A line laser curved surface calibration method and system for underwater three-dimensional visual measurement

Abstract

This invention belongs to the field of underwater 3D measurement and specifically discloses a method and system for calibrating line laser surfaces for underwater 3D visual measurement. The method includes: acquiring images of a checkerboard calibration board in multiple poses underwater using a camera (image A); determining the sealed viewport parameters on the camera's imaging path based on image A; adjusting the angle of the laser beam emitted by the laser source underwater using a galvanometer, and acquiring images of the checkerboard calibration board with line laser stripes in multiple poses using a camera (image B); determining the intersection points (surface points) between the line laser surface and the checkerboard calibration board based on the sealed viewport parameters and image B; constructing a fourth-order surface fitting error function for the line laser surface, and obtaining the coefficients in the fourth-order surface fitting error function through fitting the surface points, thus completing the underwater line laser surface calibration. This invention achieves accurate fitting of underwater line laser surfaces and can be used for high-precision underwater 3D imaging with line laser scanners.

Description

Technical Field

[0001] This invention belongs to the field of underwater three-dimensional measurement, and more specifically, relates to a line laser surface calibration method and system for underwater three-dimensional visual measurement. Background Technology

[0002] Underwater 3D measurement has wide applications in marine resource development, underwater archaeology, marine biology, and underwater robotics. Among these, underwater line laser 3D measurement is an active measurement method. By selecting a specific wavelength of line laser, the absorption and scattering effects of water on the laser can be reduced, improving image quality. Using a rotating galvanometer to project the line laser onto different locations within a scene provides a new approach to improving measurement efficiency. This method can reconstruct 3D point cloud information of a small scene without requiring additional sensor movement and registration. During underwater line laser 3D measurement, the camera, laser, and galvanometer must be integrated and sealed. The camera rays and line laser plane on the camera and laser sides will refract at the transparent viewport, causing traditional line laser triangulation reconstruction methods to fail in underwater 3D reconstruction. To achieve high-precision underwater line laser 3D measurement, the key lies in the precise calibration of the line laser plane parameters and the sealed viewport parameters, achieving refraction compensation for the camera optical path and the line laser plane optical path, and finally completing the 3D reconstruction based on the principles of underwater triangulation.

[0003] Therefore, many scholars have conducted in-depth theoretical and methodological research on high-precision underwater line laser 3D reconstruction. Based on the principles of 3D reconstruction, it can be divided into line-to-surface 3D reconstruction and line-to-line 3D reconstruction. The line-to-surface 3D reconstruction method approximates the line laser plane as a plane composed of multiple parallel laser beams. After refraction, it remains a plane, failing to consider the bending process that occurs after one refraction of a point-emission line laser plane. This method cannot completely solve the problem of underwater line laser 3D measurement accuracy being limited by the accurate modeling of the line laser surface. The line-to-line 3D reconstruction method discretizes the line laser plane into individual laser beams and performs 3D reconstruction by finding the intersection of the laser beams with the camera beams. However, this method requires calibration of the laser focal point, camera side viewport, galvanometer side viewport, and galvanometer mirror pose. Furthermore, these methods are not only cumbersome and time-consuming, but the error propagation effect during the multi-parameter calibration process can also significantly affect the final 3D reconstruction accuracy. Accordingly, there is a technical need in this field to develop a geometric model of a point-emission line laser plane that becomes a curved surface after refraction, taking into account the change in the incident angle of the line laser beam, and to propose a corresponding parametric fitting method. Summary of the Invention

[0004] In view of the above-mentioned defects or improvement needs of the existing technology, the present invention provides a linear laser surface calibration method and system for underwater three-dimensional visual measurement. Its purpose is to achieve parametric fitting of the underwater linear laser surface of the scanner, thereby achieving accurate calibration of the underwater linear laser surface.

[0005] To achieve the above objectives, according to one aspect of the present invention, a line laser surface calibration method for underwater three-dimensional visual measurement is proposed. The scanner is sealed in a housing and includes a camera, a galvanometer, and a laser. The calibration process includes the following steps:

[0006] S1. Underwater, the camera acquires images of the checkerboard calibration plate in multiple poses, i.e., image A. Based on the checkerboard corner positions on image A, the sealed viewport parameters on the camera imaging path are determined. The sealed viewport parameters include the distance d0 from the camera optical center to the viewport plane and the normal A of the viewport plane.

[0007] S2. Underwater, adjust the angle of the laser beam emitted by the laser using a galvanometer, and acquire images of the checkerboard calibration plate with line laser stripes in multiple poses using a camera, i.e., image B; determine the intersection point of the line laser surface and the checkerboard calibration plate, i.e., the surface point, based on the sealed viewport parameters and image B.

