ISO/TS 17321-4:2016

TECHNICAL ISO/TS
SPECIFICATION 17321-4
First edition
Graphic technology and
photography — Colour
characterization of digital still
cameras (DSCs) —
Part 4:
Programmable light emission system
Technologie graphique et photographie — Caractérisation de la
couleur des appareils photonumériques —
Partie 4: Système d’émission de lumière programmable
PROOF/ÉPREUVE
Reference number
ISO/TS 17321-4:2016(E)
©
ISO 2016

---------------------- Page: 1 ----------------------
ISO/TS 17321-4:2016(E)

COPYRIGHT PROTECTED DOCUMENT
© ISO 2016, Published in Switzerland
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
Ch. de Blandonnet 8 • CP 401
CH-1214 Vernier, Geneva, Switzerland
Tel. +41 22 749 01 11
Fax +41 22 749 09 47
copyright@iso.org
www.iso.org
ii © ISO 2016 – All rights reserved

---------------------- Page: 2 ----------------------
ISO/TS 17321-4:2016(E)

Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Requirements . 2
4.1 General . 2
4.2 Hardware requirements . . 4
4.2.1 General. 4
4.2.2 Operating condition . 4
4.2.3 Specifications of the system . 4
4.2.4 Time stability and long-term stability of light intensity . 8
4.3 Figure of merit for a colour target using a programmable light emission system . 9
4.3.1 General. 9
4.3.2 Terms and notations of S .
R2 10
4.3.3 Method for the calculation of S .
R2 10
4.3.4 Figure of merit .11
4.4 Report .11
Annex A (informative) Integrating sphere method and LED-driving method .13
Annex B (informative) Spectral power distribution optimization procedure for multiple LEDs.15
Annex C (informative) The need for constraints on the average values and maximum values
of S and CIEDE2000 .19
R2
Annex D (informative) Evaluation method for light source generated by a programmable
light emission system .21
Annex E (informative) S and CIEDE2000 recommendations for colorimetric image capture .24
R2
Annex F (normative) S and S calculation methods for a light emission system in which
R2L R2
spectral distribution is only obtained by measurements .26
Bibliography .28
© ISO 2016 – All rights reserved PROOF/ÉPREUVE iii

---------------------- Page: 3 ----------------------
ISO/TS 17321-4:2016(E)

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 World Trade Organization (WTO) principles in the
Technical Barriers to Trade (TBT) see the following URL: www.iso.org/iso/foreword.html.
The committee responsible for this document is ISO/TC 42, Photography.
ISO 17321 consists of the following parts, under the general title Graphic technology and photography —
Colour characterization of digital still cameras (DSCs):
— Part 1: Stimuli, metrology and test procedures
— Part 2: Considerations for determining scene analysis transforms [Technical Report]
— Part 4: Programmable light emission system [Technical Specification]
The following parts are under preparation:
— Part 3: User controls and readouts for scene-referred imaging applications [Technical Report]
iv PROOF/ÉPREUVE © ISO 2016 – All rights reserved

---------------------- Page: 4 ----------------------
ISO/TS 17321-4:2016(E)

