ISO/TR 14672:2000

TECHNICAL ISO/TR
REPORT 14672
First edition
2000-07-15
Graphic technology — Statistics of the
natural SCID images defined in ISO 12640
Technologie graphique — Statistique des données d'images en couleur
normales (SCID) définies dans l'ISO 12640
Reference number
©
ISO 2000
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but shall not
be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In downloading this
file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat accepts no liability in this
area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation parameters
were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In the unlikely event
that a problem relating to it is found, please inform the Central Secretariat at the address given below.
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means, electronic
or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or ISO's member body
in the country of the requester.
ISO copyright office
Case postale 56 � CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.ch
Web www.iso.ch
Printed in Switzerland
ii © ISO 2000 – All rights reserved

©ISO ISO/TR 14672:2000(E)
Contents Page
Foreword . . . iv
Introduction . v
1 Scope .1
2 Reference . 1
3 SCID image description .1
4 Colour distribution .3
4.1 Data value histograms and related things . 3
4.2 Average colour values .5
4.3 4 x 4 covariance matrix .5
4.4 Three-dimensional volumes .6
5 Spatial frequency characteristics .7
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.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 3.
The main task of technical committees is to prepare International Standards. Draft International Standards adopted by the
technical committees are circulated to the member bodies for voting. Publication as an International Standard requires approval
by at least 75 % of the member bodies casting a vote.
In exceptional circumstances, when a technical committee has collected data of a different kind from that which is normally
published as an International Standard (“state of the art”, for example), it may decide by a simple majority vote of its
participating members to publish a Technical Report. A Technical Report is entirely informative in nature and does not have
to be reviewed until the data it provides are considered to be no longer valid or useful.
Attention is drawn to the possibility that some of the elements of this Technical Report may be the subject of patent rights.
ISO shall not be held responsible for identifying any or all such patent rights.
ISO/TR 14672 was prepared by Technical Committee ISO TC 130, Graphic technology, Working Group 2, Prepress data
exchange.
iv
©ISO ISO/TR 14672:2000(E)
Introduction
International Standard 12640, Graphic technology — Prepress digital data exchange — CMYK standard colour image data
(CMYK/SCID), provides the digital data for a set of natural and synthetic colour images. The natural images
are intended for use in subjectively evaluating image quality as a function of image processing and/or output process. In
addition, the synthetic images are provided to allow objective measurement of process control, tone reproduction, colour
characterization, etc.
In addition, these images may be used for the statistical evaluation of the change in image content as a function of image
processing or other imaging steps. TC 130 Working Group 2 agreed to provide a set of reference statistics for these images as
a baseline for the technical community wishing to use the images for such statistical evaluation.
This statistical data was prepared by the TC 130 Japanese National Committee, and their efforts have made this technical report
possible.
v
�
TECHNICAL REPORT ISO ISO/TR 14672:2000(E)
Graphic technology —
Statistics of the natural SCID images defined in ISO 12640
1 Scope
This Technical Report provides the colour and spatial frequency distribution statistics associated with the digital image data of
International Standard 12640, Graphic technology — Prepress digital data exchange — CMYK standard colour image data
(CMYK/SCID).
2 Reference
ISO 12640:1997, Graphic technology — Prepress digital data exchange — CMYK standard colour image data (CMYK/SCID).
3 SCID image description
ISO 12640 defines a set of natural and synthetic colour images, called SCID (Standard Colour Image Data), which can be used
for evaluation of image processing algorithms or output devices, and also for colour characterization of output devices.
These images are digital files, encoded in a CMYK format. They exist in two forms, known as the primary and alternate data
sets, each of which has different resolutions and data ranges. The primary data set was created using the data encoding scheme
typical of a Colour Electronic Prepress System (CEPS). The alternate data set was created from the primary data set by re-
encoding the data in the scheme more typically used by the desktop publishing (DTP) prepress systems and defined in ISO 12639
as the P1 profile. With the primary set, tone value of 0% is encoded as data value of 28 and tone value of 100% is encoded as
data value of 228; the corresponding data values for the alternate set are 0 to 255. The primary set has an assumed resolution
of 16 pixels/mm while the alternate set has an assumed resolution of 12 pixels/mm. The alternate set was computed from the
primary set by cubic interpolation combined with a linear mapping of the data values.
The eight natural colour images of the primary set are shown as N1 to N8 in figure 1, the natural colour images of the alternate
set are denoted as N1A to N8A. When output at the assumed resolutions of 16 pixels/mm (406.4 pixels/inch) and 12 pixels/mm
(304.8 pixels/inch) the resultant size is 160mm x 128mm. The primary set of natural images are 2560 x 2048 pixels and the
alternate set are 1920 x 1536 pixels. Table 1 shows the characteristics and typical usage of the images.
These eight images were selected so that users can subjectively evaluate several aspects of image quality that are often reduced
by image processing or as a result of output device characteristics. The image quality attributes in question are described in table
1, but may be summarized as follows:
– colour reproduction;
– sharpness;
– graininess.
Although there is no established procedure for quantitative evaluation of these subjective image quality attributes, by the use
of a common set of images SCID makes it possible to conduct such subjective evaluation on a consistent basis. However, when
evaluating the application of image processing algorithms to the images some quantitative analysis is feasible and in this context
various statistical characteristics calculated from the images may prove useful. Statistics on colour and spatial frequency
distribution are such basic characteristics.
