JP5563930B2 - Non-contact three-dimensional measuring apparatus and non-contact three-dimensional measuring method - Google Patents

Description

  The present invention relates to a non-contact three-dimensional measurement apparatus and a non-contact three-dimensional measurement method for measuring the shape of an object to be measured in a non-contact manner.

  There are a passive method (passive measurement method) and an active method (active measurement method) as three-dimensional measurement methods for measuring an object to be measured. The passive method is a method of performing measurement without irradiating a target object with specific light, radio waves, or the like that assist measurement. The passive method includes, for example, a photo measurement method. The active method is a method of performing measurement using information obtained by irradiating light, radio waves, sound waves, or the like on a target object in order to perform three-dimensional measurement. The active method includes, for example, a light cutting method.

  Photo measurement, which is one of the passive methods, performs external orientation (determining the positional relationship between two cameras) from two or more captured images, and associates the same location with the feature points between the images. The three-dimensional point was calculated | required by (For example, refer patent document 1). For this reason, the size of the object to be measured is arbitrary as long as it can be seen in the photograph, and it is possible to measure from a large object to a small object.

Japanese Patent No. 3924576

Xu Tsuyoshi, “Three-dimensional CG made from photographs”, Modern Science, January 2001 31-P. 86 Edited by Toru Yoshizawa, “Optical 3D Measurement”, New Technology Communications, March 1993 28-P. 37 Kenichi Kanaya, “Image Understanding”, Morikita Publishing, November 1991, p. 48-P. 53 Satoshi Sato, “Computer Vision-Geometry of Vision”, Corona, May 1999 80-P. 103 Z.Zhang: Flexible Camera Calibration by Viewing a Plane from Unknown Orientation.Proc of 7th Int.Conference on Computer Vision, Kerkyra, Greece.pp.666-673, Sept.1999

  In photogrammetry, the association between two photographs taken by two cameras is shown in common in the two photographs, and is performed using feature points that can be identified as the same location. For this reason, if there is no feature point that can be recognized as the same location on the surface of the object to be measured, it is difficult to associate the two photographs. For example, since the surface of a plastic molded product is smooth and has no features, three-dimensional measurement using feature point correspondence by photogrammetry has not been possible.

  As described above, it is difficult to measure a shape having no features such as a smooth surface in the photo measurement which is a passive method. In addition, the light cutting method, which is an active method, has the disadvantage that the size of the object to be measured is limited by the measuring device.

  An object of the present invention is to provide a non-contact three-dimensional measurement apparatus and a non-contact three-dimensional measurement method that enable non-contact three-dimensional shape measurement of an object having no features on the surface at high speed.

A non-contact three-dimensional measurement apparatus according to the present invention is arranged at a distance from each other, and a first camera and a second camera that shoot a measurement object,
A line laser for irradiating the object to be measured with a line laser;
An orientation unit that performs external orientation for calculating the positional relationship between the first camera and the second camera;
An image processing unit that performs image processing on a first image obtained by photographing the object to be measured irradiated with the line laser with the first camera, and a second image photographed by the second camera;
With
The image processing unit
In the first image, a rectangular area including a portion where the line laser is photographed is selected as a first target area,
In the second image, a rectangular region including a portion where the line laser is photographed is selected as a second target region,
In the first target area, a portion where the line laser is imaged is extracted as first line segment data,
In the second target area, a portion where the line laser is imaged is extracted as second line segment data,
Selecting one point in the first line segment data in the first target area as a feature point;
An epipolar line corresponding to the feature point selected in the first target region on the first image is calculated in the second target region on the second image;
In the second target area on the second image, the intersection of the captured second line segment data corresponding to the line laser and the epipolar line is defined as the first target area on the first image. Determined as a corresponding point corresponding to the feature point selected in
Corresponding to the feature point and the corresponding point using the feature point in the first target region on the first image and the corresponding point in the second target region on the second image The three-dimensional coordinates of the corresponding point on the object to be measured are calculated.

  In addition, the first camera and the second camera record image data taken along a scanning line in a direction intersecting with a straight line connecting the first camera and the second camera. May be output as the first image and the second image, respectively.

  Further, the image processing unit is configured so that the long side of the rectangular region is parallel to the direction of the scanning line when the first image and the second image are recorded. Two target areas may be selected.

  Furthermore, the image processing unit can select the feature points only within the first target region and determine the corresponding points only within the second target region.

  Still further, the line laser may irradiate the object to be measured so as to intersect with a straight line connecting the first camera and the second camera.

  The line laser may be capable of scanning the line laser over the entire object to be measured.

  Furthermore, the line laser may rotate about a predetermined axis in the irradiation direction and irradiate the object to be measured with the line laser.

Non-contact three-dimensional measurement method according to the present invention,
(A) an external orientation step for performing an external orientation for calculating the positional relationship between the first camera and the second camera that are arranged at a distance from each other and photograph the object to be measured;
(B) a line laser irradiation step of irradiating the object to be measured with a line laser;
(C) The object to be measured irradiated with the line laser is photographed by the first camera and the second camera, respectively, and the first image by the first camera and the second image by the second camera are taken. An image capturing step for obtaining two images;
(D) a first target region selection step of selecting, as the first target region, a rectangular region including a portion where the line laser is captured in the first image;
(E) a second target region selection step of selecting, as the second target region, a rectangular region including a portion where the line laser is captured in the second image;
(F) a first line segment data extraction step of extracting a portion where the line laser is photographed as first line segment data in the first target area;
(G) a second line segment data extracting step of extracting a portion where the line laser is photographed as second line segment data in the second target area;
(H) a feature point selection step of selecting one point in the first line segment data in the first target region as a feature point;
(I) an epipolar line calculating step for calculating an epipolar line corresponding to the feature point selected in the first target area on the first image in the second target area on the second image;
(J) In the second target region on the second image, an intersection point of the second line segment data corresponding to the captured line laser and the epipolar line is defined as the intersection on the first image. A corresponding point determining step for determining as a corresponding point corresponding to the feature point selected in the first target region;
(K) The feature points and the correspondences using the feature points in the first target region on the first image and the corresponding points in the second target region on the second image. A three-dimensional coordinate calculation step for calculating a three-dimensional coordinate of a corresponding point on the measurement object corresponding to a point;
including.

  In the image photographing step, the image data photographed by the first camera and the second camera is taken along a scanning line in a direction intersecting with a straight line connecting the first camera and the second camera. May be recorded and output as the first image and the second image, respectively.

Furthermore, in the first target area selection step, the first target area is selected so that the long side of the rectangular area is parallel to the direction of the scanning line when the first image is recorded,
In the second target area selecting step, the second target area may be selected so that a long side of the rectangular area is parallel to a scanning line direction when the second image is recorded.

  Still further, in the feature point selection step, the feature point can be selected only within the first target region, and in the corresponding point determination step, the corresponding point can be determined only within the second target region.

  In the line laser irradiation step, the object to be measured may be irradiated with a line laser so as to intersect with a straight line connecting the first camera and the second camera.

  Further, (l) the line corresponding to the photographed line laser is repeated by repeating the steps (h) to (k) for all the points on the first line segment extracted from the first image. A three-dimensional coordinate calculation step for calculating three-dimensional coordinates for the portion may be included.

  The line laser irradiation step (b) further includes a line laser scanning step of scanning the line laser over the object to be measured, and each time the line laser is irradiated, the steps (c) to (l) ) May be repeated to scan the entire object to be measured.

  The line laser irradiation step (b) further includes a line laser scanning step of irradiating the object to be measured by rotating the line laser about a predetermined axis in an irradiation direction, and rotating the line laser to rotate the object to be measured. Each time the measurement object is irradiated, the step (c) to the step (l) may be repeated.

  Still further, in the corresponding point determining step (j), when there are two or more intersections between the epipolar line and the second line segment data, the line laser corresponds to the feature point on the first image. Irradiating by rotating about the axis of the irradiation direction to the corresponding point on the object to be measured, and setting the single intersection point of the epipolar line and the second line segment data on the second image as the feature point It may be determined as a corresponding point.

Non-contact three-dimensional measuring device according to the present invention,
A first set of camera systems including a first camera and a second camera, which are arranged at a distance from each other and photograph a measurement object;
A second set of camera systems including a third camera and a fourth camera, which are arranged at positions different from the first set of camera systems and at a distance from each other, and shoot the measurement object;
A line laser for irradiating the object to be measured with a line laser;
A first orientation unit that performs external orientation on the positional relationship between the first camera and the second camera;
A second orientation unit that performs external orientation on the positional relationship between the third camera and the fourth camera;
A first image obtained by photographing the object to be measured irradiated with the line laser with the first camera and a second image photographed with the second camera are subjected to image processing, and the line laser is emitted. An image processing unit that performs image processing on a third image obtained by photographing the measured object with the third camera and a fourth image obtained by photographing with the fourth camera;
With
The image processing unit
Selecting one point in the first line segment data corresponding to the line laser imaged on the first image as a feature point;
An epipolar line corresponding to the feature point selected on the first image is calculated on the second image;
On the second image, an intersection between the second line segment data corresponding to the captured line laser and the epipolar line is determined as a corresponding point corresponding to the feature point selected on the first image. And
Using the feature points on the first image and the corresponding points on the second image, the three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points are obtained. Calculating and obtaining a three-dimensional coordinate system on the object to be measured in the first set of camera systems;
Selecting one point in the third line segment data corresponding to the line laser imaged on the third image as a feature point;
An epipolar line corresponding to the feature point selected on the third image is calculated on the fourth image;
On the fourth image, an intersection of the fourth line segment data corresponding to the captured line laser and the epipolar line is determined as a corresponding point corresponding to the feature point selected on the third image. And
Using the feature points on the third image and the corresponding points on the fourth image, the three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points are obtained. Calculating and obtaining a three-dimensional coordinate system on the object to be measured in the second set of camera systems;
Of the three-dimensional coordinates on the measurement object in the second set of camera systems, select at least three three-dimensional coordinates corresponding to the three-dimensional coordinates on the measurement object in the first set of camera systems; For the selected at least three three-dimensional coordinates, a coordinate conversion relationship from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems is obtained, and the obtained coordinate conversion relationship And converting the three-dimensional coordinate system of the second set of camera systems into the three-dimensional coordinate system of the first set of camera systems, and based on the three-dimensional coordinate system of the first set of camera systems Obtain the three-dimensional coordinates of the measurement object.

  In addition, the image processing unit, for the selected at least three three-dimensional coordinates, a relationship of similarity conversion from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems Based on the above, the coordinate conversion relationship can be obtained.

Further, the similarity transformation from the three-dimensional coordinate system N (x, y, z) of the second set of camera systems to the three-dimensional coordinate system M (x, y, z) of the first set of camera systems is: For example, it can be represented by the following formula.
M _i = sR (N _i −t)
Mi: 3D coordinates (coordinate vector) of the first set of camera systems
Ni: 3D coordinates (coordinate vector) of the second set of camera systems
s: Scale (scalar value)
R: rotation matrix t: translation vector

  Still further, the line laser may irradiate the object to be measured so as to intersect with a straight line connecting the first camera and the second camera.

  The line laser may scan the line laser over the entire object to be measured.

  Furthermore, the line laser may rotate about a predetermined axis in the irradiation direction to irradiate the object to be measured with the line laser.

