DE102018105709A1 - Method for computer tomographic measurements of workpieces - Google Patents

Abstract

The invention relates to a method for examining a workpiece by means of computer tomography comprising at least the following steps: recording at least a first set of several radiographic images of the workpiece arranged on a turntable in a plurality of rotational positions (rotational steps) relative to a radiation source and a detector, reconstruction of a voxel volume, which has voxels with assigned voxel gray values, from the transmission images, wherein for recording a first set of transmission images, the distance between the workpiece and the source:
- Is set lower for at least some first rotational positions than for some second rotational positions, and
is set for all rotational positions so that a collision between the detector and the workpiece is avoided by taking into account the workpiece dimensions and the respective rotational position of the turntable together with the workpiece relative to the turntable and the source or the source and the detector relatively at least moved towards each other or be moved away from each other.

Description

The subject matter of an independent invention is a method for examining a workpiece and / or one or more regions (ROI region of interest) of a workpiece, preferably dimensional measurement of features on the workpiece, by means of computer tomography.

For the dimensional measurement of complex geometries (geometric features) different tactile, tactile-optical, optical or computer tomographic sensors (computed tomography or computed tomography sensors) are used. Preferably, these are operated in coordinate measuring machines (CMMs), sometimes also several sensors combined in one device (multi-sensor CMM).

Computed tomography for the dimensional measurement of workpieces is to be understood as meaning that a reconstruction of the volume information (voxel data or voxel amplitudes in the form of gray values, also called voxel volume) from the set of two-dimensional radiographs usually taken by a planar detector in several rotational positions of a workpiece Volume detected by the detector, wherein the voxel data is a measure of the local attenuation and thus the local attenuation coefficients, and at material boundaries by surface extraction methods from the voxel data measuring points or surface measurement points are generated. Measurements on the workpiece or dimensions of features or structures on the workpiece can be determined from these surface measuring points, that is to say dimensional measurements take place. The surface of the workpiece is represented for example by networking the surface points in the so-called STL (Standard Triangulation Language) format.

A computed tomography or a computed tomography (CT) sensor system generally consists of a flat detector, a radiation source, preferably X-ray source, and a mechanical axis of rotation (turntable) for rotating the workpiece to be measured in the beam cone of the detected by the detector part of the radiation source emitted radiation. In kinematic reversal, however, it is also possible to arrange the workpiece firmly and to rotate the detector and radiation source around the workpiece.

If the positioning of the object or workpiece to be measured, in particular together with the turntable, is always described below, the alternative is always to leave the object stationary and to position the X-ray detector and / or X-ray source accordingly.

The term mechanical axis of rotation serves merely to distinguish it from a mathematical axis of rotation, if a mechanical axis of rotation (also referred to as a turntable) always also allows a rotation about a mathematical axis of rotation. The use of the term rotary axis therefore refers to the mathematical axis of rotation associated with the mechanical axis of rotation, insofar as the context refers to a direction, and to the mechanical axis of rotation insofar as a device is meant. The term mechanical axis of rotation denotes no restriction on the implemented within the axis of rotation guide principle between fixed and rotatable part of the mechanical axis of rotation. Thus, both mechanically mounted, as well as air-bearing, or otherwise as hydraulically mounted, etc., means mechanical axes of rotation The mathematical axis of rotation often coincides with the physical axis of rotation of the turntable. However, both may differ if the turntable is moved during rotation. In this case, a resulting rotation of the workpiece located on the turntable about an axis of rotation, here referred to as mathematical axis of rotation, take place, which deviates from the physical axis of rotation of the turntable, as for example the EP2399237B1 describes.

In addition to area-wide detectors, line detectors are also used as detectors. These have only a single or a few (e.g., three to five) detector rows. For complete recording of radiographic information of a spatially extended workpiece workpiece and detector must be placed in several along the direction of the axis of rotation (mathematical axis of rotation) shifted positions, the displacement can also take place during the rotation (helical CT). Advantageous is the possible reduction of stray radiation artifacts. The increased time expenditure resulting from the shift is avoided by the use of area detectors. Nevertheless, the present invention or the teachings of the invention can also be implemented for line detectors. Instead of processing radiographic images, ie 2D images, the information recorded with the respective detector row (s) is processed and also referred to herein as radiographic images for the sake of simplicity.

For accurate measurement of objects or object details (ROIs - Region of Interest) computer tomography measurements are performed with high resolution. For this purpose, a detector with high Resolution, so needed a large number of pixels and / or a high magnification.

The magnification ratio is the ratio of the two distances radiation source (more precisely radiation emitting focus) to detector (Focus detector distance FDD) and radiation source to object (Focus object distance FOD). By relative displacement of the object on or together with the turntable in the direction of the radiation source (ie by reducing FOD), a higher magnification and thus a higher resolution can be achieved. This is limited by the possible collision of the object or workpiece with the radiation source in all or flat objects at least some rotational positions. Alternatively, FDD can be increased to increase the magnification and resolution. However, this results in a lower cone angle of the radiation involved in the imaging and thus a reduced beam power and larger dimensions of the computed tomography and the associated necessary and costly Strahlenschutzumhausung.

If only object details (ROIs) are to be measured accurately, but for the rest of the workpiece a lower resolution suffices, the so-called ROI-CT is used, as in the example of DE102013108152A1 described. The ROI-CT takes place with a high magnification, whereby only the respective object detail is imaged in all rotational positions on the detector. Some areas of the workpiece thus leave the recorded measurement volume in some rotational positions and are not imaged on the detector. In order to be able to carry out a reconstruction with high accuracy, the thereby missing radiographic information is otherwise procured. This can be done with the help of an additional, so-called overview CT, in which the entire workpiece is measured at a low magnification. Alternatively or additionally, however, the radiographic information can also be generated by means of forward projection on nominal data or nominal data (eg from a CAD model). This is also in the DE102013108152A1 described.

However, it is disadvantageous in the known methods that collisions with the radiation source can occur when rotating objects or workpieces that are flat, ie have significantly larger dimensions in two dimensions than in the third dimension, if the reproduction scale should be increased. Even a ROI-CT is no longer possible because the workpiece can not be completely rotated. In the alternative, a so-called axial CT or laminography measurements are then carried out, wherein the object is not or only slightly rotated and for a lateral movement of the object takes place relative to the radiation source and detector to record multiple radiographic images. However, the accuracy of these measurements is significantly lower compared to a CT or ROI-CT.

A first object of the present invention is therefore to avoid the problems of the prior art and in particular to allow a computed tomography measurement of increased accuracy, even if the object to be measured or workpiece in at least some rotational positions in adjusting the required for increased accuracy Image scale would collide with the radiation source.

In particular, an accurate ROI-CT for details (ROIs) on flat, oblate workpieces is to be made possible. The measurement resolution should be as high as possible.

To solve the invention provides that, depending on the rotational position of the workpiece to be measured, the distance between the workpiece and source is changed, so that for at least some, here referred to as first rotational positions rotational positions a larger magnification and thus a higher resolution and measurement accuracy is present than for other, here as second rotational positions designated rotational positions in which otherwise a collision between the workpiece and source would occur. The method is preferably used in a region of interest (ROI-CT) on flat, oblate workpieces, wherein the largest possible magnification is set for the highest possible number of rotational positions in order to produce a high resolution and measurement accuracy for the ROI.

• - für zumindest einige erste Drehstellungen geringer eingestellt wird, als für einige zweite Drehstellungen, wodurch sich vorzugsweise ein höherer Abbildungsmaßstab für die ersten Drehstellungen ergibt, und
• - für sämtliche Drehstellungen so eingestellt wird, dass eine Kollision zwischen Detektor und Werkstück vermieden wird,

indem unter Berücksichtigung der Werkstückabmessungen oder der erwarteten Werkstückabmessungen und der jeweiligen Drehstellung, der Drehtisch zusammen mit dem Werkstück und/oder das Werkstück relativ zum Drehtisch, und, die Quelle oder die Quelle und der Detektor, relativ zumindest aufeinander zubewegt oder voneinander weg bewegt werden.The invention provides for the solution a method for examining a workpiece and / or one or more regions (ROI region of interest) of a workpiece, preferably dimensional measurement of features on the workpiece, by means of computer tomography, comprising at least the following steps: recording at least one first set of several radiographic images of the workpiece arranged on a turntable in several rotational positions (rotational steps) relative to a radiation source (source) such as X-ray source and a detector such as X-ray detector, reconstruction of a voxel volume having voxels with assigned voxel gray values from the radiographic images and preferably determination of surface measurement points from the voxel volume by means of surface extraction methods, characterized by the fact that, to record the first set of radiographic images, the distance between the workpiece and the source:

• - is set lower for at least some first rotational positions, as for some second rotational positions, which preferably results in a higher magnification for the first rotational positions, and
• - is set for all rotational positions so that a collision between the detector and the workpiece is avoided,

by moving the turntable together with the workpiece and / or the workpiece relative to the turntable, and taking into account the workpiece dimensions or the expected workpiece dimensions and the respective rotational position, and the source or source and the detector are relatively at least moved towards or away from each other.

In particular, the invention is characterized in that the adjustment of the distance is such that in as many rotational positions, the distance is minimized, to avoid collisions a specified minimum distance between the workpiece outer contour and source is maintained.

Preferably, it is provided that the setting of the distance takes place in each case between two rotational positions or during the change of the rotational position.

In particular, the invention provides that the setting of the distance for all rotational positions is such that the specified minimum distance or another fixed distance is taken (continuous adjustment), or that the distance is set as low as possible for at least some first rotational positions and for at least some second rotational positions are set larger (gradual adjustment).

According to a particularly noteworthy proposal, it is provided that the workpiece dimensions or the expected workpiece dimensions are determined from the workpiece target data such as CAD data and / or a computed tomography measurement with a fixed distance and sufficient gap between workpiece and source to avoid collisions and reduced number of rotational positions (Fast -CT) and / or other measurement such as tactile, optical and / or computed tomography measurement.

It should also be emphasized that the distance or the position of the components source, turntable and detector relative to each other (SOUV vectors) is assigned separately for the reconstruction of each rotational position.

• - durch eine Übersichts-CT, also durch Aufnahme eines zweiten Satzes mehrerer Durchstrahlungsbilder des auf dem Drehtisch angeordneten Werkstücks in mehreren Drehstellungen (Drehschritte) relativ zur Strahlungsquelle (Quelle) wie Röntgenstrahlenquelle und dem Detektor wie Röntgendetektor, wobei der Abstand zwischen Werkstück und Quelle für alle Drehstellungen gleich eingestellt wird, und zwar so, dass in keiner der Drehstellungen eine Kollision zwischen Detektor und Werkstück auftritt, wobei der Abbildungsmaßstab geringer ist, als in den zumindest einige ersten Drehstellungen des ersten Satzes von Durchstrahlungsbildern und/oder
• - durch Vorwärtsprojektion an Soll- oder Nominaldaten wie CAD-Daten des Werkstücks.

Preferably, the invention provides that when recording the first set of transmission images, a region-of-interest CT (ROI-CT) is performed, so in at least one of the first rotational positions of the workpiece with reduced distance to the source, the workpiece is not complete is imaged on the detector and for the reconstruction the missing radiographic information is generated:

• - By an overview CT, ie by recording a second set of multiple radiographic images of the arranged on the turntable workpiece in several rotational positions (rotational steps) relative to the radiation source (source) as X-ray source and the detector such as X-ray detector, wherein the distance between the workpiece and source for all Adjusted rotational positions, and in such a way that no collision between detector and workpiece occurs in any of the rotational positions, wherein the magnification is lower than in the at least some first rotational positions of the first set of radiographic images and / or
• - by forward projection to nominal or nominal data such as CAD data of the workpiece.

It should be particularly emphasized that a plurality of first sets of transmission images, preferably of different region-of-interests, are recorded and a plurality of first sets of transmission images and / or voxel volumes reconstructed therefrom and / or surface measurement points determined from the respective voxel volumes are present in a common coordinate system and can be evaluated together.

In particular, the invention is characterized in that when taking the first or the first set of radiographic images at least some of the second rotational positions are taken, which are recorded when recording the second set of radiographic images at the same magnification.

The subject of an independent invention is a method for determining features on a workpiece, wherein the steps used in this case should be as automated as possible.

According to the prior art, several methods are known in which at least partially the steps to be performed in a typical measurement of a workpiece with a coordinate measuring machine, in particular a computer tomograph designed as a coordinate measuring machine, are automated.

In the DE 10 2017 100 594.3 The applicant, for example, optimized steps to adjust a measurement program such that certain target parameters such as measurement time, measurement precision, measurement uncertainty or structure resolution can be achieved on the basis of a simulation of the computer tomographic measurement affected here.

In the WO 2016/169 589 Methods are described for adapting a measuring program to existing tolerances of the features of the workpiece to be measured.

In the DE 10 2008 037 441 A1 The applicant describes methods in which the selection of the sensor which is most suitable for the particular measuring point to be measured is carried out automatically.

A disadvantage of the prior art, however, is that only partial solutions for individual steps of the measurement to be performed can be automated. At the first mentioned DE 10 2017 100 594.3 Only partial solutions for individual steps of the measurement, in particular the optimal selection of the measurement parameters are described and also a restriction to a special sensor technology, namely the computed tomography. Regarding the WO 2016/169 589 It should be noted that here too only individual steps of the measurement are adapted, namely the travel paths between the measuring positions. This refers practically only to the onset of tactile sensors. Likewise in the DE 10 2008 037 441 A1 only the partial solution of the selection of the optimal sensor is described. Here, for example, the adaptation of the corresponding sensor parameters to the respective measuring task is missing. In summary, it should be noted that according to the prior art always a high proportion of user intervention is necessary. This leads to a high susceptibility to errors and low automation of the entire measuring chain.

Object of the present invention is therefore to realize the greatest possible automation of the measurement. In this case, the tolerances assigned to the characteristics to be measured should be taken into account for determining the necessary measurement uncertainty. In order to achieve the necessary measurement uncertainties, u. a. the determination of suitable measuring parameters necessary. Also, the possibly limited measuring time should be considered. An application to any sensor technologies and thus the availability of the method for multi-sensor coordinate measuring machines should also be achieved. Overall, the influence of the operator should be reduced.

• • Bestückung des Koordinatenmessgerätes mit dem zu messenden Werkstück oder Werkstücken
• • Erkennung des zu messenden Werkstücks oder der zu messenden Werkstücke
• • Bereitstellung eines Messprogramms, wobei vorzugsweise ein ggf. bereits vorhandenes Messprogramm bereitgestellt wird
• • Erstellen und/oder Anpassen des Messprogramms
• • Ausführen des Messprogramms
• • Auswahl einer Auswertestrategie und Auswertung der zu messenden Merkmale
• • Zusammenfassung und Verwaltung der mittels des Messprogramms ermittelten Messergebnisse
• • Erkennung von Fehlzuständen und ggf. Eingriff in den oder die Fertigungsprozesse des Werkstücks

wobei vorzugsweise die für ein Messprotokoll vorgesehenen Messergebnisse bzw. Maße durch manuellen Bedienereingriff ausgewählt werden.To solve the invention provides a method for determining particular dimensional features on a workpiece by means of a coordinate measuring machine that preferably has a plurality of sensors, that is pronounced as a multi-sensor coordinate measuring machine or is particularly pronounced as a computed tomography, with several sensors in the sense of Multi-sensor coordinate measuring machine can be provided, in which one or any combination of several of the following automated steps takes place:

• • Equipping the coordinate measuring machine with the workpiece or workpieces to be measured
• • Detection of the workpiece to be measured or the workpieces to be measured
• • Provision of a measuring program, wherein preferably a possibly existing measuring program is provided
• • Create and / or customize the measurement program
• • Execute the measuring program
• • Selection of an evaluation strategy and evaluation of the characteristics to be measured
• • Summary and management of the measurement results determined by the measuring program
• • Detection of fault conditions and, if necessary, intervention in the manufacturing process or processes of the workpiece

wherein preferably provided for a measurement protocol measurement results or measurements are selected by manual operator intervention.

• • Identifizierung der zu messenden Merkmale und/oder Werkstückbereiche und Festlegung der benötigten Messpunkte und/oder Messpositionen
• • Auswahl des für die Merkmale bzw. den Merkmalen zugeordneten Messpunkte geeigneten, vorzugsweise idealen Sensors oder Sensoren, welche in dem jeweiligen Koordinatenmessgerät integriert sind
• • Festlegung der geeigneten Messmethode mit dem jeweiligen Sensor
• • Festlegung der geeigneten, ggf. notwendigen Verfahrbewegungen zwischen den für die Aufnahme der Messpunkte notwendigen Messpositionen
• • Festlegung der geeigneten Lage des Werkstücks und ggf. je Messpunkt veränderten Lage des Werkstücks relativ zum jeweiligen Sensor
• • Festlegung der geeigneten Messparameter mit dem jeweiligen Sensor, ggf. je Messpunkt angepassten Messparametern
• • Festlegung ggf. durchzuführender Korrekturschritte für die Auswertung der mit dem Messprogramm zunächst ermittelten Messergebnisse

wobei die Erstellung und/oder Anpassung des Messprogramms derart erfolgt, dass:

• • Die zu messenden Merkmale bestimmt werden können und
• • Die Messunsicherheit im Vergleich zu ggf. vorliegenden Toleranzen ausreichend klein ist und/oder die Messunsicherheit minimiert wird und/oder eine ggf. vorliegende Messzeitvorgabe eingehalten oder unterschritten wird und/oder
• • eines oder mehrere zuvor definierte Kriterien eingehalten oder optimiert werden, wie beispielsweise die Messpräzision.

Preferably, it is provided that the creation and / or adaptation of the measurement program comprises one or any combination of several of the following automated steps:

• • Identification of the features to be measured and / or workpiece areas and determination of the required measuring points and / or measuring positions
• Selection of the measuring points associated with the features or characteristics suitable, preferably ideal sensor or sensors, which are integrated in the respective coordinate measuring machine
• • Definition of the suitable measuring method with the respective sensor
• • Definition of the appropriate, if necessary, necessary traversing movements between the measuring positions necessary for the recording of the measuring points
• • Determination of the appropriate position of the workpiece and, if necessary, the measuring point changed position of the workpiece relative to the respective sensor
• • Determination of the suitable measuring parameters with the respective sensor, if necessary adjusted measuring parameters for each measuring point
• • Definition of any correction steps to be carried out for the evaluation of the measuring results initially determined with the measuring program

wherein the creation and / or adaptation of the measurement program is such that:

• • The features to be measured can be determined and
• • The measurement uncertainty is sufficiently small in comparison to possibly existing tolerances and / or the measurement uncertainty is minimized and / or a possibly existing measurement time specification is maintained or undershot and / or
• • one or more previously defined criteria are maintained or optimized, such as the measurement precision.

The term feature is understood in the context of the present application, the geometric expression or geometric property of workpiece sections. For example, features are circles, angles, planes, or other geometric shapes. These are dimensional, shape and / or position tolerances assigned, for example in the form of target data such as PMI or CAD data. These are also subsumed here under the term characteristic and represent the measured quantities of the features.

The details of the aforementioned inventive automated steps will be explained in more detail below.

As a first step, the workpiece to be measured must be identified and the measurement or inspection task to be solved identified. For this purpose, it is provided that the assembly of the coordinate measuring machine, for example based on CAD or STL data, for example, by a robot, rail system, pallet changer and / or on the basis of automatic detection of the workpiece to be placed on the device.

Alternatively, the manually placed on the device workpiece is only detected. In particular, when using a computer tomographic sensor is provided that with the aid of CAD data or based on a so-called fast tomography, which is characterized by the inclusion of fewer radiographic images in a reduced actual measurement number of rotational positions and resulting inaccurate reconstruction of the workpiece shape, the Workpiece recognition takes place. In particular, it is also provided according to the invention that, in the presence of a workpiece which has already been measured once after detection of the workpiece, an already existing measuring program is automatically loaded and used.

After fitting and recognizing the workpiece to be measured, the features or areas to be measured on the workpiece must be detected, at which surface measuring points or measuring points must be determined. This can be done according to the invention, for example, on the basis of the so-called PMI data (Product and Manufacturing Information) and / or a tolerance modeled in the CAD model.

The subject matter of an independent invention, which however can be combined with previously largely described idea of the largely automated measurement, is a further method and a device for determining features on a workpiece, wherein the steps used here should also run as largely automated as possible.

To check the stored in the PMI or CAD data, so tolerated features such as dimensions such as diameters or distances, shape tolerances and position tolerances (generic term: dimensional, shape and position tolerances) to roughness details are sometimes complex measurements in different areas of the workpiece necessary. The operator must apply the rules according to the standards and guidelines, ie measurement and evaluation strategies, according to the state of the art, in particular to define areas to be detected on the workpiece (measuring strategy) and link them (evaluation strategy). This also includes references to be measured and indirect references such as symmetry shops, which only arise when measuring and linking several areas. The DE 102015114715 the applicant describes a method in which appropriate measurement strategies are recorded and stored in a database.

A disadvantage of the prior art is that an automated measurement is not possible.

The object of the present invention is to guide the operator or to enable an automated measurement based on the desired data and the tolerances associated with the characteristics.

To solve the invention, the corresponding rules for the measurement of features, including the areas to be measured, the combination of each measured measuring points and the detection and linking of the direct or indirect references to be measured in advance or ongoing expand in a database and remove it automatically for the measurement.

In addition to examples of already known measurement tasks, ie tolerances on features, the database contains rules on how these are applied to similar measurement tasks, for example larger diameters or changed position of the covers. The rules include, in particular, the procedures established in accordance with the applicable standards and guidelines for determining the appropriate dimensional, geometrical and positional tolerances. The rules themselves may be stored in the database itself and / or in a measurement software using the database, which is executed on a data processing device. In particular, the measuring software may be the software (evaluation software) controlling a coordinate measuring machine or received by the coordinate measuring machine. The application of the rules and the access to the database is carried out by the measuring software or this executing data processing device. The database can also be entirely part of the measurement software or at least partially stored in a cloud. The measurement software makes it possible to automatically create a measurement program suitable for checking the tolerances originating from target data such as PMI data or CAD data and / or to interactively assist the user in the creation by making suggestions in accordance with the rules that the operator accept (verify) or modify (eg, offset measurement points due to obstacles or similar). The measurement program uses the appropriate measurement strategy for recording the required measurement points and the appropriate evaluation strategy and is performed by the measurement software on the data processing device.

The invention provides for solving a measuring software for generating a measuring program for dimensional determination of features on workpieces, in particular for comparison with derived from target data such as PMI or CAD data dimensional, shape and / or position tolerances, preferably with a coordinate measuring machine, before that This is characterized by the fact that the measurement software contains the rules, such as measurement and / or evaluation strategy, for determining the respective feature and comparison with dimensional, shape and / or positional tolerances assigned to the respective feature from target data such as PMI or CAD data and / or a database, and automatically generates a measurement program or proposes to the operator a measurement program for verification and / or modification.

In particular, the invention is characterized in that the database and / or measurement software applicable to the respective feature rules with respect to the areas to be measured on the workpiece and in particular optionally measured covers or indirect references such as symmetry and for linking the from the measured areas resulting from the measurement result such as measures.

Preferably, the invention provides that the rules are created in a META language such as DMIS.

It should also be emphasized that, based on the rules, the operator is guided in order to detect the area to be measured or to automatically set the area to be measured.

The invention is also characterized in that new rules for existing or new features of the database and / or measurement software can be added by the operator, for example based on example measurements.

The invention also provides for a solution in the form of a data processing device which is designed to execute a measurement program by the measuring software according to the invention.

Preferably, it is provided that the database is connected via an interface with the data processing device or that the database is integrated into the measurement program.

In already known and already measured workpieces, the removal of the features and associated tolerances takes place from the existing measurement program. Manual selection, for example, in addition, is also provided.

In general, for the determination of the features and / or workpiece areas which are to be measured, the setpoint data of the workpiece are consulted. From the tolerance information associated with the feature, for example, the structural resolution necessary for the realization of the respective measurement task can be derived, and thus the necessary measurement and sensor parameters.

In one embodiment of the invention, it is also provided that the desired data in the sense of pseudo target data are taken from another source, for example an already learned measuring program. It is therefore not absolutely necessary for the target data and the tolerance data to come from a common source.

In an alternative embodiment, the invention also provides, first, the raw data to evaluate automatically and to realize an automated measurement of single points and not only of complete features.

After defining the characteristics to be determined, the automatic selection of the most suitable sensors and the measuring method used is carried out. In this case, a single sensor or a combined measurement with a plurality of sensors can be provided. As a criterion for the selection, for example, the optimization of individual components of the measurement uncertainty, z. B. statistical influences, such as the measurement precision serve. But consideration can also find a possibly predetermined measuring time limit. These criteria are taken into account in the later process, for example, for the optimization of the travel distances, the measuring method and the measuring parameters.

In particular, it is also provided according to the invention to use a plurality of measuring heads. These may be different sensors or also sensors attached to a second quill, as well as different sensor configurations, for example stylus configurations of a tactile sensor.

After the determination of the sensor (s) to be used or in combination with this determination, the appropriate measuring method or method must be determined with the respective sensor. For the example of the computed tomographic measurement, this can include, for example, that the recording of the radiographic images in the movement is performed (on the fly) or only a ROI (Region of Interest) is to measure, in which case in particular the measurement uncertainty to be achieved, which thus dependent of the tolerances assigned to the characteristics, is to be considered. Further selectable measuring methods are, for example, the section-wise measurement of the workpiece in the two directions perpendicular to the main beam direction of an X-ray source assigned to the computer tomograph (referred to as X or Y raster tomography), or the so-called half-page tomography to be selected for measurement time reasons, for example, in which the axis of rotation of the Measuring table only about half of the workpiece in the detected by the detector beam cone of the X-ray source is present. Also provided may be a so-called fast CT, in which a reduced number of rotational positions for the measurement is taken or the exposure time per rotational position is shortened. The use of slit diaphragms for reducing artifacts, in particular scattered radiation, which is estimated in advance by means of simulation, for example, and the use of other scattered radiation grids between the x-ray source and workpiece and / or workpiece and detector can also be provided as measurement parameters.

When determining and adapting the measurement parameters with the respective sensor, several steps must be taken into account. This includes, for example, the determination of the suitable and, if necessary, necessary traversing movements between the measuring position necessary for receiving the measuring points. In this case, the control commands can be adapted with regard to the measuring locations, travel paths, speeds and accelerations. It is also envisaged to optimize the evaluation of the measurement results, ie to specify how, for example, individual measurement points are linked to dimensions.

Also includes in determining the evaluation strategy or the measurement parameters is the determination of the appropriate position of the workpiece or possibly depending on the measuring point to be changed position of the workpiece, as can be done for example with the linear and / or possibly existing axis of rotation. Thus, the orientation of the workpiece relative to the measuring head or sensor is taken into account or optimized. The suitable object position can for example be based on the CAD or STL data, for example by simulation. Alternatively, it is also provided to adapt the object position, ie to regulate it, for example with axes of rotation, and thereby iteratively assess the suitability of the selected object position.