N1 and N1A N2 and N2A
N3 and N3A N4 and N4A
N5 and N5A N6 and N6A
N7 and N7A N8 and N8A
Figure 1 — Reduced monochrome reproductions of the natural images
©ISO ISO/TR 14672:2000(E)
Table 1 — Natural images
Name Aspect Characteristics
N1, Portrait Portrait Used to evaluate the reproduction of human skin.
N2, Cafeteria Portrait Image with complicated geometric shapes. Suitable for evaluating the
result of image processing.
N3, Fruit Basket Landscape Image of a basket and cloth used to evaluate the reproduction of brown
colours and close texture.
N4, Wine and Landscape Image of glassware and silverware used to evaluate the reproduction
Tableware characteristics of highlight tones and neutral colours.
N5, Bicycle Portrait Image of a (penny-farthing) bicycle, resolution charts and other items
containing high detail used to evaluate the sharpness of reproduction and
the results of image processing.
N6, Orchid Landscape Image of an orchid with background vignettes used to evaluate
reproduction of highlight and shadow vignettes.
N7, Musicians Landscape Image of three girls with different skin characteristics and fine detail used
to evaluate the reproduction of different skin tones and image detail.
N8, Candle Landscape “Low-key” image of a room scene containing miscellaneous objects used
to evaluate dark colours, particularly browns and greens.
These characteristics can be used for the comparison between the data of an original image and that of a transformed or degraded
image passing through some image handling system such as an image transmission line, an imaging system, an image storage
system or an image transform and processing system. It is also possible to compute those basic quantities from a set of images
to certify validity of them if this is in doubt.
Members of the technical committee ISO TC 130 evaluated the aforesaid basic statistical quantities for the eight natural SCID
images (for both the primary and alternate set). This technical report summarizes the quantitative results.
4 Colour distribution
Statistics which represent image colour distribution are provided in this section. These were obtained by computing single
dimensional histograms, average colour values of each colour, the three-dimensional volume and covariance matrix and resultant
orthogonal matrix and principal axes. Such data were calculated for each image. The histogram data is provided graphically
(figures 5-12) and numerically (tables 5-12). The remaining statistics are provided in tables 13-20.
It should be noted that each natural image is provided with a text insert such as ISO300 or ISO400 in the image. Pixels
representing this text have a coded value of either 0 (white) or 255 (black). This text serves to distinguish between the primary
set and alternate set. It is not meaningful to include this text in the colour distribution calculation. Therefore, the calculation
was carried out only for the image area, excluding the text.
The position of the outer boundaries of the text is defined by a rectangle produced from the coordinates of two of the corners
as shown in figure 2. The position (in terms of number of pixels) of the text in each image is given in table 2 and table 3. This
was the area excluded prior to calculation of the statistical data but extended by 4 pixels in each direction to minimize any effects
arising from evaluation of the statistics after any image processing which requires calculation based on adjacent pixels. Thus
the coordinates of the corners defining the rectangle enclosing the excluded area are A(X -4,Y -4) and B(X +4,Y +4).
1 1 2 2
4.1 Data value histograms and related things
The histograms for the natural images are shown in figures 5-12. These show the frequency of occurrence of each of the dot
percent equivalent values within the image; therefore, there are four histograms for each image, one for each of the colours cyan,
magenta, yellow and black. Each of the figures has two parts; the first shows the histograms for the primary set of images and
the second part the histograms for the alternate set. The numeric data used to produce these figures are given in tables 5-12.
Figure 2 — Definition of the coordinates of the text elements
Table 2 — Position and area of the text for the primary
set of natural images (16 pixels/mm )
Image A(x ,y ) B(x ,y )
1 1 2 2
N1 (1769,39) (2008,88)
N2 (37,35) (276,84)
N3 (38,36) (277,85)
N4 (41,37) (280,86)
N5 (37,34) (276,83)
N6 (37,35) (276,84)
N7 (2286,35) (2525,84)
N8 (2070,193) (2309,242)
Table 3 — Position and area of the text for the
alternate set of natural images (12 pixels/mm )
Image A(x ,y ) B(x ,y )
1 1 2 2
N1A (1326,24) (1503,61)
N2A (26,23) (203,60)
N3A (25,22) (202,59)
N4A (31,28) (208,65)
N5A (25,23) (202,60)
N6A (25,23) (202,60)
N7A (1708,21) (1885,58)
N8A (1554,145) (1731,182)
It should be noted that although the primary set of images contains data in the ranges 0-27 and 229-255, these are not normally
differentiated by the output device since the values less than or equal to 28 are all set to 0% dot and those greater than or equal
to 228 are all set to 100% dot. In the case of the alternate set, however, all values in the range may be differentiated. This means
that when the alternate set was derived from the primary set all the values in the ranges 0-28 and 229-255 for the primary set
were mapped to 0 and 255 respectively. Thus the histograms for the alternate set tend to show very high frequencies for these
values when compared to the rest of the values within the image, or the values within the primary set.
This phenomenon has a marked effect on some of the subsequent statistics, depending on the distribution of the ‘out of range’
values in the primary images. In the case of the average value, for example, it might be expected that the same (or very similar)
©ISO ISO/TR 14672:2000(E)
dot percent value would be obtained for both sets with each of the images. However, this is rarely the case, as will be seen in
4.2, particularly for the black. A glance at the histogram data will quickly reveal why.
