JP2795790B2 - Sensor coordinate correction method for three-dimensional measuring device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for correcting sensor coordinates of an optical three-dimensional measuring device for non-contact measurement of a three-dimensional position of a three-dimensional surface.

[0002]

2. Description of the Related Art Conventionally, an optical three-dimensional measuring apparatus of this kind captures an image of a slit light illuminated portion of a three-dimensional surface by two two-dimensional image sensors such as CCDs, and calculates the slit light illuminated by an operation based on the image output. Measure the three-dimensional position of the part.

Next, an example of a conventional apparatus will be described with reference to FIGS. 2 to 6 corresponding to an embodiment of the present invention. FIG. 2 shows a measuring mechanism of the apparatus.
Is a base, 2 is a horizontal support mounted on the base 1, 3 is a three-dimensional object mounted on the support 2, 4 is a column for supporting a sensor which is erected at one corner of the base 1, 5. Is the support 4 via the fixing member 6
A non-contact measuring device attached and fixed to the upper end of the light emitting device; 7, a light projecting means of the measuring device 5; and 8a and 8b, two two-dimensional image sensors provided on the left and right sides of the light projecting device 7. , A CCD type area image sensor and the like.

The light projecting means 7 and both sensors 8a and 8b of the measuring instrument 5 are constructed as shown in FIG.
9 is a light source such as a xenon lamp with a linear slit, 10
Is a rotatable reflecting mirror forming a light projecting means 7 together with the light source 9, and irradiates a linear slit light emitted from the light source 9 at right angles to the support base 2 to the surface of the three-dimensional body 3 so as to be movable in the left-right direction. . Reference numerals 11a and 11b denote imaging units of the imaging sensors 8a and 8b, which are formed by arranging CCDs in a two-dimensional matrix of M rows and N columns. Reference numerals 12a and 12b denote condensing lenses, which form reflected light from the surfaces of the solids 3 of the sensors 8a and 8b on the imaging units 11a and 11b.

At the time of measurement, the three-dimensional 3
Then, vertical slit light sufficiently longer than the height of the solid 3 is irradiated. Also, the sensors 8a and 8b are set to have an imaging field of view such that the slit light illuminated portion of the solid 3 is photographed in an overlapping manner, and based on this setting, as shown in FIGS. , 8b, the slit light images Sa, Sb in the vertical direction of the screen (Z-axis direction) corresponding to, for example, the vertical direction of the solid 3 are formed on the imaging surfaces Fa, Fb of the imaging units 11a, 11b, and the imaging surfaces Fa, Fb , The portions of the slit light images Sa and Sb have high brightness.

Then, the left and right directions (X
(Axial direction) corresponds to the horizontal direction of the three-dimensional body 3, and the imaging output of one scanning line of each of the sensors 8 a and 8 b is formed by the light receiving output of each row of pixels of the imaging units 11 a and 11 b parallel to the X-axis direction. . Note that A ₁ ,..., A _m , A _{m + 2} ,.
_n represents the first to N scanning lines of the image pickup unit 11a, B ₁ in FIG. _{(b), ..., B m} , B m + 1, B m + 2, ... B n is the first imaging unit 11b To Nth scanning line.

The imaging outputs of the respective scanning lines of the imaging sections 11a and 11b are sequentially read out at the same timing, and are read out as shown in FIG.
Is supplied to the computer 13. This computer 13 is shown in FIG.
In the figure, reference numeral 14 denotes a clock circuit for generating a clock signal, 15a and 15b denote two signal processing circuits, and an imaging unit 11 of the sensors 8a and 8b.
a, 11b, an image pickup output of each scanning line sequentially taken in, is taken in synchronism with the clock signal, binarized at a predetermined slice level, and a digital image signal in which only the slit light images Sa, Sb become high level. To form

[0008] 16a and 16b are two counters,
In synchronization with the clock signal, distances on the imaging planes Fa and Fb from the left end reference point of each scanning line of the imaging units 11a and 11b to the slit light images Sa and Sb are counted, and distance data in the X-axis direction of each scanning line is counted. Is output. An arithmetic circuit 17 is based on distance data in the X-axis direction of each scanning line output from the counters 16a and 16b, and distance data in the Z-axis direction of each scanning line obtained from the scanning line number and the width of the scanning line. Imaging surface F
The four arithmetic operations described below are executed based on the two-dimensional coordinate values of the slit light at a and Fb and the setting of the imaging conditions such as the imaging magnification and the installation interval of the imaging sensors 8a and 8b. Non-contact measurement is performed by calculating a three-dimensional position on a scanning line.

