A method and system for calibrating the extrinsic parameters of a single line structured light system

Abstract

The application discloses a kind of single group line structure light system's external parameter calibration method and system, belong to precision testing technology and instrument measurement technical field.The method includes: clamping special-shaped calibration part and adjusting the position of line structure light sensor;Special-shaped calibration part is integrated with three types of composite features of outer cylindrical surface, plane and V-shaped groove;Establish the coordinate conversion model between sensor coordinate system and special-shaped calibration part coordinate system, model contains external parameter to be calibrated;Drive special-shaped calibration part to rotate, extract the geometric feature of three types of composite features by line structure light sensor;Based on the geometric feature, construct distance objective function, and solve external parameter by using nonlinear optimization algorithm iteratively;Based on the optimized external parameter, reconstruct the three-dimensional model of special-shaped calibration part, compare actual geometric parameter to verify calibration accuracy, and if it meets preset threshold, calibration is completed.The calibration process of the application is simple, and the system structure is simple, suitable for high-precision, high-efficiency external parameter calibration of multi-feature special-shaped parts.

Description

Technical Field

[0001] This invention relates to the field of precision testing technology and instrument measurement technology, and more specifically to a method and system for calibrating the external parameters of a single-group linear structured light system. Background Technology

[0002] With the improvement of the resolution of line structured light sensors, non-contact measurement has advantages such as speed, efficiency, and comprehensiveness, and is widely used in aerospace, new energy vehicles, and other fields. Accurate calibration is crucial to ensuring the measurement accuracy of structured light systems. Commercially available line structured light sensors typically undergo rigorous internal and external parameter calibration before leaving the factory, eliminating measurement errors inherent to the sensor itself. However, when installing line structured light sensors to build a complete measurement system, external parameter calibration becomes an important and necessary task. Even small deviations in the installation position and orientation can significantly affect measurement accuracy; therefore, accurate external parameter calibration methods are often key to improving overall measurement accuracy.

[0003] Currently, calibration components and calibration methods are classified according to their measurement strategies and application scope as follows: (1) Global measurement strategy calibration method. As shown in patent CN201810089753.2, multiple line structured light sensors are used for scanning, and the measurement data obtained by multiple line structured light sensors are spliced ​​together according to the calibration position to achieve global coverage measurement of the calibration component. Although this strategy can obtain complete data and make its calibration work not limited to the range of line structured light sensors, the system is complex, the calibration process is cumbersome, and it takes a long time. It also has high requirements for the consistency and data processing capabilities of line structured light sensors. (2) Dedicated measurement strategy calibration method. As shown in patent CN201911321041.X, a specific motion mechanism rotating platform is used in conjunction with a dedicated line structured light sensor to achieve efficient measurement of specific types of calibration components such as ring parts. Although this strategy is efficient in specific scenarios, it is generally aimed at calibration components with single features, lacks versatility, and is difficult to apply to irregularly shaped calibration components with multiple structural features. How to achieve rapid and accurate calibration of external parameters in a measurement system with only a single set of line structured light remains a pressing problem to be solved in the field of precision testing technology and instrument measurement technology. Summary of the Invention

[0004] This invention addresses the problems existing in the extrinsic parameter calibration methods of existing single-line structured light systems by providing a method and system for extrinsic parameter calibration of single-line structured light systems. It fundamentally solves the problems of difficult and cumbersome extrinsic parameter calibration of single-line structured light systems by iteratively optimizing the feature point cloud using the Levenberg-Marquardt algorithm.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] In a first aspect, embodiments of the present invention provide a method for calibrating the external parameters of a single-group linear structured light system, comprising the following steps: S1: Clamp the irregularly shaped calibration component and adjust the position and orientation of the line structured light sensor so that its laser plane covers the composite features on the surface of the irregularly shaped calibration component; the irregularly shaped calibration component integrates three types of composite features: outer cylindrical surface, plane and V-groove; S2: Establish a coordinate transformation model between the sensor coordinate system and the irregular calibration part coordinate system. This model includes the external parameters to be calibrated; the external parameters include the rotation angle (…). Translation () ); S3: Drive the irregularly shaped calibration part to rotate, collect its surface point cloud data through the line structured light sensor, and extract the geometric features of the three types of composite features; S4: Based on the geometric features, construct a distance objective function, and use the Levenberg-Marquardt nonlinear optimization algorithm to minimize the sum of squared distances from the measurement point to the three types of composite feature surfaces as the optimization objective, and iteratively solve for the extrinsic parameters; S5: Reconstruct the 3D model of the irregular calibration part based on the optimized extrinsic parameters, and verify the calibration accuracy by comparing it with the actual geometric parameters. If the preset threshold is met, the calibration is completed.

