ISO 12233:2014

Photography - Electronic still picture imaging - Resolution and spatial frequency responses

ISO 12233:2014 specifies methods for measuring the resolution and the SFR of electronic still-picture cameras. It is applicable to the measurement of both monochrome and colour cameras which output digital data or analog video signals.

Photographie — Imagerie des prises de vues électroniques — Résolution et réponses en fréquence spatiale

Fotografija - Elektronsko prikazovanje mirujočih slik - Resolucija (ločljivost) in frekvenčni odzivi v prostoru

Ta mednarodni standard določa metode za merjenje ločljivosti in frekvenčnega odziva v prostoru (SFR) elektronskih fotoaparatov za mirujoče slike. Uporablja se za merjenje monokromatskih in barvnih fotoaparatov, ki oddajajo digitalne podatke ali analogne video signale.

General Information

• Status: Withdrawn
• Publication Date: 10-Feb-2014
• Withdrawal Date: 10-Feb-2014
• ICS: 37.040.10 - Photographic equipment. Projectors
• Technical Committee: ISO/TC 42 - Photography
• Drafting Committee: ISO/TC 42/WG 18 - Electronic still picture imaging
• Current Stage: 9599 - Withdrawal of International Standard
• Start Date: 10-Jan-2017
• Completion Date: 13-Dec-2025
• Ref Project: SIST ISO 12233:2014 - Photography - Electronic still picture imaging - Resolution and spatial frequency responses

Relations

• Revised: ISO 12233:2017 - Photography - Electronic still picture imaging - Resolution and spatial frequency responses
• Effective Date: 02-Jul-2016
• Revises: ISO 12233:2000 - Photography - Electronic still-picture cameras - Resolution measurements
• Effective Date: 26-Feb-2011

Standards Content (Sample)

INTERNATIONAL ISO
STANDARD 12233
Second edition
2014-02-15
Photography — Electronic still picture
imaging — Resolution and spatial
frequency responses
Photographie — Imagerie des prises de vues électroniques —
Résolution et réponses en fréquence spatiale
Reference number
©
ISO 2014
© ISO 2014
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 2014 – All rights reserved

Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Test conditions . 5
4.1 Test chart illumination . 5
4.2 Camera framing and lens focal length setting . 5
4.3 Camera focusing . 5
4.4 Camera settings . 5
4.5 White balance. 6
4.6 Luminance and colour measurements . 6
4.7 Gamma correction . 6
5 Visual resolution measurement . 6
5.1 General . 6
5.2 Test chart. 7
5.3 Rules of judgement for visual observation . 8
6 Edge-based spatial frequency response (e-SFR) . 9
6.1 General . 9
6.2 Methodology . 9
7 Sine-based spatial frequency response (s-SFR) measurement .12
8 Presentation of results .13
8.1 General .13
8.2 Resolution .13
8.3 Spatial frequency response (SFR) .14
Annex A (informative) CIPA resolution test chart .17
Annex B (informative) Visual resolution measurement software .23
Annex C (informative) Low contrast edge SFR test chart with OECF patches .28
Annex D (normative) Edge spatial frequency response (e-SFR) algorithm .30
Annex E (normative) Sine wave star test chart .33
Annex F (normative) Sine wave Spatial Frequency Response (s-SFR) analysis algorithm .35
Annex G (informative) Colour-filtered resolution measurements .40
Annex H (informative) Units and summary metrics .42
Annex I (informative) Original test chart defined in ISO 12233:2000 .45
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 ISO/TC 42, Photography.
This second edition cancels and replaces the first edition (ISO 12233:2000), which has been technically
revised.
iv © ISO 2014 – All rights reserved

Introduction
0.1 Purpose
The spatial resolution capability is an important attribute of an electronic still-picture camera. Resolution
measurement standards allow users to compare and verify spatial resolution measurements. This
International Standard defines terminology, test charts, and test methods for performing resolution
measurements for analog and digital electronic still-picture cameras.
0.2 Technical background
For consumer digital cameras, the term resolution is often incorrectly interpreted as the number of
addressable photoelements. While there are existing protocols for determining camera pixel counts,
these should not be confused with the interpretation of resolution as addressed in this International
Standard. Qualitatively, resolution is the ability of a camera to optically capture finely spaced detail, and
is usually reported as a single valued metric. Spatial frequency response (SFR) is a multi-valued metric
that measures contrast loss as a function of spatial frequency. Generally, contrast decreases as a function
of spatial frequency to a level where detail is no longer visually resolved. This limiting frequency value
is the resolution of the camera. A camera’s resolution and its SFR are determined by a number of factors.
These include, but are not limited to, the performance of the camera lens, the number of addressable
photoelements in the optical imaging device, and the electrical circuits in the camera, which can include
image compression and gamma correction functions.
While resolution and SFR are related metrics, their difference lies in their comprehensiveness and
utility. As articulated in this International Standard, resolution is a single frequency parameter that
indicates whether the output signal contains a minimum threshold of detail information for visual
detection. In other words, resolution is the highest spatial frequency that a candidate camera can
usefully capture under cited conditions. It can be very valuable for rapid manufacturing testing, quality
control monitoring, or for providing a simple metric that can be easily understood by end users. The
algorithm used to determine resolution has been tested with visual experiments using human observers
and correlates well with their estimation of high frequency detail loss.
SFR is a numerical description of how contrast is changed by a camera as a function of the spatial
frequencies that describe the contrast. It is very beneficial for engineering, diagnostic, and image
evaluation purposes and serves as an umbrella function from which such metrics as sharpness and
acutance are derived. Often, practitioners will select the spatial frequency associated with a specified
SFR level as a modified non-visual resolution value.
In a departure from the first edition of this International Standard, two SFR measurements are described.
Additionally, the first SFR metrology method, edge-based spatial frequency response, is identical to
that described in the first edition, except that a lower contrast edge is used for the test chart. Regions
of interest (ROI) near slanted vertical and horizontal edges are digitized and used to compute the SFR
levels. The use of a slanted edge allows the edge gradient to be measured at many phases relative to the
image sensor photoelements and to yield a phase-averaged SFR response.
A second sine wave-based SFR metrology technique is introduced in this edition. Using a sine wave
modulated target in a polar format (e.g. Siemens star), it is intended to provide an SFR response that
is more resilient to ill-behaved spatial frequency signatures introduced by the image content driven
processing of consumer digital cameras. In this sense, it is intended to enable easier interpretation of
SFR levels from such camera sources. Comparing the results of the edge-based SFR and the sine based
SFR might indicate the extent to which nonlinear processing is used.
The first step in determining visual resolution or SFR is to capture an image of a suitable test chart with
the camera under test. The test chart should include features of sufficiently fine detail and frequency
content such as edges, lines, square waves, or sine wave patterns. The test chart defined in this
International Standard has been designed specifically to evaluate electronic still-picture cameras. It has
not necessarily been designed to evaluate other electronic imaging equipment such as input scanners,
CRT displays, hard-copy printers, or electro-photographic copiers, nor individual components of an
electronic still-picture camera, such as the lens.
Some of the measurements described in this International Standard are performed using digital analysis
techniques. They are also applicable with the analogue outputs of the camera by digitizing the analogue
signals if there is adequate digitizing equipment.
0.3 Methods for measuring SFR and resolution — selection rationale and guidance
This section is intended to provide more detailed rationale and guidance for the selection of the different
resolution metrology methods presented in this International Standard. While resolution metrology of
analog optical systems, by way of spatial frequency response, is well established and largely consistent
between methodologies (e.g. sine waves, lines, edges), metrology data for such systems is normally
captured under well controlled conditions where the required data linearity and spatial isotropy
assumptions hold. Generally, it is not safe to assume these conditions for files from many digital
cameras, even under laboratory capture environments. Exposure and image content dependent image
processing of the digital image file before it is provided as a finished file to the user prevents this. This
processing yields different SFR responses depending on the features in the scene or in the case of this
International Standard, the target. For instance, in-camera edge detection algorithms might specifically
operate on edge features and selectively enhance or blur them based on complex nonlinear decision
rules. Depending on the intent, these algorithms might also be tuned differently for repetitive scene
features such as those found in sine waves or bar pattern targets. Even for constrained camera settings
recommended in this International Standard, these nonlinear operators can yield differing SFR results
depending on the target feature set. Naturally, this causes confusion on which targets to use, either
alone or in combination. Guidelines for selection are offered below.
Edges are common features in naturally occurring scenes. They also tend to act as visual acuity cues
by which image quality is judged and imaging artefacts are manifested. This logic prescribed their use
for SFR metrology in the past and current editions of this International Standard. It is also why edge
features are prone to image processing in many consumer digital cameras: they are visually important.
All other imaging conditions being equal, camera SFRs using different target contrast edge features can
be significantly different, especially with respect to their morphology. This is largely due to nonlinear
image processing operators and would not occur for strictly linear imaging systems. To moderate this
behaviour, a lower contrast slanted edge feature (Figure C.1) was chosen to replace the higher contrast
version of the first edition. This feature choice still allows for acuity-amenable SFR results beyond the
half-sampling frequency and helps prevent nonlinear data clipping that can occur with high contrast
target features. It is also a more reliable rendering of visually important contrast levels in naturally
occurring scenes.
Sine wave features have long been the choice for directly calculating SFR of analogue imaging systems
and they are intuitively satisfying. They have been introduced into this edition of this International
Standard based on experiences from the edge-based approach. Because sine waves transition more
slowly than edges, they are not prone to being identified as edges in embedded camera processors.
As such, the ambiguity that image processing imposes on the SFR can be largely avoided by their use.
Alternatively, if the image processing is influenced by the absence of sharp features, more aggressive
processing might be used by the camera. A sine wave starburst test pattern (Figure 6) is adopted in
this edition. With the appropriate analysis software, a sine wave-based SFR can be calculated up to the
half-sampling frequency. For the same reasons stated above, the sine wave-based target is also of low
contrast and consistent with that of the edge-based version. An added benefit of the target’s design over
other sine targets is its compactness and bi-directional features.
All experience suggests that there is no single SFR for today’s digital cameras. Even under the strict
capture constraints suggested in this International Standard, the allowable feature sets that most
digital cameras offer prevent such unique characterization. Confusion can be reduced through complete
documentation of the capture conditions and camera setting for which the SFR was calculated. It has been
suggested that comparing edge- and sine wave-based SFR results under the same capture conditions
could be a good tool in assessing the contribution of spatial image processing in digital cameras.
Finally, at times, a full SFR characterization is simply not required, such as in end of line camera assembly
testing. Alternately, SFR might be an intimidating obstacle to those not trained in its utility. For those
in need of a simple and intuitive space domain approach to resolution using repeating line patterns, a
visual resolution metric is also provided in this second edition of this International Standard.
vi © ISO 2014 – All rights reserved