[0008] S3. Construct a quartic surface fitting error function for the line laser surface, and obtain the coefficients in the quartic surface fitting error function by fitting the surface points to complete the underwater line laser surface calibration.

[0009] As a further preferred embodiment, step S2, which determines the surface points based on the sealed viewport parameters and image B, includes the following steps:

[0010] S21. Extract the coordinates of each checkerboard corner point from image B, and then calculate the pose transformation matrix from the underwater checkerboard calibration plate coordinate system {B} to the scanner coordinate system {S}.

[0011] S22. Extract the center position p of the laser stripes on each checkerboard calibration board from image B. L And convert it into a camera ray in the air according to the camera intrinsic parameter matrix K.

[0012] S23, Based on sealed viewport parameters and camera rays in the air Determine camera rays in water

[0013] S24. Based on the camera rays in the water and pose transformation matrix The points on the surface are obtained by solving for the intersection points of the spatial ray and the plane.

[0014] As a further preferred embodiment, step S23 is based on the sealed viewport parameters and the camera rays in the air. Determine camera rays in water

[0015]

[0016] Where λ represents the camera ray in the water. The point on and The distance between them Indicates camera rays in the air Intersection with the viewport plane, The direction vector corresponding to the camera ray in the water is specifically:

[0017]

[0018]

[0019] Where μ0 represents the refractive index of air and μ1 represents the refractive index of water.

[0020] As a further preferred embodiment, in step S1, determining the sealed viewport parameters on the camera imaging path based on the checkerboard corner points on image A includes the following steps:

[0021] S11. Extract the coordinates p of each chessboard corner point from image A, and convert them into camera ray v0 in the air according to the camera intrinsic parameter matrix K:

[0022] v0 = K -1 [p 1] T

[0023] S12. Based on the fact that the camera ray v0 in the air, the normal A of the viewport plane, and the line P connecting the camera optical center and the corner points of the checkerboard are coplanar along the camera imaging path, construct the constraint equations:

[0024] v0 T (A×(RP+t))=0

[0025] Where R and t represent the rotation matrix and translation vector in the pose transformation matrix from the checkerboard coordinate system {B} to the scanner coordinate system {S}, respectively;

[0026] S13. Based on the collinearity constraint of refraction in the camera imaging path, construct the constraint equations:

[0027] (RP+t-q1)×v1=0

[0028]

[0029] Where v1 is the camera ray in the water, and q1 is the refraction point at the viewport plane;

[0030] S14. Solve the constraint equations in S12 and S13 simultaneously to obtain the sealing viewport parameters d0 and A.

[0031] As a further preferred embodiment, step S1 also includes the following steps:

[0032] S15. Using the sealed viewport parameters calculated in S14 as initial values, and with the goal of minimizing the reprojection error of the checkerboard corner points, the final values ​​of the sealed viewport parameters are obtained through optimization.

[0033] As a further preferred embodiment, the method for determining the camera intrinsic parameter matrix K is as follows:

[0034] Images of the checkerboard calibration board in multiple poses are pre-captured in the air using a camera, i.e., image C; the camera intrinsic parameter matrix K is calculated based on image C and the camera calibration algorithm.

[0035] As a further preferred embodiment, the linear laser surface refers to a surface formed by using the intersection of the laser plane in the air and the viewport plane as the guideline of the linear laser surface in the water, and the refracted laser ray as the generatrix of the linear laser surface in the water. The ruled surface established by the guideline and the generatrix is ​​the linear laser surface; this linear laser surface is a fourth-order surface.

[0036] As a further preferred embodiment, in step S3, the constructed quartic surface fitting error function J(c) for the line laser surface is... 400 ,…,c pqr ,…,c 000 )as follows:

[0037]

[0038] Among them, P i =[x i y i z i ] T Let be the coordinates of point i on the linear laser surface, and n be the total number of points on the surface to be fitted; c pqr x is the error function for fitting a quartic surface. p y q z r The coefficient of the term.

[0039] According to another aspect of the present invention, a line laser surface calibration system for underwater three-dimensional visual measurement is provided, comprising a processor for executing the above-described line laser surface calibration method for underwater three-dimensional visual measurement.