Introduction
There are many application areas such as medical imaging, cosmetics, e-commerce, sales catalogue,
fine art reproduction and artistic archive where colorimetric image capture and colorimetric image
reproduction are desired.
A high colour-fidelity imaging system using a black-and-white digital camera with rotary colour
[12] [13]
filters, and digital video cameras specified for colorimetric image capture, both of which have
the same colour sensitivity as the colour matching functions defined by CIE 1931, are available today
and fulfil these requirements. However, Reference [12] is a large-scale device which cannot be used to
capture moving objects, and Reference [13] is dedicated to motion picture use.
Digital still cameras (DSCs) are often used as convenient devices for colorimetric image capture.
Typically, DSCs do not have sensor sensitivities that are linear transforms of the colour matching
functions defined by CIE 1931. It is, therefore, necessary that a matrix conversion from DSC-image-
capture data to scene-colorimetric data be done to transform camera image data to estimates of scene
colorimetric data. Although there are several methods to derive such a matrix, a method using colour
targets is the most common when there is no data describing the DSC sensor spectral sensitivities.
1)
®
Colour targets used to derive this conversion matrix are X-Rite ColorChecker Classic , X-Rite
2)
®
ColorChecker Digital SG and others. These targets are reflective and so have a limited colour gamut
compared to scenes where the subject includes highly saturated colours. In such a case, colour targets
with highly saturated colours that can be used to derive the colour conversion matrix are very useful.
This part of ISO 17321 is applicable to light emitting devices such as inorganic or organic LEDs, quantum
dots and laser diodes.
Note that although an integrating sphere is typically used, other mechanisms would also be applicable.
A procedure using a nonlinear Generalized Reduced Gradient (GRG) algorithm is specified in this part of
ISO 17321 to minimize the square of the difference between a desired colour spectrum and the colour
spectrum of the programmable light emission system.
This part of ISO 17321-4 will make use of a metric (S ), which provides a simple and direct means
R2
to calculate the colour difference between two spectra. This criterion (S ) will be used as a method
R2
to evaluate the performance of a programmable light emission system in terms of its ability to match
a reference spectral power distribution. S and CIEDE2000 metrics are both used for colour target
R2
evaluation.
This programmable light emission system can generate arbitrary illuminants such as D55, D65 and
Illuminant A. Annex D describes evaluation metrics for light sources.
This system has several advantages as follows.
— An arbitrary smooth spectral power distribution similar to colour targets under a light source can
be produced.
— Many colour metamers can be generated easily.
— Colours with different luminance, same hue and same saturation can be generated easily.
— Colours with different saturation, same luminance and same hue can be generated easily.
— Colours with high luminance can be produced.
— Reference colour target can be provided for display systems.
®
1) ColorChecker Classic is an example of a suitable product available commercially. This information is given for
the convenience of users of this document and does not constitute an endorsement by ISO of this product.
®
2) ColorChecker Digital SG is an example of a suitable product available commercially. This information is given
for the convenience of users of this document and does not constitute an endorsement by ISO of this product.
© ISO 2016 – All rights reserved PROOF/ÉPREUVE v

---------------------- Page: 5 ----------------------
TECHNICAL SPECIFICATION ISO/TS 17321-4:2016(E)
Graphic technology and photography — Colour
characterization of digital still cameras (DSCs) —
Part 4:
Programmable light emission system
1 Scope
This part of ISO 17321 specifies requirements for a programmable light emission system to produce
various spectral radiance distributions intended for DSC colour characterization applications.
NOTE 1 Evaluation metrics are described in this part of ISO 17321. These evaluations metrics are intended
to provide “Figure of Merit (goodness)” relating to the ability of the device to produce arbitrary spectral power
distributions.
NOTE 2 This part of ISO 17321 applies to a programmable light emission system composed of LEDs. However,
it can be applied to light emitting devices such as quantum dots, organic LEDs, laser diodes and so forth.
If spiky spectral reproduction is required, it is intended that devices which have more spiky spectral
light emission be used.
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 7589, Photography — Illuminants for sensitometry — Specifications for daylight, incandescent
tungsten and printer
3 Terms and definitions
For the purposes of this document, the following terms and definitions apply.
3.1
colour matching functions
tristimulus values (3.6) of monochromatic stimuli of equal radiant power
[SOURCE: CIE Publication 17.4, 845-03-23]
3.2
colour rendering index [R]
measure of the degree to which the psychophysical colour of an object illuminated by a test illuminant
conforms to that of the same object illuminated by the reference illuminant, suitable allowance having
been made for the state of chromatic adaptation
[SOURCE: CIE Publication No. 17.4:1987, 845-02-61]
© ISO 2016 – All rights reserved PROOF/ÉPREUVE 1

