SIST EN 16016-3:2012
(Main)Non destructive testing - Radiation Methods - Computed Tomography - Part 3: Operation and interpretation
Non destructive testing - Radiation Methods - Computed Tomography - Part 3: Operation and interpretation
This European Standard specifies an outline of the operation of a CT system, and the interpretation of the results in order to provide the user with technical information to select suitable parameters.
Zerstörungsfreie Prüfung - Durchstrahlungsverfahren - Computertomographie - Teil 3: Durchführung und Auswertung
Der vorliegende Teil der Normenreihe gibt dem Anwender einen Überblick über den Betrieb eines CT-Systems und die Auswertung der Ergebnisse. Sie stellt dem Anwender technische Informationen zur Auswahl geeigneter Parameter zur Verfügung.
Essais non destructifs - Moyens utilisant les rayonnements - Tomographie informatisée - Partie 3: Fonctionnement et interprétation
La présente partie fournit à l'utilisateur une présentation générale du fonctionnement d'un système TI et l'interprétation des résultats. Elle donne à l'utilisateur des informations techniques pour sélectionner les paramètres adaptés.
Neporušitveno preskušanje - Sevalna metoda - Računalniška tomografija - 3. del: Delovanje in razlaga
Ta evropski standard opisuje delovanje sistema računalniške tomografije (CT) in razlago rezultatov z namenom zagotavljanja tehničnih podatkov uporabniku za izbiro ustreznih parametrov.
General Information
Relations
Standards Content (Sample)
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.Zerstörungsfreie Prüfung - Durchstrahlungsverfahren - Computertomographie - Teil 3: Durchführung und AuswertungEssais non destructifs - Moyens utilisant les rayonnements - Tomographie informatisée - Partie 3: Fonctionnement et interprétationNon destructive testing - Radiation Methods - Computed Tomography - Part 3: Operation and interpretation19.100Neporušitveno preskušanjeNon-destructive testingICS:Ta slovenski standard je istoveten z:EN 16016-3:2011SIST EN 16016-3:2012en,fr,de01-februar-2012SIST EN 16016-3:2012SLOVENSKI
STANDARD
SIST EN 16016-3:2012
EUROPEAN STANDARD NORME EUROPÉENNE EUROPÄISCHE NORM
EN 16016-3
August 2011 ICS 19.100 English Version
Non destructive testing - Radiation methods - Computed Tomography - Part 3: Operation and interpretation
Essais non destructifs - Méthodes par rayonnements - Tomographie numérisée - Partie 3: Fonctionnement et interprétation
Zerstörungsfreie Prüfung - Durchstrahlungsverfahren - Computertomographie - Teil 3: Durchführung und Auswertung This European Standard was approved by CEN on 29 July 2011.
CEN members are bound to comply with the CEN/CENELEC Internal Regulations which stipulate the conditions for giving this European Standard the status of a national standard without any alteration. Up-to-date lists and bibliographical references concerning such national standards may be obtained on application to the CEN-CENELEC Management Centre or to any CEN member.
This European Standard exists in three official versions (English, French, German). A version in any other language made by translation under the responsibility of a CEN member into its own language and notified to the CEN-CENELEC Management Centre has the same status as the official versions.
CEN members are the national standards bodies of Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland and United Kingdom.
EUROPEAN COMMITTEE FOR STANDARDIZATION
COMITÉ EUROPÉEN DE NORMALISATION EUROPÄISCHES KOMITEE FÜR NORMUNG
Management Centre:
Avenue Marnix 17,
B-1000 Brussels © 2011 CEN All rights of exploitation in any form and by any means reserved worldwide for CEN national Members. Ref. No. EN 16016-3:2011: ESIST EN 16016-3:2012
EN 16016-3:2011 (E) 2 Contents Page Foreword .3Introduction .41Scope .52Normative references .53Terms and definitions .54Operational procedure .54.1General .54.2CT system set-up .54.2.1General .54.2.2Geometry .54.2.3X-ray source .64.2.4Detector .64.3Reconstruction parameters .74.4Visualisation .74.5Analysis and interpretation of CT images .74.5.1General .74.5.2Feature testing/defect testing .74.5.3Dimensional testing .85Requirements for acceptable results. 105.1Image quality parameters . 105.1.1Contrast . 105.1.2Noise . 125.1.3Signal to noise ratio . 125.1.4Contrast to noise ratio . 125.1.5Spatial resolution . 135.2Suitability of testing. 155.3CT examination interpretation and acceptance criteria. 155.4Records and reports . 155.5Artefacts . 165.5.1General . 165.5.2Beam hardening artefacts . 165.5.3Edge artefacts . 175.5.4Scattered radiation . 185.5.5Instabilities . 185.5.6Ring artefacts . 185.5.7Centre of rotation error artefacts . 195.5.8Motion artefacts . 205.5.9Artefacts due to an insufficient number of projections . 215.5.10Cone beam artefacts . 21Annex A (informative)
Spatial resolution measurement using line pair gauges . 23A.1Line pair gauges . 23A.2Principle of measurement . 24Bibliography . 27 SIST EN 16016-3:2012
EN 16016-3:2011 (E) 3 Foreword This document (EN 16016-3:2011) has been prepared by Technical Committee CEN/TC 138 “Non-destructive testing”, the secretariat of which is held by AFNOR. This European Standard shall be given the status of a national standard, either by publication of an identical text or by endorsement, at the latest by February 2012, and conflicting national standards shall be withdrawn at the latest by February 2012. Attention is drawn to the possibility that some of the elements of this document may be the subject of patent rights. CEN [and/or CENELEC] shall not be held responsible for identifying any or all such patent rights. EN 16016 consists of the following parts: Non destructive testing Radiation methods Computed tomography Part 1: Terminology; Non destructive testing Radiation methods Computed tomography Part 2: Principle, equipment and samples; Non destructive testing Radiation methods Computed tomography Part 3: Operation and interpretation; Non destructive testing Radiation methods Computed tomography — Part 4: Qualification. According to the CEN/CENELEC Internal Regulations, the national standards organizations of the following countries are bound to implement this European Standard: Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland and the United Kingdom.
