JP2004085467A - 3D measuring device - Google Patents

Abstract

An object of the present invention is to provide an apparatus capable of performing three-dimensional measurement without the shape of a measurement target object affecting measurement and without moving an imaging unit.
A tertiary system including a movable stage (6) movable in the X direction with respect to a base (2) and an imaging device (5) installed at a predetermined position with respect to the base (2) and imaging an object to be measured (10). In the original measurement device 1, the imaging device 5 is installed with the depth-of-field surface 11 inclined with respect to the measurement target 10, and moves the moving stage 6 in the X direction, thereby moving the measurement target 10. It was configured to measure the shape three-dimensionally.
[Selection diagram] Fig. 1

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a three-dimensional measuring device that optically captures a measurement target and measures a cross-sectional shape or a three-dimensional shape of the measurement target.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, a triangulation method is known as a method for measuring the shape of an object (measurement target) by a non-contact method. In the triangulation method, the distance to the measurement target is determined by viewing the measurement target from different directions. In this method, as shown in FIG. 10, a light projecting unit (for example, a laser projector) for projecting light and a light receiving unit (for example, a CCD camera) for receiving the projected light have a parallax. It is arranged and three-dimensional measurement is performed. In this case, depending on the positional relationship between the light projecting unit and the light receiving unit and the shape of the measurement target, a shadow (blind spot area) may be generated at the time of measurement, and the shape of the measurement target may not be accurately measured.
[0003]
Therefore, as another method in which the shadow does not affect the measurement of the measurement target, a lens focusing method (also referred to as a focusing method) represented by focusing in a camera such as a single-lens reflex camera is known. In this method, the light projecting means and the light receiving means are arranged coaxially, and the distance between the object to be measured and the light projecting means (for example, a lens) is adjusted variably until the focus is reached.
[0004]
For example, Japanese Patent Application Laid-Open No. Hei 9-145318 discloses a method of detecting a focused position by moving a depth of field by moving a lens, and Japanese Patent Application Laid-Open No. Hei 10-38559 discloses a method of detecting a focus by moving an image sensor. There is known a method of moving a plane and detecting a focused position.
[0005]
[Problems to be solved by the invention]
However, in these methods, when the size of the measurement target is larger than the imaging range of the camera that captures the measurement target, it is necessary to move the camera that captures the measurement target. In this case, the camera position is moved with respect to the measurement target, and the camera focus is moved. Then, a method is adopted in which the camera focus is changed, several images at that time are captured, and the shape of the measurement object is measured from the optimum focus position. In other words, in this method, the camera position must be moved up and down with respect to the measurement target, and the moving member on which the measurement target is placed must be moved together. For this reason, not only the moving member but also the camera position must be moved, which complicates the structure of the three-dimensional measuring apparatus itself.
[0006]
Therefore, the present invention has been made in view of the above problems, the shape of the measurement object does not affect the measurement, the structure is simple, without moving the imaging means, the shape of the measurement object A technical object is to provide a device that can perform measurement.
[0007]
[Means for Solving the Problems]
The technical means taken to solve the above-mentioned problem includes a base, a moving means capable of moving in one direction with respect to the base, and a moving means installed at a predetermined position with respect to the base. In a three-dimensional measuring apparatus including an imaging unit that images a measurement target placed on a member and a control unit that moves the movement unit, the imaging unit includes a depth of field with respect to the measurement target. An image processing unit is provided in a state where the surface is inclined, and the control unit moves the moving unit to three-dimensionally measure the object to be measured.