[0040] In summary, compared with the prior art, the above-described technical solutions conceived by this invention mainly possess the following technical advantages:

[0041] 1. This invention fully considers the bending of the point-emitting line laser plane after one refraction, and based on the fact that the underwater line laser surface is a fourth-order ruled surface, it achieves parametric fitting of the underwater line laser surface of the scanner through the surface points; it only requires placing the line laser scanner and the checkerboard calibration plate in the water to collect data, and the line laser surface in the water at each galvanometer corner can be calibrated without the need to make additional calibration tools, thus breaking through the bottleneck of existing methods. It has the advantages of low cost, simple operation, high fitting accuracy, and wide applicability.

[0042] 2. This invention directly calibrates the underwater laser surface corresponding to each galvanometer rotation angle, without considering the assembly error between the line laser and the galvanometer, nor the viewport on the galvanometer side of the scanner. Compared with the existing technology that requires calibration of the laser focus point, camera side viewport, galvanometer side viewport, and galvanometer mirror pose, this invention is more operable, has a simpler calibration process, and avoids the propagation effect of solution errors when calibrating multiple parameters simultaneously, resulting in higher calibration accuracy and better 3D reconstruction quality.

[0043] 3. This invention presents a quartic surface fitting method based on surface points, and constructs a quartic surface fitting error function. Based on this error function, the underwater laser surface parameters can be solved, and the underwater line laser surface of the scanner can be accurately calibrated. It can be used for high-precision underwater 3D imaging of line laser scanners. Attached Figure Description

[0044] Figure 1 This is a flowchart illustrating the line laser surface calibration method for underwater three-dimensional visual measurement provided in an embodiment of the present invention.

[0045] Figure 2 This is a schematic diagram of the component setup in the line laser surface calibration method provided in this embodiment of the invention;

[0046] Figure 3 This is a schematic diagram of the structure of a scanner with a galvanometer provided in an embodiment of the present invention;

[0047] Figure 4 A schematic diagram of an underwater laser surface based on laser beams, provided for an embodiment of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0049] This invention provides a method for calibrating line laser surfaces for underwater 3D visual measurement, such as... Figure 2 As shown, the components used in the calibration process include a scanner, a checkerboard calibration plate, and a test water tank. The checkerboard calibration plate has a series of alternating black and white squares printed on it, forming a checkerboard pattern, and is preferably made of ceramic. Figure 3 As shown, the scanner includes a camera, a galvanometer, and a laser. The camera can acquire calibration images, the laser projects a line laser beam onto the target to provide strong features, and the galvanometer projects the line laser beam onto different positions on the target by rotating and reflecting it. The scanner is sealed in a housing, which mainly serves to seal and waterproof it. The housing is transparent to the positions of the camera and laser, i.e., the viewport plane.

[0050] like Figure 4 As shown, the line laser plane is composed of beams of laser light emitted from the focal point of the laser. Each beam is refracted at the viewport plane. Since each beam has a different incident angle at the viewport plane, it appears as a curved line laser surface in water.

[0051] like Figure 1 As shown, the linear laser surface calibration method includes the following steps:

[0052] S1. In the test water tank, control the scanner to acquire images of the checkerboard calibration board in multiple poses (without using a laser), i.e., image A. Determine the sealed viewport parameters on the camera imaging path based on the checkerboard corner positions on image A. The sealed viewport parameters include the distance d0 from the camera optical center to the sealed viewport plane and the normal A of the viewport plane.

[0053] S2. In the test water tank, control the scanner to acquire checkerboard calibration plate images with line laser stripes in multiple poses (in each pose, the angle of the laser beam emitted by the laser is adjusted by the galvanometer), i.e., image B; based on the sealed viewport parameters and image B, determine the intersection point of the line laser surface and the checkerboard calibration plate, i.e., the surface point.

[0054] S3. Construct a quartic surface fitting error function for the line laser surface, and obtain the coefficients in the quartic surface fitting error function by fitting the surface points to complete the underwater line laser surface calibration.

[0055] Furthermore, before step S1, the camera intrinsic parameter matrix K needs to be determined in advance:

[0056] The scanner is controlled to acquire images of a checkerboard calibration board in multiple poses in the air. Based on a camera calibration algorithm, the camera intrinsic parameter matrix of the scanner is calculated. Specifically, the camera intrinsic parameter matrix K refers to the coordinates of spatial points in the camera imaging coordinate system. c The transformation matrix that maps P to the pixel coordinates p on the camera imaging plane can be represented as follows:

[0057]

[0058] Then c P = [X] C Y C Z C ] T p = [uv] T Substituting into (1), we get:

[0059] Z C p = K c P (2)

[0060] Specifically, firstly, the calibration board is moved within the scanner's field of view, and the scanner is controlled to acquire images of the calibration board, acquiring a total of 15-20 images of the calibration board in different poses; secondly, the pixel coordinates of the checkerboard corner points in each calibration board image are extracted; then, a world coordinate system for the checkerboard calibration board is established based on the number of checkerboard corner points and the actual size of the grid, and a pinhole camera imaging model is constructed; finally, the intrinsic parameter matrix K of the camera is obtained by minimizing the reprojection error of the corner point positions.

