Method and device for calibrating an imaging optic for metrological applications

Abstract

Method for calibrating an imaging optic (24) for metrological applications, comprising the steps: - Providing at least one calibration pattern (44) with a plurality of pattern areas (46a, 46b) that are in a defined relation to each other, - Taking at least one image of at least one calibration pattern (44) through the imaging optics. - Evaluating at least one image to quantify individual properties of the imaging optics (24), and - Determining correction values ​​for a computational correction of imaging errors of the imaging optics (24) depending on the quantified individual properties, wherein at least one individual property of the imaging optics (24) is quantified on the basis of the at least one calibration pattern (44), wherein the at least one calibration pattern (44) is provided on an electronic display (38) with a plurality of matrix-arranged display pixels (40), and wherein the electronic display (38) is able to generate different calibration patterns (44) successively using the matrix-arranged display pixels (40), wherein one of the different calibration patterns (44) is displayed, characterized in that a calibration body (50) is further provided on which at least one line (54) with a defined dimension is formed,wherein a magnification factor of the imaging optics (24) is determined on the basis of at least one line (54), and wherein at least one further individual property of the imaging optics (24) is quantified on the basis of at least one calibration pattern (44).

Description

[0001] The present invention relates to a method and a device for calibrating an imaging optic for metrological applications, comprising the steps of: - Providing at least one calibration pattern with a multitude of pattern areas that are in a defined relationship to each other, - Capturing at least one image of at least one calibration pattern through the imaging optics, - Evaluating at least one image to quantify individual properties of the imaging optics, and - Determining correction values ​​for a computational correction of imaging errors of the imaging optics depending on the quantified individual properties, wherein at least one individual property of the imaging optics is quantified on the basis of at least one calibration pattern, wherein at least one calibration pattern is provided on an electronic display with a plurality of matrix-arranged display pixels, and wherein the electronic display is able to generate different calibration patterns successively using the matrix-arranged display pixels, with one of the different calibration patterns being displayed.

[0002] Such a method and a corresponding device are known, for example, from US 2007 / 0106482 A1.

[0003] A method and a device according to DE 10 2006 034 350 A1 use a so-called calibration piece to generate light-dark transitions with a known distance between them. The calibration piece is an elongated, cuboid rod with a plurality of through-holes, the opening diameters of which differ on the top and bottom of the calibration piece. Some of the through-holes taper from the top of the calibration piece to the bottom. Other through-holes taper from the bottom to the top. This calibration piece is intended to eliminate measurement errors due to sealed edges and distortion due to a lack of parallelism. The known calibration piece is used in particular for calibrating a line scan camera, which is used to determine the edge position of material webs passing under the line scan camera.

[0004] DE 10 2004 020 881 A1 discloses a method and a device for geometrically calibrating a measuring camera, wherein a defined test structure is generated using a coherent, monochromatic light source and a double slit aperture. The double slit aperture is illuminated, for example, by a light source designed as a laser. The two slits of the double slit aperture each generate a diffraction pattern, which is recorded by the measuring camera to be calibrated.

[0005] DE 195 36 297 A1 describes a method for the geometric calibration of optical 3D sensors, using a special calibration block with a black and white striped pattern. This pattern is arranged orthogonally to a stripe pattern required by the optical 3D sensor for measurement. The calibration block may also have additional signal marks or reference marks that must be within the camera's field of view.

[0006] The known calibration devices and procedures each use special calibration blocks with known properties, some of which are specifically tailored to the measuring device being calibrated. The more precise and comprehensive the calibration needs to be, the greater the effort required to provide the necessary calibration blocks. On the other hand, there is a growing desire to use imaging optics for metrological applications as flexibly as possible and, accordingly, to calibrate them with respect to a wide variety of different imaging errors. This gives rise to diverse measurement tasks that, depending on the size, type, and properties of the workpieces, should be optimized in one direction or another.

[0007] DE 10 2010 000 745 A1 describes a calibration pattern for calibrating an image measuring instrument, and in particular for calibrating a camera-based measuring device for measuring the length of an object. The calibration pattern consists of concentric or spirally arranged square rings that provide alternating light-dark transitions in the x and y directions. The respective intervals of successive light-dark and / or dark-light transitions are known and serve to calibrate the optical measuring device. Depending on how the light-dark transitions appear on an object to be measured, the appropriate transitions in the calibration pattern can be used for calibration. DE 10 2010 000 745 A1 does not describe how the proposed calibration patterns are to be implemented in practice.However, the fact that the actual distances between the light-dark transitions and / or dark-light transitions are to be used for calibration suggests that these distances must be produced with high accuracy and reproducibility, which in turn suggests a calibration object specifically manufactured for this calibration.

[0008] The aforementioned US patent 2007 / 0106482 A1 discloses a method and a device for calibrating an image acquisition system with an imaging optic, wherein a series of calibration patterns are generated on an electronic display. The electronic display can be a flat panel display, an LCD screen, or similar. Several calibration patterns are acquired with the image acquisition system. The position of the image acquisition system and the electronic display relative to each other can be changed. From the data obtained, the internal and external parameters of the image acquisition system are fully determined.