For determining the suitable measurement parameters with the respective sensor, a distinction is made according to the invention between time-dependent and / or measurement time-influencing (time variants) and measurement times independent of the measuring time (time-invariant) setting parameters. Time variant setting parameters in the case of tactile sensors are, for example, the scanning speed, in the case of computer tomographic sensors, the number of transmitted images to be averaged, the number of rotation steps, the integration time per recorded radiographic image, and in the case of image processing, for example, the number of images to be overlaid and Integration time per picture. In the case of tactile probing systems, time-invariant parameters are, for example, the stylus to be selected, in particular the stylus length and orientation, for example the set acceleration voltage, the power realized in this case, inserted prefilters (beam filters) and, for example, the illumination intensity used in the case of image processing.

For the time-invariant parameters, there are clearly optimal values regardless of the measuring time or uncertainties, which then only result in parameter settings that depend on the measuring task or tolerance. After determination of the time-invariant measurement parameters, the sequence can be adjusted to a measurement time and / or with the aid of the time-variant parameters Measurement uncertainty requirement are adjusted. If all measuring positions for the features to be measured are known, an optimum path for the various movements of the sensors and / or measuring heads can be determined from the relative position of the features relative to each other. If the model is sufficiently good, then no further iteration steps are necessary. If the model is not sufficiently good, the measurement procedure must be implemented once, in order to then adapt the measurement process to the measurement uncertainty specifications or the measurement time specification via the time-variant parameters. Here, too, it is envisaged to optimize the individual components of the measurement uncertainty (eg statistical influences), for example the precision of the measurement.

Using the example of computed tomography, this includes, for example, the optimization of the average CNR (Contrast to Noise Ratio) of the voxel gray values necessary for the determination of the necessary surface points, the minimization of the mean projection integral or the average transmission length of the projections contributing to the determination of a surface point. It is also provided that an automatic determination of the parameters to be measured and / or measurement object areas to be examined (in the example of the material inspection) and / or optimum adjustment parameters on the basis of simulations or by regulation, as mentioned above, takes place.

The setting parameters must therefore be determined in such a way that tolerances can be measured and / or the measurement uncertainty can be optimized and, if necessary, existing measurement time specifications can be maintained.

In particular, it is provided according to the invention to take into account the material properties, in particular based on the desired data, which are generated for example by means of the aforementioned rapid CT. The consideration takes place, for example, by means of simulation and the optimization of the parameters with regard to the tolerances to be maintained. With regard to computed tomography, the optimizable parameters are, for example, the magnification to be selected in conjunction with the measuring range available thereby, the distance between the measuring radiation emitting spot of the X-ray source and the detector, the acceleration voltage of the X-ray tube, the power of the X-ray tube using pre-filters (beam filters), the number of image averages and the integration time per image. In the example of image processing, the magnifications, the aperture of the optics, the integration time per image and the illumination current are available as parameters, for example.

Corresponding measurement or sensor parameters exist for all other tactile and optical sensors to which the invention also relates.

Further examples of the criteria for minimizing or reaching a given value are the average single-point measurement uncertainty, the mean measurement uncertainty of the features, the CNR (Contrast to Noise Ratio) and / or the SNR (Signal to Noise Ratio) of the gray values (for example image intensities of images or radiographic images or attenuation values in a reconstructed computed tomography volume) and the measurement time.

As a further measurement parameter that can be optimized, correction steps to be performed are to be provided. In the case of computer tomographic measurements, these are, for example, the following corrections.

It is envisaged that suitable artifact correction methods are automatically proposed or prefilters (beam filters). These are selected on the basis of, for example, simulation or previously mentioned fast CTs or on the basis of the target data. This results in the correction steps to be optimized. Also provided is the consideration of the aforementioned PMI data. With regard to computed tomography, correction methods are known, for example, as so-called autocorrection, in which selected tactile or optically recorded measuring points are compared with the locally corresponding computer tomographically recorded measuring points and the deviations are used as a network of correction values for the entire workpiece or for the measuring points associated with the characteristics be interpolated. As a further artifact correction method, a virtual auto correction is provided. This compares simulation results with and without consideration of artifacts. Also provided is a so-called empirical artifact correction. In particular, beam hardening artifacts are determined and corrected on the basis of characteristic curves. In another artifact correction method, the recording of a so-called multi-spectral CT is provided. In this case, different frequency bands of the radiation source are adjusted in order to irradiate differently intensively absorbing regions of the workpiece separately with optimized contrast, the results being evaluated in summary form.

When determining the evaluation strategy, it is provided according to the invention to evaluate the features to be measured for example on the basis of the PMI data. The selection of the measurement points to be linked for the evaluation of the corresponding features or dimensions takes place, for example, via the analysis of the resulting measurement uncertainty. Optimization criterion can be here, for example the most stable evaluation, ie the existence of the greatest repeatability. The analysis can for example be based on simulation or based on prior knowledge, for example from a database.

According to the invention, the automatic evaluation also includes the statistical evaluation of the measuring process, that is, for example also capability studies (measurement capability or measurement process suitability), the quantification of the reproducibility or the repeatability of the measurement and / or the manufacturing process, with possibly calibrated master parts repeatedly being measured To assess the deviations from the calibration value, the measurement uncertainty or the ability or suitability.

According to the invention, the summarization and management of the measurement results, ie the data management and data processing, also comprises the automatic combination of the determined measured values, in particular in connection with the measurement uncertainties present in a protocol or test report in which, for example, measured characteristics with measured values and measurement uncertainty and optionally existing images of the features are stored. From this it is possible, for example, to derive information for large data volumes (big data) for use in the production line. For example, the labeling of problems by lamps, automatic corrections of the manufacturing process such as the tool correction in conjunction with the storage of appropriate information in the appropriate database systems. The storage and / or transfer of the evaluated data is carried out automatically to a previously defined interface with an ERP system or the like.

In the following, exemplary possible and meaningful combinations of the aforementioned steps will be described. Combinations are also specified separately for use with and without a CT scanner.

• • Erkennung des zu messenden Werkstücks oder mehrerer zu messender Werkstücke, vorzugsweise anhand der Erkennung eines am jeweiligen Werkstück angebrachten Erkennungsmerkmals wie beispielsweise Barcodes oder RFID-Tag, wobei besonders bevorzugt die Erkennung mittels eines am Koordinatenmessgerät angebrachten Sensors erfolgt und
• • Zuordnung der dem Werkstück zugeordneten Solldaten wie beispielsweise PMI-Daten, zum erkannten Werkstück und
• • automatische Generierung einer Auswertestrategie basierend auf den zugeordneten Solldaten je Merkmal und daraus Erzeugung eines Messablaufs bzw. Messprogramms oder Bereitstellung eines bereits vorhandenen dem erkannten Werkstück zugeordneten Messprogramms, beispielsweise aus einer Datenbank, und
• • das Messprogramm automatisch ausgeführt wird und
• • Zusammenfassung und Verwaltung der mittels des Messprogramms ermittelten Messergebnisse, vorzugsweise Erstellung eines Messprotokolls und besonders bevorzugt Ausgabe an eine Schnittstelle zur weiteren Auswertung, vorzugsweise CAQ-Schnittstelle

und wobei vorzugsweise eine oder mehrere der folgenden Schritte automatisch erfolgt:

• • Bestückung des Koordinatenmessgerätes mit dem zu messenden Werkstück oder Werkstücken und/oder
• • automatische Auswahl einer für ein Merkmal am besten geeigneten Auswertestrategie zur Auswertung der zu messenden Merkmale insbesondere Verfahren zur Bestimmung und Verknüpfung der Oberflächenpunkte und/oder
• • Erkennung von Fehlzuständen und ggf. Eingriff in den Fertigungsprozess des Werkstücks, vorzugsweise Steuerung der einer Werkzeugmaschine bzw. Fertigungsmaschine zugeordneten Steuerung bzw. Steuersoftware.

Preferably, it is provided that the following steps are carried out:

• Recognition of the workpiece to be measured or of several workpieces to be measured, preferably based on the recognition of an attached to the workpiece detection feature such as barcodes or RFID tag, with particular preference is the detection by means of a sensor attached to the coordinate measuring machine and
• • Assignment of the setpoint data assigned to the workpiece, such as PMI data, to the detected workpiece and
• • automatic generation of an evaluation strategy based on the assigned target data per feature and therefrom generation of a measurement process or measurement program or provision of an already existing the recognized workpiece associated measurement program, for example, from a database, and
• • the measuring program is executed automatically and
• Summary and management of the measurement results determined by means of the measurement program, preferably preparation of a measurement protocol and particularly preferred output to an interface for further evaluation, preferably CAQ interface

and wherein preferably one or more of the following steps occurs automatically:

• • Equipping the coordinate measuring machine with the workpiece or workpieces to be measured and / or
• Automatic selection of an evaluation strategy which is most suitable for a characteristic for evaluating the features to be measured, in particular methods for determining and linking the surface points and / or
• • Detection of faulty conditions and, if necessary, intervention in the production process of the workpiece, preferably control of a machine tool or production machine associated control or control software.

• • Groberkennung der Geometrie und/oder Lage des zu messenden Werkstücks oder der zu erkennenden Werkstücke, indem eine Schnelltomografie oder Aufnahme und Auswertung von mehreren Durchstrahlungsbildern in mehreren Drehstellungen des Werkstücks erfolgt, wobei die Bildaufnahme mit einer verkürzten Messzeit im Vergleich zur eigentlichen Messung des Werkstücks erfolgt, vorzugsweise Messung zur Groberkennung mit geringerer Anzahl von Drehstellungen und/oder geringere Belichtungszeit je Bild und/oder geringere Anzahl von Bildmittelung realisiert wird, und/oder Erkennung der Geometrie und/oder Lage mittels eines am Werkstück befestigten Erkennungsmerkmals wie Barcode oder RFID-Tag und/oder Erkennung der Geometrie und/oder Lage durch Empfang einer Kennung über eine Schnittstelle zu beispielsweise einer Fertigungsanlage, wobei bei Vorhandensein eines Erkennungsmerkmals als die Objektlage eine vordefinierte Standardobjektlage angenommen wird oder die Objektlage durch eine zusätzliche Schnelltomografie bestimmt wird, und
• • Erstellung und/oder Anpassen des Messprogramms, wobei anhand des erkannten oder grob erkannten Werkstücks die geeigneten Messparameter des Computertomografiesensors automatisch eingestellt werden und/oder unter Berücksichtigung eines am Werkstück befestigten und erkannten Erkennungsmerkmals wie Barcode oder RFID-Tag diesem zugeordnete, zuvor definierte Messparameter eingestellt werden, wobei Messparameter insbesondere sind die Röhrenparameter wie Beschleunigungsspannung und Strom, mechanischer Vorfilter, Bildaufnahmeparameter wie Anzahl der Drehschritte, Anzahl der Bildmittelung je Drehschritt, Integrationszeit je Bild oder Geschwindigkeit der Drehbewegung und Integrationszeit je Bild und/oder Anzahl der zu überlagernden Bilder, und vorzugsweise Binning-Parameter, und vorzugsweise Abbildungsmaßstab, insbesondere Relativposition von Röntgenröhre, Röntgendetektor und Drehtisch relativ zueinander, insbesondere relativ zueinander in Abbildungsrichtung der Röntgenstrahlung und vorzugsweise quer zur Abbildungsrichtung, insbesondere zur Aufnahme senkrecht zur Röntgenstrahlrichtung versetzt aufgenommener Durchstrahlungsbilder für Region of Interest (ROI) CTs oder Raster-CTs, wobei vorzugsweise Messparameter unter Berücksichtigung der dem Werkstück zugeordneten Solldaten wie PMI-Daten, insbesondere den Merkmalen zugeordneten Toleranzen eingestellt werden, und besonders bevorzugt Festlegung der geeigneten Lage oder mehrerer geeigneter Lagen des Werkstücks auf einem Drehtisch relativ zum Computertomografiesensor (Ausrichtung des Werkstücks) auf Basis der Simulation der Computertomografie anhand der Grobgeometrie des Werkstücks oder auf Basis anderer Informationsquelle wie über eine I/O-Schnittstelle, RFID und/oder Barcode adressierte Informationsquelle), und
• • Ausführen des Messprogramms oder zumindest des Teils des Messprogramms, der die Aufnahme der Durchstrahlungsbilder, die Rekonstruktion des Voxelvolumens und vorzugsweise die Generierung von Oberflächenmesspunkten und bevorzugt die Ausrichtung des Werkstücks umfasst.

In particular, the invention is characterized in that the following steps are carried out when using a computed tomography or computer tomography sensor:

• Coarse recognition of the geometry and / or position of the workpiece to be detected or the workpieces to be detected by a rapid tomography or recording and evaluation of several radiographic images in several rotational positions of the workpiece, wherein the image acquisition takes place with a shortened measurement time compared to the actual measurement of the workpiece , Preferably measurement for coarse detection with a smaller number of rotational positions and / or lower exposure time per image and / or reduced number of image averaging is realized, and / or recognition of the geometry and / or position by means of a workpiece attached to the recognition feature such as barcode or RFID tag and / or recognition of the geometry and / or position by receiving an identifier via an interface to, for example, a production facility, wherein in the presence of a recognition feature as the object location a predefined Standard object position is assumed or the object position is determined by an additional fast tomography, and
• Creation and / or adaptation of the measuring program, whereby the suitable measuring parameters of the computed tomography sensor are automatically adjusted on the basis of the recognized or roughly recognized workpiece and / or set to a predefined measuring parameter assigned to it, taking into account a recognition feature attached to and recognized by the workpiece, such as a barcode or RFID tag in particular, the tube parameters such as acceleration voltage and current, mechanical pre-filter, image acquisition parameters such as number of rotation steps, number of image averaging per rotation step, integration time per image or speed of rotation and integration time per image and / or number of images to be overlaid, and preferably Binning parameters, and preferably magnification, in particular relative position of the X-ray tube, X-ray detector and turntable relative to each other, in particular relative to each other in the imaging direction of the X-radiation and Preferably transversely to the imaging direction, in particular for recording perpendicular to the X-ray direction offset recorded radiographic images for region of interest (ROI) CTs or raster CTs, preferably measurement parameters are adjusted taking into account the workpiece associated with desired data such as PMI data, in particular the characteristics associated tolerances, and particularly preferably determining the suitable position or multiple suitable positions of the workpiece on a turntable relative to the computed tomography sensor (alignment of the workpiece) on the basis of the simulation of computed tomography based on the rough geometry of the workpiece or on the basis of other information source such as via an I / O interface, RFID and / or barcode addressed information source), and
• • Execute the measuring program or at least the part of the measuring program that contains the Recording of the radiographic images, the reconstruction of the voxel volume and preferably the generation of surface measuring points and preferably comprises the orientation of the workpiece.

• • Groberkennung der Geometrie und/oder Lage des zu messenden Werkstücks oder der zu erkennenden Werkstücke, indem eine Schnelltomografie oder Aufnahme und Auswertung von mehreren Durchstrahlungsbildern in mehreren Drehstellungen des Werkstücks erfolgt, wobei die Bildaufnahme mit einer verkürzten Messzeit im Vergleich zur eigentlichen Messung des Werkstücks erfolgt, vorzugsweise Messung zur Groberkennung mit geringerer Anzahl von Drehstellungen und/oder geringere Belichtungszeit je Bild und/oder geringere Anzahl von Bildmittelung realisiert wird, und/oder Erkennung der Geometrie und/oder Lage mittels eines am Werkstück befestigten Erkennungsmerkmals wie Barcode oder RFID-Tag und/oder Erkennung der Geometrie und/oder Lage durch Empfang einer Kennung über eine Schnittstelle zu beispielsweise einer Fertigungsanlage, und
• • über ein Schnittstelle das dem erkannten Werkstück zugeordnete Messprogramm, welches vorzugsweise einer Standardlage des Werkstücks zugeordnet ist, eingelesen wird und
  • o das Messprogramm und die Lage des Werkstücks unverändert beibehalten wird oder
  • o die entsprechend der Schnelltomografie oder der Aufnahme und Auswertung von mehreren Durchstrahlungsbildern in mehreren Drehstellungen des Werkstücks ermittelte Lage des Werkstücks mit der Standardlage verglichen wird und gegebenenfalls vorliegende Lageabweichungen verwendet werden, um die Einstellparameter des Computertomografiesensors anzupassen, wobei entweder die Anpassung der Lage (Ausrichtung des Werkstücks) so erfolgt, dass die Standardlage eingenommen wird, oder die Anpassung der Messparameter erfolgt, wie beispielsweise Röhrenparameter wie Beschleunigungsspannung und Strom, mechanischer Vorfilter, Bildaufnahmeparameter wie Anzahl der Drehschritte, Anzahl der Bildmittelung je Drehschritt, Integrationszeit je Bild oder Geschwindigkeit der Drehbewegung und Integrationszeit je Bild und/oder Anzahl der zu überlagernden Bilder, und vorzugsweise Binning-Parameter, und vorzugsweise Abbildungsmaßstab, insbesondere Relativposition von Röntgenröhre, Röntgendetektor und Drehtisch relativ zueinander, insbesondere relativ zueinander in Abbildungsrichtung der Röntgenstrahlung und vorzugsweise quer zur Abbildungsrichtung, insbesondere zur Aufnahme senkrecht zur Röntgenstrahlrichtung versetzt aufgenommener Durchstrahlungsbilder für Region of Interest (ROI) CTs oder Raster-CTs
• • Ausführen des Messprogramms oder zumindest des Teils des Messprogramms, der die Aufnahme der Durchstrahlungsbilder, die Rekonstruktion des Voxelvolumens und vorzugsweise die Generierung von Oberflächenmesspunkten umfasst.

Preferably, the invention provides that the following steps are carried out when using a computed tomography scanner or computer tomography sensor:

• Coarse recognition of the geometry and / or position of the workpiece to be detected or the workpieces to be detected by a rapid tomography or recording and evaluation of several radiographic images in several rotational positions of the workpiece, wherein the image acquisition takes place with a shortened measurement time compared to the actual measurement of the workpiece , Preferably measurement for coarse detection with a smaller number of rotational positions and / or lower exposure time per image and / or reduced number of image averaging is realized, and / or recognition of the geometry and / or position by means of a workpiece attached to the recognition feature such as barcode or RFID tag and / or recognition of the geometry and / or position by receiving an identifier via an interface to, for example, a manufacturing facility, and
• • via an interface that the detected workpiece associated measuring program, which is preferably assigned to a standard position of the workpiece is read, and
  • o the measuring program and the position of the workpiece are kept unchanged or
  • o the position of the workpiece determined according to the rapid tomography or the recording and evaluation of several radiographic images in several rotational positions of the workpiece is compared with the standard position and possibly existing positional deviations are used to adapt the setting parameters of the computed tomography sensor, wherein either the adaptation of the position (alignment of the Workpiece) is done so that the standard position is taken, or the adjustment of the measurement parameters takes place, such as tube parameters such as acceleration voltage and current, mechanical pre-filter, image acquisition parameters such as number of rotation steps, number of image averaging per rotation step, integration time per image or speed of rotation and integration time each image and / or number of images to be superimposed, and preferably binning parameters, and preferably magnification, in particular relative position of the x-ray tube, x-ray detector and rotary ice Ch relative to each other, in particular relative to each other in the imaging direction of the X-ray and preferably transversely to the imaging direction, in particular for recording perpendicular to the X-ray direction offset shot radiographs for Region of Interest (ROI) CTs or raster CTs
• • Execution of the measuring program or at least of the part of the measuring program which includes the recording of the radiographic images, the reconstruction of the voxel volume and preferably the generation of surface measuring points.

• • Zuordnung der Solldaten, insbesondere CAD-Modell und/oder PMI-Daten des Werkstücks, entweder durch Vergleich der Messdaten bzw. zuvor ermittelten Oberflächenmesspunkte mit in einer Datenbank gespeicherten Solldaten mehrerer Werkstück oder anhand über ein Erkennungsmerkmal wie RFID-Tag, Barcode oder über eine I/O-Schnittstelle referenzierte Solldaten aus einer Datenbank oder einzelnem dem Werkstück zugeordneten Solldatensatz, und
• • Auswahl der auszuwertenden Merkmale und Einlesen der diesen zugeordneten Maß-, Form und/oder Lagetoleranzen anhand der zugeordneten Solldaten, und
• • Auswahl einer Auswertestrategie, vorzugsweise anhand einer den Merkmalen zugeordneten Regeln zur Erzeugung der Auswertestrategien enthaltenden Datenbank und
• • Auswertung der Messergebnisse zu den aus Oberflächenpunkten ermittelbaren Merkmalen und/oder aus den aus dem mittels eines Computertomografen bzw. Computertomografiesensors erzeugten Voxelvolumen ermittelbaren Merkmalen wie beispielsweise Lunkern oder Einschlüssen, Fehlstellen etc. und vorzugsweise Zusammenfassung und Verwaltung der Messergebnisse, vorzugsweise Erstellung eines Messprotokolls und besonders bevorzugt Ausgabe an eine Schnittstelle zur weiteren Auswertung, vorzugsweise CAQ-Schnittstelle, Datenbank oder ähnliches.

The subject matter of an independent invention, which, however, can also be combined with the aforementioned ideas, is a method for evaluating surface measuring points generated by means of a sensor integrated in a coordinate measuring machine, which is characterized in that the following steps are carried out:

• • Assignment of the target data, in particular the CAD model and / or PMI data of the workpiece, either by comparing the measured data or previously determined surface measuring points with target data of several workpieces stored in a database or by means of an identification feature such as RFID tag, barcode or via a I / O interface referenced setpoint data from a database or individual target data set associated with the workpiece, and
• Selection of the characteristics to be evaluated and reading of the dimensional, shape and / or positional tolerances associated therewith on the basis of the assigned nominal data, and
• Selection of an evaluation strategy, preferably based on a database containing the characteristics associated with the characteristics for generating the evaluation strategies
• Evaluation of the measurement results for the features that can be determined from surface points and / or for the features that can be determined from the voxel volume generated by a computer tomograph or computer tomography sensor, such as voids or inclusions, defects etc. and preferably summarizing and managing the measurement results, preferably preparation of a measurement protocol and especially preferably output to an interface for further evaluation, preferably CAQ interface, database or the like.

• • Erkennung des zu messenden Werkstücks oder der zu messenden Werkstücke, wobei mittels eines am Koordinatenmessgerät angebrachten Sensors das Werkstück grob gemessen wird und/oder ein am Werkstück angebrachtes Erkennungsmerkmal wie Barcode oder RFID-Tag erkannt wird und/oder die Information per I/O-Schnittstelle von einem Gerät einer Fertigungsanlage an das Messgerät gesendet wird, und
• • Zuordnung eines bereits vorhandenen Messprogramms oder automatische Erstellung eines Messprogramms anhand der dem erkannten Werkstück zugeordneten Solldaten wie beispielsweise PMI-Daten, wobei die zu messenden Merkmale identifiziert und zu messende Messpunkte am Werkstück festgelegt werden und für die zu messenden Messpunkte jeweils der geeignete Sensor festgelegt wird, wobei insbesondere die Zugängigkeit des Merkmals und/oder die dem Merkmal zugeordnete zu prüfende Toleranz und/oder die Größe und/oder die notwendige Strukturauflösung für die Erfassung des Merkmals berücksichtigt wird und die geeignete Messstrategie ausgewählt wird und die geeigneten Sensorparameter eingestellt werden, und
• • das Messprogramm ausgeführt wird und die entsprechenden Messpunkte mit dem jeweils ausgewählten Sensor aufgenommen werden und mit der ausgewählten Messstrategie die Merkmale ausgewertet werden, wobei die den einzelnen Merkmalen zugeordneten Regeln zur Erzeugung der Messstrategie vorzugsweise aus einer Datenbank oder dem bereits vorhandenen Messprogramm entnommen werden.

The invention is also characterized in that, when using a coordinate measuring machine with one or more sensors, the following automated steps take place:

• Detection of the workpiece to be measured or of the workpieces to be measured, whereby the workpiece is roughly measured by means of a sensor attached to the coordinate measuring machine and / or a recognition feature attached to the workpiece, such as a barcode or RFID tag, is detected and / or the information is transmitted via I / O. Interface is sent from a device of a manufacturing plant to the meter, and
• • Allocation of an existing measurement program or automatic creation of a measurement program based on the target data associated with the detected workpiece, such as PMI data, whereby the features to be measured are identified and measurement points to be measured are determined on the workpiece and the appropriate sensor is determined for the measurement points to be measured in particular taking into account the accessibility of the feature and / or the tolerance to be tested and / or the size and / or the necessary structure resolution for the detection of the feature and the appropriate measurement strategy is selected and the appropriate sensor parameters are set, and
• • the measuring program is executed and the corresponding measuring points are recorded with the respectively selected sensor and the features are evaluated with the selected measuring strategy, wherein the rules assigned to the individual features for generating the measuring strategy are preferably taken from a database or the already existing measuring program.

Subject of an independent invention is an apparatus and a method for selectively measuring and / or calibration of a computed tomography sensor.

Conventionally, for the calibration of computed tomography sensors normal or test specimens are used, the freestanding form elements such as balls, ball bars or corresponding arrangements of these, which are connected via pins to a common base, the base is not detected by the measuring beam of the computed tomography sensor and therefore from a long-term stable material can be made that has a high density and hardly be radiated.

As a normal or test specimens are further understood objects that embody calibrated lengths such as distances, center distances of balls or diameter of balls, so correspondingly have shaped sections as shaped elements such as spheres, holes or plane sections. In the following, therefore, the terms "sections of the specimen" or "specimen sections" are used for these form elements and optionally the term normal or specimens.

In the DE 10 2010 017 508 The Applicant is encouraged to arrange the specimens in a substrate significantly lower density than the density of the normal or specimen. The disclosure of this patent application is hereby fully incorporated by reference.

A disadvantage of the known method, however, is that a calibration of the normal or specimen, ie the embodied lengths, must be done by alternative sensors such as optical or tactile or tactile-optical sensors. For this purpose, the sections of the test specimens or of the standard must be accessible for the calibration. In the aforementioned DE 10 2010 017 508 the applicant is provided for this purpose a free space when introduced into the carrier material. The disadvantage here is that it is only possible through complex modular structure of individual substrates, to arrange sections of the test specimen or standard one above the other. As a result, deviations in the assembly are to be expected, whereby a calibration of the distances between the sections of the test specimen or normal to each other is not accurate enough and only with low Lanzeitstabilität is possible.

The object of the present invention is therefore also to provide a device and a method for calibrating or calibrating a computed tomography sensor, in which a highly accurate calibration is also realized for those test specimens or normals in which a plurality of balls are arranged one above the other and therewith are inaccessible for tactile, optical or tactile-optical calibration. It is also an object of the invention to produce a corresponding test body easily and inexpensively.

To solve the invention provides that in the substrate openings such as holes or blind holes are introduced, preferably drilled or milled, or recesses are provided, for example, by foaming or powdering or pressing in which the sections such as balls of the normal or Test specimens are introduced.

When superimposing several of these sections, according to one embodiment, spacers made, for example, of the carrier material are introduced between the sections or foaming or powdering or pressing in with the carrier material takes place, which is preferably a foam, particularly preferably a carbon foam or a powder such as carbon dust.

A particularly worth protecting idea envisages that with a separate device before foaming the sections are taken up, for example with pins starting from a base, which are detachable from the sections after foaming, in order to arrange the sections at nominal positions, before calibrating and after foaming again from the pins o.ä. to separate. This results in the advantage that the calibration can be carried out before foaming by means of conventional tactile, optical or tactile-optical sensors, as provided by an embodiment of a method according to the invention.

