ISO 12640-5:2013

Graphic technology — Prepress digital data exchange — Part 5: Scene-referred standard colour image data (RIMM/SCID)

ISO 12640-5:2013 specifies a set of standard scene-referred colour images (encoded as 16-bit RIMM RGB digital data) that can be used to evaluate transforms from a scene-referred image state to an output-referred image state (colour rendering transforms). They can be used for research, testing and assessing colour rendering transforms, in systems such as digital cameras, camera raw processing applications, colour management systems, colour profiles, and output devices such as displays and printers.

Technologie graphique — Échange de données numériques de préimpression — Partie 5: Données d'image standard en couleurs montrées en référence par scène (RIMM/SCID)

General Information

• Status: Published
• Publication Date: 10-Dec-2013
• ICS: 35.240.30 - IT applications in information, documentation and publishing
• ICS: 37.100.99 - Other standards related to graphic technology
• Technical Committee: ISO/TC 130 - Graphic technology
• Drafting Committee: ISO/TC 130 - Graphic technology
• Current Stage: 9093 - International Standard confirmed
• Start Date: 02-Jul-2024
• Completion Date: 12-Feb-2026

Overview

ISO 12640-5:2013 - Graphic technology: Prepress digital data exchange, Part 5 (RIMM/SCID) specifies a set of standard scene-referred colour image data encoded as 16-bit RIMM RGB. These reference images (natural and synthetic) are provided to support objective evaluation of colour rendering transforms that convert scene-referred images to output-referred images. The package distributed with the standard includes the PDF specification plus TIFF image files (e.g., NP01RGB…NP27RGB, NS01RGB…NS17RGB, S1RGB…S3RGB) intended for research, testing and validation.

Key topics and requirements

• Scene-referred standard images: A curated collection of images representing natural and synthetic scenes, encoded as 16-bit RIMM RGB (RIMM/SCID).
• Purpose: Evaluate and compare colour rendering transforms (scene → output) across workflows and devices.
• File formats and distribution: Images supplied as TIFF files on media (DVD/zip); installation notes emphasize preserving file structures and consulting the Readme for correct use.
• Usage constraints: Copyright-protected distribution - reproduction or transmission requires permission from ISO; users should follow licensing guidance included with the standard.
• Test consistency: Provides a common, standardized input dataset to enable reproducible testing and comparative assessments of colour pipelines.

Applications

ISO 12640-5 supports practical tasks where consistent scene-referred input data are required:

• Evaluating camera colour pipelines and raw converters for accurate scene-to-output rendering.
• Developing and validating colour management systems and transform algorithms.
• Creating or testing ICC profiles and device characterization routines for displays, printers and proofing systems.
• Research and academic studies in colour science and image processing.
• Quality assurance and regression testing for imaging software and hardware.

Who uses this standard

• Camera manufacturers and imaging-software developers (raw converters, camera pipeline engineers).
• Colour scientists and researchers conducting objective comparisons and experiments.
• Colour management specialists and prepress professionals validating profiles and rendering intents.
• Display and print device vendors performing end-to-end colour pipeline assessments.

Related standards

• Part of the broader ISO 12640 series (prepress digital data exchange).
• Commonly used alongside colour management specifications (e.g., ICC profile workflows) and ISO colourimetry references when performing end-to-end evaluations.

ISO 12640-5:2013 provides a practical, standardized image dataset (RIMM/SCID) to improve reproducibility, comparability and accuracy when testing colour rendering transforms across cameras, software and output devices.

Standards Content (Sample)

INTERNATIONAL ISO
STANDARD 12640-5
First edition
2013-12-15
Graphic technology — Prepress digital
data exchange —
Part 5:
Scene-referred standard colour image
data (RIMM/SCID)
Technologie graphique — Échange de données numériques de
préimpression —
Partie 5: Données d’image standard en couleurs montrées en
référence par scène (RIMM/SCID)
Reference number
©
ISO 2013
© ISO 2013
All rights reserved. Unless otherwise specified, 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
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail [e-mail]
Web www.iso.org
Published in Switzerland
ii © ISO 2013 – All rights reserved

Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Data description . 4
4.1 General . 4
4.2 Data set definition . 4
4.3 Image data arrangement . 4
4.4 Data colour encoding . 4
4.5 Natural images . 5
4.6 Synthetic images .19
5 Electronic data .23
Annex A (normative) Guidance for use of digital data .25
Annex B (normative) Check-sum data .27
Annex C (informative) Typical TIFF/IT file header used for image files .29
Annex D (informative) Label text insertion .31
Annex E (informative) Histogram and colour gamut .33
Bibliography .49
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 on the meaning of ISO specific terms and expressions related to conformity
assessment, as well as information about ISO’s adherence to the WTO principles in the Technical Barriers
to Trade (TBT) see the following URL: Foreword - Supplementary information
The committee responsible for this document is Technical Committee ISO/TC 130, Graphic technology.
ISO 12640 consists of the following parts, under the general title Graphic technology — Prepress digital
data exchange:
— Part 1: CMYK standard colour image data (CMYK/SCID)
— Part 2: XYZ/sRGB standard colour image data (XYZ/SCID)
— Part 3: CIELAB standard colour image data (CIELAB/SCID)
— Part 4: Wide gamut display-referred standard colour image data (Adobe RGB(1998)/SCID)
— Part 5: Scene-referred standard colour image data (RIMM/SCID)
iv © ISO 2013 – All rights reserved

Introduction
0.1 Need for standard colour image data
Standard colour image data 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, etc.
These standard, well-defined image data sets, are typical of the high quality image content commonly
encountered when capturing and printing images. Users can therefore be confident that the images
should 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 set enables users in different locations to produce comparisons without the need
to exchange images prior to reproduction.
Different applications require that the standard image data be provided in different image states
using different image encodings (see ISO 22028-1), so the user needs to select those appropriate to
the evaluation task being undertaken. While transformation of the image data to another image state
is always possible, there is, in general, no agreement amongst experts as to how this should be done.
Thus, it has been considered preferable to provide data in different image states in the various parts of
ISO 12640. The relationship between image states is shown in Figure 1 along with the applicable parts
of ISO 12640.
Figure 1 — Relationship between image states
ISO 12640-1 provides a set of 8 bits/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 might not be 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/channel, any colour transformation employed might well introduce artefacts.
ISO 12640-2 provides a set of test image data encoded both as XYZ values with each channel scaled to the
range 0-65535, and as sRGB (defined in IEC 61966-2-1), with a bit depth of 8 bits/channel. (The higher bit
depth for the XYZ encoding is necessary because of the perceptual non-uniformity of the linear colour
space.) Both sets of data are optimized for viewing on a reference sRGB display in the reference sRGB
viewing environment, and relative to CIE standard illuminant D65 for which the XYZ tristimulus values
were computed prior to scaling. The images are mainly designed to be used on systems utilizing sRGB
as the reference encoding, and as such are primarily applicable to systems for which a colour monitor
similar to the sRGB reference display is the “hub” device. Although such systems are used for consumer
photography, they are less popular in the graphic arts industry because the sRGB colour gamut is quite
different in shape from the colour gamut of typical offset printing. This difference can necessitate fairly
aggressive colour re-rendering to produce optimal prints from sRGB image data.
ISO 12640-3 provides a set of test image data with a large reflection medium colour gamut, illuminated
using illuminant D50. The bit depth of the natural images is 16 bits/channel, while the colour charts and
vi © ISO 2013 – All rights reserved