SIST EN 16016-3:2012
EN 16016-3:2011 (E) 4 Introduction This document gives guidelines for the general principles of X-ray computed tomography (CT) applicable to industrial imaging (in the context of this standard, industrial means non-medical applications); it also gives a consistent set of CT performance parameter definitions, including how these performance parameters relate to CT system specifications. This document deals with computed axial tomography and excludes other types of tomography such as translational tomography and tomosynthesis. SIST EN 16016-3:2012
EN 16016-3:2011 (E) 5 1 Scope This European Standard specifies an outline of the operation of a CT system, and the interpretation of the results in order to provide the user with technical information to select suitable parameters. 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. EN 16016-1:2011, Non destructive testing Radiation method Computed tomography Part 1: Terminology EN 16016-2:2011, Non destructive testing Radiation method Computed tomography Part 2: Principle, equipment and samples 3 Terms and definitions For the purposes of this document, the terms and definitions given in EN 16016-1:2011 apply. 4 Operational procedure 4.1 General For target-oriented CT inspection procedures, the test and measurement tasks are defined in advance with regard to the size and type of features/defects to be verified; for example, through the specification of appropriate acceptance levels and geometry deviations. In the following, the process steps of a CT application are described and information on its implementation provided.
4.2 CT system set-up 4.2.1 General The CT system set-up is oriented towards the requirements for the given task. The required spatial resolution (taking into account the tube focal spot size), contrast resolution, voxel size and the CT image quality can be derived from these requirements. The quality of the CT image is determined by different parameters, which under certain circumstances counteract each other. In the following, system parameters are described and information is provided on setting up a CT system for inspection. Due to the interactions of the different system parameters, it may be necessary to run through the set-up steps several times in order to acquire optimal data. The optimal energy is that which gives the best signal-to-noise ratio and not necessarily that which gives the clearest radiograph (the dependency of the detector efficiency on the energy is to be taken into account). However, in order to differentiate between materials of different chemical composition it may be necessary to adjust the accelerating voltage to maximise the difference in their linear attenuation coefficients. 4.2.2 Geometry The source-detector and source-object distances and thus also the beam angle used should be specified. In order to achieve high resolutions, the projection can be magnified onto the detector. The magnification is equal to the ratio of the source-detector distance to the source-object distance. Increasing source-detector SIST EN 16016-3:2012
EN 16016-3:2011 (E) 6 distance leads to a reduced intensity at the detector and thus to a reduced signal to noise ratio. Accordingly, this also applies when using detectors with improved detector resolution, which can result in a reduction of the signal-to-noise ratio due to the reduced intensity per pixel. In general, for this reason, minimisation of the source-object distance is to be preferred. In order to obtain high beam intensity at the detector, the source-detector distance should be selected so that it is as small as possible taking into account the required resolution so that the beam cone still fully illuminates the detector. In the case of 3D-CT, the (in general vertical) total cone beam angle measured parallel to the rotation axis should typically be less than 15°, but this is specimen dependant, in order to minimise reconstruction-determined (Feldkamp) distortions of the 3D model. In addition, these restrictions do not apply for the perpendicular (in general horizontal) beam angle. For a higher geometric magnification, the object must be positioned as near as possible to the source, taking into consideration the limit on sharpness imposed by focal spot size. The rotation of the object must take place at at least 180 °plus beam angle of the X-ray beam, whereby an improved data quality is the result of an increasing number of angular increments. For this reason, the object is typically turned through 360 °. Ideally, the number of angular increments should be at least sizematrix×2π where the matrix size is the number of voxels across the sample diameter or the largest dimension. For more information, refer to 5.5. In order to obtain as complete information as possible on the specimen, the requirement in general for a CT is that the object (or the interesting section of the object) is completely mapped in each projection on the detector. For large components that exceed the beam cone, a so-called measurement range extension is used. This measurement range extension is accomplished by laterally displacing either the object or the detector, recording the projection data in sequential measurements, and finally concatenating (joining) them. Under certain circumstances, it is also possible to only scan a part of the object (region-of-interest CT), which may lead to a restricted data quality in the form of so-called truncations. A possible deviation of the recording geometry (offset between the projected axis of rotation and the centre line of the image) must be corrected for in order to obtain a reconstruction which is as precise as possible. This can be achieved by careful realignment of the system or be corrected using software. 4.2.3 X-ray source At the X-ray source, the maximum beam energy and tube current are to be set such that sufficient penetration of the test object and tube power with a sufficiently small focal spot are ensured. The required voltage is determined by the maximum path length, in the material to be X-rayed in accordance with