[0008]
According to the above-described configuration, the imaging unit is provided in a state where the depth of field is inclined, and the measurement target is imaged by the oblique method. By moving the moving means in one direction in a state where the object to be measured is imaged from an oblique direction while the depth of field is inclined, image coordinates (camera coordinates) based on the depth of field are used. Can be obtained, and the camera image can be converted into the actual object coordinates to measure the shape of the measurement target three-dimensionally. This means that the imaging means does not have to be moved with respect to the measurement target even if the measurement target is larger than the imaging range of the imaging means. For this reason, the measuring device does not need a structure for moving the imaging means with respect to the measurement target, and the structure is simpler than in the related art. The three-dimensional measurement here indicates a cross-sectional shape where the depth-of-field surface and the measurement object intersect, and by moving the measurement object and repeating this operation, the cross-sectional shape on the depth-of-field surface, The three-dimensional shape of the entire measurement object can be measured.
[0009]
In this case, the image processing unit differentiates the video signal captured by the imaging unit by differentiating the video signal captured by the imaging unit and calculating the cross-sectional coordinates of the measurement target based on the differentiated video signal. Thus, a change in the brightness of the video signal can be captured according to the shape obtained from the measurement object. Based on the brightness of the differentiated video signal, it is possible to measure the cross-sectional coordinates of the measurement object on the depth of field.
[0010]
In addition, when calculating the cross-sectional coordinates, when the line is thinned by the weighted average of the differentiated video signal, the noise of the video signal differentiated by the weighted average is removed, and thinning can be performed based on this. It becomes.
[0011]
Furthermore, the image processing means converts the cross-sectional coordinates into three-dimensional data and stores the three-dimensional cross-sectional coordinates by moving the moving means by the control means. Since the height direction in three dimensions is known, the shape of the measurement target can be measured in three dimensions.
[0012]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
[0013]
FIG. 1 shows a configuration of a three-dimensional measuring device (referred to as the present device) 1 that three-dimensionally measures the shape of a measurement object 10. As shown in FIG. 1, the apparatus 1 includes a metal base 2 placed on an XY plane and a metal support 3 standing upright in the Z direction. A metal support 4 extending in the Y direction is fixed above the column 3, and an image pickup device 5 having an image pickup device inside a CCD camera, a CMOS camera, or the like is fixed above the support 4. I have. The imaging device 5 is fixed to the support unit 4 in a state where the imaging device 5 is inclined by a predetermined angle with respect to the Z direction perpendicular to the base 2.
[0014]
On the other hand, a moving stage 6 is attached to the base 2, and the moving stage 6 moves in the X direction with respect to the base 2 by the power of a hydraulic device or a motor (not shown). On the moving stage 6, a moving base plate 7 is fixed by pins or fastening members, and a measuring object 10 whose shape is to be measured three-dimensionally is mounted on the moving base plate 7. Then, by moving the moving stage 6 in the X direction by the stage control of the image processing device 9, the measuring object 10 placed on the moving stage moves in the X direction. The imaging device 5 condenses light via an optical lens (simply referred to as a lens) 8 on the way from an oblique direction to the image when the measurement object 10 moves in the X direction, and forms an internal element (for example, (A CCD element or the like). In this case, in the imaging device 5, the region in focus is determined in advance according to the distance to the measurement object 10 in terms of characteristics. In the present embodiment, a surface area in which the image pickup device provided inside the image pickup apparatus is in focus at the time of image pickup and the measurement target 10 can be recognized as an image is defined as a depth-of-field surface 11. In the state shown in FIG. 1, the imaging device 5 is attached to the measurement target 10 on the base in a state where the imaging device 5 is inclined by a predetermined angle from the vertical Z direction, and the lens 8 is passed through. Then, the depth-of-field plane 11 is inclined with respect to the measurement object 10.
[0015]
FIG. 2 shows an outline of processing of a video signal when the shape of the measurement object 10 is measured in the present apparatus 1. In the present apparatus 1, since the video signal imaged by the imaging device 5 is an analog signal, it is converted from an analog signal to a digital signal by an A / D conversion circuit provided inside the image processing device 9. After that, the digitized video signal is subjected to image processing for binarization and thinning by a process such as differentiation inside the image processing apparatus. Then, the sectional coordinates of the measurement object 10 on the depth-of-field plane 11 are calculated, and the differential value (data) obtained on the depth-of-field plane 11 is stored in a three-dimensional coordinate table.