vignettes are 8 bits/channel. 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 attained 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 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.
ISO 12640-4 provides a set of wide-gamut test image data encoded as Adobe RGB with a bit depth of
16 bits/channel. These data are optimized for viewing on a reference Adobe RGB display in the reference
Adobe RGB viewing environment (defined in the Adobe RGB (1998) Colour Image Encoding specification).
The images are designed to be used mainly on systems utilizing Adobe RGB as the reference encoding,
and as such are mainly applicable to the professional market and those systems for which the wide gamut
colour monitor is the “hub” device. Such workflows are popular among professional photographers,
and are increasingly used in the graphic arts. The Adobe RGB reference display colour gamut is closer
to typical offset printing gamuts than the sRGB reference display colour gamut. Adobe RGB encoded
images generally require much less aggressive colour re-rendering going to print than sRGB encoded
images, although this difference can necessitate colour re-rendering between Adobe RGB images and
sRGB images. The purpose of ISO 12640-4 is therefore to provide a test image data set with a larger
colour gamut than sRGB, related to the Adobe RGB wide-gamut display-referred colour space. The bit
depth of the natural images and synthetic images is 16 bits/channel.
The possible wide gamut colour encoding choices considered were Adobe RGB, opRGB (IEC 61966-2-5)
and ROMM RGB (ISO 22028-2). For ISO 12640-4, it was important that the images were well-colour-
rendered to a well-defined large gamut reference display, for which reason Adobe RGB was preferred
over the other two choices. With opRGB, the completeness of the colour rendering is left more ambiguous,
i.e. it is not as clearly output-referred, and the reference medium and viewing conditions are also
slightly different. ROMM RGB (ISO 22028-2) is clearly output-referred, but the reference medium is a
virtual reflection print (the ICC perceptual reference medium), so the image state is identical to that for
ISO 12640-3.
This part of ISO 12640 provides a set of scene-referred test image data encoded as RIMM RGB with a bit
depth of 16 bits/channel. These data are estimates of scene colorimetry obtained by capturing natural
scenes using a variety of digital cameras and transforming the captured raw camera RGB signals to
scene colorimetry estimates. The accuracy of these estimates is influenced by a number of factors
including the degree to which the camera spectral sensitivities approximate human visual system
colour matching functions, the appropriateness of the transformation from raw camera RGB signals to
colorimetry estimates, optical effects such as off-axis decrease in signal, aberrations and flare, and the
noise present in the camera signals. The transformations applied to obtain the colorimetry estimates
were general transformations, i.e. they were not optimized for the spectral characteristics of each scene.
Consequently, there can in some cases be significant errors in the estimates. The image state of these
data is scene-referred because no attempt has been made to colour render the data to produce a pleasing
reproduction on some output medium. The only processing applied to these data based on visual
evaluation was to select the scene adopted white. This was accomplished by applying gains individually
to the camera channels to achieve the desired white balance, converting to scene-referred, and then
adjusting the overall gain in a linear, scene-referred working space while viewing the image with the
example colour rendering transform specified in ISO/TS 22028-3:2012, Annex A, applied. Different
white balances can be desired in some cases for aesthetic reasons, and different overall gains can be
needed if different colour rendering transforms are used. The images provided in this part of ISO 12640
are mainly applicable for evaluating colour rendering to different output media.
0.2 Characteristics of the test images
The performance of any colour reproduction system will normally be evaluated both subjectively (by
viewing the final output image) and objectively (by measurement of control elements). This requirement
dictates 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, it is difficult within a single, relatively small, sample set to produce
elements in the scene that contain all the subtle colour differences required in test images, and that
span the full range of colours that can be encountered in real scenes. For this reason, synthetic colour
charts are also included. These colour charts are limited by the integer RIMM RGB encoding and by
the spectral locus (for areas where the RIMM RGB encoding extends outside the spectral locus). In the
future, it is proposed to develop a second set of floating point RIMM/SCID which are not limited by the
integer RIMM RGB encoding.
To obtain the images, a survey was conducted of all 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 44 natural images, two colour charts and a series of colour vignettes. The natural images
include flesh tones, hair, foliage, water, sky, flowers and other memory colours in scenes with a variety
of dynamic ranges.
0.3 File format of the digital test images
All of the images consist of pixel interleaved data (R then G then B), with the data origin at the upper
left of the image, as viewed naturally, and organized by rows. These data are included as individual files
within this part of ISO 12640. The image file format is as specified in ISO 12639 (TIFF/IT). A RIMM RGB
ICC profile meeting the requirements of ISO 15076-1 is embedded in each image file.
The images can be imported and manipulated as necessary by a wide variety of commonly used imaging
software tools and platforms in general use in the industry. (See Annex C for details of the TIFF header
and the RIMM RGB ICC profile.)
viii © ISO 2013 – All rights reserved