EN 16016-2:2011, 8.2. For the best measurement results, an attenuation ratio of approx. 1:10 should be used. That is the grey level through the sample should be about 10% of the white level (both measured with respect to the dark level). The optimal range can be achieved through the use of prefilters. It should be noted that every prefilter reduces the intensity. Prefilters have the additional advantage of reducing beam hardening, though further improvements can be made with software correction. 4.2.4 Detector The following detector settings need to be set appropriately for the sample being scanned: Exposure time (Frame rate); Number of integrations per projection; Digitisation gain and offset; Binning. If necessary, corrections for offset, gain and bad pixels (which may depend on X-ray settings) should be applied. The individual CT projection is determined by the detector properties: its geometric resolution, its sensitivity, dynamics and noise. The gain and exposure time can be adjusted together with the radiation intensity of the source so that the maximum digitised intensity does not exceed 90 % of the saturation level. SIST EN 16016-3:2012
EN 16016-3:2011 (E) 7 To reduce scattered radiation, a thin filter, grid or lamellae can be used directly in front of the detector (post-filtering). The ideal acquisition time is dependent on the required quality of the CT image and it is often limited by the time available for inspection. 4.3 Reconstruction parameters The volumetric region to be reconstructed, the size of the CT image (in terms of voxels) as well as its dynamic range (which should take into account the detector dynamic range) shall be specified. In order to achieve sufficient CT image quality, settings for the reconstruction algorithm or corrections should be optimised. The volumetric region is defined by the number of voxels along the X, Y & Z axes. 4.4 Visualisation Using volume visualisation, the CT image can be presented as a 3D object. Individual grey values can be assigned any colour and opacity values to highlight or hide materials with different X-ray densities. Zooming, scrolling, setting contrast, brightness, colour and lighting facilitate an optimal presentation of the CT image. In addition, it is possible to place user-defined sectional planes through the object in order to examine the internal structure, or to interactively visualise the CT image, for example by rotating and moving it as a 3D object. Image processing can be applied to CT images to improve feature recognition. It may not be possible to load the whole CT image at full resolution into memory at once. 4.5 Analysis and interpretation of CT images 4.5.1 General Typical features for inspection are pores, cavities, cracks, inclusions, impurities or inhomogeneous material distributions. Typical measurement tasks are obtaining dimensional properties (such as length or wall thickness) or calculating object morphology. 4.5.2 Feature testing/defect testing Features in the sample generally give rise to changes in CT grey level within the CT image. Analysis of CT images is performed by qualified personnel using software. A suitable contrast range or an automatic or manual calibration is used. The position, CT grey value and dimensions of features can be determined. Several tools are available for this, including manual ones or automatic tools such as strobe lines or gauges that engage at grey value thresholds or edges. For examining the structure and location of assembled components, a qualitative comparison of CT images without determination of the dimensions can suffice. For an automatic determination using visualisation software tools (for example fault analyses), a calibration via the specification of a grey value range is, in general, required for the sample material to be measured. The specification of the grey levels can be done manually using histograms or in an interactive manner. The detectability of features depends on the size of the feature relative to the geometric resolution and the contrast resolution compared with the contrast difference of the feature from the base material, as well as the quality of the image (signal to noise ratio, etc.) and any possible interference effects between adjacent voxels (partial volume effect). For the detectability of singular pores, cavities or cracks, their minimum extent should typically be 2 to 3 times the demagnified pixel size (at the position of the sample). SIST EN 16016-3:2012
EN 16016-3:2011 (E) 8 4.5.3 Dimensional testing 4.5.3.1 General Depending on the task, there are various methods currently in use for determining geometric features. Point-to-point distances can be manually determined in the CT slices or more complex features can be extracted with the help of analysis software.
The measurement of the geometric properties of an object using CT is an indirect procedure, in which the dimensional measurement takes place in or is derived from CT images. For this reason, in order to facilitate precise measurements, an accurate knowledge of two important variables is necessary: the precise image scale or voxel size and the boundary surface of two materials, for example the component surface (material-to-air transition), which can be determined via a CT grey value threshold in the CT image. 4.5.3.2 Determination of precise image scale The precise image scale or voxel size must be determined through the measurement of a suitable calibration standard (together with the measurement object and directly before/after the object inspection) or using a reference geometry at the object. For this, the voxel size or magnification M specified by the CT system is compared with the actual available and precisely determined (using the reference body/geometry) voxel size or magnification M*. Thus, for example, the exact voxel size can be determined with high precision via measurements without the disturbing influence of other variables (for example, the precise position of the component surface (grey value threshold) in the CT image) for the centre distances of a test piece (e.g. dumbbell, see Figure 1). In this procedure, it must be taken into account that the CT grey values of the test item can, under certain circumstances, be influenced by the accompanying reference bodies (for example, through changes to the contrast ratios, interferences and artefacts). Using the actual voxel sizes determined in this way, the visualisation software can be correspondingly scaled/corrected as regards the voxel size specified by the system.
