Surface structured light system high-precision calibration method based on double telecentric imaging
By establishing a dual telecentric imaging model and solving camera parameters, the problem of failure of traditional calibration methods in dual telecentric lens applications is solved, and high-precision calibration of surface structured light systems is achieved.
Patent Information
- Application Number
- CN202411939611.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-26
- Publication Date
- 2025-05-09
AI Technical Summary
The traditional calibration method of structured light system fails when applying a dual telecentric lens, resulting in the problem of calibration failure.
By establishing a dual telecentric imaging model, the relationship between pixel coordinates and world coordinates is determined, the conversion relationship between the image coordinate system and the camera coordinate system is established, and the homography matrix, rotation matrix and translation matrix are solved to achieve high-precision calibration.
The ambiguity problem of the rotation matrix and the incomplete problem of the translation matrix are solved, and the high-precision calibration of the dual telecentric camera is realized, which is suitable for surface structured light systems and other systems using dual telecentric lenses.
Smart Images
Figure CN119958462A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of three-dimensional measurement technology, and specifically discloses a high-precision calibration method for a surface structured light system based on dual telecentric imaging. Background Art
[0002] Surface structured light 3D measurement technology has the characteristics of non-contact, high speed and high precision, and is widely used in parts manufacturing, reverse engineering, cultural relics protection, defect detection, robot navigation and other fields. The system consists of an industrial projector and an industrial camera. The basic principle is that the projector projects the grating fringe pattern onto the surface of the object to be measured, and the camera captures the modulated fringe pattern on the surface of the object to be measured, and completes the acquisition of 3D data through phase unwrapping and model calculation.
[0003] In order to achieve high-precision 3D detection capabilities, bi-telecentric lenses are gradually applied to surface structured light 3D measurement systems. Due to the special imaging method of bi-telecentric lenses, only parallel light can pass through both the object side and the imaging side, and there is no "near big and far small" imaging relationship.
[0004] Therefore, when the structured light system uses a bi-telecentric lens, the traditional structured light system calibration method fails.
[0005] In view of this, the inventor provides a high-precision calibration method for a surface structured light system based on dual telecentric imaging to solve the above problems. Summary of the invention
[0006] The purpose of the present invention is to solve the problem of calibration failure in a traditional structured light system calibration method when a bi-telecentric lens is used.
[0007] In order to achieve the above object, the basic scheme of the present invention provides a high-precision calibration method of a surface structured light system based on double telecentric imaging, comprising the following steps:
[0008] Step S1: Establish a dual telecentric imaging model, determine the relationship between pixel coordinates and world coordinates in the dual telecentric imaging model, establish the conversion relationship between the image coordinate system and the camera coordinate system, and determine the homography matrix: Rotation matrix: And the translation matrix:
[0009] Step S2: Solve the parameters of the dual telecentric imaging model, including the complete solution of the homography matrix, the rotation matrix r 11 , r 12 , r 21 and r 22 The solution of t x and t y The solution of
[0010] Step S3: Establish a double constraint to adjust the rotation matrix r 31 and r 32 Solve accurately and determine r 33 The value of is used to determine the rotation matrix for bi-telecentric camera calibration;
[0011] Step S4: Determine the X coordinate system of the camera c O c Y c The plane passes through the current calibration plate world coordinate system O w -X w Y w Z w O w Point, so that the camera coordinate system Z c = 0, the calibration plate is moved by the motion platform, and the linear change of the calibration plate, t x The change in t y The change in t in the translation matrix is determined z To achieve the restoration of the translation matrix of the bi-telecentric camera calibration;
[0012] Step S5: According to the obtained rotation matrix and translation matrix, the three-dimensional coordinates of the characteristic points of the calibration plate in the camera coordinate system are established, and a surface structured light system model of double telecentric imaging is constructed.
[0013] Furthermore, in step S1, the conversion relationship between the established image coordinate system and the camera coordinate system is as follows:
[0014]
[0015] Where (x, y) is the coordinate of feature point P in the image coordinate system XOY, k is the magnification of the bi-telecentric lens, x is c and c is the coordinate of the feature point P in the camera coordinate system.
