Three-dimensional measuring device

Abstract

To provide a configuration that allows three-dimensional measurement of a measurement object at a higher speed by using event data.SOLUTION: A stripe pattern P projected onto a measurement object R by a projection section 20 is generated such that stripe regions Pa to Pd in which luminance changes at a predetermined rate in a horizontal direction (first direction) and the luminance does not change in a vertical direction (second direction orthogonal to the first direction) are arranged along the horizontal direction. An imaging section 30 includes an imaging element that outputs event data, including two-dimensional point data in which a position of a pixel is identified corresponding to the pixel of which luminance is changed when light is received. A measuring section 40 measures a three-dimensional shape of the measurement object R by an optical cutting method on the basis of the position of the pixel identified from the negative polarity event data output in the same time zone when the measurement object R is imaged in a unit time.SELECTED DRAWING: Figure 4

Description

Technical Field

[0004] ,

[0001] The present invention relates to a three-dimensional measurement apparatus for measuring the three-dimensional shape of a measurement object.

Background Art

[0002] Conventionally, as a three-dimensional measurement apparatus for measuring the three-dimensional shape or the like of a measurement object, for example, an apparatus using a phase shift method is known. The phase shift method is a technique for performing three-dimensional measurement on a measurement object onto which a stripe pattern image is projected by projecting a plurality of stripe pattern images with a phase shift.

[0003] Regarding the technique of performing three-dimensional measurement using the phase shift method, in order to generate an image of a measurement object at a higher speed, a three-dimensional measurement apparatus disclosed in Patent Document 1 below is known. In this three-dimensional measurement apparatus, a sine wave pattern is adopted as a predetermined stripe pattern for the phase shift method, and an event camera that outputs event data including two-dimensional point data in which the position of a pixel is specified corresponding to a pixel having a luminance change when receiving light is adopted, and an imaging image of the measurement object onto which the stripe pattern is projected is generated from the event data. Since the event camera does not output pixel information without luminance change, that is, redundant data (event data), like a conventional camera, reduction of data communication volume, reduction of image processing, etc. are realized, and information regarding the shape of the measurement object can be acquired at a higher speed. On the other hand, since the event data does not include luminance information used in the phase shift method, luminance information (stripe pattern information) is obtained based on the time difference between the generation time of event data with a positive luminance change (positive polarity event data) output at the time of light projection in pixel units and the generation time of event data with a negative luminance change (negative polarity event data) output at the time of light extinction, and three-dimensional shape measurement of the measurement object using the event data can be performed.

Prior Art Documents

Patent Documents

[0004] [Patent Document 1] Japanese Patent Publication No. 2021-067644 [Overview of the project] [Problems that the invention aims to solve]

[0005] Furthermore, the three-dimensional measuring device disclosed in Patent Document 1 projects a pattern for the light section method onto the object to be measured, which is generated such that the brightness increases at a constant rate from left to right and the brightness does not change in the vertical direction. The three-dimensional shape of the object to be measured is then measured using the light section method, which utilizes the event data output during imaging. Specifically, the position of a pixel identified from the event data output at the same time is treated as the position of the pixel irradiated by the line-shaped light in the light section method, thereby enabling three-dimensional measurement using the light section method.

[0006] However, in the measurement method that projects the pattern generated for the light section method as described above, it is necessary to change the brightness from the left edge to the right edge of the projection area. As the number of pixels in the left-right direction increases, the imaging time, and thus the processing time required for three-dimensional measurement, becomes longer.

[0007] This invention was made to solve the above-mentioned problems, and its objective is to provide a configuration that enables faster three-dimensional measurement of an object by utilizing event data. [Means for solving the problem]

[0008] To achieve the above objective, the invention described in claim 1 of the claims is: A projection unit (20) that projects a predetermined stripe pattern (P) onto the object to be measured (R), An imaging unit (30) that images the object to be measured onto which the predetermined stripe pattern is projected, A measurement unit (40) measures the three-dimensional shape of the object to be measured, which is captured by the imaging unit, A control unit (11) that controls the projection unit, A three-dimensional measuring device (40) comprising, The predetermined stripe pattern is generated such that multiple stripe regions (Pa~Pd) are arranged along the first direction, where the brightness changes by a predetermined rate in the first direction and the brightness does not change in the second direction perpendicular to the first direction. The imaging unit includes an image sensor that outputs event data including two-dimensional point data that identifies the position of a pixel corresponding to a pixel that showed a change in brightness when light was received. The measurement unit is characterized by measuring the three-dimensional shape of the object to be measured by light sectioning based on the position of pixels identified from the event data output during the same time period when the object to be measured is imaged within a unit of time. The symbols within the parentheses above indicate the correspondence with the specific means described in the embodiments described later. [Effects of the Invention]

[0009] In the invention of claim 1, a predetermined stripe pattern projected by the projection unit onto the object to be measured is generated such that multiple stripe regions are arranged along the first direction, where the brightness changes by a predetermined rate in a first direction and the brightness does not change in a second direction perpendicular to the first direction. The imaging unit includes an image sensor that outputs event data including two-dimensional point data that identifies the position of a pixel corresponding to a pixel that has changed brightness when light is received, and the measurement unit measures the three-dimensional shape of the object to be measured by the light section method based on the position of a pixel identified from the event data output during the same time period when the object to be measured is imaged within a unit of time.

[0010] As a result, a bright line connecting the pixel positions identified from event data output during the same time period is obtained for each fringe region. Therefore, the more fringe regions there are in a given fringe pattern, the shorter the imaging time, and thus the processing time required for three-dimensional measurement. Consequently, by utilizing event data, three-dimensional measurement of the object to be measured can be performed at a faster speed.

[0011] In the invention of claim 2, the measurement unit uses an event waveform generated such that the output timing of the event data output during imaging is on the vertical axis and the position of the pixel in the first direction is on the horizontal axis, to measure the three-dimensional shape of the object to be measured by the light section method based on the position of the pixel identified from the event data output during the same time period.

[0012] In the imaging unit, if a single pixel receives light from both an opaque and a transparent material, it may output two event data points for that single pixel. In such cases, the first event data points output earliest for each pixel may be a mixture of those from the opaque and transparent materials, which can hinder accurate three-dimensional measurement.

[0013] Therefore, an event waveform is generated with the output timing of the event data output during imaging as the vertical axis and the position of the pixel in the first direction as the horizontal axis. This event waveform is generated so that, for pixels imaging both an opaque and a transparent object, two event data are output for that pixel, and one event waveform is generated that is caused by the shape of the opaque object and the other event waveform is caused by the shape of the transparent object. Therefore, by using the event waveforms separated from the event waveform generated as described above, it is possible to distinguish between opaque and transparent objects and perform three-dimensional measurement.