With such a variety of methods available for measuring resolution, there are bound to be differences
in measured resolution results. To benchmark the likely variations, the committee has published the
results of a pilot study using all of the proposed techniques and how they relate to each other. These
results are provided in Reference [20].
INTERNATIONAL STANDARD ISO 12233:2014(E)
Photography — Electronic still picture imaging —
Resolution and spatial frequency responses
1 Scope
This International Standard specifies methods for measuring the resolution and the SFR of electronic
still-picture cameras. It is applicable to the measurement of both monochrome and colour cameras
which output digital data or analog video signals.
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 14524, Photography — Electronic still-picture cameras — Methods for measuring opto-electronic
conversion functions (OECFs)
3 Terms and definitions
For the purposes of this document, the following terms and definitions apply.
3.1
addressable photoelements
number of active photoelements in an image sensor
Note 1 to entry: This equals the product of the number of active photoelement lines and the number of active
photoelements per line.
3.2
aliasing
output image artefacts that occur in a sampled imaging system due to insufficient sampling
Note 1 to entry: These artefacts usually manifest themselves as moiré patterns in repetitive image features or as
jagged stair-stepping at edge transitions.
3.3
cycles per millimetre
cy/mm
spatial frequency unit defined as the number of spatial periods per millimetre
3.4
edge spread function
ESF
normalized spatial signal distribution in the linearized output of an imaging system resulting from
imaging a theoretical infinitely sharp edge
3.5
effectively spectrally neutral
having spectral characteristics which result in a specific imaging system producing the same output as
for a spectrally neutral object
3.6
electronic still-picture camera
camera incorporating an image sensor that outputs an analog or digital signal representing a still picture
Note 1 to entry: This camera may also record or store an analog or digital signal representing a still picture on a
removable media, such as a memory card or magnetic disc.
3.7
gamma correction
signal processing operation that changes the relative signal levels
Note 1 to entry: Gamma correction is performed in part to correct for the nonlinear light output versus signal
input characteristics of the display. The relationship between the light input level and the output signal level,
called the camera opto-electronic conversion function (OECF), provides the gamma correction curve shape for an
image capture device.
Note 2 to entry: The gamma correction is usually an algorithm, lookup table, or circuit which operates separately
on each colour component of an image.
3.8
horizontal resolution
resolution value measured in the longer image dimension, corresponding to the horizontal direction for
a “landscape” image orientation, typically using a vertical or near vertical oriented test-chart feature
3.9
image aspect ratio
ratio of the image width to the image height
3.10
image compression
process that alters the way digital image data is encoded to reduce the size of an image file
3.11
image sensor
electronic device that converts incident electromagnetic radiation into an electronic signal
EXAMPLE Charge coupled device (CCD) array, complementary metal-oxide semiconductor (CMOS) array.
3.12
line pairs per millimetre
lp/mm
spatial frequency unit defined as the number of equal width black and white line pairs per millimetre
3.13
line spread function
LSF
normalized spatial signal distribution in the linearized output of an imaging system resulting from
imaging a theoretical infinitely thin line
3.14
line widths per picture height
LW/PH
spatial frequency unit for specifying the width of a feature on a test chart relative to the height of the
active area of the chart
Note 1 to entry: the value in LW/PH indicates the total number of lines of the same width which can be placed edge
to edge within the height of a test target or within the vertical field of view of a camera.
Note 2 to entry: This unit is used whatever the orientation of the “feature” (e.g. line). Specifically, it applies to
horizontal, vertical, and diagonal lines.
EXAMPLE If the height of the active area of the chart equals 20 cm, a black line of 1 000 LW/PH has a width
equal to 20/1 000 cm.
2 © ISO 2014 – All rights reserved