Statistics are provided in tables 13-20 for each of the natural images in the set. The data provided are the average colour values
of each ink, the three dimensional volume, covariance matrix, resultant orthogonal matrix and principal axes. These are
described in more detail in the following subclauses.
4.2 Average colour values
The average colour values may be calculated using equation (1).
c m y k
j i j i j i j i
¯
c¯' ,¯m' ,¯y' , k'
(1)
N N N N
where
c is the value for cyan;
m is the value for magenta;
y is the value for yellow;
k is the value for black;
i  is the index value of the pixel; and
N is the number of pixels in the image.
The average values indicate the amount of each colour present in an image. Therefore, any tendency for a particular colour to
be dominant in each image can be indicated, although not categorically stated, by the average colour values. For example, the
high blue content of image N6 and N6A (Orchid) is shown by the fact that the cyan and magenta colour values are higher than
those of the yellow. However, similar values would have been obtained from an image containing large areas of cyan and
magenta and no blue.
Since no under colour removal (UCR) or gray component replacement (GCR) has been applied to these images black colour is
found only in the shadow areas, and the average value is less than that of the coloured inks. However, the magnitude of the value
provides some indication of whether an image contains dark colours. For example, the image N5 (Bicycle) contains little black
because it consists primarily of clean, but fairly colourful, and light neutral colours. Image N8 (Candle), on the other hand,
contains a lot of black because it contains many dark colours, even though many of them are fairly colourful. The black colour
quantity is independent of any colour tendency shown by the average of colour values of C, M and Y.
4.3 4 x 4 covariance matrix
The covariance matrix may be calculated using equation (2).
¯
(c&c¯) (c&c¯)(m&m¯) (c&c¯)(y&y¯) (c&c¯)(k&k)
j j j j
i i i i i i i
N N N N
¯
(m&m¯)(c&c¯) (m&m¯) (m&m¯)(y&y¯) (m&m¯)(k&k)
j i i j i j i i j i i
N N N N
[M ]'
¯
(y&y¯)(c&c¯) (y&y¯)(m&m¯) (y&y¯) (y&y¯)(k&k) (2)
j j j j
i i i i i i i
N N N N
¯ ¯ ¯ ¯
(k&k)(c&c¯) (k&k)(m&m¯) (k&k)(y&y¯) (k&k)
j i i j i i j i i j i
N N N N
The diagonal elements of the covariance matrices in tables 13-20 show the variance of data values for each colour component.
The values are related to the range of the histogram in figures 5-12. The non-diagonal elements show the correlation between
two colour components. Such a statement may not seem obvious at first, but can best be understood by considering the case
where all four separations are identical (c=m=y=k), and hence highly correlated. After normalization by dividing each of the
elements in the matrix by the square root of the product of the diagonal elements they will then give a value of 1 for each of the
non-diagonal terms. If the colour components are not identical but correlated to some degree, the normalized non-diagonal
elements will be smaller than 1. Figure 3 gives an example of a two-dimensional correlation plot with the data values c and m .
i i
If the distribution has the shape indicated as in (a), there is a high degree of correlation. If the distribution looks as indicated
in (b), the degree of correlation is much lower.
Each of the non-diagonal elements in these covariance matrices only shows the correlation between colour components in the
image and does not show the degree of colourfulness of the image or the range of colours it contains. For example, a neutral
vignette made from the three chromatic inks, and ranging from white to black, would have a very high correlation but is not at
all colourful. A similar vignette made from two chromatic inks also has a high degree of correlation, and is very colourful, but
the range of colours is very limited. A useful measure of image content is the combination of the range and colourfulness of
the colours within it. Such a measure is how widely the colours are distributed within a colour space; in other words how large
the colour gamut of the image is. Figure 3 shows this relationship in a two-dimensional space.
4.4 Three-dimensional volumes
To express this measure, it is necessary to define a volume, corresponding to the area shown in figure 3. The volume in the
colour space can be determined from the standard deviations along the principal axes of the colour space. These are denoted
by F , F , F , F which are the square roots of the diagonal elements of the covariance matrix after multiplication with the
1 2 3 4
orthogonal matrix shown in tables 13 to 20. In the example, figure 3, the F can be visualized as the diameters of the ellipse, in
the directions > and 0.  In our case, with a CMYK colour space, there are four principal axes, one for each process colour. (It
is important to note that this definition means that the relationship between these principal axes and colour is somewhat loose
since CMYK is not a true colour space. Nevertheless, it can be shown to produce a reasonable approximation to a uniform colour
space and so the following measures do prove usable.) Based on the four standard deviations, the four-dimensional volume V ,
used to show the colour gamut of an image in the full ‘colour’ space produced by the four inks, is defined as follows:
V = F × F × F × F (3)
4 1 2 3 4
However, since the colour space that humans can sense is only three-dimensional, and the black only adds a limited amount to
the colour gamut, V is not always an appropriate measure of the colour gamut. In this report, the three-dimensional volume
V is proposed as a quantity to evaluate the colour gamut of an image.
V is obtained as a product of three standard deviations, excluding the minimum standard deviation:
V = F × F × F (4)
3 1 2 3
According to the tables, the image among the eight with the largest gamut is N2 and that with the smallest is N4. These results
are consistent with subjective assessment.