[0016]
On the other hand, when the sectional coordinates in the above-described state are calculated, the moving stage 6 is moved by the stage control of the moving stage 6. Thus, while the moving stage 6 is moved in the X direction again, the section coordinates with respect to the depth of field 11 are calculated, and some section coordinates are stored. As a result, when the moving stage 6 is fixed, the shape of the measurement object 10 with respect to the depth of field 11 is such that the two-dimensional depth of field 11 is It appears only in one cross-sectional shape when passing, and the shape of the measuring object 10 can be measured only in one cross-section. However, by moving the moving stage 6 in the X direction and calculating the cross-sectional coordinates of the depth of field 11 in the X direction, the measurement object 10 and the depth of field 11 at each position in the X direction are determined. The cross-sectional shape can be measured in multiple cross-sections and the shape can be stored. In the present apparatus 1, even if the measurement target 10 is imaged by moving only one axis of the moving stage 6, the shape of the measurement target 10 can be changed even for the measurement target 10 that is larger than the imaging range of the imaging device 5. Can be measured.
[0017]
Next, an image formed on the imaging device 5 will be described from a geometrical viewpoint with reference to the explanatory diagram shown in FIG. As shown in FIG. 2, the image picked up by the imaging device 5 has camera coordinates (I, I) at which the position where the measurement object 10 intersects with the depth-of-field plane 11 (the intersection position) is the coordinate axis of the imaging device 5. (J coordinate). The video signal obtained by this image formation is subjected to A / D conversion and differential processing by the image processing device 9, and the differentiated value (differential value) is binarized based on this. Image processing for thinning is performed, and the cross-sectional shape of the measurement target 10 is measured.
[0018]
Therefore, referring to FIG. 4, the image formed on the coordinate system (camera coordinate system) of the image sensor is changed to the actual object coordinate system (world coordinate system) on which the measurement object 10 is placed. The coordinate conversion will be described.
[0019]
According to Scheimpflug's law, when the image pickup device inside the image pickup device 5 is tilted by φ around the Y axis with respect to the Z axis perpendicular to the base 2, the depth of field 11 with respect to the Z axis Around the Y axis, it is inclined by θ = tan ⁻¹ (di · tan φ / d0). In this case, d0 will be described below as the distance from the origin of the object coordinates to the center (principal point position) of the lens 8 and di: the distance from the principal point position of the lens 8 to the intersection of the imaging device 5 and the Z axis. .
[0020]
Next, a method of measuring the cross-sectional shape of the measurement object 10 performed in the image processing device 9 will be described with reference to a flowchart illustrated in FIG. In the following flowchart, the steps of the program will be described simply as “S”.
[0021]
When the imaging device 5 captures an image of the measurement object 10 placed on the moving stage 6, the imaging device 5 is fixed to the column 3 while being inclined at a predetermined angle with respect to the Z axis. Since the image pickup device 5 is attached to the support portion 4, when the image pickup device 5 picks up an image from the measurement object 10, a focused region and a non-focused region are mixed. In S1, these are subjected to A / D conversion as a video signal of the measurement object 10 and digitized. In the next step S2, in the course of the image processing, when the image from the measuring object 10 is picked up by the internal CCD element or the like, the image pickup device 5 scans, but the image is focused on the image pickup element. In order to judge whether the object is in focus (degree of focus), a differentiation process is performed for each scanning line, and a differentiation parameter represented by a differential value is obtained.
[0022]
For example, when the i-th brightness of a certain scanning line (scanning line number: j) is represented as I (i, j), the differential parameter f (i, j) is obtained by the following equation.
