A method for circumferential layout calibration of a line laser 3D camera

Abstract

This invention discloses a ring-layout calibration method for line laser 3D cameras. A polygonal prism calibration block is set up, and at least three line laser 3D cameras are uniformly arranged in a ring around the vertical cross-section of the polygonal prism calibration block, ensuring that each line laser 3D camera can observe one corner of the polygonal prism calibration block, and each line laser 3D camera is labeled. A coordinate system for the line laser 3D cameras is established, with one of the line laser 3D cameras serving as a reference camera, and its coordinate system as the reference coordinate system. By calibrating the rigid body transformation matrix between the line laser 3D cameras, the coordinate systems of the other line laser 3D cameras are aligned with the coordinate system of the reference line laser 3D camera. The rigid body transformation matrix includes a rotation matrix and a translation vector. This invention, by using a multi-camera system, can expand the field of view of line laser 3D cameras, meeting the requirement of acquiring the entire cross-sectional contour information of an object in a single image.

Description

Technical Field

[0001] This invention relates to the field of industrial machine vision 3D measurement technology, and in particular to a method for calibrating a line laser 3D camera using a circular layout. Background Technology

[0002] In the industrial field, a line laser 3D camera is a measuring instrument consisting of a camera and a laser line projector, such as... Figure 3 As shown, the principle of a line laser 3D camera is as follows: the eyepiece 2 receives diffusely reflected light from the workpiece 4 and images it on a complementary metal-oxide-semiconductor (CMOS) sensor 3. By detecting changes in the position of the laser line emitted by the laser 1, height information is obtained using triangulation. Currently, line laser 3D cameras are widely used in the measurement and positioning of height, thickness, width, radius, flatness, angle, position, shape, etc. Line laser 3D cameras have advantages such as non-contact operation, high precision, high speed, and independence from external light sources.

[0003] However, existing line laser 3D cameras have the following drawbacks when in use: due to the optical characteristics and field of view limitations of a single line laser 3D camera, it is not possible to obtain the full cross-sectional contour information of workpiece 4 in a single imaging. The measurement range of the line laser 3D camera is limited by the actual size and shape of the workpiece, and cannot meet the measurement needs of large-scale and ring-shaped layouts. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide a ring layout calibration method for line laser 3D cameras. By using a multi-camera system, the field of view of the line laser 3D camera can be expanded, which can meet the requirement of obtaining the entire cross-sectional contour information of an object in a single photograph.

[0005] To achieve the above objectives, the technical solution adopted by this invention is as follows: a ring-shaped layout calibration method for line laser 3D cameras, comprising setting up a polygonal prism calibration block, and uniformly arranging at least three line laser 3D cameras in a ring around the vertical cross-section of the polygonal prism calibration block, so that each line laser 3D camera can observe one corner of the polygonal prism calibration block, and marking the line laser 3D cameras respectively; establishing a coordinate system for the line laser 3D cameras, taking one of the line laser 3D cameras as a reference camera, and using the coordinate system of the reference line laser 3D camera as the reference coordinate system, and calibrating the rigid body transformation matrix between the line laser 3D cameras to bring the coordinate systems of the other line laser 3D cameras into the coordinate system of the reference line laser 3D camera, wherein the rigid body transformation matrix includes a rotation matrix and a translation vector.

[0006] As a preferred embodiment, the specific steps of the ring layout calibration method include: obtaining the contour line, fitting a straight line, calculating the rotation matrix, calculating the translation vector, and calculating the rigid body transformation, wherein fitting the straight line includes finding the intersection point of the straight lines and finding the direction vector of the straight lines.

[0007] As a preferred embodiment, a triangular prism calibration block is set up, and three line laser 3D cameras are evenly arranged in a ring around the vertical cross-section of the triangular prism calibration block at 120-degree intervals, so that each line laser 3D camera can observe one corner of the triangular prism calibration block. The three line laser 3D cameras are respectively labeled as the first line laser 3D camera P, the second line laser 3D camera Q, and the third line laser 3D camera M. The coordinate system of the line laser 3D cameras is set as XOZ, where X is the horizontal axis of the coordinate system of the line laser 3D cameras, Z is the vertical axis of the coordinate system of the line laser 3D cameras, and O is the origin of the coordinate system of the line laser 3D cameras. The coordinate system of the first line laser 3D camera P is used as the reference coordinate system X. w O w Z w By calibrating the rigid body transformation matrix between the line laser 3D cameras, the coordinate system of the second line laser 3D camera Q and the coordinate system of the third line laser 3D camera M are brought into the coordinate system of the first line laser 3D camera P. The rigid body transformation matrix includes a rotation matrix and a translation vector.

