Three-dimensional geometric shape measuring device and three-dimensional geometric shape measuring method

Abstract

A three-dimensional geometric shape measuring device, comprising: a preliminary measurement unit 412 that generates a plurality of preliminary measurement data indicating three-dimensional coordinates of reference points on a reference instrument; a reference data generation unit 43 that generates reference data; a calculation unit 44 that calculates a correction value based on the reference data and the preliminary measurement data that do not match the reference data; a target measurement unit 411 that generates target measurement data indicating the result of measuring measurement points of a measurement object; a correction unit 45 that corrects the target measurement data in a measurement system corresponding to the preliminary measurement data that do not match the reference data based on the correction value; and a geometric shape recognition unit 46 that uses the corrected target measurement data to recognize the geometric shape of the measurement object.

Description

Technical Field

[0001] The present disclosure relates to a three-dimensional geometric shape measurement device and a three-dimensional geometric shape measurement method for measuring the three-dimensional geometric shape of a measurement object. Background Art

[0002] The light pattern projection method uses the principle of triangulation and performs three-dimensional (3D) geometric shape measurement by projecting a stripe pattern from a projector onto a measurement object and then photographing the pattern that changes along the geometric shape of the measurement object using an optical device such as a camera. Japanese Unexamined Patent Application No. 2019-516983 describes photographing the pattern projected onto a measurement object by using a plurality of optical devices in the light pattern projection method. Summary of the Invention

[0003] Problems to be Solved by the Invention

[0004] In the method described in Japanese Unexamined Patent Application No. 2019-516983, there is a problem that the measurement results are different between the case of measuring the geometric shape of a measurement object using one of a plurality of optical devices and the case of measuring the geometric shape of the measurement object using another optical device.

[0005] Focusing on this point, an object of the present disclosure is to provide a three-dimensional (3D) geometric shape measurement device and a 3D geometric shape measurement method that can prevent inconsistency of measurement results when measuring the geometric shape of a measurement object using different optical devices.

[0006] Means for Solving the Problems

[0007] A 3D geometric shape measurement device according to a first aspect for measuring the three-dimensional geometric shape of a measurement object based on a captured image obtained by capturing the measurement object, the three-dimensional geometric shape measurement device including: a plurality of optical devices; a preliminary measurement unit that generates a plurality of preliminary measurement data indicating three-dimensional coordinates of reference points on a reference instrument by capturing the reference instrument in a plurality of measurement systems corresponding to different combinations of the plurality of optical devices; a reference data generation unit that generates reference data based on one or more of the plurality of preliminary measurement data; a calculation unit that calculates a correction value based on the reference data and the preliminary measurement data that do not match the reference data among the plurality of preliminary measurement data; a target measurement unit that generates a plurality of target measurement data indicating results of measuring measurement points of the measurement object using the plurality of measurement systems; a correction unit that corrects the target measurement data in the measurement system corresponding to the preliminary measurement data that does not match the reference data based on the correction value; and a geometric shape recognition unit that recognizes the geometric shape of the measurement object using the corrected target measurement data corrected by the correction unit.

[0008] The preliminary measurement unit can generate preliminary measurement data by photographing reference instruments installed at a plurality of installation positions, the preliminary measurement data indicating three-dimensional coordinates of reference points of the reference instruments at the plurality of installation positions, and the calculation unit can calculate correction values corresponding to the reference points at the plurality of installation positions. The reference data generation unit can generate reference data by calculating statistics of the three-dimensional coordinates of the reference points, the three-dimensional coordinates of the reference points being indicated by a plurality of preliminary measurement data corresponding to a plurality of measurement systems.

[0009] The three-dimensional geometric shape measurement device may further include an acquisition unit that acquires relative position information indicating a positional relationship of a plurality of reference points included in the reference instrument, wherein the preliminary measurement unit can generate preliminary measurement data indicating three-dimensional coordinates of the plurality of reference points, and the reference data generation unit can generate reference data by: selecting the preliminary measurement data indicating the three-dimensional coordinates of the plurality of reference points based on an error in the positional relationship indicated by the relative position information.

[0010] The preliminary measurement unit can generate preliminary measurement data indicating three-dimensional coordinates of a plurality of reference points on the reference instrument, and the reference data generation unit can generate one reference data for each reference point by selecting one of the plurality of preliminary measurement data corresponding to a plurality of measurement systems.

[0011] The reference data generation unit can select, from the plurality of preliminary measurement data corresponding to a plurality of measurement systems, the preliminary measurement data corresponding to a combination of optical devices, and using these combinations of optical devices, the triangulation accuracy of the reference points becomes optimal. The preliminary measurement unit can project a projection image including a pattern for identifying a reference point onto the reference instrument. The calculation unit can identify a function for calculating a correction value corresponding to the three-dimensional coordinates of a measurement point measured by the measurement system, and use the identified function to calculate the correction value, the measurement system corresponding to the preliminary measurement data that does not match the reference data. The calculation unit may further include a storage unit that stores a table in which (i) the three-dimensional coordinates of each grid point of the measurement space of the plurality of measurement systems (the measurement space being divided into cubic grids at equal intervals), (ii) a plurality of correction values, and (iii) an index for identifying each measurement system are associated with each other, wherein the correction unit can read from the storage unit the correction value associated with the three-dimensional coordinates of the grid point closest to the three-dimensional coordinates indicated by the target measurement data and the index indicating the measurement system corresponding to the grid point, and the correction unit can correct the preliminary measurement data based on the read correction value.

[0012] A three-dimensional geometric shape measurement method according to a second aspect of the embodiment is used to measure the three-dimensional geometric shape of a measurement object based on a captured image obtained by photographing the measurement object. The three-dimensional geometric shape measurement method includes the following steps: generating a plurality of preliminary measurement data by photographing a reference instrument with a plurality of measurement systems corresponding to combinations of different optical devices among a plurality of optical devices, where the preliminary measurement data indicates the three-dimensional coordinates of reference points on the reference instrument; generating reference data based on one or more of the plurality of preliminary measurement data; calculating a correction value based on the reference data and the preliminary measurement data that do not match the reference data among the plurality of preliminary measurement data; generating a plurality of target measurement data, where the target measurement data indicates the results of measuring measurement points of the measurement object using the plurality of measurement systems; correcting the target measurement data in the measurement system corresponding to the preliminary measurement data that does not match the reference data based on the correction value; and using the corrected target measurement data to identify the geometric shape of the measurement object.

[0013] Advantages of the present invention

[0014] According to this embodiment, the effect of preventing inconsistent measurement results when using different optical devices to measure the geometric shape of a measurement object is achieved. Description of the drawings

[0015] Figure 1A and Figure 1B Illustrates an overview of a 3D geometric shape measurement device according to an embodiment.

[0016] Figure 2 Shows the configuration of the 3D geometric shape measurement device.

[0017] Figure 3 A to Figure 3 F respectively show examples of types of projection images including binary fringe patterns projected onto a measurement object.

[0018] Figure 4 Shows the relationship with Figure 3 C to Figure 3 Examples of Gray codes corresponding to the binary fringe patterns shown in F.

[0019] Figure 5 A to Figure 5 D respectively show examples of gradient fringe patterns having a sinusoidal luminance distribution.

[0020] Figure 6 Shows the relationship between absolute projection coordinates and relative projection coordinates.

[0021] Figure 7 Shows an example of the epipolar line of a second captured image corresponding to a first captured pixel.

[0022] Figure 8A and Figure 8B each shows an example of a reference instrument.

[0023] Figure 9 Shows how to measure a reference instrument.

[0024] Figure 10 is a flowchart showing the process of calculating correction values by a 3D geometric shape measurement device.