Figure 1 Reference objects (dumbbell) 4.5.3.3 Threshold value determination In order to be able to carry out dimensional measurements, the component surface or material contact surface must be determined in the CT image. The component surface is generally derived from the transition from SIST EN 16016-3:2012
EN 16016-3:2011 (E) 9 solid object to surrounding air. The boundary surface is defined via a threshold value and is thus dependent on the materials and the X-ray settings. This threshold may be specified globally for the entire CT image as an average grey value of, for example, the material and air. This is sometimes known as the “Iso50 threshold”. A global threshold value or calibration using the Iso50 method is suitable for many measurement tasks on objects made from homogeneous materials. A global threshold is not suitable for objects made from several materials. In these cases, different thresholds should be used according to the materials either side of the boundary. Even in the case of objects made from homogeneous materials, beam hardening, scattering and other artefacts can result in local dimming or lightening in the CT image which would distort the measurement results. The grey value threshold, for example, for surfaces in the inside of the component thus frequently differs from that for surfaces on the outside of the component. The threshold can, if necessary, be determined locally from grey levels either side of the boundary. A determination of the overall component surface via locally determined threshold values, while more time consuming, is more tolerant towards contrast variations and artefact influences.
4.5.3.4 Adjustment of geometrically primitive bodies In addition to simple point-to-point operations (see 4.5.3), methods from coordinate measurement technology, such as reference geometry adjustment may be employed. In this connection, so-called geometric primitive bodies or reference elements (for example planes, cylinders, spheres or similar) are fitted, using software, to object contours of interest in the correspondingly calibrated data. At the reference elements, geometric features (for example, diameter, distances, angles, etc) are determined directly or by combining reference elements. By fitting to the typically several thousand measurement points of the corresponding data, there is thus, due to the statistic averaging and reduction of the user influence, an often much higher precision than via the manual distance measurement of two points. 4.5.3.5 Generation of geometric data So-called triangular models can be extracted from the voxels and calibrated grey value threshold. These models represent the calibrated threshold value-Iso-surface, i.e. the material surface in the form of linked triangles. The triangular model contains – as part of the extraction process precision (see below) – the geometry information on the object surface. It consists of only two types of information: the so-called vertices and the information as to which vertices belong to a triangle. The vertices are 3D points, which lie on the threshold value-Iso-surface. The quantity of all vertices is also designated a point cloud. It is initially the linking information, i.e. the information as to which three vertices in each case form a surface triangle, which defines the course of the object surface. A standard format for data exchange is the so-called STL file format (ASCII or binary and dimension-less). Alternatively, the point cloud (vertices without triangle information) can be exported, whereby in general important information on adjacent vertices is lost and if necessary must subsequently be reproduced.
The geometric quality of the generated point cloud or triangular model depends entirely on the number and position of the vertices. Since only triangles are assumed between the vertices in the triangular model, detailed surface structures, contained in the voxels, between the individual vertices can, under certain circumstances, not be represented and are thus lost. The extraction of a point cloud or a triangular model from the voxels corresponds to a scanning of the object surface. For further processing, the amount of data must in general be reduced. The quality or geometric precision of the triangular model depends on how good the triangle can reproduce the actual course of the material surface (e.g. chord error). With special software applications, a low-loss reduction of the number of triangles is aimed for. For each of these process steps, the involved losses are to be taken into account for the subsequent steps. Due to the special process conditions, the quality of the dimensional data is to be checked for plausibility and significance. SIST EN 16016-3:2012
EN 16016-3:2011 (E) 10 4.5.3.6 Nominal-actual comparison A dimensional CT application is the comparison of the recorded part (actual object) with the nominal geometry from the CAD (or other sources). After registering the CT coordinate system with the CAD coordinate system, there is the option, via the appropriate software, of comparing the geometric deviation of the CT-measured actual component with the CAD specification of the nominal geometry. The nominal-actual comparison can be carried out between the exported STL model or the point cloud and the CAD data or by directly comparing the voxels with the CAD data without previous STL or point cloud extraction. 4.5.3.7 Further processing of geometric data CT can also be used for the non-destructive determination of geometric data (reverse engineering), e.g. of prototype parts or adjacent components.
CAD models are not based on triangular models, rather on geometric primitives (e.g. cylinder) and so-called free-form surfaces. For this reason, a further processing of the geometric data in CAD systems, for example, the engineering of the surface determined from the voxels in a CAD-established model, is required. With the appropriate software, triangular models can be transferred to CAD-compatible elements (so-called reverse engineering), whereby CT-examined objects, i.e. real geometries, can again be incorporated into the CAD process. 5 Requirements for acceptable results 5.1 Image quality parameters 5.1.1 Contrast The quantity that is reconstructed in X-ray CT imaging is the linear attenuation coefficient, µ. It is measured in units of inverse length (e.g. mm-1) and is approximatel
...
