JPH045926B2 - - Google Patents

Description

[Detailed Description of the Invention] [Table of Contents] Overview...(3) Field of Industrial Application...(4) Prior Art...(4) Problems to be Solved by the Invention...(5) Problems Means for solving the problem...(6) Effect...(7) Basic principle of the invention...(8) Example () Configuration of measuring device...(15) () Measurement processing...(17) () Other embodiments ... (27) Effects of the invention ... (30) [Summary] In order to measure the spatial arrangement of objects using so-called binocular stereopsis, images obtained by two cameras are used. requires complex processing to match. The present invention controls multi-slit light generated from one multi-slit light source, uses one of the slit lights as a reference slit light, and irradiates an object with the multi-slit light and this reference slit light to determine the spatial arrangement of the object. To provide a three-dimensional measurement method of an object that can be measured in a short time with one camera.

[Industrial application field]

The present invention relates to a method for measuring the spatial arrangement of objects necessary for controlling industrial robots, etc.
More specifically, the present invention relates to a three-dimensional measurement method for an object in which the spatial arrangement of the object is measured by irradiating the object with multi-slit light and obtaining an image of the object.

[Conventional technology]

Traditionally, as a method for measuring objects three-dimensionally,
The so-called binocular stereopsis method was introduced. In this method, as shown in FIG. 11, two cameras 1 and 2 and a multi-slit light source 5 are fixedly arranged near a target object 4, and the multi-slit light source 5 is turned on to emit multi-slit light 3 onto the object. irradiate the surface of Next, a slit light projection image 9 formed by the multi-slit light 3 on the surface of the object 4 is photographed by two cameras 1 and 2. Corresponding points (in FIG. 9,
This is a method of three-dimensionally measuring an object based on the principle of triangulation by finding a point B) on the image of camera 2 relative to a point of interest A on the image of camera 1. Incidentally, images taken by cameras 1 and 2 are shown in FIG.

[Problem that the invention seeks to solve]

By the way, with this conventional method, it is difficult to make correspondences between the plurality of slit light images taken by camera 1 and the plurality of slit light images taken by camera 2 (the process of finding the corresponding point B). However, it has the disadvantage that it is complicated and takes a long processing time. Therefore, the applicant of the present invention has proposed a method for three-dimensional measurement of an object without the complicated processing of associating images taken with slit light taken by two cameras in the so-called binocular stereoscopic viewing method. In this method, two light sources, a reference slit light source and a multi-slit light source, are used, and the reference slit light source is arranged so that the slit light surface of the reference slit light coincides with one slit light surface of the multi-slit light, and the multi-slit light and the reference slit light are proposed a method for three-dimensional measurement of objects by time-sharing irradiation of
No. 107675). The present invention further improves this. That is, it is possible to obtain multi-slit light and reference slit light with one light source, and to realize three-dimensional measurement of an object with a compact device.

[Means for solving problems]

The present invention irradiates an object surface with multi-slit light that is generated by one multi-slit light source and has a plurality of slit light surfaces whose arrangement in a predetermined coordinate system is set in advance, and controls the multi-slit light. One of the slit lights is used as a reference slit light to illuminate the surface of the same object, and a plurality of slit light photographed images corresponding to the slit light projected image by the multi-slit light on the object surface and a slit by the reference slit light on the object surface are obtained. A reference slit light photographed image corresponding to the light photographed image is obtained on a predetermined image plane, a slit light photographed image corresponding to the reference slit light photographed image is identified from the plurality of slit light projected images, and the plurality of slit light photographed images are obtained. Assuming that there is no interchange between images, each slit light photographed image is associated with each slit light surface of the multi-slit light based on the relative positional relationship between this specified slit light photographed image and other slit light photographed images. , the spatial arrangement of the object in the coordinate system is measured based on the relative positional relationship of each slit light photographed image and the slit light surface associated with each slit light photographed image. This is the original measurement method.

[Effect]

According to the present invention, multi-slit light and reference slit light can be obtained by one multi-slit light source. Therefore, parts such as the LD (laser diode), LD lighting circuit, lens, and diffraction grating that make up the light source can be saved, and three-dimensional measurement of objects can be realized with a compact device. Furthermore, it is not necessary to set the positional relationship between the two light sources, and the associated man-hours for adjusting the arrangement of the light sources, calibrating, etc. can be saved.