[0007] In one embodiment, the irregular calibration element includes: an outer cylindrical surface feature, two planar features, and two V-groove features; The outer cylindrical surface has an inner hole, and the axis of the outer cylindrical surface is coaxial with the center line of the inner hole; the two planes are perpendicular to each other and symmetrically distributed around the geometric center of the irregular calibration part; the two V-grooves are respectively set on the two planes and are not completely connected, and are symmetrically distributed with the axes of the upper and lower end faces of the irregular calibration part; the angle of the V-groove is 120°.

[0008] In one embodiment, the clamping of the irregularly shaped calibration component in step S1 is achieved by the automatic centering characteristic of the three-jaw chuck, ensuring that the inner hole of the calibration component is coaxial with the rotation axis; and the line structured light sensor is rotated 90° relative to the vertical installation state and fixed, and the measurement output of the line structured light sensor is adjusted to zero.

[0009] In one embodiment, in step S2, the coordinate transformation model between the sensor coordinate system and the irregular calibration component coordinate system is established as follows:

[0010] The coordinate transformation model involves three rotations: First rotation: Represents the initial sensor coordinate system Rotate the Z-axis by an angle β; second rotation: This represents a rotation of an angle α around the X-axis of the new coordinate system formed after the first rotation; the third rotation: This indicates a rotation of an angle γ around the Z-axis of the new coordinate system formed after the second rotation; This indicates the transformation of the point to the target coordinates in the coordinate system of the irregular calibration part; This represents the measured coordinates of a point in the sensor coordinate system. This represents the distance between the line structured light sensor and the z-axis of the laser plane; This represents the distance between the line structured light sensor and the laser plane along the x-axis. The translation matrices correspond to respectively Translational pose in the direction.

[0011] In one embodiment, in step S3, the point cloud data extraction adopts a layer-by-layer slicing method, slicing the point cloud at equal intervals along the Z-axis, and identifying features such as outer cylindrical surfaces, planes, and V-grooves based on the shape of the sliced ​​point cloud.

[0012] In one embodiment, the formula for the distance objective function in step S4 is as follows:

[0013] in, k To calibrate the extrinsic parameters, due to the line structured light sensor and the laser plane... y If the axes are collinear, then Unaffected l, m, and n represent the number of measurement points for the outer cylindrical surface, plane, and V-groove, respectively. These are weighting coefficients based on the number of specially designed measurement points. These are distance functions from each measurement point to the outer cylindrical surface, the plane, and the V-groove, respectively. Indicates the first The coordinates of a measurement point belonging to the outer cylindrical surface; Indicates the first The coordinates of a measurement point belonging to the plane; Indicates the first The coordinates of the measurement point belonging to the V-groove.

[0014] In one embodiment, during the iteration process in step S4, the iteration step size is adjusted by weighting coefficients, and the average standard deviation or the maximum number of iterations is used as the convergence condition.

[0015] Secondly, embodiments of the present invention also provide an external parameter calibration system for a single-group line structured light system, comprising: a single-group line structured light system, an irregularly shaped calibration component, and a data processing unit; The single-group line structured light system includes a line structured light sensor, a linear grating, a three-axis rotary table, a linear translation stage, a three-jaw chuck, a circular grating, and a rotary table; the three-axis rotary table is used to clamp and rotate irregularly shaped calibration parts. The irregularly shaped calibration component is composed of three types of composite features: an outer cylindrical surface, a flat surface, and a V-groove. A data processing unit is configured to perform the method as described in any one of the first aspects.

[0016] As can be seen from the above technical solution, compared with the prior art, the present invention has the following technical advantages: 1. The irregularly shaped calibration part provided by the present invention can meet the actual needs of multi-feature calibration. It includes three types of composite features: outer cylindrical surface, plane and V-groove. It has the characteristics of strong transparency and wide applicability.

[0017] 2. This invention proposes a method for obtaining translational and rotational extrinsic parameters based on three types of composite features of irregularly shaped calibration parts, providing effective initial values ​​for the optimization algorithm and accelerating calibration efficiency.