3.15
linearized
digital signal conversion performed to invert the camera opto-electronic conversion function (OECF) to
focal plane exposure or scene luminance
3.16
lines per millimetre
lines/mm
spatial frequency unit defined as the number of equal width black and white lines per millimetre
Note 1 to entry: One line pair per millimetre (lp/mm) is equal to 2 lines/mm.
3.17
modulation
normalized amplitude of signal levels
Note 1 to entry: This is the difference between the minimum and maximum signal levels divided by the average
signal level.
3.18
modulation transfer function
MTF
modulus of the optical transfer function
Note 1 to entry: For the MTF to have significance, it is necessary that the imaging system be operating in an
isoplanatic region and in its linear range. Because most electronic still-picture cameras provide spatial colour
sampling and nonlinear processing, a meaningful MTF of the camera can only be approximated through the SFR.
See ISO 15529:2010.
3.19
normalized spatial frequency
spatial frequency unit for specifying resolution characteristics of an imaging system in terms of cycles
per pixel rather than in cycles/millimetre or any other unit of length
3.20
optical transfer function
OTF
two-dimensional Fourier transform of the imaging system’s point spread function
Note 1 to entry: For the OTF to have significance, it is necessary that the imaging system be operating in an
isoplanatic region and in its linear range. The OTF is a complex function whose modulus has unity value at zero
spatial frequency. (See ISO 9334). Because most electronic still-picture cameras provide spatial colour sampling
and nonlinear processing, a meaningful OTF of the camera can only be approximated through the SFR.
3.21
point spread function
PSF
normalized spatial signal distribution in the linearized output of an imaging system resulting from
imaging a theoretical infinitely small point source
3.22
resolution
measure of the ability of a camera system, or a component of a camera system, to depict picture detail
3.23
sampled imaging system
imaging system or device which generates an image signal by sampling an image at an array of discrete
points, or along a set of discrete lines, rather than a continuum of points
Note 1 to entry: The sampling at each point is done using a finite-size sampling aperture or area.
3.24
spatial frequency response
SFR
relative amplitude response of an imaging system as a function of input spatial frequency
Note 1 to entry: The SFR is normally represented by a curve of the output response to an input sinusoidal spatial
luminance distribution of unit amplitude, over a range of spatial frequencies. The SFR is divided by its value at the
spatial frequency of 0 as normalization to yield a value of 1,0 at a spatial frequency of 0.
3.24.1
edge-based spatial frequency response
e-SFR
measured amplitude response of an imaging system to a slanted-edge input, as defined in Clause 6
3.24.2
sine wave-based spatial frequency response
s-SFR
measured amplitude response of an imaging system to a range of sine wave inputs, as defined in Clause 7
3.25
spectrally neutral
exhibiting reflective or transmissive characteristics which are constant over the wavelength range of
interest
3.26
test chart
arrangement of test patterns designed to test particular aspects of an imaging system
3.27
test pattern
specified arrangement of spectral reflectance or transmittance characteristics used in measuring an
image quality attribute
3.27.1
bi-tonal pattern
pattern that is spectrally neutral or effectively spectrally neutral, and consists exclusively of two
reflectance or transmittance values in a prescribed spatial arrangement
Note 1 to entry: Bi-tonal patterns are typically used to measure resolution by visual resolution method.
3.27.2
hyperbolic wedge test pattern
bi-tonal pattern that varies continuously and linearly with spatial frequency
Note 1 to entry: A bi-tonal hyperbolic wedge test pattern is used to measure resolution by the visual resolution
method in this International Standard.
3.28
vertical resolution
resolution value measured in the shorter image dimension, corresponding to the vertical direction for a
“landscape” image orientation, typically using a horizontal or near horizontal oriented test-chart feature
3.29
visual resolution
spatial frequency at which all of the individual black and white lines of a test pattern frequency can no
longer be distinguished by a human observer
Note 1 to entry: This presumes the features are reproduced on a display or print.
4 © ISO 2014 – All rights reserved

4 Test conditions
4.1 Test chart illumination
The luminance of the test chart shall be sufficient to provide an acceptable camera output signal level.
The test chart shall be uniformly illuminated as shown in Figure 1, so that the illuminance at the
chart is within ±10 % of the illuminance in the centre of the chart at any position within the chart.
The illumination sources should be baffled to prevent direct illumination of the camera lens by the
illumination sources. The area surrounding the test chart should be of low reflectance, to minimize
flare light. The chart should be shielded from any reflected light. The illuminated test chart shall be
effectively spectrally neutral within the visible wavelengths.
Figure 1 — Test chart illumination method
4.2 Camera framing and lens focal length setting
The camera shall be positioned to properly frame the test target. The vertical framing arrows are used
to adjust the magnification and the horizontal arrows are used to centre the target horizontally. The
tips of the centre vertical black framing arrows should be fully visible and the tips of the centre white
framing arrows should not be visible. The target shall be oriented so that the horizontal edge of the
chart is approximately parallel to the horizontal camera frame line. The approximate distance between
the camera and the test chart should be reported along with the measurement results.
4.3 Camera focusing
The camera focus should be set either by using the camera autofocusing system, or by performing a series
of image captures at varying focus settings, and selecting the focus setting that provides the highest
average modulation level at a spatial frequency approximately 1/4 the camera Nyquist frequency. (In the
case of a colour camera, the Nyquist frequency is of the conceptual monochrome image sensor without
colour filter array). Auto focus accuracy is often limited and this limitation might have an impact on the
results.
4.4 Camera settings
The camera lens aperture (if adjustable) and the exposure time should be adjusted to provide a near
maximum signal level from the white test target areas. The settings shall not result in signal clipping in
either the white or black areas of the test chart, or regions of edge transitions.
Electronic still-picture cameras might include image compression, to reduce the size of the image files
and allow more images to be stored. The use of image compression can significantly affect resolution
measurements. Some cameras have switches that allow the camera to operate in various compression
or resolution modes. The values of all camera settings that might affect the results of the measurement,
including lens focal length, aperture and image quality (i.e. recording pixel number or compression)
mode (if adjustable) shall be reported along with the measurement results.
Multiple SFR measurements can be reported for different camera settings, including a setting that uses
the maximum lens aperture size (minimum f-number) and maximum camera gain.
4.5 White balance
For a colour camera, the camera white balance should be adjusted, if possible, to provide proper white
balance [equal red, green, and blue (RGB) signal levels] for the illumination light source, as specified in
ISO 14524.
4.6 Luminance and colour measurements
Resolution measurements are normally performed on the camera luminance signal. For colour cameras
that do not provide a luminance output signal, a luminance signal should be formed from an appropriate
combination of the colour records, rather than from a single channel such as green. The reader is
referred to ISO 12232 for the luminance signal calculation. Colour-filtered resolution measurements
can be performed as described in informative Annex G.
4.7 Gamma correction
The signal representing the image from an electronic still-picture camera will probably be a nonlinear
function of the scene luminance values. Since the SFR measurement is defined on a linearized output
signal, and such a nonlinear response can affect SFR values, the signal shall be linearized before the
data analysis is performed. Linearization is accomplished by applying the inverse of the camera OECF
to the output signal via a lookup table or appropriate equation. The measurement of the OECF shall be
as specified in ISO 14524, using the standard reflection camera OECF test chart or using an integrated
OECF/resolution chart.
5 Visual resolution measurement
5.1 General
The visual resolution is the maximum value of the spatial frequency in LW/PH within a test pattern that
is able to be visually distinguished. A black and white hyperbolic wedge is used as the test pattern
Because of aliasing artefacts in the high frequencies, actual resolution judgements can be ambiguous.
The objective visual resolution method cited herein using a hyperbolic wedge test pattern gives more
stable results by adopting the visual judgement rules described in 5.3 which have been used by a highly
skilled observer.
It can be measured analytically using computer analysis of captured images, as defined in Annex B. The
computer analysis method is intended to correlate with the subjective judgement of visual resolution
made by a skilled observer but is likely to yield a more consistent and objective result compared to
actual visual judgements. However, if there is a discrepancy between the results of the computer analysis
method and the judgement of a human observer, the judgement of the human observer takes priority.
6 © ISO 2014 – All rights reserved