---------------------- Page: 6 ----------------------
ISO/TS 17321-4:2016(E)

3.3
digital still camera
DSC
device which incorporates an image sensor and which produces a digital signal representing a still picture
Note 1 to entry: A digital still camera is typically a portable, hand-held device. The digital signal is usually
recorded on a removable memory, such as a solid-state memory card or magnetic disk.
[SOURCE: ISO 17321-1:2012, 3.2]
3.4
light-emitting diode
LED
semiconductor diode that emits non coherent optical radiation through stimulated emission resulting
from the recombination electrons and photons, when excited by an electric current
[SOURCE: IEC 60050-521, 521-04-39]
3.5
raw DSC image data
image data produced by or internal to a DSC that has not been processed, except for A/D conversion
and the following optional steps: linearization, dark current/frame subtraction, shading and sensitivity
(flat field) correction, flare removal, white balancing (e.g. so the adopted white produces equal RGB
values or no chrominance), missing colour pixel reconstruction (without colour transformations).
[SOURCE: ISO 17321-1:2012:3.4 — modified.]
3.6
tristimulus values
amount of the three reference colour stimuli, in a given trichromatic system, required to match the
colour of the stimulus considered
Note 1 to entry: See colour matching functions (3.1).
[SOURCE: CIE Publication 17.4, 845-03-22]
4 Requirements
4.1 General
Figure 1 shows a section of an integrating sphere. This sphere is one method to ensure good spatial
uniformity. Light emitting devices are placed at the bottom and an output window is placed on the side
to allow the mixed light to be emitted. Annex A shows a typical LED-driving method.
NOTE 1 Integrating sphere is a typical case, but other mechanisms would be applicable.
There are many kinds of light emitting devices. However, this part of ISO 17321 describes a
programmable light emission system using typical LEDs. Figure 2 shows typical spectral power
distributions of a number of LEDs. These LEDs will be intensity-modulated and mixed (integrated) to
produce a required spectral power distribution.
NOTE 2 Pulse width and interval modulation for intensity modulation is applicable.
NOTE 3 DSCs with automatic exposure control and automatic white balance cannot be applied for colour
calibration using this system.
2 PROOF/ÉPREUVE © ISO 2016 – All rights reserved

---------------------- Page: 7 ----------------------
ISO/TS 17321-4:2016(E)

4
1
2
3
Key
1 uniform light emission on the output window
2 output window
3 light emitting device array
4 integrating sphere
Figure 1 — Schematic configuration of the programmable light emission system
㻝㻚㻜
㻜㻚㻥
㻜㻚㻤
㻜㻚㻣
㻜㻚㻢
㻜㻚㻡
㻜㻚㻠
㻜㻚㻟
㻜㻚㻞
㻜㻚㻝
㻜㻚㻜
㻟㻤㻜㻠㻟㻜 㻠㻤㻜㻡㻟㻜 㻡㻤㻜㻢㻟㻜 㻢㻤㻜㻣㻟㻜
㼃㼍㼢㼑㼘㼑㼚㼓㼠㼔 㻔㼚㼙㻕
Figure 2 — Example of spectral power distributions for a chosen set of LEDs
© ISO 2016 – All rights reserved PROOF/ÉPREUVE 3
㻺㼛㼞㼙㼍㼘㼕㼦㼑㼐㻌㻵㼚㼠㼑㼚㼟㼕㼠㼥

---------------------- Page: 8 ----------------------
ISO/TS 17321-4:2016(E)