[0009] WO 2014 / 023346 A1 discloses a device for inspecting a measurement object with an optical sensor. The optical sensor comprises a lens and a camera. Furthermore, the device includes a reflector and a calibration arrangement. The reflector can optionally be placed in a beam path passing through the lens to the camera, so that the optical measurement system has an internal optical reference for calibration. WO 2014 / 023346 A1 proposes this internal optical reference as an alternative to external calibration objects, since the latter can become dirty or scratched during use.

[0010] DE 10 2013 014 475 A1 discloses a method for calibrating an optical measuring device using a calibration object, namely a so-called physical object, whose appearance is changeable and controllable and whose structure or grid is known and constant. The calibration object can be an LCD display which has a fixed and known grid.

[0011] WO 02 / 039 055 A1 discloses a further calibration system for an optical arrangement with a calibration object integrated into the optical arrangement.

[0012] US 2012 / 0287287 A1 proposes a display in the form of a so-called display screen, on which various calibration patterns can be generated, as a calibration object for a camera.

[0013] Against this background, it is an object of the present invention to provide an alternative method and a corresponding device for calibrating an imaging optic for metrological applications, which enables the most comprehensive, flexible and cost-effective calibration possible with regard to diverse measurement tasks.

[0014] According to a first aspect of the present invention, this problem is solved by a method of the type mentioned at the outset, in which a calibration body is further provided on which at least one line with a defined dimension is formed - preferably permanently - wherein a magnification factor of the imaging optics is determined on the basis of the at least one line, and wherein at least one further individual property of the imaging optics is quantified on the basis of the at least one calibration pattern.

[0015] According to a further aspect of the invention, this problem is solved by a calibration device for calibrating an imaging optic for metrological applications, comprising an electronic display with a plurality of matrix-arranged display pixels, a pattern generator which, with the aid of the matrix-arranged display pixels, is able to generate a plurality of different calibration patterns, wherein the calibration patterns each have a plurality of pattern areas which are in a defined relation to one another, and a calibration unit which evaluates at least one image of at least one of these different calibration patterns taken by the imaging optic in order to quantify individual properties of the imaging optic and to determine correction values ​​for a computational correction of imaging errors as a function of the quantified individual properties, furthermore comprising a calibration body,on which at least one line with a defined dimension - preferably permanently - is formed, wherein the calibration unit is configured to determine a magnification factor of the imaging optics on the basis of the at least one line on the transparent plate and to quantify at least one further individual property of the imaging optics on the basis of the at least one calibration pattern.

[0016] It is particularly advantageous if the new calibration device is integrated into a measuring device for measuring various objects.

[0017] The new method and the corresponding device utilize an electronic display with which a variety of different calibration patterns can be generated flexibly and variably. In principle, the electronic display can be a tablet PC or a mobile phone, particularly a smartphone. In some embodiments, the calibration target with the line is a transparent plate that is either detachably or permanently connected to the electronic display to form an electronic calibration target. For example, a standard tablet PC can be equipped with a special holder for receiving a transparent plate with the precisely defined line. In some embodiments, the display, the holder, and / or the transparent plate can be assembled by the operator of the calibration device to perform a specific calibration task.For example, a standard tablet PC could be inserted into a mechanical holder that has precisely fitting mounting points for both the tablet PC and the transparent plate. In other embodiments, the calibration object can be a spherical gauge block with multiple spheres, a ring, a partial or full sphere, or another three-dimensional body on which at least one line is implemented as an edge, circumference, and / or distance measure between two features. Furthermore, the housing of the tablet PC or smartphone, or more generally, the display, can serve as a calibration object with the defined dimensions if these dimensions are measured on-site during calibration.

[0018] The pattern generator can generate various calibration patterns on demand using an algorithm stored within the generator and / or retrieve predefined calibration patterns from an integrated memory. Generation and retrieval can be time-controlled, performed manually by the operator, and / or triggered by a signal supplied to the pattern generator, for example, by the measuring device being calibrated. Accordingly, in some embodiments, the pattern generator may have an interface with an input for supplying the trigger signal.

[0019] In all practical embodiments, the calibration block provides an absolute, known length dimension using the line. It is advantageous if the length dimension is directly or indirectly traceable to international standards for the unit of length. The known dimension makes it possible to determine the magnification factor of the imaging optics to be calibrated with high accuracy. This is desirable if the imaging optics are to be used to determine spatial dimensions of objects with high accuracy.

[0020] The magnification factor, however, is only one of numerous properties of an imaging optic that are important for metrological applications. Another individual property is, for example, the distortion of the imaging optic, which becomes visible when the image of a measured object produced by the imaging optic is pincushioned or barrel-shaped. Distortion represents a location-dependent local magnification of the imaging optic, which can differ, particularly at the edge of the field of view, from that along the optical axis. To calibrate an imaging optic with respect to its individual distortion, large-area calibration patterns are desirable, which can be generated very easily and cost-effectively using the electronic display.