5.2 Test chart
5.2.1 General
The preferred test chart for measuring the visual resolution is the CIPA resolution chart, which is shown
in Figure 2 and specified in informative Annex A.
The chart shown in Figure 2 is designed to measure cameras having a resolution of less than 2 500 LW/PH.
Nevertheless, it is possible to use the chart to measure the resolution of an electronic still camera having
a resolution greater than 2 500 LW/PH. This is accomplished by adjusting the camera to target distance,
or the focal length of the camera lens, so that the test chart active area fills only a portion of the vertical
image height of the camera. This fraction is then measured in the digital image, by dividing the number
of image lines in the camera image by the number of lines in the active chart area. The values of all test
chart features, in LW/PH, printed on the chart or specified in this International Standard, are multiplied
by this fraction, to obtain their correct values. For example, if the chart fills 1/2 of the vertical image
height of the camera, then the multiplication factor is equal to 2 and a feature labelled as 2 000 LW/PH
on the chart corresponds to 4 000 LW/PH using this chart framing.
NOTE Figure 2 includes an improved version of the test chart features originally defined in ISO 12233:2000.
5.2.2 Material
The test chart can be either a transparency that is rear illuminated, or a reflection test card that is front
illuminated. A reflection chart shall have an approximately Lambertian base material. A transparency
chart shall be rear illuminated by a diffuse source.
5.2.3 Size
The active height of reflection test charts should be no less than 20 cm. The active height of transparencies
shall be not less than 10 cm.
5.2.4 Test patterns
The test chart shall have bi-tonal patterns and should be spectrally neutral.
NOTE Bi-tonal test charts are easily manufactured and minimize the cost of producing the chart.
5.2.5 Test pattern modulation
For reflective charts, the ratio of the maximum chart reflectance, R , to the minimum chart
max
reflectance, R , for large test pattern areas should be not less than 40:1 and not greater than 80:1,
min
and shall be reported if it is outside this range. For transmissive charts, the ratio of the maximum chart
transmittance, T , to the minimum chart transmittance, T , for a large test pattern should be not
max min
less than 40:1 and not greater than 80:1, and shall be reported if it is outside this range. For a paper base
optical density of 0,10, these minimum and maximum numbers translate to optical densities of 1,7 and
2,0 respectively. Modulation ratios for the finer test chart features, relative to the ratio for large test
pattern areas, should preferably be reported by the chart manufacturer for reference.
5.2.6 Positional tolerance
The position of any test chart feature shall be reproduced with a tolerance of ±1/1 000 picture heights
(equivalent to ±1/10 % of the active test chart height). In addition, the width and duty cycle ratio of each
feature (white or black line) of the wedge pattern shall be reproduced with a tolerance of ±5 % of the
feature width.
Figure 2 — CIPA resolution test chart
5.3 Rules of judgement for visual observation
5.3.1 Rules of judgement
The viewer shall observe the following rules when judging the resolution value. These rules are intended
to achieve correct measurement value in the presence of unavoidable aliasing artefacts.
a) Beginning from the low frequency side, treat a spatial frequency as “Resolved” only when all lower
spatial frequencies are also resolved. The resolution limit is achieved at the line just before the first
occurrence of unresolved line features.
b) Treat a spatial frequency as “Not resolved” when the black and white lines appear to change polarity
or lines are blurred together to produce a reduced number of lines, compared to the number in the
test chart.
5.3.2 An example of a correct visual judgement
The boundary between the resolved and not resolved regions indicated by a dashed arrow is the
corresponding point to resolution value to be measured as shown in Figure 3.
Figure 3 — Correct application of the wedge feature interpretation
8 © ISO 2014 – All rights reserved

6 Edge-based spatial frequency response (e-SFR)
6.1 General
The edge-based spatial frequency response (e-SFR) of an electronic still-picture camera is measured
by analysing the camera data near a slanted low contrast neutral edge. The preferred test chart for
measuring e-SFR is shown in Figure 4 and specified in Annex C.
Figure 4 — Low contrast e-SFR test chart
The e-SFR measurement includes the capture of a digital image of the test chart and analysis of the
contents of the image file by a software program. This software can be accessed from www.iso.
org/12233. The SFR algorithm is defined in normative Annex D. A diagram depicting the key steps of the
SFR algorithm is shown in Figure 5.
The algorithm can automatically compute the e-SFR, using image data from a user-defined rectangular
region of the image which represents a near-vertically, or near-horizontally oriented dark to light or
light to dark edge. The algorithm will be described assuming a near-vertical edge. To measure near-
horizontally, the selected edge image data are rotated 90° before performing the calculation. Note that
a near vertical edge is used to measure a horizontal e-SFR, since the e-SFR is a measure of the image
transition across the edge, rather than along it. Likewise, a near horizontal edge is used to measure the
vertical e-SFR.
6.2 Methodology
6.2.1 Selection of the edge region of interest (ROI)
The user selects the region containing the slightly slanted edge. If the image is coloured, a luminance
record is created before the SFR calculation is performed. The result is a two-dimensional matrix of data
of values (n lines, m pixels). See item A in Figure 5.
6.2.2 Transformation into effective exposure
The image code values shall then be linearized by inverting the opto-electronic conversion function
(OECF) of the camera. The OECF shall be measured as specified in ISO 14524. Each pixel value in the ROI
is now transformed into an equivalent target reflectance value. See item B in Figure 5.
6.2.3 Estimation of the location of the edge
This is done in two steps. See items C1 and C2 in Figure 5.
6.2.3.1 Initial estimation of edge location (offset) and slope
6.2.3.2 Final estimation of edge location (offset) and slope
— Compute one-dimensional derivative
For each line of pixels perpendicular to the edge, the data are multiplied with a Hamming window vector
of the same length (m). For each line of pixels in the resulting array, the derivative of the linearized
image data are computed using a [−1/2, +1/2] finite impulse response (FIR) filter. The result is an array
which is the same size as the input ROI.
Figure 5 — Flowchart of e-SFR measurement algorithm
10 © ISO 2014 – All rights reserved

Figure 6 — Parts of element C in Figure 5
— Compute location of the edge for each line of data
The one-dimensional centroid of this derivative matrix is calculated line by line, to determine the
position of the edge on each line. The result is a vector of centroid locations (1, n).
— Estimate slope and location of the edge
A linear best-line fit to the centroid locations as a function of line number is then calculated. That is, from
Formula (1)
ym=+xb (1)
where y is the set of centroid location and x is the set of line location (1, n), compute the best-fit values for
the slope, m, and offset, b. Error messages shall be reported if any centroid is within two pixels of either
side of the input image edges, or if the edge does not contain at least 20 % modulation.
6.2.3.3 Final estimation of edge location
For each line of pixels perpendicular to the edge, the location of the centroid of the line is computed from
Formula (2),
ym=+xb (2)
11 1
where y represents the vector of centroid location computed as illustrated in Figure 6. This results in
a vector of y’ values.
The transformed image data are multiplied with a Hamming window vector of the same length (n). In
this case, the Hamming window function is centred at y’ for each line. For each line of pixels multiplied
with thus-centred Hamming window array, the derivative of the image data are computed using a [−1/2,
+1/2] finite impulse response (FIR) filter. The result is an array which is the same size as the input ROI.
6.2.3.4 Computation of final location of the edge for each line of data
The one-dimensional centroid of this derivative matrix is calculated line by line, to determine the
position of the edge on each line. The result is a vector of centroid locations (1, n).
6.2.
...