INTERNATIONAL STANDARD ISO 12640-5:2013(E)
Graphic technology — Prepress digital data exchange —
Part 5:
Scene-referred standard colour image data (RIMM/SCID)
1 Scope
This part of ISO 12640 specifies a set of standard scene-referred colour images (encoded as 16-bit RIMM
RGB digital data) that can be used to evaluate transforms from a scene-referred image state to an output-
referred image state (colour rendering transforms). They can be used for research, testing and assessing
colour rendering transforms, in systems such as digital cameras, camera raw processing applications,
colour management systems, colour profiles, and output devices such as displays and printers.
2 Normative references
The following documents, in whole or in part, are normatively referenced in this document and are
indispensable for its application. 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/TS 22028-3:2012, Photography and graphic technology — Extended colour encodings for digital
image storage, manipulation and interchange — Part 3: Reference input medium metric RGB colour image
encoding (RIMM RGB)
3 Terms and definitions
For the purposes of this document, the following terms and definitions apply.
3.1
additive RGB colour space
colorimetric colour space having three colour primaries (generally red, green and blue) such that CIE
XYZ tristimulus values can be determined from the RGB colour space values by forming a weighted
combination of the CIE XYZ tristimulus values for the individual colour primaries, where the weights are
proportional to the radiometrically linear colour space values for the corresponding colour primaries
Note 1 to entry: A simple linear 3 × 3 transformation can be used to transform between CIE XYZ tristimulus
values and the radiometrically linear colour space values for an additive RGB colour space.
Note 2 to entry: Additive RGB colour spaces are defined by specifying the CIE chromaticity values for a set of
additive RGB primaries and a colour space white point, together with a colour component transfer function.
[SOURCE: ISO 22028-1:2004, 3.3]
3.2
adopted white
spectral radiance distribution as seen by an image capture or measurement device and converted to
colour signals that are considered to be perfectly achromatic and to have an observer adaptive luminance
factor of unity; i.e. colour signals that are considered to correspond to a perfect white diffuser
Note 1 to entry: The adopted white may vary within a scene.
Note 2 to entry: No assumptions should be made concerning the relation between the adapted or adopted white
and measurements of near perfectly reflecting diffusers in a scene, because measurements of such diffusers will
depend on the illumination and viewing geometry, and other elements in the scene that may affect perception.
It is easy to arrange conditions for which a near perfectly reflecting diffuser will appear to be grey or coloured.
[SOURCE: ISO 22028-1:2004, 3.4]
3.3
colour component transfer function
CCTF
single variable, monotonic mathematical function applied individually to one or more colour channels
of a colour space
Note 1 to entry: Colour component transfer functions are frequently used to account for the nonlinear response
of a reference device and/or to improve the visual uniformity of a colour space.
Note 2 to entry: Generally, colour component transfer functions will be nonlinear functions such as a power-law
(i.e. “gamma”) function or a logarithmic function. However, in some cases a linear colour component transfer
function may be used.
[SOURCE: ISO 22028-1:2004, 3.6, modified — Abbreviated term has been added]
3.4
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 capable of being created using a particular
output device and/or medium
[SOURCE: ISO 22028-1:2004, 3.8]
3.5
colour rendering
mapping of image data representing the colour-space coordinates of the elements of a scene to output-
referred image data representing the colour-space coordinates of the elements of a reproduction
Note 1 to entry: Colour rendering generally consists of one or more of the following: compensating for differences
in the input and output viewing conditions, tone scale and gamut mapping to map the scene colours onto the
dynamic range and colour gamut of the reproduction, and applying preference adjustments.
[SOURCE: ISO 22028-1:2004, 3.11]
3.6
colour sequence
order in which the colours are stored in an image data file
3.7
orientation
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.
3.8
output-referred image state
image state associated with image data that represents the colour-space coordinates of the elements of
an image that has undergone colour rendering appropriate for a specified real or virtual output device
and viewing conditions
Note 1 to entry: When the phrase “output-referred” is used as a qualifier to an object, it implies that the object is
in an output-referred image state. For example, output-referred image data are image data in an output-referred
image state.
2 © ISO 2013 – All rights reserved