The non-contact three-dimensional measurement method according to the present invention includes the calculation of the three-dimensional coordinate system of the measurement object by the first set of camera systems and the calculation of the three-dimensional coordinate system of the measurement object by the second set of camera systems. Then, the obtained three-dimensional coordinate system of the second set of camera systems is transformed into a three-dimensional coordinate system of the first set of camera systems, and the three-dimensional coordinate system of the first set of camera systems is converted. A non-contact three-dimensional measurement method for obtaining three-dimensional coordinates of the object to be measured based on
(A) A first external orientation that performs an external orientation relating to the mutual positional relationship of the first camera and the second camera that are arranged at a distance from each other and constitute a first set of camera systems for photographing the object to be measured. Steps,
(B) a line laser irradiation step of irradiating the object to be measured with a line laser;
(C) The object to be measured irradiated with the line laser is photographed by the first camera and the second camera, respectively, and the first image by the first camera and the second image by the second camera are taken. An image capturing step by a first camera system for obtaining two images;
(D) a first line segment data extracting step of extracting a portion where the line laser is photographed as first line segment data in the first image;
(E) in the second image, a second line segment data extracting step of extracting a portion where the line laser is photographed as second line segment data;
(F) a feature point selection step of selecting one point in the first line segment data extracted from the first image as a feature point;
(G) an epipolar line calculation step of calculating an epipolar line corresponding to the feature point selected on the first image on the second image;
(H) Corresponding point determination for determining an intersection between the second line segment data extracted from the second image and the epipolar line as a corresponding point corresponding to the feature point selected on the first image. Steps,
(I) Using the feature points in the first image and the corresponding points in the second image, three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points Calculating a three-dimensional coordinate by the first set of camera systems to obtain a three-dimensional coordinate system on the measurement object in the first set of camera systems;
(J) A third camera and a fourth camera that are arranged at a distance from each other at a position different from the first set of camera systems and constitute a second set of camera systems for capturing an object to be measured. A second external orientation step for performing external orientation with respect to each other's positional relationship;
(K) a line laser irradiation step of irradiating the object to be measured with a line laser;
(L) The object to be measured irradiated with the line laser is photographed by the third camera and the fourth camera, respectively, and a third image by the third camera and a fourth image by the fourth camera are taken. An image capturing step by a second set of camera systems for obtaining four images;
(M) a third line segment data extracting step of extracting a portion where the line laser is photographed as third line segment data in the third image;
(N) a fourth line segment data extracting step of extracting, as the fourth line segment data, a portion where the line laser is photographed in the fourth image;
(O) a feature point selecting step of selecting one point in the third line segment data extracted from the third image as a feature point;
(P) an epipolar line calculation step of calculating an epipolar line corresponding to the feature point selected on the third image on the fourth image;
(Q) Corresponding point determination for determining an intersection between the fourth line segment data extracted from the fourth image and the epipolar line as a corresponding point corresponding to the feature point selected on the third image. Steps,
(R) Using the feature points in the third image and the corresponding points in the fourth image, three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points Calculating a three-dimensional coordinate by a second set of camera systems to obtain a three-dimensional coordinate system on the measurement object in the second set of camera systems;
(S) Among the three-dimensional coordinates on the object to be measured in the second set of camera systems, at least three three-dimensional coordinates corresponding to the three-dimensional coordinates on the object to be measured in the first set of camera systems are A coordinate conversion relationship calculation for selecting and calculating a coordinate conversion relationship from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems for the selected at least three three-dimensional coordinates Steps,
(T) Using the obtained coordinate transformation relationship, the three-dimensional coordinate system of the second set of camera systems is transformed into the three-dimensional coordinate system of the first set of camera systems, and the first set of Obtaining a three-dimensional coordinate of the object to be measured based on a three-dimensional coordinate system of a camera system;
including.

  In the coordinate conversion relationship calculating step, similarity conversion from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems is performed on the selected at least three three-dimensional coordinates. Based on this relationship, the coordinate conversion relationship can be obtained.

Further, the similarity transformation from the three-dimensional coordinate system N (x, y, z) of the second set of camera systems to the three-dimensional coordinate system M (x, y, z) of the first set of camera systems is: For example, it can be represented by the following formula.
M _i = sR (N _i −t)
Mi: 3D coordinates (coordinate vector) of the first set of camera systems
Ni: 3D coordinates (coordinate vector) of the second set of camera systems
s: Scale (scalar value)
R: rotation matrix t: translation vector

  Still further, in the line laser irradiation step by the first set of camera systems, the measurement object may be irradiated with a line laser so as to intersect with a straight line connecting the first camera and the second camera. .

  Further, in the line laser irradiation step by the second set of camera systems, the measurement object may be irradiated with a line laser so as to intersect with a straight line connecting the third camera and the fourth camera.

Furthermore, following the three-dimensional coordinate calculation step (i) by the first set of camera systems,
(I-2) Steps (f) to (i) are repeated for all points on the first line segment extracted from the first image, and the line laser of the object to be measured is irradiated. A three-dimensional coordinate calculation step for calculating three-dimensional coordinates for the entire line portion may be further included.

  Further, the line laser irradiation step (b) by the first set of camera systems further includes a line laser scanning step of scanning the line laser over the object to be measured, and each time the line laser is irradiated, the step The whole object to be measured may be scanned by repeating steps (c) to (i-2).

  The line laser irradiation step (b) by the first set of camera systems further includes a line laser scanning step in which the line laser is rotated about a predetermined axis in the irradiation direction to irradiate the object to be measured. The steps (c) to (i-2) may be repeated each time the laser beam is rotated to irradiate the object to be measured.

Furthermore, following the three-dimensional coordinate calculation step (s) by the second set of camera systems,
(S-2) Steps (o) to (s) are repeated for all points on the third line segment extracted from the third image, and the line laser of the object to be measured is irradiated. A three-dimensional coordinate calculation step for calculating three-dimensional coordinates for the entire line portion may be further included.

  Further, the line laser irradiation step (k) by the second set of camera systems further includes a line laser scanning step of scanning the line laser over the object to be measured, and each time the line laser is irradiated, the step The whole object to be measured may be scanned by repeating steps (l) to (s-2).

  The line laser irradiation step (k) by the second set of camera systems further includes a line laser scanning step of irradiating the object to be measured by rotating the line laser about a predetermined axis in the irradiation direction. Step (l) to step (s-2) may be repeated each time the laser beam is rotated to irradiate the object to be measured.

  Further, in the corresponding point determination step (h) by the first set of camera systems, when there are two or more intersections between the epipolar line and the second line segment data, the line laser is moved onto the first image. Irradiating by rotating about the axis in the irradiation direction to the corresponding point on the object to be measured corresponding to the feature point of the single point of the epipolar line and the second line segment data on the second image May be determined as corresponding points corresponding to the feature points.

  Still further, in the corresponding point determination step (q) by the second set of camera systems, when there are two or more intersections between the epipolar line and the fourth line segment data, the line laser is set to the third image. Irradiating by rotating about the axis of the irradiation direction to the corresponding point on the object to be measured corresponding to the feature point above, the epipolar line and the fourth line segment data are simply displayed on the fourth image. One intersection may be determined as a corresponding point corresponding to the feature point.

  In the non-contact three-dimensional measuring apparatus, the line laser may be used in combination with a plurality of color line lasers.

  Furthermore, in the non-contact three-dimensional measurement method, in the step of irradiating the line laser, a plurality of color line lasers may be used in combination as the line laser.

  According to the non-contact three-dimensional measurement apparatus and the non-contact three-dimensional measurement method according to the present invention, the line laser is irradiated onto the object to be measured, and the three-dimensional based on the principle of photo measurement even if the object to be measured has no characteristics. Coordinates can be measured. In addition, due to the coherent nature of the laser, even if the object to be measured has a shape with irregularities, the line is less likely to be blurred, and the feature points can be extracted accurately.

  Moreover, in this non-contact three-dimensional measuring apparatus and non-contact three-dimensional measuring method, the line laser image | photographed on one image of the two images which image | photographed the to-be-measured object irradiated with the line laser with two cameras. A rectangular area including the first line segment data that corresponds to is selected as the first target area, and one point of the first line segment data in the first target area is selected as the feature point. In this case, a rectangular area including the second line segment data that is a portion corresponding to the captured line laser on another image is selected as the second target area, and the first target area is included in the second target area. The epipolar line corresponding to the feature point selected in the above is calculated, and the intersection of the epipolar line and the second line segment data can be specified as the corresponding point corresponding to the feature point. This makes it possible to clearly determine the corresponding point without ambiguity compared to the case where the corresponding point is specified by the correlation with the pattern irradiated on the object to be measured, improving the accuracy of 3D measurement. Can be made. Furthermore, in the non-contact three-dimensional measurement apparatus and the non-contact three-dimensional measurement method according to the present invention, the first target region and the second target region that are necessary and sufficient for calculating the feature points and the corresponding points are used as part of the entire image. Selected. As a result, the feature points and the corresponding points are calculated only for each of the first target region and the second target region, so that the three-dimensional measurement can be performed faster than the case of calculating over the entire two images. .

  In addition, when taking into account distortion of images taken by two cameras, the coordinates of the first line segment data in the first target area and the coordinates of the second line segment data in the second target area are corrected for distortion. After that, it is possible to obtain high speed and high accuracy by calculating the above three-dimensional coordinates.

  Further, the line laser may be a line laser of any color such as infrared, red, green, and blue. Further, by using a plurality of color lasers in combination, it is possible to extract a plurality of line segments in one shooting, and to calculate three-dimensional coordinates at high speed.

It is a block diagram which shows the structure of the non-contact three-dimensional measuring apparatus which concerns on Embodiment 1 of this invention. It is the schematic which shows arrangement | positioning of a to-be-measured object and two cameras. It is the schematic which shows the 1st image image | photographed with the 1st camera. It is the schematic which shows the 2nd image image | photographed with the 2nd camera. FIG. 4 is a schematic diagram when one point in the first line segment data is selected as a feature point for the first image in FIG. 3. FIG. 5 is a schematic diagram illustrating a second image in which epipolar lines corresponding to feature points on the first image are calculated in the second image of FIG. 4. It is a conceptual diagram for demonstrating the characteristic of the non-contact three-dimensional measuring method which concerns on Embodiment 1 of this invention. It is a flowchart of the non-contact three-dimensional measuring method which concerns on Embodiment 1 of this invention. It is the schematic which shows an example in the case of selecting a 1st object area and a 2nd object area so that the long side of a rectangular area may become parallel to the direction of the scanning line of an image. It is a flowchart of the non-contact three-dimensional measuring method which concerns on the modification of Embodiment 1 of this invention. It is a flowchart of the non-contact three-dimensional measuring method which concerns on Embodiment 2 of this invention. It is a flowchart of the non-contact three-dimensional measuring method which concerns on Embodiment 3 of this invention. It is the schematic which shows the structure of the non-contact three-dimensional measuring apparatus which concerns on Embodiment 4 of this invention. It is a flowchart of the non-contact three-dimensional measuring method which concerns on Embodiment 4 of this invention. It is a flowchart of the non-contact three-dimensional measuring method which concerns on Embodiment 4 of this invention. It is the schematic which shows the 1st image image | photographed with the 1st camera of the 1st set camera system. It is the schematic which shows the 2nd image image | photographed with the 2nd camera of the 1st set camera system. It is the schematic in the case of selecting one point in 1st line segment data as a feature point about the 1st image of FIG. It is the schematic which shows the 2nd image by which the epipolar line applicable to the feature point on a 1st image was calculated in the 2nd image of FIG. It is a conceptual diagram for demonstrating the characteristic of the non-contact three-dimensional measuring method by the 1st set camera system of the non-contact three-dimensional measuring apparatus which concerns on Embodiment 4 of this invention. It is the schematic which shows the 3rd image image | photographed with the 3rd camera of the 2nd set camera system. It is the schematic which shows the 4th image image | photographed with the 4th camera of the 2nd set camera system. FIG. 22 is a schematic diagram when one point in the third line segment data is selected as a feature point for the third image in FIG. 21. FIG. 23 is a schematic diagram illustrating a fourth image in which epipolar lines corresponding to feature points on the third image are calculated in the fourth image of FIG. 22. It is a conceptual diagram for demonstrating the characteristic of the non-contact three-dimensional measuring method by the 2nd set camera system of the non-contact three-dimensional measuring apparatus which concerns on Embodiment 4 of this invention.

  A non-contact three-dimensional measuring apparatus and measuring method according to the present invention will be described below with reference to the accompanying drawings. In the drawings, substantially the same members are denoted by the same reference numerals.

(Embodiment 1)
FIG. 1 is a schematic diagram illustrating a configuration of a non-contact three-dimensional measurement apparatus 10 according to the first embodiment.
This non-contact three-dimensional measuring apparatus 10 includes two cameras 2a and 2b installed at a distance, a line laser 4 for irradiating the object to be measured 1 with a line laser, and the positions of the two cameras 2a and 2b. An orientation unit 6 that performs external orientation for calculating the relationship, a first image 20 obtained by photographing the measurement object 1 irradiated with the line laser with the first camera 2a, and a second image obtained by the second camera 2b. And an image processing unit 8 that performs image processing on the image 30. For each camera 2a, 2b, it is desirable that internal orientations (internal parameters) such as the lens center, focal length, and distortion correction coefficient are calculated in advance. Further, the photographed image is preferably an image whose distortion has been corrected using the internal orientation data, but it is not essential. Further, the line laser 4 can scan the entire surface of the measurement object 1. Further, the orientation unit 6 performs external orientation relating to the positional relationship between the first camera 2a and the second camera 2b. The image processing unit 8 selects a rectangular area including the first line segment data 22 corresponding to the captured line laser on the first image 20 as the first target area 26, and the first area 26 in the first target area 26 is selected. One point in one line segment data is selected as the feature point 24. Next, the image processing unit 8 selects a rectangular area including the second line segment data 32 corresponding to the captured line laser 4 on the second image 30 as the second target area 36 and corresponds to the feature point 24. The epipolar line 33 to be calculated is calculated in the second target area 36 on the second image 30, and the second line segment corresponding to the captured line laser in the second target area 36 on the second image 30 is calculated. The intersection of the data 32 and the epipolar line 33 is determined as a corresponding point 34 corresponding to the feature point 24 selected in the first target area 26 on the first image 20. Further, the image processing unit 8 uses the feature points 24 in the first target area 26 on the first image 20 and the corresponding points 34 in the second target area 36 on the second image 30 to use the feature points. The three-dimensional coordinates of the corresponding point 14 on the measurement object 1 corresponding to the point and the corresponding point are calculated.