[0008] As a preferred embodiment, the specific steps of the ring layout calibration method include:

[0009] S1: Obtain the outline, trigger the three line laser 3D cameras to take pictures, and each of the three line laser 3D cameras obtains one outline.

[0010] S2: Fitting straight lines. The contour lines of the three linear laser 3D cameras are fitted with two adjacent straight lines, denoted as the first straight line L1 and the second straight line L2, respectively. The equations of the first straight line L1 and the second straight line L2 are:

[0011] L1:A1X+B1Z+C1=0,

[0012] L2:A2X+B2Z+C2=0;

[0013] Where L1 is the first straight line of the line laser 3D camera, L2 is the second straight line of the line laser 3D camera, X is the horizontal axis of the coordinate system of the line laser 3D camera, Z is the vertical axis of the coordinate system of the line laser 3D camera, and A1, B1, C1, A2, B2, and C2 are constants.

[0014] Find the intersection point V of the first line L1 and the second line L2 using the equations of the line laser 3D camera. 12 Find the direction vector D1 of the first line L1 and the direction vector D2 of the second line L2. Adjust the direction vectors of the first line L1 and the second line L2 so that both lines point to the intersection point P. 12 ;

[0015] S3: Calculate the rotation matrix. The first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q capture the same face of the triangular prism calibration block. The first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q are in the reference coordinate system X. W O W Z W Given lines with the same direction, calculate the rotation angle between the first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q; let the direction vector of the first straight line PL1 to the left of the first line laser 3D camera P be D. p1 =(a p1 b p1 Let D be the direction vector of the second straight line QL2 to the right of the second line laser 3D camera Q. q2 =(a q2 b q2 The formula for calculating the rotation angle between the first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q is as follows:

[0016]

[0017] Where θ is the rotation angle between the first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q, dot() is the vector dot product operation, and D p1 Let D be the direction vector of the first straight line PL1 to the left of the first line laser 3D camera P. q2 The direction vector of the second straight line QL2 to the right of the second-line laser 3D camera Q;

[0018] Finally, the rotation matrix between the first-line laser 3D camera P and the second-line laser 3D camera Q is obtained:

[0019]

[0020] Among them, R pq Let θ be the rotation matrix between the first line laser 3D camera P and the second line laser 3D camera Q, and let θ be the rotation angle between the first line PL1 to the left of the first line laser 3D camera P and the second line QL2 to the right of the second line laser 3D camera Q.

[0021] Similarly, the rotation matrix R between the first-line laser 3D camera P and the third-line laser 3D camera M is obtained. pm And the rotation matrix R between the second-line laser 3D camera Q and the third-line laser 3D camera M. qm ;

[0022] S4: Calculate the translation vector. Each vertical section of the triangular prism calibration block is an isosceles triangle. Let the side length of the triangular prism calibration block be S. Let the coordinate system of the first linear laser 3D camera P be the reference coordinate system X. w O w Z w In step S2, the coordinates of the intersection point of the first line laser 3D camera P and the straight line are calculated as V. p (x p , z p The coordinates of the intersection point of the lines calculated by the second-line laser 3D camera Q in step S2 are V. q (x q , z q If the coordinates of the intersection point V of the second-line laser 3D camera Q are given, then... q (x q , z q The coordinates in the first-line laser 3D camera P should be:

[0023]

[0024] Among them, V′ q Let x′ be the coordinates of the intersection point of the line between the second line laser 3D camera Q and the first line laser 3D camera P. q Let z′ be the coordinates of the intersection point of the second-line laser 3D camera Q with the first-line laser 3D camera P on the horizontal axis of the coordinate system. q Let x be the coordinates of the intersection point of the lines from the second-line laser 3D camera Q with the first-line laser 3D camera P, and x be the coordinates of the intersection point on the vertical axis of the coordinate system. p Let z be the coordinate of the intersection point of the lines from the first-line laser 3D camera P on the horizontal axis of the coordinate system. p Let a be the coordinate of the intersection point of the lines from the first-line laser 3D camera P on the vertical axis of the coordinate system. p1 Let D be the direction vector of the first straight line PL1 to the left of the first line laser 3D camera P. p1 In the coordinate system, the horizontal axis coordinate is b p1 Let D be the direction vector of the first straight line PL1 to the left of the first line laser 3D camera P. p1 In the coordinate system, S is the side length of the triangular prism calibration block, located on the vertical axis.