Note 2 to entry: Output-referred image data are referred to the specified output device and viewing conditions. A
single scene can be colour-rendered to a variety of output-referred representations depending on the anticipated
output viewing conditions, media limitations, and/or artistic intents.
Note 3 to entry: Output-referred image data may become the starting point for a subsequent reproduction process.
For example, sRGB output-referred image data are frequently considered to be the starting point for the colour
re-rendering performed by a printer designed to receive sRGB image data.
[SOURCE: ISO 22028-1:2004, 3.33]
3.9
pixel
smallest discrete picture element in a digital image file
3.10
pixel interleaved
colour data organized such that the RGB colour space values for one pixel are followed by the same
sequence of colour values for the next pixel
Note 1 to entry: The specific order of colour components is determined by the ColourSequence tag as defined in
ISO 12639. Other forms of colour data interleaving are line and plane.
3.11
scene
spectral radiances of a view of the natural world as measured from a specified vantage point in space
and at a specified time
Note 1 to entry: A scene may correspond to an actual view of the natural world or to a computer-generated virtual
scene simulating such a view.
[SOURCE: ISO 22028-1:2004, 3.35]
3.12
scene-referred image state
image state associated with image data that represents estimates of the colour-space coordinates of the
elements of a scene
Note 1 to entry: When the phrase “scene-referred” is used as a qualifier to an object, it implies that the object in
a scene-referred image state. For example, scene-referred image data are image data in a scene-referred image
state.
Note 2 to entry: Scene-referred image data can be determined from raw digital still camera (DSC) image data
before colour rendering is performed. Generally, DSCs do not write scene-referred image data in image files, but
some may do so in a special mode intended for this purpose. Typically, DSCs write standard output-referred image
data where colour rendering has already been performed.
Note 3 to entry: Scene-referred image data typically represent relative scene colorimetry estimates. Absolute scene
colorimetry estimates may be calculated using a scaling factor. The scaling factor can be derived from additional
information such as the image OECF, Fnumber or ApertureValue, and ExposureTime or ShutterSpeedValve tags.
Note 4 to entry: Scene-referred image data may contain inaccuracies due to the dynamic range limitations of the
capture device, noise from various sources, quantization, optical blurring and flare that are not corrected for,
and colour analysis errors due to capture device metamerism. In some cases, these sources of inaccuracy can be
significant.
Note 5 to entry: The transformation from raw DSC image to scene-referred image data depends on the relative
adopted whites selected for the scene and the colour space used to encode the image data. If the chosen scene
adopted white is inappropriate, additional errors will be introduced into the scene-referred image data. These
errors may be correctable if the transform used to produce the scene-referred image data is known, and the
colour encoding used for the incorrect scene-referred image data has adequate precision and dynamic range.
Note 6 to entry: The scene may correspond to an actual view of the natural world, or may be a computer-
generated virtual scene simulating such a view. It may also correspond to a modified scene determined by
applying modifications to an original scene to produce a different desired scene. Any such modifications should
leave the image in a scene-referred image state, and should be done in the context of an expected colour rendering
transform.
[SOURCE: ISO 22028-1:2004, 3.36]
4 Data description
4.1 General
This part of ISO 12640 consists of 47 image data files and specifications of the content of these files
and their use as provided in this part of ISO 12640. The image file names are listed in Table 1, Table 2
and Table 3. The colour image data are encoded in RIMM RGB as specified in ISO/TS 22028-3, using
16 bits/channel and 48 bits/pixel. The image characteristics of these data are described in 4.5 and 4.6,
and the electronic data structure in Clause 5.
4.2 Data set definition
The set of standard colour image data consists of 44 natural images, captured using digital still cameras,
and three synthetic images. The primary set of natural images are identified as NP01 to NP27 and the
secondary set as NS01 to NS17, respectively. Each of them also has a descriptive name derived from the
picture content (e.g. “Falls”). The synthetic images are identified as S1, S2 and S3.
The label “ISO 12640-5 RIMM” is inserted in each image. The coordinates of the text insertion are
provided in Annex D.
4.3 Image data arrangement
The image data are pixel interleaved in the colour sequence of R then G then B (16 bits/channel) for the
images. The image data orientation corresponds to a value of 1 in TAG 274 of ISO 12639 (load from top
left, horizontally; the 0th row represents the visual top of the image and the 0th column represents the
visual left-hand side).
4.4 Data colour encoding
The image data are encoded as RIMM RGB as specified in ISO/TS 22028-3.
The Reference Input Medium Metric RGB (RIMM RGB) encoding is an extended-colour-gamut RGB
colour image encoding of the colorimetry of a scene-referred image. The colorimetry is encoded using an
additive RGB colour space associated with a hypothetical additive colour device having a specified set of
primaries, no cross-talk between the colour channels and a maximum luminance value corresponding
to 200 % of the luminance of the adopted white (i.e. a maximum luminance factor of 2,0).
There are three different precision levels specified for RIMM RGB. The images in this part of ISO 12640
are encoded as RIMM16 RGB for 16 bits/channel (48 bits/pixel) representations.
The scene-referred colorimetry has been adapted to the RIMM RGB encoding white, which has the
chromaticity of CIE Standard Illuminant D (x = 0,345 7, y = 0,358 5). This colorimetry was obtained
50 0 0
as follows.
a) Demosaic the digital camera raw image data where necessary and linearize with respect to scene
radiance, including black frame and estimated flare subtraction.
b) Determine the linear camera raw channel values R , G and B that correspond to the scene
W W W
adopted white.
4 © ISO 2013 – All rights reserved