It should be noted that all the quantities above are defined in terms of c, m, y and k coordinates. It should not be expected that
a colour gamut measure derived from CMYK space corresponds exactly to the gamut as perceived by a human observer, as
would be expected from such measures derived from CIELAB or CIELUV space.
However, the SCID images are defined as CMYK images in digital form. Their reproduced colour depends on the specific inks
and processes used for printing, and these images are intended for use by all processes. Thus there is no single set of colorimetric
data which could be specified and so there was no alternative than to use CMYK colour space for this evaluation. Tables 13-20
show the statistics on colour distribution of each image.
(a) high correlation      (b) low correlation
Figure 3 — Distribution of colour values
©ISO ISO/TR 14672:2000(E)
5 Spatial frequency characteristics
To evaluate the spatial frequency characteristics of the test images, the autocorrelation function R(a,b) defined by equation (5)
was calculated for each image.
 
fi , i−+f fi a,i + b− f
() ()
∑∑
[]xy i,i x y i
xy x
 ++a,i b
y
 
R(a,b) =
22 2
(5)
 
fi ,i−+f fi a,i+b−f
() ()
∑∑{}xy i,i{}x y i++a,i b
xy x y
 
 
where
R(a,b) is the autocorrelation function;
a is the horizontal shift from point(x,y);
b is the vertical shift from point(x,y);
i is the horizontal image coordinate; and
x
i is the vertical image co-ordinate.
y
The way the image area used to calculate the autocorrelation function was defined is shown in figure 4. The specific coordinates for each image
are given in table 4. Figures 13-20 show the autocorrelation function R(a,0) (solid line) and R(0,b) (dotted line) of the c, m, y and k
components of each picture. Each value is normalized by R(0,0), the value of for shift R(0,0) .
Figure 4 — Definition of the area used to calculate the autocorrelation function
Table 4 — Coordinates of the processed area for each image
Image i i d
x y
N1 513 769 1024
N1A 385 577 768
N2 513 769 1024
N2A 385 577 768
N3 769 513 1024
N3A 577 385 768
N4 769 513 1024
N4A 577 385 768
N5 513 769 1024
N5A 385 577 768
N6 769 513 1024
N6A 577 385 768
N7 769 513 1024
N7A 577 385 768
N8 769 513 1024
N8A 577 385 768
P i x e l s P i x e l s
C y a n M a g e n t a
200000 200000
150000 150000
100000 100000
50000 50000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
200000 200000
150000 150000
100000 100000
50000 50000
0 0
050 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 5a — Histograms for image N1
P i x e l s P i x e l s
C y a n M a g e n t a
100000 100000
75000 75000
50000 50000
25000 25000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
100000 100000
75000 75000
50000 50000
25000 25000
0 0
050 100 150 200 250 0 50 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 5b — Histograms for image N1A
©ISO ISO/TR 14672:2000(E)
P i x e l s P i x e l s
C y a n M a g e n t a
200000 200000
150000 150000
100000 100000
50000 50000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
200000 200000
150000 150000
100000 100000
50000 50000
0 0
050 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 6a — Histograms for image N2
P i x e l s P i x e l s
C y a n M a g e n t a
100000 100000
75000 75000
50000 50000
25000 25000
0 0
0 50 100 150 200 250 0 50 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
100000 100000
75000 75000
50000 50000
25000 25000
0 0
0 50 100 150 200 250 0 50 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 6b — Histograms for image N2A
P i x e l s P i x e l s
C y a n M a g e n t a
200000 200000
150000 150000
100000 100000
50000 50000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
200000 200000
150000 150000
100000 100000
50000 50000
0 0
050 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 7a — Histograms for image N3
P i x e l s P i x e l s
C y a n M a g e n t a
100000 100000
75000 75000
50000 50000
25000 25000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
100000 100000
75000 75000
50000 50000
25000 25000
0 0
050 100 150 200 250 0 50 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 7b — Histograms for image N3A