[0021] An electronic display cannot provide an absolute length measurement, or at best only with very high manufacturing costs, because the display pixels of an electronic display are subject to numerous manufacturing tolerances. The new method and the corresponding device therefore advantageously combine a simple, precise "length scale body," which in advantageous examples provides only the absolute length measurement, and an electronic display capable of generating various calibration patterns for further calibration tasks. In some advantageous embodiments, the calibration body with the length measurement can be a transparent plate on which the line is permanently applied, for example, using lithographic methods such as those employed in semiconductor technology.Alternatively or additionally, at least one line can be formed on or in the transparent plate using an electron beam writing process, for example, in a thin chromium layer. Advantageously, the calibration block is made of a material with a low coefficient of thermal expansion, such as fused silica, quartz, or Zerodur®. Zerodur® is a glass-ceramic material from Schott AG, Germany, which is known to have a very low coefficient of thermal expansion. Alternatively, the absolute length is measured on-site during calibration.

[0022] In preferred embodiments, the transparent plate has a broadband anti-reflective coating. This anti-reflective coating is advantageously effective across the entire spectral range of visible light. For infrared light, however, the transparent plate can be reflective to enable high-contrast recording of the calibration patterns through the plate and optimal detection of at least one line on the plate.

[0023] The calibration patterns that can be flexibly generated on the electronic display, while not precisely defined in their individual properties due to manufacturing tolerances of the display pixels, are nevertheless sufficiently consistent to allow calibration of the imaging optics by evaluating measurement series designed for consistency. For example, an individual calibration pattern can be captured with the imaging optics from different distances and / or different lateral positions. Since the calibration pattern itself remains the same on the electronic display, deviations in the obtained measurement results are attributable to the imaging optics and / or the relative displacement between the imaging optics and the calibration pattern.Therefore, despite the not entirely known properties of the calibration patterns, the electronic display allows for precise calibration with regard to image errors such as distortion, contrast, chromatic aberration, etc., in a relatively simple and cost-effective manner. In addition, the new calibration device uses a separate length measure for calibrating the magnification factor.

[0024] Overall, it has been shown that the new method and the corresponding device enable flexible calibration of imaging optics for metrological applications in a very cost-effective manner. The aforementioned problem has therefore been completely solved.

[0025] In a preferred embodiment of the invention, the electronic display has a top surface on which the calibration pattern is displayed, and at least one line is arranged above the calibration pattern on the top surface.

[0026] This design allows for a very cost-effective implementation, as the length scale can simply be attached to or above the electronic display. In preferred embodiments, a transparent plate has only a few lines; that is, the lines on the plate cover less than 10%, and preferably less than 2%, of the plate's transparent surface. In these embodiments, the transparent plate obstructs the view of the calibration pattern through the imaging optics only to a negligible extent. Furthermore, the view of the absolute scale through the imaging optics is unobstructed in this design; that is, the absolute scale is not affected by the display and the calibration pattern generated on it.

[0027] In another embodiment, the transparent plate serves as a cover plate for the electronic display.

[0028] In this configuration, the transparent plate with at least one line is fixed to or above the top of the electronic display. The transparent plate is an integral part of the display. This simplifies the handling of the new calibration device.

[0029] In a further embodiment, the at least one line is arranged offset from the display pixels in a top view of the calibration pattern. Advantageously, the at least one line has a width that is less than twice the width of a display pixel. In some embodiments, the width of the at least one line corresponds approximately to the lateral dimensions of a display pixel.

[0030] In this embodiment, at least one line is positioned above the display pixels of the electronic display in such a way that it does not completely cover the underlying display pixels. In a sense, the line runs between the display pixels, and depending on the size and spacing of the pixels, it may partially cover individual pixels. This embodiment has the advantage that, despite the overlying absolute scale, the calibration pattern remains virtually entirely available for quantifying at least one other individual property. The coverage of the calibration pattern by the absolute scale is reduced to a minimum.

[0031] In a further embodiment, the display pixels together define a maximum display area, wherein the imaging optics define a field of view, and wherein the electronic display is positioned at a distance from the imaging optics such that the field of view occupies a maximum of 3 / 4 of the maximum display area. Preferably, the field of view of the imaging optics is positioned largely centrally on the electronic display. Furthermore, it is preferred if the field of view of the imaging optics occupies a maximum of 2 / 3 of the maximum display area.

[0032] In this configuration, the display area of ​​the electronic display is larger than required for the imaging optics being calibrated. This allows for the generation of calibration patterns so large that the imaging optics can only image sections of these patterns. At first glance, this configuration appears disadvantageous, as it requires larger and therefore more expensive displays than are needed for calibrating the imaging optics. However, experience has shown that calibration patterns can be generated with greater homogeneity when the entire display area is not utilized. Therefore, this configuration enables high-precision calibration in a simple and ultimately cost-effective manner.

[0033] In a further embodiment, at least one image is captured with an electronic image sensor having a plurality of sensor pixels, wherein the number of display pixels is higher than the number of sensor pixels. Preferably, the ratio between the number of display pixels and the number of sensor pixels is approximately in the range of the assumed magnification factor (target magnification factor) of the imaging optics to be calibrated. In preferred embodiments, the number of display pixels is at least the magnification factor of the imaging optics higher than the number of sensor pixels, and particularly preferably, the number of display pixels is at least ten times higher than the number of sensor pixels.