[0014] In the invention of claim 3, the measurement unit uses an event waveform generated with the output timing of negative-polarity event data output during imaging as the vertical axis and the position of the pixel in the first direction as the horizontal axis to measure the three-dimensional shape of the object to be measured by the light section method based on the position of the pixel identified from the event data output during the same time period. The event waveform is generated such that, for pixels imaging both an opaque and a transparent object, when two negative-polarity event data are output for that pixel, one event waveform caused by the shape of the opaque object and the other event waveform caused by the shape of the transparent object are included. Therefore, by using one event waveform and the other event waveform separated from the event waveform generated as described above, it is possible to distinguish between opaque and transparent objects and perform three-dimensional measurement. In particular, the image sensor receives light at a high voltage in the bright state immediately before the output of negative-polarity event data, so the sensitivity is relatively low and the effect of noise is suppressed. Therefore, by using the output timing of negative-polarity event data, the measurement accuracy for three-dimensional measurement can be improved compared to when the output timing of positive-polarity event data is used.

[0015] In the invention of claim 4, the projection unit first projects a predetermined stripe pattern as a first stripe pattern, and then projects a second stripe pattern in which the predetermined proportion of brightness in the first stripe pattern is reversed. The measurement unit obtains a first event waveform generated with the output timing of the earliest output event data during imaging of the first stripe pattern as the vertical axis and the position of the pixel in the first direction as the horizontal axis, and a second event waveform generated with the output timing of the earliest output event data during imaging of the second stripe pattern as the vertical axis and the position of the pixel in the first direction as the horizontal axis. Using a composite waveform obtained by combining the first event waveform and the inverted second event waveform, the unit measures the three-dimensional shape of the object to be measured by the light section method based on the position of the pixel identified from the event data output during the same time period.

[0016] As in the invention of claim 2, when using two event data output for each pixel in a single image capture, event data that should not be output may be generated as noise, such as when event data of the same polarity is output consecutively for the same pixel after the first event data has been output, which can hinder accurate three-dimensional measurement.

[0017] In contrast, the first and second event waveforms are generated based on the output timing of the earliest event data, and are therefore unaffected by the noise mentioned above. Furthermore, since the second fringe pattern reverses the predetermined ratio of brightness and darkness in the first fringe pattern, the relative brightness and darkness of opaque and transparent objects at the same pixel position are reversed. As a result, the event data that should be output second when the first fringe pattern is projected is output first. Therefore, the composite waveform, which is a combination of the first event waveform generated when the first fringe pattern is projected and the inverted second event waveform generated when the second fringe pattern is projected, is generated to include one event waveform caused by the shape of the opaque object and the other event waveform caused by the shape of the transparent object. This makes it possible to perform three-dimensional measurement by distinguishing between opaque and transparent objects while suppressing the effects of noise.

[0018] In the invention of claim 5, the projection unit first projects a predetermined stripe pattern as a first stripe pattern, and then projects a second stripe pattern in which the predetermined proportion of brightness in the first stripe pattern is reversed. The measurement unit obtains a first event waveform generated with the output timing of the first negative-polarity event data output earliest during imaging of the first stripe pattern as the vertical axis and the position of the pixel in the first direction as the horizontal axis, and a second event waveform generated with the output timing of the first negative-polarity event data output earliest during imaging of the second stripe pattern as the vertical axis and the position of the pixel in the first direction as the horizontal axis. Using a composite waveform obtained by combining the first event waveform and the inverted second event waveform, the unit measures the three-dimensional shape of the object to be measured by the light section method based on the position of the pixel identified from the event data output during the same time period.

[0019] Since the first event waveform and the second event waveform are each generated based on the output timing of the earliest negative-polarity event data, there is no influence of the noise described above. And since the second stripe pattern reverses the predetermined ratio in the first stripe pattern with respect to light and darkness, the magnitude relationship between light and darkness of the opaque body and the transparent body at the same pixel position is reversed. Therefore, the negative-polarity event data that should be output second when the first stripe pattern is projected is output earliest. For this reason, the composite waveform obtained by synthesizing the first event waveform generated when the first stripe pattern is projected and the waveform obtained by reversing the second event waveform generated when the second stripe pattern is projected is generated so as to include one event waveform caused by the shape of the opaque body and the other event waveform caused by the shape of the transparent body. Thereby, it is possible to perform three-dimensional measurement while suppressing the influence of noise and distinguishing between the opaque body and the transparent body. In particular, in the bright state immediately before the negative-polarity event data is output, the imaging element receives light in a high-voltage state, so the sensitivity is relatively low, and the influence of noise is suppressed. For this reason, by using the output timing of the negative-polarity event data, it is possible to improve the measurement accuracy regarding three-dimensional measurement as compared with the case of using the output timing of the positive-polarity event data.

[0020] In the invention of claim 6, the projection unit further projects a stripe number specifying pattern for specifying a stripe number that distinguishes a stripe region from other stripe regions. The stripe number specifying means specifies the stripe number based on the output timing of the event data output when imaging the measurement object on which the stripe number specifying pattern is projected. For pixels for which two stripe numbers are specified because the opaque body and the transparent body are being imaged, the stripe number of the opaque body and the stripe number of the transparent body are respectively specified based on the three-dimensional shape of the measurement object assumed in advance.

[0021] In the configuration where the stripe number is specified based on the output timing of the event data output earliest at each pixel when imaging the projected stripe number specifying pattern, there is a possibility that an incorrect stripe number may be specified because two event data are output at different timings for the opaque body and the transparent body.

[0022] Therefore, on the premise that two event data may be output at different timings in pixel units when imaging the stripe number identification pattern, the stripe number identification means identifies the stripe number, thereby suppressing the incorrect identification of the stripe number based only on the output timing of the earliest output event data. For pixels for which two stripe numbers are identified because an opaque object and a transparent object are imaged, based on the three-dimensional shape of the measurement object assumed in advance, the stripe number of the opaque object and the stripe number of the transparent object are respectively identified, thereby suppressing the misidentification of the stripe numbers of the opaque object and the transparent object.

[0023] In the invention of claim 7, the projection unit further projects a first stripe number identification pattern and a second stripe number identification pattern for identifying a stripe number that distinguishes a stripe region from other stripe regions. The first stripe number identification pattern is generated and projected such that the luminance changes in a predetermined step so as to be different in the stripe region in the first direction and does not change in the second direction. The second stripe number identification pattern is generated and projected by reversing the predetermined step in the first stripe pattern with respect to light and dark. The stripe number identification means identifies the stripe number based on the output timing of the earliest output event data when imaging the measurement object projected with the first stripe number identification pattern and the output timing of the earliest output event data when imaging the measurement object projected with the second stripe number identification pattern. For pixels for which two stripe numbers are identified because an opaque object and a transparent object are imaged, based on the three-dimensional shape assumed in advance, the stripe number of the opaque object and the stripe number of the transparent object are respectively identified.