c) Multiply the linear camera raw image data channels R, G and B by 1/R , 1/G and 1/B , respectively,
W W W
so that values {1, 1, 1} are obtained for the scene adopted white. This results in white-balanced
linear camera raw image data.
d) Determine a scene analysis matrix that converts the white-balanced linear camera raw image data
to linear RIMM RGB image data, where the linear RIMM RGB image data represents estimates of
the scene colorimetry after chromatic adaptation to D50. The matrices determined are typically
camera and scene adopted white specific, and depend on the scene spectral radiance characteristics
assumed. Technical information on how to determine scene analysis matrices is provided in
ISO/TR 17321-2.
e) Apply the scene analysis matrix to the white-balanced linear camera raw image data to produce
linear RIMM RGB image data.
f) Apply the RIMM16 RGB CCTF as specified in ISO/TS 22028-3.
A RIMM RGB ICC profile is embedded in each image file to facilitate use in colour managed systems, and to
provide an example colour rendering for the images using the perceptual rendering intent. The example
colour rendering specified in ISO/TS 22028-3:2012, Annex A, is used in the perceptual transform.
The above steps do not include any compensation for the colour appearance resulting from different
scene illumination levels. Instead, the brightness value for each scene is recorded in Table 2. The
brightness values (BV) are determined from the camera aperture, shutter speed and exposure index
recorded in the raw image file metadata as shown in Formula (1).
BV=+AV TV−SV (1)
where
AV = LOG2[A ], where A is the effective f-number of the camera lens;
TV = LOG2(1/t), where t is the photosite integration (exposure) time in seconds;
SV = LOG2(EI/3), where EI is the exposure index that would have been used if the camera exposure
compensation were zero.
For example, if the exposure index is set on ISO 100 and the exposure compensation is set on minus one
stop, the effective exposure index is 200 and SV = LOG2(200/3) = 6.
The estimated midtone (18 % reflectance) scene luminance level L is calculated from the BV value as
A
shown in Formula (2).
BV
L =×3,7 2 (2)
A
4.5 Natural images
4.5.1 Description
The characteristics of the natural images (orientation, image size, brightness value (BV), average
luminance factor (Y), data range and dynamic range) are shown in Tables 1 and 2. Average Y, data range
and dynamic range are calculated using the Y value of XYZ tristimulus values of an image scaled to be
1/16 of the original size (each dimension 1/16 of the original dimension) using bicubic scaling. The data
range Y is the ratio of maximum Y value to the smallest Y value present in the image that is larger than
zero. The dynamic range is calculated by taking the ratio of the highlight Y value to the shadow Y value,
where the highlight Y value is the value corresponding to 0,999 5 on a cumulative Y histogram, and the
shadow Y value is the value corresponding to 0,005 on a cumulative Y histogram. Other methods for
highlight and shadow Y value estimation may also be used; the method used to calculate the dynamic
range values in Tables 1 and 2 is not intended to be considered a recommended method.
Table 1 shows a primary set and comprises 27 images. Table 2 shows a secondary set and consists of
17 images. It is strongly recommended that all of the images in the primary set be used and, where
necessary, additional images from the secondary set should be used to supplement this set.
The descriptive names of these images are given following the identification code. Two renderings are
shown in Figure 2: a colorimetric conversion of the scene-referred image to sRGB and a perceptual
colour rendering to sRGB.
The 44 natural images shall be interpreting as having the following characteristics:
— Resolution: 24 pixels/mm;
— Colour values: RIMM RGB data consisting of three 16-bit values;
— File format: ISO 12639:2004 (TIFF/IT);
— Label on image: “ISO 12640-5”;
— Image data orientation: load from top left, horizontally.
Table 1 — Primary set of natural images
Name Aspect, image size BV Average Y Data range Y Dynamic range Y
NP01 Falls Vertical, 2 014 × 3 040 pixels 9 1,985 2 621 177,3
NP02 Eiffel Vertical, 2 014 × 3 040 pixels 10 0,322 5 1 513 98,06
NP03 Mickey Vertical, 2 036 × 3 040 pixels 5 0,264 5 10 927 248,1
NP04 Butterfly Horizontal, 4 256 × 2 848 pixels 8 0,103 0 126,4 88,24
NP05 Threads Horizontal, 4 272 × 2 864 pixels 6 0,091 04 29 617 232 647,4
NP06 Fruits Horizontal, 4 272 × 2 864 pixels 6 0,306 0 344,2 173,4
NP07 Canal Horizontal, 4 256 × 2 848 pixels 9 0,125 6 1 145 159,4
NP08 WhiteFlowers Horizontal, 3 872 × 2 592 pixels 8 0,131 5 890,3 168,0
NP09 BarHarborPresunrise Horizontal, 4 284 × 2 408 pixels 7 0,161 7 24 647 243 237,6
NP10 BenJerrys Horizontal, 4 288 × 2 412 pixels 10 0,193 9 3 195 99,08
NP11 DelicateFlowers Horizontal, 4 288 × 2 848 pixels 7 0,342 1 252,6 47,38
NP12 DevilsBathtub Horizontal, 4 288 × 2 412 pixels 11 0,089 18 20 662 314,8
NP13 Exploratorium Horizontal, 4 288 × 2 848 pixels 10 0,155 3 16 405 121,5
NP14 GoldenGate Horizontal, 4 288 × 2 844 pixels −2 0,064 94 3 019 188,2
NP15 HancockSeedField Horizontal, 4 280 × 2 408 pixels 8 0,410 5 9 286 174 39,99
NP16 NiagaraFalls Horizontal, 4 280 × 2 408 pixels 10 0,353 4 458,2 70,59
NP17 RedwoodSunset Horizontal, 4 284 × 2 408 pixels 8 0,172 3 60 083 240 869,8
NP18 Route66Museum Horizontal, 4 288 × 2 848 pixels 3 0,411 3 11 407 44,31
NP19 SouthBranchKingsRiver Vertical, 2 844 × 4 280 pixels 9 0,106 2 64 569 951 304,5
NP20 TupperLake Horizontal, 4 288 × 2 848 pixels 9 0,235 9 60,38 25,59
NP21 Chandelier Horizontal, 3 888 × 2 592 pixels 4 0,091 99 5 178 800,0
NP22 Clock Horizontal, 3 888 × 2 592 pixels 3 0,220 4 42 953 120,1
NP23 Meat Horizontal, 3 888 × 2 592 pixels 3 0,207 1 63 245 207 220,1
NP24 Peacock Horizontal, 3 888 × 2 592 pixels 8 0,185 1 2 280 168,5
NP25 BlueFace Vertical, 2 592 × 3 888 pixels 6 0,184 1 30 925 553 239,84
NP26 PoundPake Horizontal, 3 888 × 2 592 pixels 0 0,158 5 1 031 146,2
NP27 LasVegas Horizontal, 3 888 × 2 592 pixels −3 0,091 99 3 437 414,3
6 © ISO 2013 – All rights reserved