In addition, the invention provides for the solution, a method in which the calibration by comparator measurement with a computed tomography sensor, preferably with the intended for calibration or calibration Computertomografiesenor, where a tactile, optical or tactile-optically calibrated normal, ie normal with freestanding Sections, referred to here as the second normal or specimen, is used. Preferably, the positions of the portions of the second specimen or normal are identical or nearly identical to those of the actual normal or specimen used during the comparator measurement.

As a result, calibrated lengths are produced between the sections of the standard or test specimen in an easily passable support material which, in addition to the calibration, can also be used to check the length measurement deviation described in guidelines and standards. Lengths formed from a single section, such as ball diameters, can be used to check the standard deviation. For this purpose, the sections, in particular balls are calibrated prior to insertion into the substrate with conventional tactile, optical or tactile-optical sensor. By knowing the calibrated ball diameters and ball center distances, calibrated bidirectional lengths of the standard or test specimen are also available, as required for standard-compliant checking of the bidirectional length measurement deviation.

A device according to the invention for measuring and / or calibrating a computed tomography sensor or a coordinate measuring machine with at least one computed tomography sensor or a multisensor coordinate measuring machine with at least one computed tomography sensor using a normal or test specimen which is decomposed by the x-ray radiation of the radiation source of the computed tomography sensor, the plurality of computed tomography sensors Sections, preferably balls, of the test specimen are arranged in a carrier material of significantly lower density than the density of the test specimen or standard, characterized in that in the carrier material a plurality of openings, in particular holes and / or blind holes, introduced and / or more by lathering, Einpulvern or pressing of the carrier material resulting recesses are provided, wherein each opening or recess at least a portion of the normal or specimen is introduced, and wherein the Absc hnitte inside of the normal or test specimen, and in particular not accessible for a calibration or measurement with tactile, optical and / or tactile-optical sensors.

In particular, the invention provides that at least two sections are arranged per opening, wherein a spacer, preferably made of the carrier material spacers is disposed between each two sections within an opening, more preferably outwardly directed end of the openings by lathering with the carrier material or additional spacers are closed.

In particular, the invention is characterized in that the openings are perpendicular to the surface of the carrier material introduced holes or blind holes, in particular holes or blind holes.

Preferably, the invention provides that the sections are introduced into the recesses by the sections are first arranged at predetermined desired positions and fixed by foaming, Einpulvern or pressing the carrier material, wherein the carrier material is preferably a foam such as carbon foam or a powder such as carbon powder or Carbon dust is, and wherein particularly preferably carrier material is provided after foaming, Einpulvern or pressing with a border, preferably Lamitat.

It should also be emphasized that the device has a base such as hollow cylinder and emanating from the base holding elements such as pins, preferably releasably go out, and wherein the pins are formed to receive the sections releasably, preferably by screwing and / or plug-in releasably, so after the foaming, Einpulvern or pressing of the carrier material in or around the base, the base and the holding elements separately or jointly from the sections and the carrier material are separable.

The invention is also distinguished by the fact that the base and retaining elements are arranged so that the sections arranged thereon can be measured or calibrated by means of tactile, optical and / or tactile-optical sensors before foaming, powdering or pressing in of the carrier material.

In particular, the invention is characterized in that the carrier material consists of a carbon foam or carbon powder or carbon dust, and preferably has a thermal expansion coefficient of <10 × 10 ^-6 per Kelvin.

According to a proposal to be particularly emphasized, it is provided that test piece sections form a plurality of test lengths along different spatial directions, preferably at least two or at least three spatial directions.

The invention also provides a method for measuring and / or calibrating a computed tomography sensor or a coordinate measuring machine with computed tomography sensor or a multisensor coordinate measuring machine with computed tomography sensor using the device according to the invention described above, characterized in that the calibration of the through the multiple sections of the normal or test specimen formed test lengths by the distances, in particular spherical center distances of the sections of the normal or specimen by means of a computed tomography, preferably the einzumessenden or to be calibrated computed tomography, are measured and with the determined by the same computed tomography sensor distances, in particular ball center distances, a second test body or Normals be compared or compared, the distances of the second test specimen or standard by testing body sections are formed, which are not surrounded by a carrier material and are calibrated by tactile and / or optical and / or tactile-optical sensor.

In particular, the invention is characterized in that the calibration and / or calibration by means of comparator method, wherein the sections of the second normal or specimen are arranged at approximately the same positions within the measuring range of the computed tomography sensor as sections of the actual normal or specimen ,

It is preferably provided that test lengths formed by a respective section of the standard or test specimen, such as diameter of the sections, in particular ball diameter, prior to insertion into the carrier material or before foaming, powdering or pressing in of the carrier material, preferably by means of tactile, optical and / or tactile optical sensor system are calibrated.

The invention also provides a method for measuring and / or calibrating a computed tomography sensor or a coordinate measuring machine with computed tomography sensor or a multisensor coordinate measuring machine with computed tomography sensor using the device according to the invention described above, characterized in that the calibration of the through the multiple sections Normally or specimen formed test lengths before foaming, Einpulvern or pressing of the carrier material is carried out by the distances, in particular spherical center distances of the sections of the normal or specimen by means of tactile, optical and / or tactile-optical sensors are measured.

It should also be emphasized that distances, in particular spherical center distances, before and after foaming, powdering or pressing in of the carrier material are measured by means of a computed tomography sensor, preferably the computed tomography sensor to be measured or calibrated, and deviations between the two measurements are used to correct the calibration.

The invention is also distinguished by the fact that test lengths, such as diameter of the sections, in particular ball diameter, formed by a respective section of the standard or test specimen, preferably by means of tactile, optical and / or pre-foaming, powdering or pressing in before insertion into the substrate or. or tactile-optical sensors are calibrated.

The invention also provides for a method for producing a standard or specimen according to the device according to the invention described above, which is characterized in that the sections are introduced into the recesses by first arranging the sections at predetermined desired positions and by foaming, Einpulvern or pressing the carrier material are fixed, wherein the carrier material is preferably a foam such as carbon foam or a powder such as carbon powder or carbon dust, and wherein particularly preferably carrier material after foaming, Einpulvern or pressing is provided with a border, is preferably laminated.

According to a particularly noteworthy proposal is provided that are attached to a base such as hollow cylinder holding elements such as pins, preferably releasably secured, and on the pins, the sections are releasably attached, preferably by screwing and / or plugging, and that after foaming, Einpulvern or pressing the carrier material into or around the base, the base and the holding elements separately or jointly separated from the sections and the carrier material.

The subject matter of an independent invention is an apparatus and a method for measuring a workpiece with a computer tomograph and an additional sensor.

From the WO2005119174A1 The applicant, the content of which is expressly incorporated herein by reference, a multisensor coordinate measuring machine comprising a computer tomograph and additional sensors is known. Also described are methods for correcting the measurement data taken with the computer tomograph on the basis of the measurement data recorded with the additional sensor. For this purpose, reference is made to claims 36 to 38 and 84 to 94 and related description, the disclosure of which is to be taken as fully incorporated herein.

In the DE102006022103A1 The applicant, whose contents are also expressly incorporated herein by reference, describes a method for correcting the measurement data of a computer tomograph, in which a local segmentation threshold is assigned on the basis of the measurement data of a measurement point obtained by a second method.

However, it is disadvantageous that the measurements with the computer tomograph and the additional second sensor take place one after the other, in some cases even in separate devices, whereby a re-clamping of the workpiece and the fitting of the different coordinate systems is necessary. This results in a large total measuring time. In addition, measurement deviations may occur due to drift phenomena.

It is therefore also an object of the present invention to specify an apparatus and a method for the precise and particularly rapid measurements of a workpiece. In particular, the workpiece should be measured by means of a computer tomograph and by means of at least one additional sensor. A correction of the measured data of the computer tomograph on the basis of the measured data of the additional sensor (s) is a further object of the invention.

To solve the invention provides that at least temporarily takes place a simultaneous dimensional measurement of the workpiece with the computer tomograph and the at least one additional sensor. A suitable multisensor coordinate measuring machine is equipped accordingly to allow the simultaneous measurement, and has for this purpose in particular a suitable control device. This allows a fast, parallel measurement of a workpiece with computed tomography and the outer contour with further sensor technology. This allows the rapid reduction of artefacts occurring in computed tomography by correction using the measurement data of the additional sensor (s). Also, a solution is provided to shield the additional sensor (s) from radiation emitted by the source to avoid damage.

The invention also provides a method for the dimensional measurement of a solution Workpiece with a computer tomograph and at least one tactile, optical and / or tactile-optical sensor (additional sensor) before, the computed tomography consists of at least radiation source (source), in particular X-ray source, detector, in particular X-ray detector, preferably flat X-ray detector, and the workpiece receiving and in the beam path between source and detector rotating turntable, wherein in several rotational positions of the workpiece gray scale radiographic images are taken with the detector and these are reconstructed into a voxel with gray voxel volume from the surface extraction (surface) measuring points (measuring points) are determined, which for dimensional measurement, characterized in that the workpiece is at least temporarily measured in the period with the additional sensor or additional sensors in which the plurality Drehstellu taken and taken the radiographic images, in which so the X-ray source is turned on.

• - berührungslos misst, insbesondere ein optischer Sensor ist, wobei der zusätzliche Sensor so angeordnet ist und einen ausreichend großen Arbeitsabstand aufweist, so dass er in sämtlichen Durchstrahlungsbildern nicht erfasst wird oder
• - berührungslos misst, insbesondere ein optischer Sensor ist, wobei der zusätzliche Sensor so angeordnet ist und einen ausreichend großen Arbeitsabstand aufweist, so dass in sämtlichen Durchstrahlungsbildern zwischen Werkstück und zusätzlichem Sensor zumindest ein Pixel Abstand besteht, in dem das das Werkstück umgebende Medium wie Luft abgebildet ist oder
• - berührend misst, insbesondere ein taktiler oder taktil-optischer Sensor ist, wobei in sämtlichen Durchstrahlungsbildern die Bereiche (Pixel), in denen der zusätzliche Sensor abgebildet ist, in der Rekonstruktion nicht berücksichtigt werden, oder wobei in sämtlichen Durchstrahlungsbildern oder im Voxelvolumen die Bereiche (Pixel bzw. Voxel), in denen der zusätzliche Sensor abgebildet ist, durch geeignete Grauwerte, vorzugsweise Grauwert des das Werkstück umgebenden Mediums wie Luft, ersetzt werden.

In particular, the invention is characterized in that the additional sensor

• - Non-contact measures, in particular an optical sensor, wherein the additional sensor is arranged and has a sufficiently large working distance, so that it is not detected in all radiographs or
• - Non-contact measures, in particular an optical sensor, wherein the additional sensor is arranged and has a sufficiently large working distance, so that in all radiographic images between the workpiece and additional sensor at least one pixel distance, in which the surrounding the workpiece medium such as air is or
• is touching, in particular a tactile or tactile-optical sensor, wherein in all radiographic images the regions (pixels) in which the additional sensor is imaged are not taken into account in the reconstruction, or wherein in all radiographs or in the voxel volume the regions ( Pixel or voxel), in which the additional sensor is shown, are replaced by suitable gray values, preferably gray value of the medium surrounding the workpiece, such as air.

Preferably, it is provided that the additional sensor or sensors are completely or at least partially protected, at least at the times when the source is switched on, by a shield from at least the direct radiation of the source by arranging a corresponding shield and / or the additional shield Sensor is positioned accordingly.

It should also be emphasized that the computer tomograph and the additional sensor (s) form a coordinate measuring machine, in particular a multi-sensor coordinate measuring machine, wherein the computer tomograph and all additional sensors measure in a uniform coordinate system, computer tomography and additional sensors being preferably calibrated relative to one another.

The invention is also characterized in that the one or more additional sensors are positioned by means of a separate track axis or separate measuring axis for measuring the workpiece, preferably can be positioned in a position behind the shield.

• - mit dem Computertomografen ermittelte Messpunkte anhand der mit dem oder den zusätzlichen Sensoren ermittelten Messpunkte korrigiert werden und/oder
• - Oberflächenextraktionsverfahren, insbesondere dabei verwendete Schwellwerte oder Grauwerte der Voxel des Voxelvolumens, so angepasst werden, dass mit dem Computertomografen ermittelte Oberflächenmesspunkte in dem Bereich, in dem mit dem oder den zusätzlichen Sensoren Messpunkte vorliegen, mit diesen identisch sind oder festgelegte oder einstellbare maximal zulässige Abweichung zu diesen unterschreiten,

und vorzugsweise zwischen den mit dem oder den zusätzlichen Sensoren ermittelten Messpunkten interpoliert wird.In particular, the invention provides that measured data determined by the computer tomograph, ie radiographic images, voxel volumes and / or measuring points, are corrected on the basis of the measured data of the workpiece determined by the additional sensor (s), preferably by

• - Measured with the computer tomograph measuring points are corrected on the basis of the determined with the one or more sensors measuring points and / or
• Surface extraction methods, in particular threshold values or gray values of the voxels of the voxel volume, adapted in such a way that surface measurement points determined by the computer tomograph are identical with those in the area in which the additional sensor (s) are present or fixed or adjustable maximum permissible deviation fall short of these,

and preferably interpolated between the measuring points determined with the additional sensor (s).

For example, the aforementioned first coating refers to a preferably subsequent correction, such as in the WO2005119174A1 the applicant, in particular in claims 36 to 38 and 84 to 94 and related description called.

For example, the aforementioned second painting relates to a method which is also known as direct STL reconstruction, taking into account the data of the additional sensors. An exemplary solution gives the DE102006022103A1 the applicant.

A device according to the invention comprises a multisensor coordinate measuring machine for the dimensional measurement of a workpiece comprising a computer tomograph, comprising at least a radiation source (source), in particular an X-ray source, detector, in particular an X-ray detector, preferably a surface X-ray detector, and receiving the workpiece and in the beam path between the source and the detector rotating turntable, and having at least one tactile, optical and / or tactile-optical sensor (additional sensor), and is characterized in that the multi-sensor coordinate measuring machine (10) is formed, the workpiece (4) at least temporarily in the period with the additional sensor (16) or additional sensors to be measured, in which the multiple rotational positions are taken and the transmission images are recorded, in which therefore the X-ray source (1) is turned on, in particular a suitable control device (13).

In particular, the invention provides that a shield of the one or more additional sensors is provided and arranged in front of at least the direct radiation of the source.

In particular, the invention is characterized in that additional sensor is arranged on a separate axis or measuring axis and can be positioned to measure the workpiece, preferably in a position behind the shield can be positioned.

The subject of an independent invention is a method for examining an object by means of computed tomography, in particular dimensional measurement using surface measuring points.

From the DE 10 2005 039 422 a computed tomography method is known in which the object to be measured by means of computer tomography is positioned once by a displacement device on the turntable for rotation of the component between the x-ray source and the x-ray detector so that in all the rotational positions to be taken subsequently the measurement object does not detect the region detected by the x-ray detector leaves, in particular centrally arranged in this. For this purpose, radiographic images must first be recorded in advance in several rotational positions in order to determine whether the target would leave the detected area during later rotation. This is time consuming and also allows only a single position of the object for all rotational positions set together. Although this ensures that the object can be completely detected in all rotational positions, it is not possible to define a separate position for each rotational position. In particular, the position in the direction of the connecting line between the X-ray source and the X-ray detector, that is to say the present magnification or magnification, is not influenced. As a result, it is not possible to set the magnification or the magnification for each rotational position separately.

Another object of the present invention is therefore to separately determine the position of the medium computer tomography object to be measured with regard to the X-ray tube and the X-ray detector in such a way that the object does not leave the area detected by the X-ray detector in all rotational positions and a preliminary measurement can be dispensed with to save measuring time. Preferably, the magnification or the magnification for each rotational position should be set separately as large as possible, so that an accurate measurement, in particular compliance with the necessary for a measurement task structure resolution can be done.

To solve this, the invention provides that in each rotational position the respectively recorded radiographic image or the voxel volume reconstructed on the basis of the radiographic images already taken up to this rotational position is assessed and it is determined whether the object leaves the edge region of the area detected by the detector and If necessary, repositioning takes place in the middle of the area detected by the detector. This also results in the possibility of subsequently changing the magnification such that the object completely fills the detector, ie the maximum possible enlargement or magnification is present, by positioning the object as close as possible to the X-ray detector in the direction of the connecting line between the X-ray source and the X-ray detector. Provision is made for this purpose by corresponding displacement devices, such as measuring axes for positioning the object on the turntable causing the rotation about the axis of rotation or for positioning the turntable together with the object.

If in this case the necessary for the realization of a measurement task magnification and thus voxel size and the necessary structure resolution is not achieved, provides an embodiment of the invention that in the corresponding rotational position of the image scale is further increased and a plurality of laterally offset radiographic images are taken to the object or Completely grasp the workpiece.

The invention provides for the solution, a method for dimensional measurement and / or examination of the object interior of an object, preferably a workpiece or tool by means of computed tomography, in which in several rotational positions of the object receiving turntable whose axis of rotation radiographs of the object are recorded, preferably in at least rotational positions from 0 ° to 180 ° plus cone angle of the X-ray, more preferably 0 ° to 360 °, in particular 0 ° to 360 ° plus cone angle of the X-ray radiation, wherein the object between the X-ray source and surface X-ray diffraction detector, and wherein from the radiation images, a voxel volume is reconstructed, which is examined with respect to the internal structures and / or generated from the surface measuring points, before, characterized in that the position of the object between the X-ray source and X-ray detector in the direction the connection line between the X-ray source and the center of the X-ray detector and / or in the direction perpendicular to the connecting line and perpendicular to the direction of the axis of rotation for each rotational position separately, is automatically adjusted, so that the entire object or the portion to be measured of the object is imaged completely on the X-ray detector in each rotational position by determining the position of the object on the basis of the respectively recorded radiographic image or a plurality of previously recorded radiographic images or a voxel volume (real-time voxel volume) reconstructed from all or some radiographic images already recorded at the respective time possibly adjusted.

An adjustment of the position of the object or the portion of the object to be measured in the direction of the axis of rotation is usually before the actual measurement initially for all rotational positions in the selected magnification (position of the object between the X-ray source and X-ray detector in the direction of the line connecting the X-ray source and the center of the X-ray detector) such that the image of the object or the portion of the object to be measured in the direction of the axis of rotation does not leave the X-ray detector. An adjustment of the position is only necessary for each rotational position when the magnification is increased, that is, the object is moved closer to the X-ray source, as will be described below in a particular embodiment of the invention.

In particular, the invention is characterized in that the position of the object in the respective rotational position is adjusted so that the object does not leave the area detected by the X-ray detector by changing the position of the workpiece until a final radiographic image is taken, that the image of the object or of the object section to be measured completely contains.

Preferably, it is provided that the position of the object perpendicular to the connecting line between the X-ray source and X-ray detector, in particular in the direction of the axis of rotation and / or perpendicular to the direction of the axis of rotation, is changed so that in the respective direction of the distance of the image of the object or object details Edge of the radiographic image is the same size on both sides and the displacement in the direction of the connecting line between the X-ray source and X-ray detector is carried out so that the object is arranged as close to the X-ray source, so that preferably gives the maximum possible magnification.

The possible collision between workpiece and X-ray source and / or that the workpiece leaves the area (cone) detected by the detector and / or that the maximum permissible cone angle is exceeded can have a limiting effect here. The indication of a maximum permissible cone angle can be made in order to limit artefacts, such as cone beam artifacts, resulting therefrom, in order to achieve the structural resolution and / or measurement uncertainty necessary for the performance of the measurement task.

In particular, the invention is characterized in that from the real-time voxel volume, the 3D shape and position of the object is at least determined and from the expected position of the image of the object in the radiographic image of each next rotational position is calculated and based on the position of the object set becomes.

• - weder überstrahlt noch zu dunkel ist und/oder
• - optimalen SNR (Signal to Noise Ratio) und/oder
• - optimalen CNR (Contrast to Noise Ratio) aufweist,

oder dass das aus allen oder den bis dahin aufgenommenen endgültig aufgenommenen Durchstrahlungsbildern rekonstruierte Echtzeit-Voxelvolumen bzw. Voxelvolumen

• - optimalen SNR (Signal to Noise Ratio) und/oder
• - optimalen CNR (Contrast to Noise Ratio) aufweist,

oder dass die Durchstrahlungsbilder so optimiert werden, dass die für die Messaufgabe relevanten Oberflächenmesspunkte und/oder die hieraus verknüpften Elemente und/oder die hieraus bestimmen geometrischen Merkmale eine möglichst hohe Messpräzision und/oder Messrichtigkeit aufweisen.Preferably, the invention provides that in the respective rotational position after the recording of a first transmission image and the displacement of the position of the object, the final transmission image for this rotational position is taken, preferably by evaluating the gray values of the first recorded in this rotational position radiographic image, the parameters of the X-ray source , in particular acceleration voltage and / or X-ray tube current, and / or number of image averaging and / or integration time of the respective transmission image are adapted such that the transmission image or normalized transmission image or normalized logarithmic transmission image or normalized logarithmic filtered transmission image finally recorded in the respective rotational position

• - neither overexposed nor too dark and / or
• Optimal SNR (Signal to Noise Ratio) and / or
• - has optimal CNR (Contrast to Noise Ratio),

or that of all or those recorded until then Radiographic images reconstructed real-time voxel volumes or voxel volumes

• Optimal SNR (Signal to Noise Ratio) and / or
• - has optimal CNR (Contrast to Noise Ratio),

or that the radiographic images are optimized in such a way that the surface measurement points relevant for the measurement task and / or the elements linked thereto and / or the geometric features determined therefrom have the highest possible precision and / or measurement accuracy.

The measurement accuracy denotes the systematic deviations of the mean value of an infinitely large sample from a reference value, which can be defined as the correct measured value.

The parameters here are understood to mean the parameters of the x-ray source, in particular acceleration voltage and / or x-ray tube current, but also the number of image averages and the integration time of the respective radiographic image as well as the number of projections per measurement. According to the invention, further measurement parameters can also be adapted per rotational position, such as, for example, the beam filter used or software corrections such as image filters.

The adaptation of the parameters per rotational position is described below as a self-inventive idea. The aim of the adaptation is to achieve an optimal measurement with regard to measurement precision and / or measurement uncertainty. For this purpose, the CNR should preferably be optimized in the voxel volume or final voxel volume, in particular maximized. Alternatively, the SNR can also be used. Since only radiographic images up to the respectively considered rotational position are present when adjusting the parameters, initially only one real-time voxel volume reconstructed using this can be calculated and evaluated. Alternatively, even here, a final voxel volume (also referred to as voxel volume) can be calculated and evaluated by estimating the missing radiographic images by means of simulation. If, in the course of the parameter adaptation, a reconstruction is entirely dispensed with, an alternative evaluation of the respective radiographic image in the directly recorded form or after normalization or after the subsequent logarithmization or after the subsequent filtering with respect to SNR or CNR can be undertaken ,

It is preferably provided that the position set in the reconstruction, in particular the SOUV vectors assigned to the respective radiographic image, is taken into account for each rotational position for the recording of the final radiographic image.

According to one particularly noteworthy proposal, it is provided that the position set for each rotational position for the recording of the final radiographic image and the changed parameters set for each rotational position for recording the final radiographic image are taken into account when performing an artefact correction, in particular empirical artifact correction or autocorrection or virtual autocorrection become.

• - die Relativposition zwischen Objekt, Röntgenquelle und Röntgendetektor parallel zur Verbindungslinie zwischen Röntgenquelle und Röntgendetektor wird so angepasst, dass die resultierende Strukturauflösung die vorgegebene mindestens notwendige Strukturauflösung erreicht, und dass insbesondere die dazu notwendige Voxelgröße bzw. Bereich der Voxelgröße eingehalten wird, wobei bevorzugterweise der sich dabei ergebende Kegelwinkel einen vorgegebenen maximal zulässigen Kegelwinkel nicht überschreitet, jedoch nur maximal so weit, kurz bevor das Objekt mit der Röntgenquelle kollidieren würde, und
• - in der endgültigen Position in Richtung der Verbindungslinien zwischen Röntgenquelle und Röntgendetektor wird bestimmt, ob das Objekt bzw. der zu messende Abschnitt des Objekts vollständig auf dem Röntgendetektor abgebildet wird und falls dies nicht der Fall ist, werden endgültige Durchstrahlungsbilder in mehreren senkrecht zur Richtung der Verbindungslinien zwischen Röntgenquelle und Röntgendetektor versetzten Positionen, insbesondere in Richtung der Drehachse und/oder senkrecht zur Richtung der Drehachse, so aufgenommen, dass das Objekt bzw. der zu messende Abschnitt des Objekts vollständig erfasst wird, wobei vorzugsweise die versetzt aufgenommenen Durchstrahlungsbilder zu einem resultierenden endgültigen Durchstrahlungsbild zusammengesetzt werden,

wobei besonders bevorzugt für alle Drehstellungen ein resultierendes endgültiges Durchstrahlungsbild identischer Größe erzeugt wird, indem gegebenenfalls fehlende Bereiche mit Grauwerten aufgefüllt werden, die dem das Objekt umgebendem Medium entsprechen.It should also be emphasized that the voxel size and / or structure resolution resulting from the position of the object in the direction of the connecting lines between the x-ray source and the x-ray detector is determined and compared with a predetermined value or range for the permissible voxel size or pattern resolution and, in the case of deviations from this Value or range, the following steps are performed before the final or final radiographs for the respective rotational position are recorded:

• - The relative position between the object, X-ray source and X-ray detector parallel to the line connecting X-ray source and X-ray detector is adjusted so that the resulting structure resolution reaches the predetermined minimum necessary structure resolution, and that in particular the necessary voxel size or range of voxel size is maintained, preferably resulting cone angle does not exceed a predetermined maximum allowable cone angle, but only a maximum so far, just before the object would collide with the X-ray source, and
• in the final position in the direction of the connecting lines between the X-ray source and the X-ray detector, it is determined whether the object or the portion of the object to be measured is completely imaged on the X-ray detector and, if this is not the case, final radiographic images in several directions perpendicular to the direction of Connecting lines between the X-ray source and the X-ray detector offset positions, in particular in the direction of the axis of rotation and / or perpendicular to the direction of the axis of rotation, recorded so that the object or the portion of the object to be measured is completely detected, preferably the offset recorded radiographic images to a final resulting Radiographic image are assembled

wherein particularly preferably for all rotational positions, a resulting final radiographic image of identical size is generated by possibly filling missing regions with gray values which correspond to the medium surrounding the object.

Compliance with a necessary structure resolution for a measurement task can be assessed, for example, by examining the cutoff frequency resulting from the MTF of the measurement, that is to say the convolution of all the structure resolution determining and known variables including the aperture of the X-ray detector and the focal spot size of the X-ray source this is greater than the necessary structure resolution. For the voxel size resulting, inter alia, from the magnification, this can mean that not only does it have to fall below a certain maximum permissible value, but also that it must not fall below a certain minimum permissible value. The latter results from the associated deteriorated signal-to-noise ratio (SNR) with decreasing voxel size. Falling below a maximum allowable cone angle, with which the image of the object or workpiece is performed on the X-ray detector, resulting from the increasing cone angle with the influence of artifacts, such as cone beam artifacts, and the associated deteriorated structural resolution.

The subject matter of an independent invention is a method for examining an object by means of computer tomography, in particular dimensional measurement using surface measuring points, in which the parameters or measuring parameters are adapted to each rotational position.