[0016] Further, in step S1, the relationship between the pixel coordinates and the world coordinates in the determined dual telecentric imaging model is as follows:
[0017]
[0018]
[0019] Where s is a non-zero scaling factor, α x = k / d x , α y = k / d y, M1 is the camera's intrinsic parameter matrix, which is only related to the camera hardware, M2 is the external space transformation matrix, H is the homography matrix that describes the mapping relationship between pixel coordinates and world coordinates, m and n represent row and column pixel coordinates, and m0 and n0 represent the row and column positions of the origin of the image coordinate system in the pixel coordinate system.
[0020] Further, in step S2, the rotation matrix R is solved as follows:
[0021]
[0022] For the translation matrix T, solve it as follows:
[0023] t x =(h 13 -m0) / α x
[0024] t y =(h 23 -n0) / α y .
[0025] Further, in step S3, the constraint condition 1 is specifically: according to the property that each row vector of the rotation matrix is also a unit orthogonal vector, the following formula is obtained:
[0026] r 31 × 32 =0-r 11 r 12 -r 21 r 22
[0027] Combined with the already solved r 11 , r 12 , r 21 and r 22 To determine r 31 and r 32 Are of the same or different sign.
[0028] Further, in step S3, the second constraint condition is specifically: when the feature point P te1 Compared with feature point P te2 Closer to the camera coordinate system X c O c Y c When the plane, that is, the lens plane, is determined:
[0029]
[0030] Further, in step S4, determining the translation matrix of the dual telecentric camera calibration specifically includes the following steps:
[0031] Step S41: Place the calibration plate on the stage as required, and define a plane perpendicular to the camera optical axis and passing through the feature points on the calibration plate as the X coordinate system of the camera coordinate system. c O c Y c Plane, for the calibration plate world coordinate system at the current reference position, in the transformation matrix from its world coordinate system to the camera coordinate system, the element t of the translation matrix z =0;
[0032] Step S42: Use a high-precision motion platform to drive the calibration plate to move a straight-line distance d, and use the change in the motion platform, t x The change in t y The change in t in the translation matrix is determined z , according to the world coordinates of the feature points on the calibration plate and t in the translation matrix x and t y , for the calibration plate at position j, the translation matrix change can be expressed as follows:
[0033]
[0034] In the formula, The corner mark 1 indicates the reference position of the calibration plate, and the corner mark j indicates the jth position of the calibration plate. If the movement direction of the platform is consistent with the optical axis, then t z Direction is positive otherwise negative.
[0035] The principle and effect of this scheme are:
[0036] 1. Compared with the prior art, the present invention solves the ambiguity problem of the rotation matrix by adding additional constraints of the calibration plate and analyzing the characteristics of the rotation matrix.
[0037] 2. Compared with the prior art, the present invention breaks through the problem of incomplete translation matrix in camera calibration by innovatively setting the reference plane and utilizing the motion of the precision motion platform.
[0038] 3. Compared with the prior art, the present invention establishes an accurate imaging model of a dual telecentric lens and forms a set of high-precision calibration solutions for dual telecentric cameras, which solves the problem of calibration failure in traditional structured light system calibration methods when dual telecentric lenses are used.
[0039] 4. Compared with the prior art, the present invention proposes a precise calibration scheme for system model parameters when the surface structured light system uses a double telecentric lens.
[0040] 5. Compared with the prior art, the present invention achieves high-precision calibration of the surface structured light system by relying only on a precision motion platform and a conventional calibration plate. This method is also applicable to other systems using a dual telecentric lens. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.
[0042] Figure 1 A flowchart of a high-precision calibration method for a surface structured light system based on double telecentric imaging proposed in an embodiment of the present application is shown;
[0043] Figure 2 A schematic diagram of a dual telecentric imaging model proposed in an embodiment of the present application is shown;
[0044] Figure 3 A schematic diagram showing the position and posture of the calibration plate proposed in an embodiment of the present application is shown;
[0045] Figure 4 A schematic diagram of a camera coordinate system setting reference proposed in an embodiment of the present application is shown;
[0046] Figure 5 A schematic diagram showing the position change of the calibration plate after the motion platform proposed in the embodiment of the present application moves;
[0047] Figure 6 The calibration plate image proposed in the embodiment of the present application for dual telecentric camera calibration and surface structured light system calibration shooting is shown;
[0048] Figure 7 A schematic diagram showing the reprojection error results of each image proposed in the embodiment of the present application is shown;
[0049] Figure 8 A schematic diagram of the measurement results of the calibration plate by the calibrated surface structured light system proposed in an embodiment of the present application is shown, wherein (a) is a schematic diagram of the calibration plate, and (b) is a schematic diagram of the distance error fluctuation result. DETAILED DESCRIPTION
[0050] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation mode, structure, characteristics and effects of the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.
