SIST ISO 12642-1:2008

INTERNATIONAL
STANDARD
First edition
1996-l 2-l 5
Graphic technology - Prepress digital data
Input data for characterization
exchange -
of 4-colour process printing
Technologie graphique - tkhange de don&es num&iques de
- Donnkes d’en t&e pour caractbisa tion d’impression en
prkimpression
quadrichromie
Reference number
IS0 12642:1996(E)
IS0 12642:1996(E)
Contents
Page
. . .
III
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Foreword
iv
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . 1
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2 Normative references
3 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .*.
................................................................................
4 Requirements
.................................. 2
4.1 Data set definition .
4.1.1 Basic ink value data set . .
.................................... 2
4.1.2 Extended ink value data set
..... ......................................... 7
4.1.3 User-defined data set
Colour measurement . .
4.2
.............................. 7
4.3 Data reporting .
..................................................................... 8
4.4 Data file format
Annexes
.................................. 11
A Application notes .
. . . . . . . . . . . . . . . . . . . . . . . . . . 15
B General description of keyword value file format
0 IS0 1996
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced
or utilized in any form or by any means, electronic or mechanical, including photocopying and
microfilm, without permission in writing from the publisher.
International Organization for Standardization
Case postale 56 l CH-1211 Geneve 20 l Switzerland
Printed in Switzerland
ii
IS0 12642:1996(E)
@ IS0
Foreword
IS0 (the International Organization for Standardization) is a worldwide
federation of national standards bodies (IS0 member bodies). The work of
preparing International Standards is normally carried out through IS0
technical committees. Each member body interested in a subject for which
a technical committee has been established has the right to be rep-
resented on that committee. International organizations, governmental and
non-governmental, in liaison with ISO, also take part in the work. IS0
collaborates closely with the International Electrotechnical Commission
(IEC) on all matters of electrotechnical standardization.
Draft International Standards adopted by the technical committees are
circulated to the member bodies for voting. Publication as an International
Standard requires approval by at least 75 % of the member bodies casting
a vote.
International Standard IS0 12642 was prepared by Technical Committee
lSO/TC 130, Graphic technology.
Annexes A and B of this International Standard are for information only.

@ IS0
IS0 12642:1996(E)
Introduction
General background
The technical content of this International Standard is identical to the
American National Standard lT8.7/3-1993. The ANSI document resulted
from the joint efforts of an international industry group that included
participants representing a broad range of prepress vendors, film
manufacturers, and users. This group, initially identified as the DDES
(Digital Data Exchange Standards) Committee, later became the founders
of the ANSI IT8 (Image Technology) accredited standards committee which
is responsible for electronic data exchange standards in graphic arts
prepress.
In an environment in which colour information is passed between
electronic publishing systems, it is essential for colour to be defined in an
unambiguous manner. Substantial experimental evidence enables us to
conclude that, for fovea1 vision, this can be achieved by specifying the
mixture of three linearly independent stimuli required to match that colour.
In 1931 a complete system of colour definition was developed by the CIE
(Commission Internationale de I’Eclairage) based on experimental evidence
published in the previous decade. This evidence confirmed the similarity
between observers in making such a match. That system and its
derivatives are now universally accepted for colour specification.
Many half-tone colour printing processes, however, require more than
three colourants. There are two reasons for this. Generally the gamut of
colours achievable with three printing inks is rather limited, and printing
additional inks can extend the gamut significantly. Furthermore, the
provision of extra inks can reduce the magnitude of the visual change
caused by the variability in colour and register which arises in print
production. By far the most common additional ink used is black, and four-
colour process printing is accepted as the norm for most forms of printing.
The addition of an extra ink means that the production of a colour cannot,
in general, be defined uniquely. As a result, different parts of a printed
sheet may use varying ink combinations to achieve the same colour. For
many practical purposes it is desirable to specify this combination directly,
rather than encode it by rules, and this leads to the requirement to transfer
data in a four-colour, device-specific mode. If the same data is to be used
for other applications, or even if it needs to be modified for a different set
of printing characteristics, some additional information is necessary to
enable the receiver of the data to interpret it. This International Standard
has been developed to achieve this objective. It provides a data set which
can be transmitted with an image to enable the receiver, if required, either
to transform the data into a device-independent state or correct it for a
different printing characteristic. An alternative application of the tools pro-
iv
IS0 12642:1996(E)
@ IS0
vided by this International Standard is to enable the characterization of
output systems and in this context work has been undertaken by the
committee to generate data for the major types of half-tone printing
processes which have been specified internationally. This procedure is
described in the application notes (annex A) and the data will be published
in future annexes.
The body of this International Standard defines the ink values to be used
for characterizing any four-colour (cyan, magenta, yellow, and black) half-
tone printing process (including gravure). These ink values are defined as
either digital data in a computer or half-tone tone values on film. This
requires that particular care be taken in the preparation of film to ensure
that the output device is properly “linearized” and the half-tone film values
match the numerical data in the computer file. For some applications the
film values used for linearization may be one or more generations removed
from the film produced by the film writer. The measurement procedures
and the data format to be used in determining and reporting tristimulus
values (X,Y,Z) are also included.
While the technique employed in this International Standard applies to all
output processes, the data has been optimized for four-colour half-tone
printing. For non-half-tone processes, or those which use colourants that
are significantly different from typical printing inks, the reference data file
should be determined in such a way that it provides reasonably uniform
colour differences when the data file is rendered. For a system which does
not meet the criterion, the user-optional data set could be utilized.
Suggestions for this are made in the application notes; however, they are
not part of this International Standard.
It should be noted that this International Standard does not define the
physical layout of the patches or their size. This is because any such
decision depends on the printing device to be used, and the area required
*for colour measurement. It is anticipated that a specific layout will be
produced to suit the needs of the user. However, in order to realize the
colours necessary for the measurements of specific printing processes to
be included as future annexes, it was necessary to produce a specific
layout. This layout, composed of four groups of patches, has been adopted
by both ANSI/CGATS and lSO/TC 130. Within TC 130 the digital data in the
appropriate format is contained in images S7 through SIO of the Standard
Colour Image Data (SCID), IS0 12640. For the guidance of others, this
layout is shown in figure A.I.
Technical background
Printing characteristics
Various efforts have been made over the past 20 years to reduce the
variation which occurs between printing presses. Initially, standards such
as IS0 2846 were developed to specify the colour of printing inks.
Subsequently, as a result of the lead of FOGRA/BVD in Germany,
significant effort has been made in developing specifications which define
constraints for the ink transfer onto paper. This is achieved by specifying
either the reflection density or the tristimulus values of a uniform (solid)
printed ink film, and by specifying tolerances around an optical density at
which various half-tone dot values should be reproduced. Within the
international printing community such specifications are widely recognized
and have become, in many cases, de facto printing standards. For

