Double-rotating-shaft 3D camera measuring system

The calibration process of the dual-axis 3D camera measurement system simplifies the calibration steps, improves the accuracy and efficiency of the measurement, and solves the problems of low efficiency and inaccurate results caused by the complexity of calibration in the existing technology.

CN118967829APending Publication Date: 2024-11-15SUZHOU RUIXIN TECH CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411011516.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-26
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

Existing 3D camera measurement systems are complex to calibrate, resulting in low measurement efficiency and inaccurate results, making it impossible to achieve precise measurements.

Method used

A dual-axis 3D camera measurement system was designed. Through the calibration process of steps one to nine, including determining the coordinate system center, establishing the coordinate system, predicting the camera position, constructing the transformation matrix, and rotating the camera axis, the global transformation of the camera coordinates is achieved, simplifying the calibration process.

Benefits of technology

It enables accurate and convenient calibration of the 3D camera measurement system, improves the accuracy and reliability of measurement, and enhances measurement efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118967829A_ABST
    Figure CN118967829A_ABST
Patent Text Reader

Abstract

The invention discloses a double-rotating-shaft 3D camera measurement system, which comprises a first rotating shaft, a second rotating shaft is arranged on the first rotating shaft, and a 3D camera is arranged on the second rotating shaft, and the calibration steps are as follows: step 1, determining the center of a coordinate system; 2, establishing a coordinate system; 3, pre-measuring the position of the 3D camera in the center; 4, constructing a transformation matrix; 5, determining the relation between the initial camera midpoint and the global coordinate midpoint; 6, converting the initial camera coordinates into global coordinates; 7, rotating the first rotating shaft to obtain three new coordinate axes of the camera; 8, rotating the second rotating shaft to obtain three new coordinate axes of the camera; and 9, combining the three axes and the center of the camera into a new transformation matrix. According to the technical scheme, the 3D camera measuring system is accurate and convenient in calibration, convenient to calibrate, easy to implement and high in precision, and by applying the 3D camera measuring system, the measuring accuracy and reliability are effectively improved, and meanwhile, the measuring efficiency is also effectively improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of measurement technology, in particular to a dual-rotation axis 3D camera measurement system. Background Art

[0002] Product measurement is the whole operation aimed at determining the value of the measured object. In this operation, the measured object is compared with the standard quantity that reproduces the measurement unit, and the measurement result is expressed as the ratio of the measured quantity to the unit quantity and its accuracy. According to the different ways of obtaining the measurement results, it can be divided into: direct measurement and indirect measurement. Directly obtaining the measured value or the deviation from the standard value from the reading device of the measuring instrument is called direct measurement, such as measuring the shaft diameter with a vernier caliper or outside micrometer. Indirect measurement is the measurement of the quantity that has a certain functional relationship with the measured quantity and obtaining the measured quantity according to the known functional relationship, such as obtaining the actual value of the arc radius by measuring the corresponding bow height and chord length of an arc.

[0003] During the measurement process, whether or not the product is touched can be divided into contact measurement and non-contact measurement: the measurement where the probe of the measuring instrument contacts the surface of the measured object and there is a mechanical force is contact measurement, such as the measurement of surface roughness using a light section microscope, which belongs to non-contact measurement. 3D camera measurement is a measurement method based on three-dimensional imaging technology. It achieves measurement and analysis by capturing the three-dimensional shape, size and position information of the object. 3D camera measurement is widely used in the field of industrial manufacturing for part size detection, shape analysis, assembly verification, etc. It can help manufacturers improve product quality, reduce production costs and shorten production cycles.

[0004] The existing 3D camera measurement system has complex calibration in practical use, which leads to low efficiency of 3D camera measurement, inaccurate measurement results, and inability to achieve accurate measurement and analysis of the measured products. Therefore, a dual-axis 3D camera measurement system is designed. Summary of the invention

[0005] The purpose of the present invention is to solve the above problems and designs a dual-axis 3D camera measurement system.