©ISO ISO/TR 14672:2000(E)
P i x e l s P i x e l s
C y a n M a g e n t a
200000 200000
150000 150000
100000 100000
50000 50000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
200000 200000
150000 150000
100000 100000
50000 50000
0 0
050 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 8a — Histograms for image N4
P i x e l s P i x e l s
C y a n M a g e n t a
100000 100000
75000 75000
50000 50000
25000 25000
0 0
0 50 100 150 200 250 0 50 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
100000 100000
75000 75000
50000 50000
25000 25000
0 0
0 50 100 150 200 250 0 50 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 8b — Histograms for image N4A
P i x e l s P i x e l s
C y a n M a g e n t a
200000 200000
150000 150000
100000 100000
50000 50000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s
B l a c k
P i x e l s
Y e l l o w
0 50 100 150 200 250
050 100 150 200 250
D a t a  v a l u e
D a t a  v a l u e
Figure 9a — Histograms for image N5
P i x e l s P i x e l s
C y a n M a g e n t a
100000 100000
75000 75000
50000 50000
25000 25000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
100000 100000
75000 75000
50000 50000
25000 25000
0 0
050 100 150 200 250 0 50 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 9b — Histograms for image N5A
©ISO ISO/TR 14672:2000(E)
P i x e l s P i x e l s
C y a n M a g e n t a
200000 200000
150000 150000
100000 100000
50000 50000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
200000 200000
150000 150000
100000 100000
50000 50000
0 0
050 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 10a — Histograms for image N6
P i x e l s P i x e l s
C y a n M a g e n t a
100000 100000
75000 75000
50000 50000
25000 25000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
100000 100000
75000 75000
50000 50000
25000 25000
0 0
050 100 150 200 250 0 50 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 10b — Histograms for image N6A
P i x e l s P i x e l s
C y a n M a g e n t a
200000 200000
150000 150000
100000 100000
50000 50000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s
P i x e l s
Y e l l o w B l a c k
150000 150000
100000 100000
50000 50000
0 0
050 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 11a — Histograms for image N7
P i x e l s P i x e l s
C y a n M a g e n t a
100000 100000
75000 75000
50000 50000
25000 25000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
100000 100000
75000 75000
50000 50000
25000 25000
0 0
050 100 150 200 250 0 50 100 150 200 250
Data value D a t a  v a l u e
Figure 11b — Histograms for image N7A
©ISO ISO/TR 14672:2000(E)
P i x e l s P i x e l s
C y a n M a g e n t a
200000 200000
150000 150000
100000 100000
50000 50000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
200000 200000
150000 150000
100000 100000
50000 50000
0 0
050 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 12a — Histograms for image N8
P i x e l s P i x e l s
C y a n M a g e n t a
100000 100000
75000 75000
50000 50000
25000 25000
0 0
0 50 100 150 200 250 050 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
P i x e l s P i x e l s
Y e l l o w B l a c k
100000 100000
75000 75000
50000 50000
25000 25000
0 0
050 100 150 200 250 0 50 100 150 200 250
D a t a  v a l u e D a t a  v a l u e
Figure 12b — Histograms for image N8A
Table 5a — Histogram data for N1 image

Data Number of pixels having the data value Data Number of pixels having the data value
Value Value
C M Y K C M Y K
0 270 43 398 1 061 047 64 20 404 10 031 9 447 8 612
1 37 7 43 44 649 65 15 989 8 003 8 733 9 647
2 48 2 42 50 494 66 18 255 9 336 7 273 8 335
3 32 2 41 45 314 67 20 652 8 482 8 472 9 689
4 57 8 47 47 31 68 15 152 6 752 7 842 8 591
5 62 4 82 36 761 69 17 228 9 542 7 567 8 038
6 72 8 99 41 116 70 14 659 8 104 7 551 8 410
7 80 14 101 34 029 71 16 85 8 515 8 015 7 909
8 106 13 123 32 912 72 18 581 7 272 8 757 9 238
9 110 20 184 36 002 73 16 267 8 983 8 345 8 053
10 131 29 156 37 399 74 16 412 8 048 9 464 8 161
11 168 27 233 48 236 75 19 636 8 296 10 029 8 790
12 209 28 224 73 658 76 17 200 7 731 9 455 8 368