[0051] A high-precision calibration method for a surface structured light system based on dual telecentric imaging is implemented, for example Figure 1 As shown, the following steps are included:
[0052] Step S1: Establish a dual telecentric imaging model, determine the relationship between pixel coordinates and world coordinates in the dual telecentric imaging model, establish a transformation relationship between the image coordinate system and the camera coordinate system, and determine the homography matrix, rotation matrix and translation matrix.
[0053] Specifically, the established dual telecentric imaging model is as follows: Figure 2 As shown, the relationship between the pixel coordinate system and the image coordinate system is consistent with the pinhole imaging model. And because the bi-telecentric lens can only receive parallel light, there is no relationship between near and far, so the conversion relationship between the image coordinate system and the camera coordinate system is shown in the following formula:
[0054]
[0055] Where (x, y) is the coordinate of feature point P in the image coordinate system XOY, k is the magnification of the bi-telecentric lens, x is c and c is the coordinate of the feature point P in the camera coordinate system. And because the light path is always parallel, the focus of the image is at infinity.
[0056] Depend on Figure 2 It can be obtained that the relationship between pixel coordinates and world coordinates in the dual telecentric imaging model is as follows:
[0057]
[0058] Where s is a non-zero scaling factor, α x = k / d x , α y = k / d y , M1 is the camera's intrinsic parameter matrix, which is only related to the camera hardware, M2 is the external space transformation matrix, H is the homography matrix that describes the mapping relationship between pixel coordinates and world coordinates, m and n represent row and column pixel coordinates, m0 and n0 represent the row and column positions of the origin of the image coordinate system in the pixel coordinate system;
[0059] Specifically, the rotation matrix R, translation matrix T and homography matrix H are shown as follows:
[0060]
[0061] Step S2: Solve the parameters of the dual telecentric imaging model, including the complete solution of the homography matrix, the rotation matrix r 11 , r 12 , r 21 and r 22 The solution of t x and t y The solution.
[0062] Specifically, for the homography matrix H and the camera intrinsic parameters (i.e. αx and α y ) is similar to the traditional camera imaging model and will not be described in detail in this embodiment.
[0063] For the rotation matrix R, solve it as follows:
[0064]
[0065] For the translation matrix T, solve it as follows:
[0066] t x =(h 13 -m0) / α x
[0067] t y =(h 23 -n0) / α y
[0068] Therefore, the external parameter matrix obtained by camera calibration can only determine some elements, and the remaining elements cannot be solved due to insufficient constraints.
[0069] Therefore, the rotation matrix first needs to determine r 31 With r 32 The symbol of the rotation matrix is then used to solve the remaining element r of the rotation matrix using the property that the columns of the orthogonal matrix are unit vectors and the columns are orthogonal. 13 、r 23 and r 33 :
[0070]
[0071] It can be seen that r 31 and r 32 There are four combinations in total, and only one of them matches the actual posture of the calibration plate coordinate system in the camera coordinate system.
[0072] Step S3: Establish a double constraint to adjust the rotation matrix r 31 and r 32 Solve accurately and determine r 33 The value of is used to determine the rotation matrix for bi-telecentric camera calibration.
[0073] Specifically, combined with the conclusion obtained in step S2, and based on the property that each row vector of the rotation matrix is also a unit orthogonal vector, the following formula is obtained:
[0074] r 31 × 32 =0-r 11 r 12 -r 21 r 22
[0075] Combined with the already solved r 11 , r 12 , r 21 and r 22 It can be determined that 31 and r 32 They are of the same or different signs, this is constraint one.
[0076] The present invention provides a new additional constraint by adjusting the placement posture of the calibration plate during calibration. The additional constraint is used during system calibration to ensure that any two points Z on the calibration plate are c Coordinate relationship, the specific method is Figure 3 Take the calibration plate pose shown as an example:
[0077] Select two feature points P te1 Follow P te2 , when the feature point P te1 Compared with feature point P te2 Closer to the camera coordinate system X c O c Y c When the plane, that is, the lens plane, is determined:
[0078]
[0079] In the formula, because the Z of the camera coordinate system c The axis is perpendicular to the lens plane and the direction away from the lens is positive, so when point P on the calibration plate te1 Point P te2 As you get closer to the lens plane, This is constraint condition 2.
