of different coloured balls of wool used to evaluate the repro-
2 560 × 2 048 duction of details in highly chromatic areas.
pixels
N8 Fruits Square, Image of a range of fruits and vegetables. The memory colours of
2 024 × 2 024 strawberries, oranges, lemons, green grapes, apples, pears, tomatoes
pixels and bell peppers are particularly suitable for the evaluation of the
naturalness of colour re-rendering processes.
a) N1 Bride and groom b) N2 People
c) N3 Cashew nuts d) N4 Meal
e) N5 Mandolin f) N6 Tailor scene
g) N7 Wool h) N8 Fruits
Figure 1 — Reduced size sRGB reproductions of the natural images
5.4 Synthetic images
5.4.1 Colour charts
There are eight colour charts, each of which consists of a number of colour patches that sample the
reference colour gamut. Each chart contains a number of patches at two hue angles, with each pair
separated by 180°. The design intent was to provide samples at hue angle intervals of 22,5° (from 0° to
337,5°). However, the limitations imposed by 8-bit data means that the intended hue angles can only be
approximated to within ±1°.
*
The colour charts show all the samples within the reference colour gamut at L intervals of 10 (from
* * * *
* *
L = 0 to L = 100) and C intervals of 10 (from C = 0 to the C value above the maximum C value
ab ab ab ab
*
*
for that L value from the reference colour gamut). These maximum C values are given in Table 3, and
ab
are derived from the reference colour gamut described in Annex B of this document, rounded to the
nearest value. Again, the limitations of the 8-bit data mean that the intended values shown in Table 3
*
*
can only be approximated. The L values achieved are to within ±0,2, and the C values to within ±1
ab
*
C
ab .
*
*
Because of this design, the maximum and minimum L values, and maximum C values, in the charts
ab
lie outside the reference colour gamut. This ensures that the whole of the reference colour gamut can be
evaluated (within the patch sampling limitations of the charts). The CIELAB values for each patch in the
charts are given in Annex G.
*
NOTE Applications that show the CIELAB values when these files are opened are likely to show the L values
as the integer value given in Table 3, despite the limited accuracy caused by the 8-bit resolution. Thus, these are
the values quoted in Annex G.
The size of each chart is 275 mm × 137 mm. The size of each colour patch is 10 mm × 10 mm and the
*
distance between adjacent patches is 1 mm. Each chart has a grey background of L = 80.
Figure 2 shows reduced size reproductions of the eight colour charts.
*
*
Table 3 — Colour charts: Maximum chroma value ( C ) with respect to lightness (L ) and hue
ab
angle (h )
ab
* *
Maximum C for L =
ab
h
ab
10 20 30 40 50 60 70 80 90
0° 26 52 74 91 91 82 67 47 25
22,5° 23 50 73 94 95 86 71 51 28
45° 21 45 70 91 100 99 88 67 37
67,5° 16 35 53 70 85 97 104 105 69
90° 14 30 47 62 75 88 101 112 120
112,5° 14 31 47 64 77 90 100 105 96
135° 17 37 56 76 91 99 96 81 50
157,5° 20 43 67 90 99 93 80 60 34
180° 20 41 62 81 92 86 71 52 28
202,5° 20 38 55 69 81 77 66 49 27
225° 20 38 53 65 76 72 62 47 27
247,5° 24 42 58 69 72 64 53 39 21
270° 32 55 70 75 70 61 49 34 19
292,5° 55 85 90 85 76 64 50 35 18
315° 46 85 101 97 87 73 58 41 21
337,5° 31 63 85 102 99 88 70 49 26
Figure 2 — Reduced size sRGB reproductions of the colour charts CC1 to CC8
5.4.2 Vignettes
The vignettes were built as two sets, a full and a reduced colour gamut set. All vignettes are defined for
eight hue angles, at 45° intervals between 0° and 315°, inclusive. For the full reference colour gamut set,
*
*
the first vignette at each hue angle is linearly scaled from black (L = 3,137 3, C = 0) to the maximum
ab
chroma of the reference colour gamut at the corresponding level of lightness as shown in Table 4. The
*
*
second vignette for each hue angle has the constant L value given in Table 4 from grey ( C = 0) to the
ab
* *
* *
L and maximum C values given in Table 4. The third vignette is scaled from white (L = 100, C = 0)
ab ab
*
*
to the L and maximum C values given in Table 4.
ab
The reduced reference colour gamut set consists of the same arrangement, except that the minimum
* *
and maximum lightness is restricted to L = 10 and L = 90, respectively, and the chroma values to 85 %
*
of the reference colour gamut for the same levels of lightness L as shown in Table 4.