13 281 41 306 85 412 77 15 920 8 147 13 301 7 848
14 413 49 347 114 101 78 18 890 9 638 15 543 7 519
15 499 67 512 130 473 79 15 410 8 020 21 381 9 024
16 950 62 476 111 349 80 18 006 8 957 25 941 7 230
17 1 230 161 747 150 209 81 16 422 8 933 35 696 9 084
18 1 532 197 1 078 115 731 82 14 985 9 522 53 233 8 642
19 2 754 310 1 268 108 861 83 15 200 9 766 60 900 7 221
20 3 149 525 1 569 114 065 84 16 705 11 564 62 496 7 589
21 4 957 863 2 274 107 665 85 14 717 9 566 84 686 7 910
22 4 718 1 150 2 286 106 810 86 15 687 11 170 77 933 8 675
23 9 451 1 569 2 760 105 129 87 14 566 11 430 86 948 7 676
24 9 039 2 466 3 595 98 092 88 14 276 12 124 78 694 9 028
25 11 682 2 807 3 890 106 001 89 14 544 14 065 85 409 6 951
26 16 958 4 108 5 051 65 356 90 14 824 14 202 97 997 8 752
27 17 757 5 144 5 102 89 409 91 12 964 16 184 86 962 8 799
28 15 318 6 029 5 397 70 341 92 14 992 20 327 92 062 7 330
29 18 514 5 278 7 054 65 907 93 14 446 31 448 81 827 7 880
30 21 473 6 108 7 795 52 126 94 13 807 41 794 86 842 8 212
31 17 522 6 463 7 965 43 178 95 11 249 67 960 93 417 8 798
32 17 430 5 298 8 938 30 065 96 15 305 85 176 76 728 8 325
33 19 202 6 681 8 742 35 867 97 13 723 111 841 94 724 7 346
34 22 020 6 513 7 194 27 089 98 12 293 121 375 84 275 9 004
35 19 443 5 881 10 094 25 204 99 16 910 114 070 87 522 7 465
36 25 843 7 779 10 166 22 981 100 13 564 131 825 84 504 10 042
37 26 892 6 921 9 163 22 971 101 14 988 123 686 80 810 7 233
38 26 326 8 240 9 046 20 492 102 15 487 134 460 68 328 8 137
39 31 480 8 053 10 480 15 582 103 12 927 127 381 63 980 8 351
40 27 826 7 549 8 174 18 649 104 13 364 130 361 70 020 8 935
41 25 843 8 695 10 208 13 615 105 16 183 107 165 53 519 8 865
42 26 917 7 417 9 500 16 604 106 14 863 84 470 55 285 9 003
43 28 340 10 181 9 200 12 979 107 14 414 100 273 49 660 8 351
44 30 125 9 178 8 781 14 750 108 13 859 91 913 38 171 8 960
45 22 343 9 999 10 132 13 861 109 18 348 75 726 38 905 9 731
46 25 433 7 977 8 207 11 719 110 15 045 80 094 35 855 8 947
47 23 185 10 069 8 926 13 255 111 19 737 69 994 37 759 8 770
48 22 236 9 688 10 413 12 298 112 24 529 62 268 35 567 7 802
49 25 901 9 318 8 259 12 567 113 33 496 54 049 29 450 9 135
50 20 463 10 109 8 477 12 363 114 35 358 55 283 30 335 9 465
51 19 169 8 618 9 700 11 437 115 55 165 38 225 28 941 9 351
52 23 601 9 582 7 849 10 057 116 73 897 45 318 28 672 8 661
53 20 300 8 777 9 127 12 927 117 81 134 34 866 26 040 8 885
54 20 233 8 940 9 457 9 166 118 87 199 36 499 22 326 8 777
55 18 560 8 510 7 671 10 963 119 97 666 31 155 27 780 9 435
56 21 644 8 387 9 298 10 806 120 106 314 32 260 23 699 9 828
57 18 384 7 757 7 976 10 149 121 105 432 32 401 24 853 9 877
58 18 269 9 229 9 231 9 780 122 95 274 31 144 23 243 9 125
59 19 615 7 212 7 763 10 076 123 113 588 31 846 20 132 10 169
60 18 401 8 933 8 669 9 351 124 117 750 30 365 23 400 8 261
61 18 550 7 507 7 140 9 050 125 104 160 26 052 19 670 9 743
62 16 833 8 885 8 071 9 096 126 100 827 26 448 23 341 8 902
63 18 352 7 587 7 603 9 238 127 87 430 26 451 21 385 9 393
©ISO ISO/TR 14672:2000(E)
Table 5a — Histogram data for image N1 (continued)
Data Number of pixels having the data value Data Number of pixels having the data value
Value Value
C M Y K C M Y K
128 79 266 25 177 22 536 9 642 192 18 351 27 478 60 286 13 365
129 71 068 24 263 17 694 9 708 193 21 482 34 338 47 012 14 598
130 57 214 25 153 21 920 9 610 194 19 427 41 079 53 069 13 527
131 56 356 21 205 25 557 9 515 195 22 227 53 170 63 989 16 651
132 53 796 24 785 18 721 9 914 196 18 325 65 677 55 196 16 088
133 47 619 22 815 19 796 9 367 197 22 764 76 239 53 538 16 326
134 48 550 23 477 20 938 8 634 198 21 265 96 398 68 793 16 555
135 43 203 20 471 21 659 10 710 199 25 008 90 441 47 134 13 037
136 32 062 22 443 20 272 7 830 200 21 694 71 261 45 047 15 087
137 38 041 20 445 20 650 10 287 201 24 396 58 593 35 063 11 945
138 27 849 19 563 18 442 8 192 202 24 299 55 091 25 441 9 931
139 26 808 19 636 22 517 9 394 203 23 299 39 103 19 840 7 369
140 25 193 17 650 17 585 9 555 204 27 061 41 342 13 608 5 311
141 22 340 18 126 22 010 8 900 205 25 974 39 199 11 502 3 673