[Basic principle of the invention]

Next, the basic principle of the present invention will be explained based on FIGS. 1 and 2. A camera orthogonal coordinate system O ^C -x ^C y ^C z ^C is set as a predetermined coordinate system, with the focal point of the camera 1 as the origin and the optical axis direction of the camera 1 as the z-axis. Furthermore, a multi-slit orthogonal coordinate system O ^M -x ^M y ^M z ^M with the multi-slit light source 5 as the origin is set. The relationship between the camera orthogonal coordinate system and the multi-slit orthogonal coordinate system is generally expressed by the following relational expression.

x ^M y ^M z ^M 1=h ₁₁ h ₁₂ h ₁₃ h ₁₄ h 21 h ₂₂ h ₂₃ h ₂₄ h ₃₁ h ₃₂ h ₃₃ h ₃₄ h ₄₁ h ₄₂ h _{43 h 44 x} ^C y ^C z ^C 1 ……( 1) However, h _ij (i, j = 1, 2, 3) represents the direction cosine of the angle formed between the i and j coordinate axes, and h _i4 (i =
1, 2, 3) represent the parallel movement distance.

Now, each slit light surface M of the multi-slit light 3
is a plane expressed in a multi-slit orthogonal coordinate system
Assume that x ^M =0 can be obtained by rotating by a predetermined angle with the y ^M axis as the rotation axis. If the plane x ^M = 0 is θ _j with the y ^M axis as the rotation axis, and the slit optical surface rotated by j = 1 to m is Mj, then M _j is x ^M cosθ _j −z ^M sinθ _j = 0...(2 ). If we express this in the camera orthogonal coordinate system,
Using equation (1), M _j : (h ₁₁ x ^C + h ₁₂ y ^C + h ₁₃ z ^C + h ₁₄ ) cosθ _j − (h ₃₁ x ^C
+h ₃₂ y ^C +h ₃₃ z+h ₃₄ sinθ _j =0 (j=1 to m)...(3).

Let M _G be the slit light surface S of the reference slit beam 8, and let θ _S be θ _S =θ _G (4). At this time, the slit light surface S of the reference slit light 8 is expressed as x ^S cos θ _S −z ^S sin θ _S =0 (5). Expressing this in the camera orthogonal coordinate system, using equation (1), S: (h ₁₁ x ^C + h ₁₂ y ^C + h ₁₃ z ^C + h ₁₄ ) cosθ _S − (h ₃₁ x ^C
+h ₃₂ y ^C +h ₃₃ z ^C +h ₃₄ ) sinθ _S =0...(6). Now, by irradiating m multi-slit light beams 3, a plurality of slit light projection images 9 are formed on the surface of the object 4, and by photographing them with the camera 1, the image plane c of the photographing element of the camera 1, etc. Assume that l slit light photographed images corresponding to the slit light projection image 9 are obtained above. Here, the object 4 is on the line of convenience for explanation, and the coordinates of the slit light projection image 9 on the object 4 are P _k (x ^C _pk , y ^C _pk , z ^C _PK ), (k = 1 to l),
The coordinates of the slit light photographed image corresponding to each slit light projection image 9 are I _k (x ^C _Ik , y ^C _Ik , z ^C _Ik ), (k=1
~l). Here, for simplicity, image plane C
Let be a plane parallel to the x ^C y ^C -plane z ^C = f (constant, focal length). At that time, line of sight 10 is straight line L _k (=
O ^C I _k -----→), L _k is expressed as L _k :x ^C /x ^C _Ik =y ^C /y ^C _Ik =z ^C /f (7). The three-dimensional coordinates of one of the dotted slit light projection images 9, P _k, are the lines L _k and P _k , which are the line of sight 10.
It is found as the intersection with the slit optical surface M _j corresponding to .