[0025] Thus, the coordinate system of the second-line laser 3D camera Q is obtained and transferred to the reference coordinate system X. w O w Z w The following translation relationship:

[0026]

[0027] Among them, t pq Let t be the translation vector between the first line laser 3D camera P and the second line laser 3D camera Q.x Let t be the coordinate of the translation vector between the first-line laser 3D camera P and the second-line laser 3D camera Q on the horizontal axis of the coordinate system. z Let x′ be the coordinate of the translation vector between the first-line laser 3D camera P and the second-line laser 3D camera Q on the vertical axis of the coordinate system. q Let z′ be the coordinates of the intersection point of the second-line laser 3D camera Q with the first-line laser 3D camera P on the horizontal axis of the coordinate system. q Let x be the coordinates of the intersection point of the lines from the second-line laser 3D camera Q with the first-line laser 3D camera P, and x be the coordinates of the intersection point on the vertical axis of the coordinate system. p Let z be the coordinate of the intersection point of the lines from the first-line laser 3D camera P on the horizontal axis of the coordinate system. p Let R be the coordinate of the intersection point of the lines from the first-line laser 3D camera P on the vertical axis of the coordinate system. pq Let be the rotation matrix between the first-line laser 3D camera P and the second-line laser 3D camera Q;

[0028] Similarly, the translation vector t between the first-line laser 3D camera P and the third-line laser 3D camera M is obtained. pm And the translation vector t between the second-line laser 3D camera Q and the third-line laser 3D camera M qm ;

[0029] S5: Calculate the rigid body transformation, complete the ring layout calibration, and obtain the rigid body transformation matrix by combining the rotation matrix and translation vector. For the first line laser 3D camera P and the second line laser 3D camera Q, the rigid body transformation matrix of the first line laser 3D camera P and the second line laser 3D camera Q is T. pq =[R pq , t pq For any contour line coordinate point (x1, z1) in the second-line laser 3D camera Q, the contour line coordinate point (x1, z1) after transformation is in the coordinate system of the first-line laser 3D camera P, i.e., the reference coordinate system X. w O w Z w The coordinates are:

[0030]

[0031] Where x′1 is the coordinate of any contour line point (x1, z1) in the second line laser 3D camera Q after transformation on the horizontal axis of the first line laser 3D camera P, z′1 is the coordinate of any contour line point (x1, z1) in the second line laser 3D camera Q after transformation on the vertical axis of the first line laser 3D camera P, x1 is the coordinate of any contour line point (x1, z1) in the second line laser 3D camera Q on the horizontal axis of the coordinate system, and z1 is the coordinate of any contour line point (x1, z1) in the second line laser 3D camera Q on the vertical axis of the coordinate system, Tpq Let be the rigid body transformation matrix of the first line laser 3D camera P and the second line laser 3D camera Q.

[0032] Similarly, the rigid body transformation matrix T between the first-line laser 3D camera P and the third-line laser 3D camera M is obtained. pm By performing a corresponding coordinate transformation on the contour coordinates acquired by the third-line laser 3D camera M, the transformed coordinates of the contour points in the third-line laser 3D camera M can be obtained in the coordinate system of the first-line laser 3D camera P, i.e., the reference coordinate system X. w O w Z w The coordinates.

[0033] As a preferred embodiment, in step S2, the intersection point V of the first line L1 and the second line L2 is calculated. 12 Specifically, this includes: by simultaneously solving the equations of the first line L1 and the second line L2, the intersection point V of the first line L1 and the second line L2 can be obtained. 12 ,

[0034]

[0035] Among them, V 12 Let L1 be the intersection point of the first line L1 and the second line L2, and let A1, B1, C1, A2, B2, and C2 be constants.

[0036] As a preferred embodiment, in step S2, the direction vectors of the first line L1 and the second line L2 are calculated, and the direction vectors of the first line L1 and the second line L2 are adjusted so that both the first line L1 and the second line L2 point to the intersection point P. 12 ;

[0037]

[0038]

[0039] Where D1 is the direction vector of the first line L1, D2 is the direction vector of the second line L2, and A1, B1, A2, and B2 are constants.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention solves the problem that a single line laser 3D camera cannot obtain the entire cross-sectional contour information of an object in a single photograph. First, three cameras are arranged in a ring at 120-degree intervals on the cross-section of the workpiece. Then, the spatial positional relationship between adjacent line laser 3D cameras is obtained with the help of a triangular prism calibration block. Finally, the requirement of obtaining the entire cross-sectional contour information of the workpiece in a single photograph by the three cameras can be achieved through rigid body transformation. Attached Figure Description