142 18 879 20 271 19 937 9 946 206 30 863 30 625 8 257 2 418
143 16 714 19 846 22 330 8 095 207 24 276 26 947 8 096 1 213
144 15 007 17 036 17 852 9 330 208 32 157 21 481 6 186 855
145 15 572 15 694 22 101 9 963 209 31 163 20 887 5 617 587
146 12 879 17 723 19 045 8 429 210 40 998 19 744 5 029 296
147 14 161 18 768 22 518 9 474 211 41 950 13 653 4 412 337
148 12 490 16 836 18 085 8 758 212 43 810 13 273 3 881 162
149 12 695 17 126 19 310 8 379 213 67 084 12 619 2 848 153
150 11 736 15 666 19 599 9 692 214 61 905 11 715 3 421 170
151 12 866 15 109 19 572 9 902 215 64 942 12 218 2 542 128
152 11 553 18 010 19 385 7 854 216 46 056 8 691 2 356 129
153 11 973 14 356 19 100 8 842 217 31 275 10 928 2 054 122
154 14 416 16 779 17 250 9 278 218 16 410 9 308 2 015 72
155 10 813 13 411 19 264 8 116 219 12 552 9 497 1 713 110
156 12 532 17 097 15 376 9 649 220 9 643 11 621 1 547 71
157 10 988 12 952 19 153 8 963 221 8 525 11 315 1 442 81
158 10 900 15 585 17 245 8 781 222 7 543 13 626 1 322 60
159 13 259 14 326 17 895 7 656 223 7 198 15 077 1 282 66
160 10 373 13 680 13 437 8 439 224 5 110 13 855 1 167 56
161 12 836 13 170 19 017 8 713 225 5 919 12 464 1 176 44
162 10 618 13 977 18 362 9 182 226 4 363 11 256 800 51
163 10 318 11 886 14 738 8 087 227 4 847 10 105 853 55
164 12 323 12 430 17 513 8 565 228 3 797 9 956 898 60
165 12 390 12 536 18 520 7 907 229 3 524 8 431 781 33
166 11 628 13 901 16 389 9 255 230 2 726 7 157 694 44
167 12 150 10 377 16 622 8 553 231 2 991 6 479 665 39
168 11 986 12 785 17 695 8 207 232 2 846 5 091 593 50
169 11 252 11 493 18 736 8 260 233 2 416 4 439 496 22
170 12 761 12 652 19 484 8 439 234 2 053 3 537 537 32
171 11 667 9 789 15 850 8 213 235 2 030 2 721 386 41
172 13 230 12 953 21 939 8 135 236 1 705 2 258 486 34
173 11 941 10 555 20 519 9 354 237 1 736 1 830 391 17
174 11 461 11 374 18 169 8 183 238 1 494 2 240 349 26
175 13 859 11 687 20 494 7 618 239 1 446 1 074 290 31
176 11 810 12 196 19 497 10 301 240 1 225 1 395 305 23
177 12 715 10 750 21 662 7 279 241 1 284 1 199 315 25
178 15 079 11 281 24 041 9 228 242 1 068 948 223 22
179 13 594 11 530 23 438 8 223 243 966 860 212 23
180 13 989 12 479 26 190 9 420 244 938 762 203 23
181 13 149 11 335 23 046 7 884 245 981 692 202 20
182 15 828 12 429 28 808 9 030 246 771 577 193 17
183 15 080 11 909 27 722 9 821 247 765 575 178 16
184 17 672 12 421 32 263 9 081 248 652 505 171 16
185 15 047 14 554 31 075 9 539 249 748 375 136 14
186 14 863 15 834 36 612 9 556 250 508 364 132 10
187 20 605 13 210 37 238 9 697 251 569 337 106 6
188 16 726 17 054 42 190 11 644 252 490 334 117 10
189 17 346 19 082 41 579 10 660 253 539 270 86 5
190 17 695 17 996 45 089 10 825 254 416 246 80 7
191 22 058 22 613 41 760 13 002 255 6 942 2 388 732 34
Table 5b — Histogram data for image N1A
Data Number of pixels having the data value Data Number of pixels having the data value
Value Value
C M Y K C M Y K
0 57 860 15 041 23 982 1 836 253 64 7 731 4 024 7 659 3 668
1 8 484 2 647 3 260 29 718 65 7 733 3 985 9 731 3 860
2 7 622 2 333 2 863 22 372 66 7 721 4 049 11 960 3 676
3 9 160 2 914 3 658 22 764 67 6 373 3 567 12 976 3 230
4 8 179 2 793 3 798 18 926 68 7 234 4 224 18 306 3 885
5 8 172 2 700 3 789 16 087 69 7 011 4 369 22 871 3 895
6 8 554 3 009 3 972 15 041 70 6 855 4 356 26 664 3 719
7 7 612 2 589 3 528 11 551 71 7 344 4 713 30 117 3 726
8 9 601 3 001 4 165 12 311 72 5 903 4 019 28 582 3 201
9 9 715 3 206 4 265 11 160 73 6 781 4 743 35 565 3 659
10 10 956 3 264 4 285 10 475 74 6 681 5 068 36 571 3 730
11 12 097 3 365 4 180 10 023 75 6 623 5 123 38 100 3 606
12 10 870 2 958 3 606 7 961 76 5 904 4 660 33 712 3 183
13 12 616 3 444 4 197 8 671 77 6 449 5 715 39 015 3 867
14 13 746 3 741 4 333 8 159 78 6 403 6 074 40 080 3 573
15 13 115 3 684 4 150 7 744 79 6 503 6 559 41 527 3 718
16 10 901 3 278 3 614 6 040 80 6 244 7 340 40 389 3 741
17 12 260 3 994 4 250 6 704 81 5 493 7 360 34 445 3 128
18 12 503 3 944 4 256 6 825 82 6 434 10 618 40 422 3 742
19 12 655 4 074 4 077 6 281 83 6 366 14 369 38 825 3 670
20 13 007 4 206 4 208 6 271 84 6 193 19 117 39 519 3 716