[0025] Figure 11 is a flowchart showing the process of identifying the 3D geometry of a measurement object performed by a 3D geometric shape measurement device.

[0026] Figure 12A and Figure 12B each shows how to select a measurement system in the process of generating reference data by a reference data generation unit. Detailed Description

[0027]

Outline of 3D (Three-Dimensional) Geometric Shape Measurement Device 100

[0028] Figure 1A and Figure 1B illustrate the outline of 3D geometric shape measurement device 100 according to an embodiment. Figure 1A Shows the configuration of 3D geometric shape measurement device 100. 3D geometric shape measurement device 100 includes a first imaging unit 1, a second imaging unit 2, and a projection unit 3 as optical devices. 3D geometric shape measurement device 100 includes a control unit 4 that controls various operations of the optical devices.

[0029] The projection unit 3 is a projection device having a light source such as a light-emitting diode or a laser. The projection unit 3 projects a projection image including a pattern for identifying projection coordinates onto the measurement surface of the measurement object. The projection coordinates indicate the positions of the projection pixels that make up the projection image projected by the projection unit 3. The projection coordinates can be one-dimensional coordinates indicating the vertical or horizontal position of the projection image, or two-dimensional coordinates indicating the vertical and horizontal positions of the projection image. The pattern is, for example, a stripe pattern. The number of projection units 3 is not limited to one, and 3D geometric shape measurement device 100 may include any number of projection units.

[0030] The first imaging unit 1 includes a lens 11 and an imaging element 12. The first imaging unit 1 generates a first captured image by capturing the projection image projected onto the measurement object in response to the projection unit 3 projecting the projection image onto the measurement object. The first imaging unit 1 is placed in such a way that the optical axis of the first imaging unit 1 forms a predetermined angle with the optical axis of the projection unit 3.

[0031] The second imaging unit 2 includes a lens 21 and an imaging element 22. The second imaging unit 2 generates a second captured image by capturing the projected image projected onto the measurement object in response to the projection unit 3 projecting the projected image onto the measurement object. The second imaging unit 2 is placed in such a manner that the optical axis of the second imaging unit 2 forms a predetermined angle with the optical axis of the projection unit 3. The optical axis of the second imaging unit 2 may be in the same plane as the optical axes of the first imaging unit 1 and the projection unit 3, but is not limited thereto. The control unit 4 is implemented by a computer, for example. The control unit 4 measures the 3D geometry of the measurement object based on a plurality of captured images generated by the first imaging unit 1 and the second imaging unit 2. Further, the number of imaging units is not limited to two, and the 3D geometry measurement device 100 may be provided with any number of imaging units.

[0032] Figure 1B Illustrates the error in the geometric shape measurement of the measurement object by a plurality of measurement systems of the 3D geometry measurement device 100. The 3D geometry measurement device 100 measures the 3D geometry using a plurality of measurement systems (a first measurement system and a second measurement system) corresponding to combinations of different optical devices.

[0033] When using the first measurement system, the 3D geometry measurement device 100 measures the 3D coordinates of the measurement point P by using the projection unit 3 and the first imaging unit 1. As Figure 1B shown, the 3D geometry measurement device 100 identifies which projection pixel in the projected image projected by the projection unit 3 has its light irradiating the measurement point P. In Figure 1B the example shown, the measurement point P exists on the optical path L3 through which the light of the identified projection pixel passes.

[0034] The 3D geometry measurement device 100 identifies which first captured pixel included in the first captured image captured by the first imaging unit 1 reflects the measurement point P in the measurement of the 3D coordinates of the measurement point P. The first captured pixel is the smallest unit included in the first captured image. In Figure 1B the example shown, the measurement point P exists on the optical path L1 corresponding to the identified first captured pixel. The 3D geometry measurement device 100 measures the 3D coordinates of the measurement point P by using triangulation to obtain the position of the intersection of the optical path L3 and the optical path L1.

[0035] On the other hand, when using the second measurement system, the 3D geometry measurement device 100 measures the 3D coordinates of the measurement point P by using the projection unit 3 and the second imaging unit 2. The 3D geometry measurement device 100 identifies which second captured pixel included in the second captured image captured by the second imaging unit 2 reflects the measurement point P in the measurement of the 3D coordinates of the measurement point P. The second captured pixel is the smallest unit included in the second captured image. In Figure 1BIn the example shown, the measurement point P exists on the optical path L2 corresponding to the identified second captured pixel. As described above, since the measurement point P exists on the optical path L3, the 3D geometry measurement device 100 measures the 3D coordinates of the measurement point P by obtaining the position of the intersection of the optical path L2 and the optical path L3.

[0036] When there is no error in the first measurement system and the second measurement system, since (i) the intersection between the optical path L1 and the optical path L2 and (ii) the intersection between the optical path L1 and the optical path L3 are approximately coincident with the measurement point P, therefore, as Figure 1B shown, regardless of which measurement system is used to measure the measurement point P, the measurement result of the 3D coordinates will not change. Therefore, the 3D geometry measurement device 100 can improve the measurement accuracy by using, for example, the average value of the measurement results of multiple measurement systems.

[0037] However, when there is an error in the optical path through which the light of the projection pixels of the projection unit 3 passes due to distortion of a lens (not shown) or the like, the 3D geometry measurement device 100 identifies that the measurement point P exists on the optical path L3' indicated by the dashed line in Figure 1B That is, when the measurement point P is measured by the first measurement system including the projection unit 3 and the first imaging unit 1, the 3D geometry measurement device 100 erroneously measures the 3D coordinates of the intersection P1 of the optical path L1 and the optical path L3' as the 3D coordinates of the measurement point P.

[0038] Furthermore, when the measurement point P is measured by the second measurement system including the projection unit 3 and the second imaging unit 2, the 3D geometry measurement device 100 takes the 3D coordinates of the intersection P2 of the measurement optical path L2 and the optical path L3' as the 3D coordinates of the measurement point P. When the difference between the measurement point P1 and the measurement point P2 is large, the error of the measurement result of the 3D geometry measurement device 100 becomes large.

[0039] The 3D geometry measurement device 100 corrects the errors of each measurement system in order to prevent such a reduction in the measurement accuracy of the 3D coordinates. More specifically, the 3D geometry measurement device 100 generates reference data indicating the standard of the measurement data in multiple measurement systems by pre-measuring a predetermined reference instrument.

[0040] Using this reference data, the 3D geometry measurement device 100 calculates correction values for correcting the measurement data of each of the multiple measurement systems, and corrects the measurement data in the measurement system corresponding to the calculated correction value by using the calculated correction values. In this way, the 3D geometry measurement device 100 improves the measurement accuracy of the 3D coordinates of the measurement point P and prevents the measurement results from becoming different for different measurement systems.

[0041]

Configuration of 3D Geometry Measurement Device

[0042] Figure 2 shows the internal configuration of the 3D geometric shape measurement device 100. In addition to including the first imaging unit 1, the second imaging unit 2, the projection unit 3, and the control unit 4 shown in FIG. 1, the 3D geometric shape measurement device 100 further includes a storage unit 5.

[0043] The storage unit 5 includes a storage medium, including a read-only memory (ROM), a random access memory (RAM), a hard disk, etc. The storage unit 5 stores programs to be executed by the control unit 4. The control unit 4 is, for example, a central processing unit (CPU). By executing the programs stored in the storage unit 5, the control unit 4 functions as a measurement unit 41, an acquisition unit 42, a reference data generation unit 43, a calculation unit 44, a correction unit 45, and a geometric shape recognition unit 46.