[0041] Figure 1This is a flowchart of the process of this invention;

[0042] Figure 2 This is a schematic diagram of the ring-shaped layout in this invention;

[0043] Figure 3 This is a schematic diagram of the structure of a conventional linear laser 3D camera;

[0044] The attached diagram is labeled as follows: Laser 1, Eyepiece 2, Complementary Metal-Oxide-Semiconductor 3, Workpiece 4, Triangular Prism Calibration Block 5, First Line Laser 3D Camera P, Second Line Laser 3D Camera Q, and Third Line Laser 3D Camera M. Detailed Implementation

[0045] The present invention will be further described below with reference to specific embodiments. These embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0046] Example 1:

[0047] like Figures 1-2 As shown, a ring-layout calibration method for line laser 3D cameras involves setting up a polygonal prism calibration block, and uniformly arranging at least three line laser 3D cameras in a ring around the vertical cross-section of the polygonal prism calibration block, so that each line laser 3D camera can observe one corner of the polygonal prism calibration block, and marking the line laser 3D cameras respectively; establishing a coordinate system for the line laser 3D cameras, with one of the line laser 3D cameras as a reference camera, and the coordinate system of the reference line laser 3D camera as the reference coordinate system; and calibrating the rigid body transformation matrix between the line laser 3D cameras to bring the coordinate systems of the other line laser 3D cameras into the coordinate system of the reference line laser 3D camera, wherein the rigid body transformation matrix includes a rotation matrix and a translation vector.

[0048] Preferably, the specific steps of the ring layout calibration method include: obtaining the contour line, fitting a straight line, calculating the rotation matrix, calculating the translation vector, and calculating the rigid body transformation, wherein fitting the straight line includes finding the intersection point of the straight line and finding the direction vector of the straight line.

[0049] Example 2:

[0050] Based on the preferred ring-layout calibration method for line laser 3D cameras in Embodiment 1, a triangular prism calibration block 5 is set up, and three line laser 3D cameras are uniformly arranged in a ring around the vertical cross-section of the triangular prism calibration block 5 at 120-degree intervals, so that each line laser 3D camera can observe one corner of the triangular prism calibration block 5. The three line laser 3D cameras are respectively labeled as the first line laser 3D camera P, the second line laser 3D camera Q, and the third line laser 3D camera M. The coordinate system of the line laser 3D cameras is set as XOZ, where X is the horizontal axis of the coordinate system of the line laser 3D cameras, Z is the vertical axis of the coordinate system of the line laser 3D cameras, and O is the origin of the coordinate system of the line laser 3D cameras. The coordinate system of the first line laser 3D camera P is used as the reference coordinate system X. w O w Z w By calibrating the rigid body transformation matrix between the line laser 3D cameras, the coordinate system of the second line laser 3D camera Q and the coordinate system of the third line laser 3D camera M are brought into the coordinate system of the first line laser 3D camera P. The rigid body transformation matrix includes a rotation matrix and a translation vector.

[0051] The specific steps of the ring layout calibration method include:

[0052] S1: Obtain the outline, trigger the three line laser 3D cameras to take pictures, and each of the three line laser 3D cameras obtains one outline.

[0053] S2: Fitting straight lines. The contour lines of the three linear laser 3D cameras are fitted with two adjacent straight lines, denoted as the first straight line L1 and the second straight line L2, respectively. The equations of the first straight line L1 and the second straight line L2 are:

[0054] L1:A1X+B1Z+C1=0,

[0055] L2:A2X+B2Z+C2=0;

[0056] Where L1 is the first straight line of the line laser 3D camera, L2 is the second straight line of the line laser 3D camera, X is the horizontal axis of the coordinate system of the line laser 3D camera, Z is the vertical axis of the coordinate system of the line laser 3D camera, and A1, B1, C1, A2, B2, and C2 are constants.