SLOVENSKI STANDARD
01-julij-2014
1DGRPHãþD
SIST ISO 12233:2011
)RWRJUDILMD(OHNWURQVNRSULND]RYDQMHPLUXMRþLKVOLN5HVROXFLMDORþOMLYRVWLQ
IUHNYHQþQLRG]LYLYSURVWRUX
Photography - Electronic still picture imaging - Resolution and spatial frequency
responses
Photographie - Imagerie des prises de vues électroniques - Résolution et réponses en
fréquence spatiale
Ta slovenski standard je istoveten z: ISO 12233:2014
ICS:
37.040.10 Fotografska oprema. Photographic equipment.
Projektorji Projectors
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

INTERNATIONAL ISO
STANDARD 12233
Second edition
2014-02-15
Photography — Electronic still picture
imaging — Resolution and spatial
frequency responses
Photographie — Imagerie des prises de vues électroniques —
Résolution et réponses en fréquence spatiale
Reference number
©
ISO 2014
© ISO 2014
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 2014 – All rights reserved

Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Test conditions . 5
4.1 Test chart illumination . 5
4.2 Camera framing and lens focal length setting . 5
4.3 Camera focusing . 5
4.4 Camera settings . 5
4.5 White balance. 6
4.6 Luminance and colour measurements . 6
4.7 Gamma correction . 6
5 Visual resolution measurement . 6
5.1 General . 6
5.2 Test chart. 7
5.3 Rules of judgement for visual observation . 8
6 Edge-based spatial frequency response (e-SFR) . 9
6.1 General . 9
6.2 Methodology . 9
7 Sine-based spatial frequency response (s-SFR) measurement .12
8 Presentation of results .13
8.1 General .13
8.2 Resolution .13
8.3 Spatial frequency response (SFR) .14
Annex A (informative) CIPA resolution test chart .17
Annex B (informative) Visual resolution measurement software .23
Annex C (informative) Low contrast edge SFR test chart with OECF patches .28
Annex D (normative) Edge spatial frequency response (e-SFR) algorithm .30
Annex E (normative) Sine wave star test chart .33
Annex F (normative) Sine wave Spatial Frequency Response (s-SFR) analysis algorithm .35
Annex G (informative) Colour-filtered resolution measurements .40
Annex H (informative) Units and summary metrics .42
Annex I (informative) Original test chart defined in ISO 12233:2000 .45
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 ISO/TC 42, Photography.
This second edition cancels and replaces the first edition (ISO 12233:2000), which has been technically
revised.
iv © ISO 2014 – All rights reserved

Introduction
0.1 Purpose
The spatial resolution capability is an important attribute of an electronic still-picture camera. Resolution
measurement standards allow users to compare and verify spatial resolution measurements. This
International Standard defines terminology, test charts, and test methods for performing resolution
measurements for analog and digital electronic still-picture cameras.
0.2 Technical background
For consumer digital cameras, the term resolution is often incorrectly interpreted as the number of
addressable photoelements. While there are existing protocols for determining camera pixel counts,
these should not be confused with the interpretation of resolution as addressed in this International
Standard. Qualitatively, resolution is the ability of a camera to optically capture finely spaced detail, and
is usually reported as a single valued metric. Spatial frequency response (SFR) is a multi-valued metric
that measures contrast loss as a function of spatial frequency. Generally, contrast decreases as a function
of spatial frequency to a level where detail is no longer visually resolved. This limiting frequency value
is the resolution of the camera. A camera’s resolution and its SFR are determined by a number of factors.
These include, but are not limited to, the performance of the camera lens, the number of addressable
photoelements in the optical imaging device, and the electrical circuits in the camera, which can include
image compression and gamma correction functions.
While resolution and SFR are related metrics, their difference lies in their comprehensiveness and
utility. As articulated in this International Standard, resolution is a single frequency parameter that
indicates whether the output signal contains a minimum threshold of detail information for visual
detection. In other words, resolution is the highest spatial frequency that a candidate camera can
usefully capture under cited conditions. It can be very valuable for rapid manufacturing testing, quality
control monitoring, or for providing a simple metric that can be easily understood by end users. The
algorithm used to determine resolution has been tested with visual experiments using human observers
and correlates well with their estimation of high frequency detail loss.
SFR is a numerical description of how contrast is changed by a camera as a function of the spatial
frequencies that describe the contrast. It is very beneficial for engineering, diagnostic, and image
evaluation purposes and serves as an umbrella function from which such metrics as sharpness and
acutance are derived. Often, practitioners will select the spatial frequency associated with a specified
SFR level as a modified non-visual resolution value.
In a departure from the first edition of this International Standard, two SFR measurements are described.
Additionally, the first SFR metrology method, edge-based spatial frequency response, is identical to
that described in the first edition, except that a lower contrast edge is used for the test chart. Regions
of interest (ROI) near slanted vertical and horizontal edges are digitized and used to compute the SFR
levels. The use of a slanted edge allows the edge gradient to be measured at many phases relative to the
image sensor photoelements and to yield a phase-averaged SFR response.
A second sine wave-based SFR metrology technique is introduced in this edition. Using a sine wave
modulated target in a polar format (e.g. Siemens star), it is intended to provide an SFR response that
is more resilient to ill-behaved spatial frequency signatures introduced by the image content driven
processing of consumer digital cameras. In this sense, it is intended to enable easier interpretation of
SFR levels from such camera sources. Comparing the results of the edge-based SFR and the sine based
SFR might indicate the extent to which nonlinear processing is used.
The first step in determining visual resolution or SFR is to capture an image of a suitable test chart with
the camera under test. The test chart should include features of sufficiently fine detail and frequency
content such as edges, lines, square waves, or sine wave patterns. The test chart defined in this
International Standard has been designed specifically to evaluate electronic still-picture cameras. It has
not necessarily been designed to evaluate other electronic imaging equipment such as input scanners,
CRT displays, hard-copy printers, or electro-photographic copiers, nor individual components of an
electronic still-picture camera, such as the lens.
Some of the measurements described in this International Standard are performed using digital analysis
techniques. They are also applicable with the analogue outputs of the camera by digitizing the analogue
signals if there is adequate digitizing equipment.
0.3 Methods for measuring SFR and resolution — selection rationale and guidance
This section is intended to provide more detailed rationale and guidance for the selection of the different
resolution metrology methods presented in this International Standard. While resolution metrology of
analog optical systems, by way of spatial frequency response, is well established and largely consistent
between methodologies (e.g. sine waves, lines, edges), metrology data for such systems is normally
captured under well controlled conditions where the required data linearity and spatial isotropy
assumptions hold. Generally, it is not safe to assume these conditions for files from many digital
cameras, even under laboratory capture environments. Exposure and image content dependent image
processing of the digital image file before it is provided as a finished file to the user prevents this. This
processing yields different SFR responses depending on the features in the scene or in the case of this
International Standard, the target. For instance, in-camera edge detection algorithms might specifically
operate on edge features and selectively enhance or blur them based on complex nonlinear decision
rules. Depending on the intent, these algorithms might also be tuned differently for repetitive scene
features such as those found in sine waves or bar pattern targets. Even for constrained camera settings
recommended in this International Standard, these nonlinear operators can yield differing SFR results
depending on the target feature set. Naturally, this causes confusion on which targets to use, either
alone or in combination. Guidelines for selection are offered below.
Edges are common features in naturally occurring scenes. They also tend to act as visual acuity cues
by which image quality is judged and imaging artefacts are manifested. This logic prescribed their use
for SFR metrology in the past and current editions of this International Standard. It is also why edge
features are prone to image processing in many consumer digital cameras: they are visually important.
All other imaging conditions being equal, camera SFRs using different target contrast edge features can
be significantly different, especially with respect to their morphology. This is largely due to nonlinear
image processing operators and would not occur for strictly linear imaging systems. To moderate this
behaviour, a lower contrast slanted edge feature (Figure C.1) was chosen to replace the higher contrast
version of the first edition. This feature choice still allows for acuity-amenable SFR results beyond the
half-sampling frequency and helps prevent nonlinear data clipping that can occur with high contrast
target features. It is also a more reliable rendering of visually important contrast levels in naturally
occurring scenes.
Sine wave features have long been the choice for directly calculating SFR of analogue imaging systems
and they are intuitively satisfying. They have been introduced into this edition of this International
Standard based on experiences from the edge-based approach. Because sine waves transition more
slowly than edges, they are not prone to being identified as edges in embedded camera processors.
As such, the ambiguity that image processing imposes on the SFR can be largely avoided by their use.
Alternatively, if the image processing is influenced by the absence of sharp features, more aggressive
processing might be used by the camera. A sine wave starburst test pattern (Figure 6) is adopted in
this edition. With the appropriate analysis software, a sine wave-based SFR can be calculated up to the
half-sampling frequency. For the same reasons stated above, the sine wave-based target is also of low
contrast and consistent with that of the edge-based version. An added benefit of the target’s design over
other sine targets is its compactness and bi-directional features.
All experience suggests that there is no single SFR for today’s digital cameras. Even under the strict
capture constraints suggested in this International Standard, the allowable feature sets that most
digital cameras offer prevent such unique characterization. Confusion can be reduced through complete
documentation of the capture conditions and camera setting for which the SFR was calculated. It has been
suggested that comparing edge- and sine wave-based SFR results under the same capture conditions
could be a good tool in assessing the contribution of spatial image processing in digital cameras.
Finally, at times, a full SFR characterization is simply not required, such as in end of line camera assembly
testing. Alternately, SFR might be an intimidating obstacle to those not trained in its utility. For those
in need of a simple and intuitive space domain approach to resolution using repeating line patterns, a
visual resolution metric is also provided in this second edition of this International Standard.
vi © ISO 2014 – All rights reserved

