A calibration board for 3D camera point cloud data automatic correction and fusion

Abstract

The utility model discloses a kind of calibration plate for 3D camera point cloud data automatic correction and fusion, it is related to calibration plate technical field, it includes including rectangular flat plate and the pyramid structure of m column × n row being set on the rectangular flat plate, m, n are all positive integers;Each the pyramid structure has 6 faces, the 8 vertexes formed by the intersection of 6 faces, the periphery of each the pyramid structure is equipped with the circular mark for automatically determining the pyramid structure position number and scanning direction.By the pyramid structure of m column × n row uniform array, relying on the 8 um level precision known vertex formed by the intersection of 6 faces, can provide a large number of high-precision reference points, lay accurate foundation for point cloud correction and fusion;The circular mark of the periphery of pyramid can automatically determine position number and scanning direction, small round hole auxiliary identification direction, 4-6 bit binary coding large round hole hole clear number, without manual intervention, calibration efficiency is greatly improved.

Description

Technical Field

[0001] This utility model relates to the field of calibration board technology, specifically a calibration board for automatic correction and fusion of point cloud data from 3D cameras. Background Technology

[0002] In the field of 3D vision inspection, point cloud data acquired by 3D cameras often suffers from distortion. Furthermore, when multiple cameras are used for collaborative acquisition, point cloud fusion is required to obtain complete 3D information. This process relies on a calibration board to complete point cloud correction and fusion.

[0003] Existing calibration boards mostly provide reference points through specific structures, but they generally suffer from low levels of automation: during 3D camera scanning, such as... Figure 1 The existing calibration boards shown require manual specification of the scanning direction and the position number of the reference structure on the calibration board if only a partial area is collected or if there are forward and reverse scanning situations. Only then can the correspondence between the point cloud data and the true value of the calibration board be established. Manual intervention not only reduces calibration efficiency, but also easily affects the point cloud correction accuracy and fusion effect due to human operation errors. It is difficult to meet the actual application requirements of high-precision and automated 3D detection scenarios. Therefore, this application proposes a calibration board for automatic correction and fusion of 3D camera point cloud data to solve the above problems. Utility Model Content

[0004] (a) Technical problems to be solved

[0005] To address the shortcomings of existing technologies, this invention provides a calibration board for automatic correction and fusion of point cloud data from 3D cameras, thus solving the technical problems mentioned in the background.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the technical solution adopted by this utility model is: a calibration plate for automatic correction and fusion of 3D camera point cloud data, comprising a rectangular plate and an m-column × n-row pyramid structure disposed on the rectangular plate, where m and n are both positive integers; each pyramid structure has 6 faces, namely 1 bottom face, 1 top face and 4 side faces, the 6 faces intersect to form 8 vertices, and the coordinates of all the vertices are known micrometer-level machining accuracy coordinates;

[0008] Each of the pyramid structures is surrounded by a circular marker for automatically determining the pyramid structure's location number and scanning direction. The circular marker includes:

[0009] A small circular hole for assisting in identifying the orientation of the calibration plate: the small circular hole is located in the center of the left side of the corresponding pyramid structure, and the radius of the small circular hole is 0.4mm;

[0010] A set of horizontally arranged large circular holes are used to represent the location number of the corresponding pyramid structure: the large circular holes are located below the corresponding pyramid structure, the lower side of the pyramid structure is divided into 4-6 regions, and the large circular holes are selectively arranged in each region. A 4-6 bit binary code is formed by the binary encoding rule of "having a circle" as 1 and "not having a circle" as 0, to represent the location number of the pyramid structure. The range of the location number is 1-15, 1-31 or 1-63, and the radius of the large circular holes is 0.6 mm.

[0011] Preferably, the pyramid structure set on the rectangular plate is 3 columns × 4 rows, with a total of 12 pyramids. Each pyramid structure forms a 4-bit binary code through large circular holes in 4 equally divided areas below it, corresponding to a position number in the range of 1-15.

[0012] Preferably, of the six faces of the pyramid structure, the bottom face is defined as face 1, the side facing the large circular hole is defined as face 2, the side facing the small circular hole is defined as face 3, and along the rotation direction from face 2 to face 3, the remaining two sides are defined as face 4 and face 5 respectively, and the top face is defined as face 6.