[0018] 3. This invention utilizes three types of composite features of irregularly shaped calibration parts to construct coordinate transformation equations and objective functions, and combines the Levenberg-Marquardt optimization algorithm to iteratively optimize the calibration extrinsic parameters, thereby improving the accuracy compensation of calibration rate.

[0019] 4. The calibration method provided by this invention has a simple process. Compared with the current irregularly shaped calibration parts and calibration methods, it simplifies the complex calibration system and shortens the calibration time. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0021] Figure 1 This is a flowchart of the external parameter calibration method for a single-group line structured light system provided in this embodiment of the invention; Figure 2 This is a diagram illustrating the specific implementation process of external parameter calibration for a single-group line structured light system provided in this embodiment of the invention. Figure 3 This is a schematic diagram of a single-group line structured light measurement system provided in an embodiment of the present invention; Figure 4 This is an overall structural diagram of the irregularly shaped calibration component provided in the embodiments of the present invention; Figure 5This is a schematic diagram of the output of the line structure optical sensor provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the coordinate framework of the external parameter calibration system provided in this embodiment of the invention; Figure 7 This is a schematic diagram showing the positional relationship between the sensor coordinate system and the sensor laser coordinate system provided in this embodiment of the invention; Figure 8 This is a schematic diagram of the overall contour point cloud reconstructed by the extrinsic parameter calibration method for a single-group line structured light system provided in this embodiment of the invention.

[0022] Reference numerals in the attached figures: 1-Linear structured light sensor; 2-Linear grating; 3-Three-axis rotary table; 4-Linear translation stage; 5-Three-jaw chuck; 6-Circular grating; 7-Rotary table; 8-Outer cylindrical surface; 9-Plane; 10-V-groove. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] Reference Figure 1 As shown, this embodiment of the invention discloses a method for calibrating the external parameters of a single-group line structured light system, including the following steps S1~S5: S1: Clamp the irregularly shaped calibration component and adjust the position and orientation of the line structured light sensor so that its laser plane covers the composite features on the surface of the irregularly shaped calibration component; the irregularly shaped calibration component integrates three types of composite features: an outer cylindrical surface, a plane, and a V-groove; specifically, it has one outer cylindrical surface feature, two plane features, and two V-groove features; The outer cylindrical surface has an inner hole, and the axis of the outer cylindrical surface is coaxial with the center line of the inner hole; the two planes are perpendicular to each other and symmetrically distributed around the geometric center of the irregular calibration part; the two V-grooves are respectively set on the two planes and are not completely connected, and are symmetrically distributed with the axes of the upper and lower end faces of the irregular calibration part; the angle of the V-groove is 120°.

[0025] In practical implementation, for example, the clamping of irregularly shaped calibration parts is achieved through the automatic centering characteristic of a three-jaw chuck, ensuring that the inner hole of the calibration part is coaxial with the rotating shaft; and the line structured light sensor is rotated 90° relative to the vertical installation state and fixed, and the measurement output of the line structured light sensor is adjusted to zero.

[0026] S2: Establish a coordinate transformation model between the sensor coordinate system and the irregular calibration part coordinate system, the model including the external parameters to be calibrated; S3: Drive the irregularly shaped calibration part to rotate, collect its surface point cloud data through the line structured light sensor, and extract the geometric features of the three types of composite features; S4: Based on the geometric features, construct a distance objective function, and use the Levenberg-Marquardt nonlinear optimization algorithm to minimize the sum of squared distances from the measurement point to the three types of composite feature surfaces as the optimization objective, and iteratively solve for the extrinsic parameters; S5: Reconstruct the 3D model of the irregular calibration part based on the optimized extrinsic parameters, and verify the calibration accuracy by comparing it with the actual geometric parameters. If the preset threshold is met, the calibration is complete. This step uses the actual geometric parameters of the calibration part as a benchmark, compares the relative deviation of the 3D reconstruction parameters, and compares the pose parameters of the line structured light sensor before and after calibration to verify the rationality of the optimization; generate a visualization of the reconstructed model. If the relative deviation of all composite feature parameters meets the preset acceptance threshold, the calibration is qualified; if it does not meet the threshold, return to step S4 for iterative optimization.

[0027] The calibration process of this invention is simple and the system structure is concise. It is suitable for high-precision and high-efficiency external parameter calibration of multi-feature irregular parts.