[0080] Taking feature point P 99 Taking the feature point P1 as an example, because point P 99 Closer to the lens plane (i.e. the traditional camera coordinate system X) than point P1 c O c Y c plane), then
[0081]
[0082] In the formula, because the Z of the camera coordinate system c The axis is perpendicular to the lens plane and the direction away from the lens is positive, so when point P on the calibration plate 99 When it is closer to the lens plane than point P1,
[0083] Constraints 1 and 2 can solve two unknowns. So far, r in the rotation matrix 31 and r 32 can be solved accurately, and then r can be determined13 , r 23 and r 33 The value of , get the complete rotation matrix.
[0084] It should be noted that when calibrating a dual telecentric camera, r 31 , r 32 , r 13 , r 23 , r 33 It does not participate in the calibration calculation, so the calibration plate can be placed arbitrarily. Additional constraints are required if and only if the calibration system measures model parameters.
[0085] Step S4: Determine the X coordinate system of the camera c O c Y c The plane passes through the current calibration plate world coordinate system O w -X w Y w Z w O w Point, so that the camera coordinate system Z c = 0, the calibration plate is moved by the motion platform, and the change of the motion platform, t x The change in t y The change in t in the translation matrix is determined z In order to restore the translation matrix of the bi-telecentric camera calibration.
[0086] Because telecentric lenses do not have the characteristic of "larger near and smaller far", the size of the components is the same at any position in the field of view, so for the camera coordinate system, its Z c = 0 can be at any position on the optical axis, which will make it impossible to obtain the three-dimensional coordinates of the feature points in the camera coordinate system.
[0087] In order to obtain a complete translation matrix from the world coordinate system of the calibration plate to the camera coordinate pose, the present invention accurately solves the translation matrix by optimizing the setting of the benchmark and the high-precision motion platform, including the following steps:
[0088] Step S41: Place the calibration plate on the stage as required, and define the plane perpendicular to the camera optical axis and passing through the feature point P1 on the calibration plate as the X coordinate system of the camera. c O c Y c The plane, that is, the height of the feature point P1 of the calibration plate is taken as the Z c The zero point of the axis, such as Figure 4 According to the above definition of the camera coordinate system, for the calibration plate world coordinate system at the current reference position, in the transformation matrix from the world coordinate system to the camera coordinate system, the element t of the translation matrix z =0.
[0089] Step S42: Use a high-precision motion platform to drive the calibration plate to move a straight-line distance d, and use the linear change of the calibration plate, t x The change in t y The change in t in the translation matrix is determined z .
[0090] Because the platform movement direction is consistent with the Z direction of the camera coordinate system c The axis directions cannot be absolutely parallel, and the lateral coordinates of the feature points on the calibration plate will be offset to a certain extent. The position relationship of the feature points before and after the calibration plate is moved is as follows: Figure 5 shown.
[0091] According to the world coordinates of the feature points on the calibration plate known in step S2, t in the translation matrix x and t y , the platform moves along a straight line to ensure that the rotation matrix remains unchanged, then for the calibration plate at position j, the change in its translation matrix can be expressed as follows:
[0092]
[0093] In the formula, because O c -X c Y c Z c Definition of reference position, The corner mark 1 indicates the reference position of the calibration plate, and the corner mark j indicates the jth position of the calibration plate. If the movement direction of the platform is consistent with the optical axis, then t z Direction is positive otherwise negative.
[0094] Step S5: According to the obtained rotation matrix and translation matrix, the three-dimensional coordinates of the characteristic points of the calibration plate in the camera coordinate system are established, and a surface structured light system model of double telecentric imaging is constructed.
[0095] Specifically, relative to the world coordinate system of the calibration plate at the reference position, the feature points on the calibration plate can be expressed in the camera coordinate system as follows:
[0096]
[0097] Correspondingly, the camera intrinsic parameter matrix can be used to calculate (X c , Y c , Z c ), that is, after determining the reference position, for other positions, the motion platform can be used to drive the calibration plate to translate a distance d, and then directly solve Z c Specifically, the lateral coordinates of the feature points in the camera coordinate system are calculated using the following formula:
[0098]
[0099] In the formula, and is the pixel coordinate of the i-th feature point on the j-th calibration plate position.
[0100] Therefore, for the new calibration plate position, the three-dimensional coordinates of the feature points on the calibration plate in the camera coordinate system can be solved as follows:
[0101]
[0102] In the formula, the reference position When the motion platform moves towards the camera, The sign is negative. When the motion platform moves away from the lens, The sign of is positive.