Table 2 — Secondary set of natural images
Name Aspect, image size BV Average Y Data range Y Dynamic range Y
NS01 Maple Horizontal, 3 040 × 2 014 pixels 7 0,156 3 161,6 45,45
NS02 River Horizontal, 3 040 × 2 014 pixels 9 0,258 3 64 879 055 246,9
NS03 Marker Horizontal, 4 272 × 2 864 pixels 7 0,232 5 768,7 78,85
NS04 Flowers Horizontal, 4 272 × 2 864 pixels 5 0,216 7 193,7 92,71
NS05 Fluorescence Horizontal, 4 272 × 2 864 pixels 7 0,478 4 222,1 85,92
NS06 Hotel Horizontal, 3 040 × 2 036 pixels 7 1,756 414,7 182,3
NS07 Pyramid Horizontal, 3 040 × 2 036 pixels 10 0,735 8 18,63 11,74
NS08 Nile Horizontal, 3 040 × 2 036 pixels 9 1,237 242,4 71,72
NS09 Felucca Horizontal, 3 040 × 2 036 pixels 10 0,810 5 115,5 44,46
NS10 CherryBlossom Horizontal, 3 040 × 2 036 pixels 8 0,812 6 84 845 101,1
NS11 Iris Horizontal, 3 040 × 2 036 pixels 9 1,084 8 468 419 77,28
NS12 Wharf Horizontal, 3 040 × 2 036 2 pixels 9 1,128 683,8 174,4
NS13 Hiking Horizontal, 3 872 × 2 592 pixels 9 0,125 6 1 145 107,0
NS14 CadesCove Horizontal, 4 288 × 2 412 pixels 9 0,278 3 3 834 157,0
NS15 FourCornersStorm Horizontal, 4 288 × 2 848 pixels 5 0,212 8 125,3 17,50
NS16 Sunset Horizontal, 3 872 × 2 592 pixels 6 0,145 2 6 096 789 253,7
NS17 Headlight Horizontal, 3 888 × 2 592 pixels 9 0,198 7 81 111 476,3
NP01 Falls
NP02 Eiffel
NP03 Mickey
Figure 2 — Samples of reduced size media-relative colorimetric conversion to sRGB (left) and
colour rendered sRGB (right) reproductions of scene-referred images (continued)
8 © ISO 2013 – All rights reserved

NP04 Butterfly
NP05 Threads
NP06 Fruits
NP07 Canal
Figure 2 — Samples of reduced size media-relative colorimetric conversion to sRGB (left) and
colour rendered sRGB (right) reproductions of scene-referred images (continued)
NP08 White Flowers
NP09 BarHarborPresunrise
NP10 BenJerrys
NP11 DelicateFlowers
NP12 DevilsBathtub
Figure 2 — Samples of reduced size media-relative colorimetric conversion to sRGB (left) and
colour rendered sRGB (right) reproductions of scene-referred images (continued)
10 © ISO 2013 – All rights reserved

NP13 Exploratorium
NP14 GoldenGate
NP15 HancockSeedField
NP16 NiagaraFalls
NP17 RedwoodSunset
Figure 2 — Samples of reduced size media-relative colorimetric conversion to sRGB (left) and
colour rendered sRGB (right) reproductions of scene-referred images (continued)
NP18 Route66Museum
NP19 SouthBranchKingsRiver
NP20 TupperLake
NP21 Chandelier
Figure 2 — Samples of reduced size media-relative colorimetric conversion to sRGB (left) and
colour rendered sRGB (right) reproductions of scene-referred images (continued)
12 © ISO 2013 – All rights reserved