SLOVENSKI STANDARD
oSIST prEN 16016-3:2010
01-marec-2010
1HSRUXãLWYHQHSUHLVNDYH6HYDOQDPHWRGD5DþXQDOQLãNDWRPRJUDILMDGHO
'HORYDQMHLQUD]ODJD
Non destructive testing - Radiation Methods - Computed Tomography - Part 3: Operation
and interpretation
Zerstörungsfreie Prüfung - Durchstrahlungsverfahren - Computertomographie - Teil 3:
Durchführung und Auswertung
Essais non destructifs - Moyens utilisant les rayonnements - Tomographie informatisée -
Partie 3: Fonctionnement et interprétation
Ta slovenski standard je istoveten z: prEN 16016-3
ICS:
19.100 Neporušitveno preskušanje Non-destructive testing
oSIST prEN 16016-3:2010 en,fr,de
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.
---------------------- Page: 1 ----------------------
oSIST prEN 16016-3:2010
---------------------- Page: 2 ----------------------
oSIST prEN 16016-3:2010
EUROPEAN STANDARD
DRAFT
prEN 16016-3
NORME EUROPÉENNE
EUROPÄISCHE NORM
November 2009
ICS 19.100
English Version
Non destructive testing - Radiation Methods - Computed
Tomography - Part 3: Operation and interpretation
Essais non destructifs - Moyens utilisant les rayonnements Zerstörungsfreie Prüfung - Durchstrahlungsverfahren -
- Tomographie informatisée - Partie 3: Fonctionnement et Computertomographie - Teil 3: Durchführung und
interprétation Auswertung
This draft European Standard is submitted to CEN members for enquiry. It has been drawn up by the Technical Committee CEN/TC 138.
If this draft becomes a European Standard, CEN members are bound to comply with the CEN/CENELEC Internal Regulations which
stipulate the conditions for giving this European Standard the status of a national standard without any alteration.
This draft European Standard was established by CEN in three official versions (English, French, German). A version in any other language
made by translation under the responsibility of a CEN member into its own language and notified to the CEN Management Centre has the
same status as the official versions.
CEN members are the national standards bodies of Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Estonia, Finland,
France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal,
Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland and United Kingdom.
Recipients of this draft are invited to submit, with their comments, notification of any relevant patent rights of which they are aware and to
provide supporting documentation.
Warning : This document is not a European Standard. It is distributed for review and comments. It is subject to change without notice and
shall not be referred to as a European Standard.
EUROPEAN COMMITTEE FOR STANDARDIZATION
COMITÉ EUROPÉEN DE NORMALISATION
EUROPÄISCHES KOMITEE FÜR NORMUNG
Management Centre: Avenue Marnix 17, B-1000 Brussels
© 2009 CEN All rights of exploitation in any form and by any means reserved Ref. No. prEN 16016-3:2009: E
worldwide for CEN national Members.
---------------------- Page: 3 ----------------------
oSIST prEN 16016-3:2010
prEN 16016-3:2009 (E)
Contents Page
Foreword .3
Introduction .4
1 Scope .5
2 Normative references .5
3 Terms and definitions .5
4 Operational procedure .5
4.1 CT system set-up .5
4.1.1 General .5
4.1.2 Geometry .6
4.1.3 X-ray source .6
4.1.4 Detector .6
4.2 Reconstruction parameters .7
4.3 Visualisation .7
4.4 Analysis and interpretation of CT images .7
4.4.1 Feature testing/defect testing .7
4.4.2 Dimensional testing .8
5 Requirements for acceptable results. 11
5.1 Image quality parameters . 11
5.1.1 Contrast . 11
5.1.2 Noise . 13
5.1.3 Signal to noise ratio . 13
5.1.4 Contrast to noise ratio . 14
5.1.5 Spatial resolution . 14
5.2 Suitability of testing. 16
5.3 CT examination interpretation and acceptance criteria. 16
5.4 Records and reports . 17
5.5 Artefacts . 17
5.5.1 Beam hardening artefacts . 17
5.5.2 Edge artefacts . 18
5.5.3 Scattered radiation . 19
5.5.4 Instabilities . 19
5.5.5 Ring artefacts . 20
5.5.6 Centre of rotation error artefacts . 20
5.5.7 Motion artefacts . 21
5.5.8 Artefacts due to an insufficient number of projections . 22
5.5.9 Cone beam artefacts . 22
Annex A (informative) Spatial resolution measurement using line pair gauges . 24
A.1 Line pair gauges . 24
A.2 Principle of measurement . 25
2
---------------------- Page: 4 ----------------------
oSIST prEN 16016-3:2010
prEN 16016-3:2009 (E)
Foreword
This document (prEN 16016-3:2009) has been prepared by Technical Committee CEN/TC 138 “Non-
destructive testing”, the secretariat of which is held by AFNOR.
This document is currently submitted to the CEN Enquiry.
This standard consists of the following parts, under the general title, Radiation methods – Computed
tomography :
Part 1 : Terminology ;
Part 2 : Principle, equipment and samples ;
Part 3 : Operation and interpretation ;
Part 4 : Qualification.