[0013] Preferably, the true value data of the eight vertices of the pyramid structure are obtained by a coordinate measuring machine. The method of obtaining the data is as follows: with the bottom surface of the rectangular plate as the reference surface and the lower left corner of the rectangular plate as the origin of the coordinate system, a workpiece coordinate system is established. The plane parameters (x, y, z, i, j, k) of the first to sixth faces of each pyramid structure are measured in a preset order. Then, the true value data of the eight vertices are obtained by calculating the intersection of every three adjacent faces.

[0014] Preferably, the m-column × n-row pyramid structure is uniformly distributed in an array on the rectangular plate.

[0015] (III) Beneficial Effects

[0016] The beneficial effects of this utility model are as follows:

[0017] This calibration board for automatic correction and fusion of 3D camera point cloud data uses a pyramid structure with an m-column × n-row uniform array. Relying on 8 known vertices with micrometer-level precision formed by the intersection of 6 faces, it can provide a large number of high-precision reference points, laying a precise foundation for point cloud correction and fusion. The circular markers around the pyramid can automatically determine the position number and scanning direction, small circular holes assist in direction identification, and large circular holes with 4-6 bit binary codes clearly number the holes. No manual intervention is required, which greatly improves calibration efficiency and avoids human error. Thirdly, the specific array design of 3 columns × 4 rows and the standardized vertex truth acquisition method are suitable for various detection scenarios and effectively meet the needs of high-precision and automated 3D detection. Attached Figure Description

[0018] Figure 1 This is a commonly used pyramid calibration plate;

[0019] Figure 2 This is a schematic diagram of the six faces and eight vertices of the pyramid of this utility model;

[0020] Figure 3 This is a schematic diagram of the circular marker surrounding the pyramid of this utility model;

[0021] Figure 4 This is a schematic diagram showing the location of the circular marker in the pyramid area, which is numbered 1 in this utility model.

[0022] Figure 5 This is a schematic diagram showing the positions of the circular markers numbered 2 / 4 / 8 in binary code representation of this utility model.

[0023] Figure 6 The numbers and corresponding marker circles for the 12 pyramids on the 3x4 calibration plate of this utility model;

[0024] Figure 7 This is a schematic diagram showing the true value data measured by a coordinate measuring machine for the calibration plate of this utility model;

[0025] Figure 8 This is a point cloud diagram of a single pyramid of this utility model. Detailed Implementation

[0026] The technical solutions in the embodiments of this utility model will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of this utility model, and not all embodiments. Based on the embodiments of this utility model, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this utility model.

[0027] like Figure 1-8 As shown, this utility model provides a technical solution: a calibration plate for automatic correction and fusion of 3D camera point cloud data, comprising a rectangular flat plate and an m-column × n-row pyramid structure disposed on the rectangular flat plate, where m and n are both positive integers, and the m-column × n-row pyramid structure is uniformly arrayed on the rectangular flat plate; each pyramid structure has 6 faces, such as... Figure 2As shown, the pyramid structure has one bottom surface, one top surface, and four side surfaces. The bottom surface is defined as surface 1, the side facing the large circular hole as surface 2, and the side facing the small circular hole as surface 3. Along the rotation direction from surface 2 to surface 3, the remaining two side surfaces are defined as surfaces 4 and 5, respectively, and the top surface is defined as surface 6. These six surfaces intersect to form eight vertices, and the coordinates of all vertices are known to be at the μm level for machining accuracy. The true values ​​of the eight vertices of the pyramid structure are obtained using a coordinate measuring machine (CMM). The method is as follows: using the bottom surface of the rectangular plate as the reference surface and the lower left corner of the rectangular plate as the origin of the coordinate system, a workpiece coordinate system is established. The planar parameters (x, y, z, i, j, k) of surfaces 1-6 of each pyramid structure are measured in a preset order. Then, the true values ​​of the eight vertices are calculated by intersecting every three adjacent surfaces. When obtaining the true values ​​of the pyramid corners on the calibration plate, a CMM is used according to... Figure 7 The x and y directions are shown. The bottom surface is used as the reference plane and the lower left corner is the origin of the coordinate system. The workpiece coordinate system is established. The plane parameters (x, y, z, i, j, k) of the 1st to 6th faces of each pyramid are measured in the order of #1-#8 and saved as true data. A corner point can be calculated for every three planes intersecting. A total of 8 corner point true data can be extracted for each pyramid.