4.2 Hardware requirements
4.2.1 General
This Clause is to describe a light emitting system for DSC colour characterization that uses an
integrating sphere.
4.2.2 Operating condition
The light emitting system shall be designed to operate consistently under the following ranges.
NOTE “Temperature” condition was referred from ISO 12646.
— Temperature: 18 °C to 28 °C.
— Relative humidity: 15 % to 80 %.
4.2.3 Specifications of the system
4.2.3.1 Wavelength
The wavelength range over which the combined set of the light emissive devices is evaluated shall be
380 nm to 730 nm and should be 360 nm to 830 nm.
NOTE 1 The procedure to configure an LED array to achieve required spectral power distribution and
chromaticity is described in Annex B.
NOTE 2 Evaluation metric (S ) described in 4.2.3.2 can be applied to more extended range including IR/UV
R2L
components when necessary.
4.2.3.2 Objective reference light source and calculated-reference light source
ISO 17321-1 describes that the spectral power distribution for illuminating the test target shall be
photographic daylight, D55, as defined in ISO 7589. The standard illuminant D55 shall be used as a
reference light source in this part of ISO 17321. A light source which is generated with a programmable
light emission system is obtained by minimizing the S value in Formula (1). This optimization
R2L
method is described in Annex B.
NOTE This programmable light emission system can generate other illuminants such as D65, A and so forth.
D65 is used as the default illuminant for video uses.
4.2.3.2.1 Optimization procedure of a programmable light emission system to a reference
light source
The mean of the squares of the differences between the objective reference and calculated light source
spectral power distributions (S ) is specified by:
R2L
2
 
N






()LP− EY/ Y

∑ NV
 ri Sci



 

i=1
 
S = (1)
RL2
N
N
YL= / N (2)
N ∑
ri
i=1
4 PROOF/ÉPREUVE © ISO 2016 – All rights reserved

---------------------- Page: 9 ----------------------
ISO/TS 17321-4:2016(E)

N
YV= LY/ (3)
VN∑
ri ri
i=1
N
V = 1 (4)
∑ ri
i=1
where
L is the reference light source spectrum of the i-th wavelength;
ri
E is the spectrum of the i-th wavelength calculated for the programmable light emission system;
ci
P is the scaling coefficient to adjust energy power level;
S
Y is the normalization factor on averaged power of the objective reference light source;
N
Y is the factor to compensate light source dependence using luminosity factor in case of rela-
V
tive light source value;
V is the normalized-response of the the i-th wavelength derived from the luminosity function;
ri
N is number of wavelength samples (i = 1, N).
The optimization procedure is as follows.
a) Measure the (absolute) spectral power distribution of each LED at its maximum intensity. This shall
2
be done in the same way with the same units used, for example watts/(Sr × m × nm), for each LED.
b) Using the optimization procedure described in Annex B, minimize the S value. This procedure
R2L
calculates an LED intensity coefficient b of measured spectrum intensity ε for j-th LED. A value of
j ij
P is determined as follows. When one of the LEDs will be driven at its maximum coefficient, P is
s s
set to be the maximum value of b ( j = 1, M). P is specified as the scaling coefficient for generating
j s
D55 illuminant or arbitrary light sources.
The following notations are used.
N is number of wavelength samples (i = 1, N).
M is number of LEDs.
b is intensity vector having b component ( j = 1, M).
j
  ε is measured-spectral intensity of j-th LED ( j = 1, M, i = 1, N).
ij
c) Multiply each LED’s measured spectral power distribution by the corresponding LED intensity
coefficient b , and sum all of spectral power distributions to obtain E .
j ci
d) Divide the summed spectral power distribution by P to obtain the calculated-spectral distribution
s
E of the programmable light emission system. E is an output candidate spectrum distribution
ci ci
using the programmable light emission system.
M
Eb= ε / P (5)
ci ∑ jijS
j=1
e) The calculated-reference light source spectrum distribution L′ corresponding to E is obtained by
ri ci
dividing L by P .
ri s
© ISO 2016 – All rights reserved PROOF/ÉPREUVE 5

---------------------- Page: 10 ----------------------
ISO/TS 17321-4:2016(E)