However, at this point, the correspondence between each slit light photographed image I _k (k=1 to l) and the slit light surface M _j by the multi-slit light 3, that is, I ₁ to I _l
It is unclear which slit light plane M ₁ to M _n that is irradiated onto the surface of the object 4 corresponds to. Therefore, in the present invention, the multi-slit light 3 is controlled, one of the slit lights is used as the reference slit light 8, and the reference slit light 8 and the multi-slit light 3 are irradiated in a time-sharing manner. Here, the coordinates of the slit light projection image 9 on the object 4 by the reference slit light 8 are set as P _S (x ^C _PS , y ^C _PS , z ^C _PS ), and the reference slit light image corresponding to the slit light projection image 9 is image coordinates
I _S (x ^C _IS , y ^C _IS , z ^C _IS ), z ^C _IS =f. At that time, the line of sight L _S (=O _C I _S ---→) is expressed as L _S :x ^C /x ^C _IS =y ^C /y ^C _IS =z ^C /f (8). This is a point-like slit light projection image 9.
P _S is determined as the intersection of the straight line L _S and the slit optical surface S corresponding to P _S. In other words, equation (8) or x ^C = x ^C _IS / fz ^C , y ^C = y ^C _IS / fz ^C ... (9) is obtained, and by substituting this into equation (6) and summarizing about z ^C , , {(h ₁₁ x ^C _IS /f+h ₁₂ y ^C _IS /fh ₁₃ ) cosθ _S − (h ₃₁ x ^C _IS
/f+h ₃₂ y ^C _IS /f+h ₃₃ ) sinθ _S }z ^C = −(h ₁₄ cosθ _S
−h ₃₄ sinθ _S )……(10), z ^C = −(h ₁₄ cosθ _S −h ₃₄ sinθ _S )/g……(11a) Substituting this into equation (9), x ^C = − (h ₁₄ cosθ _S −h ₃₄ sinθ _S ) x ^C _IS /fg ... (11b) y ^C = - (h ₁₄ cosθ _S - h ₃₄ sinθ _S ) y ^C _IS /fg ... (11c) is obtained.

Here, g=(h ₁₁ x ^C _IS /f+h ₁₂ y ^C _IS /f+h ₁₃ ) cosθ _S
−(h ₁₃ x ^C _IS /f+h ₃₂ y ^C _IS /f+h ₃₃ ) sinθ _S ……(12
). Equations (11a) to (11c) (hereinafter simply referred to as Equation (11)) and Equation (12) express P _S in the camera orthogonal coordinate system. By the way, since the slit light surface S formed by the reference slit light 8 coincides with one M _G of the slit light surfaces M formed by the multi-slit light 3, the slit light surface photographed by the multi-slit light 3 I _k (k=1 to l) is You are sure to find something corresponding to the reference slit light photographed image _IS inside. In this way, if the slit light photographed image specified by _IS is _Ir , _Ir is associated with the slit light surface S formed by the reference slit light 8, and therefore the slit light surface _Mk formed by the multi-slit light 3
It will be associated with After that, from the arrangement of the multi-slit light 3 and the arrangement of the slit light photographed images,
Assuming there is no interchange between the two, I _r-1 becomes M _G-1 and I _r-2
is M _G-2 ...I ₁ is M _G-r+1 , I _r+1 is M _G+1 , I _r+2 is
M _G+2 ...Il is sequentially associated with M _G+lr . From this correspondence, in equations (11) and (12), (x ^C _IS , y ^C _IS ,
Instead of θ _S ), in P ₁ (x ^C _I1 , y ^C _I1 , θ _G-r+1 )
In P ₂ , (x ^C _I2 , y ^C _I2 , θ _G-r+2 )...In P _l , (x ^C _I
The ^three _-dimensional coordinates _of the slit light projection image 9 of the object 4 according to the camera orthogonal coordinate system can be determined by substituting the following values _: becomes.

〔Example〕

Hereinafter, a first embodiment of the present invention will be described based on the drawings.

() Configuration of Measuring Apparatus An apparatus for realizing the three-dimensional measuring method of an object according to the first embodiment of the present invention is configured as shown in FIG. In the figure, 1 is a camera;
In this embodiment, a television camera is used. 11 is an analog-digital converter, 12 is an object 4
This is the usual proof that the surface is irradiated. Nos. 13 and 14 are Japanese Patent Application No. 60-284767, which is the earlier application of the applicant of the present invention.
The light source 5 is a multi-slit light source 5 as disclosed in the above issue, and is appropriately selected and used depending on the arrangement of the object 4. The multi-slit light sources 13 and 14 each generate a plurality of slit lights, and one of the slit lights has higher brightness than the other slit lights. Reference numeral 15 denotes an image processor, which includes a grayscale image processing section 16 for extracting the outline of the object 4, and a ridge point processing section 17 for extracting the center line of the linear object 4 into a line with a thickness of 1 pixel width. , a polar point processing section 18 for extracting local peak points, a binarization processing section 19 for binarizing a digital signal, and an inter-image AND operation section 20 having an inter-image AND operation function. 2
Reference numerals 1 and 22 denote image memories, which are devices for storing images obtained on the image plane C of the camera 1. 23 is a line direction image calculation unit, which
Based on the arrangement, calculations are performed to select two light source sets 13 and 14 having different directions. Reference numeral 24 denotes a correspondence calculation unit, which identifies a slit light photographed image corresponding to the standard slit light photographed image by the standard slit light 8 from among the slit light photographed images by the multi-slit light 3, and combines the slit light photographed image with the slit light surface M. This is used to calculate the correspondence between the
25 is a distance calculation unit, which calculates three-dimensional coordinates using equations (11) and (12). 26,
27 is Batsuhua. 29 is the CPU,
It is connected to all the devices mentioned above and controls and performs calculations on all the devices.