FIG. 7 is a conceptual diagram for explaining the characteristics of the non-contact three-dimensional measuring apparatus 10 according to the first embodiment. FIG. 7 shows in detail the positional relationship related to photogrammetry in the configuration of the non-contact three-dimensional measuring apparatus 10 of FIG.
<About two cameras and images>
The lens centers of the two cameras (first camera 2a and second camera 2b) are represented as point A and point B, respectively.
In the first image 20 by the first camera 2a, the first target region 26 including the first line segment data 22 which is a portion corresponding to the line laser irradiated on the measurement object 1 is selected (FIG. 3). The first target region 26 is also called a region of interest (ROI), and is selected so as to include the first line segment data in the entire first image 20. This is a necessary and sufficient range for selecting the feature points 24 to be included. One feature point 24 is selected from the first line segment data 22 in the first target region. The lens center, focal length, distortion correction coefficient, and the like of the first camera 2a are obtained by internal orientation. The internal orientation of the camera can be calculated based on the method described in Non-Patent Document 5, for example.
Further, also in the second image 30 by the second camera, the second target region 36 including the second line segment data 32 which is a portion corresponding to the line laser irradiated on the object 1 to be measured is selected (FIG. 4). ). Since the second target area 36 is selected so as to include the second line segment data 32 in the entire second image 30, the corresponding point 34 that is the intersection of the epipolar line 33 and the second line segment data 32. This is a necessary and sufficient range. In the second target region 36, a corresponding point 34 that is an intersection of the epipolar line 33 and the second line segment data 32 corresponding to the feature point 24 selected from the first line segment data is determined. Note that the lens center, focal length, distortion correction coefficient, and the like of the second camera 2b are obtained by internal orientation. The epipolar line can be calculated based on the method described in Non-Patent Document 4.

<Line laser>
The line laser 4 may be any laser that can irradiate the object 1 with a line laser. Further, it is preferable that the line laser 4 can scan the entire object to be measured 1. Alternatively, the line laser 4 may be capable of rotating a line laser. Further, the line laser 4 may use any color line laser such as infrared, red, green, and blue. Further, as the line laser 4, lasers of a plurality of colors may be used in combination. By using multiple color lasers in combination, feature points and corresponding points can be calculated for each color laser, so multiple line segments can be extracted simultaneously with a single shot, and 3D coordinates can be calculated at a higher speed. can do.

<About the object to be measured>
In addition, a corresponding point 14 on the measured object 1 corresponding to the portion 12 of the line laser irradiated on the measured object 1, the feature point 24, and the corresponding point 34 is shown. This corresponding point 14 is also shown as point C.

Note that, as shown in the schematic diagram of FIG. 9, the first camera 2a and the second camera 2b have the captured image data in a direction intersecting with a straight line connecting the first camera 2a and the second camera. An image memory for recording along the scanning lines may be provided and output as the first image 20 and the second image 30, respectively. 9 is different from the case of FIG. 7 in that the first image 20 and the second image 30 are represented in the vertical position, and the first image 20 and the second image 30 are different from each other. The difference is that the longitudinal direction intersects the straight line connecting the two cameras.
The image processing unit 8 also includes the first target region 26 and the second target so that the long sides of the rectangular region are parallel to the scanning line direction when the first image 20 and the second image 30 are recorded. Region 36 may be selected. Furthermore, the image processing unit 8 can select the feature point 24 only in the first target area 26 and determine the corresponding point only in the second target area 36.
Furthermore, the line laser 4 may irradiate the object to be measured 1 with the line laser so as to intersect a straight line connecting the first camera 2a and the second camera 2b. As shown in FIG. 9, the epipolar line extends in substantially the same direction as a straight line connecting the first camera 2a and the second camera 2b. Therefore, the epipolar line and the second line segment data 32 can be crossed by irradiating the line laser so as to cross the straight line connecting the first camera 2a and the second camera 2b as described above. .

<About photogrammetry>
a) The positional relationship between the lens center A of the first camera 2a and the lens center B of the second camera 2b and the direction of each camera are obtained by external orientation. The external orientation between the two cameras 2a and 2b can be calculated by the method described in Non-Patent Document 1, for example.
b) Further, the lens centers, focal lengths, distortion correction coefficients, etc. of the first camera 2a and the second camera 2b can be obtained by using internal orientation.
c) Next, the point C of the corresponding point 14 on the DUT 1 extends from the lens center A of the first camera 2a toward the feature point 24 in the first target region 26 on the first image 20. And on a line extending from the lens center B of the second camera 2b toward the corresponding point 34 in the second target area 36 on the second image 30. Therefore, the point C is obtained as an intersection of these by the principle of photogrammetry. Note that the three-dimensional coordinates of the corresponding point can be calculated by the method described in Non-Patent Document 1, for example.
As described above, the corresponding point 14 (C) on the object to be measured 1 can be three-dimensionally measured.

  According to the non-contact three-dimensional measuring apparatus 10 according to the first embodiment of the present invention, by irradiating the measurement object 1 with a line laser, even if the measurement object 1 has no characteristics, the third order is based on the principle of photo measurement. The original coordinates can be measured. In addition, due to the coherent nature of the laser, even if the object to be measured has a shape with irregularities, the line is less likely to be blurred, and the feature points can be extracted accurately. For this reason, the extraction accuracy of the corresponding points is increased, and the accuracy of the three-dimensional measurement points is inevitably increased.

  Further, this non-contact three-dimensional measuring apparatus 10 corresponds to a line laser imaged on one image 20 of two images obtained by imaging the object to be measured irradiated with the line laser with two cameras 2a and 2b. A rectangular area including the first line segment data 22 that is a portion to be selected is selected as the first target area 26 (FIG. 3), and one point of the first line segment data in the first target area 26 is set as the feature point 24. Selected (FIG. 5). In this case, a rectangular area including the second line segment data 32 which is a part corresponding to the captured line laser on the other image 30 is selected as the second target area (FIG. 4), and the second target area 36 is selected. An epipolar line 33 corresponding to the feature point 24 selected in the first target region 26 is calculated, and an intersection point between the epipolar line 33 and the second line segment data 32 is a corresponding point 34 corresponding to the feature point 24. (FIG. 6). The epipolar line can be calculated by the method described in Non-Patent Document 4, for example. As a result, the corresponding points can be clearly determined without including ambiguity as compared with the case where the corresponding points are specified by the correlation with the pattern irradiated on the measurement object 1, and the accuracy of the three-dimensional measurement is improved. Can be improved. Furthermore, in the non-contact three-dimensional measuring apparatus 10 according to the present invention, the first image 20 and the second image that are necessary and sufficient for calculating the feature points and the corresponding points are represented by the first image 20 and the second image. Each of 30 is selected. Accordingly, by calculating the feature points 24 and the corresponding points 34 only for the first target region 26 and the second target region 36, respectively, the three-dimensional measurement can be performed faster than the case of calculating over the entire two images. It can be carried out.

FIG. 8 is a flowchart of the non-contact three-dimensional measurement method of the present invention. This non-contact three-dimensional measurement method is executed as follows.
(A) Two cameras (a first camera 2a and a second camera 2b) that photograph the measurement object 1 are arranged at a distance from each other in a range where the measurement object 1 is shown (FIGS. 1 and 2). 2). For each camera 2a, 2b, it is desirable that internal orientations (internal parameters) such as the lens center, focal length, and distortion correction coefficient are calculated in advance. In addition, when the internal orientation of the camera is not performed, for example, the internal orientation can be obtained by a calculation method described in Non-Patent Document 5.
(B) About the 1st camera 2a and the 2nd camera 2b, external orientation regarding a mutual positional relationship is performed (S01).
With this external orientation, the positional relationship between a plurality of cameras can be known. For example, as a method described in Non-Patent Document 1, external orientation can be performed by extracting eight or more previously known corresponding points around the measurement object 1.

(C) The line laser 4 is irradiated to the object 1 to be measured 1 (S02) (FIG. 1). By using a line laser, it is possible to measure three-dimensional coordinates based on the principle of photo measurement even if the non-measurement object 1 has no characteristics. In addition, due to the coherent nature of the laser, even if the object to be measured has a shape with irregularities, the line is less likely to be blurred, and the feature points can be extracted accurately.
(D) The measurement object 1 irradiated with the line laser is photographed by the first camera 2a and the second camera 2b, respectively, and the first image (FIG. 3) by the first camera 2a and the second A second image (FIG. 4) is obtained by the camera 2b (S03).
(E) In the first image 20, a rectangular area including a portion where the line laser is photographed is selected as the first target area 26 (S04) (FIG. 3).
(F) In the second image 30, a rectangular area including a portion where the line laser is photographed is selected as the second target area 36 (S05) (FIG. 4).
(G) In the first target area 26, a portion corresponding to the captured line laser is extracted as the first line segment data 22 (S06) (FIG. 3).
(H) In the second target area 36, a portion corresponding to the captured line laser is extracted as the second line segment data 32 (S07) (FIG. 4).

(I) One point in the first line segment data 22 in the first target region 26 is selected as the feature point 24 (S08) (FIG. 5). Note that the feature points 24 are sequentially selected over the entire first line segment data 22, and each time the feature point 24 is selected, the following (j) to (l) are repeated, whereby the first line segment 22 Three-dimensional measurement can be performed for the whole.
(J) The epipolar line 33 corresponding to the feature point 24 is calculated in the second target region 36 on the second image (S09) (FIG. 6). The epipolar line is a virtual line connecting the lens center of the first camera 2a and a point on the measurement object 1 corresponding to the feature point 24 on the first image 20 to the second camera. This is the projected line segment. How the epipolar line 33 is calculated in the second target region 36 on the second image 30 depends on the internal orientation of the first camera 2a and the second camera 2b, the first camera 2a and the second camera 30b. It can be determined by external orientation that calculates the positional relationship between the two cameras 2b. The epipolar line can be calculated by the method described in Non-Patent Document 4, for example. There is a corresponding point corresponding to the feature point 24 somewhere on the epipolar line 33 calculated in the second target region 36 on the second image 30.
(K) The intersection of the second line segment data 32 and the epipolar line 33 in the second target region 36 is determined as a corresponding point 34 corresponding to the feature point 24 selected in the first target region 26 (S10) (FIG. 6).
(L) Three-dimensional of the corresponding point 14 on the measurement object 1 corresponding to the feature point and the corresponding point using the feature point 24 in the first target region 26 and the corresponding point 34 in the second target region 36 Coordinates are calculated (S11). The three-dimensional coordinates of the corresponding point 14 can be calculated by the method described in Non-Patent Document 1, for example.
(M) Steps (i) to (l) are repeated for all points on the first line segment in the first target region 26 on the first image 20, and the line laser of the object to be measured is Three-dimensional coordinates are calculated for the irradiated line portion (S12).
Through the above steps, the three-dimensional coordinates of the corresponding point 14 of the measurement object 1 corresponding to the feature point 24 and the corresponding point 34 can be obtained.

  According to the non-contact three-dimensional measurement method according to the first embodiment of the present invention, the line laser is irradiated onto the object 1 to be measured, and the 3D based on the principle of photo measurement even if the object 1 has no characteristics. Coordinates can be measured. In addition, due to the coherent nature of the laser, even if the object to be measured has a shape with irregularities, the line is less likely to be blurred, and the feature points can be extracted accurately. For this reason, the extraction accuracy of the corresponding points is increased, and the accuracy of the three-dimensional measurement points is inevitably increased.

  Moreover, in this contacted three-dimensional measurement method, it corresponds to the line laser image | photographed on one image 20 among the two images which image | photographed the to-be-measured object irradiated with the line laser with the two cameras 2a and 2b. A rectangular region including the first line segment data 22 that is a part is selected as the first target region 26, and one point of the first line segment data 22 in the first target region 26 is selected as the feature point 24. In this case, a rectangular area including the second line segment data 32 that is a part corresponding to the captured line laser on the other image 30 is selected as the second target area 36, and the above-described second target area 36 is filled with the above-described area. The epipolar line 33 corresponding to the feature point 24 selected in the first target region 26 is calculated, and the intersection of the epipolar line 33 and the second line segment data 32 is specified as the corresponding point 34 corresponding to the feature point 24. be able to. As a result, the corresponding points can be clearly determined without including ambiguity as compared with the case where the corresponding points are specified by the correlation with the pattern irradiated on the measurement object 1, and the accuracy of the three-dimensional measurement is improved. Can be improved. Furthermore, in the non-contact three-dimensional measurement method according to the present invention, first and second target areas necessary and sufficient for calculating feature points and corresponding points are selected from the first image and the second image, respectively. doing. As a result, the feature points and the corresponding points are calculated only for each of the first target region and the second target region, so that the three-dimensional measurement can be performed faster than the case of calculating over the entire two images. .