[0006] To achieve the above objective, the technical solution of the present invention is a dual-axis 3D camera measurement system, comprising a first axis, a second axis is arranged on the first axis, a 3D camera is arranged on the second axis, and the calibration steps are as follows:

[0007] Step 1: Determine the center of the coordinate system;

[0008] Step 2: Establish a coordinate system;

[0009] Step 3: Pre-measure the position of the 3D camera at the center;

[0010] Step 4: Construct the transformation matrix;

[0011] Step 5: Determine the relationship between the initial camera midpoint and the global coordinate midpoint;

[0012] Step 6: Convert the initial camera coordinates to global coordinates;

[0013] Step 7: Rotate the first axis to get three new coordinate axes of the camera;

[0014] Step 8: Then rotate the second axis to obtain three new coordinate axes of the camera;

[0015] Step 9: Combine the three axes of the camera and the center into a new transformation matrix.

[0016] As a further description of the technical solution, in step one, the axis centers of the first rotation axis and the second rotation axis are defined as the centers of the global space coordinate system.

[0017] As a further description of the present technical solution, in step 2, the initial camera position is oriented in the same direction as axis 1 and in the same direction as axis 2 in the horizontal direction. The x-axis CX of its coordinate system is the z-axis (0,0,1) of the global coordinate system, the y-axis of its coordinate system is the x-axis (1,0,0) of the CY global coordinate system, and the z-axis of its coordinate system is the y-axis (0,1,0) of the CZ global coordinate system.

[0018] As a further description of the technical solution, in step three, according to the camera size, the position of the camera center Center in the center of the global coordinate system is pre-measured as (dx, dy, dz).

[0019] As a further description of the technical solution, in step 4, the following transformation matrix can be constructed according to the initial camera posture:

[0020]

[0021] As a further description of the technical solution, in step 5, the relationship is as follows: Global = RP;

[0022] Global is the point coordinate in the global coordinate system, R is the transformation matrix constructed by the initial camera pose, and P is the coordinate of the initial camera midpoint.

[0023] As a further description of the technical solution, in step six, according to Global=RP, the initial camera coordinates are converted into global coordinates.

[0024] As a further description of the technical solution, in step 7, the first axis of rotation is essentially a rotation of θ° around the z-axis.

[0025] According to the Rodrigues rotation formula, vector P rotates around A by angle θ to obtain a new vector

[0026] P'=P*cosθ+(A×P)sinθ+A(A·P)(1-cosθ)

[0027] Therefore, we can get three new coordinate axes of the camera

[0028] The new X-axis CX1 is CX*cosθ+(A1×CX)sinθ+A1(A1·CX)(1-cosθ); where CX is the original camera X-axis, θ is the first rotation angle, and A1 is the standard unit z-axis;

[0029] The same can be said

[0030] The new Y axis CY1 is CY*cosθ+(A1×CY)sinθ+A1(A1·CY)(1-cosθ)

[0031] The new Z axis CZ1 is CZ*cosθ+(A1×CZ)sinθ+A1(A1·CZ)(1-cosθ)

[0032] During rotation, the camera center also changes in the spatial coordinate system.

[0033] Center1 is Center*cosθ+(A1×Center)sinθ+A1(A1·Center)(1-cosθ)

[0034] Where Center is the original camera center and A1 is the standard unit z-axis.

[0035] As a further description of the technical solution, in step eight, the second axis rotation is essentially a rotation of β° around the current camera y axis Cy1.

[0036] At this time, according to the Rodrigues rotation formula, the three new coordinate axes of the camera and the new camera center can be obtained;

[0037] The new X-axis CX2 is CX1*cosβ+(A2×CX1)sinβ+A2(A2·CX1)(1-cosβ), where CX1 is the X-axis of the camera after one rotation, β is the second rotation angle, and A2 is the current camera y-axis Cy1.