[0024] When using two event data output for each pixel in one imaging as in the invention of claim 6, due to reasons such as the case where event data of the same polarity is continuously output at the same pixel after the first event data is output, event data that should not be output originally may occur as noise and may prevent the accurate identification of the stripe number.

[0025] ​​In contrast, the stripe number identification means identifies the stripe number based on the output timing of the event data that is output earliest when the first stripe number identification pattern is imaged and the output timing of the event data that is output earliest when the second stripe number identification pattern is imaged. Therefore, the aforementioned noise does not affect the identification of the stripe number. Furthermore, since the second stripe number identification pattern reverses the brightness and darkness of a predetermined stage in the first stripe number identification pattern, the event data that should be output second when the first stripe number identification pattern is projected is output earliest. As a result, it is possible to suppress the effects of noise and prevent misidentification of stripe numbers between opaque and transparent objects. [Brief explanation of the drawing]

[0026] [Figure 1] This is a block diagram showing the schematic configuration of a three-dimensional measuring device according to the first embodiment. [Figure 2] Figure 2(A) is an explanatory diagram illustrating a fringe pattern for the light section method according to the first embodiment, and Figure 2(B) is an explanatory diagram illustrating the relationship between the light emission time and the position of pixels in the left-right direction for projecting the fringe pattern of Figure 2(A). [Figure 3] This is an explanatory diagram illustrating the image generated using event data output with the stripe pattern shown in Figure 2 projected onto a plane. Figure 3(A) shows the state 10 μs after the projection of the stripe pattern in Figure 2 begins, Figure 3(B) shows the state 20 μs later, and Figure 3(C) shows the state 30 μs later. [Figure 4] This is an explanatory diagram illustrating the image generated using event data output with the stripe pattern shown in Figure 2 projected onto a spherical object being measured. Figure 4(A) shows the state 10 μs after the projection of the stripe pattern in Figure 2 begins, Figure 4(B) shows the state 20 μs later, and Figure 4(C) shows the state 30 μs later. [Figure 5]This diagram illustrates a comparison between the output timing of negative event data when capturing a stripe pattern consisting of four stripe regions, based on the pixel position in the left-right direction, and the output timing of negative event data when capturing a stripe pattern consisting of one stripe region. [Figure 6] Figure 6(A) is an explanatory diagram illustrating a stripe pattern for the light section method according to a modified example of the first embodiment, and Figure 6(B) is an explanatory diagram illustrating the relationship between the light emission time and the position of pixels in the left-right direction for projecting the stripe pattern of Figure 6(A). [Figure 7] This is an explanatory diagram illustrating the state of measuring an object that has a transparent plate in front of the blackboard in the second embodiment. [Figure 8] Figure 8(A) is an explanatory diagram illustrating the stripe pattern of a triangular wave; Figure 8(B) is an explanatory diagram illustrating the event waveform generated when only the blackboard is imaged; Figure 8(C) is an explanatory diagram illustrating the event waveform generated when only the transparent plate is imaged; and Figure 8(D) is an explanatory diagram illustrating the event waveform generated when an object being measured has a transparent plate in front of the blackboard. [Figure 9] Figure 9(A) is an explanatory diagram illustrating the event waveform portion used for three-dimensional measurement when the presence of a transparent plate is not considered, and Figure 9(B) is an explanatory diagram illustrating the measurement results using the event waveform portion of Figure 9(A). [Figure 10] These are explanatory diagrams illustrating two types of stripe patterns in the third embodiment; Figure 10(A) shows the first stripe pattern, and Figure 10(B) shows the second stripe pattern. [Figure 11] Figure 11(A) is an explanatory diagram illustrating the first event waveform generated when imaging the first fringe pattern shown in Figure 10(A); Figure 11(B) is an explanatory diagram illustrating the measurement results using the first event waveform from Figure 11(A); Figure 11(C) is an explanatory diagram illustrating the second event waveform generated when imaging the second fringe pattern shown in Figure 10(B); and Figure 11(D) is an explanatory diagram illustrating the measurement results using the second event waveform from Figure 11(C). [Figure 12]This is an explanatory diagram illustrating the composite waveform, which is created by combining the first event waveform and the inverted second event waveform. [Figure 13] This is an explanatory diagram illustrating the pattern for identifying the stripe number in the fourth embodiment. Figure 13(A) shows the light emission time in the stripe region of stripe number "1", Figure 13(B) shows the light emission time in the stripe region of stripe number "2", Figure 13(C) shows the light emission time in the stripe region of stripe number "3", and Figure 13(D) shows the light emission time in the stripe region of stripe number "4". [Figure 14] This is an explanatory diagram illustrating the relationship between the event waveforms generated when measuring an object with a transparent plate in front of a blackboard, the stripe number identified by one event waveform due to the shape of the blackboard, and the stripe number identified by the other event waveform due to the shape of the transparent plate. [Figure 15] Figure 15(A) is an explanatory diagram illustrating the case where one negative event data is output when the stripe numbers are the same for the blackboard and the transparent plate, and Figure 15(B) is an explanatory diagram illustrating the case where two negative event data are output when the stripe numbers are different for the blackboard and the transparent plate. [Figure 16] This is an explanatory diagram illustrating the stripe number identification pattern in the fifth embodiment. Figure 16(A) shows the first stripe number identification pattern, and Figure 16(B) shows the second stripe number identification pattern. [Figure 17] This is an explanatory diagram illustrating the predetermined ratios used to generate the striped pattern. [Modes for carrying out the invention]

[0027] [First Embodiment] Hereinafter, a first embodiment of the three-dimensional measuring device of the present invention will be described with reference to the drawings. The three-dimensional measuring device 10 according to this embodiment is a device for measuring the three-dimensional shape of an object R, and as shown in Figure 1, it is configured to include a control unit 11 that controls the overall system, a projection unit 20 that projects a predetermined stripe pattern onto the object R, an imaging unit 30 that images the object R onto which the predetermined stripe pattern has been projected, and a measuring unit 40 that measures the three-dimensional shape of the object R from the image. The three-dimensional measuring device 10 configured in this way measures the three-dimensional shape of an object R, such as a workpiece, which moves at high speed relative to the hand, for example, when it is attached to the hand of a robot. Here, relative movement refers to the relative movement between the movement of the three-dimensional measuring device 10 attached to the hand of the robot and the high-speed movement of the object R. If the position of the three-dimensional measuring device 10 is fixed, the relative movement is the movement of the object R.

[0028] The control unit 11 is mainly composed of a microcontroller and has a CPU, system bus, input / output interface, etc., and together with a storage unit consisting of ROM, RAM, non-volatile memory, etc., it constitutes an information processing device. In addition to programs related to robot control, the storage unit has programs related to the control of the projection unit 20 and programs for executing control processing using the three-dimensional measurement results from the measurement unit 40, etc., which are pre-stored in a manner that can be executed by the control unit 11.