() Measurement processing Assume a linear object as the object to be measured. First, as shown in FIG. 1, a camera 1 and a multi-slit light source 5 are fixedly placed near a linear object 4 of interest. The multi-slit light source 5 emits the multi-slit light 3 and the reference slit light 8 as described in the basic principle of the invention, and is selected from the multi-slit light sources 13 and 14 depending on the position of the object 4.
The process will be explained below according to the flowchart shown in FIG. The object 4 is illuminated by lighting the normal illumination 12 placed near the object 4 (41). However, if the object 4 and the background can be clearly distinguished, it is not necessary to turn on the light. Next, the usual proof 12
Object 4 illuminated by is photographed by camera 1,
The outline of the object 4 is captured on an image plane C of a photographing element or the like. Thereafter, the photographed image is converted into a digital signal having gray levels by the analog-to-digital converter 11 (80), and the signal is further input to an image processor. Image processor 1
The signal input to 5 is sent to the grayscale image processing unit 16
Then, as shown in the flowchart of FIG. 5, smoothing processing or Laplacian processing 81 or smoothing processing and Laplacian processing 82 is performed. next,
The signal is subjected to ridge point processing 83 in the ridge point processing section 17, and further subjected to binarization processing 84 in the binarization processing section 19. If object 4
When there is a lot of whisker-like noise on the center line, after binarization processing 84, noise removal processing 85 is performed to remove the noise.

Through the processing described above, the center line of the object 4 is extracted as a line with a thickness of 1 pixel width, and a storage process 86 is performed in the image memory 21 as the image 30 (42). Next, the linear direction calculation unit 23 calculates the arrangement direction of the linear object 4 from the obtained data. The method is to first find the starting point L _S (x _s ^C , y _s ^C ) of the object 4 on the image plane C using the camera orthogonal coordinate system.
and the end point L _e (x _e ^C , y _e ^C ). If the object 4 to be treated here is an elongated linear member that does not have a very complicated shape, the angle θ indicating the direction in which the object 4 is arranged is determined from L _s and L _e . That is, the direction θ is θ=arc tany _e ^C −y _s ^C /x _e ^C −x _s ^C. As a result, a direction vector l=(cos θ, sin θ) indicating the arrangement of the object 4 is determined. Next, using this direction vector l, light sources A and B arranged in different directions are selected. The method is to calculate the absolute value of the inner product _{l→・S A −} → and _the inner product l→・_SB The absolute value of -→ is calculated, the magnitude of both is compared, and the light source with the smaller absolute value of the inner product is selected (43). This is to improve measurement resolution by obtaining more slit light projection images 9 for the multi-slit light 3. However, it is not necessary to select only one of the light sources 13 and 14, and measurement may be performed twice from different directions using both light sources 13 and 14.