[0044] The measurement unit 41 uses the projection unit 3 to project a projection image including a predetermined light pattern onto a measurement object or the like. The measurement unit 41 generates captured images captured by the first imaging unit 1 and the second imaging unit 2 by capturing the projection image projected onto the measurement object or the like. The measurement unit 41 generates a captured image by capturing a predetermined reference instrument. The measurement unit 41 includes a target measurement unit 411 and a preliminary measurement unit 412.

[0045] The target measurement unit 411 generates a plurality of target measurement data indicating measurement results by measuring measurement points of the measurement object using a plurality of measurement systems. The plurality of measurement systems are, for example, the following first measurement system to fourth measurement system.

[0046] First measurement system: the first imaging unit 1 and the projection unit 3

[0047] Second measurement system: the second imaging unit 2 and the projection unit 3

[0048] Third measurement system: the first imaging unit 1, the second imaging unit 2, and the projection unit 3

[0049] Fourth measurement system: the first imaging unit 1 and the second imaging unit 2

[0050]

Measurement according to the first measurement system

[0051] Using the first measurement system, the target measurement unit 411 measures the measurement points of the measurement object by combining one imaging unit and one projection unit. The target measurement unit 411 projects a projection image including a predetermined light pattern onto the measurement object through the projection unit 3. Figure 3 A to Figure 3 F respectively show examples of the types of projection images including binary stripe patterns projected by the target measurement component 411. Figure 3 A to Figure 3 The black areas in F represent non-projection areas where the projection unit 3 does not project light, while the white areas represent light projection areas where the projection unit 3 projects light.

[0052] Figure 3 A shows the standard pattern (i.e., all - black pattern) adopted when light is not projected onto any part of the measurement object. Figure 3 B shows the standard pattern (i.e., all - white pattern) adopted when light is projected onto the entire measurement object. Figure 3 C to Figure 3 F show binary fringe patterns, which are composed of a light - projected area and a non - projected area, and in these binary fringe patterns, fringes with different widths for each projected image are arranged in the same direction.

[0053] Figure 3 C to Figure 3 The fringe patterns shown in C to F correspond to Gray codes and are used to identify the projection coordinates that indicate the positions of the projected pixels of the projection image corresponding to the first captured pixel in the captured image. Figure 4 shows the one corresponding to Figure 3 C to Figure 3 F. An example of the Gray code corresponding to the binary fringe pattern is shown. By associating 0 in the Gray code with the non - projected area and 1 with the light - projected area, the binary fringe patterns shown in C to Figure 3 C to Figure 3 F are generated.

[0054] Figure 3 A to Figure 3 F and Figure 4 each position in the x - direction is represented by a code value, which is a combination of the digits 0 or 1 at each position in the Gray code. Figure 4 The position 0 corresponds to the code value "0000", the position 1 corresponds to the code value "0001", and the position 15 corresponds to the code value "1000".

[0055] The target measurement unit 411 projects a projection image including a gradient fringe pattern with a sine - shaped luminance distribution onto the measurement object. Figure 5 A to Figure 5 D respectively show examples of gradient fringe patterns with a sine - shaped luminance distribution. Although Figure 3 C to Figure 3 F the binary fringe patterns shown are binary images composed of a black area and a white area, in Figure 5 A to Figure 5 D the gradient fringe patterns shown, the shading changes sinusoidally from the white area to the black area along the width direction of the fringes. Figure 5 A to Figure 5 D the interval between the fringes in the gradient fringe pattern is constant, and the spatial frequency of the fringes in the gradient fringe pattern is, for example, Figure 3 four times the spatial frequency of the binary fringe pattern in F.

[0056] Figure 5 A through Figure 5 The gradient stripe patterns from A to D are different from each other in that the phases of the sine waves indicating the luminance distributions differ from each other by 90 degrees, and their luminance distributions are the same in other respects. The target measurement unit 411 projects a total of ten projection images: Figure 3 A and Figure 3 the two standard patterns shown in B, Figure 3 C through Figure 3 the four binary stripe patterns shown in F, and Figure 5 A through Figure 5 the four gradient stripe patterns shown in D. Figure 5 The gradient stripe patterns from A to D and Figure 5 the stripe patterns shown in A through Figure 3 F are used together to identify the projection coordinates. Figure 3 The target measurement unit 411 generates a first captured image using a first measurement system, where the projection image projected onto the measurement object is captured by the first imaging unit 1. The target measurement unit 411 identifies the projection coordinates corresponding to the first captured pixels of the first captured image based on the patterns included in the first captured image. For example, the target measurement unit 411 analyzes the change in the shadow in the patterns included in the first captured image, thereby identifying the projection coordinates indicating the position of the projection pixel corresponding to the first captured pixel included in the first captured image.

[0057] The target measurement unit 411 calculates, for each pixel, the average value of the luminance value when the all - black pattern shown in

[0058] A is projected and the luminance value when the all - white pattern shown in Figure 3 B is projected as the median value. Similarly, with respect to the first captured image captured when the binary stripe patterns from Figure 3 C to Figure 3 F are projected onto the measurement object, the target measurement unit 411 identifies the code value of each first captured pixel by comparing the luminance value of each first captured pixel in the four first captured images with the corresponding median value. By identifying the code value, the target measurement unit 411 can identify which binary stripe within the binary stripes projected toward different positions is reflected at the pixel position of each first captured pixel. The target measurement unit 411 identifies which of the positions from position 1 to position 15 shown in Figure 3 F that each first captured pixel included in the first captured image is included in. Figure 4 is shown.

[0059] Further, the target measurement unit 411 identifies the phase of the sine wave corresponding to the first captured pixel in the first captured image when a gradient stripe pattern having a sine luminance distribution is projected onto the measurement object, and identifies the projection coordinates based on the identified phase. Since the gradient stripe pattern of the projection image has periodicity, multiple projection pixels in the projection image have the same projection coordinates. Hereinafter, the periodic projection coordinates in the projection image are also referred to as relative projection coordinates. In addition, the projection coordinates uniquely determined in the projection image are also referred to as absolute projection coordinates.

[0060] Figure 6 Shows the relationship between the absolute projection coordinates and the relative projection coordinates. Figure 6 The vertical axis of represents the projection coordinates. Figure 6 The horizontal axis of indicates the position of the projection pixels in the width direction of the stripes included in the projection image. The width direction is the direction orthogonal to the direction in which the stripes extend. As Figure 6 shown by the solid line in, the relative projection coordinates have periodicity. The relative projection coordinates show the same value for each period of repetition of the gradient stripe pattern having a sine luminance distribution. On the other hand, as Figure 6 shown by the diagonally extending dashed line in, the absolute projection coordinates are uniquely determined in the projection image.

[0061] The target measurement unit 411 identifies the relative projection coordinates corresponding to the first captured pixel by analyzing the shadow of the gradient stripe pattern. The target measurement unit 411 identifies which position among positions 0 to 15 the first captured pixel corresponds to based on the Gray code indicated by the binary stripe pattern. The target measurement unit 411 identifies the absolute projection coordinates corresponding to the first captured pixel based on the relative position indicated by the relative projection coordinates in the position identified by the Gray code. The target measurement unit 411 uses (i) the two-dimensional coordinates of the first captured pixel and (ii) the identified one-dimensional absolute projection coordinates to identify the 3D coordinates of the measurement point on the measurement object corresponding to the first captured pixel by using the triangulation principle.

[0062] The target measurement unit 411 may project multiple projection images with different directions of the stripes (i.e., the direction in which the stripes extend), and identify the projection coordinates corresponding to the first captured pixel for each projection image with a different stripe direction. The target measurement unit 411 may project multiple projection images with different stripe directions onto the measurement object, and identify the projection coordinates corresponding to the first captured pixel for each projection image with a different stripe direction.