[0023]
f (i, j) = | -I (i-1, j) + 2 * I (i, j) -I (i + 1, j) |
Further, when the value of the in-focus differential parameter is defined as an evaluation function, the conditional expression that the evaluation function value of f (i, j) is larger than k (k: threshold value for in-focus) is satisfied. Filter processing (for example, binarization, weighted averaging, etc.) is performed for only those that satisfy the conditional expression (f (i, j)> k), and thinning is performed in S3 using the evaluation function values as evaluation function values. I have to.
[0024]
In this embodiment, the thinning data g (j) is calculated from the following equation based on the evaluation function, and the cross-sectional coordinates on the depth of field (depth of field) are calculated. 11 and the position at which the measurement object 10 intersects).
[0025]
g (j) = Σ (i * f (i, j)) / Σf (i, j)
In this case, an evaluation function value can be obtained for the captured image by another method described below. For example, assuming that the brightness is I (i, j) with respect to the camera coordinates (i, j) in the camera coordinate system, the evaluation function value f (i, j) is a Prussian filter, a Sobel filter, a Roberts filter, or the like. Can be used. In the case of the Prussian filter, the evaluation function value f (i, j) can be obtained from the following equation.
[0026]
f (i, j) = | -I (i, j-1) -I (i-1, j) + 4 * I (i, j) -I (i + 1, j) -I (i, j + 1) |
In the case of a Sobel filter, the evaluation function value f (i, j) can be obtained from the following equation.
[0027]
fx (i, j) =-I (i-1, j-1) + I (i + 1, j-1)
-2 * I (i-1, j) + 2 * I (i + 1, j)
−I (i−1, j + 1) + I (i + 1, j + 1)
fy (i, j) =-I (i-1, j-1) -2 * I (i, j-1) -I (i + 1, j-1)
+ I (i-1, j + 1) + 2 * I (i, j + 1) + I (i + 1, j + 1)
f (i, j) = | fx (i, j) | + | fy (i, j) |
Further, in the case of the Roberts filter, the evaluation function value f (i, j) can be obtained from the following equation.
[0028]
f (i, j) = | f (i + 1, j + 1) -f (i, j) | + | f (i + 1, j) -f (i, j + 1) |
When converting the data of the camera coordinate system imaged by the imaging device 5 to the object coordinate system, the following calculation method is used. That is, as shown in FIG. 3, the principal point position of the lens 8 is a point O, an intersection point between the measurement object 10 and the depth of field 11 is a point Q, and a perpendicular line lowered from the point Q to the Z axis and the Z axis Is the point P, the distance from the origin of the object coordinates to the point Q is Δ, the image of the point Q picked up on the image sensor 3 is the point Q ′, the perpendicular drawn from the point Q ′ to the Z axis, and the Z axis. Is defined as a point P ′, and the distance of the point Q ′ from the intersection of the imaging device 5 and the Z axis is defined as δ, from the relationship ΔOPQ∝ΔOP′Q ′,
Δsinθ / (d0−Δcosθ) = δsinφ / (di + δcosφ)
Δ (di · sin θ + sin θ · δcosφ + δsinφ · cosθ) = d0 · δsinφ
By using the addition theorem of the trigonometric function, the following equation is established.
[0029]
Δ (di · sin θ + δ sin (θ + φ)) = d0 · δ sin φ
Δ = δ · D1 / (D2 + δ)
(However, D1 = d0 · sinφ / sin (θ + φ) and D2 = di · sinθ / sin (θ + φ)). Therefore, for the point Q (x, y, z), z = Δcosθ. Here, if the moving stage 6 is moved by x0 in the −X direction from the measurement start point, the coordinates of x can be obtained from the relational expression x = Δsinθ + x0. As for the Y coordinate, as shown in FIG. 4, the intersection between the perpendicular line drawn from the point Q to the XZ plane and the XZ plane is defined as the point R, and the intersection point between the perpendicular line drawn from the point Q to the XZ plane and the XZ plane is calculated. If the distance from the point Q to the point R is further defined as Λ and the distance λ from the point Q to the point R ′ is defined as the point R, the following equation is established from the relationship of ΔOQR∝ΔOQR ′.