*
NOTE The 8-bit resolution necessitates some of the C values being rounded.
ab
The size of the vignettes is 250 mm in length and 10 mm in width. The vignettes are calculated in
*
4 096 steps. They are arranged to a set on a grey background with L = 80. The size of one set is
308 mm × 263 mm.
Figure 3 shows a reduced size reproduction of the vignette CV1.
Table 4 — Maximum and reduced chroma values and corresponding level of lightness for 8 hue
angles
Hue angle
h
Parameter ab
0° 45° 90° 135° 180° 225° 270° 315°
*
Lightness (L ) 48 52 95 65 50 51 37 31
*
Maximum chroma ( C
ab
92 101 123 100 92 75 76 100
)
* *
Colour values (a , b ) (92, 0) (71, 71) (0, 123) (−71, 71) (−92, 0) (−53, −53) (0, −76) (71, −71)
*
78 86 105 85 78 64 65 85
Reduced chroma ( C )
ab
* *
Colour values (a , b ) (78, 0) (61, 61) (0, 105) (−60, 60) (−78, 0) (−45, −45) (0, −65) (60, −60)
Figure 3 — Reduced size sRGB reproduction of the maximum chroma vignette CV1
5.4.3 Synthetic image format
The colour charts and vignettes are written in the PDF 1.4 format.
6 Electronic data
6.1 Image file characteristics
Image data are contained in 18 data files that are included in this document. File names correspond to
the image IDs as described in 5.3 and 5.4. Table 5 shows the file name, size, colour values and descriptive
name of each data file, as well as the pixel height and width of each image. The file size shown represents
the file as recorded and includes headers, etc. Image height and width are provided for raster files. The
checksums given in Annex C shall be used to check the data integrity.
The restrictions on the use of these image data files shall be in accordance Annex A.
Table 5 — Image file characteristics
File size Height Width Colour values
File name Descriptive name
* * *
bytes pixels pixels L , a , b
N1.TIF 31 458 304 2 048 2 560 Three 16-bit values Bride and groom
N2.TIF 31 458 304 2 048 2 560 Three 16-bit values People
N3.TIF 31 458 304 2 560 2 048 Three 16-bit values Cashew nuts
N4.TIF 31 458 304 2 048 2 560 Three 16-bit values Meal
N5.TIF 31 458 304 2 560 2 048 Three 16-bit values Mandolin
N6.TIF 31 458 304 2 048 2 560 Three 16-bit values Tailor scene
N7.TIF 31 458 304 2 048 2 560 Three 16-bit values Wool
N8.TIF 24 580 480 2 024 2 024 Three 16-bit values Fruits
CC1.PDF 12 507 — — Three 8-bit values Chart h 0°/180°
ab
CC2.PDF 14 001 — — Three 8-bit values Chart h 22,5°/202,5°
ab
CC3.PDF 13 869 — — Three 8-bit values Chart h 45°/225°
ab
CC4.PDF 13 982 — — Three 8-bit values Chart h 67,5°/247,5°
ab
CC5.PDF 12 377 — — Three 8-bit values Chart h 90°/270°
ab
CC6.PDF 14 115 — — Three 8-bit values Chart h 112,5°/292,5°
ab
CC7.PDF 14 057 — — Three 8-bit values Chart h 135°/315°
ab
CC8.PDF 14 186 — — Three 8-bit values Chart h 157,5°/337,5°
ab
CV1.PDF 90 015 — — Three 8-bit values Vignette maximal chroma
CV2.PDF 89 213 — — Three 8-bit values Vignette reduced chroma
6.2 Image file formats
The natural images N1.TIF to N8.TIF are recorded as 16-bit CIELAB data in accordance with ISO 12639
(also readable by TIFF 6.0 extended, photometric interpretation tag 8, CIELAB, signed encoding). The
colour charts and vignettes CC1.PDF to CC8.PDF and CV1.PDF and CV2.PDF are recorded as PDF files in
accordance with PDF 1.4.
Annex D shows the file header of image N1.TIF.
Annex A
(normative)
Guidance for use of digital data
A.1 General
To ensure that these images can be used successfully for the testing and comparisons for which they
are intended, all use shall conform to the procedures and guidelines described in A.2 and A.3.
A.2 Guidelines for use
A.2.1 Reproduction
All reproductions of these images shall contain an annotation identifying this document as the data
source and shall retain the colour space identifier included in the image data.