[0034] A high pixel density in the pattern-generating display makes it easier to quantify the contrast gradient at light-dark edges or the "washing out" caused by the imaging optics. Determining the position of an edge is of great importance in modern image processing. The more precisely the position of an edge can be determined, the more accurately a measurement object can be measured. Accordingly, it is advantageous if the new calibration device can generate edge gradients that appear "continuous" to the electronic image sensor, even though they are generated using discrete display pixels. This criterion is easily met in the present design. It is particularly advantageous if the edges of at least one line on the transparent plate are so sharply defined that the threshold values ​​for edge detection can be calibrated using the line edge.The preferred method uses the edges of at least one line to calibrate the thresholds for edge detection. This design is advantageous because it enables cost-effective and comprehensive calibration of imaging optics for metrological applications.

[0035] In a further embodiment, the imaging optics have an optical axis and a field edge, wherein at least one line is positioned between the optical axis and the field edge. In some embodiments, the at least one line forms a circular ring that is positioned approximately midway between the optical axis of the imaging optics and its field edge.

[0036] Alternatively, in other embodiments, the at least one line could be positioned on or in the region of the optical axis or at the very edge of the field of view of the imaging optics. The preferred embodiment, however, has the advantage that the distortion, quantifiable using the at least one calibration pattern, is centered on a mean magnification factor, thereby minimizing the relative variation of the position-dependent magnification. This embodiment therefore enables calibration with higher accuracy while maintaining the same calibration effort. Furthermore, the design of the at least one line as a circular ring allows for the simple, direction-dependent determination and correction of scale errors in the imaging optics.

[0037] In a further embodiment, a reducing projection optic is arranged at the image-side end of the imaging optics.

[0038] This design is advantageous because the calibration pattern generated using the display pixels is reduced in size on the electronic image sensor, effectively reducing the "pixelation" of the electronic display and resulting in smoother contrast gradients. Therefore, this design provides easily optimized calibration patterns for flexible and highly accurate calibration of the imaging optics with respect to contrast blurring and edge detection thresholds.

[0039] In a further embodiment, each individual display pixel has a distinct emission characteristic, and an edge-shaped intermediate element is arranged between the at least one calibration pattern and the imaging optics. This intermediate element modifies the emission characteristic of the individual display pixels. It is particularly advantageous if the plate-shaped intermediate element reduces the emission characteristic of the individual display pixels, meaning that the light emitted by the display pixels is concentrated into a narrower solid angle than without the plate-shaped intermediate element. In some embodiments, the plate-shaped intermediate element is a filter disc or film, such as those used in notebooks and tablet PCs to hinder unauthorized side-viewing (so-called viewing-angle-restricting films, vicuity filters).In other embodiments, the plate-shaped intermediate element can have a plurality of microlenses and / or be a perforated aperture, wherein in these cases the plate-shaped intermediate element is also advantageously designed in such a way that it concentrates the emission characteristics of the individual display pixels in a direction perpendicular to the surface of the display.

[0040] This design has the advantage of achieving a higher light yield towards the imaging optics being calibrated. This allows for the creation of calibration patterns with very finely graduated contrast gradients, which is beneficial for flexible and precise calibration.

[0041] In a further embodiment, at least two images of the at least one calibration pattern are captured successively by the imaging optics, whereby the at least one calibration pattern is shifted and / or distorted between the capture of the at least two images on the electronic display.

[0042] In this embodiment, the position of the calibration pattern on the electronic display is changed "electronically" by shifting the calibration pattern generated on the display through appropriate control of the individual display pixels. The position of the display relative to the imaging optics, however, remains constant. Since the calibration pattern itself remains consistent, this embodiment advantageously allows the intrinsic distortion of the electronic display to be determined from the captured images. Advantageously, the shifting of all image elements of the calibration pattern occurs in increments of the display's division error. Alternatively or additionally, the distortion of the contrast pattern on the display can be amplified "electronically" in a determined manner by modifying the calibration pattern using a mathematically known function.By performing a suitable adjustment calculation between the known distortion and the observed distortion in the recorded images, the optical aberrations leading to the distortion and the display aberrations can be advantageously separated.

[0043] Furthermore, this design has the advantage that location-dependent imaging errors of the imaging optics can be quantified in a very simple and cost-effective way using the consistent calibration pattern.

[0044] In a further embodiment, the imaging optics are part of a measuring device with a defined measuring volume, and the electronic display is permanently arranged in the defined measuring volume.

[0045] In this configuration, the electronic display is an integrated part of the measuring device, whose imaging optics are to be calibrated using the electronic display. This configuration has the advantage that the calibration can be easily and conveniently repeated at any time during operation of the measuring device. It is particularly advantageous if an individual calibration is performed before each new measurement process.

[0046] In a further embodiment, at least one calibration pattern is generated on the electronic display depending on the object being measured.

