JPH0763527A - Form measuring device - Google Patents

Abstract

PURPOSE:To calculate the three-dimensional coordinate of a measuring surface from the slippage between respective stripes of stripe pattern images on a standard plane and the measuring surface and the position (x), (y) of the picture element of the measuring surface, and precisely perform a form measurement at high speed. CONSTITUTION:A standard plane is preliminarily set instead of a subject 9 to be measured, and a change-over switch 4 is connected to (a)-side to store 5 a stripe pattern image. The standard plane is vertical to the optical axis of a television camera 2, and the distances to respective lens centers of the camera 2 and a projecting device 1 are equally set. At a measurement, the subject 9 is placed instead of the standard plane, and the switch 4 is connected to (b)-side to photograph 2 the stripe pattern image. The stripe pattern 11a of the subject 9 and the stored 5 stripe pattern 1b of the standard plane show thick stripe images, and the two images are compared to each other by a slippage measuring circuit 6 to measure the x-directional slippage between the strips. The x, y-coordinate of the picture element of the pattern image of the subject 9 is also measured 7. A form arithmetic circuit 8 calculates and outputs the space coordinate of a measuring point by use of the slippage and the x,y-coordinate of the picture element.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for measuring the flatness of a three-dimensional shape of an industrial product such as a steel pipe or a shaped steel or a plate-like product such as a steel plate.

[0002]

2. Description of the Related Art The so-called light-section method is well known as a method for measuring a three-dimensional shape, but it is only possible to measure the shape of one cross section by one image pickup, and the calculation formula is complicated and it takes a long calculation time. There's a problem. Further, the unevenness amount of the object is much smaller than the distance from the light source or the TV camera to the object, but in the light-section method, the unevenness amount or shape of the There is also a problem that the measurement accuracy is not very good. As one of the improved methods, there is a method of projecting a stripe pattern consisting of many stripes. With this method, the shape of many cross sections can be measured with one image capture, but the calculation time is long and the measurement accuracy is low. The problem of remains. Japanese Patent Application No. 2-77606 discloses an apparatus capable of improving the measurement accuracy and shortening the calculation time, and this apparatus will be briefly described below with reference to FIG.

In FIG. 4A, a bright and dark pattern is projected obliquely on the reference plane 16 and the object 17 at an angle of θ with respect to the reference plane 16 by a combination of a collimator light source 13 for projecting a parallel light beam and a liquid crystal shutter 14. The projected and projected light and dark patterns are picked up by the television camera 15. The liquid crystal shutter 14 can change the bright and dark pattern to be projected by changing the voltage application pattern to the liquid crystal electrode, and can perform single slit light projection, multiple slit light projection, or coded band illumination. Has become. Coded strip illumination is an illumination method that can efficiently measure the shape of 2 ^N cross-sections by N times of imaging. IEEE, 1984 “RANGE-IMAGING SYSTEM FOR 3-DO
BJECT RECOGNITION by S. INOKUCHI, K. SATO and F. M
It has been published on "ATSUDA". When a plurality of slit lights are projected and the object 17 is not present and only the reference plane 16 is present, the image captured by the television camera 15 is as shown in FIG. 4B, and the object 17 is placed on the reference plane 16. The image at that time is as shown in FIG. 18 in FIG. 4C is an object 17
The shadow of the object formed by the projection being blocked by
By comparing the two images in FIG. 4C and counting the number of slits corresponding to the shadow 18 of the object,
The shadow length s of the object in (a) can be obtained. Since light is projected everywhere on the reference plane at an angle of θ, the height z of the object is calculated by the following equation (1).

[Equation 1] Can be calculated easily.

As described above, the apparatus disclosed in Japanese Patent Application No. 2-77606 uses the conventional optical cutting method in order to obtain the height of the object by comparing the reference plane and the object which are close to each other in distance. Compared with the above, the calculation formula is (1)
The calculation time is short because it is as simple as the formula. Furthermore, if the coded strip illumination is used, the shapes of a large number of cross sections can be efficiently obtained with a small number of times of imaging, so that the three-dimensional shape measurement can be performed in a short time in combination with the short calculation time.