A.2.2 Modification
Any images created by modification of these data (derivative images) shall also have a visible identifier
added within the image. The accompanying material shall include a tabulation of the steps used to
modify the image data including all editing steps used as well as any data rescaling or interpolation.
A.2.3 Colour manipulation
Any colour or tonal manipulation of these images shall be restricted to “global” colour changes only.
A.2.4 Cropping
Cropping of these images shall be permitted so long as the appropriate image colour space identifier is
included as part of, or with, the images.
A.3 Guidelines for distribution and sharing
A.3.1 General
Many of the intended uses of these images require that they be used at several locations and/or by
several participants in test programs. The following uses have been interpreted to be acceptable and
allowed by ISO.
A.3.2 For-profit sale
Neither the data, nor images printed from these data, shall be sold “for-profit” except as defined in A.3.3.
A.3.3 Test and evaluation packages
It shall be permitted to include the data corresponding to these images, or derivations of these images,
as part of test and evaluation packages to be sold or provided free of charge where an authentic copy of
this document is included as part of the complete package.
NOTE It is recognized that certain test and evaluation packages that will make use of these images might
need to embed the data to be used within other data processing procedures. The inclusion of an authentic copy
of this document, obtained from the appropriate standards agency, as part of the package allows the inclusion of
similar or derived data as needed within the package.
A.3.4 Test and evaluation programs
Copies of these data files, or derivative files, may be exchanged between participants in test and
evaluation programs. The sponsoring organization shall be capable of showing ownership of an
authentic copy of this document.
A.3.5 Reports
It shall be permitted to display these images as part of the report of test programs, or in advertisements,
as long as the organization sponsoring the display is in possession of an authentic copy of this document.
Annex B
(informative)
Definition of the reference colour gamut
B.1 Development of the reference colour gamut
The reference colour gamut defined for this document was derived from three different colour gamuts
developed quite independently, but for similar reasons. The first was developed within ISO/TC 130 in
order to provide an estimate of the gamut of surface colours for the purposes of this document. The
second was produced by Hewlett-Packard as the gamut of colours produced by all colour printers,
during their own product development work, and offered to the International Color Consortium (ICC)
as a reference colour gamut for the Profile Connection Space for perceptual renderings defined in
ICC.1:2004-04. It was also offered to ISO/TC 130 for consideration in the definition of the reference
gamut required for this document. During the development of this standard another useful gamut was
brought to the attention of the committee. This is a gamut known as PhotoGamutRGB, which is based
on measurement of the results obtained from silver halide printers used for producing photographic
prints from digital photographs.
On review it became clear that, although there were some differences, there was also considerable
similarity between these gamuts. So, it was agreed by the committee that the data from the three
sources could be reconciled in order to produce the single reference colour gamut defined for this
document. This reconciliation is described in B.5. However, for information purposes, the derivation of
each of the gamuts is briefly described.
B.2 Gamut of surface colours
The initial specification for the reference colour gamut for this document was obtained by finding the
maximum gamut obtainable with surface colours, according to the published data available to the
[8]
committee in 1998. The main sources of this data were the Pointer gamut, data for Pantone colours,
[6]
the SOCS data and data for photographic media available to the committee. Some additional samples
were also measured.
[8]
In 1980, Pointer published an investigation of the gamut of real surface colours , in which colour data
was analysed from the following sources:
— 768 colours from the Munsell Limit Color Cascade;
— 310 colours from the Matte Munsell Atlas;
— 1 393 colours from ink and paint samples, textiles, coloured plastics and papers (measured in the
[8]
investigation );
— 1 618 colours describing flower colours (tabulated by the Royal Horticultural Society).
Altogether, the colour coordinates of 4 089 colours were available.
[8]
Pointer combined these colour data and published the maximum chroma value at 36 hue angles
and 16 lightness levels. However, the Pointer gamut data refer to CIE standard illuminant C, whereas
the committee needed the gamut with respect to illuminant D50. The Pointer gamut boundary data
was therefore converted to that for D50 using the Bradford chromatic adaptation conversion used in
CIECAM97s. The D50 Pointer data were then combined with other colour data exhibiting high chroma
values. In particular, the data for 1 025 Pantone colours, a series of new colour data measured from
printed samples and the colorimetric data from the SOCS data set (see ISO/TR 16066) were included.
From these XYZ data the colour gamut was calculated as a convex hull and then transformed into the
CIELAB colour space. Table B.1 shows the maximum chroma value obtainable with surface colours for
36 hue angles and 19 lightness levels calculated as a convex hull for illuminant D50 and the 1931, 2°,
...