[0028] The following is combined with Figure 2 The method of the present invention will be further described in detail below: First, the single-line structured light measurement system and the anisotropic calibration component involved in this invention are described below, with reference to... Figure 3 As shown, a single-group line structured light system includes: a line structured light sensor 1, a linear grating 2, a three-axis rotary table 3, a linear translation stage 4, a three-jaw chuck 5, a circular grating 6, and a rotary table 7.

[0029] like Figure 4 As shown, the irregular calibration part is composed of three types of composite features: an outer cylindrical surface 8, a plane 9, and a V-groove 10. The surface features of the irregular calibration part are those used for calibration: including an outer cylindrical surface feature with an inner hole, the axis of which is coaxial with the center line of the inner hole. It includes two planar features, which are perpendicular to each other and symmetrically distributed around the geometric center of the irregular calibration part; It also includes two V-groove features, which are set on two planes and are not completely connected. They are symmetrically distributed with the axes of the upper and lower end faces of the irregular calibration part; the coaxiality of the inner and outer cylinders is 0.003mm, the included angle of the two planes is 90°, the V-groove angle is 120°, and the diameter of the outer cylinder is 68mm.

[0030] A flowchart illustrating the specific method for calibrating the external parameters of a single-group linear structured light system can be found in [link to flowchart documentation]. Figure 2The specific process includes the following steps: W1: Based on the automatic centering characteristics of the three-jaw chuck 5, the three jaws of the three-jaw chuck 5 are engaged with the inner hole of the irregular calibration part to complete the clamping of the irregular calibration part, and then the position and attitude of the line structured light sensor 1 are adjusted.

[0031] First, the three-axis rotary table 3 is adjusted to move the line structured light sensor 1 to a suitable position, ensuring that its laser plane can completely cover the composite features distributed along the axial direction of the irregular calibration part; then, the line structured light sensor 1 is rotated 90° relative to its conventional vertical installation state and fixed, so that its installation posture parameters are used as initial parameters; finally, a zeroing operation is performed, that is, the measurement output of the line structured light sensor 1 is adjusted to zero to ensure the accuracy of the initial measurement state; W2: Establish the initial calibration system coordinate frame and connect the sensor coordinate system. Convert to coordinate system of irregular calibration part Coordinate transformation; specifically including: W2.1: First, establish the initial sensor coordinate system. The coordinate system is defined in the initial pose of the online structured light sensor 1 when it has not performed any rotational motion, and its origin is... Located at the optical center of the laser plane of the line structured light sensor 1, The initial measurement reference direction of the laser beam of the axis-and-line structured light sensor 1 is consistent, and the sensor coordinate system is established similarly. Simultaneously, establish a coordinate system for the irregularly shaped calibration parts. Its coordinate origin The initial sensor coordinate system is located at the center point of the inner hole on the lower end face of the irregularly shaped calibration component. Sensor coordinate system Coordinate system for irregularly shaped calibration parts All are established according to the right-hand rule; such as Figure 5 The diagram shown is a schematic of the output of a line structured light sensor. The coordinate system in the diagram is the coordinate system of the irregularly shaped calibration component. When the structured light from the sensor line hits the V-groove, it outputs the V-shaped line below.

[0032] W2.2: The core of external parameter calibration is determining the coordinates from the initial sensor coordinate system. coordinate system of irregular calibration parts The coordinate transformation relationship is given by the rotation matrix. Translation matrix Composition, in which Rotation angle Angular poses corresponding to yaw, pitch, and roll, respectively; translation matrix Corresponding to The direction of translation pose; the initial sensor coordinate system is determined by measuring the initial sensor state, which establishes the initial state of a process.

[0033] Specifically, based on the calibration system coordinate framework established in W2.1, the sensor coordinate system is determined. Coordinate system of irregularly shaped calibration parts The coordinate transformation relationship between the two is as follows: (1) in, The point is in the sensor coordinate system The measured coordinates below Its coordinate system in the irregular calibration part The coordinates below; A schematic diagram of the coordinate frame of the external parameter calibration system is shown below. Figure 6 As shown in the figure, the sensor coordinate system is displayed. Coordinate system of irregularly shaped calibration parts , indicating the sensor coordinate system Transform into the coordinate system of irregular calibration parts The illustrated process.