[0030]
Λ / (d0−Δcosθ) = λ / (di + δcosφ)
Λ = λ · (d0−Δcosθ) / (di + δcosφ)
Therefore, y = Λ = λ · (d0−Δcos θ) / (di + δcos φ)
Assuming that the camera coordinate of the point Q is I (δ, λ), the object equation (x, y, z) of the point Q in the three-dimensional object coordinate system is represented by S4 Is required. The three-dimensional coordinates obtained in this manner are stored in the three-dimensional coordinate table for data storage in S5. Such processing is performed for all the scanning lines that form an image on the imaging device 5, and the cross-sectional coordinates are stored and stored in the three-dimensional coordinate table in S5. In this case, one image data corresponding to the position of the moving stage 6 is stored in the three-dimensional coordinate table. Then, in the next S6, the moving stage 6 is moved by a predetermined amount (in a few cm or linear state) in the X direction. After that, the section coordinates are calculated again for all the scanning lines from the positional relationship between the depth-of-field surface 11 and the measurement object 10 and stored in the three-dimensional coordinate table.
[0031]
Then, in S7, it is determined whether the measurement of the measurement object 10 has been completed. The end of the measurement on the measuring object 10 means a position where the moving stage 6 is sequentially moved in the X direction with respect to the depth of field 11, and the object to be measured 10 passes through the depth of field 11 and is no longer hit. By moving the moving stage 6 up to this point, it is determined whether the measurement has been completed. Here, when the measurement of the measurement object 10 is not completely completed, the process returns to S1, and the above-described processing from S1 is repeated. However, when the measurement is completed, the three-dimensional measurement of the measurement object 10 is performed. finish.
[0032]
Image data obtained through such a process is stored in a predetermined storage area of the three-dimensional coordinate table. In this case, as shown in FIG. 6, each coordinate point of the camera coordinates stores therein a differentiated evaluation function value f (i, j) of the video signal, and based on this value, the depth of field is determined. Above, the evaluation function value g (x, y, z) is stored in the three-dimensional coordinate table. At this time, the evaluation function value f (i, j) becomes a large value (for example, 7, 8 in FIG. 6) at the position where the image of the measurement target 10 is in focus of the imaging device 5. In the present embodiment, the two-dimensional depth-of-field plane 11 is moved in the X direction by moving the moving stage 6 in the X direction while fixing the position of the imaging device 5 while the imaging device 5 is tilted. The same state as the state made is created, whereby the cross-sectional shape of the measurement object 10 at which the depth-of-field plane 11 and the measurement object 10 intersect can be represented by thin lines in object coordinates as shown in FIG. Thus, the cross-sectional shape of the measurement object 10 can be measured by thinning in the object coordinate system.
[0033]
That is, a plurality of evaluation function values are sequentially stored in a predetermined area of a three-dimensional coordinate table in the x-direction in two-dimensional camera coordinates, and g (x, y, z) is stored as three-dimensional data. The evaluation function values are stored three-dimensionally as shown in FIG. 8, and the shape of the measuring object 10 can be measured three-dimensionally.
[0034]
In this case, the three-dimensional point in the z direction is the maximum value of the evaluation function g (x, y, z) at the X and Y coordinates, the weighted average of z, g (x, y, z), z , G (x, y, z) or a Gaussian function.
[0035]
According to this method, there is no blind spot area in the measurement of the measurement object 10. Then, since the imaging device 5 does not move during shape measurement, the reliability of shape measurement is improved. In addition, since measurement can be performed while the measurement target 10 is moving, the measurement of the measurement target 10 can be performed at high speed.