3
---------------------- Page: 5 ----------------------
oSIST prEN 16016-3:2010
prEN 16016-3:2009 (E)
Introduction
This document gives guidelines for the general principles of X-ray computed tomography (CT) applicable to
industrial imaging (in the context of this standard, industrial means non-medical applications); it also gives a
consistent set of CT performance parameter definitions, including how these performance parameters relate
to CT system specifications. This document deals with computed axial tomography and excludes other types
of tomography such as translational tomography and tomosynthesis.
4
---------------------- Page: 6 ----------------------
oSIST prEN 16016-3:2010
prEN 16016-3:2009 (E)
1 Scope
This part gives the user an outline of the operation of a CT system, and the interpretation of the results. It will
provide the user with technical information to select suitable parameters.
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.
prEN 16016-1, Non Destructive Testing – Radiation method – Computed tomography - Terminology.
prEN 16016-2, Non Destructive Testing – Radiation method – Computed tomography - Operation and
interpretation.
3 Terms and definitions
For the purposes of this document, the terms and definitions given in prEN 16016-1 apply.
4 Operational procedure
For target-oriented CT inspection procedures, the test and measurement tasks are defined in advance with
regard to the size and type of features/defects to be verified; for example, through the specification of
appropriate acceptance levels and geometry deviations. In the following, the process steps of a CT application
are described and information on its implementation provided.
4.1 CT system set-up
4.1.1 General
The CT system set-up is oriented towards the requirements for the given task. The required spatial resolution
(taking into account the tube focal spot size), contrast resolution, voxel size and the CT image quality can be
derived from these requirements. The quality of the CT image is determined by different parameters, which
under certain circumstances counteract each other.
In the following, system parameters are described and information is provided on setting up a CT system for
inspection. Due to the interactions of the different system parameters, it may be necessary to run through the
set-up steps several times in order to acquire optimal data.
The optimal energy is that which gives the best signal-to-noise ratio and not necessarily that which gives the
clearest radiograph (the dependency of the detector efficiency on the energy is to be taken into account).
However, in order to differentiate between materials of different chemical composition it may be necessary to
adjust the accelerating voltage to maximise the difference in their linear attenuation coefficients.
5
---------------------- Page: 7 ----------------------
oSIST prEN 16016-3:2010
prEN 16016-3:2009 (E)
4.1.2 Geometry
The source-detector and source-object distances and thus also the beam angle used should be specified. In
order to achieve high resolutions, the projection can be magnified onto the detector. The magnification is
equal to the ratio of the source-detector distance to the source-object distance. Increasing source-detector
distance leads to a reduced intensity at the detector and thus to a reduced signal to noise ratio. Accordingly,
this also applies when using detectors with improved detector resolution which can result in a reduction of the
signal-to-noise ratio due to the reduced intensity per pixel. In general, for this reason, minimisation of the
source-object distance is to be preferred.
In order to obtain a high beam intensity at the detector, the source-detector distance should be selected so
that it is as small as possible taking into account the required resolution so that the beam cone still fully
illuminates the detector. In the case of 3D-CT, the (in general vertical) cone beam angle parallel to the rotation
axis should typically be less than 15°, but this is specimen dependant, in order to prevent reconstruction-
determined (Feldkamp) distortions of the 3D model. In addition, these restrictions do not apply for the
perpendicular (in general horizontal) beam angle. For a higher geometric magnification, the object must be
positioned as near as possible to the source taking into consideration the geometric blurriness (focal spot
size). The rotation of the object must take place at least at 180 ° plus beam angle of the X-ray beam, whereby
an improved data quality is the result of an increasing number of angular increments. For this reason, the
object is typically turned through 360 °. Ideally the number of angular increments should be at least
π 2× matrix size where the matrix size is the number of voxels across the sample diameter or the largest
dimension. For more information, refer to clause 5.5.
In order to obtain as complete information as possible on the specimen, the requirement in general for a CT is
that the object (or the interesting section of the object) is completely mapped in each projection on the
detector. For large components that exceed the beam cone, a so-called measurement range extension is
used. This measurement range extension is accomplished by laterally displacing either the object or the
detector, recording the projection data in sequential measurements, and finally concatenating (joining) them.
Under certain circumstances, it is also possible to only scan a part of the object (region-of-interest CT), which
may lead to a restricted data quality in the form of so-called truncations.
A possible deviation of the recording geometry (offset between the projected axis of rotation and the centre
line of the image) must be corrected for in order to obtain a reconstruction which is as precise as possible,
This can be achieved by careful realignment the system or be corrected using software.
4.1.3 X-ray source
At the X-ray source, the maximum beam energy and tube current are to be set such that sufficient penetration
of the test object and tube power with a sufficiently small focal spot are ensured. The required voltage is
determined by the maximum path length, in the material to be X-rayed according to 8.2 of prEN 16016-2. For
the best measurement results, an attenuation ratio of approx. 1:10 should be used. That is the grey level
through the sample should be about 10% of the white level (both measured with respect to the dark level).