[0028] Each pyramid structure is surrounded by a circular marker for automatically determining its position number and scanning direction. The circular marker includes: a small circular hole for assisting in identifying the calibration plate direction; this small circular hole is located centered on the left side of the corresponding pyramid structure, and its radius is 0.4 mm; and a set of horizontally arranged large circular holes for indicating the position number of the corresponding pyramid structure; these large circular holes are located below the corresponding pyramid structure, which is divided into 4-6 equal regions. Each region selectively contains one of these large circular holes, and a 4-6 bit binary code is formed using a binary encoding rule of 1 for "circle present" and 0 for "no circle present" to represent the position number of the pyramid structure. The position number ranges from 1-15, 1-31, or 1-63, and the radius of each large circular hole is 0.6 mm. Figure 4 As shown, a small circle for identifying the direction is located in the center of the left side of the pyramid, and a set of large circles for determining the position number is located on the bottom side of the pyramid. Divide the bottom side of the pyramid into four equal parts to obtain four regions. Each region may or may not have a circle, thus forming a 4-bit binary code to represent the pyramid number. The 4-bit code can obtain the pyramid number range 1-15. If there are many pyramids, the bottom side of the pyramid can also be divided into five or six equal parts to obtain a five-bit binary code (1-31) or a six-bit binary code (1-63).

[0029] The pyramid structure set on the rectangular plate is 3 columns × 4 rows, with a total of 12 pyramids. Each pyramid structure forms a 4-bit binary code through large circular holes in 4 equally divided areas below it, corresponding to a position number in the range of 1-15.

[0030] To achieve automated calibration, a set of horizontally arranged circular holes below the pyramid indicates the location number of each pyramid. A smaller circular hole with a radius of 0.4 mm, centered on the left side of the pyramid, is used to assist in identifying the orientation of the calibration plate. Figure 3 It indicates the location of the circular markers surrounding the pyramids. Figure 4 This is an example of the location of the large circular hole (0.6mm), numbered 1. The large circular hole is located at the lower left corner of the pyramid. By combining the location of the large circular hole with binary data, we can deduce that this pyramid is numbered 1. Figure 5 This is a diagram showing the positions of the numbered circles on pyramids numbered 2 / 4 / 8;

[0031] Figure 6 It is a 3x4 calibration board with 12 pyramids on it. When the 3D camera scans, each pyramid can be numbered by the surrounding marker circle. Even if the 3D camera only scans part of the calibration board area, it can automatically obtain the number of each pyramid.

[0032] When obtaining the true values ​​of the corner points of the pyramid on the calibration plate, a coordinate measuring machine is used according to... Figure 7 The x and y directions are shown. The bottom surface is used as the reference plane and the lower left corner is the origin of the coordinate system. The workpiece coordinate system is established. The plane parameters (x, y, z, i, j, k) of the 1st to 6th faces of each pyramid are measured in the order of #1-#8 and saved as true data. A corner point can be calculated for every three intersecting planes. A total of 8 corner point true data can be extracted for each pyramid.

[0033] When obtaining the measured values ​​of corner points in a pyramid point cloud, take a single pyramid in the point cloud as an example, and refer to... Figure 8 Each pyramid has 6 faces: the bottom face is face 1, the side facing the large circle is face 2, the side facing the small circle is face 3, the remaining two sides are face 4 and face 5 in this rotational order, and the top face is face 6. Every three planes intersect to calculate a corner point, and a total of 8 corner points can be extracted from each pyramid.

[0034] The operational steps for this application are as follows:

[0035] Place the calibration plate with an m-column × n-row uniform array pyramid structure stably, ensuring that the bottom surface of the rectangular plate is horizontal; start the 3D camera to scan the calibration plate. There is no need to limit the scanning area, and the camera can collect part or all of the point cloud data of the calibration plate.