[0029] The projection unit 20 is a so-called DLP projector, controlled by the control unit 11, which projects a predetermined stripe pattern, as described later, by reflecting light from a light source using DMD elements. The DMD elements are arrays of tiny mirrors corresponding to each pixel of the image projected onto the screen, and are configured to switch the light emitted onto the screen ON / OFF in microsecond increments by changing the angle of each mirror. Therefore, each mirror can be switched from reflection OFF to reflection ON to a projection state, and from reflection ON to reflection OFF to a power-off state. In other words, the projection unit 20 functions to project a predetermined stripe pattern by controlling the ON / OFF of the reflection of incident light by the DMD, which is an array of multiple mirrors, for each mirror, as controlled by the control unit 11. Therefore, by changing the gradation (brightness) of the reflected light depending on the ratio of the time each mirror is ON and OFF, it becomes possible to display gradation based on the image data of the projected image. In this embodiment, the projection unit 20 is equipped with a mirror corresponding to k × l pixels (for example, 1140 × 912) where the upper left is (1, 1) and the lower right is (k, l).

[0030] In this configuration, the longer the emission time (time from reflection ON to reflection OFF) of a single pulse emission that is emitted once within a unit time allocated for each emission state, the brighter the emission state becomes. Therefore, the emission state can be identified according to the emission time. For example, if red (R), green (G), and blue (B) light are provided as incident light for the DMD element, the R-color emission state, where R light is emitted when reflected by the mirror, the G-color emission state, where G light is emitted when reflected by the mirror, and the B-color emission state, where B light is emitted when reflected by the mirror, are repeated at a predetermined short period, and the emission time for each is individually adjusted, making it possible to project a color image. For this reason, the control unit 11 functions to set the ON / OFF timing of reflection within a unit time for each mirror according to a predetermined stripe pattern, which will be described later.

[0031] The imaging unit 30 is a so-called event camera and is equipped with an image sensor that outputs event data (specifically, two-dimensional point data, time, and polarity of brightness change) including two-dimensional point data that identifies the position of a pixel corresponding to a pixel that has changed brightness when light is received. The imaging unit 30 is configured to generate an image from the event data output from the image sensor. For this reason, for each pixel in the image, the imaging unit 30 outputs event data of positive polarity (positive brightness change) when a brightness change occurs due to receiving light, and outputs event data of negative polarity (negative brightness change) when a brightness change occurs due to the disappearance of that light. By plotting the two-dimensional point data of multiple event data output within a certain period as points on a predetermined plane, image data of the object to be measured R can be generated, and the imaging unit 30 is configured to output the image data or event data (two-dimensional point data, time, polarity of brightness change) generated in this way to the measurement unit 40.

[0032] The measurement unit 40 is controlled by the control unit 11 and measures the three-dimensional shape of the object R to be measured by the light section method based on an image captured by the imaging unit 30 of the object R to be measured, in which a predetermined stripe pattern is projected from the projection unit 20.

[0033] Therefore, the projection unit 20 projects a predetermined fringe pattern for the light section method (hereinafter also referred to as fringe pattern P). In this embodiment, the fringe pattern P is generated such that multiple fringe regions are arranged along the first direction, where the brightness changes at a predetermined rate in the first direction and the brightness does not change in the second direction perpendicular to the first direction. Specifically, the fringe pattern P is generated such that, like the sawtooth wave illustrated in Figures 2(A) and 2(B), four fringe regions Pa to Pd are arranged along the left-right direction (first direction), and each fringe region Pa to Pd is generated such that the brightness increases at a constant rate from left to right in the left-right direction (first direction) and the brightness does not change in the up-down direction (second direction). As can be seen from Figure 2(B), by increasing the emission time of pixels on the right side for each fringe region, the fringe pattern P shown in Figure 2(A) is generated and projected.

[0034] In the imaging unit 30, which images the surface onto which the generated stripe pattern P is projected, the emission lines connecting the positions of pixels identified from the negative-polarity event data output during the same time period are captured as vertical lines for each stripe region Pa to Pd, moving from left to right within a unit of time, like a video. This is because the light emission ends earlier as you approach the left edge of the stripe region because it becomes darker, and later as you approach the right edge of the stripe region because it becomes brighter, while the timing of the end of light emission is almost the same in the vertical direction.

[0035] Specifically, for example, when the imaging unit 30 captures an image 10 μs after the projection unit 20 starts projecting the stripe pattern P onto a plane, the emission lines Sa to Sd are captured in a vertical line shape near the left edge of each stripe region Pa to Pd, as shown in Figure 3(A). Subsequently, for example, after 20 μs, the emission lines Sa to Sd are captured to move to the right, as shown in Figure 3(B), and after 30 μs, the emission lines Sa to Sd are captured to move further to the right, as shown in Figure 3(C).

[0036] Emitting lines (lines connecting the positions of pixels identified from negative-polarity event data) that are imaged in a linear fashion can be used as linear laser beams in the light section method.

[0037] The following describes the three-dimensional shape measurement of a spherical object R using the light section method, using the emission lines Sa to Sd captured by the imaging unit 30 when the aforementioned stripe pattern P is projected onto the object R as an example, with reference to Figures 4(A) to (C). For convenience, the object R is shown as a dashed line in Figures 4(A) to (C).

[0038] Immediately after imaging the fringe pattern P on the object R, in the fringe regions Pb and Pc, the emission lines Sb and Sc are imaged in such a way that they deform according to the shape of the object R. Then, 10 μs after the start of imaging, in Pb and Pc, as shown in Figure 4(A), the emission lines Sb and Sc are imaged in such a way that they move to the right relative to the object R, deforming according to the relationship between the position of the emission lines Sb and Sc and the shape of the object R. Subsequently, 20 μs and 30 μs after the start of imaging, in Pb and Pc, as shown in Figures 4(B) and (C), the emission lines Sb and Sc are imaged in such a way that they move further to the right relative to the object R, deforming according to the relationship between the position of the emission lines Sb and Sc and the shape of the object R. On the other hand, in the fringe regions Pa and Pd, as can be seen from Figures 4(A) to (C), the emission lines Sa and Sd are imaged in such a way that they move to the right in a line in the vertical direction, so as not to overlap with the object R.

[0039] Therefore, the measurement unit 40 can measure the three-dimensional shape of the object R by light section method based on the position of each emission line Sa to Sd captured when the object R is imaged within the above unit time, that is, the position of the pixel identified from the negative polarity event data output during the same time period.

[0040] As described above, in the three-dimensional measuring device 10 according to this embodiment, the stripe pattern P projected by the projection unit 20 onto the object to be measured R is generated such that multiple stripe regions Pa to Pd are arranged along the left-right direction, where the brightness changes by a predetermined rate in the left-right direction (first direction) and the brightness does not change in the up-down direction (second direction perpendicular to the first direction). The imaging unit 30 is equipped with an image sensor that outputs event data including two-dimensional point data that identifies the position of a pixel corresponding to a pixel that has changed brightness when light is received, and the measurement unit 40 measures the three-dimensional shape of the object to be measured R by the light section method based on the position of a pixel identified from the negative-polarity event data output during the same time period when the object to be measured R is imaged within a unit time.