The optimal range can be achieved through the use of prefilters. It should be noted that every prefilter reduces
the intensity. Prefilters have the additional advantage of reducing beam hardening, though further
improvements can be made with software correction .
4.1.4 Detector
The following detector settings need to be set appropriately for the sample being scanned:
Exposure time (Frame rate);
Number of integrations per projection;
Digitisation gain and offset;
Binning.
6
---------------------- Page: 8 ----------------------
oSIST prEN 16016-3:2010
prEN 16016-3:2009 (E)
If necessary, corrections for offset, gain and bad pixels (which may depend on X-ray settings) should be
applied.
The individual CT projection is determined by the detector properties: its geometric resolution, its sensitivity,
dynamics and noise. The gain and exposure time can be adjusted together with the radiation intensity of the
source so that the maximum digitised intensity does not exceed 90% of the saturation level.
To reduce scattered radiation, a thin filter, grid or lamellae can be used directly in front of the detector (post-
filtering).
The ideal acquisition time is dependent on the required quality of the CT image and it is often limited by the
time available for inspection.
4.2 Reconstruction parameters
The volumetric region to be reconstructed, the size of the CT image (in terms of voxels) as well as its dynamic
range (which should take into account the detector dynamic range) shall be specified. In order to achieve
sufficient CT image quality, settings for the reconstruction algorithm or corrections should be optimised.
The volumetric region is defined by the number of voxels along the X, Y & Z.
4.3 Visualisation
Using volume visualisation, the CT image can be presented as a 3D object. Individual grey values can be
assigned any colour and opacity values to highlight or hide materials with different X-ray densities. Zooming,
scrolling, setting contrast, brightness, colour and lighting facilitate an optimal presentation of the CT image. In
addition, it is possible to place user-defined sectional planes through the object in order to examine the
internal structure, or to interactively visualise the CT image, for example by rotating and moving it as a 3D
object. Image processing can be applied to CT images to improve feature recognition.
It may not be possible to load all of the CT image at full resolution into memory at once.
4.4 Analysis and interpretation of CT images
Typical features for inspection are pores, cavities, cracks, inclusions, impurities or inhomogeneous material
distributions.
Typical measurement tasks are obtaining dimensional properties (such as length or wall thickness) or
calculating object morphology.
4.4.1 Feature testing/defect testing
Features in the sample generally give rise to changes in CT grey level within the CT image. Analysis of CT
images is performed by qualified personnel using software. A suitable contrast range or an automatic or
manual calibration is used. The position, CT number and dimensions of features can be determined. Several
tools are available for this, including manual ones or automatic tools such as strobe lines or gauges that
engage at grey value thresholds or edges. For examining the structure and location of assembled
components, a qualitative comparison of CT images without determination of the dimensions can suffice.
For an automatic determination using visualisation software tools (for example fault analyses), a calibration via
the specification of a grey value range is, in general, required for the sample material to be measured. The
specification of the grey levels can be done manually using histograms or in an interactive manner.
7
---------------------- Page: 9 ----------------------
oSIST prEN 16016-3:2010
prEN 16016-3:2009 (E)
The detectability of features depends on the size of the feature relative to the geometric resolution and the
contrast resolution compared with the contrast difference of the feature from the base material, as well as the
quality of the image (signal to noise ratio, etc.) and any possible interference effects between adjacent voxels
(partial volume effect). For the detectability of singular pores, cavities or cracks, their minimum extent should
typically be 2 to 3 times the demagnified pixel size (at the position of the sample).
4.4.2 Dimensional testing
Depending on the task, there are various methods currently in use for determining geometric features. Point-
to-point distances can be manually determined in the CT slices or more complex features can be extracted
with the help of analysis software.
The measurement of the geometric properties of an object using the CT is an indirect procedure, in which the
dimensional measurement takes place in or is derived from CT images. For this reason, in order to facilitate
precise measurements, an accurate knowledge of two important variables is necessary:
the precise image scale or voxel size and
the boundary surface of two materials, for example the component surface (material-to-air transition),
which can be determined via a CT grey value threshold in the CT image.
4.4.2.1 Determination of precise image scale
The precise image scale or voxel size must be determined through the measurement of a suitable calibration
standard (together with the measurement object and directly before/after the object inspection) or using a
reference geometry at the object. For this, the voxel size or magnification M specified by the CT system is
compared with the actual available and precisely determined (using the reference body/geometry) voxel size
or magnification M*. Thus, for example, the exact voxel size can be determined with high precision via
measurements without the disturbing influence of other variables (for example, the precise position of the
component surface (grey value threshold) in the CT image) for the centre distances of a test piece (e.g.
dumbbell, see Figure 1). In this procedure, it must be taken into account that the CT grey values of the test
item can, under certain circumstances, be influenced by the accompanying reference bodies (for example,
through changes to the contrast ratios, interferences and artefacts). Using the actual voxel sizes determined in
this way, the visualisation software can be correspondingly scaled/corrected as regards the voxel size
specified by the system.