[0034] In this embodiment, Figure 7 The sensor coordinate system in this invention and sensor laser coordinate system The positional relationship, the sensor laser coordinate system is denoted as . It is the origin of the sensor's laser coordinate system, located at the optical center of the laser plane of the line structured light sensor; The initial measurement reference direction of the axis-side structured light sensor is established. The axis is established along the extension direction of the laser stripe. The axes are determined according to the right-hand rule. Sensor coordinate system. and sensor laser coordinate system It can be seen that the coordinates of the two do not coincide, and there is a and The distance. That is, due to the limitations of the measurement system's processing and assembly, the actual coordinate transformation relationship has the following problems: ① The line structured light sensor 1 and the laser plane z The axis has a certain distance ② Line structured light sensor 1 and laser plane x The axis has a certain distance ③ Line structured light sensor 1 and laser plane y Collinear axes, i.e. ; To address the above issues, the coordinate transformation formula is revised as follows: (2) In equation (2), the coordinate transformation model has three rotations: First rotation: Represents the initial sensor coordinate system The Z-axis is rotated by an angle β (yaw angle); the second rotation: This indicates a rotation of an angle α (pitch angle) around the X-axis of the new coordinate system formed after the first rotation; the third rotation: This represents a rotation of an angle γ (roll angle) around the Z-axis of the new coordinate system formed after the second rotation. This is due to the line structured light sensor 1 and the laser plane... y The axes are collinear, therefore the extrinsic translation matrix needs to be solved. Unaffected, the number of solution parameters is simplified from 6 to 5, i.e. .

[0035] W3: Acquisition of point cloud data on the surface of irregularly shaped calibration parts, followed by extraction of three types of composite features for calibration; specifically including: W3.1: Running a single-group linear structured light system, its circular grating 6 rotates circumferentially at a uniform speed around an irregularly shaped calibration component. The rotation signal triggers the probe to acquire data, obtaining the position of the irregularly shaped calibration component in the sensor coordinate system. The point cloud data collected includes point cloud data with three types of composite features; W3.2: Employs a layer-by-layer feature extraction method to extract all point clouds in the coordinate system of the irregular calibration part. z The slices are equidistant on the axis. The circular point cloud on the slices corresponds to the outer cylindrical surface feature; the straight line point cloud on the slices corresponds to the two planar features; the broken straight line point cloud on the slices corresponds to the V-groove feature. The outer cylindrical surface is used to determine the rotation axis, while the two planes and the V-groove determine the position relative to the rotation axis.

[0036] W4: Based on the Levberg-Marquardt nonlinear optimization algorithm, the external parameters of the three types of composite features are optimized. The scaling strategy is determined by the weight coefficients. The iteration is terminated by the average standard deviation and the specified number of iterations to obtain the final calibration system framework. Specifically, five extrinsic parameters are determined by optimizing the sum of squares of the distances from the measurement points to the fitted three types of composite feature surfaces using the least squares method. Therefore, the objective function is the sum of squares of the distances from the measurement points to each calibration feature: (3) in, k To calibrate the external parameters, l, m, and n represent the number of measurement points for the outer cylindrical surface, plane, and V-groove. The weighting coefficients are based on a specially designed number of measurement points. The distance functions from each measurement point to the outer cylindrical surface, the plane, and the V-groove are as follows: Indicates the first The coordinates of a measurement point belonging to the outer cylindrical surface; Indicates the first The coordinates of a measurement point belonging to the plane; Indicates the first The coordinates of the measurement points belonging to the V-groove; (4) (5) (6) In equation (4-6), x , y , z These are the coordinates of the measured point. It is the coordinate of a reference point on the axis of the outer cylindrical surface. It is the direction vector of the axis of the outer cylindrical surface. These are the coefficients of the plane equations of two planes. These are the plane equation coefficients of the V-groove; According to the Levenberg-Marquardt method, Least squares optimization is used as the objective function: (7) in, f Coordinates and parameters k The function, i for N The serial number of each sampling point For the first i The corresponding true value of each point; For residuals; The iterative formula is expressed as: (8) in, These are the calibration extrinsic parameters for the later and earlier iterations during the iteration process. These are the weighting coefficients. I It is the identity matrix. J The Jacobian matrix contains right The partial derivatives, each row includes the partial derivatives of the corresponding feature function with respect to the five calibrated extrinsic parameters at each policy point; the Levenberg-Marquardt method determines the weight coefficients. A scaling measurement strategy is used to ensure continued convergence after approaching the target during the iteration process, and an adjustment coefficient is selected. Weighting coefficient The scaling measurement strategy is expressed as follows: (9) Subscript v and v+ 1 represents the number of iterations; Calculate the mean standard deviation in each iteration. : (10) Among them, when Less than 0.1 Alternatively, the iteration can be terminated when the number of iterations exceeds 10,000, and the final calibration extrinsic parameters can be output. k Obtain the coordinate system of the irregular calibration part coordinates below : (11) in:

[0037]

[0038] in, The coordinates output by the line structured light sensor 1 The rotation angle of the three-jaw chuck.