[0005]

However, even when the coded strip illumination is used in the above apparatus, it is necessary to take a plurality of times of imaging, and it is not possible to measure a large number of cross sections by a single imaging. There are also problems that it is difficult to measure a large area because a parallel beam is used, and the lighting device is expensive because the bright and dark patterns are switched by the liquid crystal shutter. Furthermore, the number of slits corresponding to the shadow 18 of the object in FIG. 4C does not strictly correspond to the length s of the shadow of the object in FIG. Since the straight line connecting U corresponds to the length s'corresponding to the point V intersecting with the reference plane 16, there is a problem that the value obtained by the above equation (1) is not accurate. As described above, the conventional apparatus has a defect in any of the number of times of imaging, calculation time, measurement accuracy, measurement area, and apparatus price, and no apparatus can satisfy all of them.

The present invention has been made in order to solve the above problems, and can easily measure a large area, does not require expensive parts such as a liquid crystal shutter, and has a large number of cross-sections by one-time imaging. It is an object of the present invention to provide a three-dimensional shape measuring device which can measure the shape of the object, can be executed at high speed with a simple calculation formula, and has high measurement accuracy.

[0007]

In order to solve the above problems, the shape measuring apparatus of the present invention is a projection apparatus for projecting a fringe pattern characterized so that the order of fringes can be discriminated on a reference plane and a measurement plane. A television camera that captures the projected fringe pattern from a direction different from the projection direction; an image memory that stores the fringe pattern image of the reference plane captured by the television camera; and an image memory that stores the image. A displacement amount measuring circuit that compares the striped pattern image of the reference plane with the striped pattern image of the measurement surface captured by the television camera to measure the displacement amount between the stripes of the same order,
A shape calculation circuit that calculates the three-dimensional coordinates of the measurement surface from the displacement amount and the x and y positions of the pixels of the striped pattern image of the measurement surface is provided, and the calculation of the three-dimensional coordinates can be performed at high speed with a simple mathematical expression. In order to do so, the optical axis of the television camera is arranged to be perpendicular to the reference plane, and the lens center of the projection device and the lens center of the television camera are both arranged equidistant from the reference plane.

[0008]

The principle of measuring the shape by the apparatus of the present invention will be described below with reference to FIG. In FIG. 3, P is the lens center of the projection device, and Q is the lens center of the television camera, both of which are separated from the reference plane 10 by h. 12 is the light receiving surface of the television camera, and O'is its center. O'Q is the optical axis of the television camera, and OO 'is perpendicular to the reference plane 10 where O is the point where the extension line intersects the reference plane 10. The light receiving surface 12, PQ, and the reference plane 10 are parallel to each other. Reference plane 1 corresponding to both ends A'and B'of the light-receiving surface 12
The points on 0 are A and B, respectively. In the following, the coordinate system is based on O as the origin, the OA direction is the x direction, the OQ direction is the z direction, and the direction perpendicular to the paper surface is the y direction. To Further, the projected stripe pattern is assumed to be a pattern in which light and dark are repeated in the x direction. It is assumed that the existing striped light exits from P, reaches the point C on the reference plane 10, and forms an image at the point C ′ on the light receiving surface 12. When the measuring object 9 is present instead of the reference plane 10, the light beam PC illuminates D on the measuring object 9, and D is imaged on D ′ of the light receiving surface 12. Let E be the point where D'D or its extension intersects the reference plane 10. If the coordinates of D are (x, y, z), the following equation (2)

[Equation 2] From which z is

[Equation 3] And the following equation

[Equation 4] Is obtained. In the equation (4), AB is the length in the x direction of the range of the reference plane 10 captured by the television camera, so that it can be accurately grasped in advance, and h and QP can also be measured in advance. C'D '/ A'B' is x on the image
It is the ratio of the number of pixels of stripe deviation in the direction to the total number of pixels in the x direction on the light receiving surface. Since the total number of pixels in the x-direction on the light-receiving surface is known, the height z of point D can be calculated by measuring the amount of fringe shift in the x-direction in pixel units. For the x coordinate,

[Equation 5] Because

[Equation 6] Is obtained. Here, since O'D '/ A'B' is determined by the pixel position in the x direction at the point D ', equation (6) can be calculated from the stripe shift amount in the x direction and the x position of the pixel. In order to obtain the y-coordinate, the following formula (7) in which AB of the numerator is replaced by w and O′D ′ / A′B ′ is replaced by m / n in the formula (6) while keeping the denominator

[Equation 7] Can be calculated by Here, w is the length of the reference plane 10 in the y direction of the range imaged by the television camera, m is the y position on the image of the point being measured, and the scanning line measured from the center of the image. Is a number and n is the number of all scan lines. As can be seen from this equation, the y coordinate can be calculated from the amount of fringe shift in the x direction and the y position of the pixel.