[0057] Find the intersection point V of the first line L1 and the second line L2 using the equations of the line laser 3D camera. 12 Find the direction vector D1 of the first line L1 and the direction vector D2 of the second line L2. Adjust the direction vectors of the first line L1 and the second line L2 so that both lines point to the intersection point P. 12 ;

[0058] S3: Calculate the rotation matrix. The first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q capture the same face of the triangular prism calibration block 5. The first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q are in the reference coordinate system X. W O W Z W Given lines with the same direction, calculate the rotation angle between the first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q; let the direction vector of the first straight line PL1 to the left of the first line laser 3D camera P be D. p1 =(a p1 b p1 Let D be the direction vector of the second straight line QL2 to the right of the second line laser 3D camera Q. q2 =(a q2 b q2 The formula for calculating the rotation angle between the first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q is as follows:

[0059]

[0060] Where θ is the rotation angle between the first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q, dot() is the vector dot product operation, and D p1 Let D be the direction vector of the first straight line PL1 to the left of the first line laser 3D camera P. q2 The direction vector of the second straight line QL2 to the right of the second-line laser 3D camera Q;

[0061] Finally, the rotation matrix between the first-line laser 3D camera P and the second-line laser 3D camera Q is obtained:

[0062]

[0063] Among them, R pq Let θ be the rotation matrix between the first line laser 3D camera P and the second line laser 3D camera Q, and let θ be the rotation angle between the first line PL1 to the left of the first line laser 3D camera P and the second line QL2 to the right of the second line laser 3D camera Q.

[0064] Similarly, the rotation matrix R between the first-line laser 3D camera P and the third-line laser 3D camera M is obtained. pm And the rotation matrix R between the second-line laser 3D camera Q and the third-line laser 3D camera M. qm ;

[0065] S4: Calculate the translation vector. Each vertical section of the triangular prism calibration block 5 is an isosceles triangle. Let the side length of the triangular prism calibration block 5 be S. Let the coordinate system of the first linear laser 3D camera P, i.e., the reference coordinate system X, be... w O w Z w In step S2, the coordinates of the intersection point of the first line laser 3D camera P and the straight line are calculated as V. p (x p , z p The coordinates of the intersection point of the lines calculated by the second-line laser 3D camera Q in step S2 are V. q (x q , z q If the coordinates of the intersection point V of the second-line laser 3D camera Q are given, then... q (x q , z q The coordinates in the first-line laser 3D camera P should be:

[0066]

[0067] Among them, V′ q Let x′ be the coordinates of the intersection point of the line between the second line laser 3D camera Q and the first line laser 3D camera P. q Let z′ be the coordinates of the intersection point of the second-line laser 3D camera Q with the first-line laser 3D camera P on the horizontal axis of the coordinate system. q Let x be the coordinates of the intersection point of the lines from the second-line laser 3D camera Q with the first-line laser 3D camera P, and x be the coordinates of the intersection point on the vertical axis of the coordinate system. p Let z be the coordinate of the intersection point of the lines from the first-line laser 3D camera P on the horizontal axis of the coordinate system. p Let a be the coordinate of the intersection point of the lines from the first-line laser 3D camera P on the vertical axis of the coordinate system. p1 Let D be the direction vector of the first straight line PL1 to the left of the first line laser 3D camera P. p1 In the coordinate system, the horizontal axis coordinate is b p1 Let D be the direction vector of the first straight line PL1 to the left of the first line laser 3D camera P. p1 In the coordinate system, S is the side length of the triangular prism calibration block 5, located on the vertical axis.

[0068] Thus, the coordinate system of the second-line laser 3D camera Q is obtained and transferred to the reference coordinate system X. w O w Z w The following translation relationship:

[0069]

[0070] Among them, t pq Let t be the translation vector between the first line laser 3D camera P and the second line laser 3D camera Q.x Let t be the coordinate of the translation vector between the first-line laser 3D camera P and the second-line laser 3D camera Q on the horizontal axis of the coordinate system. z Let x′ be the coordinate of the translation vector between the first-line laser 3D camera P and the second-line laser 3D camera Q on the vertical axis of the coordinate system. q Let z′ be the coordinates of the intersection point of the second-line laser 3D camera Q with the first-line laser 3D camera P on the horizontal axis of the coordinate system. q Let x be the coordinates of the intersection point of the lines from the second-line laser 3D camera Q with the first-line laser 3D camera P, and x be the coordinates of the intersection point on the vertical axis of the coordinate system. p Let z be the coordinate of the intersection point of the lines from the first-line laser 3D camera P on the horizontal axis of the coordinate system. p Let R be the coordinate of the intersection point of the lines from the first-line laser 3D camera P on the vertical axis of the coordinate system. pq Let be the rotation matrix between the first-line laser 3D camera P and the second-line laser 3D camera Q;

[0071] Similarly, the translation vector t between the first-line laser 3D camera P and the third-line laser 3D camera M is obtained. pm And the translation vector t between the second-line laser 3D camera Q and the third-line laser 3D camera M qm ;