[0063]

Measurement according to the second measurement system

[0064] In the second measurement system, the target measurement unit 411 measures the 3D coordinates of the measurement points of the measurement object in a manner similar to the first measurement system by combining the second imaging unit 2 and the projection unit 3.

[0065]

Measurement according to the third measurement system

[0066] In the third measurement system, the target measurement unit 411 measures the measurement points of the measurement object by combining the first imaging unit 1, the second imaging unit 2, and the projection unit 3. The target measurement unit 411 identifies the projection coordinates in a manner similar to the first measurement system, and the projection coordinates indicate the position of the projection pixel corresponding to the first captured pixel included in the first captured image. The target measurement unit 411 identifies the projection coordinates indicating the position of the projection pixel corresponding to the second captured pixel included in the second captured image.

[0067] Figure 7 An example of the epipolar line E of the second captured image corresponding to the first captured pixel A is shown. BA Assume that the focal point of the lens 11 of the first imaging unit 1 is O1. The optical path corresponding to the first captured pixel A is a straight line extending from the focal point O1 to the measurement point MP, and the straight line indicating this optical path and projected on Figure 7 the image plane of the second imaging unit 2 on the left side is the epipolar line E. BA Due to the nature of geometric constraints, the second captured pixel B (which corresponds to the measurement point MP in the same way as the first captured pixel A) in the image plane of the second imaging unit 2 on the left side is at any position on the epipolar line E. Figure 7 The target measurement unit 411 reads the arrangement information indicating the arrangements of the first imaging unit 1 and the second imaging unit 2 pre-stored in the storage unit 5, and identifies the epipolar line E of the second captured image corresponding to the first captured pixel A based on the read arrangement information. BA The arrangement information is information indicating, for example, the focal length, position, and orientation. BA As

[0068] shown, the target measurement unit 411 selects the second captured pixel B from the second captured pixels located on the identified epipolar line E. Figure 7 The second captured pixel B corresponds to the absolute projection coordinates in the same way as the first captured pixel A. The target measurement unit 411 measures the 3D coordinates of the common measurement point MP corresponding to the first captured pixel A and the selected second captured pixel B by using the triangulation principle for the first captured pixel A and the selected second captured pixel B. BA

[0069]

Measurement according to the fourth measurement system

[0070] In the fourth measurement system, the target measurement unit 411 identifies the correspondence between the first captured pixels of the first captured image and the second captured pixels of the second captured image. When the measurement object has feature points such as texture and edges, the target measurement unit 411 measures the 3D coordinates of the feature points without projecting a projection image onto the measurement object.

[0071] The target measurement unit 411 identifies the first captured pixels corresponding to the feature points of the measurement object. Further, the target measurement unit 411 identifies the second captured pixels corresponding to the same feature points. The target measurement unit 411 uses (i) the first captured pixels corresponding to the feature points and (ii) the second captured pixels corresponding to the same feature points to measure the 3D coordinates of the feature points by using the triangulation principle.

[0072]

Stripe patterns in two directions

[0073] It should be noted that the target measurement unit 411 is not limited to the example of generating target measurement data by using only a projection image including a stripe pattern extending in one direction. For example, the target measurement unit 411 can generate target measurement data by projecting a projection image including a stripe pattern extending in the horizontal direction onto the measurement object in addition to projecting a projection image including a stripe pattern extending in the vertical direction onto the measurement object. The target measurement unit 411 projects a projection image including a stripe pattern extending in the horizontal direction onto the measurement object to identify both the vertical projection coordinates and the horizontal projection coordinates corresponding to the captured pixels. In this case, the target measurement unit 411 uniquely identifies the projection pixels corresponding to the captured pixels without using epipolar lines.

[0074]

Measurement of the reference instrument

[0075] The preliminary measurement unit 412 generates a plurality of preliminary measurement data indicating the 3D coordinates of the reference points on the reference instrument by photographing a predetermined reference instrument in a plurality of measurement systems corresponding to combinations of different optical devices among the plurality of optical devices. A plurality of reference points are arranged on the reference instrument. For example, the preliminary measurement unit 412 uses the same method as the target measurement unit 411 to measure the 3D coordinates of the reference points by using each of the first to fourth measurement systems.

[0076] Figure 8A and Figure 8B Each shows an example of the reference instrument. Figure 8A and Figure 8B Each shows the reference instrument as viewed from above. Figure 8A Shows a reference instrument with a plurality of black circles arranged in an orderly state longitudinally and transversely. Figure 8B Shows a grid reference instrument. When using Figure 8AWhen using the reference instrument shown, the preliminary measurement unit 412 uses the center of the black circle as the reference point and generates preliminary measurement data indicating the 3D coordinates of the reference point. When using Figure 8B the reference instrument shown, the preliminary measurement unit 412 determines each vertex of the white and black squares forming the checkerboard pattern as the reference point and generates preliminary measurement data indicating the 3D coordinates of the determined reference points. The preliminary measurement data contains information indicating the 3D coordinates of a plurality of reference points on the reference instrument.

[0077] Figure 9 FIG. shows how to measure the reference instrument 51. In Figure 9 the example, the reference instrument 51 is a flat member, but is not limited thereto, and any shape can be used as the reference instrument. The preliminary measurement unit 412 photographs the reference instrument 51 more than once. In the measurement, the reference instrument 51 is successively installed at a plurality of installation positions, and the coordinates of these installation positions are different in the depth direction (as shown by the arrow in Figure 9 ). The installation positions of the reference instrument 51 do not need to be precisely determined and can be schematic. Moreover, the 3D coordinates of the installation positions of the reference instrument 51 do not need to be measured in advance, and the reference instrument 51 can be installed at any installation position. Therefore, since there is no need to prepare an instrument for positioning the reference instrument 51, another measuring instrument, etc., the measurement of the reference instrument 51 can be easily performed at low cost in the 3D geometry measuring device 100.

[0078] The preliminary measurement unit 412 photographs the reference instrument 51 in a state where the reference instrument 51 is placed at each installation position. The preliminary measurement unit 412 generates preliminary measurement data indicating the 3D coordinates of the reference points of the reference instrument 51 at a plurality of installation positions based on the photographed images of the reference instrument 51 installed at the plurality of installation positions. Moreover, the example of setting a plurality of installation positions by parallelly moving the reference instrument 51 so that the coordinates in the depth direction are different from each other is not limiting, and a plurality of installation positions can also be set by changing the posture of the reference instrument 51 so that the 3D coordinates are different from each other.

[0079] Moreover, the preliminary measurement unit 412 is not limited to the example of pre-measuring the 3D coordinates of the reference points arranged on the reference instrument 51. For example, the preliminary measurement unit 412 can project a projection image including a marker at a specific position to be used as a reference point onto the reference instrument 51 with the projection unit 3. The preliminary measurement unit 412 can determine the position where the marker is projected as the reference point and measure the 3D coordinates of the reference point.

[0080] Moreover, the preliminary measurement unit 412 can project a projection image including a pattern for identifying a reference point onto the reference instrument 51. The projection image including the pattern for identifying the reference point is, for example, a projection image including a stripe pattern extending in one direction. The preliminary measurement unit 412 can acquire a first captured image and a second captured image obtained by photographing the projection image projected onto the reference instrument 51. In this case, the preliminary measurement unit 412 identifies the projection coordinates corresponding to the first captured pixel and the projection coordinates corresponding to the second captured pixel in the same manner as the target measurement unit 411, and uses the correspondence relationship among the first captured pixel, the second captured pixel, and the projection image to measure the 3D coordinates of a specific position on the reference instrument 51. The preliminary measurement unit 412 can use this specific position as a reference point. In other words, the preliminary measurement unit 412 can also project a stripe pattern onto the reference instrument 51 and use a certain position commonly measured by multiple measurement systems as a reference point.