According to the state of the art, identical measurement parameters, in particular parameters of the x-ray tube, are used for all rotational positions in a computed tomography. This results in the disadvantage that the measurement parameters can not be adapted to the different irradiation conditions per rotational position, in particular maximum transmission length or distribution of the available transmission lengths, with respect to an accurate measurement, in particular measurement precision or measurement uncertainty.

Another object of the present invention is to adjust the parameters or measurement parameters per rotational position so that a particularly accurate measurement takes place.

To solve the invention provides that the parameters or measurement parameters are set separately for each rotational position. The optimization criterion here is the achievable measurement precision or measurement uncertainty or measurement accuracy, which is preferably achieved by optimal, that is to say maximum CNR in the reconstructed voxel volume. An alternative optimization criterion is maximum SNR or CNR in the respective radiographic image.

• - weder überstrahlt noch zu dunkel ist und/oder
• - optimalen SNR (Signal to Noise Ratio) und/oder
• - optimalen CNR (Contrast to Noise Ratio) aufweist,

oder so angepasst werden, dass das aus allen oder den bis dahin aufgenommenen Durchstrahlungsbildern rekonstruierte Echtzeit-Voxelvolumen bzw. Voxelvolumen

• - optimalen SNR (Signal to Noise Ratio) und/oder
• - optimalen CNR (Contrast to Noise Ratio) aufweist,

oder dass die Durchstrahlungsbilder so optimiert werden, dass die für die Messaufgabe relevanten Oberflächenmesspunkte und/oder die hieraus verknüpften Elemente und/oder die hieraus bestimmen geometrischen Merkmale eine möglichst hohe Messpräzision und/oder Messrichtigkeit aufweisen.The invention provides a solution for a method for dimensional measurement and / or examination of the object interior of an object, preferably a workpiece or tool by means of computed tomography, are taken in several rotational positions of the object receiving turntable about the axis of rotation radiographic images of the object, preferably in at least rotational positions from 0 ° to 180 ° plus the cone angle of the X-ray radiation, particularly preferably 0 ° to 360 °, in particular 0 ° to 360 ° plus cone angle of the X-radiation, wherein the object is between the X-ray source and X-ray detector having a pronounced surface, and wherein from the transmission images Voxel volume is reconstructed, which is examined with respect to the internal structures and / or generated from the surface measuring points, before that it is characterized in that for each rotational position, the parameters of the X-ray source, in particular acceleration voltage and / or Rön tgenröhrenstrom, and / or number of image averaging and / or integration time of the respective radiographic image to be adjusted so that the recorded in the respective rotational position radiographic image or normalized radiographic image or normalized logarithmic radiographic image or normalized logarithmic filtered radiographic image

• - neither overexposed nor too dark and / or
• Optimal SNR (Signal to Noise Ratio) and / or
• - has optimal CNR (Contrast to Noise Ratio),

or be adapted in such a way that the real-time voxel volume or voxel volume reconstructed from all or the transmission images taken up to then

• Optimal SNR (Signal to Noise Ratio) and / or
• - has optimal CNR (Contrast to Noise Ratio),

or that the radiographic images are optimized in such a way that the surface measurement points relevant for the measurement task and / or the elements linked thereto and / or the geometric features determined therefrom have the highest possible precision and / or measurement accuracy.

It should be emphasized that the assessment of the overexposure makes sense only on the non-normalized image.

As already explained above, the parameters here are understood to mean the parameters of the x-ray source, in particular acceleration voltage and / or x-ray tube current, but also the number of image averages and the integration time of the respective radiographic image. According to the invention, further measurement parameters can also be adapted per rotational position, such as, for example, the beam filter used or software corrections such as image filters. With regard to the implementation of the adaptation, reference is made to previously explained relationships. The number of rotational steps can also be adjusted according to the invention.

After a particularly emphasized proposal is provided that in the respective rotational position first a first radiation image is recorded and preferably gray values, in particular SNR or CNR of the respective radiographic image or real-time voxel volume or voxel volume are evaluated, then adjusted the parameters and then recorded the final radiographic image or that the parameters are determined and adjusted on the basis of a simulation of the respective radiographic image or real-time voxel volume or voxel volume, and preferably the final radiographic image is taken directly.

Preferably, the invention provides that the parameters are determined and set on the basis of a previously created "Detectability Map" and preferably directly the final radiographic image is taken.

The creation of a so-called "Detectability Map" is described below as a self-inventive idea. The method described there can be combined with the idea described here.

The subject matter of an independent invention is a method for examining an object by means of computed tomography, in particular dimensional measurement using surface measuring points, in which the magnification per rotation position is optionally adjusted.

In accordance with the prior art, a fixed trajectory for the rotation or movement of the object to be measured by means of computer tomography is determined in advance. This includes either a fixed magnification or a trajectory in which the magnification varies accordingly. Here, the shape of the object is taken into account only conditionally. In particular, the magnification and the trajectory are determined so that the object or the portion of the object to be measured is imaged completely on the X-ray detector in all rotational positions.

In the WO2005119174A1 the applicant is a method is described in which is measured at an increased magnification, wherein the object is completely detected that perpendicular to the line connecting the X-ray source, more accurate focal point of the X-ray source, and X-ray detector, more precisely center of the detector, staggered several radiographic images are taken. But here, too, an identical magnification is used for all rotational positions.

However, the problem is not solved that, depending on the shape of the object, a maximum or optimum voxel size and thus a predetermined or the best possible measurement precision or measurement uncertainty and the structural resolution required for the measurement task are achieved. This is therefore another object of the present invention.

To solve the invention, the magnification, so adjust the position of the object or to be measured portion of the object in the direction of the connecting lines between the X-ray source and X-ray detector per rotational position accordingly. If complete detection on the X-ray detector by a transmission image is therefore not possible, a plurality of transmission images offset perpendicularly to the connecting line are recorded and possibly combined. The increased image size resulting for this case is achieved in one embodiment of the invention for all rotational positions by refilling missing regions with gray values of the medium surrounding the object.

• - die Relativposition zwischen Objekt, Röntgenquelle und Röntgendetektor parallel zur Verbindungslinie zwischen Röntgenquelle und Röntgendetektor wird so angepasst, dass die resultierende Strukturauflösung die vorgegebene mindestens notwendige Strukturauflösung erreicht, und dass insbesondere die dazu notwendige Voxelgröße bzw. Bereich der Voxelgröße eingehalten wird, wobei bevorzugterweise der sich dabei ergebende Kegelwinkel einen vorgegebenen maximal zulässigen Kegelwinkel nicht überschreitet, jedoch nur maximal so weit, kurz bevor das Objekt mit der Röntgenquelle kollidieren würde, und
• - in der endgültigen Position in Richtung der Verbindungslinien zwischen Röntgenquelle und Röntgendetektor wird bestimmt, ob das Objekt bzw. der zu messende Abschnitt des Objekts vollständig auf dem Röntgendetektor abgebildet wird und falls dies nicht der Fall ist, werden endgültige Durchstrahlungsbilder in mehreren senkrecht zur Richtung der Verbindungslinien zwischen Röntgenquelle und Röntgendetektor versetzten Positionen, insbesondere in Richtung der Drehachse und/oder senkrecht zur Richtung der Drehachse, so aufgenommen, dass das Objekt bzw. der zu messende Abschnitt des Objekts vollständig erfasst wird.

The invention provides a solution for a method for dimensional measurement and / or examination of the object interior of an object, preferably a workpiece or tool by means of computed tomography, are taken in several rotational positions of the object receiving turntable about the axis of rotation radiographic images of the object, preferably in at least rotational positions from 0 ° to 180 ° plus the cone angle of the X-ray radiation, particularly preferably 0 ° to 360 °, in particular 0 ° to 360 ° plus cone angle of the X-radiation, wherein the object is between the X-ray source and X-ray detector having a pronounced surface, and wherein from the transmission images Voxel volume is reconstructed, which is examined with respect to the internal structures and / or generated from the surface measurement points, before that thereby characterized in that each vertex size and / or structure resolution resulting from the position of the object in the direction of the connecting lines between X-ray source and X-ray detector is determined and compared with a predetermined value or range for the permissible voxel size or structure resolution and in case of deviations from the value or The following steps are performed before the final or final radiographic images for the respective rotational position are recorded:

• - The relative position between the object, X-ray source and X-ray detector parallel to the line connecting X-ray source and X-ray detector is adjusted so that the resulting structure resolution reaches the predetermined minimum necessary structure resolution, and that in particular the necessary voxel size or range of voxel size is maintained, preferably resulting cone angle does not exceed a predetermined maximum allowable cone angle, but only a maximum so far, just before the object would collide with the X-ray source, and
• in the final position in the direction of the connecting lines between the X-ray source and the X-ray detector, it is determined whether the object or the portion of the object to be measured is completely imaged on the X-ray detector and, if this is not the case, final radiographic images in several directions perpendicular to the direction of Connecting lines between the X-ray source and the X-ray detector offset positions, in particular in the direction of the axis of rotation and / or perpendicular to the direction of the axis of rotation, so that the object or the section of the object to be measured is completely detected.

It should also be emphasized that the transmitted radiation images recorded per rotational position are combined to form a final radiographic image resulting in each rotational position and a resulting final radiographic image of identical size is generated for all rotational positions by possibly filling up missing regions with gray values corresponding to the medium surrounding the object in which the voxel volume is reconstructed from all the resulting final transmission images.

The subject matter of an independent invention is a method for examining an object by means of computed tomography, in particular dimensional measurement using surface measuring points, the measuring points being assigned a noise, a measuring precision and / or a component for measuring uncertainty.

It is also an object of the invention to apply the method, instead of for computed tomography, to further measuring methods in which there is an intensity distribution of a measuring signal at an edge transition or media transition. This is the case for example with optical sensors, in particular image processing sensors, optical distance sensors such as laser distance sensors, preferably according to the principle of the foucault cutting edge, chromatic or chromatic-confocal distance sensors or confocal distance sensors.

According to the prior art, the noise, the measurement precision or other components for measuring uncertainty measured by computer tomography measuring point (surface measuring point), here referred to as dimensional noise, by elaborate simulation of the influencing variables or by assessing the scatter in repetitive measurements according to VDI / VDE Guideline 2630-2.1. Both methods are extremely time consuming.

A method for assigning a local quality measure is in the WO2016 / 066265 in which only the global contrast (even if referred to as "local signal-to-noise ratio") is taken into account and the measurement time is to be optimized. The global contrast in the present invention is intended to denote the gray value difference between the voxels on both outer sides of an edge transition between the object and the surrounding medium, but not in the immediate vicinity of the measurement point. The procedure of WO2016 / 066265 Although it is proposed to serve as a basis for the determination of the measurement uncertainty or measurement accuracy. However, a concrete solution for this is not specified. The quality measures specified there are not suitable for specifying the noise, the measurement precision or other components for measuring uncertainty. In the WO2016 / 042105 In addition, a direction-dependent (normal and tangential) quality measure is described.

A further object of the present invention is therefore to specify a less time-consuming and accurate method with which a measurement point can be assigned a noise, a measurement precision or another component for measuring uncertainty due to random measurement deviations during the computed tomographic measurement of an object.

To solve this, the invention provides that, for this purpose, the edge location is interpolated or extrapolated from the noise of the voxel gray values contributing to the edge location determination and from the gray value difference of the voxel gray values contributing to the edge location determination, in particular the gray value difference of the voxel gray values (local contrast, inner edge voxels), the dimensional noise of the coordinates of a measurement point is calculated, which is used as noise, measurement precision or component of the measurement uncertainty.

The invention provides a solution for a method for dimensional measurement of an object, preferably a workpiece or tool by means of computed tomography, are taken in several rotational positions of the object receiving turntable about the axis of rotation radiographic images of the object, preferably in at least rotational positions of 0 ° to 180 ° plus cone angle of the X-ray radiation, particularly preferably 0 ° to 360 °, in particular 0 ° to 360 ° plus cone angle of the X-ray radiation, wherein the object is between the X-ray source and X-ray detector having a pronounced surface, and wherein a voxel volume is reconstructed from the transmission images, from the Surface measuring points (measuring points) are generated, wherein the voxel volume has a plurality of voxels whose center is assigned a gray value (Voxelgrauwert) before that is characterized in that the noise and / or the measurement precision and / or at least one component of the Measurement uncertainty of a measuring point is calculated from the noise of the voxel gray values, which is assigned to the voxels (inner edge voxels) surrounding the respective measuring point and used for the edge location detection, in particular directly adjacent to the respective measuring point. In a particular embodiment, it is also provided that in addition the edge length of the corresponding voxels and / or the differences of the voxel gray values of the corresponding voxels are included in the calculation.

The noise of a measuring point is the variation of the position of the measuring point in space, which is to be assumed by random deviations in the measurement. The noise of the voxel gray values and the noise of the gray values of the pixels of a transmission image (image gray values) denote the corresponding gray value variations due to random measurement deviations. Variations also result from systematic measurement deviations. However, these should preferably be corrected before the noise of the invention is treated.

The inner edge voxels are the voxels that lie in the vicinity of the location of the respective measuring point and / or are used for the calculation of this. These can be the two voxels directly adjacent to the measurement point, but also several voxels depending on the edge location detection method used.

For example, after a simple edge location detection process, a 2x2x2 environment may be considered, ie the eight voxels in the immediate neighborhood between which the location of the measurement point is subvoxel accurate by simple interpolation. This also depends on the voxel size. The voxels comprising the edge course are also referred to here as edge voxels. The gray value difference of the voxels (outer edge voxels) present between the two outer regions of the edge can also be referred to as global contrast. The gray value difference of the inner edge voxels is referred to here as local contrast. If the local contrast is greater than or equal to the global contrast, the voxels are very likely too large and there is undersampling, or the image is too blurred or the pattern resolution too low, so the resulting component for measurement uncertainty, measurement precision or noise is very high becomes big or even infinite.

The measurement precision is here assumed to be proportional to the ratio of voxel size and local CNR (contrast to noise ratio). The local CNR results from the ratio of local contrast and noise. Therefore, the basic idea of the present invention is to assign a noise to the inner edge voxels, which thus have the local contrast, from which the dimensional noise for the measuring point is then derived.

The voxel size or voxel edge length is determined by the voxel center distance resulting from the pixel pitch of the x-ray detector and the magnification. The voxel or the voxel center is assigned the reconstructed gray value, which is also referred to here as the voxel center value or voxel gray value.

In particular, the invention is characterized in that the calculation provides that along the vertical (normal) to the object surface by the measuring point or for all three spatial directions separately present the gray values of the voxel gray values of, in possibly the respective spatial direction, adjacent inner edges -Voxel, is converted into a dimensional noise value, which enters into the noise or the measurement precision or the component of the measurement uncertainty of the measuring point, wherein preferably the gray value curve of the voxel gray values along the vertical or respective spatial direction around the measuring point, in particular its increase in the measuring point , particularly preferably the voxel edge length, is taken into account.

The voxel edge length is implicitly included in the calculation, as well as the increase in the gray value curve around or in the measuring point. The gray-scale difference between the inner edge voxels becomes local contrast in this context designated. The larger the voxel edge length and the lower the local contrast, the flatter the gray value curve becomes and, as a result, the larger the dimensional noise of the measuring point becomes. The voxel size has no influence on the slope of the gray value curve. The local contrast, however, increases with increasing voxel size.

However, the "lever length" increases, which in turn results in increased uncertainty. The lower the noise of the voxel gray values used to determine the edge location, the lower the uncertainty and the smaller the dimensional noise of the measurement point.

The viewing of the vertical is effected, for example, in a method in which a normal to the expected surface, for example derived from setpoint data such as a CAD model, is formed and the gray value profile along the normal is examined by the measuring point. A space-direction dependent method is, for example, the well-known marching cube algorithm.

If a contribution to the noise or the measurement precision or the component of the measurement uncertainty of the measurement point is calculated for each spatial direction, the contributions must be offset to a total value that represents a noise or a measurement precision or a component of the measurement uncertainty in the room. The summary of the contributions, for example, by adding the squares and subsequent rooting, ie a spatial overlay, take place. The exact design depends on the edge location detection method used and may also be more complicated. The calculation of the noise, the measurement precision or the component of the measurement uncertainty of the measurement point is preferably carried out by quadratic error propagation on the basis of the mathematical relationship underlying the method used for determining the surface points.

For the calculation of the dimensional noise value, it is preferably provided that the intersections of the envelopes around the inner edge voxels provided with the noise values are used with the local threshold value associated with the respective measuring point or with the global threshold value.

In particular, the invention is distinguished by the fact that the uncertainty of the threshold value is taken into account in the calculation of the dimensional noise value, preferably the points of intersection with the envelope furthest in relation to the measuring point are used.

Preferably, the invention provides that as the noise of the inner edge voxels, the maximum or average variation (maximum or average noise) of the gray level voxel gray values along the respective spatial direction or along the normal to the object surface through the measurement point with respect to a smoothed or filtered Gray value course (compensation curve) is used, wherein preferably the gray value profile is considered at least in the region of the material transition, particularly preferably the gray value curve in the region of the material transition and / or the gray value curve in a lying in both directions after the material transition range determinable size.

The smoothed or filtered gray value curve is calculated from the measured gray value curve and is a compensation curve, from which the respective measuring point is determined with subvoxel accuracy, namely between the inner edge voxels, using the threshold value. However, other methods of edge detection are contemplated in which additional noise effects are created by the method that must be considered and transmitted to the inner edge voxels. The calculation of the smoothed or filtered gray value curve can also be carried out by different methods, for example by linear fit or by Gaussfit.

According to the invention, the inner edge voxels are assigned the noise which, as described above, is present along the gray value profile through the measuring point. It is assumed that the noise does not change abruptly, ie adjacent noise can be assigned to the same voxels. The variation of the gray values from voxel to voxel is thus assumed to be the noise of a voxel in the gray scale range per se. Two alternative methods of estimating the noise of a voxel are described below.

It should also be emphasized that the noise of the inner edge voxels of one or more adjacent measurement points is included for the noise of the inner edge voxels of a measurement point or that for the noise and / or the measurement precision and / or at least one component of the measurement uncertainty of a measurement point the noise and / or the measurement precision and / or at least one component of the measurement uncertainty of one or more adjacent measurement points is included.

The invention is also characterized in that, as the noise of the inner edge voxels, the variation (noise) of the voxel gray values of the inner edge voxels is used the reconstruction of simulated radiographic images.

In particular, the noise assumed for at least some measurement parameters and component properties is transmitted to the voxel gray values by means of simulation. Preferably, Monte Carlo simulations are used in which the input parameters are varied and the variation (noise) of the output values (eg, voxel gray values) is calculated.

According to a particularly preferred proposal, the normalized, logarithmized and filtered, preferably high-pass filtered, image data are reconstructed.

In particular, the invention is characterized in that the gray values of the pixels of the radiographic images (image gray values) are respectively assigned the associated irradiated length and preferably a compensation characteristic is formed which approximates the relationship between image gray values and irradiated length, and that each irradiated length is the Then, as the noise of the inner edge voxels, the maximum or mean or weighted variation (noise) of the associated image gray values of all the irradiated lengths associated with the image gray values of the pixels corresponding to the image gray values Reconstruction of the respective inner edge voxels were used.

The transmission length associated with the image gray values is obtained using the reconstructed volume data and backward projection, or roughly by a target data set of the object and forward projection. Under certain circumstances, different image gray values are assigned identical transmission lengths, which results in a variation of the image gray values which are assigned to a transmission length. This variation is proportionally transferred to all voxels in which the corresponding transmission length, or exactly the pixel that has been assigned to this transmission length, is used. Since each voxel is reconstructed from a plurality of pixels, the noise of all pixels used for its reconstruction must be taken into account as the noise of the voxel. These pixels are assigned different transmission lengths. The noise of pixels with transmission lengths, which occur more frequently, are thus weighted more heavily.

This solution according to the invention initially applies only to monomaterials. In the case of multi-material objects, a linear combination of the different materials that are included in the respective transmission length must be taken into account.

According to a particularly preferred proposal, the gray values of the pixels of the radiographic images (image gray values) are not the image data recorded directly with the X-ray detector, but the normalized, logarithmized and filtered, preferably high-pass filtered, image data are used.

Systematic measurement deviations are particularly preferably corrected before the assignment of the associated irradiated length to the image gray values.

It is also an object of the present invention to advantageously use the method described above in other measuring methods in which there is an intensity distribution of a measuring signal at an edge transition or media transition. In the above-mentioned optical sensors such as image processing sensors or optical distance sensors, such as laser distance sensors, preferably according to the principle of foucault cutting edge, chromatic or chromatic-confocal distance sensors or confocal distance sensors there is usually no voxel volume. The location of the surface points is determined from intensity distributions at the edge transition present in 1D or 2D. Image processing sensors have 2D image data, and so-called search rays that are laid through the edge give the intensity of the pixels of the image over the edge. Alternatively, the edge detection takes place in the context of a so-called contour image processing. For each edge measurement point determined therefrom, an intensity profile of the image pixels can be derived from the image data via the edge. Corresponding intensity profiles can also be generated with optical distance sensors. If these detect the surface only punctiform or linear, a measuring movement over the edge is carried out for this purpose. In all cases, the observation of the intensity gradients of the pixel gray values over the edge corresponds to the gray value curves of the voxels in the case of computed tomography.

To solve the above-mentioned considerations are limited by one spatial direction or in the case of separate consideration of the spatial directions limited only to the available two spatial directions. Thus, a gray value curve of the edge pixels is considered instead of the edge voxel.

Preferably, the invention thus provides that the method is applied to any measuring method in which an intensity distribution of a measuring signal is present at an edge transition or media transition or can be generated where pixel data (pixel gray values) is considered instead of voxel data (voxel gray values)

The subject matter of an independent invention is a method for examining an object by means of computed tomography, in particular dimensional measurement using surface measuring points, in which pre-radiation images are recorded before the actual measurement.

A known method for adjusting the parameters provided for a computer tomography describes the EP2665035A2 the applicant. Therein, the adjustment is made with respect to the image quality of one or more pre-recorded radiographic images. Here, the measurement parameters are adjusted so that in particular the contrast in the radiographic images is optimized. Since several radiographic images are considered separately here, an optimization can not be carried out simultaneously for all radiographic images.

Of the DE102017100594.3 The Applicant is a method in which the optimization of the measurement parameters with respect to the measurement precision, measurement accuracy, measurement uncertainty and measurement accuracy is done. Here, a complex model equation of the entire measurement chain is required.

A further object of the present invention is therefore to provide a simple method for determining suitable measurement parameters for a computer tomography, in which all radiographic images are taken into account.

To solve the invention provides that from all previously recorded radiographic images (Vorabdurchstrahlungsbilder) a so-called "detectability map" is determined, which detects the transmission ratios and thus attenuations from all available transmission lengths during the complete rotation in the actual actual computed tomography.

• - mit einem Satz an vorgegebenen Bedingungen verglichen und gegebenenfalls eine Rückmeldung an den Bediener erfolgt, wenn eine oder mehrere Bedingungen nicht eingehalten werden, und/oder
• - aus der Detectability Map die vorgegebenen Bedingungen erfüllende und vorzugsweise optimale Messparameter für die eigentliche computertomografische Messung bestimmt werden.

The invention provides for the solution a method for the dimensional measurement of an object, preferably a workpiece or tool by means of computer tomography, are taken in several rotational positions of the object receiving turntable about the axis of rotation radiographic images of the object, preferably in at least rotational positions of 0 ° to 180 ° plus cone angle of the X-ray radiation, particularly preferably 0 ° to 360 °, in particular 0 ° to 360 ° plus cone angle of the X-ray radiation, wherein the object is between the X-ray source and X-ray detector having a pronounced surface, and wherein a voxel volume is reconstructed from the transmission images the surface measuring points (measuring points) are generated, wherein the voxel volume has a plurality of voxels whose center is assigned a gray value (voxel gray value), characterized in that before the actual measurement radiographic images of the object in rotational positions from 0 ° to 180 ° plus cone angle of the X-ray radiation (Vorabdurchstrahlungsbilder) are recorded and from these a so-called "Detectability Map" is determined, and

• - compared with a set of given conditions and, where appropriate, feedback to the operator if one or more conditions are not met, and / or
• - From the Detectability Map the predetermined conditions fulfilling and preferably optimal measurement parameters for the actual computer tomographic measurement are determined.

The recording of the preliminary radiation images may be, for example, an image acquisition as in the case of the high-speed tomography already described above. The "Detectability Map" detects the transmission ratios occurring during the pre-measurement on the basis of the gray values occurring in the radiographic images and thus attenuations, and serves the estimation of the workpiece transmission and thus the derivation of the optimal measurement parameters with respect to at least one of the variables to be optimized noise, structure resolution, precision measurement, measurement accuracy and / or measurement uncertainty of the measurement points or features for the subsequent actual measurement. To make a complete estimate, at least one reconstruction would be necessary. Alternatively, however, it is also intended to realize the optimization with respect to the noise (SNR or CNR) present in the transmission images. Then the assessment can only be made on the radiographs. The previously mentioned measurement parameters include, for example, the magnification to be selected in conjunction with the measurement range available thereby, the distance between the radiation emanating the measuring radiation spot (focal spot) of the X-ray source and the detector, the acceleration voltage of the X-ray tube, the current and thus the power of the X-ray tube, the Prefilter to be used (beam filter), the number of image averaging and the integration time per radiographic image.

The step of optimizing the measurement parameters is preceded by the fact that the existing conditions in the transmission images, ie the "detectability map", are compared with a set of predetermined conditions (boundary conditions). Predefinable conditions are, for example, the intensity limits of the gray values in the transmission images in order to avoid overexposure and underexposure, the available measuring range resulting from the Measurement required resulting required minimum structure resolution and the maximum allowable measurement time. If these boundary conditions are not met, feedback is preferably sent to the operator, and suitable measurement parameters are derived which fulfill the given conditions, preferably optimal measurement parameters. From the measurement parameters set during the recording of the "Detectability Map" and the radiographic images thus determined, a relationship is thus formed which predicts how the radiographic images change with changed measurement parameters and whether the given conditions are then met. Optimum measurement parameters are present when the given conditions are optimally fulfilled, ie, for example, the smallest possible measurement uncertainty results, without exceeding a predetermined maximum measurement time, for example. If several conditions are to be optimized, a quality function is formed from the conditions to be optimized and optimized.

In particular, the invention is characterized in that one or more transmission length images are simulated from the preliminary radiation images and the target data such as the CAD model of the object or from the preliminary voxel volume reconstructed from the preliminary radiation images, the measurement parameters being used for the actual measurement in the simulation ,

The use of the measurement parameters for the actual measurement here means that the simulation is carried out with different measurement parameters and the or the transmission length images are then examined with regard to the quantities to be optimized. By transmission length image is meant here the simulated transmission image resulting from a forward projection on the desired data or on the pre-voxel volume. The simulation can be done in several rotational positions resulting in multiple radiographic length images.

Preferably, it is provided that it is detected during the recording of Vorabdurchstrahlungsbilder per rotational position, if the object leaves the area detected by the X-ray detector and optionally the relative position between the object and X-ray source and X-ray detector in one, two or all three spatial directions is adjusted so that the Object is completely detected, with the resulting relative positions are stored in the form of the SOUV vectors.