Next, the normal illumination 12 is turned off and the selected multi-slit light source 13 or 14 is turned on to irradiate the surface of the object 4 with the multi-slit light 3 (44). Here, the G-th slit light in the multi-slit light 3 is defined as a reference slit light 8, and its light surface is defined as a reference slit light surface S. Multi-slit light 3
The emission line 7 crosses the surface of the object 4 by
is drawn, and a multi-slit light projection image 9 is formed on the surface of the object 4. Multi-slit light source 13,
14, the one disclosed in Japanese Patent Application No. 60-284767 is used, so the multi-slit light projection image 9 obtained at this time consists of a reference slit light beam of higher brightness and other slit lights. A camera 1 photographs a bright line 7 including a slit light projection image. The photographed line image 37 is converted into a digital signal having gray levels by the analog-to-digital converter 11 shown in FIG. 3 (90),
Furthermore, the signal is input to the image processor 15. The signal input to the image processor 15 is
The gradation image processing unit 16 performs smoothing processing or Laplacian processing 91 or smoothing processing and Laplacian processing 92 as shown in the flowchart of FIG.
will be done. Next, the signal is processed by the ridge point processing section 17.
Ridge point processing 93 is performed at , and further binarization processing 95 is performed at the binarization processing section 19 to create a line image 3 by the reference slit light 8 with even higher brightness.
7 is extracted and stored 96 in the image memory 22 as an image 31 (45). The threshold level in the binarization processing section is set to a level high enough to extract only the reference slit light with high brightness from the multi-slit light. At this time, instead of the ridge point processing 93 performed by the ridge point processing section 17, the extreme point processing 94 performed by the extreme point processing section 18 may be applied. In this case, points having local peaks are extracted rather than the center line of the line image 37 formed by the reference slit light 8. Particularly, when the reference slit light photographed image 38 is extracted as one pixel, the integration process shown in FIG. 8 is unnecessary, which is efficient.

Next, the image data is read out from the image memory 21 and the image memory 22 at the same time, and the image data is
The AND operation unit 20 combines the image 30 and the image 31.
By performing an AND operation, the reference slit light photographed image 3 is obtained.
Extract 8. The reference slit light photographed image 38 is
Even if the object 4 is linear, there may generally be cases where a plurality of points are gathered. Therefore, those coordinates are expressed as B _j (x ^C , y ^C )j=1~n in the camera orthogonal coordinate system.
and store it in the buffer 26 (46).
The flow of these processes is from 100 to 10 in Figure 7.
2.

Next, the threshold level is lowered and the multi-slit optical image 39 is extracted (47). This process is similar to the case where the reference slit light photographed image 38 is obtained only with a low threshold level. The processing flow is described in FIG. 4 from 47 to 50, and corresponds to 44 to 46 for the reference slit light 8. That is, a plurality of bright lines 7 are drawn across the surface of the object 4 by the multi-slit light 3, and a plurality of slit light projection images 9 are formed on the surface of the object 4. A plurality of bright lines 7 including a plurality of slit light projection images 9 are photographed by the camera 1, and the image plane C of the photographing element of the camera 1, etc.
A plurality of line images 37 including a plurality of slit light photographed images 39 corresponding to a plurality of slit light projection images 9 formed by the multi-slit light 3 are obtained. The photographed line image 37 is processed in the same manner as in the case of the reference slit light 8, and the center line of the line image 37 by the multi-slit light 3 is extracted, and storage processing 96 is performed in the image memory 22 as the image 33 (48).

Next, the image data is read out from the image memory 21 and the image memory 22 at the same time, and the image data is
The AND operation unit 20 combines the image 30 and the image 33.
A plurality of slit light photographed images 39 are extracted by performing an AND operation. Each slit light photographed image 39
Even if the object 4 is linear, it is generally a collection of multiple points, and the coordinates of the obtained slit light photographed image 39 are expressed as C _j (x ^C ,
y ^C ), j=1 to k, and stored in the buffer 27 (49). Note that the other processing performed in the case of the reference slit light 8 is the multi-slit light 3.
In the case of , the corresponding processing is similarly performed.
After that, turn off the multi-slit light source 5 (50).

Further, assuming that there is no exchange between the slit light images 39, a slit light image 39 corresponding to the reference slit light image 38 of the reference slit light 8 is selected from among the plurality of slit light images 39 of the multi-slit light 3. Identify. To this end, the mapping calculation unit 24 performs mapping processing as shown in FIG. First, data B _j (x ^C ,
y ^C ), taking the average B(x, y) of j=1 to n,
The coordinates of the reference slit light photographed image 38 are made representative (61). That is, B (x ^C , y ^C )=1/n _o 〓 ^j=1 B _j (x ^C , y ^C ).

Next, this reference slit light photographed image 38 and each slit light photographed image 3 by the multi-slit light 3 are
Find the distance and direction from 9 (62). That is,
The distance d _j and the angle η _j on the image plane C are d _j = |C _j (x ^C , y ^C )−B(x ^C , y ^C ) | η _j = arc tanC _j (x ^C )− B( ^xC )/ _Cj ( ^yC )-B( ^yC )j=1~k. Next, rearrange C _j (x ^C , y ^C ) from the smallest d _j and set it as D _j (x ^C , y ^C ) (63). In D _j (x ^C , y ^C ), _a certain distance r
Those smaller than are classified as class 0 (65). Next, for cases where the sign of η _j is positive,
Classification is performed using the cluster method of No. 235898, and classes 1, 2, and 3 are classified from those closest to B (x ^C , y ^C ).
... (68). Similarly, the same process is applied to the negative signs of η _j , and the classes closest to B(x ^C , y ^C ) are classified as −1, −2, −3, and so on. When the ridge point processing 93 is performed, the slit light photographed image 39 may be extracted with a size of several pixels, so the average within the same class is taken to represent it (69). Let them be E _p (x ^C , y ^C ), p=1 to l.

