SIST ISO 13322-1:2006

INTERNATIONAL ISO
STANDARD 13322-1
First edition
2004-12-01

Particle size analysis — Image analysis
methods —
Part 1:
Static image analysis methods
Analyse granulométrique — Méthodes par analyse d'images —
Partie 1: Méthodes par analyse d'images statiques




Reference number
ISO 13322-1:2004(E)
©
ISO 2004

---------------------- Page: 1 ----------------------
ISO 13322-1:2004(E)
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.


©  ISO 2004
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland

ii © ISO 2004 – All rights reserved

---------------------- Page: 2 ----------------------
ISO 13322-1:2004(E)
Contents Page
Foreword. iv
Introduction . v
1 Scope. 1
2 Normative references . 1
3 Terms, abbreviated terms, definitions, and symbols. 1
3.1 Terms, abbreviated terms and definitions. 1
3.2 Symbols . 3
4 Sample preparation demands for method description . 4
4.1 General recommendations. 4
4.2 Suggested preparation methods. 5
5 Image capture. 6
5.1 General. 6
5.2 Procedures . 7
5.3 Operating conditions for an image capture instrument. 7
6 Microscopy and image analysis . 8
6.1 General. 8
6.2 Size classes and magnification . 9
6.3 Counting procedure. 9
7 Calculation of the particle size results . 13
8 Test report. 13
Annex A (normative) Study on the sample size required for the estimation of mean particle
diameter . 15
Annex B (normative) Operating magnification. 34
Annex C (normative) Resolution and sizing limits for typical objective lenses . 35
Annex D (informative) Flow chart showing a typical image analysis method . 36
Annex E (informative) Statistical tests of mean and variance — Analysis of variance and multiple
comparisons. 37
Bibliography . 39

© ISO 2004 – All rights reserved iii

---------------------- Page: 3 ----------------------
ISO 13322-1:2004(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.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
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.
ISO 13322-1 was prepared by Technical Committee ISO/TC 24, Sieves, sieving and other sizing methods,
Subcommittee SC 4, Sizing by methods other than sieving.
ISO 13322 consists of the following parts, under the general title Particle size analysis — Image analysis
methods:
 Part 1: Static image analysis methods
 Part 2: Dynamic image analysis methods
iv © ISO 2004 – All rights reserved

---------------------- Page: 4 ----------------------
ISO 13322-1:2004(E)
Introduction
The purpose of this part of ISO 13322 is to give guidance for a measurement description and its validation
when determining particle size by image analysis.
Image analysis is a technique used in very different applications on image material with variations in material
properties. Hence, it is not relevant to describe an exact standard method for determination of particle size by
image analysis. The aim of this part of ISO 13322 is limited to give a standardized description of the technique
used and a standardized validation.
This part of ISO 13322 includes methods of calibration verification using a certified standard graticule as a
reference or by using certified standard particles. It is sensible to make some measurements on particles, or
other reference objects, of known size so that the likely systematic uncertainties introduced by the equipment
can be calculated.
This part of ISO 13322 gives a recommendation for a precise description of the distribution including the
number of analyzed particles and an analysis window to make sure that the obtained information is valid.
Measurement of particle-size distributions by microscopy methods is apparently simple, but because only a
small amount of sample is examined, considerable care has to be exercised in order to ensure that the
analysis is representative of the bulk sample. This can be demonstrated by splitting the original sample and
making measurements on three or more parts. Statistical analysis of the data, for example using the Student's
t-test, will reveal whether the samples are truly representative of the whole.
Errors introduced at all stages of the analysis from sub-division of the sample to generation of the final result
add to the total uncertainty of measurement and it is important to obtain estimates for the uncertainty arising
from each stage. Indications where this is required are given at the appropriate point in the method.
Because of the diverse range of equipment and sample preparation expertise available, it is not intended to
give a prescriptive procedure where use of individual methods does not jeopardize the validity of the data.
However, essential operations are identified to ensure that measurements made conform to this part of
ISO 13322 and are traceable.
© ISO 2004 – All rights reserved v