8
---------------------- Page: 10 ----------------------
oSIST prEN 16016-3:2010
prEN 16016-3:2009 (E)
Figure 1 – Reference objects (dumbbell)
4.4.2.2 Threshold value determination
In order to be able to carry out dimensional measurements, the component surface or material contact surface
must be determined in the CT image. The component surface is generally derived from the transition from
solid object to surrounding air. The boundary surface is defined via a threshold value and is thus dependent
on the materials and the recording parameters (in particular contrast) (edge finding problem in the CT image).
A consistent method consists in specifying, globally for the entire data record, the surface via an averaged
grey value threshold (Iso50 threshold: Average of the determined, global grey values of, for example, material
and air). A global threshold value or calibration using the Iso50 method is suitable for objects made from a
homogeneous material and supplies already acceptable measured values for many measuring tasks.
The global definition of the grey value threshold incorrectly describes the component surface for objects made
from several materials, as here in principal different threshold values must be taken into account. The
component surface is also usually incorrectly described, due to a global grey value threshold, for
homogeneous objects with complex geometries, for example, cavities. Artefacts like, for example, edge
artefacts, beam hardening or scattered radiation can result in local dimming or lightening in the CT image,
which would distort the measurement results. The grey value threshold, for example, for surfaces in the inside
of the component thus frequently differs from that for surfaces on the outside of the component. In the case of
spatially limited problems, the Iso50 threshold value can, if necessary, be determined via calibration on the
respective edge, whereby a local threshold value can be manually defined and the corresponding surface can
frequently be described more precisely than with the global threshold value.
A determination of the overall component surface via the respective locally determined threshold values is in
contrast more time consuming but also more tolerant towards contrast variations and artefact influences. In
this connection, via the corresponding software tools, the edges can be analyzed locally in the CT image and
the grey value threshold, which defines the component surface/contact surface, is in each case locally
determined.
9
---------------------- Page: 11 ----------------------
oSIST prEN 16016-3:2010
prEN 16016-3:2009 (E)
4.4.2.3 Adjustment of geometrically primitive bodies
In addition to simple point-to-point operations (see 4.4.2), methods from coordinate measurement technology,
such as reference geometry adjustment may be employed. In this connection, so-called geometric primitive
bodies or reference elements (for example planes, cylinders, spheres or similar) are fitted, using software, to
object contours of interest in the correspondingly calibrated data. At the reference elements, geometric
features (for example, diameter, distances, angles, etc) are determined directly or by combining reference
elements. By fitting to the typically several thousand measurement points of the corresponding data, there is
thus, due to the statistic averaging and reduction of the user influence, an often much higher precision than
via the manual distance measurement of two points.
4.4.2.4 Generation of geometric data
So-called triangular models can be extracted from the voxels and calibrated grey value threshold. These
models represent the calibrated threshold value-Iso-surface, i.e. the material surface in the form of linked
triangles. The triangular model contains – as part of the extraction process precision (see below) – the
geometry information on the object surface. It consists of only two types of information: the so-called vertices
and the information as to which vertices belong to a triangle. The vertices are 3D points, which lie on the
threshold value-Iso-surface. The quantity of all vertices is also designated a point cloud. It is initially the linking
information, i.e. the information as to which three vertices in each case form a surface triangle, which defines
the course of the object surface.
A standard format for data exchange is the so-called STL file format (ASCII or binary and dimension-less).
Alternatively, the point cloud (vertices without triangle information) can be exported, whereby in general
important information on adjacent vertices is lost and if necessary must subsequently be reproduced.
The geometric quality of the generated point cloud or triangular model depends entirely on the number and
position of the vertices. Because only triangles are assumed between the vertices in the triangular model,
detailed surface structures, contained in the voxels, between the individual vertices can, under certain
circumstances, not be represented and are thus lost.
The extraction of a point cloud or a triangular model from the voxels corresponds to a scanning of the object
surface. For further processing, the amount of data must in general be reduced. The quality or geometric
precision of the triangular model depends on how good the triangle can reproduce the actual course of the
material surface (e.g. chord error). With special software applications, a low-loss reduction of the number of
triangles is aimed for.
For each of these process steps, the involved losses are to be taken into account for the subsequent steps.
Due to the special process conditions, the quality of the dimensional data is to be checked for plausibility and
significance.
4.4.2.5 Nominal-actual comparison
A dimensional CT application is the comparison of the recorded part (actual object) with the nominal geometry
from the CAD (or other sources). After registering the CT coordinate system with the CAD coordinate system,
there is the option, via the appropriate software, of comparing the geometric deviation of the CT-measured
actual component with the CAD specification of the nominal geometry. The nominal-actual comparison can be
carried out between the exported STL model or the point cloud and the CAD data or by directly comparing the
voxels with the CAD data without previous STL or point cloud extraction.
10
---------------------- Page: 12 ----------------------
oSIST prEN 16016-3:2010
prEN 16016-3:2009 (E)
4.4.2.6 Further processing of geometric data
CT can also be used for the non-destructive determination of geometric data (reverse engineering), e.g. of
prototype parts or adjacent components.
CAD models are not based on triangular models, rather on geometric primitives (e.g. cylinder) and so-called
free-form surfaces. For this reas
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.