[0041] As a result, a line of light (Sa~Sd) connecting the positions of pixels identified from negative-polarity event data output during the same time period is obtained for each fringe region (Pa~Pd). Therefore, the more fringe regions there are in a predetermined fringe pattern projected from the projection unit 20, the shorter the imaging time required to image that fringe pattern P can be.

[0042] For example, a striped pattern consisting of only one striped region where the brightness changes at a constant rate in the left-right direction is projected so that the emission time gradually increases towards the right edge. Therefore, when this striped pattern is projected onto a plane and imaged, the output timing of negative polarity event data becomes later towards the right edge (see dashed line Lb in Figure 5), resulting in a longer imaging time (corresponding to the symbol Tb in Figure 5) for imaging this striped pattern. On the other hand, when a striped pattern P consisting of four striped regions Pa to Pd is imaged, as in this embodiment, the output timing of negative polarity event data occurs simultaneously in all four striped regions (see solid line La in Figure 5), allowing the above imaging time (corresponding to the symbol Ta in Figure 5) to be reduced to 1 / 4.

[0043] In this way, the imaging time required to capture the stripe pattern, that is, the processing time required for three-dimensional measurement, can be shortened. As a result, by using event data, three-dimensional measurement of the object R can be performed at a faster speed.

[0044] Furthermore, the predetermined fringe pattern for the light section method projected from the projection unit 20 is not limited to being generated as a sawtooth wave as exemplified in Figures 2(A) and (B). For example, it may be generated as a triangular wave as exemplified in Figures 6(A) and (B), such that multiple fringe regions are arranged along the first direction, where the brightness changes at a predetermined rate in the first direction and the brightness does not change in the second direction perpendicular to the first direction.

[0045] [Second Embodiment] Next, a three-dimensional measuring device according to a second embodiment of the present invention will be described with reference to the drawings. This second embodiment differs from the first embodiment in that it distinguishes between opaque and transparent objects and measures them three-dimensionally using the light section method. Therefore, components that are substantially the same as those in the first embodiment are denoted by the same reference numerals, and their descriptions are omitted.

[0046] In the imaging unit 30, if light from both an opaque and a transparent material is received for a single pixel, two negative-polarity event data may be output for that single pixel. In such cases, the first negative-polarity event data output for each pixel may be a mixture of those from the opaque and transparent materials, which can hinder accurate three-dimensional measurement.

[0047] For example, consider a case where, as shown in Figure 7, a transparent plate R2 is placed in front of a blackboard R1, and a triangular wave fringe pattern P (a fringe pattern in which the ratio of brightness change between adjacent fringes is reversed in terms of brightness and darkness) as shown in Figure 8(A) is projected onto the object R to be measured, and the three-dimensional shape of the object R is measured using the light section method as described above. Note that the blackboard R1 corresponds to an example of an "opaque object," and the transparent plate R2 can correspond to an example of a "transparent object."

[0048] In this case, if only the blackboard R1 is imaged, the waveform (hereinafter also referred to as the event waveform) is generated with the output timing of the negative polarity event data output at the time of imaging as the vertical axis and the position of the pixels in the left-right direction (first direction) as the horizontal axis, as exemplified in Figure 8(B). Also, if only the transparent plate R2 is imaged, the event waveform is generated as exemplified in Figure 8(C). Thus, the event waveform generated when imaging an object R with the transparent plate R2 in front of the blackboard R1 is as shown in Figure 8(D). Note that in Figure 8 and Figure 9 described later, the event waveform generated due to the shape of the blackboard R1 is shown by a solid line, and the event waveform generated due to the shape of the transparent plate R2 is shown by a dashed line.

[0049] If the presence of the transparent plate R2 is not considered, three-dimensional measurement will be performed using the first negative polarity event data output earliest at each pixel. In this case, three-dimensional measurement will be performed using the event waveform portion related to the blackboard R1 and the event waveform portion related to the transparent plate R2, as shown by the thick lines in Figure 9(A). As a result, the blackboard R1 and transparent plate R2 that should be measured as shown in Figure 7 will be measured with parts of the blackboard R1 and transparent plate R2 missing, as shown in Figure 9(B).

[0050] Therefore, assuming that the transparent body is measuring an object R located in front of an opaque body, the measurement unit 40 in this embodiment uses the event waveform generated as described above to measure the three-dimensional shape of the object R using the light section method based on the position of pixels identified from the event data output during the same time period.

[0051] Specifically, the measurement unit 40 generates an event waveform with the output timing of negative-polarity event data output during imaging as the vertical axis and the position of the pixels in the left-right direction (first direction) as the horizontal axis. For pixels imaging both opaque and transparent objects, two negative-polarity event data are output for that pixel. Therefore, as illustrated in Figure 8(D), the event waveform is generated to include one event waveform caused by the shape of the opaque object (see solid line waveform in Figure 8(D)) and the other event waveform caused by the shape of the transparent object (see dashed line waveform in Figure 8(D)). Because opaque and transparent objects are separated, one event waveform and the other event waveform can be easily separated. Therefore, by utilizing the one event waveform and the other event waveform separated from the event waveform generated as described above, three-dimensional measurement can be performed by distinguishing between opaque and transparent objects. In particular, generally, in the bright state immediately before the output of negative-polarity event data, the image sensor receives light at a high voltage, resulting in relatively low sensitivity and suppression of noise effects. Therefore, by utilizing the output timing of negative-polarity event data, it is possible to improve the measurement accuracy related to three-dimensional measurement compared to using the output timing of positive-polarity event data.

[0052] [Third Embodiment] Next, a three-dimensional measuring device according to a third embodiment of the present invention will be described with reference to the drawings. This third embodiment differs from the second embodiment in that it distinguishes between opaque and transparent objects and performs three-dimensional measurement using the light section method, based on the premise of projecting two types of stripe patterns. Therefore, components that are substantially the same as those in the second embodiment are denoted by the same reference numerals, and their descriptions are omitted.

[0053] As in the second embodiment described above, when using two negative-polarity event data output for each pixel in a single image capture, event data that should not be output may be generated as noise, hindering accurate three-dimensional measurement, for example, when event data of the same polarity is output consecutively for the same pixel after the first negative-polarity event data has been output.