[0047] In this configuration, the calibration pattern used depends on the object being measured. Accordingly, two different calibration patterns are generated on the display for two different objects. This configuration advantageously utilizes the flexibility offered by the new method and the new calibration device. For example, if a relatively small object is being measured, it can be advantageous to calibrate the imaging optics with a calibration pattern optimized for the area around the optical axis of the imaging optics. Larger imaging errors at the field edge of the imaging optics are then accepted as a benefit, since they are of minor importance for the measurement task.However, if a larger object is being measured, the image of which completely fills the field of view of the imaging optics, it is advantageous to optimize the calibration across the entire field of view, even if this leads to greater measurement uncertainties in the region of the optical axis. The present embodiment makes advantageous use of the possibilities of the new method and the new calibration device by employing a dynamic calibration adapted to the respective measurement situation.

[0048] It is understood that the features mentioned above and those to be explained below can be used not only in the combinations specified, but also in other combinations or on their own, without leaving the scope of the present invention.

[0049] Exemplary embodiments of the invention are shown in the drawing and are explained in more detail in the following description. They show: Fig. 1 a coordinate measuring machine with an integrated calibration device according to a preferred embodiment of the invention, Fig. 2 a schematic representation to illustrate preferred embodiments of the new method and the new calibration device, Fig. 3 a simplified representation of an image taken by the imaging optics to be calibrated, with a line structure that forms an absolute scale, and Fig. 4 a schematic representation showing a preferred positioning of at least one line across the display pixels of an electronic display.

[0050] In Fig. Reference numeral 10 is a coordinate measuring machine with a preferred embodiment of the new calibration device. The coordinate measuring machine 10 has a workpiece holder 12, which here is designed in the form of a cross table. The cross table has a lower part 14, which is movably mounted on guide rails (not shown here). The lower part 14 can move on the guide rails along an axis, which here is designated as the y-axis. An upper part 16 is arranged on the lower part 14, which is movably mounted on guide rails 18 along a second axis. The second axis of movement is here designated as the x-axis. The x- and y-axes are arranged orthogonally to each other, so that the workpiece holder 12 can move in two mutually orthogonal spatial directions to position a workpiece (not shown here) within a defined measuring volume.

[0051] The coordinate measuring machine 10 also has a column 20 on which a quill 22 is mounted for vertical movement. The vertical direction of movement of the quill 22 is referred to here as the z-axis. An imaging optic 24, which is part of an optical sensor used to measure workpieces, is arranged on the quill 22. Typically, the optical sensor includes a camera that captures an image of the object being measured through the imaging optic 24. This image is evaluated using image processing methods. In addition, the positions of the workpiece holder 12 along the x- and y-axes, as well as the position of the quill 22 along the z-axis, are usually evaluated in order to establish a spatial relationship between the features of the object being measured that were determined using image processing.

[0052] In this embodiment, the coordinate measuring machine 10 further comprises a tactile sensor on which a stylus 26 is arranged. The stylus 26 allows the coordinate measuring machine 10 to probe selected measuring points on a measuring object in order to determine the spatial coordinates of the probed measuring points within the measuring volume defined by the axes of motion.

[0053] Reference numeral 28 designates an evaluation and control unit that controls the movements of the workpiece holder 12 and the quill 22. Furthermore, the evaluation and control unit evaluates the signals supplied by the sensors and determines the desired properties of the measured object based on a user-defined measurement program. Reference numeral 30 designates a control panel through which an operator can input data into the evaluation and control unit.

[0054] The coordinate measuring machine 10 shown here is an example of a measuring device in which the new method and the new calibration device can be advantageously used. However, the new method and the new calibration device are not limited to the coordinate measuring machine 10. They can also be used with other measuring devices that differ from the illustration in Fig. 1. have a different design. Furthermore, the new method and the new calibration device can be advantageously used with measuring devices that determine the surface properties of a measured object, such as roughness, gloss level, or similar characteristics. In other words, the new method and the new calibration device are not limited to measuring devices that determine the geometric properties of a measured object.

[0055] According to a preferred embodiment, the coordinate measuring machine 10 here includes an integrated calibration device 32 (see Fig. 2), which includes an electronic display 38. In the preferred embodiment, the electronic display 38 is permanently arranged in the measuring volume of the coordinate measuring machine 10, so that calibration of the imaging optics 24 can be performed at any time, even within a measurement sequence or between successive measurement sequences. In other embodiments, the display 38 can be mounted on a workpiece pallet, such as those used for the automated feeding of workpieces to the measuring device. In this case, it is advantageous if the workpiece pallet has electrical contacts via which the display 38 can communicate with the evaluation and control unit 28 of the measuring device.

[0056] In the illustrated embodiment, the electronic display 38 is attached to the upper part 16 of the workpiece holder 12, and the upper part 16 can be moved along its axis of movement until the imaging optics 24 are positioned vertically above the electronic display 38. Subsequently, a calibration pattern provided by the electronic display 38 can be received and evaluated by the imaging optics 24 to be calibrated in order to determine individual properties of the imaging optics 24 and, depending on these properties, to calculate correction values ​​with which imaging errors of the imaging optics 24 can be computationally corrected.