[0081] Furthermore, the preliminary measurement unit 412 can project the projection coordinates corresponding to the first captured image and the projection coordinates corresponding to the second captured pixel by projecting a projection image including a stripe pattern extending in the vertical direction and a projection image including a stripe pattern extending in the horizontal direction onto the measurement object. In this case, the preliminary measurement unit 412 can obtain the correspondence relationship between the first captured pixel and the second captured pixel by using the projection image including the stripe pattern extending in the vertical direction and the projection image including the stripe pattern extending in the horizontal direction without using epipolar lines. Therefore, the preliminary measurement unit 412 can reduce the risk of reducing the measurement accuracy due to errors that may occur when identifying epipolar lines.

[0082] The acquisition unit 42 acquires relative position information indicating the positional relationship between a plurality of reference points included in the reference instrument 51 from the storage unit 5. It is assumed that the positional relationship between the plurality of reference points has been measured in advance using another measuring instrument. The positional relationship of the plurality of reference points is, for example, the two-dimensional coordinates of the reference points other than the origin when one reference point of the reference instrument 51 is assumed as the origin.

[0083]

Generation of Reference Data

[0084] The reference data generation unit 43 generates reference data based on one or more of a plurality of preliminary measurement data corresponding to a plurality of measurement systems. The reference data is data for matching the respective measurement data of the plurality of measurement systems with the measurement data of other measurement systems.

[0085] For example, assume that the distance from the projection unit 3 to the position of a specific reference point in the first measurement system is shown as 199 mm, the distance from the projection unit 3 to the position of the same reference point in the second measurement system is 201 mm, and the distance from the projection unit 3 to the position of the same reference point in the third measurement system is 200 mm. In this case, for example, the reference data generation unit 43 matches the measurement data of each measurement system by determining 199 mm measured using the first measurement system as the reference data, performing a correction of subtracting 2 mm from the data measured using the second measurement system, and performing a correction of subtracting 1 mm from the data measured using the third measurement system. The method of correcting the distance has been described as an example, but in practice, it is desirable to correct the 3D coordinates.

[0086] For example, the reference data generation unit 43 calculates statistics such as the average value and median of the 3D coordinates of the same reference point included in a plurality of preliminary measurement data corresponding to different measurement systems. The reference data generation unit 43 may generate reference data in which the calculated statistic is the 3D coordinate of the reference point. For example, the reference data generation unit 43 calculates the average value of the 3D coordinates of the same reference point measured using the first measurement system, the second measurement system, and the third measurement system. The reference data generation unit 43 generates reference data in which the calculated average value is the 3D coordinate of the reference point.

[0087] Based on the relative position information acquired by the reference acquisition unit 42, the reference data generation unit 43 can obtain the error between the positional relationship indicated by the relative position information and the positional relationships of the plurality of reference points indicated by the preliminary measurement data, and select any one of the preliminary measurement data based on the obtained error.

[0088] For example, when one reference point of the reference instrument 51 is the origin, the reference data generation unit 43 obtains (i) the relative coordinates of the reference points other than the origin in the relative position information, and (ii) the relative coordinates of the reference points other than the origin in the preliminary measurement data. For each reference point, the reference data generation unit 43 obtains the error between the relative coordinates of the reference points other than the origin in the relative position information and the corresponding reference point in the preliminary measurement data. The reference data generation unit 43 obtains statistics of the errors for each measurement system, such as the sum or standard deviation of the errors at the plurality of reference points of the reference instrument 51, and selects the preliminary measurement data of the measurement system in which the statistic such as the average value of the obtained errors is the smallest among the plurality of measurement systems. The relative distance from the origin can be used instead of the relative coordinates from the origin.

[0089] In addition, when arranging multiple reference points on a single plane in the reference instrument 51, the reference data generation unit 43 identifies a virtual plane using the least squares method or the like such that the sum of the distances from the virtual plane to the multiple reference points on the reference instrument 51 indicated by the preliminary measurement data is minimized. The reference data generation unit 43 identifies such a plane for multiple preliminary measurement data corresponding to different measurement systems. The reference data generation unit 43 calculates, for each measurement system, the sum of the distances between (i) the identified plane and (ii) the multiple reference points on the reference instrument 51 in the preliminary measurement data corresponding to the identified plane. The reference data generation unit 43 may select the preliminary measurement data of the measurement system in which the calculated sum of the distances is the smallest.

[0090] With this configuration, the reference data generation unit 43 can use the preliminary measurement data of the measurement system that most accurately reproduces the plane where the multiple reference points arranged on the reference instrument 51 are located. The reference data generation unit 43 generates reference data based on the selected preliminary measurement data. For example, the reference data generation unit 43 uses the selected preliminary measurement data as the reference data.

[0091] When the deviation between multiple preliminary measurement data corresponding to multiple measurement systems is equal to or greater than a predetermined value, the reference data generation unit 43 does not need to generate reference data and may display an error indicating a large deviation between the preliminary measurement data on a display unit (not shown). In this way, the reference data generation unit 43 can prevent a decrease in the measurement accuracy of the 3D geometry due to defects such as deviations in the orientation of optical devices.

[0092]

Calculation of correction value

[0093] The calculation unit 44 calculates a correction value based on (i) the preliminary measurement data that does not match the reference data and (ii) the reference data generated by the reference data generation unit 43. When the reference data generation unit 43 uses the preliminary measurement data corresponding to a single measurement system as the reference data, the preliminary measurement data that does not match the reference data is, for example, the preliminary measurement data measured by a measurement system different from the measurement system corresponding to the reference data.

[0094] The calculation unit 44 assumes that the reference data generation unit 43 uses the preliminary measurement data measured by the third measurement system as the reference data. The calculation unit 44 identifies the 3D coordinates indicating a predetermined reference point in the first measurement system different from the third measurement system, and identifies the 3D coordinates indicating the same reference point in the reference data. The calculation unit 44 obtains the difference ΔC(x 1,i,j ,y 1,i,j ,z 1,i,j ) between the two 3D coordinates using Equation 1 below.

[0095] ΔC(x 1,i,j ,y1,i,j , z 1,i,j ) = [x 1,i,j , y 1,i,j , z 1,i,j - [x r,i,j , y r,i,j , z r,i,j ··· Equation 1

[0096] In Equation 1, the "1" in "x 1,i,j " is an index indicating the first measurement system, "i" is an index for identifying a reference point (i = 1, 2, etc.), and "j" is an index for identifying the installation position of the reference instrument 51 (j = 1, 2, etc.). [x 1,i,j , y 1,i,j , z 1,i,j are the 3D coordinates indicated by the preliminary measurement data measured by the first measurement system. [x r,i,j , y r,i,j , z r,i,j are the 3D coordinates indicated by the reference data.

[0097] The calculation unit 44 calculates a correction value for correcting the target measurement data measured by the first measurement system based on the obtained difference ΔC(x 1,i,j , y 1,i,j , z 1,i,j ). As an example, the calculation unit 44 uses the difference ΔC(x 1,i,j , y 1,i,j , z 1,i,j ) as the correction value, but the correction value can be calculated using a known method based on the difference ΔC(x 1,i,j , y 1,i,j , z 1,i,j ).