When the polar point processing 94 is performed and each slit light photographed image 39 is extracted by one pixel, it is not necessary to calculate the average value. Sort E _p (x ^C , y ^C ) from the smallest class (70). At this time, p=r
If is class 0, E _r (x ^C , y ^C ) corresponds to the reference slit light photographed image 38 , and by this processing, the reference slit light photographed image 38 is extracted from the slit light photographed image 39 . This means that the corresponding one has been identified. Therefore, the illumination angle _{θ S} of the slit light surface S by the reference slit light 8
coincides with the slit optical surface M _G by the G-th multi-slit beam 3, so E _r (x ^C , y ^C ) is θ _G
It will be associated with Then, regarding the other slit light photographed images 39, assuming that there is no replacement, E _r+1 (x ^C , y ^C ) becomes θ _G+1 , E _r+2
(x ^C , y ^C ) is θ _G+2 , ...El (x ^C , y ^C ) is θ _G+lr , and similarly E _r-1 (x ^C , y ^C ) is θ _G-1 , E _{r- 2} (x ^C , y ^C ) is associated with θ _G-2 , and...E ₁ (x ^C , y ^C ) is associated with θ _G-r+1 (71). In the above, for simplicity, the coordinates on the image plane C are expressed using the camera orthogonal coordinate system, but in general, an image plane coordinate system may be set on the image plane C and the coordinates may be expressed using the coordinates. In that case, a process 72 is required to change the image plane coordinate system to the camera orthogonal coordinate system. In this way, each slit light photographed image 39 is associated with each slit light plane M _j of the multi-slit light 3 based on the relative positional relationship between the specified slit light photographed image 39 and other slit light photographed images 39. Become (51). The coordinates and angle of the slit light photographed image 39 on each image plane C obtained in this way are expressed as (x ^C _IS , y ^C _IS ,
θ _S ), and after finding the relationship between the camera coordinate system and the multi-slit coordinate system in advance by calibration etc., (h _ij (i, j = 1 to
4)), the coordinates of the slit light projection image 9 and the distance between the coordinates can be found (73,
52), three-dimensional measurement of object 4 will be performed.

In this embodiment, the object 4 is assumed to be linear, but the object 4 is not limited to a linear object, but can also be applied to a general three-dimensional object. On this occasion,
The slit light projection image on the object is processed as a linear image (in the embodiment, it was processed as a dot-shaped image).

() Other Examples In the above example, as a multi-slit light source, one slit light among a plurality of slit lights was made to have a much higher brightness than the other slit lights, and this was used as a reference slit light. Second aspect of the present invention
In this embodiment, a shutter is provided in front of the multi-slit light source, and the shutter realizes the irradiation of only the reference slit light and the irradiation of the entire multi-slit light in a time-division manner.

The apparatus according to the second embodiment of the present invention is constructed as shown in FIG. 9, and the measurement process is performed as shown in the flow chart of FIG. In FIG. 9, the difference from the configuration in FIG. 3 is that the multi-slit light source 1
3 and 14 are respectively provided with shutters 53 and 54 to form a light source set. A multi-slit light source having such a shutter mechanism has been proposed in Japanese Patent Application No. 1983-1983, which was proposed as an earlier application by the applicant of the present invention.
Disclosed in No. 284768. That is, shutter 5
When 3 and 54 are turned ON, one slit light in the multi-slit light (reference slit light)
All other slit lights are blocked by the shutter, and only one reference slit light is irradiated onto the object. Further, the shutters 53 and 54 are opened by being turned off, and all the slit lights (multi-slit lights) are irradiated onto the object. The shutters 53 and 54 can be controlled ON/OFF by a host computer or a manual switch.