Reference numeral 18 denotes a storage unit for storing the coordinate values of the three-dimensional position calculated by the arithmetic circuit 17, reference numeral 19 denotes a display condition setting unit, and reference numeral 20 denotes a recognition circuit. The dimensions, surface condition, shape, and the like of the solid 3 are identified from each coordinate value, and the coordinate values of the storage unit 18 and the display signal of the identification result are output to the display means 21 in FIG. 5 and displayed on the screen.

Next, the measurement operation of the operation circuit 17 will be described. This measurement calculation is performed, for example, in the same manner as the calculation described in the specification and drawings attached to the application filed by the present applicant (Japanese Patent Application Nos. 60-4408 and 4409).

That is, when the imaging fields of view of the two sensors 8a and 8b are completely equal and the left end of the imaging field coincides with the Z axis, as shown in FIG. The light of (x, y, z) is transmitted to both sensors 8a and 8b.
Center points P (a, ₀ , z), Q of the lenses 12a, 12b
When an image is formed through (b, ₀ , z), a line segment connecting each image forming point and the center point P or Q forms an arbitrary point c on the Y axis perpendicular to the X axis and the Z axis. XZ plane and the point U (d,
If they intersect at c, z) and V (e, c, z), the point G is determined as the intersection of the line segment passing through the points P and U and the line segment passing through the points Q and V.

Then, based on the coordinate values of the points P, Q, U and V, the X and Y axis components x and y of the point G are obtained from the following equations (1) and (2).

[0013]

(Equation 1)

[0014]

(Equation 2)

In the two equations, ba is the distance L between the image pickup units 11a and 11b of the sensors 8a and 8b, and d and e are the image pickup surfaces determined by the magnification of the lenses 12a and 12b and the mounting positions of the sensors 8a and 8b. This is the position of the point G in the X-axis direction at Fa and Fb. Then, d, e are determined as the distance data from the reference point d ₀ of the left end of the imaging surface Fa, Fb. Also, a,
b and c are the mounting positions of the sensors 8a and 8b, the lenses 12a and
This is a constant set by the magnification of 12b or the like.

If the constants a and c in the X and Y axis directions based on the magnification of the lenses 12a and 12b, the positions of the sensors 8a and 8b, and the like are Kx and Ky, the X and Y axis components (coordinate values) of the point G will be described. x and y are obtained from the following equations (3) and (4).

[0017]

(Equation 3)

[0018]

(Equation 4)

Further, the Z-axis component (coordinate value) z of the point G is
The scanning line number r of the point G, the number of scanning lines, the width, and the lens 1
Based on the coefficient Kz determined by the magnification of 1a and 11b,
It is obtained from the following equation (5).

[0020]

(Equation 5)

Then, Kx, Ky,
L and Kz in the equation (5) are constants, d and e are obtained as distance data in the X-axis direction obtained by the counters 16a and 16b, and r is obtained by counting clock signals. Kx, Ky, Kz, position information of the slit light irradiation position S on the imaging surfaces Fa and Fb formed by the distance data of the counters 16a and 16b and the count data of the clock signal, and the setting data of the imaging conditions.
Based on the L data, the four arithmetic operations of the equations (3) to (5) are executed to calculate the three-dimensional position (x, y, z) of the point G.

Further, the calculation of the three-dimensional position is performed on each point of the slit light irradiation part of each scanning line, and the three-dimensional position of the slit light irradiation part of the solid 3 is calculated and measured. The rotation of the reflecting mirror 10 of the light projecting means 7 is controlled by a control means (not shown), and the rotation of the reflecting mirror 10 causes the irradiation position of the slit light of the light projecting means 7 to be X within the overlapping visual field of the two sensors 8a and 8b. Move in the axial direction.

Then, based on the movement of the irradiation position, the arithmetic circuit 17 calculates and measures the three-dimensional position of each point on the surface of the solid 3 in the overlapping visual field of the sensors 8a and 8b. The coordinate values of the three-dimensional position of each point on the three-dimensional surface obtained by this measurement are supplied to a recognition circuit 20 via a storage unit 18, and the circuit 20 supplies each point based on the setting conditions of a setting unit 19. , The surface state, the shape, etc. of the solid 3 are identified from the coordinate values of the three-dimensional position of the three-dimensional position.
And displayed on the screen.

When the measurement is performed on the entire surface of the three-dimensional object 3, for example, Japanese Patent Application Nos. 60-4408 and 440
As described in the specification and the drawings attached to the application for the application No. 9, a plurality of columns corresponding to the columns 4 are erected around the base 1, and each column has the same non- The device is formed with a contact measuring device.