  In the image photographing step, the image data photographed by the first camera 2a and the second camera 2b are scanned along a scanning line in a direction intersecting with a straight line connecting the first camera 2a and the second camera 2b. You may record and output as the 1st image 20 and the 2nd image 30, respectively.

  In the first target region selection step, the first target region 26 may be selected so that the long side of the rectangular region is parallel to the direction of the scanning line when the first image 20 is recorded. Further, in the second target area selection step, the second target area 36 may be selected so that the long side of the rectangular area is parallel to the direction of the scanning line when the second image 30 is recorded.

  Furthermore, in the feature point selection step, the feature point 24 can be selected only in the first target region 26, and in the corresponding point determination step, the corresponding point 34 can be determined only in the second target region 36.

  Further, in the line laser irradiation step, the line laser may be irradiated to the measurement object 1 so as to intersect with a straight line connecting the first camera 2a and the second camera 2b.

(Modification)
FIG. 10 is a flowchart of a non-contact three-dimensional measurement method according to a modification of the first embodiment. The non-contact three-dimensional measurement method of this modification is different from the non-contact three-dimensional measurement method according to Embodiment 1 in FIG. 8 in that the line laser is scanned over the entire object to be measured.
Since (a) and (b) are the same as the non-contact three-dimensional measurement method according to Embodiment 1, the description thereof is omitted.
(C) A line laser is scanned over the DUT 1 (S13). Here, the line laser 4 is irradiated in an arbitrary direction, and the entire object to be measured 1 is sequentially scanned.
Since (d) to (k) are the same as the non-contact three-dimensional measurement method according to Embodiment 1, the description thereof is omitted.
(L) It is confirmed whether the entire measurement object 1 has been scanned (S14). When the entire object to be measured 1 is not scanned, (c) to (k) are repeated to perform three-dimensional measurement on the entire object to be measured 1.
Through the above steps, three-dimensional measurement can be performed over the entire object to be measured 1.

(Embodiment 2)
FIG. 11 is a flowchart of the non-contact three-dimensional measurement method according to the second embodiment. In the non-contact three-dimensional measurement method of this modification, as compared with the non-contact three-dimensional measurement method according to Embodiment 1 in FIG. It differs in the point which irradiates.
Since (a) and (b) are the same as the non-contact three-dimensional measurement method according to Embodiment 1, the description thereof is omitted. In the second embodiment, the line laser irradiation in step (c) is performed as follows.
(C) The line laser is rotated about a predetermined axis to irradiate the measurement object 1 (S15). In the modification of the first embodiment, the line laser is scanned in an arbitrary direction and scanned over the entire object to be measured 1. However, the line laser 4 is also rotated about a predetermined axis to measure the object 1 The whole can be scanned.
Since (d) to (m) are the same as the non-contact three-dimensional measurement method according to Embodiment 1, the description thereof is omitted.
(N) It is confirmed whether the line laser is rotated at all angles for irradiation (S16). When the entire object to be measured 1 is not scanned, (c) to (k) are repeated to perform three-dimensional measurement on the entire object to be measured 1.
Through the above steps, three-dimensional measurement can be performed over the entire object to be measured 1.

(Embodiment 3)
FIG. 12 is a flowchart of the non-contact three-dimensional measurement method according to the third embodiment. In the non-contact three-dimensional measurement method of this modification, when compared with the non-contact three-dimensional measurement method according to Embodiment 1 in FIG. 8, the intersection of the second line segment data 32 and the epipolar line 33 is a single point in (i). When there are two or more intersections instead of one, the line laser is different in that the line laser is irradiated while being rotated around the corresponding point 14 on the measurement object 1 corresponding to the feature point 24.
Since (a) to (j) are the same as the non-contact three-dimensional measurement method according to Embodiment 1, the description thereof is omitted.
Before step (k), it is confirmed on the second image 30 whether there is a single intersection between the epipolar line 33 and the second line segment 32 (S17). When there is one intersection point, (k) to (m) of the non-contact three-dimensional measurement method according to Embodiment 1 are performed, and the third order of the corresponding point 14 of the measurement object 1 corresponding to the feature point and the corresponding point Original coordinates can be obtained.

  On the other hand, when there are a plurality of intersection points, for example, there are a plurality of intersection points when the second line segment data 32 and the epipolar line 33 corresponding to the imaged line laser extend along the same direction. Cases arise. In this case, the line laser 4 is irradiated while being rotated around the corresponding point 14 on the measurement object 1 corresponding to the feature point 24 (S18). In this case, it is preferable to rotate the line laser 4 so that the epipolar line 33 and the line direction of the line laser form an angle. More preferably, the line laser 4 is rotated so that the epipolar line 33 and the line direction of the line laser are orthogonal to each other. Further, the corresponding point 14 of the measurement object 1 corresponding to the feature point 24 is rotated so as to hold the feature point 24. Thereafter, (d) to (j) can be repeated so that the intersection of the epipolar line 33 and the second line segment 32 becomes a single point.

  As a result, even when the epipolar line 33 and the second line segment data 32 are substantially parallel, for example, the corresponding point 34 cannot be specified as one, the line laser 4 is rotated, and the epipolar line 33 and the second line are rotated. A single corresponding point 34 is obtained as an intersection where the minute data 32 intersects, and the corresponding point can be clearly identified, and the accuracy of the three-dimensional measurement can be improved.

  If no intersection between the epipolar line 33 and the second line segment data 32 is obtained, the corresponding point 14 on the measurement object 1 corresponding to the feature point 24 selected on the first image 20 is obtained. This means that it does not appear in the second target area 36 on the second image 30. In this case, the range of the measurement object 1 included in the first target area 26 on the first image 20 from the first camera 2a and the second image 30 from the second camera 2b. It is considered that the range of the measured object 1 that is common with the range of the measured object 1 included in the second target region 36 is small or does not exist. Therefore, the distance between the first camera 2a and the second camera 2b is reduced, and the first target area 26 on the first image 20 and the second target area 36 on the second image 30 are shared. It is preferable that the range of the measurement object 1 is obtained.

(Embodiment 4)
FIG. 13 is a schematic diagram illustrating a configuration of the non-contact three-dimensional measurement apparatus 100 according to the fourth embodiment. Compared with the non-contact three-dimensional measuring apparatus 100 according to the first embodiment, the non-contact three-dimensional measuring apparatus 100 includes a first set of cameras in addition to a first set of camera systems including two cameras 2a and 2b. The camera system further includes a second set of camera systems including two cameras (a third camera 2c and a fourth camera 2d) that capture the object 1 to be measured from a location different from the system. The second set of camera systems may be physically prepared separately from the first set of camera systems. However, after the first set of camera systems is shot and processed at the first position, The pair of camera systems may be moved as they are to a location different from the initial position, and shooting and image processing may be performed as the second set of camera systems. In addition, this non-contact three-dimensional measuring device may include at least one line laser 4. Alternatively, two sets of line lasers may be prepared separately for the first set of camera systems and the second set of camera systems.

  FIG. 20 is a conceptual diagram illustrating the characteristics of the non-contact three-dimensional measurement method using the first set of camera systems of the non-contact three-dimensional measurement apparatus 100 according to the fourth embodiment. FIG. 20 shows in detail the positional relationship related to photogrammetry by the first set of camera systems in the configuration of the non-contact three-dimensional measuring apparatus 100 of FIG. FIG. 25 is a conceptual diagram illustrating features of the non-contact three-dimensional measurement method using the second set of camera systems of the non-contact three-dimensional measurement apparatus 100 according to the fourth embodiment. FIG. 25 shows in detail the positional relationship related to photogrammetry by the second set of camera systems in the configuration of the non-contact three-dimensional measuring apparatus 100 of FIG.

Further, the non-contact three-dimensional measurement apparatus 100 is different from the non-contact three-dimensional measurement apparatus according to the first embodiment in the image processing method in the image processing unit. Specifically, this image processing unit is different from the first embodiment in that the target region (ROI) is not selected from each image. Note that the target region (ROI) may be selected from each image as in the first embodiment. In addition, the image processing unit performs the following processing.
(1) First, as in the first embodiment, a three-dimensional coordinate system on the object to be measured in the first set of camera systems is obtained. For example, the three-dimensional coordinate by the first set of camera systems is M (x, y, z).
(2) Next, a three-dimensional coordinate system on the object to be measured in the second set of camera systems is obtained. For example, let N (x, y, z) be a three-dimensional coordinate by the second set of camera systems.
(3) Thereafter, among the three-dimensional coordinates on the object to be measured in the second set of camera systems, at least three points corresponding to the three-dimensional coordinates on the object to be measured in the first set of camera systems, that is, at least three points Select the 3D coordinates of the marker. For example, the three-dimensional coordinate of the i-th marker by the first set of camera systems is A _i (x, y, z), and the three-dimensional coordinate of the i-th marker by the second set of camera systems is B _i (x, y, z).
(4) For at least three selected three-dimensional coordinates, a coordinate conversion relationship from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems is obtained. This three-dimensional coordinates B _i of i-th marker by the second set of camera system _(x, y, z) of the first set of camera system according to the i-th three-dimensional coordinates A _{i (x} markers, y, The coordinate transformation relationship corresponding to z) is obtained. This is because, for at least three selected three-dimensional coordinates, the three-dimensional coordinate system M (x, y, z) of the first camera system is changed from the three-dimensional coordinate system N (x, y, z) of the second camera system. Based on the relation of similarity transformation to z), the coordinate transformation relation can be obtained. More specifically, the similarity transformation is represented by the following formula, for example.
M _i = sR (N _i −t)
Mi: 3D coordinates (coordinate vector) of the first set of camera systems
Ni: 3D coordinates (coordinate vector) of the second set of camera systems
s: Scale (scalar value)
R: rotation matrix t: translation vector (5) and the obtained coordinate transformation relationship, in the case of the above example, using the scale s, rotation matrix R, translation vector t, etc., the three-dimensional coordinates of the second set of camera systems The system N (x, y, z) is coordinate-transformed into a three-dimensional coordinate system M (x, y, z) of the first set of camera systems, and the measurement target is based on the three-dimensional coordinate system of the first set of camera systems. Get the 3D coordinates of an object.

  For example, by placing three or more markers 60 on the object 1 to be measured, the three-dimensional coordinates common to the three-dimensional coordinates in the first camera system and the three-dimensional coordinates in the second camera system are common. As coordinates, the three-dimensional coordinates of the marker 60 can be made to correspond (FIG. 13). Note that it is not necessary to place a marker when at least three or more points having a clear correspondence between the three-dimensional coordinates of the first camera system and the three-dimensional coordinates of the second camera system can be used. In addition, it is preferable to use at least three markers that are not arranged on the same straight line.

  In the non-contact three-dimensional measurement apparatus 100 according to the fourth embodiment, not only the first set of camera systems but also the second set of camera systems is used. For example, in the case of a large object to be measured such as a mountain, only one surface of the mountain can be observed with only the first set of camera systems. However, according to the non-contact three-dimensional measurement apparatus according to Embodiment 4, the second set With this camera system, it is possible to observe a portion that cannot be observed in the shadow with the first set of camera systems. Thereafter, the entire three-dimensional coordinate system can be integrated by converting the three-dimensional coordinate system of the second set of camera systems into the first set of three-dimensional coordinate systems.

  The non-contact three-dimensional measurement apparatus according to the fourth embodiment includes any one of the configurations of the non-contact three-dimensional measurement apparatus according to the first to third embodiments or a plurality of these configurations. May be.