[0103] When the eight-parameter model proposed by Da Feipeng et al. is adopted and a traditional pinhole imaging lens is used, the surface structured light three-dimensional imaging model is as follows:
[0104]
[0105] In the formula, μ m and μ n represents the row and column pixel size, f c represents the focal length, m and n represent the row and column pixel coordinates, m0 and n0 represent the row and column positions of the origin of the image coordinate system in the pixel coordinate system, and a i (1≤i≤8) represents the system model parameters, and θ represents the absolute phase corresponding to the surface structured light.
[0106] However, when a double telecentric lens is used, the eight-parameter model fails, so it needs to be optimized to adapt to the surface structured light system under the double telecentric imaging model. First, based on the method described above, the three-dimensional coordinates of the feature points of the calibration plate in the camera coordinate system are obtained, and the parameters of the surface structured light system in the eight-parameter model are solved by combining the absolute phase information of the grating stripes corresponding to each feature point. Then, the pixel coordinates and absolute phase are brought into the model to realize the measurement of the surface morphology of the parts.
[0107] Therefore, when a dual telecentric lens is used, for the surface structured light 3D measurement system, the expression of the 3D point is as follows:
[0108]
[0109] In the formula, m x = k / μ m , m y = k / μ n .
[0110] The calibration process is realized by using a precision motion platform, and the pictures taken are as follows: Figure 6As shown in the figure, the first 9 images are used to calibrate the surface structured light system, and all images are used to calibrate the dual telecentric camera. Based on the above calibration images and platform motion, the calibration parameters of the surface structured light system based on dual telecentric imaging are m x =m y =40.018, m0=1223.945, n0=1024.001, a1=3.244×10-02, a2=-2.47×10-04, a3=2.448×10-02 , a4=9.991×10-01, a5=7.214×10-07, a6=2.342×10-06, a7=5.613×10-05, a8=1.306×10-02.
[0111] After camera calibration based on dual telecentric imaging, the reprojection error of each image is as follows: Figure 7 As shown in the figure, in order to test the accuracy of the coded grating surface topography measurement system built in this paper, the calibration plate is measured continuously for multiple times. Among them, the calibration plate is located in the system calibration space, and any corner of the calibration plate is raised and placed randomly.
[0112] like Figure 8 As shown in (a), the calibration plate is made of ceramic material, and the whole is a black background with white circle array distribution. The center distance is 2mm, the processing accuracy is ±1μm, and the definitions of the long side, short side, and diagonal are shown in the figure. The phase at the center position is solved by using the center recognition algorithm and the grating fringe encoding and decoding method, and the spatial position coordinates of the center on the calibration plate are calculated according to the system calibration parameters. After measuring the calibration plate 20 times using the surface structured light system calibrated by the table of the present invention, the distance error fluctuations of the short side, long side, and diagonal corresponding to the point cloud are as follows: Figure 8 (b) as shown.
[0113] analyze Figure 8 (b) It can be seen that in the system calibration space, after 20 tests, the average errors of the 16mm short side, 20mm long side and 25.612mm diagonal on the calibration plate are ±2.1μm, ±2.5μm and ±2.1μm respectively, and the maximum errors are 6.5μm, 3.8μm and 3.5μm respectively. Therefore, for the center coordinates of the calibration plate, the distance error after the system measurement is kept within ±6.5μm.
[0114] The present invention solves the ambiguity problem of the rotation matrix by adding additional constraints of the calibration plate and analyzing the characteristics of the rotation matrix.
[0115] At the same time, the present invention breaks through the incompleteness of the translation matrix in camera calibration by innovatively setting the reference plane and using the motion of the precision motion platform. It also forms a set of high-precision calibration solutions for dual telecentric cameras by establishing an accurate imaging model of dual telecentric lenses, solving the problem of calibration failure in traditional structured light system calibration methods when dual telecentric lenses are used.
[0116] Moreover, the present invention only relies on a precision motion platform and a conventional calibration plate to achieve high-precision calibration of the surface structured light system. Of course, the high-precision calibration method of the surface structured light system based on double telecentric imaging provided by the present invention is also applicable to other systems using double telecentric lenses.
[0117] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.