[0038] The same can be said

[0039] The new Y axis CY2 is CY1*cosβ+(A2×CY1)sinβ+A2(A2·CY1)(1-cosβ) The new Z axis CY2 is CZ1*cosβ+(A2×CZ1)sinβ+A2(A2·CZ1)(1-cosβ)

[0040] New Camera Center

[0041] Center2 is Center1*cosβ+(A2×Center1)sinβ+A2(A2·Center1)(1-cosβ).

[0042] As a further description of the technical solution, in step nine, the new transformation matrix is ​​as follows:

[0043]

[0044] At this time, the global coordinates and the camera coordinates have the following relationship

[0045] Global=R2P

[0046] Global is the point coordinate in the global coordinate system, R2 is the transformation matrix constructed by the camera pose, and P is the coordinate of the camera midpoint

[0047] According to this formula, the camera coordinates after dual-axis rotation can be converted to global coordinates

[0048] Finally, both the initial camera and the camera after dual-axis rotation are transformed into the global coordinate system, thus achieving dual-camera calibration.

[0049] The beneficial effect is that the 3D camera measurement system of the technical solution is accurate and convenient to calibrate, easy to calibrate, and has high precision. By using this 3D camera measurement system, the accuracy and reliability of measurement are effectively improved through precise calibration, and at the same time, the efficiency of measurement is also effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic diagram of the installation of the dual-axis 3D camera of the present invention.

[0051] In the figure, 1, first rotating shaft; 2, second rotating shaft; 3, 3D camera; DETAILED DESCRIPTION

[0052] First of all, the original intention of the design of the present invention is explained. 3D camera measurement is widely used in the field of industrial manufacturing for size detection, shape analysis, assembly verification, etc. of parts. It can help manufacturers improve product quality, reduce production costs and shorten production cycles. The existing 3D camera measurement system is complex to calibrate during practical use, which leads to low efficiency of 3D camera measurement, inaccurate measurement results, and inability to achieve accurate measurement and analysis of the measured product. Therefore, the present invention designs a dual-axis 3D camera measurement system.

[0053] The present invention will be described in detail below in conjunction with the accompanying drawings. Figure 1As shown, a dual-axis 3D camera measurement system includes a first axis 1, a second axis 2 is arranged on the first axis 1, and a 3D camera 3 is arranged on the second axis 2. The calibration steps are as follows:

[0054] Step 1: Determine the center of the coordinate system; in step 1, the axis of the first rotation axis 1 and the second rotation axis 2 is determined as the center of the global space coordinate system.

[0055] Step 2: Establish a coordinate system; in step 2, let the initial camera position be in the same direction as axis 1, and the horizontal direction be in the same direction as axis 2, the x-axis CX of its coordinate system is the z-axis (0,0,1) of the global coordinate system, the y-axis of its coordinate system is the x-axis (1,0,0) of the CY global coordinate system, and the z-axis of its coordinate system is the y-axis (0,1,0) of the CZ global coordinate system.

[0056] Step 3: Pre-measure the position of the 3D camera at the center; in step 3, based on the camera size, the position of the camera center Center at the center of the global coordinate system is measured in advance as (dx, dy, dz).

[0057] Step 4: Construct a transformation matrix; in step 4, the following transformation matrix can be constructed according to the initial camera posture:

[0058]

[0059] Step 5: Determine the relationship between the initial camera midpoint and the global coordinate midpoint; in step 5, the relationship is as follows: Global = RP;

[0060] Global is the point coordinate in the global coordinate system, R is the transformation matrix constructed by the initial camera pose, and P is the coordinate of the initial camera midpoint.

[0061] Step 6: Convert the initial camera coordinates to global coordinates; In step 6, according to Global=RP, the initial camera coordinates are converted to global coordinates.

[0062] Step 7: The first axis 1 rotates to obtain three new coordinate axes of the camera; in step 7, the first axis rotates, which is essentially a rotation of θ° around the z axis.