[0039] W5: Using the actual geometric parameters of the calibration component as a benchmark, compare the relative deviations of the 3D reconstruction parameters, and at the same time compare the pose parameters of the line structured light sensor 1 before and after calibration to verify the rationality of the optimization; generate a visualization of the reconstruction model; if the relative deviations of all composite feature parameters meet the preset acceptance threshold, the calibration is qualified; if it does not meet the standard, return to W4 for iterative optimization.

[0040] Figure 8 To reconstruct three types of composite point clouds using the extrinsic parameter calibration method of a single-group structured light system, three types of composite features are extracted layer by layer. The extrinsic parameters of the three types of composite features are optimized according to the Levenberg-Marquardt nonlinear optimization algorithm. The scaling strategy is determined by the weight coefficients, and the iteration is terminated by the mean standard deviation and a specified number of iterations to obtain the final calibration system framework. The reconstruction model is then visualized, intuitively demonstrating the practicality of the method in the field of extrinsic parameter calibration and reconstruction.

[0041] Based on the same inventive concept, this embodiment of the invention also provides an external parameter calibration system for a single-group line structured light system, including: a single-group line structured light system, an irregularly shaped calibration component, and a data processing unit; The single-group line structured light system includes a line structured light sensor 1, a linear grating 2, a three-axis rotary table 3, a linear translation stage 4, a three-jaw chuck 5, a circular grating 6, and a rotary table 7; the three-axis rotary table 3 is used to clamp and rotate irregularly shaped calibration parts. The irregularly shaped calibration part consists of three composite features: an outer cylindrical surface 8, a flat surface 9, and a V-groove 10. The data processing unit is configured as follows: 1) Store the actual geometric parameters of the irregular calibration element; clamp the irregular calibration element and adjust the position and orientation of the line structured light sensor so that its laser plane covers the composite features of the surface of the irregular calibration element; 2) Establish a coordinate transformation model between the sensor coordinate system and the irregular calibration part coordinate system, the model including the external parameters to be calibrated; 3) Drive the irregularly shaped calibration part to rotate, collect its surface point cloud data through a line structured light sensor, and extract the geometric features of three types of composite features; 3) Construct a distance objective function based on geometric features, and use the Levenberg-Marquardt nonlinear optimization algorithm to minimize the sum of squared distances from the measurement point to the three types of composite feature surfaces, and iteratively solve for the extrinsic parameters. 4) Reconstruct the 3D model of the irregular calibration part based on the optimized extrinsic parameters, and verify the calibration accuracy by comparing it with the actual geometric parameters. If the preset threshold is met, the calibration is completed.

[0042] Among them, the line structured light sensor is posed by a three-axis rotary table 3 and a linear translation table 4, and is fixed so that its laser plane is rotated 90° relative to the conventional vertical installation state, so as to fully cover the composite features of the irregular calibration part distributed along the axis. The irregularly shaped calibration component is clamped by a three-jaw chuck 5 through its inner hole. The three-jaw chuck 5 has an automatic centering characteristic to ensure that the axis of the inner hole of the irregularly shaped calibration component is coaxial with the rotation axis of the rotary table 7. The rotary table 7, driven by the circular grating 6, drives the three-jaw chuck 5 and the irregularly shaped calibration component to rotate circumferentially at a uniform speed. The rotation signal triggers the structured light sensor to collect the contour point cloud data of the irregularly shaped calibration component surface.