With such a variety of methods available for measuring resolution, there are bound to be differences
in measured resolution results. To benchmark the likely variations, the committee has published the
results of a pilot study using all of the proposed techniques and how they relate to each other. These
results are provided in Reference [20].
INTERNATIONAL STANDARD ISO 12233:2014(E)
Photography — Electronic still picture imaging —
Resolution and spatial frequency responses
1 Scope
This International Standard specifies methods for measuring the resolution and the SFR of electronic
still-picture cameras. It is applicable to the measurement of both monochrome and colour cameras
which output digital data or analog video signals.
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 14524, Photography — Electronic still-picture cameras — Methods for measuring opto-electronic
conversion functions (OECFs)
3 Terms and definitions
For the purposes of this document, the following terms and definitions apply.
3.1
addressable photoelements
number of active photoelements in an image sensor
Note 1 to entry: This equals the product of the number of active photoelement lines and the number of active
photoelements per line.
3.2
aliasing
output image artefacts that occur in a sampled imaging system due to insufficient sampling
Note 1 to entry: These artefacts usually manifest themselves as moiré patterns in repetitive image features or as
jagged stair-stepping at edge transitions.
3.3
cycles per millimetre
cy/mm
spatial frequency unit defined as the number of spatial periods per millimetre
3.4
edge spread function
ESF
normalized spatial signal distribution in the linearized output of an imaging system resulting from
imaging a theoretical infinitely sharp edge
3.5
effectively spectrally neutral
having spectral characteristics which result in a specific imaging system producing the same output as
for a spectrally neutral object
3.6
electronic still-picture camera
camera incorporating an image sensor that outputs an analog or digital signal representing a still picture
Note 1 to entry: This camera may also record or store an analog or digital signal representing a still picture on a
removable media, such as a memory card or magnetic disc.
3.7
gamma correction
signal processing operation that changes the relative signal levels
Note 1 to entry: Gamma correction is performed in part to correct for the nonlinear light output versus signal
input characteristics of the display. The relationship between the light input level and the output signal level,
called the camera opto-electronic conversion function (OECF), provides the gamma correction curve shape for an
image capture device.
Note 2 to entry: The gamma correction is usually an algorithm, lookup table, or circuit which operates separately
on each colour component of an image.
3.8
horizontal resolution
resolution value measured in the longer image dimension, corresponding to the horizontal direction for
a “landscape” image orientation, typically using a vertical or near vertical oriented test-chart feature
3.9
image aspect ratio
ratio of the image width to the image height
3.10
image compression
process that alters the way digital image data is encoded to reduce the size of an image file
3.11
image sensor
electronic device that converts incident electromagnetic radiation into an electronic signal
EXAMPLE Charge coupled device (CCD) array, complementary metal-oxide semiconductor (CMOS) array.
3.12
line pairs per millimetre
lp/mm
spatial frequency unit defined as the number of equal width black and white line pairs per millimetre
3.13
line spread function
LSF
normalized spatial signal distribution in the linearized output of an imaging system resulting from
imaging a theoretical infinitely thin line
3.14
line widths per picture height
LW/PH
spatial frequency unit for specifying the width of a feature on a test chart relative to the height of the
active area of the chart
Note 1 to entry: the value in LW/PH indicates the total number of lines of the same width which can be placed edge
to edge within the height of a test target or within the vertical field of view of a camera.
Note 2 to entry: This unit is used whatever the orientation of the “feature” (e.g. line). Specifically, it applies to
horizontal, vertical, and diagonal lines.
EXAMPLE If the height of the active area of the chart equals 20 cm, a black line of 1 000 LW/PH has a width
equal to 20/1 000 cm.
2 © ISO 2014 – All rights reserved