There is also an apparatus in which the slit light is formed on the support 2 as horizontal (parallel) linear light. The optical three-dimensional measuring apparatus can change the measuring position by moving the irradiation position of the slit light without moving the measuring device 5, and can perform measurement by simple four arithmetic operations. There is an advantage that measurement can be performed in a shorter time than the apparatus of the method.

Further, since the slit light irradiation portion of the solid 3 is imaged from two directions by the pair of sensors 8a and 8b, accurate measurement can be performed even when the solid 3 is small and when the surface of the solid 3 has irregularities.

Further, since the measurement is carried out in a non-contact manner, even if the solid 3 is flexible and easily deformed, such as rubber, it can be easily measured. In addition, since the linear slit light is used. Alternatively, weak light having a low energy density may be used, and no distortion occurs in the three-dimensional body 3 due to illumination heat when illumination is used.

[0028]

SUMMARY OF THE INVENTION The conventional optical type 3
In the case of the dimension measuring device, the two-dimensional coordinate axes (X axis, Z axis) with respect to the irradiation light (slit light) on the imaging surfaces Fa, Fb are changed by the sensors 8a, 8b due to the mounting position error (deviation) of the imaging sensors 8a, 8b. If there is a deviation between them, the computations of Equations 3 to 5 are performed using position information including an error based on this deviation, and a measurement error occurs. In order to improve the measurement accuracy by minimizing the displacement of the coordinate axes between the sensors 8a and 8b, conventionally, complicated and advanced adjustment work of the mounting position of the sensors 8a and 8b is required, and the adjustment is mechanically performed. There is a limit associated with accuracy.

Therefore, there is a problem that the adjustment operation cannot be simplified and high-precision measurement cannot be performed. The present invention
An object of the present invention is to enable high-accuracy measurement without mechanical adjustment of the mounting position of an image sensor.

[0030]

In order to achieve the above object, a method of correcting a sensor coordinate of a three-dimensional measuring apparatus according to the present invention comprises the steps of: A two-dimensional image sensor is used to take an image, and a difference between two-dimensional coordinate axes of the imaging surfaces of both sensors is detected based on a difference between position information of the two spot lights on the imaging surfaces of the two sensors, based on a result of the detection. A correction value of the other coordinate axis of both sensors whose coordinate axes are coincident with each other with reference to one coordinate axis of both sensors is obtained. The displacement of the coordinate axis is corrected by calculation.

[0031]

In the sensor coordinate correcting method of the three-dimensional measuring apparatus according to the present invention having the above-described configuration, the difference between the position information of the two spot light spots for the marker points on the imaging surfaces of the two imaging sensors is obtained. From this, the displacement of the two-dimensional coordinate axis of the imaging surface corresponding to the mounting position error of both sensors is detected. Further, based on the result of this detection, a correction value of the other coordinate axis of both sensors is obtained based on one coordinate axis of both sensors.

Then, the position information of the other sensor at the time of measurement is corrected by this correction value, and the three-dimensional position of the three-dimensional surface is measured in a non-contact manner by correcting the displacement of the coordinate axes by calculation. Therefore, the complicated and sophisticated mechanical adjustment work of the mounting positions of the two sensors as in the related art can be omitted, and high-precision measurement without limit due to mechanical accuracy or the like can be performed.

[0033]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment will be described with reference to FIGS. First, in the measurement mechanism of FIGS. 2 and 3,
The difference from the prior art is that, when adjusting the mounting position error of the sensors 8a and 8b, the lever operation or the like causes the slit light between the light source 9 of the light projecting means 7 and the reflecting mirror 10 to pass through the optical path shown in FIG. The point is that the spot light forming filter 23 having the holes 22a and 22b formed therein or the spot light forming filter 25 having the through hole 24 shown in FIG.

Since the filter 25 also uses the through hole 24 as the through holes 22a and 22b of the filter 23, the position of the through hole 24 is changed, for example, by a manual operation at right angles to the optical path. The filter 2 is disposed between the light source 9 and the reflecting mirror 10.
When 3 or 25 is interposed, the slit light of the light source 9 is processed into spot light at two marker points in the vertical direction (Z-axis direction) of the solid 3 in the overlapping visual field of the imaging sensors 8a and 8b, and the two spot lights are used. Is irradiated on the solid 3.

Further, the two spot lights are used by the imaging sensor 8.
a and 8b, and the imaging output of each scanning line of both sensors 8a and 8b is supplied to the computer 13 of FIGS. At this time, based on the mounting position error between the sensors 8a and 8b, the positions of the spot lights on the imaging surfaces Fa and Fb of the sensors 8a and 8b are shifted as shown in FIGS. 1A and 1B, for example.