As described above, if the x and y positions of the pixels of the stripe of the measurement object image and the amount of deviation in the x direction when this stripe is compared with the stripe of the reference plane image of the same order, It is possible to calculate the three-dimensional coordinates of an object. In this way, the shape of the cross section can be measured along the stripe, but since there are many stripes, it is possible to obtain shape data for multiple cross sections from one measurement object image. Formula (4), Formula (6),
The expression (7) is not an approximate expression but a strict calculation expression, but since all of them are simple arithmetic operations, they can be executed at extremely high speed. Alternatively, various deviation amounts and pixel x
The spatial coordinate calculation corresponding to the position and the y position is performed and the calculation result is tabulated, so that it can be executed at a higher speed. Further, this method is good in measurement accuracy because the reference plane and the measurement object are compared with each other in terms of distance, and measurement is performed with the reference plane as a reference. Further, as is apparent, since the projection device does not illuminate parallel light but spreads with distance, it is easy to measure a large area, and since it only projects a stripe pattern, it can be realized at low cost.

[0010]

FIG. 1 is a schematic view showing the apparatus configuration of an embodiment of the present invention. In the figure, reference numeral 1 is a projection device for projecting a stripe pattern, and a device such as a slide projector can be used. In this embodiment, one stripe is thick in order to be able to determine the order of the projected stripe,
11 is the thick stripe. Reference numeral 2 is a television camera for capturing the projected stripe pattern, and 3 is a television monitor. Prior to the measurement, a reference plane (not shown) is set in advance in place of the measuring object 9 and the stripe pattern image at that time is stored in the image memory 5. At this time, changeover switch 4
Is connected to a. The reference plane is perpendicular to the optical axis of the television camera 2 and is set so that the distance to the lens center of the television camera 2 and the lens center of the projection device 1 are equal. When performing the measurement, the measuring object 9 is placed instead of the reference plane, and the stripe pattern image at that time is captured by the television camera 2. At this time, the changeover switch is connected to the side b.

FIG. 2 (a) is a striped pattern image of the measuring object 9 at this time, and FIG. 2 (b) is a striped pattern image of the reference plane stored in the image memory 5, 11a and 11b.
Indicates a thick striped image. The two images are compared by the shift amount measuring circuit 6 and the shift amount of the stripes of the same order in the x direction is measured. At the same time, with respect to the pattern image of the measurement object, the x, y coordinates of the pixel are measured by the x, y position measuring circuit 7. In this embodiment, since the order of the stripes can be discriminated by the thick stripe 11, it is possible to definitely compare the stripes of the reference plane and the stripes of the measurement object with the same order. Reference numeral 8 denotes a shape calculation circuit, which uses the displacement amount measured by the displacement amount measuring circuit 6 and the x, y coordinates of the pixel measured by the x, y position measuring circuit 7 to obtain formulas (4) and (6). The spatial coordinates of the measurement point are calculated and output according to the equation (7).

In the present embodiment, one thick stripe is used in order to distinguish the order of stripes, but the method is not necessarily limited to this. For example, the stripe pattern may be colored and a color camera may be used.

[0013]

As is clear from the above description, according to the apparatus of the present invention, it is possible to obtain the shapes of a large number of cross sections by one image pickup, and the calculation can be executed at a high speed. It will be possible. Further, since the reference plane and the object to be measured, which are close to each other in terms of distance, are compared with each other and the calculation formula is strict, the measurement accuracy is high. Furthermore, since expensive parts such as a liquid crystal shutter are not required and parallel light is not used, it is possible to easily measure a large area. As described above, the device of the present invention can be applied to applications where the conventional device cannot be applied in terms of measurement accuracy, measurement time, device price, object size, and the like, and the effect is extremely great.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing a device configuration of an embodiment of the present invention.

FIG. 2 is a diagram showing a striped pattern image on a measurement object and a striped pattern image on a reference plane, which are imaged by a television camera with the configuration of FIG.