---------------------- Page: 5 ----------------------
INTERNATIONAL STANDARD ISO 13322-1:2004(E)

Particle size analysis — Image analysis methods —
Part 1:
Static image analysis methods
1 Scope
This part of ISO 13322 is applicable to the analysis of images for the purpose of determining particle size
distributions. The particles are appropriately dispersed and fixed on an optical or electron microscope sample
stage such as glass slides, stubs, filters, etc. Image analysis can recover particle images directly from
microscopes or from photomicrographs.
Even though automation of the analysis is possible, this technique is basically limited to narrow size
distributions of less than an order of magnitude. A standard deviation of 1,6 of a log-normal distribution
corresponds to a distribution of less than 10:1 in size. Such a narrow distribution requires that over 6 000
particles be measured in order to obtain a repeatable volume-mean diameter. If reliable values are required
for percentiles, e.g. D or other percentiles, at least 61 000 particles must be measured. This is described in

90
Annex A.
2 Normative references
The following referenced documents are indispensable for the application of this document. For dated
references, only the edition cited applies. For undated references, the latest edition of the referenced
document (including any amendments) applies.
ISO 9276-1, Representation of results of particle size analysis — Part 1: Graphical representation
ISO 9276-2, Representation of results of particle size analysis — Part 2: Calculations of average particle
sizes/diameters and moments from particle size distributions
3 Terms, abbreviated terms, definitions and symbols
3.1 Terms, abbreviated terms and definitions
For the purposes of this document, the following definitions apply.
3.1.1
view field
field which is viewed by a viewing device, e.g. optical microscope or electron scanning microscope
3.1.2
measurement frame
field in a view field in which particles are counted for image analysis
NOTE The set of measurement frames composes the total measurement field.
© ISO 2004 – All rights reserved 1

---------------------- Page: 6 ----------------------
ISO 13322-1:2004(E)
3.1.3
binary image
digitized image consisting of an array of pixels, each of which has a value of 0 or 1, whose values are
normally represented by dark and bright regions on the display screen or by the use of two distinct colours
3.1.4
edge finding
one of many edge detection methods used to detect transition between objects and background
3.1.5
Euler number
number of objects minus the number of holes inside the objects, which describes the connectedness of a
region, not its shape
NOTE A connected region is one in which all pairs of points can be connected by a curve lying entirely in the region.
If a complex two-dimensional object is considered to be a set of connected regions, where each one can have holes, the
Euler number for such an object is defined as the number of connected regions minus the number of holes. The number of
holes is one less than the connected regions in the set compliment of the object. It is important to report the Euler number
together with the connectivity applied, i.e., 4-connectivity or 8-connectivity.
3.1.6
Feret diameter
distance between two parallel tangents on opposite sides of the image of a particle
3.1.7
equivalent circular diameter
ecd
diameter of a circle having the same area as the projected image of the particle
NOTE It is also known as the Haywood Diameter.
3.1.8
grey image
image in which multiple grey level values are permitted for each pixel
3.1.9
image analysis
processing and data reduction operation which yields a numerical or logical result from an image
3.1.10
numerical aperture
NA
product of the refractive index of the object space and the sine of the semi-aperture of the cone of rays
entering the entrance pupil of the objective lens from the object point
3.1.11
pixel
picture element
individual sample in a digital image that has been formed by uniform sampling in both the horizontal and
vertical directions
3.1.12
segmentation
〈noun〉 part into which something can be divided; subdivision or section
3.1.13
segmentation
〈verb〉 act of dividing something into segments
2 © ISO 2004 – All rights reserved