[0054] Therefore, in this embodiment, the projection unit 20 first projects a predetermined stripe pattern as the first stripe pattern P1, and then projects a second stripe pattern P2 which is obtained by reversing the predetermined ratio of brightness in the first stripe pattern P1. Specifically, for example, after projecting the first stripe pattern P1 generated as shown in Figure 10(A), the second stripe pattern P2 generated as shown in Figure 10(B) is projected. As shown in Figures 10(A) and (B), the stripes P2a to P2h of the second stripe pattern P2 are generated by reversing the left-right ratio of the change in brightness in the left-right direction (first direction) with respect to brightness by reversing the ratio of brightness change in the left-right direction (first direction).

[0055] The measurement unit 40 then determines a first event waveform W1, which is generated with the output timing of the first negative-polarity event data output earliest during imaging of the first fringe pattern P1 as the vertical axis and the position of the pixel in the first direction as the horizontal axis. Subsequently, the measurement unit 40 determines a second event waveform W2, which is generated with the output timing of the first negative-polarity event data output earliest during imaging of the second fringe pattern P2 as the vertical axis and the position of the pixel in the first direction as the horizontal axis. Then, the measurement unit 40 uses a composite waveform W, which is a combination of the first event waveform W1 and the inverted second event waveform W2 obtained as described above, to measure the three-dimensional shape of the object to be measured R by the light section method based on the position of the pixel identified from the event data output during the same time period.

[0056] Specifically, for example, when imaging an object R with a transparent plate R2 in front of a blackboard R1 as shown in Figure 7, the first event waveform W1 when imaging the first stripe pattern P1 shown in Figure 10(A) is generated as shown in Figure 11(A). This first event waveform W1 is generated according to the shape of a part of the blackboard R1 and a part of the shape of the transparent plate R2, as shown in Figure 11(B).

[0057] Subsequently, the second event waveform W2, generated during imaging of the second fringe pattern P2 shown in Figure 10(B), is generated as shown in Figure 11(C). This second event waveform W2 is generated according to the shape of the remaining portion of the blackboard R1 and the shape of the remaining portion of the transparent plate R2, as shown in Figure 11(D).

[0058] The first event waveform W1 and the second event waveform W2 are generated based on the output timing of the earliest negative-polarity event data, and are therefore unaffected by the noise mentioned above. Furthermore, since the second fringe pattern P2 reverses the predetermined ratio of brightness in the first fringe pattern P1, the relative brightness (time) of the blackboard R1 and the transparent plate R2 at the same pixel position is reversed. As a result, the negative-polarity event data that should be output second during the projection of the first fringe pattern P1 is output first.

[0059] Therefore, as shown in Figure 12, the composite waveform W, which is created by combining the first event waveform W1 generated when the first fringe pattern P1 is projected and the second event waveform W2 generated when the second fringe pattern P2 is projected and then inverted, is generated so that it includes one event waveform caused by the shape of the opaque object and the other event waveform caused by the shape of the transparent object. By utilizing the one event waveform and the other event waveform separated from the event waveform generated as described above, it is possible to perform three-dimensional measurement while suppressing the effects of noise and distinguishing between opaque and transparent objects. Note that if no noise is generated, the composite waveform W shown in Figure 12 and the event waveform shown in Figure 8(D) will be substantially the same waveform.

[0060] [Fourth Embodiment] Next, a three-dimensional measuring device according to a fourth embodiment of the present invention will be described with reference to the drawings. This fourth embodiment differs from the second embodiment in that it accurately identifies the stripe numbers of the opaque material and the transparent material, respectively, and distinguishes between the opaque and transparent materials for three-dimensional measurement using the light section method. Therefore, components that are substantially the same as those in the second embodiment are given the same reference numerals, and their descriptions are omitted.

[0061] When imaging a stripe pattern consisting of multiple stripe regions, it is necessary to identify which stripe region is being imaged for each pixel. Therefore, in order to identify a stripe number that distinguishes a stripe region from other stripe regions, the projection unit 20 further projects a stripe number identification pattern onto the measurement target R that has been imaged with the above-mentioned stripe pattern. The measurement unit 40 performs a stripe number identification process to identify the stripe number based on the output timing at which negative polarity event data is output when the measurement target R onto which the stripe number identification pattern is projected is imaged. The measurement unit 40 that performs the stripe number identification process may correspond to an example of a "stripe number identification means".

[0062] Specifically, for example, in a stripe pattern with four stripe regions numbered "1" to "4" from left to right, the stripe number identification pattern is projected such that, in the stripe region with stripe number "1", the light emission time ends 1 ms from the start of projection, as illustrated in Figure 13(A); in the stripe region with stripe number "2", the light emission time ends 2 ms from the start of projection, as illustrated in Figure 13(B); in the stripe region with stripe number "3", the light emission time ends 3 ms from the start of projection, as illustrated in Figure 13(C); and in the stripe region with stripe number "4", the light emission time continues until the end of projection, as illustrated in Figure 13(D). As a result, in the stripe number identification process performed by the measurement unit 40, for example, the stripe number of the pixel that outputs the first negative polarity event data 1 ms after the start of projection of the stripe number identification pattern can be identified as "1", and the stripe number of the pixel that outputs the first negative polarity event data 2 ms after the start of projection can be identified as "2". The illumination time corresponding to each stripe region can be appropriately changed according to the measurement target R, measurement environment, etc.

[0063] However, in a configuration where the fringe number is identified based on the output timing of the earliest negative-polarity event data output at each pixel, as described above, when imaging the projected fringe number identification pattern, there is a possibility that an incorrect fringe number may be identified because two negative-polarity event data are output at different timings for the opaque and transparent objects.

[0064] Therefore, in the fringe number identification process of this embodiment, the fringe number is identified on the premise that when the fringe number identification pattern is captured, two negative polarity event data may be output at different timings on a pixel-by-pixel basis.

[0065] Specifically, for example, when measuring an object R with a transparent plate R2 in front of a blackboard R1, a striped pattern of four stripes of triangular waves is projected, and when an event waveform is generated that includes one event waveform caused by the shape of the blackboard R1 (see solid line waveform in Figure 14) and the other event waveform caused by the shape of the transparent plate R2 (see dashed line waveform in Figure 14), the aforementioned stripe number identification pattern (see Figures 13(A) to (D)) is projected.

[0066] In this case, as can be seen from Figure 14, at pixel G1, the stripe number is the same "1" on both the blackboard R1 and the transparent plate R2. Therefore, as shown in Figure 15(A), a change in brightness occurs 1 ms after the start of projection of the stripe number identification pattern, causing a positive event data to be generated, followed 1 ms later by the output of the first negative event data. Thus, when one negative event data is output during imaging of the stripe number identification pattern, one stripe number is identified for that pixel, preventing the identification of incorrect stripe numbers on the blackboard R1 and the transparent plate R2.