LL' = / P (6)
ri ri S
L′ is used as the reference light source for optimization in 4.3.2 and 4.3.3.
ri
The use of Y is optional. In cases where the luminosity function is applied independently (i.e. Y is not
V V
used), the value of Y should be set to 1,0.
V
NOTE 1 If normalization component L /Y is not used, S values of illuminants with strong spectral peaks
ri N R2L
such as fluorescent lamps are smaller than those values of flat-like illuminants such as illuminant D55 and D65.
The maximum intensity of each light emitting device of the programmable light emission system can be
determined beforehand from its hardware specification. It is recommended that these values are used
as maximum constraint conditions for the optimization method described in Annex B in order to achieve
the maximum intensity of each light emitting device. In this case, the intensity value is the maximum
energy level of the corresponding light emitting device. In practice, the intensity values of light emitting
devices should be set to lower levels than the maximum energy possible when reproducing spectra of
objects colours. Intensity values for every light emitting devices for illuminant D55 can be determined
using the optimization method with these constraint conditions.
NOTE 2 Exposure level and white balance setting for DSC can be achieved using light source power
distribution above.
4.2.3.2.2 Evaluation for calculated light source
The evaluation for calculated light source of the programmable light emission system (S ) is specified
R2L
by Formula (1).
Annex F describes the calculation procedure of S if both relative spectral distribution of a reference
R2L
light source and absolute spectral distribution of a measured light source are given.
4.2.3.3 Size, luminance, uniformity and angular characteristic
4.2.3.3.1 Output window
The output window (see Figure 1) shall be at least 50 mm in diameter.
4.2.3.3.2 Minimum luminance
2
The minimum luminance of the output window shall be greater than 40 cd/m and should be greater
2
than 80 cd/m when simulating various light sources including fluorescent and LED light sources.
2
NOTE 80 cd/m provides a luminance that allows an object illuminated by the source to be photographed
satisfactorily at a distance of 50 cm using a DSC with an ISO speed of 200, an aperture of F5,6, a shutter speed
of 1/30 s.
4.2.3.3.3 Uniformity
The luminance measured at the centre and at 8 points evenly spaced around the circumference of the
output window at 45° intervals shall differ by no more than ±2 %. The luminance measurements are
made normal to the plane of the output window at each measurement point.
6 PROOF/ÉPREUVE © ISO 2016 – All rights reserved

---------------------- Page: 11 ----------------------
ISO/TS 17321-4:2016(E)

0
315
45
90
270 C
135
225
180
Figure 3 — Measurement points on the output window with every 45°
Figure 3 shows measurement points on the output window to calculate uniformity characteristics.
Uniformity is defined by the Formula (7):
YY−
k ave
DY = ×100 (7)
u
Y
ave
where, Y is the average luminance of Y and Y . Y is the luminance at the centre point and Y is luminance
ave c k c k
at the circumference point for k = 0°, 45°, 90°, 135°, 180°, 225°, 270°, 315°. ∆Y shall be within ±2 %.
4.2.3.3.4 Angular characteristics
The luminance, when measured within a 10° cone angle to the normal line of the centre of the output
window, shall differ by no more than ±2 %.
An integrating sphere, similar to that shown in Annex A, should be used in order to ensure uniformity
across the output window. Figure 4 shows angular characteristic measurement method. It is the typical
method for a colour target having an arrangement of an output window and light sources shown in
Figure 1.
+θ
Y
+θ
Y
nor
Y
-θ −θ
Output window
Figure 4 — Angular characteristics measurement method
© ISO 2016 – All rights reserved PROOF/ÉPREUVE 7

---------------------- Page: 12 ----------------------
ISO/TS 17321-4:2016(E)