---------------------- Page: 7 ----------------------
ISO 13322-1:2004(E)
3.1.14
threshold
grey level value which is set to discriminate objects of interest from background
3.2 Symbols
δ error
θ half-angle subtended by the particle at the objective lens
λ wavelength, expressed in micrometres
µ refractive index of the surrounding medium
ϕ shape factor
A projected area of particle i
i
d minimum feature length
H horizontal calibration factor
cal
K constant numerically determined by the confidence limit
N number of particles to be measured
n numbers of particles of size X
i i
P probability
P
probability that particle i exists in the measurement frame (also called Miles-Lantuejoul factor)
i
V vertical calibration factor
cal
V relative volume of particle i
i
X diameter of spherical particle i
A
X area equivalent diameter of particle i
Ai
X horizontal Feret diameter of object
F1
X vertical Feret diameter of object
F2
X dimension of particle i
i
X longest dimension of particle i, also called maximum Feret diameter
imax
X shortest dimension of particle i, also called minimum Feret diameter
imin
X lower limit of a class interval
LIL
X mean of X
mean i
X upper limit of a class interval
UIL
X horizontal dimension of object
1
© ISO 2004 – All rights reserved 3

---------------------- Page: 8 ----------------------
ISO 13322-1:2004(E)
X horizontal dimension, expressed in micrometres
1,m
X horizontal dimension, expressed in pixels
1,p
X vertical dimension of object
2
X vertical dimension, expressed in micrometres
2,m
X vertical dimension, expressed in pixels
2,p
Z horizontal side length of the rectangular measurement frame
1
Z vertical side length of the rectangular measurement frame
2
4 Sample preparation demands for method description
4.1 General recommendations
4.1.1 General
The following recommendations provide a sampling of standard microscopy practices.
NOTE See References [4], [5] and [10] for additional suggestions.
4.1.2 Sample subdivision
As only a small amount is needed to prepare a sample, the whole sample shall be subdivided in a manner that
ensures that the portion taken is representative of the whole.
The method used to subdivide the sample is likely to be dictated by the sample preparation method and will
be decided by the laboratory performing the analysis.
Provided that the sample is well dispersed by the method and that there is no segregation of particles by size,
the choice of method is left to the expertise of the laboratory, since any specialized equipment required by a
particular method might not be available to all.
4.1.3 Touching particles
The number of particles touching each other should be minimized.
It is a prime requirement of the method that measurements shall be made on isolated particles. There should
be as few particles as possible touching each other. Touching particles measured as one particle without a
proper separation will introduce error.
4.1.4 Particle distribution
There should be an adequate distribution of particles on the sample support. The whole area of the
preparation should be examined to ascertain whether there is noticeable segregation of particles (by size).
Statistical comparison of the results on a frame-by-frame basis will test for uniform distribution of particles.
This procedure is detailed in Clause 7.
4.1.5 Sample preparation
Electron microscope samples should be coated with a thin layer of metal (e.g. Au, Au/Pd, Pt/Pd) to reduce
charging effects.
4 © ISO 2004 – All rights reserved