[0043] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0044] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for calibrating the extrinsic parameters of a single-group linear structured light system, characterized in that, Includes the following steps: S1: Clamp the irregularly shaped calibration component and adjust the position and orientation of the line structured light sensor so that its laser plane covers the composite features on the surface of the irregularly shaped calibration component; the irregularly shaped calibration component integrates three types of composite features: outer cylindrical surface, plane and V-groove; S2: Establish a coordinate transformation model between the sensor coordinate system and the irregular calibration part coordinate system. This model includes the external parameters to be calibrated; the external parameters include the rotation angle (…). Translation () ); S3: Drive the irregularly shaped calibration part to rotate, collect its surface point cloud data through the line structured light sensor, and extract the geometric features of the three types of composite features; S4: Based on the geometric features, construct a distance objective function, and use the Levenberg-Marquardt nonlinear optimization algorithm to minimize the sum of squared distances from the measurement point to the three types of composite feature surfaces as the optimization objective, and iteratively solve for the extrinsic parameters; S5: Reconstruct the 3D model of the irregular calibration part based on the optimized extrinsic parameters, and verify the calibration accuracy by comparing it with the actual geometric parameters. If the preset threshold is met, the calibration is completed.

2. The method as described in claim 1, characterized in that, The irregularly shaped calibration component includes: an outer cylindrical surface feature, two planar features, and two V-groove features; The outer cylindrical surface has an inner hole, and the axis of the outer cylindrical surface is coaxial with the center line of the inner hole; the two planes are perpendicular to each other and symmetrically distributed around the geometric center of the irregular calibration part; the two V-grooves are respectively set on the two planes and are not completely connected, and are symmetrically distributed with the axes of the upper and lower end faces of the irregular calibration part; the angle of the V-groove is 120°.

3. The method as described in claim 1, characterized in that, In step S1, the clamping of the irregularly shaped calibration component is achieved through the automatic centering characteristic of the three-jaw chuck, ensuring that the inner hole of the calibration component is coaxial with the rotation axis; and the line structured light sensor is rotated 90° relative to the vertical installation state and fixed, and the measurement output of the line structured light sensor is adjusted to zero.

4. The method as described in claim 1, characterized in that, In step S2, the coordinate transformation model between the sensor coordinate system and the irregular calibration component coordinate system is established as follows: The coordinate transformation model involves three rotations: First rotation: Represents the initial sensor coordinate system Rotate the Z-axis by an angle β; second rotation: This represents a rotation of an angle α around the X-axis of the new coordinate system formed after the first rotation; the third rotation: This indicates a rotation of an angle γ around the Z-axis of the new coordinate system formed after the second rotation; This indicates the transformation of the point to the target coordinates in the coordinate system of the irregular calibration part; This represents the measured coordinates of a point in the sensor coordinate system. This represents the distance between the line structured light sensor and the z-axis of the laser plane; This represents the distance between the line structured light sensor and the laser plane along the x-axis. The translation matrices correspond to respectively Translational pose in the direction.

5. The method as described in claim 1, characterized in that, In step S3, the point cloud data extraction adopts a layer-by-layer slicing method, which slices the point cloud at equal intervals along the Z-axis and identifies the features of the outer cylindrical surface, plane and V-groove based on the shape of the sliced ​​point cloud.

6. The method as described in claim 1, characterized in that, In step S4, the formula for the distance objective function is as follows: in, k To calibrate the extrinsic parameters, due to the line structured light sensor and the laser plane... y If the axes are collinear, then Unaffected l, m, and n represent the number of measurement points for the outer cylindrical surface, plane, and V-groove, respectively. These are weighting coefficients based on the number of specially designed measurement points. These are distance functions from each measurement point to the outer cylindrical surface, the plane, and the V-groove, respectively. Indicates the first The coordinates of a measurement point belonging to the outer cylindrical surface; Indicates the first The coordinates of a measurement point belonging to the plane; Indicates the first The coordinates of the measurement point belonging to the V-groove.

7. The method as described in claim 1, characterized in that, During the iteration process in step S4, the iteration step size is adjusted by weighting coefficients, and the average standard deviation or the maximum number of iterations is used as the convergence condition.

8. An external parameter calibration system for a single-group linear structured light system, characterized in that, include: Single-line structured light system, irregularly shaped calibration component, and data processing unit; The single-group line structured light system includes a line structured light sensor (1), a linear grating (2), a three-axis rotary table (3), a linear translation stage (4), a three-jaw chuck (5), a circular grating (6), and a rotary table (7); the three-axis rotary table (3) is used to clamp and rotate irregularly shaped calibration parts; The irregular calibration part is composed of three composite features: an outer cylindrical surface (8), a plane (9), and a V-groove (10); A data processing unit for performing the method as described in any one of claims 1 to 7.

Images