[0057] Fig. Figure 2 shows the new calibration device in further detail. The display 38 has a plurality of pixels 40. A pattern generator 42 uses the display pixels 40 to create a calibration pattern 44, which has a plurality of pattern areas 46a, 46b. In the embodiment shown here, the pattern areas 46a, 46b are wavy lines that are in a defined relationship to each other. In principle, the calibration pattern 44 can have other pattern areas 46, such as concentric rings, square rings, stripes, grid structures, or combinations of such pattern elements. In the preferred embodiments of the new method, it is not important that the defined relationships between the individual pattern areas are known exactly. Rather, it is sufficient that the relationships between the pattern areas remain consistent over a period of time required for the calibration process.To calibrate the imaging optics 24, one or more calibration patterns 44 are recorded through the imaging optics 24 with a camera 48 and evaluated with regard to the consistency between the pattern areas.

[0058] In some preferred embodiments, the distance d between the imaging optics 24 (more precisely, between the front glass surface of the imaging optics 24) and the calibration pattern 44 on the display 38 is varied from one image capture to the next, so that the calibration pattern 44 is captured from different distances. Since the pattern areas 46a, 46b remain consistent with each other across the different images of the measurement series, variations between the pattern areas across the individual images of the measurement series are attributable to individual properties of the imaging optics 24 and / or the axes of motion of the coordinate measuring machine 10. Accordingly, in the preferred embodiments, individual properties of the imaging optics 24 are determined from a plurality of images of one or more calibration patterns 44, which are captured within a measurement series at different distances and / or viewing angles.

[0059] In some embodiments, the display 38 is a commercially available tablet PC and the pattern generator 42 is an integrated microprocessor of the tablet PC. In these embodiments, the calibration patterns 44 are supplied to the pattern generator 42 externally via a suitable interface (for example, a USB interface), in particular from the evaluation and control unit 28 of the coordinate measuring machine 10. This evaluation and control unit is located in Fig. 2 is only shown schematically at reference number 28.

[0060] In other embodiments, the pattern generator 42 can be a processor of the display 38, on which a special program for generating calibration patterns is executed. For example, the program can be an app that the manufacturer of the coordinate measuring machine 10 (more generally: the manufacturer of the measuring device) provides for calibrating its device.

[0061] The evaluation of mutually consistent pattern areas of a calibration pattern across a series of measurements enables the quantification of various individual properties of the imaging optics 24, such as distortion, contrast or contrast washout, color drift, etc. However, for calibrating the individual magnification factor, it is desirable to have a precisely known absolute scale. This is only achievable with considerable effort using commercially available displays, since in this case the individual properties of the display 38 are of great importance and these individual properties are generally unknown.

[0062] For this reason, in this embodiment, the new calibration device 32 has, in addition to the display 38, a transparent plate 50, which is attached here to the top surface 52 of the display 38. In some embodiments, the transparent plate can be attached to the top surface 52 of the display 38 by adhesive bonding or by pressure bonding. Preferably, the transparent plate 50 is attached to the top surface of the display 38 only at specific points, so that the display 38 and the transparent plate 50 can move relative to each other. In preferred embodiments, the plate 50 is made of fused silica, quartz, Zerodur® from Schott AG, or another transparent material with low thermal expansion. In contrast, the display 38 can consist of one or more materials that have a significantly higher thermal expansion compared to the plate 50.

[0063] In other embodiments (not shown here), the display 38 can have a mechanical holder or be itself arranged in a mechanical holder, the holder having additional precisely fitting retaining elements for the transparent plate. In all these cases, the transparent plate 50 is arranged above the calibration pattern 44 in a top view, the calibration pattern 44 remaining visible through the plate 50 (not shown here for clarity).

[0064] On the transparent plate 50, at least one line with an exactly known dimension is permanently formed. In the illustrated preferred embodiment, the plate 50 has several lines 54 on its upper surface, which here form, on the one hand, a circular ring with an exactly known diameter and, on the other hand, a square with exactly known longitudinal sides. In preferred embodiments, the lines 54 are formed on the plate 50 using an electron beam writing process and / or lithography processes such as those known from semiconductor manufacturing.

[0065] The lines 54 form a precisely known absolute scale, which is used in the preferred calibration method to quantify the magnification factor of the imaging optics 24. Furthermore, in the preferred embodiments, a light-dark threshold for edge detection can be calibrated by evaluating the light-dark transition generated by the imaging optics 24 in the recorded images at the edges of the lines 54.

[0066] In the embodiment shown here, the lines 54 form an absolute scale without graduations. This is possible because the distortion of the imaging optics 24 can be quantified using the calibration patterns 44 on the display 38. A single absolute length measurement is therefore sufficient for quantifying the magnification factor. Furthermore, the absence of graduations has the advantage that the view of the calibration pattern 44 is only minimally obstructed. However, in contrast to the illustration shown here, the plate 50 can have several absolute length measurements of different lengths in order to optimally calibrate various magnification factors of an imaging optics 24 with a zoom function.