21 10 162 3 564 3 491 5 260 85 5 926 26 822 39 519 3 790
22 11 139 4 180 4 031 6 049 86 5 335 29 279 33 688 3 207
23 11 017 4 197 3 997 5 854 87 6 319 40 967 39 040 3 827
24 10 645 4 195 4 099 5 752 88 6 219 49 959 40 701 3 617
25 9 350 3 626 3 524 5 171 89 6 077 52 544 39 591 3 824
26 10 733 4 268 4 149 5 712 90 5 530 47 022 33 817 3 203
27 10 302 4 226 3 981 5 480 91 6 566 55 170 38 701 3 762
28 9 839 4 204 3 869 5 437 92 6 561 57 298 37 768 4 015
29 9 443 4 181 4 084 5 070 93 6 553 59 313 36 498 3 952
30 8 396 3 664 3 332 4 428 94 6 617 59 871 33 239 3 880
31 9 886 4 023 3 913 4 998 95 5 515 51 518 27 332 3 372
32 9 300 3 942 4 005 5 099 96 6 377 58 251 29 824 3 793
33 9 032 3 980 3 966 4 648 97 6 338 55 748 28 953 3 834
34 8 986 3 966 3 738 4 805 98 6 776 49 932 26 122 4 033
35 7 704 3 368 3 334 4 084 99 6 888 45 994 25 067 4 060
36 9 027 3 729 4 075 4 650 100 5 723 36 861 19 453 3 354
37 8 866 3 815 3 775 4 691 101 6 741 41 633 21 028 4 052
38 8 379 3 821 3 925 4 460 102 6 667 39 707 18 966 4 010
39 7 357 3 206 3 140 3 834 103 7 408 37 412 17 546 3 966
40 8 636 3 725 3 703 4 338 104 6 379 30 735 14 573 3 479
41 8 417 3 815 3 756 4 243 105 7 769 34 824 16 737 4 179
42 8 478 3 886 3 699 4 387 106 9 073 31 182 15 985 4 049
43 7 994 3 880 3 623 4 173 107 10 647 29 036 15 241 3 969
44 7 045 3 265 3 178 3 509 108 13 382 26 387 14 256 4 049
45 8 222 3 737 3 624 4 106 109 13 747 20 818 12 083 3 462
46 8 595 3 961 3 822 4 000 110 19 778 22 892 13 579 4 184
47 8 101 3 797 3 706 4 198 111 25 501 19 584 12 823 4 101
48 8 370 3 982 3 696 4 051 112 30 883 18 783 12 465 4 132
49 7 107 3 248 3 152 3 457 113 34 952 17 279 12 054 4 109
50 8 209 3 743 3 754 3 998 114 33 662 14 157 10 075 3 510
51 7 760 3 765 3 724 3 959 115 41 656 15 810 11 614 4 090
52 7 636 3 930 3 721 3 874 116 45 040 15 240 11 663 4 266
53 6 508 3 287 3 271 3 397 117 46 149 14 933 10 966 4 167
54 7 745 3 830 3 827 3 906 118 40 031 12 278 9 193 3 685
55 7 668 3 736 3 776 3 805 119 46 685 14 232 10 714 4 216
56 7 853 3 759 3 896 3 872 120 47 579 14 054 10 139 4 295
57 7 825 3 879 3 836 3 692 121 49 352 13 835 9 920 4 251
58 6 886 3 240 3 512 3 158 122 52 481 13 600 10 151 4 374
59 8 011 3 782 4 282 3 709 123 43 496 11 091 8 543 3 550
60 8 035 3 822 4 458 3 827 124 48 217 12 432 9 650 4 295
61 7 899 3 749 4 791 3 669 125 45 374 12 085 9 781 4 163
62 7 853 3 978 5 476 3 732 126 40 922 11 877 9 720 4 226
63 6 694 3 376 5 422 3 199 127 33 085 10 203 8 570 3 823
©ISO ISO/TR 14672:2000(E)
Table 5b — Histogram data for image N1A (continued)
Data Number of pixels having the data value Data Number of pixels having the data value
Value Value
C M Y K C M Y K
128 33 642 11 332 9 763 4 268 192 5 387 4 594 9 103 3 362
129 30 621 10 943 9 426 4 390 193 6 261 5 277 10 976 3 981
130 27 231 10 880 9 526 4 319 194 6 629 5 397 11 308 3 970
131 25 954 10 463 9 776 4 286 195 6 621 5 510 11 706 4 050
132 21 494 9 041 8 312 3 729 196 6 908 5 620 12 308 4 038
133 24 046 10 712 9 448 4 462 197 5 871 4 868 10 944 3 465
134 22 333 10 394 9 398 4 216 198 7 004 5 765 13 291 4 044
135 21 432 10 088 9 342 4 136 199 7 377 5 925 14 044 4 202
136 19 437 9 842 9 458 4 250 200 7 245 6 385 14 558 4 248
137 15 193 8 340 8 114 3 712 201 7 705 6 736 15 751 4 376
138 16 119 9 660 9 242 4 040 202 6 507 5 844 13 932 3 917
139 16 098 9 502 9 370 4 257 203 8 127 7 000 17 085 4 481
140 13 854 9 115 9 045 4 106 204 8 058 7 717 17 999 4 745
141 11 297 7 516 7 838 3 665 205 8 086 8 148 18 382 4 803
142 12 084 8 762 9 358 4 182 206 7 396 7 699 16 596 4 241
143 10 866 8 445 9 129 4 193 207 8 574 9 595 20 686 5 229
144 10 221 8 615 9 330 4 327 208 8 897 10 894 20 926 5 406
145 9 005 8 526 9 191 4 221 209 8 623 12 520 22 727 5 779
146 7 123 7 258 7 951 3 653 210 9 104 14 440 22 910 5 961
147 7 521 8 533 9 534 3 990 211 7 975 14 702 20 747 5 335
148 6 907 7 881 9 097 4 124 212 9 407 20 394 25 303 6 708
149 6 680 7 718 9 338 4 284 213 9 510 24 654 25 838 6 723
150 6 445 7 990 9 424 4 137 214 9 186 29 653 26 127 7 124
151 5 221 6 808 8 037 3 442 215 9 877 34 349 26 087 7 245
152 6 084 7 832 9 287 4 087 216 8 438 32 314
...