In FIG. 1, P1a and P2a indicate spot light images on the imaging surface Fa, and P1b and P2b indicate spot light images on the imaging surface Fb corresponding to the spot light images P1a and P2a. When the filter 25 is used, the spot light images P1a and P1b and the spot light images P2a and P2b are separately photographed and combined.

As is clear from the comparison between FIGS. 1A and 1B, the displacement of the X-axis and Z-axis based on the mounting position error of the sensors 8a and 8b causes, for example, a spot light image P
2a, P2b both the M scan line A _m, is located on the B _m, the spot light image P1a, P1b the first M-1 scan line A
_{m −1} , located at the (M−2) th scanning line B _m−2 . Next, the electronic computer 13 is different from the conventional computer in that the computer 13 has a sensor processing unit as well as a measurement processing unit similar to the conventional computer.

At the time of adjusting the mounting position error of the sensors 8a and 8b, for example, when a command for coordinate adjustment processing is given to the control terminal 26 of FIGS. 5 and 6 in conjunction with the movement of the filter 23 or 25, FIG. As shown in FIG.
7 is supplied. With this supply, the arithmetic circuit 17 shifts from the normal measurement processing to the sensor coordinate adjustment processing.

FIG. 1 obtained at the time of this adjustment
The imaging output of each scanning line of the imaging planes Fa and Fb in (a) and (b) is processed by the signal processing circuits 15a and 15b and the counters 16a and 16b of the electronic computer 13 in the same manner as in normal measurement. By this processing, the counter 16a outputs distance data in the X-axis direction of the spot light images P1a and P2a to the arithmetic circuit 17, and the counter 16b outputs the spot light images P1b and P2a.
The distance data in the X-axis direction of P2b is output to the arithmetic circuit 17.

The operation circuit 17 includes a counter 16a,
The position information of the spot light at the two marker points on the imaging surfaces Fa and Fb formed by the distance data in the X-axis direction 16b and the count data (scanning lines) of the clock signal of the clock circuit 14 is obtained, and the difference between the sensors is obtained. Is detected as a shift between the X and Z axes based on the mounting position error of the sensors 8a and 8b.
Further, based on the detection result, for example, with reference to the coordinate axis of the sensor 8a, the coordinate axis of the sensor 8b coincides with the coordinate axis of the sensor 8a, and the coordinate axis of the sensor 8b at which the position information of both spot lights on the imaging surfaces Fa and Fb becomes equal. Is obtained.

[0041] That is, FIGS. 1 (a) the same figure the same with reference FIG scan lines A ₁ to A _n of the sensor 8a (c) and (d)
X-axis to correct the scanning line B ₁ .about.B _n before correction indicated by the solid line of the sensor 8b the dashed scan lines B ₁ '~B _n' in the figure, obtaining a correction value of the Z-axis. Then, this correction value is stored in, for example, a non-volatile correction memory provided in the arithmetic circuit 17, and the sensor coordinate adjustment processing is ended to return to the normal measurement processing.

Thereafter, the filter 23 or 2 is operated by lever operation or the like.
When the specimen 5 is moved out of the optical path of the slit light and shifts to the normal measurement, the three-dimensional body 3 is irradiated with the same slit light as in the related art.
Then, the slit light is photographed from two directions by the sensors 8a and 8b, and the imaging output of each scanning line including the mounting position error of the two sensors 8a and 8b is output to the computer 13 as in the related art.
Supplied to

Further, the computer 13 includes the signal processing circuit 1
5a, 15b, counters 16a, 16b, and a clock circuit 14 determine the position information of the slit light on the imaging surfaces Fa, Fb such as distance data in the X direction of each scanning line in the same manner as in the prior art. To supply. At this time, the arithmetic circuit 17 calculates and corrects the position information of the slit light on the imaging surface Fb based on the stored two-dimensional correction value, and the sensors 8a and 8b of the position information of the slit light on the imaging surfaces Fa and Fb. The error based on the mounting position error is eliminated.

Then, based on the corrected position information of the slit light and the setting of the photographing conditions of the sensors 8a and 8b, the slit light on the surface of the three-dimensional object 3 is calculated from the above-mentioned equations (3) to (5) in the same manner as before. The three-dimensional position of the irradiated part is calculated and measured.
Further, based on the result of the measurement, the recognition circuit 20 identifies the surface state, shape, and the like of the solid 3, and the identification result is displayed on the display unit 21 on a screen.