[0067] In further embodiments, the absolute length dimension can be a housing edge of the display or another defined dimension on a separate calibration body. In some embodiments, the absolute length dimension can be determined during calibration by using the previously calibrated positioning axes of the coordinate measuring machine 10 to determine the absolute length dimension. The stylus 26 can advantageously be used for this purpose on the coordinate measuring machine 10.

[0068] As in Fig. As shown schematically in Figure 2, the imaging optics 24 has an optical axis 56 which, in preferred embodiments of the new method, is largely orthogonal to the display 38 and the plate 50. Furthermore, the imaging optics 24 has a field of view which is Fig. 2 is indicated at reference numeral 58. In preferred embodiments, the display 38 is positioned with the plate 50 at a distance d from the imaging optics 24, which is selected such that the imaging optics 24 only captures a portion of the display 38, but not the entire display 38. In the illustration in Fig. 2 corresponds approximately to the lateral extent of the plate 50, as captured by the imaging optics 24. In other embodiments, however, the plate 50, like the display 38, can extend beyond or be smaller than the field of view 58 of the imaging optics 24. In preferred embodiments, the field of view of the imaging optics 24 is positioned approximately in the center of the display 38, and the distance d is selected such that the field of view 58 occupies a maximum of 3 / 4, and preferably a maximum of 2 / 3, of the maximum display area 60, which is defined by the totality of all display pixels 40.

[0069] As in Fig. As indicated in Figure 2, in preferred embodiments of the method, the calibration pattern 44 and the lines 54 are recorded by a camera 48 with an image sensor having a plurality of sensor pixels 62. In the preferred embodiments, the image sensor is the camera which, together with the imaging optics 24 and optionally other elements, forms the optical sensor of the coordinate measuring machine 10. In principle, however, it is conceivable that an imaging optics 24 is coupled with a special calibration camera to perform the calibration.

[0070] In preferred embodiments, the number of display pixels 40 of the display 38 is many times greater than the number of sensor pixels 62 of the image sensor used for calibration. Preferably, the number of display pixels 40 is at least ten times greater than the number of sensor pixels 62, but at least by the assumed magnification factor of the imaging optics 24.

[0071] Alternatively or additionally, a reducing projection optic 64 can be used in further embodiments. In this embodiment, the projection optic 64 is arranged at the image-side end of the imaging optic 24. The projection optic 64 reduces the real image generated by the imaging optic 24 and projects it onto the image sensor. As a result, the display pixels 40 of the display 38 appear smaller in the captured image, and contrast gradients at light-dark edges appear smoother.

[0072] Furthermore, in preferred embodiments, an intermediate element that modifies the emission characteristics of the display pixels 40 can be arranged between the display 38 and the imaging optics 24. In some embodiments, the intermediate element 66 is a so-called privacy filter, such as those offered under the trade name Vicuity filter for notebooks or the like. In other embodiments, the intermediate element 66 can comprise a plurality of microlenses (not shown here), with one microlens arranged over each display pixel 40. The intermediate element 66 can also include a perforated circuit board, with one through-hole (not shown here) of the circuit board arranged over each display pixel 40.

[0073] In all preferred embodiments, the intermediate element 66 focuses the light emitted by the display pixels 40 towards the imaging optics 24 in order to obtain a high image contrast.

[0074] Fig. Figure 3 shows a simplified representation of an image of line 54 on the transparent plate 50, as produced by the imaging optics 24 to be calibrated. The optical axis of the imaging optics 24 is symbolically represented at reference numeral 56. The outer circle 68 represents the field edge, that is, the outer boundary of the field of view 58. As can be seen in Fig. As can be seen in Figure 3, the line 54, which forms the absolute scale for calibrating the magnification factor of the imaging optics 24, is located approximately midway between the optical axis 56 and the field edge 68 in this embodiment. Generally, it is preferred if the at least one line 54 lies in the middle third between the optical axis 56 and the field edge 68. Thus, if the radius from the optical axis 56 to the field edge 68 is divided into three equal sections, as indicated by reference numeral 70, the at least one line 54 is located in the middle section in the preferred embodiments. In this way, the position-dependent magnification or distortion is centered around a mean magnification factor, and the relative variation of the position-dependent magnification is reduced.

[0075] Furthermore, in the preferred embodiments, at least one line 54 is positioned such that its center of gravity or axis of symmetry 72 lies between the display pixels 40, as shown in Fig. 4 is simplified. In other words, the positioning of the at least one line 54 in the preferred embodiments is chosen such that as few as possible, and ideally no display pixels 40, are completely obscured by the at least one line 54.

[0076] The edges 74 of the at least one line 54 are formed on the plate 50 with a light-dark transition that is as abrupt as technically possible. Therefore, in the preferred embodiments, the at least one line 54 enables not only the calibration of the magnification factor of the imaging optics 24, but also the determination of optimal threshold values ​​for edge detection and determination of the edge position.