Claims
1. A high-precision calibration method for a surface structured light system based on dual telecentric imaging, characterized in that: The steps include: Step S1: Establish a dual telecentric imaging model, determine the relationship between pixel coordinates and world coordinates in the dual telecentric imaging model, establish the conversion relationship between the image coordinate system and the camera coordinate system, and determine the homography matrix: Rotation matrix: And the translation matrix: Step S2: Solve the parameters of the dual telecentric imaging model, including the complete solution of the homography matrix, the r in the rotation matrix 11 , r 12 , r 21 and r 22 The solution of t x and t y The solution of Step S3: By establishing a double constraint on the rotation matrix r 31 and r 32 Solve accurately and determine r 33 The value of is used to determine the rotation matrix for bi-telecentric camera calibration; Step S4: Determine the X coordinate system of the camera c O c Y c The plane passes through the current calibration plate world coordinate system O w -X w Y w Z w O w Point, so that the camera coordinate system Z c = 0, the calibration plate is moved by the motion platform, and the linear change of the calibration plate, t x The change in t y The change in t in the translation matrix is determined z To determine and recover the translation matrix of the bi-telecentric camera calibration; Step S5: According to the obtained rotation matrix and translation matrix, the three-dimensional coordinates of the characteristic points of the calibration plate in the camera coordinate system are established, and a surface structured light system model of double telecentric imaging is constructed.
2. According to the high-precision calibration method of the surface structured light system based on double telecentric imaging according to claim 1, it is characterized in that: In step S1, the conversion relationship between the established image coordinate system and the camera coordinate system is as follows: Where (x, y) is the coordinate of feature point P in the image coordinate system XOY, k is the magnification of the bi-telecentric lens, x c and c is the coordinate of the feature point P in the camera coordinate system.
3. The high-precision calibration method of a surface structured light system based on dual telecentric imaging according to claim 2, characterized in that: In step S1, the relationship between the pixel coordinates and the world coordinates in the determined dual telecentric imaging model is shown in the following formula: Where s is a non-zero scaling factor, α x = k / d x , α y = k / d y , M1 is the camera's intrinsic parameter matrix, which is only related to the camera hardware, M2 is the external space transformation matrix, H is the homography matrix that describes the mapping relationship between pixel coordinates and world coordinates, m and n represent row and column pixel coordinates, and m0 and n0 represent the row and column positions of the origin of the image coordinate system in the pixel coordinate system.
4. The high-precision calibration method for a surface structured light system based on dual telecentric imaging according to claim 3, characterized in that: In step S2, the rotation matrix R is solved as follows: For the translation matrix T, solve it as follows: t x =(h 13 -m0) / α x t y =(h 23 -n0) / a y 。 5. The high-precision calibration method of a surface structured light system based on dual telecentric imaging according to claim 4, characterized in that: In step S3, the first constraint condition is specifically: according to the property that each row vector of the rotation matrix is also a unit orthogonal vector, the following formula is obtained: r 31 ×r 32 ＝0-r 11 r 12 -r 21 r 22 Combined with the already solved r 11 , r 12 , r 21 and r 22 To determine r 31 and r 32 Are of the same or different sign.
6. The high-precision calibration method for a surface structured light system based on dual telecentric imaging according to claim 5, characterized in that: In step S3, the second constraint condition is specifically: when the feature point P te1 Compared with feature point P te2 Closer to the camera coordinate system X c O c Y c When the plane, that is, the lens plane, is determined:
7. The high-precision calibration method for a surface structured light system based on dual telecentric imaging according to claim 6, characterized in that: In step S4, determining the translation matrix of the dual telecentric camera calibration specifically includes the following steps: Step S41: Place the calibration plate on the stage as required, and define a plane perpendicular to the camera optical axis and passing through the feature points on the calibration plate as the X coordinate system of the camera coordinate system. c O c Y c Plane, for the calibration plate world coordinate system at the current reference position, in the transformation matrix from its world coordinate system to the camera coordinate system, the element t of the translation matrix z =0; Step S42: Use a high-precision motion platform to drive the calibration plate to move a straight-line distance d, and use the change in the motion platform, t x The change in t y The change in t in the translation matrix is determined z , according to the world coordinates of the feature points on the calibration plate and t in the translation matrix x and t y , for the calibration plate at position j, the translation matrix change can be expressed as follows: In the formula, The corner mark 1 indicates the reference position of the calibration plate, and the corner mark j indicates the jth position of the calibration plate. If the movement direction of the platform is consistent with the optical axis, then t z Direction is positive otherwise negative.