[0072] S5: Calculate the rigid body transformation, complete the ring layout calibration, and obtain the rigid body transformation matrix by combining the rotation matrix and translation vector. For the first line laser 3D camera P and the second line laser 3D camera Q, the rigid body transformation matrix of the first line laser 3D camera P and the second line laser 3D camera Q is T. pq =[R pq , t pq For any contour line coordinate point (x1, z1) in the second-line laser 3D camera Q, the contour line coordinate point (x1, z1) after transformation is in the coordinate system of the first-line laser 3D camera P, i.e., the reference coordinate system X. w O w Z w The coordinates are:

[0073]

[0074] Where x′1 is the coordinate of any contour line point (x1, z1) in the second line laser 3D camera Q after transformation on the horizontal axis of the first line laser 3D camera P, z′1 is the coordinate of any contour line point (x1, z1) in the second line laser 3D camera Q after transformation on the vertical axis of the first line laser 3D camera P, x1 is the coordinate of any contour line point (x1, z1) in the second line laser 3D camera Q on the horizontal axis of the coordinate system, and z1 is the coordinate of any contour line point (x1, z1) in the second line laser 3D camera Q on the vertical axis of the coordinate system, Tpq Let be the rigid body transformation matrix of the first line laser 3D camera P and the second line laser 3D camera Q.

[0075] Similarly, the rigid body transformation matrix T between the first-line laser 3D camera P and the third-line laser 3D camera M is obtained. pm By performing a corresponding coordinate transformation on the contour coordinates acquired by the third-line laser 3D camera M, the transformed coordinates of the contour points in the third-line laser 3D camera M can be obtained in the coordinate system of the first-line laser 3D camera P, i.e., the reference coordinate system X. w O w Z w The coordinates.

[0076] Specifically, in step S2, the intersection point V of the first line L1 and the second line L2 is calculated. 12 Specifically, this includes: by simultaneously solving the equations of the first line L1 and the second line L2, the intersection point V of the first line L1 and the second line L2 can be obtained. 12 ,

[0077]

[0078] Among them, V 12 Let L1 be the intersection point of the first line L1 and the second line L2, and let A1, V1, C1, A2, V2, and C2 be constants.

[0079] Specifically, in step S2, the direction vectors of the first line L1 and the second line L2 are calculated, and the direction vectors of the first line L1 and the second line L2 are adjusted so that both the first line L1 and the second line L2 point to the intersection point P. 12 ;

[0080]

[0081]

[0082] Where D1 is the direction vector of the first line L1, D2 is the direction vector of the second line L2, and A1, B1, A2, and B2 are constants.

[0083] This invention solves the problem that the optical characteristics and field of view limitations of a single line laser 3D camera prevent the acquisition of all cross-sectional contour information of the workpiece 4 in a single imaging operation; at the same time, the measurement range of a single line laser 3D camera is limited by the actual size and shape of the workpiece 4, and cannot meet the measurement needs of large-scale and ring-shaped layouts.

[0084] This invention expands the line laser 3D field of view by using multiple line laser 3D cameras. First, three line laser 3D cameras are arranged in a ring at 120-degree intervals on the cross section of the workpiece 4. Then, the spatial positional relationship between the cameras is obtained with the help of the triangular prism calibration block 5. Finally, the requirement of obtaining the entire cross section contour information of the workpiece 4 in one photograph by the three line laser 3D cameras can be realized through rigid body transformation, providing a solution for subsequent measurement, display and other functions.