[0063] According to the Rodrigues rotation formula, vector P rotates around A by angle θ to obtain a new vector

[0064] P'=P*cosθ+(A×P)sinθ+A(A·P)(1-cosθ)

[0065] Therefore, we can get three new coordinate axes of the camera

[0066] The new X-axis CX1 is CX*cosθ+(A1×CX)sinθ+A1(A1·CX)(1-cosθ); where CX is the original camera X-axis, θ is the first rotation angle, and A1 is the standard unit z-axis;

[0067] The same can be said

[0068] The new Y axis CY1 is CY*cosθ+(A1×CY)sinθ+A1(A1·CY)(1-cosθ)

[0069] The new Z axis CZ1 is CZ*cosθ+(A1×CZ)sinθ+A1(A1·CZ)(1-cosθ)

[0070] During rotation, the camera center also changes in the spatial coordinate system.

[0071] Center1 is Center*cosθ+(A1×Center)sinθ+A1(A1·Center)(1-cosθ)

[0072] Where Center is the original camera center and A1 is the standard unit z-axis.

[0073] Step 8: Then the second axis 2 rotates to obtain three new coordinate axes of the camera; in step 8, the second axis 2 rotates, which is essentially a rotation of β° around the current camera y axis Cy1

[0074] At this time, according to the Rodrigues rotation formula, the three new coordinate axes of the camera and the new camera center can be obtained;

[0075] The new X-axis CX2 is CX1*cosβ+(A2×CX1)sinβ+A2(A2·CX1)(1-cosβ), where CX1 is the X-axis of the camera after one rotation, β is the second rotation angle, and A2 is the current camera y-axis Cy1.

[0076] The same can be said

[0077] The new Y axis CY2 is CY1*cosβ+(A2×CY1)sinβ+A2(A2·CY1)(1-cosβ) The new Z axis CY2 is CZ1*cosβ+(A2×CZ1)sinβ+A2(A2·CZ1)(1-cosβ)

[0078] New Camera Center

[0079] Center2 is Center1*cosβ+(A2×Center1)sinβ+A2(A2·Center1)(1-cosβ).

[0080] Step 9: Combine the three axes and the center of the camera into a new transformation matrix. In step 9, the new transformation matrix is ​​as follows:

[0081]

[0082] At this time, the global coordinates and the camera coordinates have the following relationship

[0083] Global=R2P

[0084] Global is the point coordinate in the global coordinate system, R2 is the transformation matrix constructed by the camera pose, and P is the coordinate of the camera midpoint

[0085] According to this formula, the camera coordinates after dual-axis rotation can be converted to global coordinates

[0086] Finally, both the initial camera and the camera after dual-axis rotation are transformed into the global coordinate system, thus achieving dual-camera calibration.

[0087] The 3D camera measurement system of the present technical solution is accurate and convenient in calibration, easy to implement, and has high precision. By using this 3D camera measurement system, the accuracy and reliability of measurement are effectively improved through precise calibration, and at the same time, the efficiency of measurement is also effectively improved.

[0088] The above technical solutions only reflect the preferred technical solutions of the technical solutions of the present invention. Some changes that may be made to certain parts thereof by technicians in this technical field all reflect the principles of the present invention and fall within the protection scope of the present invention.

Claims

1. A dual-axis 3D camera measurement system, characterized in that: The device comprises a first rotating shaft (1), a second rotating shaft (2) is arranged on the first rotating shaft (1), a 3D camera (3) is arranged on the second rotating shaft (2), and the calibration steps are as follows: Step 1: Determine the center of the coordinate system; Step 2: Establish a coordinate system; Step 3: Pre-measure the position of the 3D camera at the center; Step 4: Construct the transformation matrix; Step 5: Determine the relationship between the initial camera midpoint and the global coordinate midpoint; Step 6: Convert the initial camera coordinates to global coordinates; Step 7: Rotate the first axis (1) to obtain three new coordinate axes of the camera; Step 8: Then rotate the second axis (2) to obtain three new coordinate axes of the camera; Step 9: Combine the three axes of the camera and the center into a new transformation matrix.

2. A dual-axis 3D camera measurement system according to claim 1, characterized in that: In the step 1, the axis centers of the first rotation axis (1) and the second rotation axis (2) are defined as the centers of the global space coordinate system.