NP22 Clock
NP23 Meat
NP24 Peacock
NP25 BlueFace
Figure 2 — Samples of reduced size media-relative colorimetric conversion to sRGB (left) and
colour rendered sRGB (right) reproductions of scene-referred images (continued)
NP26 PoundCake
NP27 LasVegas
NS01 Maple
NS02 River
Figure 2 — Samples of reduced size media-relative colorimetric conversion to sRGB (left) and
colour rendered sRGB (right) reproductions of scene-referred images (continued)
14 © ISO 2013 – All rights reserved

NS03 Marker
NS04 Flowers
NS05 Fluorescence
NS06 Hotel
Figure 2 — Samples of reduced size media-relative colorimetric conversion to sRGB (left) and
colour rendered sRGB (right) reproductions of scene-referred images (continued)
NS07 Pyramid
NS08 Nile
NS09 Felucca
NS10 CherryBlossom
Figure 2 — Samples of reduced size media-relative colorimetric conversion to sRGB (left) and
colour rendered sRGB (right) reproductions of scene-referred images (continued)
16 © ISO 2013 – All rights reserved

NS11 Iris
NS12 Wharf
NS13 Hiking
NS14 CadesCove
Figure 2 — Samples of reduced size media-relative colorimetric conversion to sRGB (left) and
colour rendered sRGB (right) reproductions of scene-referred images (continued)
NS15 FourCornersStorm
NS16 Sunset
NS17 Headlight
Figure 2 — Samples of reduced size media-relative colorimetric conversion to sRGB (left) and
colour rendered sRGB (right) reproductions of scene-referred images
4.5.2 Rendering
In Figure 2, the images on the left side are scene-referred colorimetry scaled to the sRGB dynamic range
and encoded as sRGB. They were produced by applying the following steps.
a) Convert from (nonlinear) RIMM16 RGB to linear RIMM RGB by inverting the RIMM16 RGB CCTF
specified in ISO/TS 22028-3. A RIMM16 RGB code value of 65535 will map to a linear RIMM RGB
value of 2.
b) Divide the linear RIMM RGB values by 2 to normalize them to the 0-1 sRGB encoding range.
c) Convert from the RIMM RGB white point and primaries to the sRGB white point and primaries as
shown in Formula (3).
R 2,,03426 −−0 72738 0,30688 R
    
sRGBlin RRIMMlin
    
G = −−−0,,22873 1 23161 0,00288 G (3)
sRGBlin RIMMlin
    
B  −−0,,00850 0 15331 1,16181 B 
 sRGBlin   RIMMlin
NOTE 1 The matrix in Formula (3) was obtained by multiplying the linear Bradford adapted D50 XYZ to
sRGB matrix with the inverse RIMM conversion matrix from ISO/TS 22028-3.
18 © ISO 2013 – All rights reserved

d) Clip the normalized linear sRGB values to the 0-1 sRGB encoding range in each channel to limit the
colour gamut to that of the sRGB reference display.
e) Apply the 8-bit sRGB CCTF as specified in IEC 61966-2-1.
NOTE 2 The scene-referred conversion can be accomplished using the ISO/TS 22028-3_RIMM-RGB-exCR.icc
profile as the source profile, the sRGB_v4_ICC_preference.icc profile as the destination profile, and the media-
relative colorimetric rendering intent with black point compensation on.
NOTE 3 The output-referred conversion can be accomplished by using the ISO/TS 22028-3_RIMM-RGB-
...

INTERNATIONAL ISO
STANDARD 12640-5
First edition
2013-12-15
Graphic technology — Prepress digital
data exchange —
Part 5:
Scene-referred standard colour image
data (RIMM/SCID)
Technologie graphique —Échange de données numériques de
préimpression —
Partie 5: Données d'image standard en couleurs montrées en référence
par scène (RIMM/SCID)
Reference number
©
ISO 2013
This DVD contains:
1) the publication ISO 12640-5:2013 in portable document format (PDF), which can be viewed using
Adobe® Acrobat® Reader;
2) image files NP01RGB.tif to NP27RGB.tif and NS01RGB.tif to NS17RGB.tif, which correspond to the
natural images described in Tables 1 and 2 and depicted in Figure 2;
3) image files S1RGB.tif to S3RGB.tif, which correspond to the synthetic images described in Table 3
and depicted in Figure 3.
Adobe and Acrobat are trademarks of Adobe Systems Incorporated.
©  ISO 2013
All rights reserved. Unless required for installation or otherwise specified, no part of this DVD may be reproduced, stored in a retrieval
system or transmitted in any form or by any means without prior permission from ISO. Requests for permission to reproduce this product
should be addressed to
ISO copyright office  Case postale 56  CH-1211 Geneva 20  Switzerland
Internet [e-mail]
Reproduction may be subject to royalty payments or a licensing agreement.
Violators may be prosecuted.
Published in Switzerland
ii © ISO 2013 – All rights reserved

Installation
If this
...