14 and 15 are flowcharts of the non-contact three-dimensional measurement method according to the fourth embodiment. This non-contact three-dimensional measurement method performs calculation of the three-dimensional coordinate system of the object to be measured by the first set of camera systems and calculation of the three-dimensional coordinate system of the object to be measured by the second set of camera systems. The obtained three-dimensional coordinate system of the second set of camera systems is transformed into the three-dimensional coordinate system of the first set of camera systems, and the object to be measured based on the three-dimensional coordinate system of the first set of camera systems is converted It is characterized by obtaining three-dimensional coordinates.
FIG. 16 is a schematic diagram showing a first image 20 taken by the first camera 2a of the first set of camera systems. FIG. 17 is a schematic diagram illustrating a second image 30 captured by the second camera 2b of the first set of camera systems. FIG. 18 is a schematic diagram when one point in the first line segment data 22 is selected as the feature point 24 for the first image 20 in FIG. 16. FIG. 19 is a schematic diagram showing the second image 30 in which the epipolar line 33 corresponding to the feature point 24 on the first image 20 is calculated in the second image 30 of FIG.
(A) For the first camera 2a and the second camera 2b that are arranged at a distance from each other and that constitute the first set of camera systems for photographing the object 1 to be measured, external orientation relating to the mutual positional relationship is performed ( S21).
(B) The line laser is irradiated to the measurement object 1 in the first set of camera systems (S22) (FIG. 20).
(C) The measurement object 1 irradiated with the line laser is photographed by the first camera 2a and the second camera 2b, respectively, and the first image 20 (FIG. 16) by the first camera 2a and the second The second image 30 (FIG. 17) obtained by the camera 2b is obtained (S23).
(D) In the first image 20, a portion corresponding to the captured line laser is extracted as the first line segment data 22 (S24) (FIG. 16).
(E) In the second image 30, a portion corresponding to the captured line laser is extracted as second line segment data 32 (S25) (FIG. 17).
(F) One point in the first line segment data 22 extracted from the first image 20 is selected as the feature point 24 (S26) (FIG. 18).
(G) The epipolar line 33 corresponding to the feature point 24 selected on the first image 20 is calculated on the second image 30 (S27) (FIG. 19).
(H) The intersection of the second line segment data 32 extracted from the second image 30 and the epipolar line 33 is determined as a corresponding point 34 corresponding to the feature point 24 selected on the first image 20 (S28). (FIG. 19).
(I) Three-dimensional of the corresponding point 14 on the measurement object 1 corresponding to the feature point 24 and the corresponding point 34 using the feature point 24 in the first image 20 and the corresponding point 34 in the second image 30. The coordinates are calculated to obtain a three-dimensional coordinate system on the measurement object 1 in the first set of camera systems (S29) (FIG. 20).

FIG. 21 is a schematic diagram showing a third image 40 taken by the third camera 2c of the second set of camera systems. FIG. 22 is a schematic diagram showing a fourth image 50 taken by the fourth camera 2d of the second set of camera systems. FIG. 23 is a schematic diagram when one point in the third line segment data 42 is selected as the feature point 44 for the third image 40 in FIG. FIG. 24 is a schematic diagram showing the fourth image 50 in which the epipolar line 53 corresponding to the feature point 44 on the third image 40 is calculated in the fourth image 50 of FIG.
(J) A third camera 2c and a fourth camera 2d, which are arranged at a distance from each other at positions different from the first set of camera systems and constitute a second set of camera systems for photographing the object 1 to be measured. Are subjected to external orientation with respect to each other's positional relationship (S30).
(K) Irradiate the measurement object 1 with a line laser (S31)
(L) The measurement object 1 irradiated with the line laser is photographed by the third camera 2c and the fourth camera 2d, respectively, and the third image 40 (FIG. 21) by the third camera 2c and the fourth A fourth image 50 (FIG. 22) obtained by the camera 2d is obtained (S32).
(M) In the third image 40, a portion corresponding to the captured line laser is extracted as third line segment data 42 (S33) (FIG. 21).
(N) In the fourth image 50, a portion corresponding to the captured line laser is extracted as the fourth line segment data 52 (S34) (FIG. 22).
(O) One point in the third line segment data 42 extracted from the third image 40 is selected as the feature point 44 (S35) (FIG. 23).
(P) The epipolar line 53 corresponding to the feature point 44 selected on the third image 40 is calculated on the fourth image 50 (S36) (FIG. 24).
(Q) The intersection of the fourth line segment data 52 extracted from the fourth image 50 and the epipolar line 53 is determined as a corresponding point 54 corresponding to the feature point 44 selected on the third image 40 (S37). (FIG. 24).
(R) Using the feature point 44 in the third image 40 and the corresponding point 54 in the fourth image 50, the three-dimensional coordinates of the corresponding point on the measurement object 1 corresponding to the feature point 44 and the corresponding point 54 To obtain a three-dimensional coordinate system on the object to be measured in the second set of camera systems (S38) (FIG. 25).

(S) Of the three-dimensional coordinates on the measurement object 1 in the second set of camera systems, select at least three three-dimensional coordinates corresponding to the three-dimensional coordinates on the measurement object in the first set of camera systems; For at least three selected three-dimensional coordinates, a coordinate conversion relationship from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems is obtained (S39). This three-dimensional coordinates B _i of i-th marker by the second set of camera system _(x, y, z) of the first set of camera system according to the i-th three-dimensional coordinates A _{i (x} markers, y, The coordinate transformation relationship corresponding to z) is obtained. This is because, for at least three selected three-dimensional coordinates, the three-dimensional coordinate system M (x, y, z) of the first camera system is changed from the three-dimensional coordinate system N (x, y, z) of the second camera system. Based on the relation of similarity transformation to z), the coordinate transformation relation can be obtained. More specifically, the similarity transformation is represented by the following formula, for example.
M _i = sR (N _i −t)
Mi: 3D coordinates (coordinate vector) of the first set of camera systems
Ni: 3D coordinates (coordinate vector) of the second set of camera systems
s: Scale (scalar value)
R: rotation matrix t: translation vector (t) The obtained coordinate transformation relationship, for example, the scale s, the rotation matrix R, the translation vector t, etc. The coordinate transformation is performed to the three-dimensional coordinate system of the first set of camera systems, and the three-dimensional coordinates of the measurement object based on the three-dimensional coordinate system of the first set of camera systems are obtained (S40).
As described above, the three-dimensional coordinates of the measurement object based on the three-dimensional coordinate system of the first set of camera systems can be obtained.

The step (s) will be further described.
Below, the relational expression of similarity transformation is listed again.
M _i = sR (N _i −t)
Mi: 3D coordinates (coordinate vector) of the first set of camera systems
Ni: 3D coordinates (coordinate vector) of the second set of camera systems
s: Scale (scalar value)
R: rotation matrix t: translation vector An example of how to calculate the scale s, the rotation matrix R, and the translation vector t related to the coordinate transformation is given.

(A) How to obtain the scale Each of the coordinate vectors between the i-th and i + 1-th three-dimensional coordinates A (x, y, z) of the marker by the first set of camera systems is subtracted, respectively. When the coordinate vector is subtracted between the i-th and i + 1-th three-dimensional coordinates B (x, y, z) of the corresponding marker by the camera system to obtain the square of the magnitude of the vector on both sides, From the relational expression of the similarity transformation, the following formula is obtained.
(A _i −A _{i + 1} ) ² = s ² (B _i −B _{i + 1} ) ²
In the above equation, the translation vector t can be deleted, and the whole can be a scalar only. Thereby, the scale s can be calculated. Since the number of markers is three or more, three or more different scale s values can be obtained, and the average value can be obtained as the scale s.

(B) How to Obtain Rotation Matrix R First, the similarity transformation formula is modified as follows.
M _i = sR (N _i −t) = R (sN _i −st) = R (sN _i ) −R (st)
Next, the marker is subtracted from i + 1 to i-th.
M _{i + 1} −M _i = (R (sN _{i + 1} ) −R (st)) − (R (sNi) −R (st))
M _{i + 1} −M _i = R (sN _{i + 1} ) −R (sN _i )
= R ((sN _{i + 1} )-(sN _i ))
here,
m _i = M _{i + 1} −M _i
n _i = sN _{i + 1} −sN _i
Is replaced by the following formula:
m _i = Rn _i

の条件を満たす回転行列Ｒを求めることである。なお、上記最小二乗法による回転行列Ｒは、通常の方法によって得られる。その求め方の詳細については、例えば、非特許文献３等に詳しい。 When there are only three markers, the rotation matrix R is uniquely determined from the above equation, but when there are multiple markers, it is necessary to consider that measurement errors are included in the three-dimensional coordinates of each marker. There is. Thus, by obtaining a rotation matrix R that minimizes the sum of squares of the subtraction vector (m _i −Rn _i ) from the above equation, the error in coordinate transformation can be minimized. In particular,

Is to obtain a rotation matrix R that satisfies the following condition. The rotation matrix R by the least square method is obtained by a normal method. The details of how to find it are detailed in Non-Patent Document 3, for example.

となる。なお、Ｒ^Ｔは、回転行列Ｒの転置行列である。複数のマーカのそれぞれについて並進ベクトルｔ_ｉが得られるので、その平均値を全体の並進ベクトルｔとする。
以上によって、座標変換関係のスケールｓ、回転行列Ｒ、並進ベクトルｔのそれぞれが得られる。なお、上記の座標変換関係のスケールｓ、回転行列Ｒ、並進ベクトルｔの求め方は一例であってこれらに限られない。 (C) How to determine the translation vector t

It becomes. ^RT is a transposed matrix of the rotation matrix R. Since translation vector t _i is obtained for each of a plurality of markers, and the average value and the overall translation vector t.
As described above, each of the coordinate transformation-related scale s, rotation matrix R, and translation vector t is obtained. The method for obtaining the coordinate conversion-related scale s, rotation matrix R, and translation vector t is an example, and is not limited thereto.

  In the non-contact three-dimensional measurement apparatus and the non-contact three-dimensional measurement method according to Embodiment 4, not only the first set of camera systems but also the second set of camera systems is used. For example, in the case of a large object to be measured such as a mountain, only one surface of the mountain can be observed with only the first set of camera systems, but according to the non-contact three-dimensional measurement apparatus and method according to the fourth embodiment, With the two sets of camera systems, it is possible to observe a portion that cannot be observed in the shadow of the first set of camera systems. Thereafter, the three-dimensional coordinate system of the entire object to be measured can be integrated by performing coordinate conversion of the three-dimensional coordinate system of the second set of camera systems into the first set of three-dimensional coordinate systems.

  According to the non-contact three-dimensional measurement apparatus and the non-contact three-dimensional measurement method according to the present invention, even when an object having no feature on the surface is set as the measurement object, one of the two images is selected. The corresponding point on the other image can be clearly identified for the feature point. Therefore, the non-contact three-dimensional measurement apparatus and the non-contact three-dimensional measurement method according to the present invention are useful for non-contact three-dimensional measurement using an object having no feature on the surface as an object to be measured. Furthermore, in this method, since the arrangement of the camera, the laser, and the like can be arbitrarily arranged, it is possible to easily cope with a change in the size of the measurement object. In addition, since cameras, lasers, and the like have relatively low power consumption, they can be driven by batteries and have excellent portability.

DESCRIPTION OF SYMBOLS 1 Measured object 2a 1st camera 2b 2nd camera 2c 3rd camera 2d 4th camera 4 Line laser 6 Orientation part 6a 1st orientation part 6b 2nd orientation part 8 Image processing part 10 Non-contact three-dimensional measurement Device 12 Irradiated Line Laser 14 Corresponding Point 20 First Image 22 First Line Segment Data 24 Feature Point 26 First Target Area 30 Second Image 32 Second Line Segment Data 33 Epipolar Line 34 Corresponding Point 36 Second Target Region 40 Third image 42 Third line segment data 44 Feature point 50 Fourth image 52 Fourth line segment data 54 Corresponding point 60 Marker 100 Non-contact three-dimensional measuring device

Claims (29)