In particular, the invention is characterized in that the relative position between object and X-ray source and X-ray source (SOUV vectors) set during the recording of the preliminary radiation images are stored and used for the actual measurement or based on the stored SOUV vectors SOUV vectors can be set for the actual measurement taking into account the ideal magnification set for the actual measurement.

The stored SOUV vectors serve as starting values for the actual measurement and identify the object position, as already described for rapid tomography. Should the o.g. Simulation that the resulting from the stored SOUV vectors magnification is not sufficient to achieve an optionally predetermined target value for a size, such as the structure resolution is also provided to adjust the SOUV vectors and, for example, set a magnification, in which the object would leave the area detected by the X-ray detector. Nevertheless, in order to capture the object completely, laterally offset images of the object take place.

• - ideale Integrationszeit je Durchstrahlungsbild, vorzugsweise indem weder zu helle noch zu dunkle Pixel auftreten, und/oder
• - ideale Anzahl der Bildmittelungen und/oder
• - idealer Abbildungsmaßstab

für die eigentliche Messung bestimmt werden und/oder Größen für die Beurteilung der Signalqualität wie minimales, mittleres oder maximales CNR (Contrast to noise ratio) bestimmt werden, aus denen sich zumindest eine der zu optimierenden Größen ergibt.Preferably, the invention provides that from the Vorabdurchstrahlungsbildern or from the transmission length images of the or the measurement parameters

• - Ideal integration time per radiographic image, preferably by neither too bright nor too dark pixels occur, and / or
• - ideal number of image averages and / or
• - ideal magnification

be determined for the actual measurement and / or sizes for the assessment of the signal quality as minimum, average or maximum CNR (Contrast to noise ratio) are determined, which results in at least one of the sizes to be optimized.

In the case of using the pre-radiation images to determine the ideal integration time, the brightest and darkest pixels must be viewed and the integration time changed so that neither too bright (overexposed) nor too dark (underexposed or noisy) gray levels exist. If also modified beam parameters (X-ray voltage, power, pre-filter) are included in the optimization, the expected gray values result from the forward projected radiographic length images.

By way of example, it is provided that the pre-radiographic image or radiographic length image is searched for with the darkest pixel, from which the radiographic length present is simulated, and thus the longest radiographic length is found. This then serves for forward projection with the respective measurement parameters.

Whether changed beam parameters must be used results from the evaluation of the signal quality, such as the CNR, and the quantities of noise, structure resolution, measurement precision, measurement accuracy and / or measurement uncertainty of the measurement points or characteristics which can be derived from this. Influence of this also has the number of image averages. Here, however, the most frequently specified maximum measuring time must also be taken into account. Another factor influencing the quantities to be optimized is the imaging scale. This is limited by the possible exit of the object from the measuring range, as already explained above, and the associated need for several laterally offset images and thus again the measuring time and possible collisions between the object and the X-ray source. Under these constraints, the maximum or required magnification can be selected. Since the optimization in terms of the sizes is also done by adjusting the other measurement parameters, this is a multi-dimensional optimization task.

• - das im Objekt vorliegende Material bestimmt wird, vorzugsweise indem die Schwächungswerte ermittelt und daraus der Absorptionskoeffizient bestimmt und zusammen mit dem vorliegenden Strahlspektrum insbesondere vorliegender Spannung das Material zumindest grob bestimmt wird und/oder
• - diverse Korrekturwerte für die Simulation bestimmt werden.

It should also be emphasized that from the preliminary radiographs or from the radiographic length images

• the material present in the object is determined, preferably by determining the attenuation values and determining therefrom the absorption coefficient and at least roughly determining the material together with the present beam spectrum, in particular the present stress, and / or
• - Various correction values for the simulation are determined.

In the material determination, the physical relationship is used that the characteristic of the absorption (related attenuation values from the gray values of the radiographic images and irradiated length) depends on the material. With the aid of known relationships (characteristic curves), the present characteristic can at least roughly be assigned to a material.

The simulation named here comprises at least the above-mentioned simulation of the transmission (forward projection). Here, several boundary parameters and correction factors are taken into account, which can be taken from the pre-radiation images or from the transmission length images, such as detector tilt, focal spot size, etc.

The following embodiment illustrates the method. The operator sets first starting values for the measuring parameters (in particular magnification, x-ray voltage, power, pre-filter, number of rotational positions, integration time and number of image averaging) according to his knowledge. He chooses the magnification so that the object probably does not leave the measuring range. By selecting a corresponding button in the user interface of the device control, the device automatically picks up the "Detectability Map", ie pre-radiation images in rotational positions from 0 ° to 180 ° plus cone angle. As a result, all transmission lengths, which also occur during a complete rotation (0 ° to 360 ° plus cone angle), are detected at least once. Based on this data, it is checked whether the object leaves the measuring range at the selected magnification. If this is the case, the object position can be adapted accordingly or object positions are defined which serve for the acquisition of laterally offset radiographic images, by which the object is completely detected. If this is not the case, a higher magnification can also be set automatically. This can be limited by a maximum permissible cone angle of the radiation through which the object is to be detected. On the basis of the "Detectability Map" is also checked whether pixels of Vorabdurchstrahlungsbilder are under or overexposed. If this is the case, the measurement parameters are adjusted accordingly. The evaluation of the then changing pixel gray values takes place on the basis of the simulated transmission length images. The optimization of the measurement parameters then takes place with respect to at least one of the variables noise, structure resolution, measurement precision, measurement accuracy and / or measurement uncertainty of the measurement points or features for the subsequent actual measurement.

Subject of an independent invention is a method for examining a workpiece or workpiece area by means of computer tomography.

According to the prior art, the settings of the recording parameters (voltage, current or power, pre-filter, etc.) used in a computed tomography are made for the X-ray source, recording parameters (integration time, number of image averages) of the detector and number of rotational positions depending on the object to be irradiated in particular its absorption on the basis of material and dimensions. Regardless of the magnification FDD / FOD (FDD = focus detector distance, FOD = focus object distance, focus = position of the X-ray emitting focal spot of the X-ray tube) is set so that, for example, the measurement object can be completely detected in a tomography or in such a way that the structural resolution limit (voxel) resulting from the detector resolution and the magnification Structure resolution limit) corresponding to the desired size for measuring the structures of the workpiece.

The disadvantage here, however, is that under certain circumstances resulting from the recording parameters of the X-ray source focal spot size leads to a significantly poorer structural resolution, so that the effective detector resolution may be too high. Due to the small amount of light present here, which falls on one pixel of the detector, there is the disadvantage that there are worse measurement results with regard to the signal-to-noise ratio.

A further object of the present invention is therefore to realize a measurement which is as accurate as possible, in particular with low noise, taking into account the present focal spot size and the resulting structure resolution limit (focal spot texture resolution limit). In addition, the measurement time should be as low as possible.

As a solution, the invention provides that the structure resolution limit (voxel structure resolution limit) resulting from the effective detector resolution and the image scale is adapted to the resolution limit (focal spot texture resolution limit) caused by the focal spot.

For the given workpiece size (which determines the size of the required measuring range), it is preferable to determine the optimum combination of FDD and FOD (and thus the optimal magnification) with the structural resolution to be achieved.

When the image scale is changed, influencing factors on the structure resolution, for example the effective pixel and focal spot size (in relation to the voxel size, ie in the object plane) are changed. The magnification must be chosen such that these influencing factors lead to the minimum structural resolution required for the inspection or measuring task or the maximum possible structural resolution (taking into account the maximum permissible measuring time). It does not make sense to arbitrarily reduce the influence on the structure resolution by a single influencing variable, if a different influencing variable results in a much stronger influence on the structure resolution. The aim is thus to minimize or at least reduce to a certain extent a number of influences on the structure resolution so that a predetermined structure resolution is achieved. In other words, the size of the influence by the influencing variables and the difference between the magnitudes of the respective influence on the structure resolution should be as small as possible without the point spread function dependent on the influencing variables when superimposing the influencing variables (eg convolution) to a smaller one as the at least necessary structure resolution or alternatively leads to the smallest possible structure resolution. The advantages of this procedure can be understood from the following examples.

Case 1: Effective focal spot> Effective pixel size: The magnification should be reduced to match the effective pixel size to the effective focal spot size. This is done by reducing FDD, resulting in an increase in intensity and effective cone angle. It is also possible to reduce the resolution by binning in order to reduce the effective pixel size. Also, the FOD can be increased. This moves the workpiece closer to the detector and reduces the effective cone angle.

Case 2: Effective focal spot <effective pixel size: Increase the magnification to match the effective pixel size to the effective focal spot size. This is done either by increasing FDD or by decreasing FOD. In both cases, a higher structure resolution arises. Increasing the FDD reduces the effective cone angle and the cone-beam artifacts that depend on it. Reducing the FOD increases the measured photon energy of the measurement radiation (X-ray radiation) or the SNR of the measured local attenuation in the volume (with the voxel size remaining the same).

In general, this method, while maintaining the required structural resolution for the measurement task, either optimizes the SNR of the measured local attenuation and the measurement precision of the surface or interface points determined from the volume data or reduces systematic uncertainty influences such as cone beam artifacts.

Regardless of this and in general, the following should be successful: maximizing the voxel size while at the same time achieving the maximum possible or at least necessary structure resolution for the set setting parameters in order to minimize the computation time and the memory requirement. Or, adjusting the effective focal spot size to the effective pixel size by defocusing the focal spot or changing the magnification.

For this purpose, the invention provides in a first embodiment that the detector pixel size is virtually enlarged by binning. The advantage of using binning methods is that having virtual pixels a higher amount of light, and thus higher signal to noise ratio SNR arise. The resulting lower detector resolution is justifiable, since a corresponding restriction by the focal spot size already exists. Also advantageous is the presence of a smaller number of pixels, which reduces the computation time in the reconstruction and subsequent evaluation steps, as well as the reduced memory required to store the reconstructed data.

According to a second alternative, it is provided that in the reconstruction the calculation of a correspondingly adapted, in particular enlarged, voxel grid takes place. This also gives the advantage of less computing time and less space required.

A further alternative provides that the magnification is reduced by reducing the focus-detector distance (FDD). Due to the "square-law", this increases the intensity detected per pixel. Advantageous is a resulting increased signal to noise ratio.

In another alternative, it is intended to increase the magnification by reducing the focus-to-object distance (FOD). This may result in the case that the workpiece is no longer completely imaged on the detector. In this case, a plurality of computed tomography measurements of the workpiece are carried out in positions of the workpiece offset laterally relative to the main beam direction of the x-ray source relative to the source and detector. The comparison of the focal spot resolution limit and the voxel resolution limit here serves as a plausibility check of this measurement method.

The invention provides a method for adjusting the resulting from the effective detector resolution and magnification structural resolution (voxel-structure resolution limit) in a computer tomography for examining a workpiece and / or one or more regions (ROI region of interest) of a workpiece, preferably dimensional measurements of features on the workpiece, wherein the computed tomography comprises at least the following steps: recording multiple radiographic images of the arranged on a turntable workpiece in several rotational positions (rotational steps) relative to a radiation source (source) such as X-ray source and a detector such as X-ray detector, in particular flat X-ray detector, reconstruction of a voxel volume from the radiographic images and preferably determination of surface measurement points from the voxel volume by means of surface extraction method that characterized auszeic hets that from the focal spot size resulting from the source measurement data required for the measurement, the workpiece resolution resultant limit (focal spot texture resolution limit) is determined and the voxel texture resolution limit is changed toward the focal point pattern resolution limit; preferably almost identical adapted to this.

In particular, the invention is characterized in that, for adaptation of the voxel structure resolution limit, a plurality of neighboring detector pixels are combined by binning, whereby the effective detector resolution is reduced.

Preferably, it is provided that, in order to adapt the voxel structure resolution limit, the reconstruction of a voxel volume takes place with a correspondingly adapted, preferably enlarged, voxel grid.

In particular, the invention is characterized in that, to adjust the voxel texture resolution limit, the magnification is reduced by decreasing FDD (focus detector distance), preferably keeping FOD (focus-to-object distance) constant, or by increasing FOD ,

Preferably, the invention contemplates that to accommodate the voxel texture resolution limit, the magnification is increased by decreasing FOD (focus-to-object distance), with FDD (focus-detector distance) preferably remaining constant or by increasing FDD, particularly Preferably, the workpiece is detected in a plurality of tomographies, in which the workpiece is offset to the source and the detector laterally with respect to the main radiation direction of the radiation source, in particular in the direction of the axis of rotation of the turntable and / or in the direction perpendicular thereto.

In particular, the invention is characterized in that, when there is too little (high) intensity in the transmission images, the power of the radiation source is increased (decreased), thereby changing the focal spot pattern resolution limit, and the voxel pattern resolution limit is directed toward the focal spot. Structure resolution limit is changed, preferably by reduction (magnification) of the image scale.

The subject matter of an independent invention is a method for the reconstruction of a computer tomography for the examination, in particular dimensional measurement of a workpiece.

According to the state of the art in computed tomography for industrial metrology, the reconstruction is carried out for the entire voxel volume with a constant voxel size. One reason for this is that the detectors used here have an isotropic behavior, ie have equidistantly distributed pixels in both directions. In addition, the focal spot geometry of the X-ray tube is also assumed to be isotropic. In reality, however, both assumptions do not fully agree so that measurement errors can occur. From detector resolution and magnification, possibly also from the focal spot size, a meaningful voxel size can be derived. By back projection of the filtered, logarithmized and normalized radiographic image data, the voxel gray values are determined for the selected voxel grid.

Deviating from this suggests the DE 10 2012 102 709 A1 The applicant proposes to subdivide the voxel volume into first and second subregions, wherein the first subregions comprise the regions of the workpiece in which material transitions or the transition between material and air is present. In this area, work is done with a lower voxel size in order to achieve the highest possible resolution and thus high measuring accuracy. The remaining areas designated as second areas are reconstructed with a larger voxel size, for example, to save measurement time and storage space. Two separate voxel volumes are generated.

However, there is the problem that in different areas of the material transitions, ie the areas where measurement points for linking to dimensions for the determination of features exist, may not always the same accuracy for the performance of the measurement task is needed because different large Tolerances may be present.

Equally unresolved is the consideration of different steep gradients of the gray values of the voxels at edge transitions, as may occur, for example, due to altered focal spot size or other artifacts in computed tomography.

It is also unresolved to combine voxel data from computed tomography measurements with different magnifications to a common voxel volume, as they occur in the region of interest computed tomography (ROI) CT, when at least two measurements with different magnification are performed. The so-called overview tomography, which detects the entire component width perpendicular to the direction of the axis of rotation, has a larger voxel size than the one or more high-resolution detail tomographies.

If voxel volumes of different voxel sizes are present, the dimensional evaluation of measures has hitherto been carried out, in that the measurement points to be linked were carried out by separate evaluation (surface extraction) of the respective voxel volume. The same applies to the examination of the interior of the material. Again, the multiple voxel volumes were examined separately.

Another object of the present invention is therefore to provide a voxel size which is locally adapted to the measurement task, in order to be able to measure accurately enough in accordance with the required measurement task and to save computing time and storage space on the one hand.

It is also an object of the invention, especially in the area of the workpiece edges, to locally adjust a voxel resolution adapted to the expected course of the voxel gray values, in particular the gradients of the edge transition.

It is also an object of the invention to be able to evaluate voxel volumes of different voxel sizes together in order, for example, to carry out a material inspection of the entire interior of the material or to generate surface points (surface measuring points) with a uniform algorithm, in particular to examine the entire object in layers.

To solve this, the invention provides that the reconstruction of the voxel volume takes place in regions with adaptable voxel size and thus resolution (structure resolution). In particular, it is necessary to adjust the voxel size continuously. The adjustment can be made by selecting the appropriate areas where the measurement points are to determine the dimensions for individual features, and determining the required resolution and thus voxel size by the operator. Alternatively, the determination of the ranges and voxel sizes can also be carried out automatically, for example, depending on the desired data and associated PMI data, in which the measurement tasks, assigned tolerances and resulting structure resolutions are defined. In the reconstruction, the backprojection of the radiographic image data is preferably carried out for each voxel center position chosen in this way and thus any voxel center position.

A voxel itself has no size as such. The voxel size is the voxel center distance, ie the distance between two adjacent positions having a voxel gray value. It can be different in different directions. According to the invention, therefore, a variable voxel distribution or voxel density or Voxelraster proposed, so there is an inhomogeneity in the Voxelmittenabständen.

The invention also provides for a solution in that the voxel size is locally adapted separately to the expected respective edge course, in particular gradients. The edge profile can be estimated by the operator's expertise, by a simulation of the CT, or by an additional high-speed CT from which a coarse voxel volume is reconstructed.

To solve this, the invention also contemplates combining differently resolved sets of transmission images (eg, from a ROI CT) into an inhomogeneously resulting voxel volume by either separately reconstructing the individual sets and assembling the reconstructed voxel volumes into the resulting voxel volume, or a joint reconstruction of all sentences into the resulting inhomogeneous voxel grid. The material inspection or determination of surface measurement points, particularly preferably layer-by-layer examination of the voxel volume to determine dimensions, can then take place on this resulting voxel volume.

The invention provides for a solution for a reconstruction method for a computed tomography for examining a workpiece, in particular for the dimensional determination of features on a workpiece comprising at least the following steps: recording at least a first set of multiple radiographic image of the arranged on a turntable workpiece in several rotational positions (rotational steps ) relative to a radiation source (source) such as an X-ray source and a detector such as X-ray detector, preferably X-ray area detector, reconstruction of a voxel volume having voxels with assigned voxel gray values from the transmission images, and preferably determination of surface measurement points from the voxel volume by surface extraction methods characterized thereby in that the size of the voxels (voxel size) within the reconstructed voxel volume is locally different, ie the voxel volume is inhomogeneous, in particular In particular, the reconstruction takes place in regions with different voxel sizes, or voxel volumes of different voxel sizes are combined to form a resulting voxel volume, or the different voxel sizes are generated in a reconstructed voxel volume, for example by resampling and / or combining voxels and / or interpolation methods.

In particular, the invention is characterized in that the voxel size is infinitely adjustable for the plurality of regions, in particular from voxel to voxel, that is, the voxel center distance.

Preferably, it is provided that the adjustment of the voxel size, in particular automatically, depending on the required for the solution of the respective measurement task structure resolution, in particular depending on the local curvature of each considered surface area occurs.

In particular, the invention is characterized in that the voxel size is determined in regions depending on the respective measurement task assigned to the respective feature, in particular depending on the workpiece associated PMI data and preferably assigned tolerances, is particularly preferably set automatically, in particular only for the areas around the surface measurement points needed for the features to be measured are set to a respective smaller voxel size compared to the remaining voxel volume.

Preferably, the invention provides that the determination of the different areas is carried out by operator selection and / or automatically based on the location of the required surface measurement points, ie the transitions between two different materials or a material and the ambient medium such as air, the position of the surface measuring points is roughly determined based on target data such as CAD data and / or on the basis of a previously performed higher voxel size quick reconstruction based on the first set of transmission images, the higher voxel size being realized by reconstructing binned detector pixels, or on a previously performed fast tomography using a separate set is reconstructed from radiographic images.

It should also be emphasized that the voxel size in the region of the material transitions is set to be locally adapted to the expected course of the voxel gray values, in particular the gradients of the edge transition, preferably wherein the expected profile is determined by a computer tomography simulation, in particular taking into account the respective focal spot size, and / or by a quick CT and / or by default by the operator.

The invention is also distinguished by the fact that a resulting voxel volume with regions of different voxel size is generated and made available for further evaluation, wherein preferably a material inspection and / or a determination of surface measuring points, particularly preferably a layer-by-layer examination of the voxel volume for the determination of Measurements are made taking at least two sets of radiographic images, the reconstruction of each set resulting in different voxel sizes, preferably because the sets were taken at different magnifications, such as ROI-CT, and either reconstructing the individual sets separately and reconstructing the sets reconstructed voxel volumes are assembled to the resulting voxel volume, or a joint reconstruction of all sets into the resulting inhomogeneous voxel volume occurs.

In particular, the invention is characterized in that different regions of a voxel volume are reconstructed according to a defined sequence, in particular selected regions are temporally reconstructed before other regions, wherein the selected regions are preferably the regions from which surface measurement points are determined for the linkage to dimensions, wherein particularly preferably areas are layers perpendicular to the direction of the axis of rotation.

As a result, the evaluation of the measurements can already be carried out before the entire reconstruction is completed. The idea described here also applies as a separate invention, detached from the presence of an inhomogeneous voxel volume.

According to a particularly noteworthy proposal it is envisaged that in the radiographic images local areas will be summarized (binned) before the reconstruction, which in the reconstruction contribute to areas of the voxel volume for which a smaller voxel size is provided in comparison to other areas of the voxel volume.

The resulting smaller amount of data faster data processing, especially reconstruction is possible. Which image areas in the radiographic images contribute to which areas in the voxel volume can be determined beforehand, for example by simulating the CT. This must be determined separately for each radiograph. This idea described here is alternatively considered as a separate invention, detached from the presence of an inhomogeneous voxel volume and that is because, alternatively, a homogeneous voxel volume with a constant voxel size can be achieved by the reconstruction of this directly or subsequently (eg by resampling or interpolation) generated. It is then simply used for multiple voxels on the identical binned pixels of the radiograph. This leaves the advantage of the smaller amount of data for the radiographic images.

The invention relates to a computed tomography device with a radiation protection hood and integrated in the radiation protection turntable, X-ray detector and X-ray source.

In the known devices for computed tomography essentially two types of X-ray sources are to be distinguished: so-called "closed" x-ray tubes and so-called "open" or openable x-ray tubes.

"Closed" x-ray tubes are used with reflection target. Due to their design, these X-ray tubes can not be opened to replace wearing parts. The necessary vacuum is generated once by the manufacturer and is maintained over the life of the system. Maintenance is not possible. A high voltage generator is directly connected to the X-ray tube in such systems. Due to the design of these tubes they are limited to a voltage range of usually 130 kV, in exceptional cases 150 kV. Although a compact design is possible with reflection target tubes, the design-related principle leads to a poor ratio between performance and resolution.

The second design is used in particular for X-ray tubes in the range greater than 150 kV. Here usually the "open" construction principle is used. The vacuum is generated during operation via a vacuum pump. The high voltage generator is connected to the tube via a cable and is located outside the X-ray machine. Turntable, X-ray detector and X-ray source are arranged in a radiation protection housing such that a high voltage power supply of the X-ray source is realized by a generator disposed outside the radiation protection housing, between generator and X-ray source, a high voltage cable is used. This design is used in particular for X-ray tubes with transmission target. Wear parts such as filaments and target can be changed easily. Maintenance of the X-ray tube is possible. On the other hand, the "open" design of these X-ray tubes requires frequent maintenance, which is associated with relatively high costs. The cable connection between the X-ray tube and the generator must also be regularly maintained. The disadvantage here is that the high voltage cable for connection to the vacuum tube of the X-ray source must be performed by a surrounding the vacuum tube insulator and this surrounding jacket. This required rubber sleeves to maintain the vacuum in the vacuum chamber act insulating, however, the formation of cracks can form air chambers, resulting in a rollover and thus the destruction of at least the Lead cable. Corresponding rubber sleeves must therefore be greased regularly. Another disadvantage is that the construction with a separate generator requires a relatively large footprint around the generator (dimensions of side lengths of about 0.5 meters) in order to arrange this. Transmission tubes are not yet manufactured as "closed" systems due to wear. However, they offer a better ratio of performance to resolution. This is currently being purchased at a higher maintenance cost. In particular, standard filaments are also equipped with a filament at least in transmission tubes, so that high current densities are present within the thread, resulting in high wear.

The object of the present invention is to provide a compact and low-wear computed tomography device, which allows both the easy replacement of wearing parts such as filament and target, and has a particularly long life. In addition, a fast measurement with high resolution and accuracy should be possible.

To solve the invention provides a device in which the radiation protection housing in addition to turntable, X-ray detector and X-ray source and the generator is arranged, preferably flanged to the X-ray source. The x-ray source is designed as a transmission tube with a transmission target, in particular a "open" or openable tube suitable for maintenance, preferably with a particularly high maximum power and a particularly high maximum acceleration voltage, in particular power> 40 W and acceleration voltage> 150 kV. Transmission target tubes offer the advantage of high performance and measurement speed with a small focal spot due to the execution of the target as a transmission target and thus high resolution. This makes it possible to measure accurately and at the same time quickly. This is especially true for plastic workpieces with high glass fiber content and metal components.

As transmission tubes according to the invention optionally those with maximum acceleration voltage of 130 kV and 160 kV or greater and maximum power up to about 80 W (eg., At 160 kV) are provided. This makes it possible to measure even dense materials with large transmission lengths. In order to realize a modular concept it is intended to construct and arrange X-ray source and X-ray detectors easily exchangeable.

Corresponding open transmission tubes have a long service life due to easy exchangeability of wearing parts such as the filament or the target. In order to further reduce wear and / or maintenance cycles, filaments with a plurality of filaments or filaments are provided with an otherwise identical electron current.

By flanging the generator to the transmission tube corresponding plugs and to greasy rubber sleeves account, so that corresponding tubes need not be completely replaced after a few years compared to the prior art.

It is also envisaged to arrange the device provided for evacuating the vacuum tube, in particular the vacuum pump in the radiation protection housing. The arrangement preferably takes place directly on the vacuum tube.

The implementation of X-ray source and generator as a unit, preferably as a unit together with the vacuum pump results in low operating costs and short downtime by repair or maintenance, and low acquisition costs.

According to the invention, it is further provided to design the computed tomography device as a coordinate measuring machine, that is to say to use drives and measuring systems in the form of measuring axes for relative positioning of turntable, X-ray detector and X-ray source, and possibly further conventional sensors known from coordinate measuring technology for realizing a multi-sensor coordinate measuring machine in the radiation protection housing to integrate. For this purpose, the further sensors are equipped with separate measuring axes or at least partially use the measuring axes already present for relative positioning of the x-ray tube, the x-ray detector and the turntable. In order to measure very precisely, it is also intended to use air-bearing rotary tables.

In the context of the invention is also intended to design the generator in a particularly compact design. This is possible in particular if an open transmission tube is used as the tube. In connection with the flanging to the transmission tube, this results in the possibility of integrating the generator in the radiation protection housing, without having to use a particularly large radiation protection housing. This results in a compact design, low overall mass and easy transportability.

The computed tomography device is designed in particular for measuring workpieces. In a first embodiment it is provided that in particular the dimensional measurement of features on the surface of workpieces is provided. For this purpose, a control and evaluation unit connected to the computed tomography device is provided and adapted to record radiographic images in several relative rotational positions between the workpiece and the unit of X-ray detector and X-ray source, which are reconstructed into a voxel volume, from which in turn surface measuring points for the determination of geometric features on the workpieces be derived. In a second alternative or additional embodiment, the control and evaluation unit is also designed to inspect the material or to examine the voxel volume representing the interior of the material for example voids, inclusions or the like.

The invention provides for solving a computed tomography device, comprising a radiation protection housing, within which at least one turntable, at least one X-ray detector, preferably flat X-ray detector, and at least one X-ray source comprising at least vacuum tube, filament and target, is characterized in that the X-ray source has a transmission target, that is designed as a transmission tube, and that the X-ray source is preferably designed as a so-called open so openable tube with exchangeable consumable parts such as filament and / or target, wherein the required for supplying the X-ray source with high voltage generator with within the radiation protection housing is arranged.