3.15
linearized
digital signal conversion performed to invert the camera opto-electronic conversion function (OECF) to
focal plane exposure or scene luminance
3.16
lines per millimetre
lines/mm
spatial frequency unit defined as the number of equal width black and white lines per millimetre
Note 1 to entry: One line pair per millimetre (lp/mm) is equal to 2 lines/mm.
3.17
modulation
normalized amplitude of signal levels
Note 1 to entry: This is the difference between the minimum and maximum signal levels divided by the average
signal level.
3.18
modulation transfer function
MTF
modulus of the optical transfer function
Note 1 to entry: For the MTF to have significance, it is necessary that the imaging system be operating in an
isoplanatic region and in its linear range. Because most electronic still-picture cameras provide spatial colour
sampling and nonlinear processing, a meaningful MTF of the camera can only be approximated through the SFR.
See ISO 15529:2010.
3.19
normalized spatial frequency
spatial frequency unit for specifying resolution characteristics of an imaging system in terms of cycles
per pixel rather than in cycles/millimetre or any other unit of length
3.20
optical transfer function
OTF
two-dimensional Fourier transform of the imaging system’s point spread function
Note 1 to entry: For the OTF to have significance, it is necessary that the imaging system be operating in an
isoplanatic region and in its linear range. The OTF is a complex function whose modulus has unity value at zero
spatial frequency. (See ISO 9334). Because most electronic still-picture cameras provide spatial colour sampling
and nonlinear processing, a meaningful OTF of the camera can only be approximated through the SFR.
3.21
point spread function
PSF
normalized spatial signal distribution in the linearized output of an imaging system resulting from
imaging a theoretical infinitely small point source
3.22
resolution
measure of the ability of a camera system, or a component of a camera system, to depict picture detail
3.23
sampled imaging system
imaging system or device which generates an image signal by sampling an image at an array of discrete
points, or along a set of discrete lines, rather than a continuum of points
Note 1 to entry: The sampling at each point is done using a finite-size sampling aperture or area.
3.24
spatial frequency response
SFR
relative amplitude response of an imaging system as a function of input spatial frequency
Note 1 to entry: The SFR is normally represented by a curve of the output response to an input sinusoidal spatial
luminance distribution of unit amplitude, over a range of spatial frequencies. The SFR is divided by its value at the
spatial frequency of 0 as normalization to yield a value of 1,0 at a spatial frequency of 0.
3.24.1
edge-based spatial frequency response
e-SFR
measured amplitude response of an imaging system to a slanted-edge input, as defined in Clause 6
3.24.2
sine wave-based spatial frequency response
s-SFR
measured amplitude response of an imaging system to a range of sine wave inputs, as defined in Clause 7
3.25
spectrally neutral
exhibiting reflective or transmissive characteristics which are constant over the wavelength range of
interest
3.26
test chart
arrangement of test patterns designed to test particular aspects of an imaging system
3.27
test pattern
specified arrangement of spectral reflectance or transmittance characteristics used in measuring an
image quality attribute
3.27.1
bi-tonal pattern
pattern that is spectrally neutral or effectively spectrally neutral, and consists exclusively of two
reflectance or transmittance values in a prescribed spatial arrangement
Note 1 to entry: Bi-tonal patterns are typically used to measure resolution by visual resolution method.
3.27.2
hyperbolic wedge test pattern
bi-tonal pattern that varies continuously and linearly with spatial frequency
Note 1 to entry: A bi-tonal hyperbolic wedge test pattern is used to measure resolution by the visual resolution
method in this International Standard.
3.28
vertical resolution
resolution value measured in the shorter image dimension, corresponding to the vertical direction for a
“landscape” image orientation, typically using a horizontal or near horizontal oriented test-chart feature
3.29
visual resolution
spatial frequency at which all of the individual black and white lines of a test pattern frequency can no
longer be distinguished by a human observer
Note 1 to entry: This presumes the features are reproduced on a display or print.
4 © ISO 2014 – All rights reserved

4 Test conditions
4.1 Test chart illumination
The luminance of the test chart shall be sufficient to provide an acceptable camera output signal level.
The test chart shall be uniformly illuminated as shown in Figure 1, so that the illuminance at the
chart is within ±10 % of the illuminance in the centre of the chart at any position within the chart.
The illumination sources should be baffled to prevent direct illumination of the camera lens by the
illumination sources. The area surrounding the test chart should be of low reflectance, to minimize
flare light. The chart should be shielded from any reflected light. The illuminated test chart shall be
effectively spectrally neutral within the visible wavelengths.
Figure 1 — Test chart illumination method
4.2 Camera framing and lens focal length setting
The camera shall be positioned to properly frame the test target. The vertical framing arrows are used
to adjust the magnification and the horizontal arrows are used to centre the target horizontally. The
tips of the centre vertical black framing arrows should be fully visible and the tips of the centre white
framing arrows should not be visible. The target shall be oriented so that the horizontal edge of the
chart is approximately parallel to the horizontal camera frame line. The approximate distance between
the camera and the test chart should be reported along with the measurement results.
4.3 Camera focusing
The camera focus should be set either by using the camera autofocusing system, or by performing a series
of image captures at varying focus settings, and selecting the focus setting that provides the highest
average modulation level at a spatial frequency approximately 1/4 the camera Nyquist frequency. (In the
case of a colour camera, the Nyquist frequency is of the conceptual monochrome image sensor without
colour filter array). Auto focus accuracy is often limited and this limitation might have an impact on the
results.
4.4 Camera settings
The camera lens aperture (if adjustable) and the exposure time should be adjusted to provide a near
maximum signal level from the white test target areas. The settings shall not result in signal clipping in
either the white or black areas of the test chart, or regions of edge transitions.
Electronic still-picture cameras might include image compression, to reduce the size of the image files
and allow more images to be stored. The use of image compression can significantly affect resolution
measurements. Some cameras have switches that allow the camera to operate in various compression
or resolution modes. The values of all camera settings that might affect the results of the measurement,
including lens focal length, aperture and image quality (i.e. recording pixel number or compression)
mode (if adjustable) shall be reported along with the measurement results.
Multiple SFR measurements can be reported for different camera settings, including a setting that uses
the maximum lens aperture size (minimum f-number) and maximum camera gain.
4.5 White balance
For a colour camera, the camera white balance should be adjusted, if possible, to provide proper white
balance [equal red, green, and blue (RGB) signal levels] for the illumination light source, as specified in
ISO 14524.
4.6 Luminance and colour measurements
Resolution measurements are normally performed on the camera luminance signal. For colour cameras
that do not provide a luminance output signal, a luminance signal should be formed from an appropriate
combination of the colour records, rather than from a single channel such as green. The reader is
referred to ISO 12232 for the luminance signal calculation. Colour-filtered resolution measurements
can be performed as described in informative Annex G.
4.7 Gamma correction
The signal representing the image from an electronic still-picture camera will probably be a nonlinear
function of the scene luminance values. Since the SFR measurement is defined on a linearized output
signal, and such a nonlinear response can affect SFR values, the signal shall be linearized before the
data analysis is performed. Linearization is accomplished by applying the inverse of the camera OECF
to the output signal via a lookup table or appropriate equation. The measurement of the OECF shall be
as specified in ISO 14524, using the standard reflection camera OECF test chart or using an integrated
OECF/resolution chart.
5 Visual resolution measurement
5.1 General
The visual resolution is the maximum value of the spatial frequency in LW/PH within a test pattern that
is able to be visually distinguished. A black and white hyperbolic wedge is used as the test pattern
Because of aliasing artefacts in the high frequencies, actual resolution judgements can be ambiguous.
The objective visual resolution method cited herein using a hyperbolic wedge test pattern gives more
stable results by adopting the visual judgement rules described in 5.3 which have been used by a highly
skilled observer.
It can be measured analytically using computer analysis of captured images, as defined in Annex B. The
computer analysis method is intended to correlate with the subjective judgement of visual resolution
made by a skilled observer but is likely to yield a more consistent and objective result compared to
actual visual judgements. However, if there is a discrepancy between the results of the computer analysis
method and the judgement of a human observer, the judgement of the human observer takes priority.
6 © ISO 2014 – All rights reserved