[0077] Preferred calibration patterns generated with the display 38 have continuous curved and / or wavy lines. These are indeed "step-like" or "stair-like" due to the discrete display pixels 40. However, due to the high number of display pixels 40 compared to the number of sensor pixels 62 and / or due to the use of the reducing projection optics 64, the lines of the preferred calibration patterns appear sufficiently continuous to perform the new calibration.

Claims

[1] Method for calibrating an imaging optic (24) for metrological applications, comprising the steps: - Providing at least one calibration pattern (44) with a plurality of pattern areas (46a, 46b) that are in a defined relation to each other, - Taking at least one image of at least one calibration pattern (44) through the imaging optics. - Evaluating at least one image to quantify individual properties of the imaging optics (24), and - Determining correction values ​​for a computational correction of imaging errors of the imaging optics (24) depending on the quantified individual properties, wherein at least one individual property of the imaging optics (24) is quantified using at least one calibration pattern (44), wherein the at least one calibration pattern (44) is provided on an electronic display (38) with a plurality of matrix-arranged display pixels (40), and wherein the electronic display (38) is able to generate different calibration patterns (44) successively using the matrix-arranged display pixels (40), with one of the different calibration patterns (44) being displayed. characterized by, furthermore, a calibration body (50) is provided on which at least one line (54) with a defined dimension is formed, wherein a magnification factor of the imaging optics (24) is determined on the basis of the at least one line (54), and wherein at least one further individual property of the imaging optics (24) is quantified on the basis of the at least one calibration pattern (44). [2] Method according to claim 1, characterized by , that the calibration body (50) is a transparent plate on which at least one line (54) is permanently formed. [3] Method according to claim 2, characterized by , that the transparent plate is a cover plate of the electronic display (38). [4] Method according to any one of claims 1 to 3, characterized by, that the electronic display (38) has a top surface (52) on which the calibration pattern (44) is displayed, and that at least one line (54) is arranged on the top surface (52) above the calibration pattern (44). [5] Method according to any one of claims 1 to 4, characterized by , that at least one line (54) in a top view of the calibration pattern (44) is arranged offset from the display pixels (40). [6] Method according to any one of claims 1 to 5, characterized by , that the display pixels (40) together define a maximum display area (60), wherein the imaging optics (24) spans a field of view (58), and wherein the electronic display (38) is positioned at a distance (d) to the imaging optics (24) such that the field of view (58) occupies a maximum of 3 / 4 of the maximum display area (60). [7] Method according to any one of claims 1 to 6, characterized by, that at least one image is taken with a camera (48) which has a large number of sensor pixels (62), where the number of display pixels (40) is higher than the number of sensor pixels (62). [8] Method according to any one of claims 1 to 7, characterized by , that the imaging optics (24) has an optical axis and a field edge (68), wherein at least one line (54) is positioned between the optical axis (56) and the field edge (68). [9] Method according to any one of claims 1 to 8, characterized by , that a reducing projection optic (64) is arranged at the image-side end of the imaging optic (24). [10] Method according to any one of claims 1 to 9, characterized by, that the individual display pixels (40) each have a radiation characteristic, and that a plate-shaped intermediate element (66) is arranged between the one calibration pattern (44) and the imaging optics (24), which modifies the radiation characteristic of the individual display pixels (40). [11] Method according to any one of claims 1 to 10, characterized by , that at least two images of the one calibration pattern (44) are taken successively by the imaging optics (24), wherein the one calibration pattern (44) is shifted and / or distorted between the taking of the at least two images on the electronic display (38). [12] Method according to any one of claims 1 to 11, characterized by , that the imaging optics (24) is part of a measuring device (10) with a defined measuring volume and that the electronic display (38) is permanently arranged in the defined measuring volume. [13] Method according to any one of claims 1 to 12, characterized by , that the imaging optics (24) is part of a measuring device (10) for measuring different objects, wherein one calibration pattern (44) is generated on the electronic display (38) depending on an object to be measured. [14] Calibration device for calibrating an imaging optic (24) for metrological applications, comprising an electronic display (38) with a plurality of matrix-arranged display pixels (40), a pattern generator (42) which, with the aid of the matrix-arranged display pixels (40), is able to generate a plurality of different calibration patterns (44), wherein the calibration patterns (44) each have a plurality of pattern areas (46a, 46b) which are in a defined relation to each other, and a calibration unit (28) which evaluates at least one image of at least one of these different calibration patterns (44) taken up by the imaging optic (24) in order to quantify individual properties of the imaging optic (24) and to determine correction values ​​for a computational correction of imaging errors depending on the quantified individual properties, characterized bya calibration body (50) on which at least one line (54) with a defined dimension is formed, wherein the calibration unit (28) is configured to determine a magnification factor of the imaging optics (24) on the basis of the at least one line (54) and to quantify at least one further individual property of the imaging optics (24) on the basis of the at least one calibration pattern (44). [15] Measuring device for measuring various objects, comprising a workpiece holder (12) for positioning an object to be measured, an imaging optic (24) that produces an image of the object, and an evaluation unit (28) for evaluating the image and determining properties of the object, characterized by a calibration device (32) according to claim 14.

Images