In particular, the invention is characterized in that the generator and the X-ray source are connected directly to each other, wherein the generator is flanged directly to the X-ray source and / or vacuum tube, in particular a separate pluggable cable connection and connector implementation omitted, generator and X-ray source thus form a unit.

Preferably, it is provided that the vacuum pump provided for evacuating the vacuum tube is arranged inside the radiation protection housing, is preferably arranged directly on the vacuum tube, particularly preferably generator, X-ray source and vacuum pump thus form a unit.

According to a proposal to be particularly emphasized, it is provided that the control unit provided for controlling the computer tomography device, in particular the electronic unit, is designed at least for controlling at least one of the turntable, x-ray detector, x-ray source, generator, vacuum tube and / or optionally existing measuring axes, directly below, in particular directly below Computer tomography device is arranged and forms a unit with this.

The computed tomography device, in particular the radiation protection housing and components arranged therein, are arranged in a special embodiment on a subframe that contains the control unit. This offers the advantage that the structure is space-saving and easy to transport as a unit. This also results in a small footprint. Special cable shielding can be omitted as they run inside the unit. Only be led to the outside cables for power and for connection to a control computer, such as USB cable.

It should also be emphasized that the X-ray source is designed for high maximum powers and high maximum acceleration voltages, in particular powers greater than 40 W and acceleration voltages greater than 150 kV

After a particularly emphasized proposal is provided that the turntable is carried out air bearing.

In particular, the invention is characterized in that the filament of the vacuum tube has more than one filament, in particular two or three filaments.

Preferably, it is provided that X-ray detector and / or X-ray source are designed and arranged interchangeably.

In particular, the invention is characterized in that the radiation protection housing is at least partially modularly constructed from a plurality of lead plates and / or several mineral cast elements. This results in a simplified, easily adaptable structure.

It should also be emphasized that the measuring volume, that is to say the area penetrated by the X-ray detector and heat sources such as generator are arranged spatially separated within the radiation protection housing, and an additional separation by means of an at least partially air-permeable bulkhead and / or fans are arranged such that Air flow of measuring volume in the direction of the heat sources can be generated. The bulkhead serves to shield the measurement volume against heat from the heat sources inside the radiation protection housing. By way of example, an air flow firstly passes through the first outside fan into the measuring volume through the first fan, and from there through the bulkhead, which is at least partially permeable to air, further in the direction of the heat sources, past them and out through the housing through the second fan. First and second fans are arranged accordingly and executed in their effect (direction of rotation). This will be temporal created constant temperature conditions inside the housing, which can be measured more accurately.

Preferably, it is provided that the components turntable, X-ray detector and X-ray source to each other in the radiation protection housing fixed, are arranged without mechanical adjustment axes.

According to a proposal to be particularly emphasized, it is provided that measuring axes are arranged relative to one another in the radiation protection housing for the relative movement of the components of the turntable, the x-ray detector and the x-ray source.

In particular, the invention is characterized in that the components turntable, X-ray detector and X-ray source to each other in radiation protection housing with automatically adjustable measuring axes and / or manually adjustable axes for movement, preferably the turntable, along the central beam direction of the X-ray radiation are arranged to adjust the magnification ,

• - je Drehstellung zu jeweils einem resultierenden Durchstrahlungsbild zu verknüpfen und aus den resultierenden Durchstrahlungsbildern der mehreren Drehstellungen mittels Rekonstruktion einen resultierenden Volumendatensatz (Voxelvolumen) zu erstellen oder
• - zu jeweils einem separaten Volumendatensatz (Voxelvolumen) zu rekonstruieren und die mehreren Volumendatensätze zu einem resultierenden Volumendatensatz zu verknüpfen,

wobei vorzugsweise aus dem resultierenden Volumendatensatz Oberflächenmesspunkte ermittelt werden oder aus den Volumendatensätzen Oberflächenmesspunkte ermittelt und verknüpft werden, wobei sämtliche Oberflächenmesspunkte in einem gemeinsamen Koordinatenmesssystem vorliegen.It should also be emphasized that the components turntable, X-ray detector and X-ray source to each other in the radiation protection housing with automatically adjustable measuring axes and / or with manually adjustable axes, for movement, in one or both perpendicular to the mean beam direction of the X-ray directions are arranged around the section to be measured or to set a section of a workpiece emanating from the turntable, in particular to sequentially set and tomograph different section or cutout of the workpiece (recording of transmission images in different rotational positions of the turntable) and thereby recorded radiographic images of at least two of the sections and / or cutouts

• - Link each rotational position to a respective radiographic image and from the resulting radiographic images of the multiple rotational positions by reconstruction to create a resulting volume data set (voxel volume) or
• - to reconstruct a separate volume data set (voxel volume) and to link the multiple volume data sets into a resulting volume data set,

wherein surface measurement points are preferably determined from the resulting volume data set or surface measurement points are determined and linked from the volume data sets, wherein all surface measurement points are present in a common coordinate measuring system.

As the plurality of sections, portions of the workpiece which are offset from each other in the direction of the rotation axis of the measuring table are referred to. In this case, the transmission images or the volume data or the surface measurement points can be linked, since a reconstruction for each section is separately possible with high accuracy.

As the plurality of cutouts, portions of the workpiece are designated which are offset from each other perpendicular to the direction of the rotation axis of the measuring table. In this case, the radiographic images of the multiple cutouts per rotational position must first be linked to resulting radiographic images in order to achieve a reconstruction with high accuracy. It is according to the invention, however, also intended to individually reconstruct the cutouts and to accept corresponding occurring artifacts.

Covering the plurality of to be measured areas of the workpiece for the selected magnification in the direction of the axis of rotation of the measuring table and transverse to the detected by the X-ray detector beam cone emitted by the X-ray source measurement radiation, so in both directions offset from each other radiographic images are taken in each case a plurality of rotational positions. As explained for several sections, the transmission images of all regions (sections and sections) are preferably first linked to resulting radiographic images and then reconstructed.

It is particularly preferably provided to thereby detect and measure the entire workpiece in a preferably selectable, particularly preferably as high a scale as possible, in particular to measure surface measuring points. It is also intended to detect only individual sections and / or cut-outs which are adjacent or separate from one another, that is to say have a spacing or clearance relative to one another.

Due to the mobility of the components turntable, X-ray detector and X-ray source to each other in the direction and in the two directions transverse to the main beam direction of the measuring radiation (X-radiation), it is also provided according to the invention to perform so-called ROI CT's (Region of Interest computed tomography) and multi-ROI CT's , In a ROI-CT, a single section of the workpiece is tomographed on a high magnification scale. In a multi-ROI-CT, a plurality of cutouts, which are mostly spaced apart from each other, are sequentially detected in each case with a high-resolution tomography. Both methods have missing transmission information of workpiece areas, which are not in all rotational positions on the X-ray detector are displayed, otherwise provided for the reconstruction. This can be done by a second tomography with a correspondingly lower magnification or mathematically by forward projection on the target data of the workpiece.

It is also provided according to the invention that the recording of the radiographic images takes place in the several rotational positions during the rotational movement of the rotary table, that is, in particular, the rotational movement for the image recording is not stopped. To reduce motion blur, the image is taken with a short exposure or integration time. To avoid associated quality limitations, the number of rotation positions in which the images are taken is increased. To increase the image quality (signal-to-noise ratio or contrast-to-noise ratio), it is also preferably provided to superimpose a plurality of successively recorded images.

Preferably, it is provided that the relative to the relative movement of the components turntable, X-ray detector and X-ray source to each other arranged in the radiation protection housing measuring axes are designed to operate as a Cartesian coordinate measuring machine with conventional sensors suitable.

It should also be emphasized that in the radiation protection housing one or more other conventional sensors such as tactile sensors, optical sensors, especially image processing sensors or optical distance sensors, in particular laser distance sensors, such as Foucault distance sensors and / or chromatic confocal distance sensors and / or interferometric distance sensors and / or white light sensors , and / or tactile-optical sensors are integrated, and form together with the computed tomography device, a multi-sensor coordinate measuring machine, wherein the further sensors are connected to separate measuring axes or with each other for the relative movement of the components rotary table, X-ray detector and X-ray source existing measuring axes.

Another object of the invention is to realize a further simplification of the measurement with a computed tomography device, in particular automatic measurement of several workpieces, with a compact design.

To solve is according to a further inventive idea, which is claimed independently or in combination with the aforementioned ideas, provided to integrate a parking station in the radiation protection housing of the computed tomography device. This accommodates several workpieces or pallets with one or more workpieces, which can optionally be placed automatically on the turntable to be measured successively. This results in the advantage that the radiation protection housing can be closed and the X-ray source can remain switched in between the measurements of several workpieces. As a result, an automated measurements without user intervention, for example, overnight, possible. Time savings also result because warm-up phases of the x-ray source are obsolete. For moving the respective workpiece or the respective pallet, it is preferably provided to use the measurement axes which are present anyway, for example the measuring axes assigned to the turntable. However, it is alternatively or additionally provided for this purpose to use separate positioning systems such as grippers o. Ä. To make the loading. According to a further alternative proposal, it is provided to carry out the loading by means of a gripper which can be attached to the measuring axes for the further sensors.

The solution is provided according to an independent or combinable idea that a computed tomography device has a parking station with multiple places, which is integrated in a radiation protection housing, wherein in each place a workpiece or one or more workpieces receiving workpiece holding device (pallet) automatically from the turntable can be stored and on the turntable is receivable by the respective place by measuring axes preferably assigned to the turntable are formed to move on the turntable placed workpiece or pallet to a selectable place of the parking station and to couple with this and decoupled, for example by means magnetic locking or by means of positive locking generating movement, wherein the parking station outside the range of the X-ray detected by the X-ray detector is disposed, preferably between the X-ray source and X-ray detector in the range of movement of the measuring axis n.

According to a particularly noteworthy proposal, it is provided that the turntable is designed to accommodate and jointly rotate a plurality of workpieces and / or to receive a workpiece holding device (pallet) receiving a plurality of workpieces, whereby a plurality of workpieces can be measured simultaneously with the computed tomography device.

Subject of an independent invention is a method for examining a workpiece preferably dimensional measurement of features on the workpiece, by means of computed tomography, in particular artifacts are reduced or corrected.

For the dimensional measurement of complex geometries (geometric features) different tactile, tactile-optical, optical or computer tomographic sensors (computed tomography or computed tomography sensors) are used. Preferably, these are operated in coordinate measuring machines (CMMs), sometimes also several sensors combined in one device (multi-sensor CMM).

Computer tomography for the dimensional measurement of workpieces is to be understood as meaning that a reconstruction of the volume information (voxel data or voxel amplitudes in the form of gray values, also referred to as voxel volume or volume data) is calculated from the set of two-dimensional radiographic images recorded in a plurality of rotational positions of a workpiece. for the volume detected by the detector, wherein the voxel data is a measure of the local attenuation and thus the local attenuation coefficients, and at material boundaries by surface extraction methods from the voxel data measuring points or surface measurement points are generated. Measurements on the workpiece or dimensions of features or structures on the workpiece can be determined from these surface measuring points, that is to say dimensional measurements take place. The surface of the workpiece is represented for example by networking the surface points in the so-called STL (Standard Triangulation Language) format.

A computed tomography or computed tomography (CT) sensor system generally consists of a flat detector, a radiation source, preferably an X-ray source, and a turntable for rotating the workpiece to be measured in the beam cone of the detector detected part of the radiation emitted by the radiation source around a (mathematical) axis of rotation. In kinematic reversal, however, it is also possible to arrange the workpiece firmly and to rotate the detector and radiation source around the workpiece. There are both mechanically stored, as well as air-bearing, or otherwise such as hydraulically mounted, etc., turntables provided. The mathematical axis of rotation often coincides with the physical axis of rotation of the turntable. However, both may differ if the turntable is moved during rotation. In this case, a resulting rotation of the workpiece located on the turntable can take place about a deviating from the physical axis of rotation of the turntable virtual axis of rotation, as for example the EP2399237B1 the applicant describes.

In addition to area-wide detectors, line detectors are also used as detectors. These have only a single or a few (e.g., three to five or even ten) detector rows. For complete recording of radiographic information of a spatially extended workpiece workpiece and detector must be placed in several along the direction of the axis of rotation (mathematical axis of rotation) shifted positions, the displacement can also take place during the rotation (helical CT). Advantageous is the possible reduction of stray radiation artifacts. The increased time expenditure resulting from the shift is avoided by the use of area detectors. Nevertheless, the present invention or the teachings of the invention can also be implemented for line detectors. Instead of processing radiographic images, ie 2D images, the information recorded with the respective detector row (s) is processed and also referred to herein as radiographic images for the sake of simplicity.

When scanning a workpiece by means of computed tomography, a large number of physical effects occur, such as absorption, scattering and beam hardening. By non-corrected deviations in the geometric arrangement of the radiation source, workpiece, the workpiece rotating turntable and detector to each other, the cone beam geometry of the received from the detector part of the emitted radiation from the radiation source and other deviations in the properties of these components such as thermal and temporal drift of the measuring radiation emitting focal spot of the radiation source, tumbling the axis of rotation of the turntable occur more errors in the measurement. These effects are collectively called artifacts. For the correction of artifacts in the measurement of workpieces with computer tomography, various methods are known.

For the correction of beam hardening artifacts, a characteristic-based correction is known in which the characteristic curves are corrected from the gray values of the transmission images and the assigned transmission lengths and corrected radiation patterns corrected for the corrected characteristic are used to reconstruct the volume data. The method is also called "Empirical Artifact Correction".

A correction method in which artifacts are calculated and corrected by simulating one or more physical effects, preferably beam hardening, scattered radiation and / or cone beam artifacts, preferably by the correction from the comparison of simulated radiographic images, volume data or surface measurement points is derived with and without including the physical effects is in the WO2013167616A2 by the assignee and is referred to as "Virtual Artifact Correction".

Also known is a so-called multi-spectral tomography (MSP-CT), in which at least two tomographies with different spectrum of the X-ray tube are made and the results of the tomographies are combined. The variation of the spectrum is effected by changing the acceleration voltage of the X-ray source and possibly adjusting the beam filter directly in front of the X-ray source.

A multi-slice tomography, in which at least two tomographies with different position or orientation of the workpiece relative to the source and detector take place and the results of the tomographies are combined is known. The change in position is effected by a pivoting movement of the workpiece on the turntable. In the following, "position or orientation" according to the invention in addition to the angular position and the position, ie translational position of the workpiece relative to the X-ray tube, detector and turntable axis to be meant.

It is from the WO2005119174A1 The applicant is also known to perform several measurements in staggered positions of the workpiece relative to the unit of X-ray source and detector to capture the workpiece piecewise.

In the known half-side tomography, only one side of the workpiece is detected, as a result of which workpieces can be measured whose lateral extent is greater than the area of the measuring radiation detected by the detector. This is associated with deviations in the measurement accuracy.

However, the two last named methods have not heretofore been intended to correct artifacts or to implement multi-slice tomography. It has also been found that none of the aforementioned methods taken alone is accurate enough to meet the highest accuracy requirements, in particular to correct all possible artifacts. It also lacks the flexibility to detect artifacts and then select the appropriate correction method.

The object of the present invention is therefore to avoid the problems of the prior art and in particular to enable a particularly accurate artifact correction in computed tomography.

To solve the invention, in a first embodiment, a plurality of artifact correction methods and / or multiple tomography methods are executed mixed or cascaded. This makes it possible to select the method which is sensitive in each case for the present artifacts, it being possible, in particular for locally present artifacts, to select the suitable correction method, for example a suitable position in a multi-slice tomography.

• - Mehrspektren-Tomografie (MSP-CT), bei der mindestens zwei Tomografien mit unterschiedlichem Spektrum der Röntgenröhre erfolgen und die Ergebnisse der Tomografien kombiniert werden, wobei entweder Durchstrahlungsbilder je Drehstellung oder je Tomografie rekonstruierte Volumendaten kombiniert werden,
• - Mehrlagen-Tomografie, bei der mindestens zwei Tomografien mit unterschiedlicher Lage bzw. Orientierung des Werkstücks relativ zu Quelle und Detektor erfolgen und die Ergebnisse der Tomografien kombiniert werden, wobei entweder Durchstrahlungsbilder je Drehstellung oder je Tomografie rekonstruierte Volumendaten kombiniert werden,
• - Mehrbrennfleck-Tomografie, bei der mindestens zwei Tomografien mit unterschiedlicher Brennfleckgröße der Röntgenröhre, insbesondere durch Änderung der Röntgenröhrenleistung, und/oder Brennflecklage auf dem Target der Röntgenröhre durchgeführt werden und die Ergebnisse der Tomografien kombiniert werden, wobei entweder Durchstrahlungsbilder je Drehstellung oder je Tomografie rekonstruierte Volumendaten kombiniert werden,
• - Korrektur-Verfahren, bei dem die Kennlinien aus den Grauwerten der Durchstrahlungsbilder und den zugeordneten Durchstrahlungslängen korrigiert wird und mit der korrigierten Kennlinie korrigierte Durchstrahlungsbilder berechnet werden (empirische Artefaktkorrektur), die zur Rekonstruktion der Volumendaten verwendet werden,
• - Korrektur-Verfahren, bei dem Artefakte durch Simulation einer oder mehrerer bei der Computertomografie, insbesondere der Durchstrahlung auftretender physikalischer Effekte, vorzugsweise Strahlaufhärtung, Streustrahlung und/oder Kegelstrahlartefakte berechnet und korrigiert werden (Virtuelle Artefaktkorrektur), vorzugsweise indem die Korrektur aus dem Vergleich simulierter Durchstrahlungsbilder, Volumendaten oder Oberflächenmesspunkten mit und ohne Einbeziehen der physikalischen Effekte abgeleitet wird,

und vorzugsweise nacheinander in beliebiger Reihenfolge durchgeführt werden, und wobei die ggf. notwendige Kombination der Ergebnisse der mehreren Computertomografien erfolgt indem:

• - je Satz von Durchstrahlungsbildern ein Voxelvolumen rekonstruiert wird und aus den mehreren Voxelvolumen ein resultierendes Voxelvolumen erzeugt wird oder
• - die mehreren Sätze von Durchstrahlungsbildern, vorzugsweise je Drehstellung, zu resultierenden Durchstrahlungsbildern kombiniert und diese zu einem Voxelvolumen rekonstruiert werden oder
• - zumindest zwei Voxelvolumen kombiniert werden, wobei zumindest eines der Voxelvolumen aus einem Satz resultierenden Durchstrahlungsbildern rekonstruiert wird, wobei die resultierenden Durchstrahlungsbilder aus zumindest zwei Sätzen von Durchstrahlungsbildern erzeugt werden,

und wobei die ggf. notwendige Anwendung der Korrektur-Verfahren erfolgt, indem:

• - der oder die jeweiligen Sätze von Durchstrahlungsbildern durch die empirische Artefaktkorrektur und/oder virtuelle Artefaktkorrektur korrigiert werden oder
• - jeweils aus den mehreren Voxelvolumen oder aus dem resultierenden Voxelvolumen durch Rückprojektion ermittelte Durchstrahlungsbilder durch die empirische Artefaktkorrektur und/oder virtuelle Artefaktkorrektur korrigiert und anschließend zu jeweils einem korrigierten Voxelvolumen bzw. dem korrigierten resultierenden Voxelvolumen rekonstruiert werden oder
• - jeweils die mehreren Voxelvolumen oder das resultierende Voxelvolumen durch die virtuelle Artefaktkorrektur korrigiert werden oder
• - die Oberflächenmesspunkte durch die virtuelle Artefaktkorrektur korrigiert werden

und aus dem Voxelvolumen oder dem resultierenden Voxelvolumen oder dem korrigierten Voxelvolumen oder dem korrigierten resultierenden Voxelvolumen zur dimensionellen Messung des Werkstücks die Oberflächenmesspunkten ermittelt werden und/oder das Voxelvolumen oder resultierende Voxelvolumen oder korrigierte Voxelvolumen oder korrigierte resultierende Voxelvolumen auf Lunker, Einschlüsse, Defekte oder ähnliches untersucht wird.The invention provides for a method for examining a workpiece, preferably dimensional measurement of features on the workpiece, by means of computed tomography, comprising at least the following steps: recording at least a first set of several radiographic images of the workpiece arranged on a turntable in several rotational positions (rotational steps) relative to a radiation source (source) such as X-ray source and a detector such as X-ray detector (first computed tomography) reconstruction of a voxel volume (volume data) having voxels with assigned voxel gray values from the transmission images and preferably determination of surface measurement points from the voxel volume by surface extraction methods characterized in that at least two of the following computed tomography methods (second or further computed tomography or tomography, wherein further sets of radiographic image in several rotational positions) or correction procedures for reducing and / or correcting artifacts are combined:

• - multispectral tomography (MSP-CT) involving at least two tomographies of different X-ray tube spectrum and combining the results of the tomographs, combining either radiographic images per rotational position or volume data reconstructed per tomography,
• Multilayer tomography, in which at least two tomographies with different position or orientation of the workpiece relative to source and detector take place and the results of the tomographs are combined, combining either radiographic images per rotational position or volume data reconstructed per tomography,
• Multi-focal point tomography, in which at least two tomographies with different focal spot size of the X-ray tube, in particular by changing the X-ray tube power, and / or focal spot on the target of the X-ray tube are performed and the results of the tomographs are combined, wherein either radiographic images per rotational position or per tomography reconstructed Volume data are combined,
• - Correction method in which the characteristics of the gray values of the radiographic images and the associated transmission lengths is corrected and corrected with the corrected characteristic corrected radiographic images (empirical artifact correction), which are used to reconstruct the volume data,
• Correction method in which artifacts are calculated and corrected by simulation of one or more physical effects, preferably beam hardening, scattered radiation and / or cone beam artifacts occurring in computed tomography, (virtual artifact correction), preferably by the correction from the comparison of simulated radiographic images , Volume data or surface measurement points are derived with and without the physical effects,

and preferably one after the other in any order, and where necessary the necessary combination of the results of the several computerized tomographies is carried out by:

• a voxel volume is reconstructed for each set of transmission images and a resulting voxel volume is generated from the plurality of voxel volumes, or
• - The plurality of sets of transmission images, preferably per rotational position, combined to resulting radiographic images and these are reconstructed into a voxel volume, or
• at least two voxel volumes are combined, wherein at least one of the voxel volumes is reconstructed from a set of resulting transmission images, wherein the resulting transmission images are generated from at least two sets of transmission images,

and where necessary the application of the correction procedures is carried out by:

• - the respective sets of radiographic images are corrected by the empirical artifact correction and / or virtual artifact correction, or
• - Each of the multiple voxel volume or from the resulting voxel volume determined by backprojection radiographic images are corrected by the empirical artifact correction and / or virtual artifact correction and then reconstructed to a respective corrected voxel volume or the corrected resulting voxel volume or
• in each case the multiple voxel volumes or the resulting voxel volumes are corrected by the virtual artifact correction or
• - the surface measurement points are corrected by the virtual artifact correction

and from the voxel volume or resulting voxel volume or corrected voxel volume or corrected voxel volume for dimensional measurement of the workpiece, the surface measurement points are determined and / or the voxel volume or resulting voxel volumes or corrected voxel volumes or corrected resulting voxel volumes for voids, inclusions, defects or the like are examined becomes.

In multi-focal spot tomography, different focal spot sizes and the associated different possible beam power in one case (small focal spot size) allow details to be better resolved and, in the other case (high beam power), difficult-to-irradiate areas can be made more visible. When reconstructing the different states are to be considered. If regions in the radiographic images are overexposed or too dark, the invention provides that these are masked out or an automatic adaptation of the image acquisition time, number of image averages and / or degree of local or spatial superimposition, ie overlap of the radiographic images takes place.

The invention also provides to change the position of the focal spot on the target. This results in a changed position of the focal spot and thus the source with respect to the workpiece and the detector and the method can be used to vary the position in a multi-slice tomography. This is particularly useful for large imaging scales, since the angle at which the workpiece is irradiated and thus the respective irradiated length changes.

In addition, the change in the position of the focal spot on the target can also result in a changed spectrum of the emitted measuring radiation. Therefore, it is also provided according to the invention to use this as an embodiment of a multi-spectral tomography.

In particular, the invention is characterized in that for the multiple layers of a multi-layer tomography in each case an MSP-CT is performed, first the results per layer and then for the multiple layers are combined.

Preferably, it is provided that in each case a multi-slice tomography is performed for the multiple spectra of an MSP-CT, wherein First, the results per spectrum and then combined for the multiple spectra.

In particular, the invention provides that an empirical artifact correction or a virtual artifact correction is performed beforehand.

According to a proposal to be particularly emphasized, it is provided that before combining volume data sets, these are fitted to one another, that is, they are present in an identical voxel grid, preferably by using resampling methods.

The combination of the volume data records takes place z. B. also by finding a suitable linear combination of the two volume data sets, as is known for example for MSP-CT.

• - Variation der Röntgenspannung bzw. Beschleunigungsspannung der Röntgenröhre und/oder
• - Variation des Materials des die beim Auftreffen von Elektronen erzeugenden Targets der Röntgenröhre und/oder
• - Variation des Materials und/oder der Dicke des zur Filterung der von der Röntgenröhre abgegebenen Messstrahlung verwendeten Strahlfilters, welcher vor dem Auftreffen der Messstrahlung auf das Werkstück angeordnet ist und/oder
• - Variation des Materials und/oder der Dicke des zur Filterung der von der Röntgenröhre abgegebenen Messstrahlung verwendeten Strahlfilters, welcher nach dem Durchdringen des Werkstücks mit der Messstrahlung und vor dem Auftreffen auf den Detektor angeordnet ist und/oder
• - Variation des Materials und/oder der Dicke der Szintillatorschicht des Detektors und/oder
• - Variation der Brennflecklage auf dem Target der Röntgenröhre.

It should also be emphasized that the variation of the spectrum of the X-ray tube in MSP-CT is done by:

• Variation of the x-ray voltage or acceleration voltage of the x-ray tube and / or
• - Variation of the material of the electron-generating targets of the X-ray tube and / or
• Variation of the material and / or the thickness of the beam filter used for filtering the measuring radiation emitted by the X-ray tube, which is arranged on the workpiece prior to the impact of the measuring radiation and / or
• Variation of the material and / or the thickness of the beam filter used for filtering the measuring radiation emitted by the X-ray tube, which is arranged after penetrating the workpiece with the measuring radiation and before hitting the detector and / or
• Variation of the material and / or the thickness of the scintillator layer of the detector and / or
• - Variation of the focal spot on the target of the X-ray tube.

Preferably, the invention provides that a combined multi-slice tomography and MSP-CT is performed, wherein at least a first tomography in a first position with a first spectrum and a second tomography in a deviating from the first position second position with one of the first spectrum deviating second spectrum is carried out, preferably using a simulation of the position-dependent transmission lengths first layer is selected so that as many short transmission lengths are present and second layer is chosen so that as many long transmission lengths are present, the first spectrum in comparison to the second spectrum lower on the Workpiece and / or the detector incident beam power and / or acceleration voltage, particularly preferably by varying the spectrum according to claim 6, comprising.

To solve the invention sees in a further independent form, which is optionally combinable with the above ideas, however, that artifacts are detected and corrected by comparing the measured radiographic images with generated by forward projection radiographic images.