[0085] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A ring layout calibration method of a line laser 3D camera, characterized in that: Set up a polygonal prism calibration block, and arrange at least three line laser 3D cameras in a uniform ring around the vertical cross-section of the polygonal prism calibration block, so that each line laser 3D camera can observe one corner of the polygonal prism calibration block, and mark the line laser 3D cameras respectively; establish the coordinate system of the line laser 3D cameras, take one of the line laser 3D cameras as the reference camera, and the coordinate system of the reference line laser 3D camera as the reference coordinate system. By calibrating the rigid body transformation matrix between the line laser 3D cameras, make the coordinate system of the other line laser 3D cameras uniformly into the coordinate system of the reference line laser 3D camera, where the rigid body transformation matrix includes rotation matrix and translation vector; The specific steps of the ring layout calibration method include: obtaining the contour line, fitting a straight line, calculating the rotation matrix, calculating the translation vector, and calculating the rigid body transformation, wherein fitting the straight line includes finding the intersection point of the straight line and finding the direction vector of the straight line; A triangular prism calibration block is set up, and three line laser 3D cameras are evenly arranged in a ring around the vertical section of the calibration block at 120-degree intervals, so that each line laser 3D camera can observe one corner of the calibration block. The three line laser 3D cameras are labeled as line laser 3D camera P, line laser 3D camera Q, and line laser 3D camera M, respectively. The coordinate system of the line laser 3D cameras is set as XOZ, where X is the horizontal axis of the coordinate system, Z is the vertical axis of the coordinate system, O is the origin of the coordinate system, and the coordinate system of line laser 3D camera P is used as the reference coordinate system X. w O w Z w By calibrating the rigid body transformation matrix between the line laser 3D cameras, the coordinate system of the second line laser 3D camera Q and the coordinate system of the third line laser 3D camera M are brought into the coordinate system of the first line laser 3D camera P. The rigid body transformation matrix includes a rotation matrix and a translation vector. The specific steps of the ring layout calibration method include: S1: Obtain the outline, trigger the three line laser 3D cameras to take pictures, and each of the three line laser 3D cameras obtains one outline. S2: Fitting straight lines. The contour lines of the three linear laser 3D cameras are fitted with two adjacent straight lines, denoted as the first straight line L1 and the second straight line L2, respectively. The equations of the first straight line L1 and the second straight line L2 are: L1: A1X+B1Z+C1=0, L2: A2X+B2Z+C2=0; Where L1 is the first straight line of the line laser 3D camera, L2 is the second straight line of the line laser 3D camera, X is the horizontal axis of the coordinate system of the line laser 3D camera, Z is the vertical axis of the coordinate system of the line laser 3D camera, and A1, B1, C1, A2, B2, and C2 are constants. Find the intersection point V of the first line L1 and the second line L2 using the equations of the line laser 3D camera. 12 Find the direction vector D1 of the first line L1 and the direction vector D2 of the second line L2. Adjust the direction vectors of the first line L1 and the second line L2 so that both lines point to the intersection point P. 12 ; S3: Calculate the rotation matrix. The first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q capture the same face of the triangular prism calibration block. The first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q are in the reference coordinate system X. W O W Z W Given lines with the same direction, calculate the rotation angle between the first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q; let the direction vector of the first straight line PL1 to the left of the first line laser 3D camera P be D. p1 =(a p1 b p1 Let D be the direction vector of the second straight line QL2 to the right of the second line laser 3D camera Q. q2 =(a q2 ,b q2 The formula for calculating the rotation angle between the first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q is as follows: Where θ is the rotation angle between the first straight line PL1 to the left of the first line laser 3D camera P and the second straight line QL2 to the right of the second line laser 3D camera Q, dot() is the vector dot product operation, and D p1 Let D be the direction vector of the first straight line PL1 to the left of the first line laser 3D camera P. q2 The direction vector of the second straight line QL2 to the right of the second-line laser 3D camera Q; Finally, the rotation matrix between the first-line laser 3D camera P and the second-line laser 3D camera Q is obtained: wherein R pq is a rotation matrix between the first line laser 3D camera P and the second line laser 3D camera Q, and θ is a rotation angle of the first straight line PL1 on the left side of the first line laser 3D camera P and the second straight line QL2 on the right side of the second line laser 3D camera Q. Similarly, the rotation matrix R between the first line laser 3D camera P and the third line laser 3D camera M pm , and the rotation matrix R between the second line laser 3D camera Q and the third line laser 3D camera M qm are obtained. S4: Calculate the translation vector. Each vertical section of the triangular prism calibration block is an isosceles triangle. Let the side length of the triangular prism calibration block be S. Let the coordinate system of the first linear laser 3D camera P be the reference coordinate system X. w O w Z w In step S2, the coordinates of the intersection point of the first line laser 3D camera P and the straight line are calculated as V. p (x p , z p The coordinates of the intersection point of the lines calculated by the second-line laser 3D camera Q in step S2 are V. q (x q , z q