A first camera and a second camera which are arranged at a distance from each other and shoot a measurement object;
A line laser that irradiates the object to be measured by rotating a line laser about a predetermined axis in an irradiation direction; and
For the first camera and the second camera, an orientation unit that performs external orientation to calculate the mutual positional relationship;
An image processing unit that performs image processing on a first image obtained by photographing the object to be measured irradiated with the line laser with the first camera, and a second image photographed by the second camera;
With
The image processing unit
In the first image, a rectangular area including a portion where the line laser is photographed is selected as a first target area,
In the second image, a rectangular region including a portion where the line laser is photographed is selected as a second target region,
In the first target area, a portion where the line laser is imaged is extracted as first line segment data,
In the second target area, a portion where the line laser is imaged is extracted as second line segment data,
Selecting one point in the first line segment data in the first target area as a feature point;
Calculating an epipolar line corresponding to the feature point selected in the first target region on the first image in the second target region on the second image;
In the second target area on the second image, the intersection of the captured second line segment data corresponding to the line laser and the epipolar line is defined as the first target area on the first image. Determined as a corresponding point corresponding to the feature point selected in
Corresponding to the feature point and the corresponding point using the feature point in the first target region on the first image and the corresponding point in the second target region on the second image Calculate the three-dimensional coordinates of the corresponding point on the object to be measured,
A line on which the line laser of the object to be measured is irradiated by repeating the process of calculating the three-dimensional coordinates of the corresponding points for all points on the first line segment extracted from the first image. A non-contact three-dimensional measuring device that calculates three-dimensional coordinates for a part,
The non-contact three-dimensional measurement apparatus according to claim 1, wherein the image processing unit repeats the image processing every time the line laser is rotated about a predetermined axis in an irradiation direction to irradiate the object to be measured. A first camera and a second camera which are arranged at a distance from each other and shoot a measurement object;
A line laser that irradiates the object to be measured by rotating a line laser about an axis in an irradiation direction; and
For the first camera and the second camera, an orientation unit that performs external orientation to calculate the mutual positional relationship;
An image processing unit that performs image processing on a first image obtained by photographing the object to be measured irradiated with the line laser with the first camera, and a second image photographed by the second camera;
With
The image processing unit
In the first image, a rectangular area including a portion where the line laser is photographed is selected as a first target area,
In the second image, a rectangular region including a portion where the line laser is photographed is selected as a second target region,
In the first target area, a portion where the line laser is imaged is extracted as first line segment data,
In the second target area, a portion where the line laser is imaged is extracted as second line segment data,
Selecting one point in the first line segment data in the first target area as a feature point;
Calculating an epipolar line corresponding to the feature point selected in the first target region on the first image in the second target region on the second image;
In the second target area on the second image, the intersection of the captured second line segment data corresponding to the line laser and the epipolar line is defined as the first target area on the first image. Determined as a corresponding point corresponding to the feature point selected in
Corresponding to the feature point and the corresponding point using the feature point in the first target region on the first image and the corresponding point in the second target region on the second image A non-contact three-dimensional measuring device that calculates three-dimensional coordinates of a corresponding point on the object to be measured,
In the process of determining corresponding points in the image processing unit, when there are two or more intersections between the epipolar line and the second line segment data, the line laser corresponds to the feature point on the first image. Irradiating by rotating about the axis of the irradiation direction to the corresponding point on the object to be measured, and setting the single intersection point of the epipolar line and the second line segment data on the second image as the feature point It is determined as a corresponding point corresponding to
Non-contact 3D measuring device.   The first camera and the second camera each include an image memory that records captured image data along a scanning line in a direction intersecting with a straight line connecting the first camera and the second camera. The non-contact three-dimensional measurement apparatus according to claim 1, wherein the non-contact three-dimensional measurement apparatus outputs the first image and the second image, respectively.   The image processing unit includes the first target region and the second target so that a long side of the rectangular region is parallel to a direction of a scanning line when the first image and the second image are recorded. The non-contact three-dimensional measuring apparatus according to claim 3, wherein a region is selected.   The non-processing according to any one of claims 1 to 4, wherein the image processing unit selects the feature points only within the first target region and determines the corresponding points only within the second target region. Contact three-dimensional measuring device.   The said line laser irradiates the said to-be-measured object with a line laser so that it may cross | intersect the straight line which connects a said 1st camera and a said 2nd camera, The non-according to any one of Claim 1 to 5 Contact three-dimensional measuring device. (A) an external orientation step for performing external orientation on the positional relationship between the first camera and the second camera, which are arranged at a distance from each other and shoot the object to be measured;
(B) a line laser irradiation step of irradiating the object to be measured with a line laser;
(C) The object to be measured irradiated with the line laser is photographed by the first camera and the second camera, respectively, and the first image by the first camera and the second image by the second camera are taken. An image capturing step for obtaining two images;
(D) a first target region selection step of selecting, as the first target region, a rectangular region including a portion where the line laser is captured in the first image;
(E) a second target region selection step of selecting, as the second target region, a rectangular region including a portion where the line laser is captured in the second image;
(F) a first line segment data extraction step of extracting a portion where the line laser is photographed as first line segment data in the first target area;
(G) a second line segment data extracting step of extracting a portion where the line laser is photographed as second line segment data in the second target area;
(H) a feature point selection step of selecting one point in the first line segment data in the first target region as a feature point;
(I) an epipolar line calculating step for calculating an epipolar line corresponding to the feature point selected in the first target area on the first image in the second target area on the second image;
(J) In the second target region on the second image, an intersection point of the second line segment data corresponding to the captured line laser and the epipolar line is defined as the intersection on the first image. A corresponding point determining step for determining as a corresponding point corresponding to the feature point selected in the first target region;
(K) The feature points and the correspondences using the feature points in the first target region on the first image and the corresponding points in the second target region on the second image. A three-dimensional coordinate calculation step for calculating a three-dimensional coordinate of a corresponding point on the measurement object corresponding to a point;
(L) Steps (h) to (k) are repeated for all points on the first line segment extracted from the first image, and the line laser of the object to be measured is irradiated. A three-dimensional coordinate calculation step for calculating three-dimensional coordinates for the line part;
A non-contact three-dimensional measurement method including
The line laser irradiation step (b) further includes a line laser scanning step in which the line laser is rotated about a predetermined axis in an irradiation direction to irradiate the object to be measured, and the line laser is rotated to rotate the object to be measured. The non-contact three-dimensional measurement method characterized by repeating the step (c) to the step (l) every time irradiation is performed. (A) an external orientation step for performing external orientation on the positional relationship between the first camera and the second camera, which are arranged at a distance from each other and shoot the object to be measured;
(B) a line laser irradiation step of irradiating the object to be measured with a line laser;
(C) The object to be measured irradiated with the line laser is photographed by the first camera and the second camera, respectively, and the first image by the first camera and the second image by the second camera are taken. An image capturing step for obtaining two images;
(D) a first target region selection step of selecting, as the first target region, a rectangular region including a portion where the line laser is captured in the first image;
(E) a second target region selection step of selecting, as the second target region, a rectangular region including a portion where the line laser is captured in the second image;
(F) a first line segment data extraction step of extracting a portion where the line laser is photographed as first line segment data in the first target area;
(G) a second line segment data extracting step of extracting a portion where the line laser is photographed as second line segment data in the second target area;
(H) a feature point selection step of selecting one point in the first line segment data in the first target region as a feature point;
(I) an epipolar line calculating step for calculating an epipolar line corresponding to the feature point selected in the first target area on the first image in the second target area on the second image;
(J) In the second target region on the second image, an intersection point of the second line segment data corresponding to the captured line laser and the epipolar line is defined as the intersection on the first image. A corresponding point determining step for determining as a corresponding point corresponding to the feature point selected in the first target region;
(K) The feature points and the correspondences using the feature points in the first target region on the first image and the corresponding points in the second target region on the second image. A three-dimensional coordinate calculation step for calculating a three-dimensional coordinate of a corresponding point on the measurement object corresponding to a point;
A non-contact three-dimensional measurement method including
In the corresponding point determination step (j), when there are two or more intersections between the epipolar line and the second line segment data, the line laser is detected as the feature point corresponding to the feature point on the first image. Rotate and irradiate a corresponding point on the measurement object with respect to the irradiation direction axis, and a single intersection between the epipolar line and the second line segment data corresponds to the feature point on the second image. A non-contact three-dimensional measurement method, characterized in that it is determined as a corresponding point.   In the image photographing step, image data photographed by the first camera and the second camera is recorded along a scanning line in a direction intersecting with a straight line connecting the first camera and the second camera. The contactless three-dimensional measurement method according to claim 7 or 8, wherein the first image and the second image are output as the first image and the second image, respectively. In the first target area selection step, the first target area is selected so that the long side of the rectangular area is parallel to the direction of the scanning line when the first image is recorded,
The second target region is selected in the second target region selection step so that a long side of the rectangular region is parallel to a scanning line direction when the second image is recorded. The non-contact three-dimensional measurement method described.   The feature point is selected only in the first target region in the feature point selection step, and the corresponding point is determined only in the second target region in the corresponding point determination step. The non-contact three-dimensional measurement method according to any one of the above.   The line laser irradiation step irradiates the object to be measured with a line laser so as to intersect with a straight line connecting the first camera and the second camera. Non-contact 3D measurement method. A first set of camera systems including a first camera and a second camera, which are arranged at a distance from each other and photograph a measurement object;
A second set of camera systems including a third camera and a fourth camera, which are arranged at positions different from the first set of camera systems and at a distance from each other, and shoot the measurement object;
A line laser that irradiates the object to be measured by rotating a line laser about a predetermined axis in an irradiation direction; and
A first orientation unit that performs external orientation on the positional relationship between the first camera and the second camera;
A second orientation unit that performs external orientation on the positional relationship between the third camera and the fourth camera;
A first image obtained by photographing the object to be measured irradiated with the line laser with the first camera and a second image photographed with the second camera are subjected to image processing, and the line laser is emitted. An image processing unit that performs image processing on a third image obtained by photographing the measured object with the third camera and a fourth image obtained by photographing with the fourth camera;
With
The image processing unit
Selecting one point in the first line segment data corresponding to the line laser imaged on the first image as a feature point;
An epipolar line corresponding to the feature point selected on the first image is calculated on the second image;
On the second image, an intersection between the second line segment data corresponding to the captured line laser and the epipolar line is determined as a corresponding point corresponding to the feature point selected on the first image. And
Using the feature points on the first image and the corresponding points on the second image, the three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points are obtained. Calculate
A line on which the line laser of the object to be measured is irradiated by repeating the process of calculating the three-dimensional coordinates of the corresponding points for all points on the first line segment extracted from the first image. Obtaining a three-dimensional coordinate system on the object to be measured in the first set of camera systems for the entire portion;
Selecting one point in the third line segment data corresponding to the line laser imaged on the third image as a feature point;
An epipolar line corresponding to the feature point selected on the third image is calculated on the fourth image;
On the fourth image, an intersection of the fourth line segment data corresponding to the captured line laser and the epipolar line is determined as a corresponding point corresponding to the feature point selected on the third image. And
Using the feature points on the third image and the corresponding points on the fourth image, the three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points are obtained. Calculating and obtaining a three-dimensional coordinate system on the object to be measured in the second set of camera systems;
Of the three-dimensional coordinates on the measurement object in the second set of camera systems, select at least three three-dimensional coordinates corresponding to the three-dimensional coordinates on the measurement object in the first set of camera systems; For the selected at least three three-dimensional coordinates, a coordinate conversion relationship from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems is obtained, and the obtained coordinate conversion relationship And converting the three-dimensional coordinate system of the second set of camera systems into the three-dimensional coordinate system of the first set of camera systems, and based on the three-dimensional coordinate system of the first set of camera systems A non-contact three-dimensional measuring device for obtaining a three-dimensional coordinate of an object to be measured,
The image processing unit
In the first set of camera systems, every time the line laser is rotated about a predetermined axis in the irradiation direction and irradiated to the object to be measured, the entire line portion irradiated with the line laser of the object to be measured is applied. On the other hand, a non-contact three-dimensional measuring apparatus characterized by repeating the process of obtaining a three-dimensional coordinate system on the object to be measured in the first set of camera systems. A first set of camera systems including a first camera and a second camera, which are arranged at a distance from each other and photograph a measurement object;
A second set of camera systems including a third camera and a fourth camera, which are arranged at positions different from the first set of camera systems and at a distance from each other, and shoot the measurement object;
A line laser that irradiates the object to be measured by rotating a line laser about a predetermined axis in an irradiation direction; and
A first orientation unit that performs external orientation on the positional relationship between the first camera and the second camera;
A second orientation unit that performs external orientation on the positional relationship between the third camera and the fourth camera;
A first image obtained by photographing the measurement object irradiated with the line laser with the first camera;
The second image photographed by the second camera is subjected to image processing, and the third image obtained by photographing the measurement object irradiated with the line laser with the third camera and the fourth camera. An image processing unit that performs image processing on the captured fourth image;
With
The image processing unit