3. A dual-axis 3D camera measurement system according to claim 1, characterized in that: In step 2, the initial camera position is oriented in the same direction as axis 1 and in the same direction as axis 2. The x-axis CX of its coordinate system is the z-axis (0,0,1) of the global coordinate system, the y-axis of its coordinate system is the x-axis (1,0,0) of the CY global coordinate system, and the z-axis of its coordinate system is the y-axis (0,1,0) of the CZ global coordinate system.

4. The dual-axis 3D camera measurement system according to claim 1, characterized in that: In the step three, according to the camera size, the position of the camera center Center in the center of the global coordinate system is measured in advance as (dx, dy, dz).

5. The dual-axis 3D camera measurement system according to claim 1, characterized in that: In step 4, the following transformation matrix can be constructed according to the initial camera posture:

6. The dual-axis 3D camera measurement system according to claim 1, characterized in that: In step 5, the relationship is as follows: Global = RP; Global is the point coordinate in the global coordinate system, R is the transformation matrix constructed by the initial camera pose, and P is the coordinate of the initial camera midpoint.

7. The dual-axis 3D camera measurement system according to claim 1, characterized in that: In the step six, according to Global=RP, the initial camera coordinates are converted into global coordinates.

8. The dual-axis 3D camera measurement system according to claim 1, characterized in that: In step 7, the first rotation axis (1) rotates, which is essentially a rotation of θ° around the z-axis. According to the Rodrigues rotation formula, vector P rotates around A by angle θ to obtain a new vector P'=P*cosθ+(A×P)sinθ+A(A·P)(1-cosθ) Therefore, we can get three new coordinate axes of the camera The new X-axis CX1 is CX*cosθ+(A1×CX)sinθ+A1(A1·CX)(1-cosθ); where CX is the original camera X-axis, θ is the first rotation angle, and A1 is the standard unit z-axis; The same can be said The new Y axis CY1 is CY*cosθ+(A1×CY)sinθ+A1(A1·CY)(1-cosθ) The new Z axis CZ1 is CZ*cosθ+(A1×CZ)sinθ+A1(A1·CZ)(1-cosθ) During rotation, the camera center also changes in the spatial coordinate system. Center1 is Center*cosθ+(A1×Center)sinθ+A1(A1·Center)(1-cosθ) where Center is the original camera center and A1 is the standard unit z-axis.

9. The dual-axis 3D camera measurement system according to claim 1, characterized in that: In step 8, the second axis (2) rotates, which is essentially a rotation of β° around the current camera y-axis Cy1. At this time, according to the Rodrigues rotation formula, the three new coordinate axes of the camera and the new camera center can be obtained; The new X-axis CX2 is CX1*cosβ+(A2×CX1)sinβ+A2(A2·CX1)(1-cosβ), where CX1 is the X-axis of the camera after one rotation, β is the second rotation angle, and A2 is the current camera y-axis Cy1. The same can be said The new Y axis CY2 is CY1*cosβ+(A2×CY1)sinβ+A2(A2·CY1)(1-cosβ) The new Z axis CY2 is CZ1*cosβ+(A2×CZ1)sinβ+A2(A2·CZ1)(1-cosβ) New Camera Center Center2 is Center1*cosβ+(A2×Center1)sinβ+A2(A2·Center1)(1-cosβ).

10. The dual-axis 3D camera measurement system according to claim 1, characterized in that: In step nine, the new transformation matrix is ​​as follows: At this time, the global coordinates and the camera coordinates have the following relationship Global=R2P Global is the point coordinate in the global coordinate system, R2 is the transformation matrix constructed by the camera pose, and P is the coordinate of the camera midpoint According to this formula, the camera coordinates after dual-axis rotation can be converted to global coordinates Finally, both the initial camera and the camera after dual-axis rotation are transformed into the global coordinate system, thus achieving dual-camera calibration.

Citation Information

Cited By

  • Refueling taper sleeve center positioning method based on binocular vision of deep learning

    CN121414854A