[0098] When the reference data generation unit 43 generates reference data by averaging the 3D coordinates of the reference points indicated by a plurality of preliminary measurement data corresponding to different measurement systems, the preliminary measurement data used to generate the reference data may not match the reference data. Therefore, the calculation unit 44, when generating reference data by averaging the 3D coordinates of the reference points indicated by a plurality of preliminary measurement data, can calculate the correction value based on the difference between (i) the 3D coordinates of the reference points indicated by the preliminary measurement data used to generate the reference data and (ii) the 3D coordinates of the same reference points indicated by the reference data.

[0099] The calculation unit 44 calculates correction values corresponding to reference points of the reference instrument 51 installed at a plurality of installation positions. For example, when the preliminary measurement unit 412 generates three preliminary measurement data corresponding to three installation positions with different coordinates in the depth direction, the calculation unit 44 generates correction values corresponding to each of the three preliminary measurement data. The calculation unit 44 associates (i) an index of the installation position of the reference instrument 51, (ii) an index for identifying the measurement system used to perform the measurement of the reference instrument 51, (iii) an index for identifying the reference point, and (iv) the correction value with each other, and stores them in the storage unit 5.

[0100] Further, the calculation unit 44 can identify a function for calculating a correction value corresponding to the 3D coordinates of a measurement point measured by a measurement system corresponding to preliminary measurement data that does not match the reference data. For example, the calculation unit 44 can identify a polynomial function for converting target measurement data measured by the first measurement system into a correction value corresponding to the target measurement data by referring to the relationship between the preliminary measurement data measured by the first measurement system and the corresponding correction value, using a known technique such as the least squares method. In this case, the function is stored in the storage unit 5 in advance as part of a program or the like, and the calculation unit 44 identifies a plurality of coefficients of the function stored in the storage unit 5. The function for converting the target measurement data into the correction value [Δc m,x Δc m,y Δc m,z can be represented by Equation 2 below. In Equation 2, (x m , y m , z m ) are the 3D coordinates of the measurement point. In the function of Equation 2, correction values corresponding to each component of the 3D coordinates are calculated. [Δc m,x Δc m,y Δc m,z = [f x (x m , y m , z m ) f y (x m , y m , z m ) f z (x m , y m , z m )] ··· Equation 2 For example, by dividing the function in the Z direction (depth direction), more than one such function can be prepared. By doing so, the calculation unit 44 can use a function with higher precision as each polynomial function to calculate the correction value, and thus obtain a correction result with higher precision.

[0101] Examples of the calculation unit 44 generating correction values corresponding to reference points have been described. The calculation unit 44 can calculate correction values corresponding to positions other than the reference points at which the preliminary measurement data was acquired by interpolating based on multiple correction values corresponding to multiple reference points. The calculation unit 44 can divide the measurement space of each measurement system at equal intervals with a cubic grid, calculate correction values corresponding to each grid point by interpolating based on multiple correction values corresponding to multiple reference points, and store a table associating the 3D coordinates of the grid points, the index for identifying the measurement system, and the calculated correction values in the storage unit 5. By calculating the correction values of each grid point, the calculation unit 44 can perform correction when correcting the measurement result of the 3D coordinates of the measurement object by using the correction value of the grid point closest to the measured coordinate value.

[0102]

Correction of Target Measurement Data

[0103] The correction unit 45 corrects the target measurement data of the measurement system corresponding to the preliminary measurement data that does not match the reference data based on the correction value. As an example, when the reference data generation unit 43 generates reference data by selecting the preliminary measurement data measured by the third measurement system, the correction unit 45 corrects the target measurement data of the first measurement system different from the third measurement system. The correction unit 45 reads from the storage unit 5 the correction value associated with: (i) the index of the installation position where the difference between the coordinate in the depth direction indicated by the target measurement data and the coordinate in the depth direction of the reference point is the smallest, (ii) the index indicating the first measurement system, and (iii) the index of the reference point whose 3D coordinates are closest to the 3D coordinates indicated by the target measurement data. The correction unit 45 corrects the target measurement data measured by the first measurement system based on the read correction value.

[0104] The correction unit 45 can read from the storage unit 5 the correction value associated with the 3D coordinates of the grid point closest to the 3D coordinates indicated by the target measurement data and the index indicating the measurement system corresponding to the grid point, and correct the target measurement data based on the read correction value. The correction unit 45 can read from the storage unit 5 the correction values associated with the 3D coordinates of multiple grid points relatively close to the 3D coordinates indicated by the target measurement data, and obtain the correction value corresponding to the target measurement data through an interpolation process using the read multiple correction values. Moreover, when obtaining a function for converting the target measurement data into a correction value, the correction unit 45 can use this function to convert the target measurement data into a correction value and correct the target measurement data based on the converted correction value.

[0105] The geometric shape recognition unit 46 uses the target measurement data corrected by the correction unit 45 to recognize the geometric shape of the measurement object. The geometric shape recognition unit 46 recognizes the 3D geometric shape of the measurement object by connecting the 3D coordinates of the measurement points indicated by the plurality of corrected target measurement data. The geometric shape recognition unit 46 can recognize the geometric shape of the measurement object by connecting the 3D coordinates indicated by the plurality of target measurement data corresponding to different measurement systems.

[0106] There may be a case where the correction unit 45 does not correct the target measurement data of the measurement system corresponding to the reference data. Therefore, when using the target measurement data corresponding to the reference data, it is assumed that the geometric shape recognition unit 46 can recognize the geometric shape of the measurement object by using the target measurement data not corrected by the correction unit 45. The geometric shape recognition unit 46 can recognize the geometric shape of the measurement object by connecting (i) the 3D coordinates of the measurement points indicated by the uncorrected target measurement data corresponding to the reference data and (ii) the 3D coordinates of the measurement points indicated by the corrected target measurement data.

[0107]

Processing procedure for calculating correction value

[0108] Figure 10 is a flowchart showing the processing procedure for calculating the correction value by the 3D geometric shape measuring device 100. When the operation receiving unit (not shown) of the 3D geometric shape measuring device 100 receives an operation indicating calibration of each measurement system from a user, this processing procedure starts.

[0109] First, the preliminary measurement unit 412 generates captured images of a predetermined reference instrument 51 using a plurality of measurement systems (S101). The preliminary measurement unit 412 measures the 3D coordinates of the reference points and generates preliminary measurement data indicating the measured 3D coordinates (S102). It is assumed that the preliminary measurement unit 412 generates a plurality of preliminary measurement data corresponding to a plurality of measurement systems in S102. The reference data generation unit 43 generates reference data based on one or more of the generated plurality of preliminary measurement data (S103).

[0110] The calculation unit 44 calculates the difference between (i) the preliminary measurement data of the measurement systems that do not match the measurement system of the reference data among the plurality of measurement systems and (ii) the reference data generated by the reference data generation unit 43. The calculation unit 44 calculates a correction value for correcting the target measurement data measured by the measurement system corresponding to the preliminary measurement data that does not match the reference data based on this difference (S104), and ends the processing.

[0111]

Processing procedure for recognizing three-dimensional geometric shape

[0112] Figure 11It is a flowchart showing the process of identifying the 3D geometry of a measurement object performed by the 3D geometry measurement device 100. This process starts when the operation reception unit of the 3D geometry measurement device 100 receives an instruction from the user to start measuring the 3D geometry.

[0113] First, the target measurement unit 411 projects a projection image including a predetermined light pattern onto the measurement object or the like using the projection unit 3 (S201). The target measurement unit 411 generates a first captured image captured by the first capture unit 1 by capturing the projection image projected onto the measurement object (S202). The target measurement unit 411 generates target measurement data indicating the measurement result by measuring the measurement points of the measurement object (S203). The correction unit 45 corrects the target measurement data based on the correction value (S204). The geometry recognition unit 46 recognizes the geometry of the measurement object based on the target measurement data corrected by the correction unit 45 (S205).