Selecting one point in the first line segment data corresponding to the line laser imaged on the first image as a feature point;
An epipolar line corresponding to the feature point selected on the first image is calculated on the second image;
On the second image, an intersection between the second line segment data corresponding to the captured line laser and the epipolar line is determined as a corresponding point corresponding to the feature point selected on the first image. And
Using the feature points on the first image and the corresponding points on the second image, the three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points are obtained. Calculating and obtaining a three-dimensional coordinate system on the object to be measured in the first set of camera systems;
Selecting one point in the third line segment data corresponding to the line laser imaged on the third image as a feature point;
An epipolar line corresponding to the feature point selected on the third image is calculated on the fourth image;
On the fourth image, an intersection of the fourth line segment data corresponding to the captured line laser and the epipolar line is determined as a corresponding point corresponding to the feature point selected on the third image. And
Using the feature points on the third image and the corresponding points on the fourth image, the three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points are obtained. Calculate
A line on which the line laser of the object to be measured is irradiated by repeating the process of calculating the three-dimensional coordinates of the corresponding points for all points on the third line segment extracted from the third image. Obtaining a three-dimensional coordinate system on the object to be measured in the second set of camera systems for the entire portion;
Of the three-dimensional coordinates on the measurement object in the second set of camera systems, select at least three three-dimensional coordinates corresponding to the three-dimensional coordinates on the measurement object in the first set of camera systems; For the selected at least three three-dimensional coordinates, a coordinate conversion relationship from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems is obtained, and the obtained coordinate conversion relationship And converting the three-dimensional coordinate system of the second set of camera systems into the three-dimensional coordinate system of the first set of camera systems, and based on the three-dimensional coordinate system of the first set of camera systems Get the 3D coordinates of the measurement object,
A non-contact three-dimensional measuring device,
The image processing unit
In the second set of camera systems, every time the line laser is rotated about a predetermined axis in the irradiation direction to irradiate the object to be measured, the entire line portion irradiated with the line laser of the object to be measured is applied. On the other hand, the non-contact three-dimensional measurement apparatus characterized by repeating the process of obtaining a three-dimensional coordinate system on the measurement object in the second set of camera systems. A first set of camera systems including a first camera and a second camera, which are arranged at a distance from each other and photograph a measurement object;
A second set of camera systems including a third camera and a fourth camera, which are arranged at positions different from the first set of camera systems and at a distance from each other, and shoot the measurement object;
A line laser that irradiates the object to be measured by rotating a line laser about an axis in an irradiation direction; and
A first orientation unit that performs external orientation on the positional relationship between the first camera and the second camera;
A second orientation unit that performs external orientation on the positional relationship between the third camera and the fourth camera;
A first image obtained by photographing the measurement object irradiated with the line laser with the first camera;
The second image photographed by the second camera is subjected to image processing, and the third image obtained by photographing the measurement object irradiated with the line laser with the third camera and the fourth camera. An image processing unit that performs image processing on the captured fourth image;
With
The image processing unit
Selecting one point in the first line segment data corresponding to the line laser imaged on the first image as a feature point;
An epipolar line corresponding to the feature point selected on the first image is calculated on the second image;
On the second image, an intersection between the second line segment data corresponding to the captured line laser and the epipolar line is determined as a corresponding point corresponding to the feature point selected on the first image. And
Using the feature points on the first image and the corresponding points on the second image, the three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points are obtained. Calculating and obtaining a three-dimensional coordinate system on the object to be measured in the first set of camera systems;
Selecting one point in the third line segment data corresponding to the line laser imaged on the third image as a feature point;
An epipolar line corresponding to the feature point selected on the third image is calculated on the fourth image;
On the fourth image, an intersection of the fourth line segment data corresponding to the captured line laser and the epipolar line is determined as a corresponding point corresponding to the feature point selected on the third image. And
Using the feature points on the third image and the corresponding points on the fourth image, the three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points are obtained. Calculating and obtaining a three-dimensional coordinate system on the object to be measured in the second set of camera systems;
Of the three-dimensional coordinates on the measurement object in the second set of camera systems, select at least three three-dimensional coordinates corresponding to the three-dimensional coordinates on the measurement object in the first set of camera systems; For the selected at least three three-dimensional coordinates, a coordinate conversion relationship from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems is obtained, and the obtained coordinate conversion relationship And converting the three-dimensional coordinate system of the second set of camera systems into the three-dimensional coordinate system of the first set of camera systems, and based on the three-dimensional coordinate system of the first set of camera systems Get the 3D coordinates of the measurement object,
A non-contact three-dimensional measuring device,
In the process of determining the corresponding points of the first set of camera systems in the image processing unit, when there are two or more intersections between the epipolar line and the second line segment data, the line laser is changed to the first laser. Irradiating by rotating about the axis of the irradiation direction to the corresponding point on the measurement object corresponding to the feature point on the image of the image, and the epipolar line and the second line segment data on the second image, A single intersection point is determined as a corresponding point corresponding to the feature point,
Non-contact 3D measuring device. A first set of camera systems including a first camera and a second camera, which are arranged at a distance from each other and photograph a measurement object;
A second set of camera systems including a third camera and a fourth camera, which are arranged at positions different from the first set of camera systems and at a distance from each other, and shoot the measurement object;
A line laser that irradiates the object to be measured by rotating a line laser about an axis in an irradiation direction; and
A first orientation unit that performs external orientation on the positional relationship between the first camera and the second camera;
A second orientation unit that performs external orientation on the positional relationship between the third camera and the fourth camera;
A first image obtained by photographing the object to be measured irradiated with the line laser with the first camera and a second image photographed with the second camera are subjected to image processing, and the line laser is emitted. An image processing unit that performs image processing on a third image obtained by photographing the measured object with the third camera and a fourth image obtained by photographing with the fourth camera;
With
The image processing unit
Selecting one point in the first line segment data corresponding to the line laser imaged on the first image as a feature point;
An epipolar line corresponding to the feature point selected on the first image is calculated on the second image;
On the second image, an intersection between the second line segment data corresponding to the captured line laser and the epipolar line is determined as a corresponding point corresponding to the feature point selected on the first image. And
Using the feature points on the first image and the corresponding points on the second image, the three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points are obtained. Calculating and obtaining a three-dimensional coordinate system on the object to be measured in the first set of camera systems;
Selecting one point in the third line segment data corresponding to the line laser imaged on the third image as a feature point;
An epipolar line corresponding to the feature point selected on the third image is calculated on the fourth image;
On the fourth image, an intersection of the fourth line segment data corresponding to the captured line laser and the epipolar line is determined as a corresponding point corresponding to the feature point selected on the third image. And
Using the feature points on the third image and the corresponding points on the fourth image, the three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points are obtained. Calculating and obtaining a three-dimensional coordinate system on the object to be measured in the second set of camera systems;
Of the three-dimensional coordinates on the measurement object in the second set of camera systems, select at least three three-dimensional coordinates corresponding to the three-dimensional coordinates on the measurement object in the first set of camera systems; For the selected at least three three-dimensional coordinates, a coordinate conversion relationship from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems is obtained, and the obtained coordinate conversion relationship And converting the three-dimensional coordinate system of the second set of camera systems into the three-dimensional coordinate system of the first set of camera systems, and based on the three-dimensional coordinate system of the first set of camera systems A non-contact three-dimensional measuring device for obtaining a three-dimensional coordinate of an object to be measured,
In the process of determining the corresponding points of the second set of camera systems in the image processing unit, when there are two or more intersections between the epipolar line and the fourth line segment data, the line laser is changed to the third laser. Irradiating by rotating about the axis of the irradiation direction to the corresponding point on the measurement object corresponding to the feature point on the image of the image, and the epipolar line and the fourth line segment data on the fourth image A non-contact three-dimensional measuring apparatus characterized by determining a single intersection point as a corresponding point corresponding to the feature point.   The image processing unit is based on the similarity transformation relationship from the three-dimensional coordinate system of the second camera system to the three-dimensional coordinate system of the first camera system for the selected at least three three-dimensional coordinates. The non-contact three-dimensional measurement apparatus according to claim 13, wherein a coordinate conversion relationship is obtained. The similarity transformation from the three-dimensional coordinate system N (x, y, z) of the second set of camera systems to the three-dimensional coordinate system M (x, y, z) of the first set of camera systems is expressed by the following equation: The non-contact three-dimensional measuring apparatus according to claim 17 represented by:
M _i = sR (N _i −t)
Mi: 3D coordinates (coordinate vector) of the first set of camera systems
Ni: 3D coordinates (coordinate vector) of the second set of camera systems
s: Scale (scalar value)
R: rotation matrix t: translation vector   The line laser irradiates the object to be measured with a line laser so as to intersect a straight line connecting the first camera and the second camera. Contact three-dimensional measuring device. After the calculation of the three-dimensional coordinate system of the object to be measured by the first set of camera systems and the calculation of the three-dimensional coordinate system of the object to be measured by the second set of camera systems, the obtained second set The three-dimensional coordinate system of the camera system is transformed into the three-dimensional coordinate system of the first set of camera systems to obtain the three-dimensional coordinates of the object to be measured based on the three-dimensional coordinate system of the first set of camera systems. A non-contact three-dimensional measurement method,
(A) A first external orientation that performs an external orientation relating to the mutual positional relationship of the first camera and the second camera that are arranged at a distance from each other and constitute a first set of camera systems for photographing the object to be measured. Steps,
(B) a line laser irradiation step of irradiating the object to be measured with a line laser;
(C) The object to be measured irradiated with the line laser is photographed by the first camera and the second camera, respectively, and the first image by the first camera and the second image by the second camera are taken. An image capturing step by a first camera system for obtaining two images;
(D) a first line segment data extracting step of extracting a portion where the line laser is photographed as first line segment data in the first image;
(E) in the second image, a second line segment data extracting step of extracting a portion where the line laser is photographed as second line segment data;
(F) a feature point selection step of selecting one point in the first line segment data extracted from the first image as a feature point;
(G) an epipolar line calculation step of calculating an epipolar line corresponding to the feature point selected on the first image on the second image;
(H) Corresponding point determination for determining an intersection between the second line segment data extracted from the second image and the epipolar line as a corresponding point corresponding to the feature point selected on the first image. Steps,
(I) Using the feature points in the first image and the corresponding points in the second image, three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points Calculating a three-dimensional coordinate by the first set of camera systems to obtain a three-dimensional coordinate system on the measurement object in the first set of camera systems;
(I-2) Steps (f) to (i) are repeated for all points on the first line segment extracted from the first image, and the line laser of the object to be measured is irradiated. A three-dimensional coordinate calculation step for calculating a three-dimensional coordinate for the entire line portion,
(J) A third camera and a fourth camera that are arranged at a distance from each other at a position different from the first set of camera systems and constitute a second set of camera systems for capturing an object to be measured. A second external orientation step for performing external orientation with respect to each other's positional relationship;
(K) a line laser irradiation step of irradiating the object to be measured with a line laser;
(L) The object to be measured irradiated with the line laser is photographed by the third camera and the fourth camera, respectively, and a third image by the third camera and a fourth image by the fourth camera are taken. An image capturing step by a second set of camera systems for obtaining four images;
(M) a third line segment data extracting step of extracting a portion where the line laser is photographed as third line segment data in the third image;
(N) a fourth line segment data extracting step of extracting, as the fourth line segment data, a portion where the line laser is photographed in the fourth image;
(O) a feature point selecting step of selecting one point in the third line segment data extracted from the third image as a feature point;
(P) an epipolar line calculation step of calculating an epipolar line corresponding to the feature point selected on the third image on the fourth image;
(Q) Corresponding point determination for determining an intersection between the fourth line segment data extracted from the fourth image and the epipolar line as a corresponding point corresponding to the feature point selected on the third image. Steps,
(R) Using the feature points in the third image and the corresponding points in the fourth image, three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points Calculating a three-dimensional coordinate by a second set of camera systems to obtain a three-dimensional coordinate system on the measurement object in the second set of camera systems;
(S) Among the three-dimensional coordinates on the object to be measured in the second set of camera systems, at least three three-dimensional coordinates corresponding to the three-dimensional coordinates on the object to be measured in the first set of camera systems are A coordinate conversion relationship calculation for selecting and calculating a coordinate conversion relationship from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems for the selected at least three three-dimensional coordinates Steps,
(T) Using the obtained coordinate transformation relationship, the three-dimensional coordinate system of the second set of camera systems is transformed into the three-dimensional coordinate system of the first set of camera systems, and the first set of Obtaining a three-dimensional coordinate of the object to be measured based on a three-dimensional coordinate system of a camera system;
A non-contact three-dimensional measurement method including
The line laser irradiation step (b) by the first set of camera systems further includes a line laser scanning step of irradiating the object to be measured by rotating the line laser about a predetermined axis in an irradiation direction, The non-contact three-dimensional measurement method characterized by repeating the step (c) to the step (i-2) each time the object to be measured is rotated and irradiated. After the calculation of the three-dimensional coordinate system of the object to be measured by the first set of camera systems and the calculation of the three-dimensional coordinate system of the object to be measured by the second set of camera systems, the obtained second set The three-dimensional coordinate system of the camera system is transformed into the three-dimensional coordinate system of the first set of camera systems to obtain the three-dimensional coordinates of the object to be measured based on the three-dimensional coordinate system of the first set of camera systems. A non-contact three-dimensional measurement method,