[0036]
Next, a modified example of the image processing performed by the image processing device 9 shown in FIG. 5 will be described. FIG. 5 shows a method of obtaining the depth-of-field surface of the measurement object 10 by thinning it in the object coordinate system. However, in the processing shown in FIG. 7, the measurement object 10 is measured three-dimensionally in the object coordinate system. Is what you do. Thus, FIG. 7 will be described.
[0037]
In the image processing in FIG. 7, the processing performed at the beginning of S11 and S12 is the same as S1 and S2 shown in FIG. Thereafter, in step S13, for the differential value (evaluation function value) of the imaged video signal having a fixed threshold value or more, the sectional coordinates in the object coordinate system are calculated by the same method as in step S4 shown in FIG. When the sectional coordinates have been calculated, the sectional coordinates calculated in S14 are stored in a three-dimensional coordinate table (evaluation function table) as shown in FIG.
[0038]
In this case, one image data corresponding to the position of the moving stage 6 is stored in the three-dimensional coordinate table. Then, in the next S15, the moving stage 6 is moved in the X-axis direction by a predetermined amount (every several cm or continuously). After that, the section coordinates are calculated again for all the scanning lines from the positional relationship between the depth-of-field surface 11 and the measurement object 10 and stored in the three-dimensional coordinate table.
[0039]
Then, in S16, it is determined whether the measurement of the measurement target object 10 has been completed. The end of the measurement on the measurement target 10 means that the moving stage 6 is sequentially moved in the X direction with respect to the depth of field 11, and the measurement target 10 passes through the depth of field 11 and is no longer hit. By moving the moving stage 6 up to this point, it is determined whether the measurement has been completed. Here, if the measurement of the measurement object 10 has not been completely completed, the process returns to S1, and the above-described process from S1 is repeated. However, if the measurement has been completed, the process proceeds to S17 and subsequent processes.
[0040]
In S17, as shown in FIG. 9, one h (x, y, z) is calculated from a plurality of evaluation function values g (x, y, z) for the same XY coordinates, and in S18, three-dimensional coordinates are calculated. To save. Thereafter, in S19, it is determined whether or not the height calculation has been completed for all the XY coordinates. If the height of the measuring object 10 has not been completely calculated yet, the process returns to S17, and the processing from S17 to S19 is performed. The process is repeated, but when the height calculation is completed for all the XY coordinates, the three-dimensional measurement of the measurement object 10 is completed.
[0041]
Image data obtained through such a process is stored in a predetermined storage area of the three-dimensional coordinate table. In this case, for each coordinate point of the camera coordinates, the sectional coordinates are calculated from the differentiated evaluation function value f (i, j) of the video signal, and the evaluation function value g (x, y, z) is stored in the three-dimensional coordinate table. At this time, the evaluation function value f (i, j) becomes a large value at the position where the image of the measurement object 10 is in focus at the imaging device 5. In the present embodiment, the two-dimensional depth-of-field plane 11 is moved in the X direction by moving the moving stage 6 in the X direction while fixing the position of the imaging device 5 while the imaging device 5 is tilted. It is possible to create the same state as the state that has been made, and to express the cross-sectional shape of the measurement object 10 where the depth-of-field plane 11 and the measurement object 10 intersect in a three-dimensional numerical form as shown in FIG. Thus, the entire shape of the measuring object 10 can be measured by digitizing it in the object coordinate system.
[0042]
If the three-dimensional measurement is performed by the above-described present apparatus 1 in the processing shown in FIG. 7, the three-dimensional measurement can be performed by moving the moving stage 6 in the X direction while the position of the imaging device 5 is fixed. In this case, there is no blind spot area in the shape measurement of the measurement target 10. Further, according to this measuring method, since the imaging device 5 does not move, the imaging device 5 can be prevented from being blurred at the time of imaging, and the reliability in measurement is improved. In addition, even when the measurement object 10 is flowing in a line such as a factory, for example, a three-dimensional shape can be measured at high speed.