FIG. 3 is an explanatory diagram for explaining the measurement principle of the present invention.

FIG. 4 is an explanatory diagram showing an example of a conventional measuring device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Projection device 2 Television camera 3 Television monitor 4 Changeover switch 5 Image memory 6 Deviation amount measuring circuit 7 x, y position measuring circuit 8 Shape calculation circuit 9 Measurement object 10 Reference plane 11 Thick stripes 11a, 11b Thick stripe image 12 Light reception Surface 13 Collimator light source 14 Liquid crystal shutter 15 Television camera 16 Reference plane 17 Object 18 Shadow of object

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] June 15, 1994

[Procedure Amendment 1]

[Document name to be amended] Statement

[Correction target item name] 0008

[Correction method] Change

[Correction content]

[0008]

The principle of measuring the shape by the apparatus of the present invention will be described below with reference to FIG. In FIG. 3, P is the lens center of the projection device, and Q is the lens center of the television camera, both of which are separated from the reference plane 10 by h. 12 is the light receiving surface of the television camera, and O'is its center. O'Q is the optical axis of the television camera, and OO 'is perpendicular to the reference plane 10 where O is the point where the extension line intersects the reference plane 10. The light receiving surface 12, PQ, and the reference plane 10 are parallel to each other. Reference plane 1 corresponding to both ends A'and B'of the light-receiving surface 12
The points on 0 are A and B, respectively. In the following, the coordinate system is based on O as the origin, the OA direction is the x direction, the OQ direction is the z direction, and the direction perpendicular to the paper surface is the y direction. To Further, the projected stripe pattern is assumed to be a pattern in which light and dark are repeated in the x direction. It is assumed that the existing striped light exits from P, reaches the point C on the reference plane 10, and forms an image at the point C ′ on the light receiving surface 12. When the measuring object 9 is present instead of the reference plane 10, the light beam PC illuminates D on the measuring object 9, and D is imaged on D ′ of the light receiving surface 12. Let E be the point where D'D or its extension intersects the reference plane 10. If the coordinates of D are (x, y, z), the following equation (2)

[Equation 2] From which z is

[Equation 3] And the following equation

[Equation 4] Is obtained. In the equation (4), AB is the length in the x direction of the range of the reference plane 10 captured by the television camera, so that it can be accurately grasped in advance, and h and QP can also be measured in advance. C'D '/ A'B' is x on the image
It is the ratio of the number of pixels of stripe deviation in the direction to the total number of pixels in the x direction on the light receiving surface. Since the total number of pixels in the x-direction on the light-receiving surface is known, the height z of point D can be calculated by measuring the amount of fringe shift in the x-direction in pixel units. For the x coordinate,

[Equation 5] Because

[Equation 6] Is obtained. Here, since O'D '/ A'B' is determined by the pixel position in the x direction at the point D ', equation (6) can be calculated from the stripe shift amount in the x direction and the x position of the pixel. In order to obtain the y-coordinate, the following formula (7) in which AB of the numerator is replaced by w and O′D ′ / A′B ′ is replaced by m / n in the formula (6) while keeping the denominator

[Equation 7] Can be calculated by Here, w is the length of the reference plane 10 in the y direction of the range imaged by the television camera, m is the y position on the image of the point being measured, and the scanning line measured from the center of the image. Is a number and n is the number of all scan lines. As can be seen from this equation, the y coordinate can be calculated from the amount of fringe shift in the x direction and the y position of the pixel.

Claims (1)

[Claims] 1. A projection device for projecting a striped pattern characterized so that the order of the striped lines can be discriminated on a reference plane and a measurement surface, and a television camera for imaging the projected striped pattern from a direction different from the projection direction. And an image memory for storing the stripe pattern image of the reference plane imaged by the television camera, the stripe pattern image of the reference plane stored in the image memory, and the stripes on the measurement surface imaged by the television camera. A three-dimensional coordinate of the measurement surface is calculated from a deviation amount measurement circuit that compares the pattern image and measures the deviation amount between stripes of the same order, and the deviation amount and the x and y positions of the pixels of the stripe pattern image on the measurement surface. The television camera has an optical axis perpendicular to the reference plane, and the lens center of the projection device and the lens center of the television camera are both equidistant from the reference plane. A shape measuring device characterized in that the shape measuring device is arranged as follows.