(A) A first external orientation that performs an external orientation relating to the mutual positional relationship of the first camera and the second camera that are arranged at a distance from each other and constitute a first set of camera systems for photographing the object to be measured. Steps,
(B) a line laser irradiation step of irradiating the object to be measured with a line laser;
(C) The object to be measured irradiated with the line laser is photographed by the first camera and the second camera, respectively, and the first image by the first camera and the second image by the second camera are taken. An image capturing step by a first camera system for obtaining two images;
(D) a first line segment data extracting step of extracting a portion where the line laser is photographed as first line segment data in the first image;
(E) in the second image, a second line segment data extracting step of extracting a portion where the line laser is photographed as second line segment data;
(F) a feature point selection step of selecting one point in the first line segment data extracted from the first image as a feature point;
(G) an epipolar line calculation step of calculating an epipolar line corresponding to the feature point selected on the first image on the second image;
(H) Corresponding point determination for determining an intersection between the second line segment data extracted from the second image and the epipolar line as a corresponding point corresponding to the feature point selected on the first image. Steps,
(I) Using the feature points in the first image and the corresponding points in the second image, three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points Calculating a three-dimensional coordinate by the first set of camera systems to obtain a three-dimensional coordinate system on the measurement object in the first set of camera systems;
(J) A third camera and a fourth camera that are arranged at a distance from each other at a position different from the first set of camera systems and constitute a second set of camera systems for capturing an object to be measured. A second external orientation step for performing external orientation with respect to each other's positional relationship;
(K) a line laser irradiation step of irradiating the object to be measured with a line laser;
(L) The object to be measured irradiated with the line laser is photographed by the third camera and the fourth camera, respectively, and a third image by the third camera and a fourth image by the fourth camera are taken. An image capturing step by a second set of camera systems for obtaining four images;
(M) a third line segment data extracting step of extracting a portion where the line laser is photographed as third line segment data in the third image;
(N) a fourth line segment data extracting step of extracting, as the fourth line segment data, a portion where the line laser is photographed in the fourth image;
(O) a feature point selecting step of selecting one point in the third line segment data extracted from the third image as a feature point;
(P) an epipolar line calculation step of calculating an epipolar line corresponding to the feature point selected on the third image on the fourth image;
(Q) Corresponding point determination for determining an intersection between the fourth line segment data extracted from the fourth image and the epipolar line as a corresponding point corresponding to the feature point selected on the third image. Steps,
(R) Using the feature points in the third image and the corresponding points in the fourth image, three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points Calculating a three-dimensional coordinate by a second set of camera systems to obtain a three-dimensional coordinate system on the measurement object in the second set of camera systems;
(S) Among the three-dimensional coordinates on the object to be measured in the second set of camera systems, at least three three-dimensional coordinates corresponding to the three-dimensional coordinates on the object to be measured in the first set of camera systems are A coordinate conversion relationship calculation for selecting and calculating a coordinate conversion relationship from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems for the selected at least three three-dimensional coordinates Steps,
(S-2) Steps (o) to (s) are repeated for all points on the third line segment extracted from the third image, and the line laser of the object to be measured is irradiated. A three-dimensional coordinate calculation step for calculating a three-dimensional coordinate for the entire line portion,
(T) Using the obtained coordinate transformation relationship, the three-dimensional coordinate system of the second set of camera systems is transformed into the three-dimensional coordinate system of the first set of camera systems, and the first set of Obtaining a three-dimensional coordinate of the object to be measured based on a three-dimensional coordinate system of a camera system;
A non-contact three-dimensional measurement method including
The line laser irradiation step (k) by the second set of camera systems further includes a line laser scanning step of irradiating the object to be measured by rotating the line laser about a predetermined axis in the irradiation direction, The non-contact three-dimensional measurement method characterized by repeating the step (l) to the step (s-2) every time the object to be measured is rotated and irradiated. After the calculation of the three-dimensional coordinate system of the object to be measured by the first set of camera systems and the calculation of the three-dimensional coordinate system of the object to be measured by the second set of camera systems, the obtained second set The three-dimensional coordinate system of the camera system is transformed into the three-dimensional coordinate system of the first set of camera systems to obtain the three-dimensional coordinates of the object to be measured based on the three-dimensional coordinate system of the first set of camera systems. A non-contact three-dimensional measurement method,
(A) A first external orientation that performs an external orientation relating to the mutual positional relationship of the first camera and the second camera that are arranged at a distance from each other and constitute a first set of camera systems for photographing the object to be measured. Steps,
(B) a line laser irradiation step of irradiating the object to be measured with a line laser;
(C) The object to be measured irradiated with the line laser is photographed by the first camera and the second camera, respectively, and the first image by the first camera and the second image by the second camera are taken. An image capturing step by a first camera system for obtaining two images;
(D) a first line segment data extracting step of extracting a portion where the line laser is photographed as first line segment data in the first image;
(E) in the second image, a second line segment data extracting step of extracting a portion where the line laser is photographed as second line segment data;
(F) a feature point selection step of selecting one point in the first line segment data extracted from the first image as a feature point;
(G) an epipolar line calculation step of calculating an epipolar line corresponding to the feature point selected on the first image on the second image;
(H) Corresponding point determination for determining an intersection between the second line segment data extracted from the second image and the epipolar line as a corresponding point corresponding to the feature point selected on the first image. Steps,
(I) Using the feature points in the first image and the corresponding points in the second image, three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points Calculating a three-dimensional coordinate by the first set of camera systems to obtain a three-dimensional coordinate system on the measurement object in the first set of camera systems;
(J) A third camera and a fourth camera that are arranged at a distance from each other at a position different from the first set of camera systems and constitute a second set of camera systems for capturing an object to be measured. A second external orientation step for performing external orientation with respect to each other's positional relationship;
(K) a line laser irradiation step of irradiating the object to be measured with a line laser;
(L) The object to be measured irradiated with the line laser is photographed by the third camera and the fourth camera, respectively, and a third image by the third camera and a fourth image by the fourth camera are taken. An image capturing step by a second set of camera systems for obtaining four images;
(M) a third line segment data extracting step of extracting a portion where the line laser is photographed as third line segment data in the third image;
(N) a fourth line segment data extracting step of extracting, as the fourth line segment data, a portion where the line laser is photographed in the fourth image;
(O) a feature point selecting step of selecting one point in the third line segment data extracted from the third image as a feature point;
(P) an epipolar line calculation step of calculating an epipolar line corresponding to the feature point selected on the third image on the fourth image;
(Q) Corresponding point determination for determining an intersection between the fourth line segment data extracted from the fourth image and the epipolar line as a corresponding point corresponding to the feature point selected on the third image. Steps,
(R) Using the feature points in the third image and the corresponding points in the fourth image, three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points Calculating a three-dimensional coordinate by a second set of camera systems to obtain a three-dimensional coordinate system on the measurement object in the second set of camera systems;
(S) Among the three-dimensional coordinates on the object to be measured in the second set of camera systems, at least three three-dimensional coordinates corresponding to the three-dimensional coordinates on the object to be measured in the first set of camera systems are A coordinate conversion relationship calculation for selecting and calculating a coordinate conversion relationship from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems for the selected at least three three-dimensional coordinates Steps,
(T) Using the obtained coordinate transformation relationship, the three-dimensional coordinate system of the second set of camera systems is transformed into the three-dimensional coordinate system of the first set of camera systems, and the first set of Obtaining a three-dimensional coordinate of the object to be measured based on a three-dimensional coordinate system of a camera system;
A non-contact three-dimensional measurement method including
In the corresponding point determination step (h) by the first set of camera systems, when there are two or more intersections between the epipolar line and the second line segment data, the line laser is set on the first image. A single intersection point of the epipolar line and the second line segment data on the second image by irradiating by rotating about the axis of the irradiation direction to the corresponding point on the measurement object corresponding to the feature point Is determined as a corresponding point corresponding to the feature point, a non-contact three-dimensional measurement method. After the calculation of the three-dimensional coordinate system of the object to be measured by the first set of camera systems and the calculation of the three-dimensional coordinate system of the object to be measured by the second set of camera systems, the obtained second set The three-dimensional coordinate system of the camera system is transformed into the three-dimensional coordinate system of the first set of camera systems to obtain the three-dimensional coordinates of the object to be measured based on the three-dimensional coordinate system of the first set of camera systems. A non-contact three-dimensional measurement method,
(A) A first external orientation that performs an external orientation relating to the mutual positional relationship of the first camera and the second camera that are arranged at a distance from each other and constitute a first set of camera systems for photographing the object to be measured. Steps,
(B) a line laser irradiation step of irradiating the object to be measured with a line laser;
(C) The object to be measured irradiated with the line laser is photographed by the first camera and the second camera, respectively, and the first image by the first camera and the second image by the second camera are taken. An image capturing step by a first camera system for obtaining two images;
(D) a first line segment data extracting step of extracting a portion where the line laser is photographed as first line segment data in the first image;
(E) in the second image, a second line segment data extracting step of extracting a portion where the line laser is photographed as second line segment data;
(F) a feature point selection step of selecting one point in the first line segment data extracted from the first image as a feature point;
(G) an epipolar line calculation step of calculating an epipolar line corresponding to the feature point selected on the first image on the second image;
(H) Corresponding point determination for determining an intersection between the second line segment data extracted from the second image and the epipolar line as a corresponding point corresponding to the feature point selected on the first image. Steps,
(I) Using the feature points in the first image and the corresponding points in the second image, three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points Calculating a three-dimensional coordinate by the first set of camera systems to obtain a three-dimensional coordinate system on the measurement object in the first set of camera systems;
(J) A third camera and a fourth camera that are arranged at a distance from each other at a position different from the first set of camera systems and constitute a second set of camera systems for capturing an object to be measured. A second external orientation step for performing external orientation with respect to each other's positional relationship;
(K) a line laser irradiation step of irradiating the object to be measured with a line laser;
(L) The object to be measured irradiated with the line laser is photographed by the third camera and the fourth camera, respectively, and a third image by the third camera and a fourth image by the fourth camera are taken. An image capturing step by a second set of camera systems for obtaining four images;
(M) a third line segment data extracting step of extracting a portion where the line laser is photographed as third line segment data in the third image;
(N) a fourth line segment data extracting step of extracting, as the fourth line segment data, a portion where the line laser is photographed in the fourth image;
(O) a feature point selecting step of selecting one point in the third line segment data extracted from the third image as a feature point;
(P) an epipolar line calculation step of calculating an epipolar line corresponding to the feature point selected on the third image on the fourth image;
(Q) Corresponding point determination for determining an intersection between the fourth line segment data extracted from the fourth image and the epipolar line as a corresponding point corresponding to the feature point selected on the third image. Steps,
(R) Using the feature points in the third image and the corresponding points in the fourth image, three-dimensional coordinates of the corresponding points on the measurement object corresponding to the feature points and the corresponding points Calculating a three-dimensional coordinate by a second set of camera systems to obtain a three-dimensional coordinate system on the measurement object in the second set of camera systems;
(S) Among the three-dimensional coordinates on the object to be measured in the second set of camera systems, at least three three-dimensional coordinates corresponding to the three-dimensional coordinates on the object to be measured in the first set of camera systems are A coordinate conversion relationship calculation for selecting and calculating a coordinate conversion relationship from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems for the selected at least three three-dimensional coordinates Steps,
(T) Using the obtained coordinate transformation relationship, the three-dimensional coordinate system of the second set of camera systems is transformed into the three-dimensional coordinate system of the first set of camera systems, and the first set of Obtaining a three-dimensional coordinate of the object to be measured based on a three-dimensional coordinate system of a camera system;
A non-contact three-dimensional measurement method including
In the corresponding point determination step (q) by the second set of camera systems, when there are two or more intersections between the epipolar line and the fourth line segment data, the line laser is set on the third image. A single intersection point of the epipolar line and the fourth line segment data on the fourth image by irradiating by rotating about the axis of the irradiation direction to the corresponding point on the measurement object corresponding to the feature point the determined as the corresponding point corresponding to the feature point, a non-contact three-dimensional measurement method.   In the coordinate transformation relationship calculating step, the similarity transformation relationship from the three-dimensional coordinate system of the second set of camera systems to the three-dimensional coordinate system of the first set of camera systems for the selected at least three three-dimensional coordinates The non-contact three-dimensional measurement method according to any one of claims 20 to 23, wherein a coordinate conversion relationship is obtained based on the method. The similarity transformation from the three-dimensional coordinate system N (x, y, z) of the second set of camera systems to the three-dimensional coordinate system M (x, y, z) of the first set of camera systems is expressed by the following equation: The non-contact three-dimensional measurement method according to claim 24, represented by:
M _i = sR (N _i −t)
Mi: 3D coordinates (coordinate vector) of the first set of camera systems
Ni: 3D coordinates (coordinate vector) of the second set of camera systems
s: Scale (scalar value)
R: rotation matrix t: translation vector   26. The line laser irradiation step by the first set of camera systems, the line laser is irradiated to the object to be measured so as to intersect a straight line connecting the first camera and the second camera. The non-contact three-dimensional measurement method according to any one of the above.   27. The line laser irradiation step by the second set of camera systems, the line laser is irradiated to the object to be measured so as to intersect with a straight line connecting the third camera and the fourth camera. The non-contact three-dimensional measurement method according to any one of the above.   The non-contact three-dimensional measurement apparatus according to any one of claims 1 to 6, and 13 to 19, wherein the line laser uses a plurality of color line lasers in combination.   The non-contact three-dimensional measurement according to any one of claims 7 to 12, and 20 to 27, wherein in the step of irradiating the line laser, a line laser of a plurality of colors is used in combination as the line laser. Method.

Images