---------------------- Page: 9 ----------------------
ISO 13322-1:2004(E)
Samples should be examined as soon as possible after preparation, and should contain an expiration date.
The sample preparation method used should be fully described in the final particle size analysis report by
giving quantitative details of the nominal masses, volumes and compositions of particles and products used at
each stage of the preparation procedure.
4.1.6 Number of particles to be counted
The number of particles measured should be determined based on the particle-size distribution and the
desired confidence limits. Assuming the particles are log-normally distributed, the required number (N) of
particles with a given error ()δ and a given confidence limit is estimated in accordance with Equation (1):
log N = −2 log δ + K (1)
where K is numerically determined by the confidence limit, particle distribution and other parameters; see
References [1] and [2].
NOTE See Annex A for detailed information.
4.2 Suggested preparation methods
Several methods can be investigated for preparing samples for measurement. The following methods may be
used. They are based on the assumption that a representative sample be used to give an adequate dispersion
of the particles and a sharply contrasted image.
4.2.1 Camphor-naphthalene (C-N) method
This method uses a eutectic mixture of 60 % mass fraction camphor and 40 % mass fraction naphthalene that
melts at 32 °C and sublimates rapidly in a vacuum. To prepare the sample, a 1 g sample of the particles to be
counted is kneaded by hand inside a plastic bag with the requisite amount of the C-N eutectic mixture. When
the particles sample is fully disaggregated and well dispersed in the C-N by the heat of the hand, the plastic
bag is cooled to solidify the resulting mixture. Small lumps of this solid mixture are then transferred to a
microscope slide resting on a warm plate. The sample, when melted, is flattened under a cover-slip that is
afterwards removed to allow the C-N eutectic to sublimate under vacuum.
This technique was found to give good dispersion of irregular quartz particles and has the advantages that the
particles are viewed in air, which results in a good contrast in the refractive index, and that the slides do not
age. However, tests with glass beads have been unsuccessful, as the particles segregate on the slide, do not
stick well and tend to roll off, making the method unusable; see Reference [3].
4.2.2 Paste-dilution method
A sample of about 1 g of particles is mixed with a viscous liquid (gelatine, sucrose or glycerol in water,
collodion in amyl acetate) on a watch glass with a spatula to give a thick paste, thus ensuring mechanical
disaggregation and dispersion. A sample of the paste is then taken with the point of a spatula and diluted in
the same viscous liquid to a concentration such that, after homogenization, one drop of the resulting
suspension, flattened under a cover-slip, will give the required number of particles on a microscope slide, that
is, about 20 particles per view frame. Depending on the choice of liquid, the slides can have only a temporary
life or might be able to be stored indefinitely. Using glycerol, this method has been successful for glass beads.
It gives a good uniform dispersion and a reasonably contrasted image. The use of a cover-slip aids resolution
with high-magnification objectives. However, the slides tend to dry out within an hour or so and repeat counts
with the same slide are not possible; see Reference [4].
© ISO 2004 – All rights reserved 5

---------------------- Page: 10 ----------------------
ISO 13322-1:2004(E)
4.2.3 Filtration methods
4.2.3.1 Powder or dry suspensions
A 1 g sample of particles is suspended in a suitable liquid and dispersed. A given volume of this suspension is
then filtered to dryness on a suitable membrane. The concentration of the suspension and the membrane area
of filtration are such that the particles are deposited in the required concentration for counting (about
20 particles per measurement frame). After air-drying, the membrane is cut into small sections which are
attached by their edges to a microscope slide using an acetone-resistant glue (e.g. cyanoacrylate or “super-
glue”). The gluing is to prevent the membrane from shrinking. The slide is then put in a closed container on a
support over a free surface of liquid acetone, whose vapour renders the membrane transparent for viewing
and particle-counting. The method has the advantage that the particles are viewed in air giving a good
contrast in refractive index. Tests indicate that to avoid the membrane re-opacifying, it is preferable to perform
the exposure to acetone very slowly over several hours; see Reference [5].
4.2.3.2 Liquid suspensions
A known volume of suspension, typically 100 ml, is vacuum-filtered, as described below, through a membrane
of compatible material and known pore size, typically 0,8 µm cellulose nitrate for mineral oils. Particles larger
than the pore size should appear well scattered across the membrane with little or no overlap. If the number of
particles is too great and overlapping is excessive, the test should be repeated with a smaller known volume
of suspension. Conversely, if the number of particles is too few, a greater volume of known amount should be
1)
used. The vacuum arrangement, for example a Millipore filtration system, consists of a membrane holder
attached to an open flask, with a vacuum pump attached below the filter holder. A separate spray container
with an integral filter attachment, typically 0,45 µm, is used with a compatible solvent to wash down the sides
of the open flask to ensure that all particles are collected on the membrane for analysis, and to remove the
liquid from the suspension, leaving a reasonably dry membrane for examination. The membrane should be
examined as soon as possible; if there is a delay, it can be inserted between two pre-cleaned microscope
slides. Appropriate glue for making the membrane transparent may be used; see Reference [6].
4.2.4 Dry deposition method
Particles may be prepared for counting by dry deposition onto a slide covered with double-sided transparent
adhesive tape. Care shall be taken that all the particles in a given sample effectively stick on the slide, so as
to ensure that there is no selective capture of particles by size. A microscope slide is positioned in the bottom
of a vacuum chamber having a volume of about 1 l. A conical metal plug is fitted in the top of the chamber and
the particles to be analyzed are placed in a groove all around the plug. When the vacuum is released by lifting
the plug, the particles are sucked as a cloud into the chamber and fall on the slide. Adhesion on the slide may
be enhanced by using double-sided tape or a film of grease. This method also has the advantage that
particles are viewed in air, resulting in a good contrast in refractive index.
5 Image capture
5.1 General
Particle-size data can be influenced by specific parameters affecting the image formation process. It is
possible to distort the reported size, particularly of the smallest particles, by using inappropriate image-capture
conditions, e.g. magnification, illumination, etc. Distortion in the image might arise from a number of causes,
but its presence and effect on the image can be measured by selecting a recognizable object at a number of
points and orientations in a frame of view. It is important to note that the measurements made provide only
two-dimensional, X and Y, information. The imaging instrument should be set up and operated in accordance
with its manufacturer's recommendations considering the conditions given below.