[0114]

Effects of the 3D Geometry Measurement Device of the Present Embodiment

[0115] According to the present embodiment, the correction unit 45 corrects the target measurement data for multiple measurement systems using the reference data. Therefore, the correction unit 45 can prevent a decrease in measurement accuracy caused by errors in the measurement system.

[0116] Since the geometry recognition unit 46 can perform measurements by connecting the target measurement data of multiple measurement systems with high precision, even if there are parts where the target measurement data cannot be obtained by a certain measurement system due to shadows, saturation, etc., the target measurement data of another measurement system can be used to recognize the geometry of the measurement object. Therefore, the geometry recognition unit 46 can recognize the geometry of the measurement object more efficiently and accurately.

[0117]

Variant Example

[0118] In the present embodiment, an example is described in which the reference data generation unit 43 generates reference data including the 3D coordinates of multiple reference points using the preliminary measurement data corresponding to a single measurement system. However, the present disclosure is not limited to this. The reference data generation unit 43 can also generate a certain reference data by selecting one from multiple preliminary measurement data corresponding to different measurement systems for each reference point. For example, the reference data generation unit 43 can select the preliminary measurement data corresponding to the measurement system corresponding to the combination of optical devices where the parallax becomes the largest from multiple preliminary measurement data corresponding to different measurement systems, so that the accuracy of triangulation for the reference point will be the highest.

[0119] Figure 12A and Figure 12BEach shows how to select a measurement system when the reference data generation unit 43 generates reference data. Figure 12A and Figure 12B Each shows the reference instrument 51 as viewed from above. In Figure 12A and Figure 12B 's example, in addition to the first imaging unit 1 and the second imaging unit 2, the 3D geometric shape measurement device further includes a third imaging unit 300. In Figure 12A and Figure 12B 's example, the projection unit 3 is omitted. For example, when using the reference points on the reference instrument 51, the preliminary measurement unit 412 can use the following fourth to sixth measurement systems to generate preliminary measurement data.

[0120] Fourth measurement system: the first imaging unit 1 and the second imaging unit 2

[0121] Fifth measurement system: the first imaging unit 1 and the third imaging unit 300

[0122] Sixth measurement system: the second imaging unit 2 and the third imaging unit 300

[0123] Figure 12A shows how to select the measurement system corresponding to the reference point at the right end of the lowermost stage of the reference instrument 51. The reference data generation unit 43 selects the measurement system corresponding to the combination of optical devices where the parallax of the reference point becomes the largest. When the reference data generation unit 43 selects the fourth measurement system, the parallax formed by the first imaging unit 1 and the second imaging unit 2 for the reference point at the right end of the lowermost stage is α1. On the other hand, when the reference data generation unit 43 selects the fifth measurement system, the parallax formed by the first imaging unit 1 and the third imaging unit 300 for the reference point at the right end of the lowermost stage is α2, and this parallax is the largest among the fourth to sixth measurement systems. Therefore, when generating the reference data corresponding to the reference point at the right end of the lowermost stage, the reference data generation unit 43 selects the preliminary measurement data measured by the fifth measurement system.

[0124] Figure 12B shows how to select the measurement system corresponding to the reference point at the left end of the uppermost stage of the reference instrument 51. When the reference data generation unit 43 selects the fourth measurement system, the parallax formed by the first imaging unit 1 and the second imaging unit 2 for the reference point at the left end of the uppermost stage is β1. On the other hand, when the reference data generation unit 43 selects the sixth measurement system, the parallax formed by the second imaging unit 2 and the third imaging unit 300 for the reference point at the left end of the uppermost stage is β2, and this parallax is the largest among the fourth to sixth measurement systems. Therefore, when generating the reference data corresponding to the reference point at the left end of the uppermost stage, the reference data generation unit 43 selects the preliminary measurement data measured by the sixth measurement system. The reference data generation unit 43 generates a single reference data by combining the preliminary measurement data selected for each reference point of the reference instrument 51.

[0125] With this configuration, the reference data generation unit 43 generates reference data using the preliminary measurement data measured by the measurement system in which the parallax of each reference point is the largest, so that the measurement accuracy of the 3D coordinates can be improved by using triangulation.

[0126] In the case of using the reference points projected onto the reference instrument 51 by one or more projection units, the reference data generation unit 43 is not limited to an example of selecting the preliminary measurement data measured by the measurement system for each reference point, in which the parallax formed by the pair of two imaging units is the largest. For example, the reference data generation unit 43 can consider the parallax formed by the pair of two imaging units and the parallax formed by the pair of the projection unit or the imaging units, so as to select the preliminary measurement data measured by the measurement system including the pair with the largest parallax for each reference point.

[0127] If the specifications of the imaging units are different from each other, or the distances from the imaging units to the reference points are different from each other, there may be a case where the measurement system with the largest parallax does not match the measurement system with the best measurement accuracy. The reference data generation unit 43 can read a table associating the approximate values of the 3D coordinates of the reference points with the measurement system having the best measurement accuracy from the storage unit 5, and by referring to this table, select the preliminary measurement data of the measurement system having the best measurement accuracy.

[0128] Furthermore, the reference data generation unit 43 can also select the preliminary measurement data of the measurement system having the best measurement accuracy based on the relative distance from the origin to the reference point. For example, when measuring the reference instrument 51, the reference data generation unit 43 designates an arbitrary reference point on the reference instrument 51 as the origin. The reference data generation unit 43 measures the distance from the identified origin to the reference point C which is a reference point other than the origin. On the other hand, the reference data generation unit 43 reads a pre-stored value from the storage unit 5 as the distance from the origin to the reference point C. The reference data generation unit 43 can also select the preliminary measurement data of the measurement system with the smallest difference between the measured distance and the distance read from the storage unit 5 as the preliminary measurement data with the best measurement accuracy for the reference point C.

[0129] The present disclosure is described based on the exemplary embodiments. The technical scope of the present disclosure is not limited to the scope described in the above embodiments, and various changes and modifications can be made within the scope of the present disclosure. For example, the specific embodiments of the distribution and integration of the devices are not limited to the above embodiments, and all or part of them can be configured with any units that are functionally or physically dispersed or integrated. Further, the new exemplary embodiments generated by any combination of them are included in the exemplary embodiments of the present disclosure. Further, the effects of the new exemplary embodiments brought by the combination also have the effects of the original exemplary embodiments.

[0130]

Description of Reference Numerals

[0131] 1 First shooting unit

[0132] 2 Second shooting unit

[0133] 3 Projection unit

[0134] 4 Control unit

[0135] 5 Storage unit

[0136] 11 Lens

[0137] 12 Imaging element

[0138] 21 Lens

[0139] 22 Imaging element

[0140] 41 Measurement unit

[0141] 42 Acquisition unit

[0142] 43 Reference data generation unit

[0143] 44 Calculation unit

[0144] 45 Correction unit

[0145] 46 Geometric shape recognition unit

[0146] 100 Three-dimensional geometric shape measurement device

[0147] 300 Third shooting unit

[0148] 411 Target measurement unit

[0149] 412 Preliminary measurement unit

Claims

1. A three-dimensional geometric shape measuring device for measuring the three-dimensional geometric shape of a measurement object based on a captured image obtained by photographing the measurement object, the three-dimensional geometric shape measuring device comprising: a plurality of optical devices; a preliminary measurement unit that generates a plurality of preliminary measurement data by photographing a reference instrument in a plurality of measurement systems corresponding to combinations of different ones of the plurality of optical devices, the plurality of preliminary measurement data indicating three-dimensional coordinates of a plurality of reference points on the reference instrument; a reference data generation unit that generates, for each of the reference points, a reference data by selecting one of the plurality of preliminary measurement data corresponding to the plurality of measurement systems based on one or more of the plurality of preliminary measurement data; a calculation unit that calculates a correction value based on the reference data and the preliminary measurement data among the plurality of preliminary measurement data that do not match the reference data; a target measurement unit that generates a plurality of target measurement data, the plurality of target measurement data indicating results of measuring measurement points of the measurement object using the plurality of measurement systems; a correction unit that corrects the target measurement data in the measurement system corresponding to the preliminary measurement data that does not match the reference data based on the correction value; and a geometric shape recognition unit that recognizes the geometric shape of the measurement object using the corrected target measurement data corrected by the correction unit.