In this case, since the measurement error based on the mounting position error of the sensors 8a and 8b is corrected by calculation, the complicated and high-level adjustment work of the mounting position of the sensors 8a and 8b as in the related art is omitted, and the mechanical accuracy is reduced. Without any adjustment limitations
High-precision measurement can be performed. In the above embodiment, the spot light at the two marker points is formed from the slit light by using the filter 23 or 25. However, the light source 9 of the light projecting means 7 is used as a spot light source, and the light source 9 and the reflecting mirror are used as shown in FIG. A movable convex cylindrical lens 27 for forming slit light is provided between the light source 9 and the reflecting mirror 10, and the lens 27 is moved out of the optical path between the light source 9 and the reflecting mirror 10 during adjustment. A spot light at the marker point may be formed. In this case, at the time of normal measurement, the spot light may be processed into slit light by interposing the lens 27 in the optical path between the light source 9 and the reflecting mirror 10.

When measuring the entire surface of the solid 3, for example, a plurality of columns corresponding to the columns 4 are erected around the base 1, and a non-contact measuring device similar to the measuring device 5 is provided on each column. May be provided to form the device.

[0047]

Since the present invention is configured as described above, the following effects can be obtained. Two spot light images P1a, P2a, P1b, P2 for marker points on the imaging surfaces Fa, Fb of the two two-dimensional imaging sensors 8a, 8b.
The displacement of the two-dimensional coordinate axes of the imaging surfaces Fa and Fb corresponding to the mounting position error of the two sensors 8a and 8b is detected from the difference between the sensors of the position information of the two sensors 8a and 8b.
A correction value for the other coordinate axis is determined with reference to one of the coordinate axes a and 8b, and the correction values for both sensors 8a and 8b during measurement are obtained.
The position information of the other of the sensors 8a and 8b is corrected in a non-contact manner by correcting the position information of the other of the sensors 8b and calculating the displacement of the coordinate axes by non-contact. High-precision measurement can be performed without an adjustment limit due to mechanical accuracy and the like, without performing a complicated adjustment operation.

[Brief description of the drawings]

FIGS. 1A to 1D are explanatory views of correction of a sensor coordinate correcting method of a three-dimensional measuring apparatus according to an embodiment of the present invention.

FIG. 2 is a perspective view of a measuring mechanism of the three-dimensional measuring apparatus according to one embodiment of the present invention.

FIG. 3 is a diagram illustrating the configuration of the non-contact measurement device of FIG. 2;

4 (a) and 4 (b) are explanatory diagrams of an imaging output at the time of measurement by the two two-dimensional imaging sensors of FIG. 3;

FIG. 5 is a block diagram of a circuit unit that processes the imaging output of FIG. 3;

FIG. 6 is a detailed block diagram of a part of FIG. 5;

FIG. 7 is an explanatory diagram of a measurement process of the arithmetic circuit in FIG. 6;

8 (a) and 8 (b) are front views of one example and another example of a spot light forming filter provided in the light projecting means of the non-contact measuring device of FIG. 2;

9 is a configuration explanatory view of another example of the light projecting means of the non-contact measurement device of FIG. 2;

[Explanation of symbols]

3 3D 8a, 8b Two-dimensional imaging sensor Fa, Fb Imaging surface P1a, P1b, P2a, P2b Spot light image for marker point

Continuation of the front page (56) References JP-A-4-184203 (JP, A) JP-A-5-28246 (JP, A) JP-A-2-243236 (JP, A) JP-A-6-207810 (JP) JP-A-4-118504 (JP, A) JP-A-63-217214 (JP, A) JP-A-5-26639 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB G01B 11/00-11/30

Claims (1)

(57) [Claims] 1. An irradiation light on a three-dimensional surface is photographed from two directions by two two-dimensional imaging sensors, and position information of the irradiation light on an imaging surface of each of the two sensors and setting of imaging conditions of the two sensors are used. In a three-dimensional measuring device for non-contact measurement of a three-dimensional position of a three-dimensional surface by a calculation based on the three-dimensional surface, a spot light for a marker point applied to two points of the three-dimensional surface is photographed by the two sensors, and the imaging surfaces of the two sensors A difference between two-dimensional coordinate axes of the imaging surfaces of the two sensors is detected from a difference between the sensors of the position information of the two spot lights, and based on a result of the detection, one coordinate axis of the two sensors is used as a reference. A correction value of the other coordinate axis of the two sensors, in which the coordinate axes of the two sensors coincide, is calculated, and the positional information of the other sensor at the time of measurement is corrected by the correction value to calculate the displacement of the coordinate axes of the two sensors. Supplement Sensor coordinate correction method of the three-dimensional measuring apparatus, characterized by.