[0036] After scanning, the system automatically identifies the circular markers around each pyramid. The scanning direction of the calibration plate is determined through the small circular hole with a radius of 0.4mm in the center on the left. Through the large circular hole with a radius of 0.6mm at the bottom, the system parses the 4-6 bit binary code according to the rule of "1 if there is a circle, 0 if there is no circle" to obtain the position number of each pyramid. For example, the 3-column × 4-row calibration plate corresponds to numbers 1-15 through 4 bits of code.

[0037] The system calls up the pre-stored vertex true value data, which is obtained by using a coordinate measuring machine to measure the parameters of pyramid faces 1-6 and calculate the coordinates of 8 vertices with the bottom surface of the flat plate as the reference surface and the lower left corner as the origin. At the same time, it fits each pyramid face 1-6 from the point cloud data and calculates the measured values ​​of 8 vertices by intersecting every three adjacent faces.

[0038] Based on the pyramid position numbering, the correspondence between the vertex true value and the measured value is established, and the transformation relationship between the point cloud and the calibration plate is calculated. Using this relationship, the point cloud with distortion is corrected. If it is acquired by multiple cameras, multiple sets of point clouds can be fused to obtain accurate and complete 3D point cloud data.

[0039] In the description of this utility model, it should be understood that the indicated orientation or positional relationship is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing this utility model and simplifying the description, and is not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this utility model.

[0040] In this utility model, unless otherwise explicitly specified and limited, for example, it can be a fixed connection, a detachable connection, or an integral part; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components or an interaction between two components. Unless otherwise explicitly limited, those skilled in the art can understand the specific meaning of the above terms in this utility model according to the specific circumstances.

[0041] Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A calibration board for automatic correction and fusion of point cloud data from 3D cameras, characterized in that: It includes a rectangular plate and an m-column × n-row pyramid structure set on the rectangular plate, where m and n are both positive integers; each pyramid structure has 6 faces, namely 1 bottom face, 1 top face and 4 side faces, and the 6 faces intersect to form 8 vertices; Each of the pyramid structures is surrounded by a circular marker for automatically determining the pyramid structure's location number and scanning direction. The circular marker includes: A small circular hole for assisting in identifying the orientation of the calibration plate: the small circular hole is located in the center of the left side of the corresponding pyramid structure, and the radius of the small circular hole is 0.4mm; A set of horizontally arranged large circular holes are used to represent the location number of the corresponding pyramid structure: the large circular holes are located below the corresponding pyramid structure, the lower side of the pyramid structure is divided into 4-6 regions, and the large circular holes are selectively arranged in each region. A 4-6 bit binary code is formed by the binary encoding rule of "having a circle" as 1 and "not having a circle" as 0, to represent the location number of the pyramid structure. The range of the location number is 1-15, 1-31 or 1-63, and the radius of the large circular holes is 0.6 mm.

2. The calibration board for automatic correction and fusion of 3D camera point cloud data according to claim 1, characterized in that: The rectangular plate has 12 pyramid structures arranged in 3 columns and 4 rows. Each pyramid structure forms a 4-bit binary code through large circular holes in 4 equally divided areas below it, corresponding to a position number in the range of 1-15.

3. A calibration board for automatic correction and fusion of point cloud data from a 3D camera according to claim 1, characterized in that: Of the six faces of the pyramid structure, the bottom face is defined as face 1, the side facing the large circular hole is defined as face 2, the side facing the small circular hole is defined as face 3, and the remaining two sides along the rotation direction from face 2 to face 3 are defined as face 4 and face 5 respectively, and the top face is defined as face 6.

4. A calibration board for automatic correction and fusion of 3D camera point cloud data according to claim 1, characterized in that: The true data of the eight vertices of the pyramid structure were obtained by a coordinate measuring machine. The method of obtaining the data was as follows: the bottom surface of the rectangular plate was used as the reference surface, and the lower left corner of the rectangular plate was used as the origin of the coordinate system. The workpiece coordinate system was established, and the plane parameters (x, y, z, i, j, k) of the 1-6 faces of each pyramid structure were measured in a preset order. The true data of the eight vertices were then obtained by calculating the intersection of every three adjacent faces.

5. A calibration board for automatic correction and fusion of 3D camera point cloud data according to claim 1, characterized in that: The m-column × n-row pyramid structure is uniformly distributed on the rectangular plate.

Images