[0067] On the other hand, at pixel G2, the stripe number for blackboard R1 is "1" and the stripe number for transparent plate R2 is "2". As shown in Figure 15(B), the first negative event data is output when the first brightness change occurs 1 ms after the start of projection, and then the second negative event data is output when the second brightness change occurs 2 ms after the start of projection. In this way, when two negative event data are output at different timings for a single pixel during imaging of the stripe number identification pattern, two stripe numbers, "1" and "2", are identified as candidates. When two stripe numbers are identified as candidates in this way, the stripe numbers for blackboard R1 and transparent plate R2 are identified based on the pre-predicted three-dimensional shape of the object to be measured R. Information regarding the three-dimensional shape of the object to be measured R can be stored in advance in the memory unit of the control unit 11.

[0068] In other words, the measurement results are compared between the case where the stripe number on blackboard R1 is "1" and the stripe number on transparent plate R2 is "2," and the case where the stripe number on blackboard R1 is "2" and the stripe number on transparent plate R2 is "1." The system then determines which measurement result is closer to the pre-defined three-dimensional shape of the object R being measured. In the example above, the measurement result when the stripe number on blackboard R1 is "1" and the stripe number on transparent plate R2 is "2" is determined to be closer to the three-dimensional shape of the object R being measured, and the stripe number on blackboard R1 is identified as "1" and the stripe number on transparent plate R2 as "2."

[0069] Thus, in the fringe number identification process, by identifying the fringe number based on the premise that two negative-polarity event data may be output at different timings for each pixel when the fringe number identification pattern is captured, it is possible to suppress the incorrect identification of the fringe number that would otherwise occur if the output timing were based solely on the output timing of the earliest output negative-polarity event data. Furthermore, for pixels where two fringe numbers are identified because both an opaque and a transparent object is being captured, the fringe numbers of the opaque object and the transparent object are identified separately based on the pre-determined three-dimensional shape of the object being measured, thereby suppressing the misidentification of fringe numbers between opaque and transparent objects.

[0070] [Fifth Embodiment] Next, a three-dimensional measuring device according to a fifth embodiment of the present invention will be described with reference to the drawings. This fifth embodiment differs from the fourth embodiment in that it accurately identifies the stripe numbers of the opaque and transparent objects, respectively, by projecting two types of stripe number identification patterns, and then distinguishes between the opaque and transparent objects for three-dimensional measurement using the light section method. Therefore, components that are substantially the same as those in the fourth embodiment are given the same reference numerals, and their descriptions are omitted.

[0071] As in the fourth embodiment described above, when using two negative-polarity event data output for each pixel in a single image capture, event data that should not be output may be generated as noise, such as when event data of the same polarity is output consecutively for the same pixel after the first negative-polarity event data has been output, which can hinder the accurate identification of the fringe number.

[0072] Therefore, in this embodiment, the projection unit 20 further projects a first stripe number identification pattern PP1 and a second stripe number identification pattern PP2 for identifying stripe numbers that distinguish stripe regions from other stripe regions. The first stripe number identification pattern PP1 has a brightness that is consistent with the stripe region in the left-right direction (first direction). eachThe pattern is generated and projected so that it changes in predetermined stages in a different manner, and the brightness does not change in the vertical direction (second direction). The second fringe number identification pattern PP2 is generated and projected so that the predetermined stages in the first fringe number identification pattern PP1 are reversed in terms of brightness.

[0073] Specifically, for example, if the first fringe number identification pattern PP1 is generated and projected as shown in Figure 16(A) because its emission time is controlled as shown in Figures 13(A) to (D), then the second fringe number identification pattern PP2 will be generated and projected as shown in Figure 16(B). That is, if the first fringe number identification pattern PP1 is generated and projected so that the fringe regions PP1a, PP1b, and PP1c become darker in that order, then the second fringe number identification pattern PP2 will be generated and projected so that the fringe region PP2d corresponds to fringe region PP1d, fringe region PP2c corresponds to fringe region PP1c, and fringe region PP2b corresponds to fringe region PP1b become darker in that order.

[0074] Then, in the stripe number identification process performed by the measurement unit 40, the stripe number is identified based on the output timing of the negative polarity event data that is output earliest when the measurement target R onto which the first stripe number identification pattern PP1 is projected is imaged, and the output timing of the negative polarity event data that is output earliest when the measurement target onto which the second stripe number identification pattern PP2 is projected is imaged.

[0075] Specifically, for each pixel, the fringe number is identified from the output timing of negative polarity event data when the first fringe number identification pattern PP1 is projected, and the fringe number is identified from the output timing of negative polarity event data when the second fringe number identification pattern PP2 is projected. When the first fringe number identification pattern PP1 is projected, the fringe numbers are identified from left to right in the order of "1", "2", "3", and "4", and when the second fringe number identification pattern PP2 is projected, the fringe numbers are identified from right to left in the order of "4", "3", "2", and "1". If the identified fringe numbers match, that fringe number is identified as the fringe number of that pixel.

[0076] On the other hand, if there are pixels with different identified fringe numbers, the transparent object is assumed to be measuring the object R located in front of the opaque object, and the two fringe numbers are identified as candidates, similar to the fourth embodiment described above. For example, in the pixel G2 described above, fringe number "1" is identified when the first fringe number identification pattern PP1 is projected, and fringe number "2" is identified when the second fringe number identification pattern PP2 is projected. This is because the order of the identified fringe numbers is reversed between the projection of the first fringe number identification pattern PP1 and the projection of the second fringe number identification pattern PP2, with the order reversed from left to right.

[0077] Thus, when two stripe numbers are identified as candidates, the stripe numbers for the opaque object and the transparent object are determined based on the pre-defined three-dimensional shape of the object R to be measured, as described above.

[0078] Since the fringe number is identified based on the output timing of the first negative-polarity event data output when imaging with the first fringe number identification pattern PP1 and the output timing of the first negative-polarity event data output when imaging with the second fringe number identification pattern PP2, the aforementioned noise does not affect the identification of the fringe number. Furthermore, since the second fringe number identification pattern PP2 reverses the brightness and darkness of a predetermined stage in the first fringe number identification pattern PP1, the second negative-polarity event data that should be output when projecting the first fringe number identification pattern PP1 is output first. Therefore, it is possible to suppress the effects of noise while preventing the misidentification of fringe numbers between opaque and transparent objects.

[0079] The present invention is not limited to the embodiments described above, and may be further embodied as follows, for example. (1) The striped regions constituting the striped pattern P described above are not limited to four, but may be two, three, or five or more. Similarly, the striped regions constituting the first striped pattern P1 and the second striped pattern P2 described above are not limited to eight, but may be two to seven, or nine or more. In this case, the striped number identification patterns such as the first striped number identification pattern PP1 and the second striped number identification pattern PP2 can be generated to have the same number of striped regions.

[0080] (2) The three-dimensional measuring device 10 is not limited to moving while mounted on the robot's hand to measure the three-dimensional shape of a relative moving object, but may also be used in a fixed state to measure the three-dimensional shape of an object moving along a conveyor line.