2. A three-dimensional geometric shape measuring device for measuring the three-dimensional geometric shape of a measurement object based on a captured image obtained by photographing the measurement object, the three-dimensional geometric shape measuring device comprising: a plurality of optical devices; an acquisition unit that acquires relative position information indicating the positional relationship of a plurality of reference points included in a reference instrument; a preliminary measurement unit that generates a plurality of preliminary measurement data by photographing the reference instrument in a plurality of measurement systems corresponding to combinations of different ones of the plurality of optical devices, the plurality of preliminary measurement data indicating three-dimensional coordinates of the plurality of reference points on the reference instrument; a reference data generation unit that calculates a statistic of a plurality of errors for each of the plurality of measurement systems, and generates reference data by selecting the preliminary measurement data corresponding to the measurement system selected based on the calculated statistic, the plurality of errors being a plurality of errors between the coordinates of a plurality of reference points other than the origin when one reference point of the reference instrument is used as the origin, identified based on the plurality of positional relationships indicated by the plurality of preliminary measurement data, and the coordinates of the plurality of reference points other than the origin identified based on the positional relationship indicated by the relative position information; a calculation unit that calculates a correction value based on the reference data and the preliminary measurement data among the plurality of preliminary measurement data that do not match the reference data; a target measurement unit that generates a plurality of target measurement data, the plurality of target measurement data indicating results of measuring measurement points of the measurement object using the plurality of measurement systems; A correction unit that corrects target measurement data in the measurement system corresponding to the preliminary measurement data that does not match the reference data based on the correction value; and A geometric shape recognition unit that uses the corrected target measurement data corrected by the correction unit to recognize the geometric shape of the measurement object.

3. The three-dimensional geometric shape measurement device according to claim 1 or 2, wherein the preliminary measurement unit generates the preliminary measurement data indicating the three-dimensional coordinates of the reference points of the reference instrument at the plurality of installation positions by photographing the reference instrument installed at the plurality of installation positions, and the calculation unit calculates the correction value corresponding to the reference points at the plurality of installation positions.

4. The three-dimensional geometric shape measurement device according to claim 1 or 2, wherein the reference data generation unit generates the reference data by calculating the statistic of the three-dimensional coordinates of the reference points indicated by the plurality of preliminary measurement data corresponding to different plurality of measurement systems.

5. The three-dimensional geometric shape measurement device according to claim 1 or 2, further comprising: An acquisition unit that acquires relative position information indicating the positional relationship of the plurality of reference points included in the reference instrument, wherein the preliminary measurement unit generates the preliminary measurement data indicating the three-dimensional coordinates of the plurality of reference points, and the reference data generation unit generates the reference data by selecting the preliminary measurement data indicating the three-dimensional coordinates of the plurality of reference points based on the error from the positional relationship indicated by the relative position information.

6. The three-dimensional geometric shape measurement device according to claim 1 or 2, wherein the preliminary measurement unit generates the preliminary measurement data indicating the three-dimensional coordinates of the plurality of reference points on the reference instrument, and the reference data generation unit generates one reference data for each of the reference points by selecting one of the plurality of preliminary measurement data corresponding to different plurality of measurement systems.

7. The three-dimensional geometric shape measurement device according to claim 6, wherein the reference data generation unit selects the preliminary measurement data corresponding to the combination of the optical devices for which the triangulation accuracy for the reference point becomes optimal from the plurality of preliminary measurement data corresponding to different plurality of measurement systems.

8. The three-dimensional geometric shape measurement device according to claim 1 or 2, wherein the preliminary measurement unit projects a projection image including a pattern for identifying the reference point onto the reference instrument.

9. The three-dimensional geometric shape measurement device according to claim 1 or 2, wherein the calculation unit identifies a function for calculating the correction value corresponding to the three-dimensional coordinates of the measurement point measured by the measurement system corresponding to the preliminary measurement data that does not match the reference data, and uses the identified function to calculate the correction value.

10. The three-dimensional geometric shape measurement device according to claim 1 or 2, wherein The calculation unit further includes a storage unit storing a table in which (i) three-dimensional coordinates of each grid point of a measurement space divided by a cubic grid at equal intervals of the plurality of measurement systems, (ii) a plurality of the correction values, and (iii) indices for identifying each of the measurement systems are associated with each other, where the correction unit reads, from the storage unit, the correction value associated with the three-dimensional coordinates of the grid point closest to the three-dimensional coordinates indicated by the target measurement data and the index indicating the measurement system corresponding to the grid point, and corrects the preliminary measurement data based on the read correction value.

11. A three-dimensional geometric shape measurement method for measuring a three-dimensional geometric shape of a measurement object based on a captured image obtained by photographing the measurement object, the three-dimensional geometric shape measurement method including the following steps: Generating a plurality of preliminary measurement data indicating three-dimensional coordinates of a plurality of reference points on a reference instrument by photographing the reference instrument with a plurality of measurement systems corresponding to combinations of different optical devices among a plurality of optical devices; Generating one reference data for each of the reference points by selecting, based on one or more of the plurality of preliminary measurement data, one of the plurality of preliminary measurement data corresponding to the plurality of measurement systems; Calculating a correction value based on the reference data and the preliminary measurement data among the plurality of preliminary measurement data that do not match the reference data; Generating a plurality of target measurement data indicating results of measuring measurement points of the measurement object using the plurality of measurement systems; Correcting the target measurement data in the measurement system corresponding to the preliminary measurement data that does not match the reference data based on the correction value; And Identifying the geometric shape of the measurement object using the corrected target measurement data.

12. A three-dimensional geometric shape measurement method for measuring a three-dimensional geometric shape of a measurement object based on a captured image obtained by photographing the measurement object, the three-dimensional geometric shape measurement method including the following steps: Obtaining relative position information indicating a positional relationship of a plurality of reference points included in a reference instrument; Generating a plurality of preliminary measurement data indicating three-dimensional coordinates of the plurality of reference points on the reference instrument by photographing the reference instrument with a plurality of measurement systems corresponding to combinations of different optical devices among a plurality of optical devices; Calculating a statistic of a plurality of errors for each of the plurality of measurement systems, and generating reference data by selecting the preliminary measurement data corresponding to the measurement system selected based on the calculated statistic, the plurality of errors being a plurality of errors between coordinates of a plurality of reference points other than an origin when one reference point of the reference instrument is used as the origin, identified based on a plurality of positional relationships indicated by the plurality of preliminary measurement data, and coordinates of the plurality of reference points other than the origin, identified based on the positional relationship indicated by the relative position information; Calculate a correction value based on the reference data and the preliminary measurement data among the multiple preliminary measurement data that do not match the reference data; Generate multiple target measurement data indicating the results of measuring the measurement points of the measurement object using the multiple measurement systems; Correct the target measurement data in the measurement system corresponding to the preliminary measurement data that does not match the reference data based on the correction value; And Use the corrected target measurement data to identify the geometry of the measurement object.

Images