5.2 Test chart
5.2.1 General
The preferred test chart for measuring the visual resolution is the CIPA resolution chart, which is shown
in Figure 2 and specified in informative Annex A.
The chart shown in Figure 2 is designed to measure cameras having a resolution of less than 2 500 LW/PH.
Nevertheless, it is possible to use the chart to measure the resolution of an electronic still camera having
a resolution greater than 2 500 LW/PH. This is accomplished by adjusting the camera to target distance,
or the focal length of the camera lens, so that the test chart active area fills only a portion of the vertical
image height of the camera. This fraction is then measured in the digital image, by dividing the number
of image lines in the camera image by the number of lines in the active chart area. The values of all test
chart features, in LW/PH, printed on the chart or specified in this International Standard, are multiplied
by this fraction, to obtain their correct values. For example, if the chart fills 1/2 of the vertical image
height of the camera, then the multiplication factor is equal to 2 and a feature labelled as 2 000 LW/PH
on the chart corresponds to 4 000 LW/PH using this chart framing.
NOTE Figure 2 includes an improved version of the test chart features originally defined in ISO 12233:2000.
5.2.2 Material
The test chart can be either a transparency that is rear illuminated, or a reflection test card that is front
illuminated. A reflection chart shall have an approximately Lambertian base material. A transparency
chart shall be rear illuminated by a diffuse source.
5.2.3 Size
The active height of reflection test charts should be no less than 20 cm. The active height of transparencies
shall be not less than 10 cm.
5.2.4 Test patterns
The test chart shall have bi-tonal patterns and should be spectrally neutral.
NOTE Bi-tonal test charts are easily manufactured and minimize the cost of producing the chart.
5.2.5 Test pattern modulation
For reflective charts, the ratio of the maximum chart reflectance, R , to the minimum chart
max
reflectance, R , for large test pattern areas should be not less than 40:1 and not greater than 80:1,
min
and shall be reported if it is outside this range. For transmissive charts, the ratio of the maximum chart
transmittance, T , to the minimum chart transmittance, T , for a large test pattern should be not
max min
less than 40:1 and not greater than 80:1, and shall be reported if it is outside this range. For a paper base
optical density of 0,10, these minimum and maximum numbers translate to optical densities of 1,7 and
2,0 respectively. Modulation ratios for the finer test chart features, relative to the ratio for large test
pattern areas, should preferably be reported by the chart manufacturer for reference.
5.2.6 Positional tolerance
The position of any test chart feature shall be reproduced with a tolerance of ±1/1 000 picture heights
(equivalent to ±1/10 % of the active test chart height). In addition, the width and duty cycle ratio of each
feature (white or black line) of the wedge pattern shall be reproduced with a tolerance of ±5 % of the
feature width.
Figure 2 — CIPA resolution test chart
5.3 Rules of judgement for visual observation
5.3.1 Rules of judgement
The viewer shall observe the following rules when judging the resolution value. These rules are intended
to achieve correct measurement value in the presence of unavoidable aliasing artefacts.
a) Beginning from the low frequency side, treat a spatial frequency as “Resolved” only when all lower
spatial frequencies are also resolved. The resolution limit is achieved at the line just before the first
occurrence of unresolved line features.
b) Treat a spatial frequency as “Not resolved” when the black and white lines appear to change polarity
or lines are blurred together to produce a reduced number of lines, compared to the number in the
test chart.
5.3.2 An example of a correct visual judgement
The boundary between the resolved and not resolved regions indicated by a dashed arrow is the
corresponding point to resolution value to be measured as shown in Figure 3.
Figure 3 — Correct application of the wedge feature interpretation
8 © ISO 2014 – All rights reserved

6 Edge-based spatial frequency response (e-SFR)
6.1 General
The edge-based spatial frequency response (e-SFR) of an electronic still-picture camera is measured
by analysing the camera data near a slanted low contrast neutral edge. The preferred test chart for
measuring e-SFR is shown in Figure 4 and specified in Annex C.
Figure 4 — Low contrast e-SFR test chart
The e-SFR measurement includes the capture of a digital image of the test chart and analysis of the
contents of the image file by a software program. This software can be accessed from www.iso.
org/12233. The SFR algorithm is defined in normative Annex D. A diagram depicting the key steps of the
SFR algorithm is shown in Figure 5.
The algorithm can automatically compute the e-SFR, using image data from a user-defined rectangular
region of the image which represents a near-vertically, or near-horizontally oriented dark to light or
light to dark edge. The algorithm will be described assuming a near-vertical edge. To measure near-
horizontally, the selected edge image data are rotated 90° before performing the calculation. Note that
a near vertical edge is used to measure a horizontal e-SFR, since the e-SFR is a measure of the image
transition across the edge, rather than along it. Likewise, a near horizontal edge is used to measure the
vertical e-SFR.
6.2 Methodology
6.2.1 Selection of the edge region of interest (ROI)
The user selects the region containing the slightly slanted edge. If the image is coloured, a luminance
record is created before the SFR calculation is performed. The result is a two-dimensional matrix of data
of values (n lines, m pixels). See item A in Figure 5.
6.2.2 Transformation into effective exposure
The image code values shall then be linearized by inverting the opto-electronic conversion function
(OECF) of the camera. The OECF shall be measured as specified in ISO 14524. Each pixel value in the ROI
is now transformed into an equivalent target reflectance value. See item B in Figure 5.
6.2.3 Estimation of the location of the edge
This is done in two steps. See items C1 and C2 in Figure 5.
6.2.3.1 Initial estimation of edge location (offset) and slope
6.2.3.2 Final estimation of edge location (offset) and slope
— Compute one-dimensional derivative
For each line of pixels perpendicular to the edge, the data are multiplied with a Hamming window vector
of the same length (m). For each line of pixels in the resulting array, the derivative of the linearized
image data are computed using a [−1/2, +1/2] finite impulse response (FIR) filter. The result is an array
which is the same size as the input ROI.
Figure 5 — Flowchart of e-SFR measurement algorithm
10 © ISO 2014 – All rights reserved

Figure 6 — Parts of element C in Figure 5
— Compute location of the edge for each line of data
The one-dimensional centroid of this derivative matrix is calculated line by line, to determine the
position of the edge on each line. The result is a vector of centroid locations (1, n).
— Estimate slope and location of the edge
A linear best-line fit to the centroid locations as a function of line number is then calculated. That is, from
Formula (1)
ym=+xb (1)
where y is the set of centroid location and x is the set of line location (1, n), compute the best-fit values for
the slope, m, and offset, b. E
...