1)
Millipore is an example of a suitable product available commercially. This information is given for the convenience of

users of this part of ISO 13322 and does not constitute an endorsement by ISO of this product.

6 © ISO 2004 – All rights reserved

---------------------- Page: 11 ----------------------
ISO 13322-1:2004(E)
5.2 Procedures
At each operating condition used for the analysis, carry out the following steps.
a) Select a recognizable object in the image.
b) Place the feature in the centre and at the corners of the field of view in turn and measure its horizontal
length X .
( )
1
c) Rotate the sample stage 90 degrees and repeat the measurements ( X ) .
2
d) Record the values of X and X with the final result.
1 2
e) Calibrate the imaging instrument prior to the examination of samples using a certified graticule or
equivalent.
f) If possible, mount the traceable calibration graticule together with the specimen in the imaging instrument.
g) Select the magnification in accordance with Annex B or Annex C and set the corresponding illumination
and imaging conditions.
h) Place the calibration grating in the field of view, select a suitable area and focus it.
i) Obtain the image to be analyzed and then capture it either digitally or by use of a suitable photographic
image.
j) Record a significant number of measurement frames for each sample by scanning the sample in a raster
pattern as indicated in Figure 1. Once this operation is started, no changes to the operating conditions
should be made.

Figure 1 — Sample raster pattern
k) At the end of measurement, place the calibration graticule in the field of view and check the calibration
once more. The comparison of two calibration images taken at the beginning and the end of the
examination will provide a measure of the variability in instrument magnification.
l) Report the calibration constants obtained before and after the analysis together with the precise details of
the microscope settings (working distance, spot size, electron microscope magnification, etc.).
5.3 Operating conditions for an image capture instrument
5.3.1 General
There are various imaging systems used for particle sizing. The setup for particle sizing using an electron or
optical microscope is briefly described in 5.3.2 and 5.3.3.
© ISO 2004 – All rights reserved 7

---------------------- Page: 12 ----------------------
ISO 13322-1:2004(E)
5.3.2 Operating conditions for an electron microscope
The following special conditions are required for measuring particle size using an electron microscope:
a) image contrast mode: used to adjust the desired peak signal level;
b) accelerating voltage: set according to the material to be measured;
c) specimen position: The sample working distance specified by the electron microscope
manufacturer for high resolution imaging should be selected. The
sample should be mounted flat on the specimen holder with the stage
tilt set to zero;
d) dynamic focus and tilt correction: both of these controls should be switched off.
e) Select the operating magnification after reference to Annex B. The total magnification is the product of
electron microscope magnification and any image analyzer transfer magnification.
5.3.3 Operating conditions for an optical microscope
For bright-field images in the light microscope which are commonly used for particle sizing, the minimum
feature length, d, expressed in micrometres, distinguishable in monochromatic light is given by Equation (2):
0,6λ
d = (2)
µ sinθ
where
λ is the wavelength, expressed in micrometres;
θ is the half angle, expressed in degrees, subtended by the particle at the objective lens;
µ is the refractive index of the surrounding medium.
The theoretical lower limit is approximately 0,2 µm, but the diffraction halo around the particle gives a gross
overestimate of size. Special attention should be given to range of particle size to be measured, then to the
measurem
...