[0043]
【The invention's effect】
According to the present invention, an image of an object to be measured is taken in an oblique direction while the depth of field is inclined, and the moving means is moved in one direction, so that the image coordinates ( Conversion from camera coordinates) to actual object coordinates can be performed, and three-dimensional measurement of the measurement object can be performed. In this case, even if the measurement target is larger than the imaging range of the imaging unit, the imaging unit does not have to be moved with respect to the measurement target, so that the structure of the three-dimensional measurement apparatus can be simplified as compared with the related art.
[0044]
In this case, the image processing unit differentiates the video signal captured by the imaging unit by differentiating the video signal captured by the imaging unit and calculating the cross-sectional coordinates of the measurement target based on the differentiated video signal. Thus, a change in the brightness of the video signal can be detected. Then, based on the brightness of the differentiated video signal, the cross-sectional coordinates of the measurement object on the depth of field can be obtained.
[0045]
In addition, the calculation of the cross-sectional coordinates is performed by performing weighted averaging of the differentiated video signal, and by performing weighted averaging, the noise of the differentiated video signal can be removed. If thinning is performed based on a video signal, reliability in shape measurement can be improved.
[0046]
Further, the image processing means three-dimensionally converts and stores the cross-sectional coordinates, and by moving the moving means by the control means, sequentially stores the three-dimensional cross-sectional coordinates. Since the height direction of the object to be measured can be determined in three dimensions, the shape of the object to be measured can be measured in three dimensions.
[Brief description of the drawings]
FIG. 1 is a configuration diagram of a three-dimensional measurement apparatus according to an embodiment of the present invention.
FIG. 2 is an explanatory diagram illustrating an image of a measurement target imaged by an image sensor illustrated in FIG. 1;
FIG. 3 is an explanatory diagram geometrically showing a positional relationship between the image sensor shown in FIG. 1 and an object to be measured.
FIG. 4 is a diagram illustrating a relationship between the image sensor and the measurement target illustrated in FIG. 1 geometrically, and is an explanatory diagram when converting from two-dimensional coordinates to three-dimensional coordinates.
5 is a flowchart for obtaining a cross-sectional shape of a measurement object of the three-dimensional measuring apparatus shown in FIG.
FIG. 6 is an explanatory diagram showing values of an evaluation function at camera coordinates captured by the image sensor illustrated in FIG. 1;
7 is a flowchart for obtaining a three-dimensional shape of a measurement target of the three-dimensional measuring device shown in FIG.
FIG. 8 is an explanatory diagram showing values of an evaluation function on a depth of field.
FIG. 9 is an explanatory diagram showing values of an evaluation function in three dimensions.
FIG. 10 is a configuration diagram of a conventional three-dimensional measuring device.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 3D measuring device 2 Base 3 Support 5 Imaging device (imaging means)
6. Moving stage (moving means)
9. Image processing device (image processing means)
10 Measurement object 11 Depth of field

Claims (4)

A base,
A moving means movable in one direction with respect to the base;
An imaging unit that is installed at a predetermined position with respect to the base and captures an image of a measurement target placed on the moving member,
In a three-dimensional measurement device comprising: a control unit that moves the moving unit,
The imaging unit is provided in a state where a depth-of-field plane is inclined with respect to the measurement target, and moves the movement unit by the control unit to three-dimensionally measure the measurement target. A three-dimensional measuring device comprising image processing means. The three-dimensional image processing apparatus according to claim 1, wherein the image processing unit differentiates a video signal captured by the imaging unit, and calculates a cross-sectional coordinate of the measurement target based on the differentiated video signal. Measuring device. The three-dimensional measurement apparatus according to claim 2, wherein the calculation of the cross-sectional coordinates is performed by thinning the weighted average of the differentiated video signal. 2. The image processing unit according to claim 1, wherein the cross-sectional coordinates are three-dimensionally converted and stored, and the control unit moves the moving unit to sequentially store the three-dimensional cross-sectional coordinates. The three-dimensional measuring device according to claim 3.

Images