YY−
±nθ or
ΔY = ×100 (8)
θ
Y
nor
Where, Y is luminance measured along the axis normal to the output window, Y or Y is luminance
nor +θ -θ
measured along the axis which is inclined +θ or -θ to the normal axis, respectively. Maximum θ is 5° of
arc. ∆Y shall be within ±2 %.
4.2.4 Time stability and long-term stability of light intensity
The following method shows how to calculate the time stability of light intensity.
Step 1: Generate CIE Illuminant D55 using the light emitting system.
Step 2: Allow the system to warm-up for an optimum time (e.g. 35 min) to reach stable state.
Step 3: Measure spectral power distribution every five minutes
Step 4: Repeat measurement of the spectral power distribution (step 3) at least seven times.
Step 5: Calculate ρ as shown in Formula (9) and repeat for i = 1 to N to obtain averaged spectral
avi
power distribution ρ .
av
M
ρ
()
∑
ij
j=1
ρ = (9)
avi
M
where
ρ is spectral intensity of i-th wavelength at j-th iteration ( j = 1, M);
ij
ρ is time-averaged-spectral intensity of i-th wavelength;
avi
ρ is spectral power distribution having ρ components, where i = 1, N;
av avi
N is the number of wavelength samples (i = 1, N);
M is the number of the measurement iterations.
The wavelength range to be evaluated shall be 380 nm to 730 nm and should be 380 nm to 780 nm.
NOTE 1 M = 7 would be appropriate.
Step 6: Calculate tristimulus values X Y Z of light source using ρ and CIE colour matching functions.
av av av av
Step 7: Calculate tristimulus values X Y Z of light source using ρ (i = 1, N) and CIE colour matching
j j j ij
functions.
NOTE 2 X Y Z can be obtained by direct measurement with an appropriate instrument. X Y Z are calculated
j j j av av av
by averaging operation of X Y Z ( j = 1, M).
j j j
Step 8: Calculate CIEDE2000 using X Y Z and X Y Z .
j av av av j j j
NOTE 3 CIEDE2000 is calculated under illuminant D55 condition.
8 PROOF/ÉPREUVE © ISO 2016 – All rights reserved

---------------------- Page: 13 ----------------------
ISO/TS 17321-4:2016(E)

Step 9: Calculate CIEDE2000 , a time average of CIEDE2000 , as shown in Formula (10):
av j
M
CIEDE2000
()
∑ j
j=1
CIEDE2000 = (10)
av
M
Step 10: Calculate standard deviation SD. This standard deviation means reproducibility of the light
emitting system.
M
2
CIEDE2000 −CIEDE2000
()
∑
jav
j=1
2
SD = (11)
M
Step 11: Calculate maximum CIEDE2000 of CIEDE2000
max j
Figure 5 shows X Y Z .
j j j
Requirements for reproducibility and maximum CIEDE2000 are described in Table 1.
NOTE 4 An optimum time to reach stability is 35 min in Figure 5.
tristimulus value Y tristimulus value Z
tristimulus value X
140 140 130
130 130 120
120 120 110
X Y Z
110 110 100
Xj Yj Zj
100 100 90
90 80
90
05 10 15 20 25 30 35 40 45 50 55 60 65 05 10 15 20 25 30 35 40 45 50 55 60 65 05 10 15 20 25 30 35 40 45 50 55 60 65
time (minutes) time (minutes)
time (minutes)
Figure 5 — X Y Z for measuring time stability of light source
j j j
Table 1 — Requirements for time stability of light intensity
SHALL SHOULD
Reproducibility (SD) ∆E (CIEDE2000) ≤ 0,3 ∆E (CIEDE2000) ≤ 0,1
00 00
Maximum CIEDE2000 ∆E (CIEDE2000) ≤ 0,6 ∆E (CIEDE2000) ≤ 0,2
00 00
NOTE 5 The spectral power distribution of LEDs is very dependent on temperature and so it is very important
that the temperature of LED array remains stable over time.
In order to determine the long-term stability, the tristimulus values of (X ,Y ,Z ) of the light source shall
j j j
be measured once a month following the warm-up period (e.g. 35 min) and recorded as described in 4.4.
Annex A shows one method to maintain stability over time using pulse width modulation in order to
avoid spectral shift of LEDs with current change.
4.3 Figure of merit for a colour target using a programmable light emission system
4.3.1 General
A figure of merit for the programmable light emission system can be represented by the combination of
...