@ IS0
IS0 12642:1996(E)
magazine and periodical printing, SWOP (in the USA) and FIPP (in Europe)
are widely recognized standards. For commercial printing, the specifica-
tions of FOGRA and PIRA are widely known in Europe. Specifications are
also evolving for newspaper and heat-set web production. Future annexes
to this International Standard may contain the calorimetric tristimulus values
corresponding to these percent dot values when printed in accordance
with a number of such printing specifications. Such data can be used as
the basis for the conversion between ink values and tristimulus values.
It should be noted that any characterization of the process takes account
of all steps involved in print production. Thus it includes production of the
separations, any contacting operations which may be required and
platemaking. All of the printing specifications as referred to above include
recommendations for maintaining consistency of such operations to
ensure that validity of a characterization is maintained.
For characterizing printing conditions which differ from the published
specifications, two options exist. Either the large palette of colours can be
printed and measured, or the process can be modelled analytically. The
analytical modelling approach has the advantage that it requires far fewer
colour measurements; the disadvantages lie in the accuracy of prediction.
For many applications, a satisfactory compromise is achieved by using
modelling for the modification of published data. This is discussed in more
detail in the application notes.
Choice of colour palette
It is generally agreed that measurement of a reasonably large number of
colours is preferred for accurate characterization of any printing process. It
is not possible to be precise about how many colours are required; the
number will depend on many factors including the accuracy of colour
rendition required, the uniformity of spacing of the samples in terms of
colour, the type of modelling process used, and any nonlinear
However, practical
characteristics of a specific printing process.
experience suggests that measuring all combinations of six levels each for
cyan, magenta, yellow, and black, preferably weighted towards lower half-
tone dot values, will frequently prove adequate. Generally, for higher levels
of black, the number of samples may be considerably reduced since the
colour difference between samples is very small. With the addition of
single colour scales which contain extra values to assist in defining local
nonlinearity,
the accuracy obtained for most printing processes is
adequate.
A reduced-size data set may be used if:
- a less accurate characterization is adequate;
- the process ca n be modelled accurately by one of the well-kn own
mod els listed in the application notes;
- the aim of the measurement is to seek small corrections to an already
accurate characterization.
The advantages of this approach are that the measurement effort is
substantially lower and that the file size of the data is greatly reduced. This
can be advantageous when images are compressed although, in general,
even the larger file is small compared to most images
The proposal accepted for this International Standard defines a colour
palette consisting of 928 combinations of cyan, magenta, yellow, and black
vi
IS0 12642:1996(E)
@ IS0
ink values. It is this palette (hereafter called the extended ink value data
set) which has been measured to provide colour characterization data on
the major printing specifications.
Where such an extensive set of data is not required, a subset of this
palette which consists of 182 colours (hereafter called the basic ink value
data set) is specified. It provides data suited to a variety of modelling
methods and generally provides excessive data for any specific method. It
is sufficient for almost all published modelling methods.
For a characterization which cannot be achieved with the data sets defined
in this International Standard, provision is made for a user-optional set of
any size. The format of the data is defined in this International Standard.
It is anticipated that the basic data set will be the default file supplied in
the header of image files to be exchanged, and that by prior agreement,
one of the larger palettes may be provided when required. It is the intent
of ANSI lT8/CGATS and of lSO/TC 130 to
...