If the coordinates of the intersection point V of the second-line laser 3D camera Q are given, then... q (x q , z q The coordinates in the first-line laser 3D camera P should be: Among them, V′ q Let x′ be the coordinates of the intersection point of the line between the second line laser 3D camera Q and the first line laser 3D camera P. q Let z′ be the coordinates of the intersection point of the lines from the second-line laser 3D camera Q with the first-line laser 3D camera P on the horizontal axis of the coordinate system. q Let x be the coordinates of the intersection point of the lines from the second-line laser 3D camera Q with the first-line laser 3D camera P, and x be the coordinates of the intersection point on the vertical axis of the coordinate system. p Let z be the coordinate of the intersection point of the lines from the first-line laser 3D camera P on the horizontal axis of the coordinate system. p Let a be the coordinate of the intersection point of the lines from the first-line laser 3D camera P on the vertical axis of the coordinate system. p1 Let D be the direction vector of the first straight line PL1 to the left of the first line laser 3D camera P. p1 In the coordinate system, the horizontal axis coordinate is b p1 Let D be the direction vector of the first straight line PL1 to the left of the first line laser 3D camera P. p1 In the coordinate system, S is the side length of the triangular prism calibration block, located on the vertical axis. Thus the translation relationship of the coordinate system of the second line laser 3D camera Q to the reference coordinate system X w O w Z w is obtained: Among them, t pq Let t be the translation vector between the first line laser 3D camera P and the second line laser 3D camera Q. x Let t be the coordinate of the translation vector between the first-line laser 3D camera P and the second-line laser 3D camera Q on the horizontal axis of the coordinate system. z Let x′ be the coordinate of the translation vector between the first-line laser 3D camera P and the second-line laser 3D camera Q on the vertical axis of the coordinate system. q Let z′ be the coordinates of the intersection point of the lines from the second-line laser 3D camera Q with the first-line laser 3D camera P on the horizontal axis of the coordinate system. q Let x be the coordinates of the intersection point of the lines from the second-line laser 3D camera Q with the first-line laser 3D camera P, and x be the coordinates of the intersection point on the vertical axis of the coordinate system. p Let z be the coordinate of the intersection point of the lines from the first-line laser 3D camera P on the horizontal axis of the coordinate system. p Let R be the coordinate of the intersection point of the lines from the first-line laser 3D camera P on the vertical axis of the coordinate system. pq Let be the rotation matrix between the first-line laser 3D camera P and the second-line laser 3D camera Q; Similarly, the translation vector t between the first-line laser 3D camera P and the third-line laser 3D camera M is obtained. pm And the translation vector t between the second-line laser 3D camera Q and the third-line laser 3D camera M qm ; S5: Calculate the rigid body transformation, complete the ring layout calibration, and obtain the rigid body transformation matrix by combining the rotation matrix and translation vector. For the first line laser 3D camera P and the second line laser 3D camera Q, the rigid body transformation matrix of the first line laser 3D camera P and the second line laser 3D camera Q is T. pq =[R pq ,t pq For any contour line coordinate point (x1, z1) in the second-line laser 3D camera Q, the contour line coordinate point (x1, z1) after transformation is in the coordinate system of the first-line laser 3D camera P, i.e., the reference coordinate system X. w O w Z w The coordinates are: Where x′1 is the coordinate of any contour line point (x1, z1) in the second line laser 3D camera Q after transformation on the horizontal axis of the first line laser 3D camera P, z′1 is the coordinate of any contour line point (x1, z1) in the second line laser 3D camera Q after transformation on the vertical axis of the first line laser 3D camera P, x1 is the coordinate of any contour line point (x1, z1) in the second line laser 3D camera Q on the horizontal axis of the coordinate system, and z1 is the coordinate of any contour line point (x1, z1) in the second line laser 3D camera Q on the vertical axis of the coordinate system, T pq Let be the rigid body transformation matrix of the first line laser 3D camera P and the second line laser 3D camera Q. Similarly, the rigid body transformation matrix T between the first-line laser 3D camera P and the third-line laser 3D camera M is obtained. pm By performing a corresponding coordinate transformation on the contour coordinates acquired by the third-line laser 3D camera M, the transformed coordinates of the contour points in the third-line laser 3D camera M can be obtained in the coordinate system of the first-line laser 3D camera P, i.e., the reference coordinate system X. w O w Z w The coordinates.

2. The ring layout calibration method of a line laser 3D camera according to claim 1, characterized in that: In step S2, the intersection point V of the first line L1 and the second line L2 is calculated. 12 Specifically, this includes: by simultaneously solving the equations of the first line L1 and the second line L2, the intersection point V of the first line L1 and the second line L2 can be obtained. 12 , Among them, V 12 Let L1 be the intersection point of the first line L1 and the second line L2, and let A1, B1, C1, A2, B2, and C2 be constants.

3. The ring layout calibration method of a line laser 3D camera according to claim 1 or 2, characterized in that: In step S2, the direction vectors of the first line L1 and the second line L2 are calculated, and the direction vectors of the first line L1 and the second line L2 are adjusted so that both the first line L1 and the second line L2 point to the intersection point F. 12 ; Where D1 is the direction vector of the first line L1, D2 is the direction vector of the second line L2, and A1, B1, A2, and B2 are constants.

Images