Particularly noteworthy is therefore an inventive method for examining a workpiece, preferably dimensional measurement of features on the workpiece by computer tomography comprising at least the following steps: Recording at least a first set of multiple radiographic images of the arranged on a turntable workpiece in several rotational positions (rotational steps) relative to a source of radiation (source) such as an X-ray source and a detector such as X-ray detector (first computed tomography), reconstruction of a voxel volume (volume data) having voxels with assigned voxel gray values from the transmission images and preferably determination of surface measurement points from the voxel volume by means of surface extraction methods preferably one of the preceding ideas, which is characterized in that in a multi-slice tomography or multi-spectral tomography recorded radiographic image it is compared with front-projection radiographic images of the same position and rotational position of the workpiece, wherein the forward projection is on target data such as CAD data and / or volume data reconstructed from at least one set of radiographic images, and wherein the comparison identifies the image components in the acquired radiographic images which have artifacts, wherein preferably the identified image components are replaced by the corresponding image components of the forward projection generated transmission images.

However, as a further independent solution, which is optionally combinable with the aforementioned ideas, the invention provides that a multi-spectral tomography is performed.

• - Variation der Röntgenspannung bzw. Beschleunigungsspannung der Röntgenröhre und/oder
• - Variation des Materials des die beim Auftreffen von Elektronen erzeugenden Targets der Röntgenröhre und/oder
• - Variation des Materials und/oder der Dicke des zur Filterung der von der Röntgenröhre abgegebenen Messstrahlung verwendeten Strahlfilters, welcher vor dem Auftreffen der Messstrahlung auf das Werkstück angeordnet ist und/oder
• - Variation des Materials und/oder der Dicke des zur Filterung der von der Röntgenröhre abgegebenen Messstrahlung verwendeten Strahlfilters, welcher nach dem Durchdringen des Werkstücks mit der Messstrahlung und vor dem Auftreffen auf den Detektor angeordnet ist und/oder
• - Variation des Materials und/oder der Dicke der Szintillatorschicht des Detektors und/oder
• - Variation der Brennflecklage auf dem Target der Röntgenröhre.

In particular, the invention is therefore also characterized by a solution according to the invention comprising a method for examining a workpiece, preferably dimensional measurement of features on the workpiece, by means of computer tomography comprising at least the following steps: recording at least a first set a plurality of radiographic images of the workpiece arranged on a turntable in several rotational positions (rotational steps) relative to a radiation source (source) such as X-ray source and a detector such as X-ray detector (first computed tomography or first tomography), reconstruction of a voxel volume (volume data) having voxels with assigned voxel gray values , From the radiographic images and preferably determination of surface measurement points from the voxel volume by surface extraction method, preferably one of the preceding ideas, which is characterized in that in a multi-spectral tomography (MSP-CT), in which at least two tomographies with different spectrum of the X-ray tube and the results of the tomographs are combined, combining either radiographic images per rotational position or volume data reconstructed per tomography, the variation of the spectrum of the x-ray tube being carried out by:

• Variation of the x-ray voltage or acceleration voltage of the x-ray tube and / or
• - Variation of the material of the electron-generating targets of the X-ray tube and / or
• Variation of the material and / or the thickness of the beam filter used for filtering the measuring radiation emitted by the X-ray tube, which is arranged on the workpiece prior to the impact of the measuring radiation and / or
• Variation of the material and / or the thickness of the beam filter used for filtering the measuring radiation emitted by the X-ray tube, which is arranged after penetrating the workpiece with the measuring radiation and before hitting the detector and / or
• Variation of the material and / or the thickness of the scintillator layer of the detector and / or
• - Variation of the focal spot on the target of the X-ray tube.

To solve the invention provides in a further independent expression, which is however optionally combined with the above ideas before that a multi-layer tomography is performed.

• - Verschiebung des Werkstücks, vorzugsweise gemeinsam mit dem Drehtisch, oder gemeinsame Verschiebung der Quelle und des Detektors in Richtung der Drehachse des Drehtisches und/oder in der Richtung senkrecht zur Richtung der Drehachse des Drehtisches, die senkrecht zur mittleren Strahlrichtung der Quelle verläuft,
• - Verschiebung des Werkstücks gemeinsam mit dem Drehtisch oder auf dem Drehtisch, oder gemeinsame Verschiebung der Quelle und des Detektors, in der Richtung senkrecht zur Richtung der Drehachse des Drehtisches, die senkrecht zur mittleren Strahlrichtung der Quelle verläuft, wobei in zumindest einer Lage des Werkstücks dieses um eine außerhalb der Mittelachse des Werkstücks liegende Drehachse gedreht wird, Werkstück also mit unterschiedlicher Exzentrizität bezüglich einer festen Drehachse tomografiert wird, wobei vorzugsweise für den Fall der gemeinsamen Relativ-Verschiebung von Drehtisch und Werkstück in Bezug auf Quelle und Detektor während der Drehung des Drehtisches dieser oder Quelle und Detektor so auf einer Kreisbahn um die feste Drehachse bewegt werden, dass das Werkstück um die feste Drehachse taumelt,
• - Verschiebung von Quelle und Detektor relativ zueinander in Richtung der Drehachse des Drehtisches und/oder in der Richtung senkrecht zur Richtung der Drehachse des Drehtisches, die senkrecht zur mittleren Strahlrichtung der Quelle verläuft und/oder in Richtung der mittleren Strahlrichtung der Quelle,
• - Verschiebung des Werkstücks gemeinsam mit dem Drehtisch, oder gemeinsame Verschiebung der Quelle und des Detektors, in der Richtung senkrecht zur Richtung der Drehachse des Drehtisches, die senkrecht zur mittleren Strahlrichtung der Quelle verläuft, wobei zwei Positionen eingenommen werden, in denen jeweils zumindest eine Seite oder Hälfte des Werkstücks auf dem Detektor abgebildet wird (Halbseiten-Tomografie), also Bereiche des Werkstücks den vom Detektor erfassten Bereich in zumindest einigen Drehstellungen verlassen, wobei Verschiebung bevorzugt geringer als Werkstückausdehnung in Verschieberichtung gewählt wird, so dass sich ein zumindest geringfügiger Überlapp ergibt, wobei besonders bevorzugt je Halbseiten-Tomografie zusätzlich veränderte Lage bzw. Orientierung eingestellt wird durch eine Veränderung der Schwenkstellung des Werkstücks auf dem Drehtisch und/oder eine Veränderung der Schwenkstellung des das Werkstück aufnehmenden Drehtisches,
• - Aufnahme des ersten Satzes von Durchstrahlungsbildern in ersten Drehstellungen in einem ersten Winkelbereich kleiner als eine volle Umdrehung, insbesondere in einem Winkelbereich von 0° bis einer halben Umdrehung (180°) zzgl. Kegelwinkel der von der Röntgenröhre abgegebenen Messstrahlung (erste Halbwinkel-Tomografie), und Aufnahme des zweiten Satzes von Durchstrahlungsbildern in zweiten Drehstellungen in einem zweiten Winkelbereich kleiner als eine volle Umdrehung der sich zumindest teilweise von dem ersten Winkelbereich unterscheidet, insbesondere in einem Winkelbereich von 180° bis einer halben Umdrehung (360°) zzgl. Kegelwinkel der von der Röntgenröhre abgegebenen Messstrahlung (zweite Halbwinkel-Tomografie), wobei besonders bevorzugt je Halbwinkel-Tomografie zusätzlich veränderte Lage bzw. Orientierung eingestellt wird durch eine Veränderung der Schwenkstellung des Werkstücks auf dem Drehtisch und/oder eine Veränderung der Schwenkstellung des das Werkstück aufnehmenden Drehtisches,
• - Variation der Brennflecklage auf dem Target der Röntgenröhre.

According to a particularly noteworthy proposal, the invention therefore also provides a method for examining a workpiece, preferably dimensional measurement of features on the workpiece, by means of computed tomography comprising at least the following steps: recording at least a first set of several radiographic images of the workpiece arranged on a turntable in a plurality of rotational positions (Rotation steps) relative to a radiation source (source) such as X-ray source and a detector such as X-ray detector (first computed tomography or first tomography), reconstruction of a voxel volume (volume data) having voxels with assigned voxel gray values, from the transmission images and preferably determination of surface measurement points from the Voxel volume by surface extraction method, preferably according to one of the preceding ideas, which is characterized in that one or more other sets of transmission B The results of the tomographies for reducing and / or correcting artifacts are combined, wherein either radiographic images per rotational position or volume data reconstructed per tomography are combined , wherein the respectively changed position or orientation is adjusted by one of the following variants or by any combination of these:

• Displacement of the workpiece, preferably together with the turntable, or joint displacement of the source and the detector in the direction of the axis of rotation of the turntable and / or in the direction perpendicular to the direction of the axis of rotation of the turntable, which is perpendicular to the mean beam direction of the source,
• - Displacement of the workpiece together with the turntable or on the turntable, or common displacement of the source and the detector, in the direction perpendicular to the direction of the axis of rotation of the turntable, which is perpendicular to the mean beam direction of the source, wherein in at least one position of the workpiece this order a rotation axis lying outside the central axis of the workpiece is rotated, that is workpiece is tomographed with different eccentricity with respect to a fixed axis of rotation, preferably in the case of relative displacement of rotary table and workpiece with respect to source and detector during rotation of the turntable Source and detector are moved on a circular path about the fixed axis of rotation, that the workpiece wobbles about the fixed axis of rotation,
• Displacement of source and detector relative to each other in the direction of the axis of rotation of the turntable and / or in the direction perpendicular to the direction of the axis of rotation of the turntable, which is perpendicular to the mean beam direction of the source and / or in the direction of the mean beam direction of the source,
• - Movement of the workpiece together with the turntable, or common displacement of the source and the detector, in the direction perpendicular to the direction of the axis of rotation of the turntable, which is perpendicular to the mean beam direction of the source, wherein two positions are occupied, in each case at least one side or half of the workpiece is imaged on the detector (half-side tomography), ie areas of the workpiece leave the area detected by the detector in at least some rotational positions, wherein displacement is preferably selected to be less than workpiece extension in the direction of displacement, so that an at least slight overlap results, wherein particularly preferably each half-side tomography additionally changed position or orientation is adjusted by a change in the pivot position of the workpiece on the turntable and / or a change in the pivot position of the workpiece receiving turntable,
• - Recording of the first set of radiographic images in first rotational positions in a first angular range smaller than a full revolution, in particular in an angular range of 0 ° to half a revolution (180 °) plus cone angle of the radiation emitted by the X-ray tube (first half-angle tomography) , and recording the second set of transmission images in second rotational positions in a second angular range smaller than a full revolution which at least partially differs from the first angular range, in particular in an angular range of 180 ° to one-half revolution (360 °) plus the cone angle of the X-ray tube emitted measuring radiation (second half-angle tomography), particularly preferably each half-angle tomography additionally changed position or orientation is adjusted by changing the pivot position of the workpiece on the turntable and / or a change in the pivot position of the workpiece record the turntable,
• - Variation of the focal spot on the target of the X-ray tube.

As a result of the displacement of the workpiece in the direction of the axis of rotation of the turntable, this also makes it possible, in particular, to reduce cone beam artifacts, since the upper and lower end of the workpiece can be measured in the center of the beam and thus influenced by cone beam artifacts.

As a result of the displacement of the source and the detector relative to one another in the beam direction, this also results in particular in the possibility that the cone angle is chosen differently, that is to say the transmission angle varies, as a result of which cone beam artifacts have a different effect.

By the tomography with different eccentricity with respect to a fixed in the beam center fixed axis of rotation, for example, a first tomography, once again results in changed radiographic directions and lengths for the multi-slice tomography. It is tomographed at the same magnification. In the first embodiment, the workpiece is arranged once in the center and once eccentrically, ie eccentrically on the turntable, and wobbles about the axis of rotation of the stationary turntable. It is also intended to set only different eccentricities, ie to dispense with the central tomography (eccentricity = 0). In the second embodiment, the workpiece is arranged centrally on the turntable, but the turntable itself wobbles about a virtual axis of rotation, namely the imaginary fixed axis of rotation. This allows the stepless adjustment of the eccentricity without an additional displacement device, such as XY stage on the turntable, for adjusting the eccentricity of the workpiece on the turntable, since there are usually measuring axes for moving the turntable in the respective directions perpendicular to the direction of the axis of rotation of the turntable.

It should also be emphasized that the respectively changed position or orientation for a multi-slice tomography is set by a change in the pivot position of the workpiece on the turntable and / or a change in the pivot position of the turntable receiving the workpiece.

Preferably, the invention provides that each changed position or orientation is set for a multi-slice tomography, based on a simulation of the position-dependent transmission lengths first position is chosen so that as many short transmission lengths are present and second layer is chosen so that as many long transmission lengths are present, tomography preferably in the first position with the first spectrum and tomography in the second position with deviating from the first spectrum second spectrum, wherein the first spectrum compared to the second spectrum lower incident on the workpiece and / or the detector beam power and / or acceleration voltage having.

However, as a further independent solution, which can be combined optionally with the aforementioned ideas, the invention provides that the position and / or spectrum are alternately varied with the rotation steps.

Therefore, according to an idea according to the invention, a method for examining a workpiece, preferably dimensional measurement of features on the workpiece, by computer tomography comprising at least the following steps: recording at least a first set of several radiographic images of the workpiece arranged on a turntable in a plurality of rotational positions ( Rotation steps) relative to a radiation source (source) such as X-ray source and a detector such as X-ray detector (first computed tomography), reconstruction of a voxel volume (volume data) having voxels with assigned voxel gray values from the transmission images and preferably detection of surface measurement points from the voxel volume by surface extraction method, according to preferably one of the preceding ideas, which is characterized in that a multi-spectral tomography (MSP-CT) or a multi-layer tomography od a combined multi-spectral tomography (MSP-CT) and multi-slice tomography in which at least a first tomography in a first position with a first spectrum and a second tomography in a deviating from the first position second position with a different from the first spectrum second spectrum is carried out, wherein the change of the position and / or the spectrum in each case after one or a group of successively set rotational steps, preferably alternately, wherein group successively set rotational steps each less than a full revolution comprises, and wherein initially radiographic images with identical spectrum and identical position can be reconstructed separately to volume data, which are then combined to form a resulting voxel volume.

The initial reconstruction of respective subsets and subsequent correction has the advantage that the measurement is less sensitive to drift phenomena.

Further details, advantages and features of the invention will become apparent not only from the claims, the features to be taken from them - alone and / or in combination - but also from the following description of the figures.

• 1 eine Prinzipdarstellung des erfindungsgemäßen Verfahrens mit verändertem Abstand zwischen Werkstück und Quelle eines Computertomografen, abhängig von der jeweiligen Drehstellung, insbesondere für den Fall einer Region of Interest-Computertomografie (ROI-CT),
• 2 eine Prinzipdarstellung der erfindungsgemäßen Vorrichtung eines Multisensor-Koordinatenmessgerätes mit Computertomograf und zusätzlichem Sensor
• 3 eine Prinzipdarstellung des erfindungsgemäßen Verfahrens zur Bestimmung des Rauschens und/oder der Messpräzision und/oder zumindest einer Komponente der Messunsicherheit eines Messpunktes ohne Berücksichtigung der Unsicherheit des Schwellwertes,
• 4 eine Prinzipdarstellung des erfindungsgemäßen Verfahrens zur Bestimmung des Rauschens und/oder der Messpräzision und/oder zumindest einer Komponente der Messunsicherheit eines Messpunktes mit Berücksichtigung der Unsicherheit des Schwellwertes und
• 5 die erfindungsgemäße Computertomografievorrichtung mit Strahlenschutzgehäuse.

It shows:

• 1 a schematic representation of the method according to the invention with a different distance between the workpiece and the source of a computer tomograph, depending on the respective rotational position, in particular in the case of a region of interest computed tomography (ROI-CT),
• 2 a schematic diagram of the inventive device of a multi-sensor coordinate measuring machine with computed tomography and additional sensor
• 3 a schematic diagram of the inventive method for determining the noise and / or the measurement precision and / or at least one component of the measurement uncertainty of a measuring point without taking into account the uncertainty of the threshold,
• 4 a schematic diagram of the inventive method for determining the noise and / or the measurement precision and / or at least one component of the measurement uncertainty of a measuring point with consideration of the uncertainty of the threshold and
• 5 the computed tomography device according to the invention with radiation protection housing.

Based on 1 the process of the invention is exemplified. Between the radiation source 1 with radiation emitting focal spot 1a and the detector 2 is the object to be measured like workpiece 4 on a turntable 3 arranged. The physical axis of rotation 3a of the turntable points into the drawing plane. Hatched represents the highly accurate area to be measured 5 of the workpiece 4 ie the region of interest (ROI). In the case of the ROI computed tomography, also referred to here as recording the first set of several radiographic images, the workpiece becomes 3 , especially the ROI 5 irradiated in several rotational positions, particularly preferably in rotational positions over an angular range of 360 ° plus the cone angle of the source 1 delivered and from the detector 2 detected measuring radiation 7 , Exemplary is the workpiece 4 in the rotational positions 4 - 1 . 4 - 2 . 4 - 3 and 4 - 4 shown. As a rule, radiographic images are recorded in several hundred to a few thousand rotational positions.

To the ROI 5 to be able to measure with high accuracy, is provided according to the invention to record in as many rotational positions radiographic images with high resolution, ie large imaging scale (FDD / FOD), in particular FOD set as small as possible. In the example of 1 this is for the rotational positions in the angular range between the positions 4 - 4 about 4-2 to 4-3 clockwise possible, also referred to here as the first rotational positions. The turntable 3 is in the position 3 - 2 whose magnification is the ratio of the length of the lines FDD (Focus detector distance) to FOD2 (Focus object distance). Another turning would collide the workpiece 4 with the source 1 to lead. To avoid this collision, the turntable becomes 3 with the workpiece 4 for the remaining rotational positions between the positions 4 - 3 to 4 - 4 clockwise, here as well second rotational positions referred to the position 3 - 1 along the line 6 towards the detector 2 shifted, in which then the reduced magnification FDD / FOD1 is present.

Preference is also intermediate positions for the turntable 3 between the positions 3 - 2 and 3 - 1 provided, in extreme cases, a separate setting for each rotational position. The intermediate positions are chosen so that in each case no collision between the workpiece 4 and source 1 occurs or a predetermined minimum distance (safety distance) remains. This results in a continuous adjustment of the image scale. Due to the associated expense, a stepwise adjustment is alternatively provided, wherein at least 2 stages are provided, as in the example of 1 is shown. As a boundary condition, however, it must always be observed that the ROI 5 within the beam cone 7 remains, ie in all rotational positions on the detector 2 is shown. As a result, in addition to possible collisions, the workpiece 4 not be moved arbitrarily close to the detector.

To complete a reconstruction of the workpiece 4 To realize, according to a particular embodiment of the invention is also intended to perform an overview CT, in which the workpiece 4 with a small magnification (FDD / FOD3) is completely measured, so in any of the rotational positions of the beam cone 7 leaves. This measurement, also referred to here as a recording of the second set of several radiographic images, takes place further in the line 6 towards the detector 2 to shifted position 3 - 3 ,

The 2 shows a multi-sensor coordinate measuring machine according to the invention 10 that at least a computer tomograph 11 , a controller 13 , an additional sensor 12 , here for example a non-contact optical sensor, the measuring axes 15 and a shield such as lead plate 14 includes. The control 13 is for the at least temporarily simultaneous measurement with the computer tomograph 11 and the additional sensor 12 educated. This means that the measurement with the additional sensor 16 among other things, in the period in which the multiple rotational positions of the workpiece 4 taken and the radiographic images with the X-ray detector 2 be recorded. During this period, the X-ray source is 1 switched on. By taking the multiple rotational positions, the outer contour of the workpiece 4 with the additional sensor 12 be recorded. At least secondarily means that the measurement with the additional sensor does not necessarily take place in all rotational positions or during the recording of the radiographic images themselves.

The computer tomograph 11 has the X-ray source 1 , the X-ray detector 2 and the turntable 3 that's about the axis of rotation 3a is rotatable and the workpiece 4 picks up. From the X-ray source 1 becomes the measuring radiation 7a delivered, with the proportion 7 detected by the detector 2 and thus for measuring the workpiece 4 is available. To the additional sensor 12 before the direct radiation 7a to protect, this is at least when the radiation 7a the X-ray source 1 is turned on, through the shield 14 covered. This is the sensor 12 preferably by means of the measuring axes 15 in the position shown 16 positionable. Alternatively, it is also provided, the shield 14 self-positionable to design. Through the measuring axes 15 can the sensor 12 be positioned so that the workpiece 4 , For example, the outer contour, can be detected, for example, measuring points on the surface in the transmitted light or reflected light are measured, for example, the roundness is measured.

The measurement data of the additional sensor ( 12 ) are thus immediately corrected for the measurement data of the computer tomograph ( 10 ) to disposal.

Based on 3 and 4 an expression of the method according to the invention is exemplified. Shown is the course of the gray values 100 the voxel (voxel gray values) in the region of an edge, ie the transition between the object and the surrounding medium, such as air in an exemplary direction, for example a spatial direction. The course between any two media, for example two materials in a multi-material component corresponds to the fundamentally same course. Plotted are the gray values 100 along the voxel grid 101 , The abscissa denotes the location s, the ordinate the gray value in gv (gray value) of, for example, 0 to 255 ( 8th bit resolution).

In the gray value course 100 we have a compensation curve 102 included, for example, a Gaussian curve that represents a smoothed or filtered gray scale gradient. From the intersection of the compensation curve 102 with the threshold gray value or threshold value 103 the position of the measuring point results 104 in the direction s. The threshold 103 may be local and, for example, the location of the inflection point of the compensation curve 102 or from the maximum and minimum gray value of the gray value curve 100 be derived in the area of the edge, or set globally, wherein before the application of the global threshold value, all gray values corresponding to the local gray scale gradients are adapted such that the application of the global threshold results in the identical location s for the respective measurement point, as is the case with the local threshold of the case would. In 3 First, it is assumed that there is no uncertainty associated with the applied threshold.

To the voxel gray values 100 around is now the upper envelope 105a and the lower envelope 105b adjusted so that the variation, so the noise of the voxel gray values 100 detected, so is included. This is an estimate of the gray-scale variation from voxel to voxel. This variation becomes all voxels 100 assigned as noise in the gray area, as exemplified by the areas 106 . 107 and 108 is marked. These areas are not centered on the already noisy voxel gray value, but on the compensation curve 102 so that the noise is not considered twice. The noise of the voxel gray values 100 So it's through the envelopes 105 estimated. For the inner edge voxels 100a and 100b results in the noise 108 respectively. 107 , Will the through the envelopes 105 limited area in the measuring point 104 considered, in the direction s, the width 110 results through the intersections 109 of the threshold 103 with the envelopes 105 , 110 designates the dimensional noise of the measuring point 104 ,

Alternatively, envelopes 105 also only in the two border areas 111 and 112 definable size, which lie in both directions after the actual material transition (the actual edge) are formed. The resulting gray-scale noise is then applied to the inner boundary voxels 100a and 100b transmitted, again represented by the areas 107 and 108 , These are connected linearly according to a simple procedure and result in an adapted envelope in the area of the inner edge voxels 100a and 100b from which the dimensional noise 110 of the measuring point 104 is derived. As a result, the possibly difficult to estimate variation of the gray values of the inner edge voxels 110a and 100b itself does not enter the envelopes 105 one. However, alternative methods are also provided where the variation of the gray levels of the inner edge voxels 110a and 100b is estimated otherwise, for example by simulation.

In 4 In addition, an uncertainty for the applied threshold 103 taken into account, represented by the range of variation between the lines 103a (upper threshold limit) and 103b (lower threshold limit). To calculate the then enlarged dimensional noise 110 become the intersections 109 the upper envelope 105a with the lower threshold 103b and the lower envelope 105b with the upper threshold limit 103a used. Generalized are the intersections 109 the threshold limits 103a and 103b with the envelopes 105 taken in the direction s furthest from the measuring point 104 lie away.

In the 5 is the computed tomography device according to the invention 1 shown. This includes the radiation protection housing 2 , in which at least the workpiece 4 receiving and rotating turntable 3 (preferably air-bearing), that of the X-ray source 5 discharged and from the workpiece 4 attenuated X-radiation 12 receiving x-ray detector 10 that the X-rays 12 emitting X-ray source 5 and the for the high voltage power supply of the X-ray source 5 designed generator 6 , And preferably for evacuation of the vacuum tube 5b provided vacuum pump 5c. To conventional power supply ( 220 V) of the generator 6 is this over a line 6a with a power supply outside the radiation protection housing 2 connected. The generator 6 generated high voltage is through the contacts 7 to the directly flanged X-ray tube 5 , in particular its vacuum tube 5b transfer.

At the X-ray source 5 In particular, it is a transmission tube in an open design, the long life and easy replacement of wearing parts such as the filament yarns 5a , here according to the invention with three filaments, for example, in order to reduce the current density and thus the wear, and that with the filament 5a emitted electrons (e ^- ) bombarded and the X-ray 12 generating target 11 allows. In particular, it is the transmission tube 5 a tube with high maximum power, in particular greater than 40 W, and high maximum acceleration voltage, preferably greater than 150 kV.

For relative positioning of turntable 3 , X-ray detector 10 and x-ray tube 5 to each other are measuring axes 8th provided here, for example, for each of the three components in three spatial directions, indicated by the corresponding three arrows, are formed. However, alternative embodiments are also provided in which only individual or in each case two spatial directions for one or two of the components are X-ray tubes 5 , Turntable 3 and x-ray detector 10 are provided.

Further shown is an example of an additional sensor 9 , here embodied as a tactile sensor. This can be done with not shown own measuring axes relative to the workpiece 4 be designed to be movable, or it will, as shown, inter alia, the measuring axes 8th that is the X-ray detector 10 are assigned, for the movement of the sensor 9 relative to the workpiece 4 used to take a measurement of the workpiece 4 to realize with the sensor 9. Through the measuring axes 8th is the Computed tomography apparatus 1 designed as a coordinate measuring machine, in connection with other sensors, such as the sensor shown here by way of example 9 as a multi-sensor coordinate measuring machine.

For measuring the workpiece 4 the invention provides that a control and evaluation unit, not shown here, is provided which controls the rotation of the workpiece 4 by means of the turntable 3 controls an axis running vertically in the image plane. In different rotational positions of the workpiece 4 be with the x-ray detector 10 Transmitted radiation images, which are reconstructed by the connected control and evaluation, in particular evaluation unit such as PC, to a voxel volume. From this, surface points for the determination of features, ie for the dimensional measurement of the workpiece 4 derived. Additionally or alternatively, it is also provided to solve material inspection tasks from the voxel volume, in particular the examination of the interior of the workpiece for voids and inclusions or the like.