In FIG. 10, the difference from the flow diagram in FIG. 4 is that the shutter 53 or 54 is turned on when the multi-slit light source 13 or 14 is turned on in step
Then, in step 47, the shutter 53 or 54 is turned off to obtain the multi-slit light 3 and irradiate the surface of the object 4 with the multi-slit light 3. It is.

In the first and second embodiments described above, the measurement method described above may not be applicable because the correspondence cannot be established unless the object to be measured is irradiated with the reference slit light. In this case, the problem can be solved by using a mechanism that allows the light source itself to be moved. For example, a light source may be mounted on an XY stage, a robot, or the like.

Further, in the second embodiment, the light source may be fixed and the shutter may be moved to move the position of the reference slit light among the multi-slit lights. Furthermore, the position of the reference slit light may be changed by dividing the shutter itself into a plurality of segments corresponding to each slit light of the multi-slit light and controlling each of the segments. If such segments are used, the reference slit light can be coded and irradiated, allowing more advanced three-dimensional measurement.

〔Effect of the invention〕

As explained above, according to the present invention, by controlling the multi-slit light, it is possible to substantially obtain the multi-slit light and the reference slit light with one light source, and the multi-slit light and/or the reference slit light can be substantially obtained using one light source. When an object is irradiated with slit light, three-dimensional measurement of the object can be performed based on the obtained image and the relationship between each slit light. Therefore, in an apparatus that implements the method according to the present invention, it is possible to measure an object using one camera and one multi-slit light source, and the processing thereof becomes simple. In addition, since the present invention focuses only on the photographed image on the object, it can be measured without depending on the contrast with the background, the unevenness of the background, etc., so it can be widely applied to wiring automation etc. It is something that can be done.

[Brief explanation of drawings]

Fig. 1 is an explanatory diagram of the principle of the present invention, Fig. 2 is a diagram of the basic principle of the invention, Fig. 3 is an overall configuration diagram of the first embodiment of the invention, and Fig. 4 is a diagram of the first embodiment of the invention. Flowchart of processing, FIG. 5 is a flowchart of object contour extraction processing, FIG. 6 is a flowchart of line image extraction processing using reference slit light or multi-slit light, FIG. 7 is a flowchart of reference slit image or slit light photographed image extraction processing, Fig. 8 is a flow chart of the correspondence calculation process, Fig. 9 is an overall configuration diagram of the second embodiment of the present invention, Fig. 10 is a flow chart of the process of the second embodiment of the present invention, and Fig. 11 is a conventional object A spatial layout diagram related to the three-dimensional measurement method of
FIG. 12 is a diagram showing images taken by cameras 1 and 2. 1, 2...Camera, 3...Multi-slit light,
4...Object, 5...Multi-slit light source, 9...
Slit light projection image, 8...Reference slit light, 38
...Reference slit light photographed image, 39...Slit light photographed image.

Claims (1)

[Claims] 1. Irradiating the surface of an object with multi-slit light generated by one multi-slit light source and having a plurality of slit light surfaces whose arrangement in a predetermined coordinate system is set in advance, and controlling the multi-slit light. Then, one of the slit lights is used as a reference slit light to illuminate the surface of the same object, and a plurality of slit light photographed images corresponding to the slit light projected image by the multi-slit light on the object surface and a reference slit light on the same object surface are obtained. A reference slit light photographed image corresponding to the slit light projected image by light is obtained on a predetermined image plane, a slit light photographed image corresponding to the reference slit light photographed image is identified from the plurality of slit light photographed images, and the plurality of slit light photographed images are identified. Assuming that there is no interchange between the slit light images, each slit light image and each slit light surface of the multi-slit light are determined based on the relative positional relationship between this identified slit light image and other slit light images. The spatial arrangement of the object in the above coordinate system is measured based on the relative positional relationship of each slit light photographed image and the slit light surface associated with each slit light photographed image. A method for three-dimensional measurement of objects. 2 The brightness of one slit light of the multi-slit light is made higher than the other slit lights, and a plurality of slit light photographed images corresponding to the slit light projected image by the multi-slit light are obtained at a high level of the threshold value in the binarization process. and, on the other hand, obtain one slit light photographed image serving as a reference corresponding to a slit light projected image by high brightness slit light at a low threshold level in the binarization level. The method described in section. 3 The multi-slit light generated by the multi-slit light source can be shielded by a shutter except for one slit light that should serve as a reference, and the single slit light that serves as a reference and the multi-slit light can be separated in a time-sharing manner by turning on/off the shutter. The method according to claim 1, wherein the method is as follows.