[0081] (3) The three-dimensional measuring device 10 may be configured such that the projection unit 20 and imaging unit 30 and the measuring unit 40 are separate entities, and the measuring unit 40 is configured as an information processing terminal capable of wireless or wired communication with the projection unit 20 and imaging unit 30.

[0082] (4) In each of the above embodiments, three-dimensional measurement of the object to be measured R is performed using the output timing of negative polarity event data. However, the three-dimensional measurement of the object to be measured R may also be performed using the output timing of positive polarity event data, provided that the brightness change of a predetermined stripe pattern projected from the projection unit 20 is appropriately adjusted.

[0083] (5) In each of the above embodiments, the stripe pattern projected from the projection unit 20 is not limited to being generated as illustrated in Figures 2 and 6, etc., but is not limited to being generated in such a way that multiple stripe regions are arranged along the first direction, where the brightness changes at a predetermined rate in the first direction and the brightness does not change in the second direction perpendicular to the first direction. Furthermore, the predetermined rate may differ for each stripe region, provided that it is set in advance. Specifically, the predetermined rate may be set, for example as illustrated in Figure 17, such that for stripe number "1" (first period), it increases linearly in the first period, increases logarithmically in the next period, decreases linearly in the last period, and decreases linearly for stripe number "2" (second period). [Explanation of Symbols]

[0084] 10… Three-dimensional measuring device 11…Control Unit 20…Projection section 30…Imaging Unit 40... Measuring unit (means for identifying stripe numbers) P... Stripe pattern P1...First stripe pattern P2...Second stripe pattern Pa~Pd,P1a~P1h,P2a~P2h...stripe area PP1…Pattern for identifying the first stripe number PP2…Second stripe number identification pattern R...Object to be measured R1…Blackboard (non-transparent) R2…Transparent plate (transparent body) Sa~Sd…Emission line W...Composite event waveform W1…First event waveform W2…Second event waveform

Claims

1. A projection unit that projects a predetermined stripe pattern onto the object to be measured, An imaging unit that images the object to be measured onto which the predetermined stripe pattern is projected, A measurement unit that measures the three-dimensional shape of the object to be measured, which is captured by the imaging unit, A control unit for controlling the projection unit, A three-dimensional measuring device comprising, The predetermined stripe pattern is generated such that multiple stripe regions are arranged along the first direction, where the brightness changes by a predetermined rate in the first direction and the brightness does not change in the second direction perpendicular to the first direction. The imaging unit includes an image sensor that outputs event data including two-dimensional point data that identifies the position of a pixel corresponding to a pixel that showed a change in brightness when light was received. The three-dimensional measuring device is characterized in that the measurement unit measures the three-dimensional shape of the object to be measured by the light section method based on the position of pixels identified from the event data output during the same time period when the object to be measured is imaged within a unit of time.

2. The three-dimensional measuring device according to claim 1, characterized in that the measurement unit uses an event waveform generated such that the output timing of the event data output during imaging is the vertical axis and the position of the pixel in the first direction is the horizontal axis, and measures the three-dimensional shape of the object to be measured by the light section method based on the position of the pixel identified from the event data output during the same time period.

3. The image sensor is configured to output positive event data in the case of a brightness change that causes it to become brighter, and negative event data in the case of a brightness change that causes it to become darker. The three-dimensional measuring device according to claim 1, characterized in that the measurement unit uses an event waveform generated such that the output timing of the negative polarity event data output during imaging is the vertical axis and the position of the pixel in the first direction is the horizontal axis, and measures the three-dimensional shape of the object to be measured by the light section method based on the position of the pixel identified from the event data output during the same time period.

4. The projection unit first projects the predetermined stripe pattern as the first stripe pattern, and then projects a second stripe pattern obtained by reversing the predetermined ratio of light and dark in the first stripe pattern. The three-dimensional measuring device according to claim 1, characterized in that the measuring unit determines a first event waveform generated such that the output timing of the event data output earliest when the first fringe pattern is imaged is the vertical axis and the position of the pixel in the first direction is the horizontal axis, and a second event waveform generated such that the output timing of the event data output earliest when the second fringe pattern is imaged is the vertical axis and the position of the pixel in the first direction is the horizontal axis, and uses a composite waveform obtained by combining the first event waveform and the inverted second event waveform to measure the three-dimensional shape of the object to be measured by the light section method based on the position of the pixel identified from the event data output during the same time period.

5. The image sensor is configured to output positive event data in the case of a brightness change that causes it to become brighter, and negative event data in the case of a brightness change that causes it to become darker. The projection unit first projects the predetermined stripe pattern as the first stripe pattern, and then projects a second stripe pattern obtained by reversing the predetermined ratio of light and dark in the first stripe pattern. The three-dimensional measuring device according to claim 1, characterized in that the measuring unit determines a first event waveform generated such that the output timing of the negative polarity event data output earliest when the first fringe pattern is imaged is the vertical axis and the position of the pixel in the first direction is the horizontal axis, and a second event waveform generated such that the output timing of the negative polarity event data output earliest when the second fringe pattern is imaged is the vertical axis and the position of the pixel in the first direction is the horizontal axis, and uses a composite waveform obtained by combining the first event waveform and the inverted second event waveform to measure the three-dimensional shape of the object to be measured by the light section method based on the position of the pixel identified from the event data output during the same time period.

6. The projection unit further projects a stripe number identification pattern for identifying a stripe number that distinguishes the stripe region from other stripe regions. The measurement unit includes a stripe number identification means that identifies the stripe number based on the output timing of the event data output when the object to be measured is imaged with the stripe number identification pattern projected onto it. The three-dimensional measuring apparatus according to any one of claims 2 to 5, characterized in that the stripe number identification means identifies the stripe number of the non-transparent object and the stripe number of the transparent object, respectively, based on the pre-predicted three-dimensional shape of the object to be measured, for pixels in which two stripe numbers are identified because an opaque object and a transparent object are being imaged.

7. The projection unit further projects a first stripe number identification pattern and a second stripe number identification pattern for identifying stripe numbers that distinguish the stripe region from other stripe regions. The first stripe number identification pattern is generated and projected such that the brightness changes in predetermined steps in the first direction so that it differs for each stripe region, and the brightness does not change in the second direction. The second stripe number identification pattern is generated and projected such that the predetermined stages in the first stripe pattern are reversed in terms of brightness and darkness. The measurement unit includes a stripe number identification means that identifies the stripe number based on the output timing of the event data that is output earliest when the object to be measured is imaged with the first stripe number identification pattern projected onto it, and the output timing of the event data that is output earliest when the object to be measured is imaged with the second stripe number identification pattern projected onto it. The three-dimensional measuring apparatus according to claim 4 or 5, characterized in that the stripe number identification means identifies the stripe number of the non-transparent body and the stripe number of the transparent body, respectively, based on a predetermined three-dimensional shape, for pixels in which two stripe numbers are identified because an opaque body and a transparent body are being imaged.

Images