Also provided is a parking station 14 with several places 15 in which automatically workpiece 4 or these receiving brackets 13 , which can also be designed as multiple workpieces receiving pallets, can be stored and resumed. Preferably, the loading and unloading of the turntable takes place 3 with the workpiece to be measured 4 by moving the measuring axes 8th that the turntable 3 assigned. Alternatively, it is also provided, separate, not shown here means such as grippers or robots in the radiation protection housing 2 to arrange. The loading is also provided by means of grippers mounted in place of the further sensor, such as magnetic grippers.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• EP 2399237 B1 [0006, 0303]
• DE 102013108152 A1 [0010]
• DE 102017100594 [0026, 0029, 0194]
• WO 2016/169589 [0027, 0029]
• DE 102008037441 A1 [0028, 0029]
• DE 102015114715 [0039]
• DE 102010017508 [0081, 0082]
• WO 2005119174 A1 [0106, 0117, 0153, 0310]
• DE 102006022103 A1 [0107, 0118]
• DE 102005039422 [0123]
• WO 2016/066265 [0161]
• WO 2016/042105 [0161]
• EP 2665035 A2 [0193]
• DE 102012102709 A1 [0236]
• WO 2013167616 A2 [0307]

Claims (26)

Method for examining a workpiece and / or one or more regions (ROI region of interest) of a workpiece, preferably dimensional measurement of features on the workpiece, by means of computed tomography comprising at least the following steps: Recording at least a first set of multiple radiographic images on a turntable arranged workpiece in several rotational positions (rotational steps) relative to a radiation source (source) such as X-ray source and a detector such as X-ray detector, reconstruction of a voxel volume having voxels with assigned voxel gray values, from the radiographic images and preferably determination of surface measurement points from the voxel volume by means of surface extraction method, characterized in that, to accommodate the first set of transmission images, the distance between the workpiece and the source: - is set lower for at least some first rotational positions than for some e second rotational positions, which preferably results in a higher magnification for the first rotational positions, and - is set for all rotational positions so that a collision between the detector and workpiece is avoided by taking into account the workpiece dimensions or the expected workpiece dimensions and the respective rotational position, the Turntable together with the workpiece and / or the workpiece relative to the turntable, and, the source or the source and the detector, relatively at least moved towards each other or away from each other. Method for determining features, in particular dimensional features, on a workpiece with a coordinate measuring machine, preferably a multi-sensor coordinate measuring device, in particular by means of a computer tomography machine which is pronounced as a coordinate measuring machine, wherein one or any combination of several of the following automated steps takes place: • Equipping the coordinate measuring machine with the workpiece or workpieces to be measured • Detection of the workpiece to be measured or the workpieces to be measured • Provision of a measurement program, wherein preferably a possibly existing measuring program is provided • Create and / or customize the measurement program • Execute the measuring program • Selection of an evaluation strategy and evaluation of the characteristics to be measured • Summary and management of the measurement results determined by the measuring program • Detection of faulty conditions and, if necessary, intervention in the production process or processes of the workpiece, wherein preferably the measurement results or measurements provided for a measurement protocol are selected by manual operator intervention. Method for evaluating surface measuring points generated by means of a sensor integrated in a coordinate measuring machine according to Claim 2 , characterized in that the following steps are carried out: • Assignment of the target data, in particular CAD model and / or PMI data of the workpiece, either by comparing the measurement data or previously determined surface measurement points stored in a database target data multiple workpiece or using a Identification feature such as RFID tag, barcode or referenced via an I / O interface target data from a database or individual workpiece assigned to the target data, and • selection of the features to be evaluated and reading the associated dimensional, shape and / or position tolerances on the basis of the assigned target data , and • selection of an evaluation strategy, preferably based on a database containing the characteristics associated with the characteristics for generating the evaluation strategies and evaluation of the measurement results for the features that can be determined from surface points and / or from those using a computer tomograph or computer tomography ors generated voxel volume traceable features such as voids or inclusions, flaws etc. and preferably summarizing and managing the measurement results, preferably creation of a measurement protocol and more preferably output to an interface for further evaluation, preferably CAQ interface, database or the like. Measurement software for generating a measurement program for the dimensional determination of features on workpieces, in particular for comparison with dimensional, shape and / or position tolerances originating from target data such as PMI or CAD data, preferably with a coordinate measuring machine, preferably according to at least one of Claims 2 to 3 Characterized in that the measuring software includes the rules, such as measurement and / or evaluation strategy for the determination of each feature and comparison with from target data, such as PMI or CAD data derived assigned to the respective characteristic dimensions, shape and / or position tolerances, and / or a database, and automatically generates a measurement program or proposes to the operator a measurement program for verification and / or modification. Data processing device designed to execute a measurement program by the measuring software according to Claim 4 , Apparatus for measuring and / or calibrating a computed tomography sensor or a coordinate measuring machine with at least one computed tomography sensor or a multisensor coordinate measuring machine with at least one computed tomography sensor using a normal body which is decomposed by the x-ray radiation of the radiation source of the computed tomography sensor, wherein the plurality of sections, preferably balls, the test specimen are arranged in a carrier material of significantly lower density than the density of the specimen or normal, characterized in that in the carrier material a plurality of openings, in particular holes and / or blind holes, introduced and / or more by lathering, Einpulvern or pressing the Carrier material resulting recesses are provided, wherein each opening or recess at least a portion of the normal or test specimen is introduced, and wherein the sections in the interior of the standard are as or test specimen, and in particular are not accessible for calibration or measurement with tactile, optical and / or tactile-optical sensor. Method for measuring and / or calibrating a computed tomography sensor or a coordinate measuring machine with a computed tomography sensor or a multisensor coordinate measuring machine with computed tomography sensor using the apparatus according to FIG Claim 6 , characterized in that the calibration of the test lengths formed by the plurality of sections of the standard or specimen by the distances, in particular spherical center distances of the sections of the normal or specimen by means of a computed tomography sensor, preferably the einzumessenden or to be calibrated computed tomography sensor, are measured and The distances of the second test body or normal are formed by test specimen sections, which are not surrounded by a carrier material and by means of tactile and / or with the determined by the same computed tomography sensor distances, in particular ball center distances, a second specimen or standard optical and / or tactile-optical sensors are calibrated. Method for measuring and / or calibrating a computed tomography sensor or a coordinate measuring machine with a computed tomography sensor or a multisensor coordinate measuring machine with computed tomography sensor using the apparatus according to FIG Claim 6 , characterized in that the calibration of the test lengths formed by the plurality of sections of the standard or test specimen prior to foaming, Einpulvern or pressing of the carrier material by the distances, in particular spherical center distances of the sections of the normal or specimen by means of tactile, optical and / or tactile-optical sensors are measured. Method for the dimensional measurement of a workpiece with a computer tomograph, consists of at least the radiation source (source), in particular X-ray source, detector, in particular X-ray detector, preferably flat X-ray detector, and the workpiece receiving and rotating in the beam path between the source and detector turntable, wherein in several rotational positions of the Workpiece images containing gray scale radiographs are taken with the detector and these are reconstructed into a voxel with voxel volume having gray values, from which surface measurement points (measurement points) are used by surface extraction methods, which are used for dimensional measurement, and for dimensional measurement with at least one tactile, optical and / or tactile-optical sensor (additional sensor), characterized in that the workpiece measured at least temporarily in the period with the additional sensor or additional sensors n, in which the multiple rotational positions are taken and the transmission images are taken, in which the X-ray source is thus switched on. Multisensor coordinate measuring machine (10) for the dimensional measurement of a workpiece (4), comprising a computer tomograph (11), consists of at least radiation source (source) (1), in particular X-ray source, detector (2), in particular X-ray detector, preferably flat X-ray detector, and the workpiece (4) receiving and in the beam path between the source (1) and detector (2) rotating turntable (3), and having at least one tactile, optical and / or tactile-optical sensor (additional sensor) (12), characterized in that the multi-sensor coordinate measuring machine (10) is designed to measure the workpiece (4) at least temporarily during the period of time with the additional sensor (16) or additional sensors, in which the plurality of rotational positions are taken and the transmission images are recorded, ie X-ray source (1) is turned on, in particular a suitable control device (13). Method for the dimensional measurement and / or examination of the object interior of an object, preferably a workpiece or tool by means of computed tomography, are taken in several rotational positions of the object receiving turntable about the axis of rotation radiographic images of the object, preferably in at least rotational positions of 0 ° to 180 ° plus cone angle of the X-radiation, particularly preferably 0 ° to 360 °, in particular 0 ° to 360 ° plus the cone angle of the X-ray radiation, wherein the object is located between the X-ray source and the surface X-ray detector, and wherein from the transmission images a voxel volume is reconstructed, which is examined with respect to the inner structures and / or are generated from the surface measurement points, characterized in that the position the object between the X-ray source and the X-ray detector in the direction of the line connecting the X-ray source and the center of the X-ray detector and / or in the direction perpendicular to the connecting line and perpendicular to the direction of the rotation axis for each rotational position is set separately, so that the entire object or to be measured Section of the object in each rotational position is completely imaged on the X-ray detector by using the respective recorded radiographic image or more previously recorded radiographic images or one of all or eini gene at the particular time already recorded radiographic images reconstructed voxel volume (real-time voxel volume) determines the location of the object and optionally adjusted. Method for the dimensional measurement and / or examination of the object interior of an object, preferably a workpiece or tool by means of computed tomography, are taken in several rotational positions of the object receiving turntable about the axis of rotation radiographic images of the object, preferably in at least rotational positions of 0 ° to 180 ° plus the cone angle of the X-ray radiation, particularly preferably 0 ° to 360 °, in particular 0 ° to 360 ° plus cone angle of the X-radiation, wherein the object is located between the X-ray source and the surface X-ray detector, and wherein a voxel volume is reconstructed from the radiographic images, with respect to the inner structures is examined and / or are generated from the surface measuring points, characterized in that for each rotational position, the parameters of the x-ray source, in particular acceleration voltage and / or x-ray tube current, and / or number of center of the image lungs and / or integration time of the respective radiographic image are adapted so that the recorded in the respective rotational position radiograph or normalized radiographic image or normalized logarithmic radiograph or normalized logarithmic filtered radiographic image - is neither overexposed nor too dark and / or - optimal SNR (Signal to Noise Ratio ) and / or - optimal CNR (Contrast to Noise Ratio), or adapted so that the real-time voxel volume or voxel volume reconstructed from all or the previously recorded radiographic images - optimal SNR (Signal to Noise Ratio) and / or - optimal CNR (Contrast to Noise Ratio), or that the radiographic images are optimized so that the relevant for the measurement task surface measurement points and / or the elements associated therefrom and / or the geometric characteristics determined therefrom aufwei as high a precision and / or measurement accuracy sen. Method for the dimensional measurement and / or examination of the object interior of an object, preferably a workpiece or tool by means of computed tomography, are taken in several rotational positions of the object receiving turntable about the axis of rotation radiographic images of the object, preferably in at least rotational positions of 0 ° to 180 ° plus the cone angle of the X-ray radiation, particularly preferably 0 ° to 360 °, in particular 0 ° to 360 ° plus cone angle of the X-radiation, wherein the object is located between the X-ray source and the surface X-ray detector, and wherein a voxel volume is reconstructed from the radiographic images, with respect to the inner structures is examined and / or are generated from the surface measuring points, characterized in that each rotational position resulting from the position of the object in the direction of the connecting lines between the X-ray source and X-ray detector voxel size and / or structure resolution is determined and compared with a predetermined value or range for the permissible voxel size or structure resolution and, in the case of deviations from the value or range, the following steps are performed before the final transmission image (s) is recorded for the respective rotational position: the relative position between object, X-ray source and X-ray detector parallel to the connecting line between X-ray source and X-ray detector is adjusted so that the resulting structure resolution reaches the predetermined minimum required structure resolution, and that in particular the necessary voxel size or range of voxel size is maintained, which preferably thereby resulting cone angle does not exceed a predetermined maximum allowable cone angle, but only a maximum so far, just before the object would collide with the X-ray source, and - in the final position in Richtu ng the connecting lines between X-ray source and X-ray detector is determined whether the If the object or the portion of the object to be measured is completely imaged on the X-ray detector, and if this is not the case, final radiographic images are formed in a plurality of positions offset perpendicular to the direction of the connecting lines between the X-ray source and the X-ray detector, in particular in the direction of the axis of rotation and / or perpendicular to Direction of the rotation axis, recorded so that the object or the measured portion of the object is completely detected. Method for the dimensional measurement of an object, preferably a workpiece or tool by means of computed tomography, in which a plurality of rotational positions of an object receiving turntable about the axis of rotation radiographic images of the object are taken, preferably in at least rotational positions from 0 ° to 180 ° plus cone angle of the X-ray, particularly preferably 0 ° to 360 °, in particular 0 ° to 360 ° plus cone angle of the X-ray radiation, wherein the object is between the X-ray source and X-ray detector having a pronounced surface, and wherein a voxel volume is reconstructed from the transmission images from which surface measuring points (measuring points) are generated , wherein the voxel volume has a plurality of voxels whose center is assigned a gray value (voxel gray value), characterized in that the noise and / or the measurement precision and / or at least one component of the measurement uncertainty of a measurement point is calculated w is derived from the noise of the voxel gray values which are associated with the voxels (inner edge voxels) surrounding the respective measurement point and used for edge location detection, in particular directly adjacent to the respective measurement point, wherein preferably the edge length of the corresponding voxels and / or the differences the voxel gray values of the corresponding voxels are included in the calculation. Method for the dimensional measurement of an object, preferably a workpiece or tool by means of computed tomography, in which a plurality of rotational positions of an object receiving turntable about the axis of rotation radiographic images of the object are taken, preferably in at least rotational positions from 0 ° to 180 ° plus cone angle of the X-ray, particularly preferably 0 ° to 360 °, in particular 0 ° to 360 ° plus cone angle of the X-ray radiation, wherein the object is between the X-ray source and X-ray detector having a pronounced surface, and wherein a voxel volume is reconstructed from the transmission images from which surface measuring points (measuring points) are generated , wherein the voxel volume has a plurality of voxels whose center is assigned a gray value (voxel gray value), characterized in that before the actual measurement, radiographic images of the object in rotational positions from 0 ° to 180 ° plus cone angle of the x-ray genstrahlen (Vorabdurchstrahlungsbilder) are recorded and from these a so-called "Detectability Map" is determined, and - compared with a set of predetermined conditions and optionally a response to the operator if one or more conditions are not met, and / or - the Detectability Map the predetermined conditions fulfilling and preferably optimal measurement parameters for the actual computer tomographic measurement are determined. Method for adjusting the structure resolution (voxel texture resolution limit) resulting from the effective detector resolution and the magnification in a computer tomography for examining a workpiece and / or one or more areas of a workpiece, preferably dimensional measurements of features on the workpiece A workpiece, wherein the computer tomography comprises at least the following steps: recording a plurality of radiographic images of the workpiece arranged on a turntable in several rotational positions (rotational steps) relative to a radiation source (source) such as X-ray source and a detector such as X-ray detector, in particular flat X-ray detector, reconstruction of a voxel volume the radiographic images and preferably determination of surface measuring points from the voxel volume by means of surface extraction method, characterized in that from the focal spot size, the from the sample acquisition parameters required for the measurement, determining the pattern resolution limit (focal spot texture resolution boundary) on the workpiece, and changing the voxel texture resolution boundary toward the focal spot texture resolution boundary, preferably adjusting it almost identically. Reconstruction method for a computed tomography for examining a workpiece, in particular for the dimensional determination of features on a workpiece, comprising at least the following steps: Recording at least a first set of multiple radiographic image of the arranged on a turntable workpiece in several rotational positions (rotational steps) relative to a radiation source (source ), such as X-ray source and a detector such as X-ray detector, preferably X-ray area detector, reconstruction of a voxel volume comprising voxels with assigned voxel gray values from the transmission images, and preferably determination of surface measurement points from the voxel volume by surface extraction methods, characterized in that the size of the voxels (voxel size) is locally different within the reconstructed voxel volume, so the voxel volume is therefore inhomogeneous, in particular where the reconstruction areawise with different voxel sizes, or voxel volumes of different voxel sizes are combined into a resulting voxel volume, or regions of different voxel size are generated in a reconstructed voxel volume, for example by resampling and / or combining voxels and / or interpolation methods. Method according to preferably Claim 17 characterized in that different regions of a voxel volume are reconstructed according to a predetermined order, in particular selected regions are temporally reconstructed before other regions, the selected regions preferably being the regions from which surface measurement points are determined for linkage to dimensions, more preferably regions Layers are perpendicular to the direction of the axis of rotation. Method according to preferably one of the preceding Claims 17 to 18 , characterized in that in the radiographic images local areas are summarized (binned) prior to the reconstruction, which in the reconstruction contribute to areas of the voxel volume for which a smaller voxel size is provided in comparison to other areas of the voxel volume. Computed tomography device (1) comprising a radiation protection housing (2), within which at least one turntable (3), at least one X-ray detector (10), preferably flat X-ray detector, and at least one X-ray source (5) comprising at least vacuum tube (5b), filament (5a ) and target (11), characterized in that the X-ray source (5) has a transmission target (11), that is designed as a transmission tube, and that the X-ray source (5) preferably as a so-called open so openable tube with exchangeable wear parts such For example, filament (5a) and / or target (11) is carried out, wherein the generator (6) required for supplying the x-ray source (5) with high voltage is arranged inside the radiation protection housing (2). Device according to preferably Claim 20 , characterized in that a computed tomography device (1) comprises a parking station (14) with a plurality of places (15) integrated in a radiation protection housing (2), wherein in each place (15) a workpiece (4) or one or more Workpieces (4) receiving workpiece holding device (13) (pallet) automatically from the turntable (3) off and from the respective place (15) on the turntable (3) is receivable by preferably the turntable (3) associated measuring axes (8 ) are arranged to move on the turntable (3) placed workpiece (4) or pallet (13) to a selectable place (15) of the parking station (14) and to couple with this and decoupled, for example by means of magnetic locking or by means of form-fitting generating movement, wherein the parking station (14) outside of the range of the X-ray detector (10) detected X-radiation (12) is arranged, preferably between the X-ray source (5) and Rön Detector (10) in the range of movement of the measuring axes (8). Method for examining a workpiece, preferably dimensional measurement of features on the workpiece, by computer tomography comprising at least the following steps: recording at least a first set of several radiographic images of the arranged on a turntable workpiece in several rotational positions (rotational steps) relative to a radiation source (source) such as X-ray source and a detector such as X-ray detector (first computed tomography or first tomography), reconstruction of a voxel volume (volume data) having voxels with assigned voxel gray values, from the transmission images and preferably determination of surface measurement points from the voxel volume by surface extraction method, characterized in that at least two of Following computed tomography method (second or more computed tomography or tomography, with other sets of radiographic images taken in several rotational positions who and / or correction of artifacts are combined: - multi-spectral tomography (MSP-CT), wherein at least two tomographies with different spectrum of the X-ray tube are made and the results of the tomographs are combined, using either radiographic images Multilayer tomography, in which at least two tomographies with different position or orientation of the workpiece relative to the source and detector take place and the results of the tomographs are combined, either radiographs per rotational position or per tomography reconstructed volume data are combined, - multi-focal spot tomography, in which at least two tomographies with different focal spot size of the X-ray tube, in particular by changing the X-ray tube performance, and / or focal spot are performed on the target of the X-ray tube and the results of the tomographies are combined, combining either radiographic images per rotational position or volume data reconstructed per tomography, correction method in which the characteristic curves are corrected from the gray values of the radiographic images and the assigned radiographic lengths and radiographs corrected with the corrected characteristic curve are calculated (empirical artifact correction) used to reconstruct the volume data, - correction method in which artifacts are simulated by simulating one or more of the Computed tomography, in particular the radiation occurring physical effects, preferably beam hardening, scattered radiation and / or cone beam artifacts are calculated and corrected (virtual artifact correction), preferably by the correction of the comparison of simulated radiographic images, volume data or surface measurement points is derived with and without the physical effects, and preferably one after the other in any desired order, and wherein the possibly necessary combination of the results of the plurality of computed tomographies is carried out by: - a voxel volume is reconstructed for each set of transmission images and a resulting voxel volume is generated from the plurality of voxel volumes or - the plurality of sets of transmission images, preferably each rotational position, combined to resulting radiographic images and these are reconstructed into a voxel volume or - at least two voxel volume combined who wherein at least one of the voxel volumes is reconstructed from a set of resulting radiographic images, the resulting radiographic images being generated from at least two sets of radiographic images, and where necessary applying the correction methods by: - the or each of the sets of Radiation images are corrected by the empirical artifact correction and / or virtual artifact correction or - respectively determined from the multiple voxel volume or from the resulting voxel volume by backprojection radiographic images by the empirical artifact correction and / or virtual artifact correction and then to each corrected voxel volume or the corrected resulting Voxel volumes are reconstructed or - each of the multiple voxel volumes or the resulting voxel volume is corrected by the virtual artifact correction, or - the surface measurement points through the virtual artifact correction is corrected and the surface measurement points are determined from the voxel volume or resulting voxel volume or corrected voxel volume or corrected voxel volume for dimensional measurement of the workpiece and / or the voxel volume or resulting voxel volumes or corrected voxel volumes or corrected resulting voxel volumes for voids, inclusions , Defects or the like is examined. Method for examining a workpiece, preferably dimensional measurement of features on the workpiece, by computer tomography comprising at least the following steps: recording at least a first set of several radiographic images of the arranged on a turntable workpiece in several rotational positions (rotational steps) relative to a radiation source (source) such as X-ray source and a detector such as X-ray detector (first computed tomography or first tomography), reconstruction of a voxel volume (volume data) having voxels with assigned voxel gray values, from the transmission images and preferably determination of surface measurement points from the voxel volume by surface extraction method, preferably Claim 22 Characterized in that recorded in a multi-layer tomography or more spectra tomography radiographic images are compared with generated by forward projection radiographic images of the same position and rotational position of the workpiece, wherein the forward projection reconstructed to set data such as CAD data and / or of at least of a set of radiographic images Volume data is carried out, and wherein the comparison identifies the image components in the recorded radiographic images having artifacts, wherein preferably the identified image components are replaced by the corresponding image components of the forward projection generated radiographic images. Method for examining a workpiece, preferably dimensional measurement of features on the workpiece, by computer tomography comprising at least the following steps: recording at least a first set of several radiographic images of the arranged on a turntable workpiece in several rotational positions (rotational steps) relative to a radiation source (source) such as X-ray source and a detector such as X-ray detector (first computed tomography or first tomography), reconstruction of a voxel volume (volume data) having voxels with assigned voxel gray values, from the transmission images and preferably determination of surface measurement points from the voxel volume by surface extraction method, preferably one of the preceding Claims 22 to 23 , characterized in that in a multi-spectral tomography (MSP-CT), in which at least two tomographies with different spectrum of the X-ray tube and the results of the tomographs are combined, wherein either radiographic images per rotational position or per tomography reconstructed volume data are combined, the variation the spectrum of the X-ray tube is effected by: - Variation of the X-ray voltage or acceleration voltage of the X-ray tube and / or - Variation of the material of the target of the electron-generating X-ray tube targets and / or - Variation of the material and / or the thickness of the for filtering the X-ray tube delivered Measuring radiation used beam filter, which is arranged prior to the impingement of the measuring radiation on the workpiece and / or - variation of the material and / or the thickness of the used for filtering the emitted radiation from the X-ray tube beam filter, which after penetrating the workpiece with the measuring radiation and before the impact on the detector is arranged and / or - variation of the material and / or the thickness of the scintillator layer of the detector and / or - variation of the focal spot on the target of the X-ray tube. Method for inspecting a workpiece, preferably dimensional measurement of Characteristics on the workpiece, by means of computer tomography comprising at least the following steps: recording at least a first set of several radiographic images of the arranged on a turntable workpiece in several rotational positions (rotational steps) relative to a radiation source (source) such as X-ray source and a detector such as X-ray detector (first computed tomography or first tomography), reconstruction of a voxel volume (volume data) having voxels with assigned voxel gray values from the transmission images and preferably determination of surface measurement points from the voxel volume by means of surface extraction methods, preferably one of the preceding ones Claims 22 to 24 characterized in that one or more further sets of radiographic images are taken at each changed position or orientation of the workpiece relative to the source and detector (multi-slice tomography) and the results of the tomographies are combined to reduce and / or correct artifacts, wherein either the radiographic images per rotational position or volume data reconstructed per tomography are combined, the respectively changed position or orientation being adjusted by one of the following variants or by any combination of these: displacement of the workpiece, preferably together with the rotary table, or joint displacement of the source and the detector in the direction of the axis of rotation of the turntable and / or in the direction perpendicular to the direction of the axis of rotation of the turntable, which is perpendicular to the mean beam direction of the source, - displacement of the workpiece together with the turntable or on the turn isch, or common displacement of the source and the detector, in the direction perpendicular to the direction of the axis of rotation of the turntable, which is perpendicular to the mean beam direction of the source, wherein in at least one position of the workpiece this is rotated about an axis lying outside the central axis of the workpiece axis of rotation Thus, in the event of joint relative displacement of turntable and workpiece with respect to source and detector during rotation of the turntable this or source and detector so on a circular orbit around the fixed Be moved axis of rotation, that the workpiece wobbles about the fixed axis of rotation, - displacement of the source and detector relative to each other in the direction of the axis of rotation of the turntable and / or in the direction perpendicular to the direction of the axis of rotation of the turntable, which is perpendicular to the mean beam direction of the source and / or in the direction of the mean beam direction of the source, - displacement of the workpiece together with the turntable, or common displacement of the source and the detector, in the direction perpendicular to the direction of the rotation axis of the turntable, which is perpendicular to the mean beam direction of the source, wherein two positions are taken, in each of which at least one side or half of the workpiece is imaged on the detector (half-side tomography), so leave areas of the workpiece detected by the detector area in at least some rotational positions, with displacement preferably less than workpiece extension in the direction chosen is so that there is an at least slight overlap, wherein particularly preferably each half-side tomography additionally changed position or orientation is adjusted by a change in the pivot position of the workpiece on the turntable and / or a change in the pivot position of the d receiving the first set of radiographic images in first rotational positions in a first angular range smaller than a full revolution, in particular in an angular range of 0 ° to half a revolution (180 °) plus cone angle of the measuring radiation emitted by the X-ray tube ( first half-angle tomography), and recording the second set of transmission images in second rotational positions in a second angular range smaller than a full revolution which at least partially differs from the first angular range, in particular in an angular range of 180 ° to one-half revolution (360 °). plus cone angle of the emitted radiation from the X-ray tube (second half-angle tomography), particularly preferably each half-angle tomography additionally changed position or orientation is adjusted by a change in the pivot position of the workpiece on the turntable and / or a change in Schw Adjustment of the turntable receiving the workpiece, - Variation of the focal spot position on the target of the X-ray tube. Method for examining a workpiece, preferably dimensional measurement of features on the workpiece, by computer tomography comprising at least the following steps: recording at least a first set of several radiographic images of the arranged on a turntable workpiece in several rotational positions (rotational steps) relative to a radiation source (source) such as X-ray source and a detector such as X-ray detector (first computed tomography or first tomography), reconstruction of a voxel volume (volume data) having voxels with assigned voxel gray values, from the transmission images and preferably determination of surface measurement points from the voxel volume by surface extraction method, preferably one of the preceding Claims 22 to 25 Characterized, in that a multi-spectra imaging (MSP-CT), or a multi-layer tomography, or a combined multi-spectra tomography (MSP-CT) and multi-layer tomography in which at least a first tomography in a first position with a first spectrum and a second tomography takes place in a second position deviating from the first position with a second spectrum deviating from the first spectrum, wherein the change in position and / or spectrum is in each case according to one or a group of successively set rotational steps, preferably alternating, where successively set rotational steps each comprising less than one full revolution, and wherein first transmission images with identical spectrum and identical position are each reconstructed separately to volume data, which are then combined to form a resulting voxel volume.

Images