[0061] Further, in step S1, determining the sealed viewport parameters on the camera imaging path based on the checkerboard corner positions on image A includes the following steps:

[0062] S11. Extract the two-dimensional coordinates p of each chessboard corner point based on image A, and convert them into camera ray v0 in the air according to the camera intrinsic parameter matrix K:

[0063] v0 = K -1 [p 1] T (3)

[0064] S12. Based on the coplanar relationship of the lines in the camera imaging path, namely the camera ray v0 in the air, the normal A of the viewport plane, and the line P connecting the optical center and the corner points of the checkerboard, construct the coplanar constraint equation:

[0065] v0 T (A×(RP+t))=0 (4)

[0066] Where R and t represent the rotation matrix and translation vector in the pose transformation matrix from the checkerboard coordinate system {B} to the scanner coordinate system {S}, respectively;

[0067] S13. Based on the collinearity constraint of refraction in the camera imaging path, that is, the lines connecting the camera ray v1 in the water, the checkerboard corner point, and the refraction point q1 at the viewport plane are collinear, it can be expressed as:

[0068] (RP+t-q1)×v1=0 (5)

[0069] Where q1 can be represented as:

[0070]

[0071] S14. Solve the equations (3)-(6) simultaneously to obtain the sealed viewport parameters d0 and A on the camera's refractive imaging path;

[0072] S15. Using the sealed viewport parameters calculated in S14 as initial values, and with the goal of minimizing the reprojection error of the checkerboard corner points, the final values ​​of the sealed viewport parameters are obtained through optimization.

[0073] Specifically, first, the scanner and the checkerboard calibration plate are placed in the test water tank. Within the scanner's field of view, the camera is controlled to acquire 15-20 images of the calibration plate in different poses. Second, the pixel coordinates of the checkerboard corner points in each calibration plate image are extracted. Finally, the sealed viewport parameters of the scanner are obtained by minimizing the reprojection error of the corner point positions, namely the distance d0 from the camera optical center to the sealed viewport plane and the normal A of the viewport plane.

[0074] Further, in step S2, determining the surface points based on the sealed viewport parameters and image B includes the following steps:

[0075] S21. Based on image B, extract the two-dimensional coordinates of each checkerboard corner point. Then, based on the coplanar constraints and collinearity constraints in the camera imaging path, solve for the pose transformation matrix from the underwater checkerboard calibration plate coordinate system {B} to the scanner coordinate system {S}. It includes the rotation matrix R and the translation vector t:

[0076]

[0077] S22. Extract the center position p of the laser stripes on each calibration plate. L And convert it into camera rays in the air according to the camera imaging model. Represented as:

[0078]

[0079] S23. The camera rays in the water were calculated based on Snell's law of refraction and the calibrated sealed viewport parameters. It can be represented as:

[0080]

[0081] Where λ represents the camera ray in the water. The point on and The distance between them Indicates camera rays in the air The intersection with the viewport plane can be represented as:

[0082]

[0083]

[0084] in, Let μ0 represent the direction vector corresponding to the camera ray in the water, μ1 represent the refractive index of air, and μ0 represent the refractive index of water.

[0085] S24. Based on the camera rays in the water and pose transformation matrix The points on the surface are obtained by solving for the intersection points of the spatial ray and the plane;

[0086] Specifically, first, based on the pose transformation matrix Obtain the parameters of the calibration plate plane:

[0087]

[0088] Where, n plane d represents the normal to the plane of the calibration plate. plane This represents the distance from the calibration plate plane to the origin of the coordinate system. Represented by the pose transformation matrix The rotation matrix formed by the first three rows and the first three columns Represented by the pose transformation matrix The translation vector formed by the first three rows of the fourth column. Representing the rotation matrix The third column;

[0089] Then, by combining (13) and (12), we can obtain the three-dimensional coordinates of the intersection points of the camera rays and the checkerboard pattern in the water, that is, the points {P1,P2,P3,...,P} on the underwater laser surface. i ,...,P n}

[0090] Furthermore, based on the point-out line laser planar refraction model, it can be deduced that the line laser surface is a quartic surface; the construction process of the point-out line laser planar refraction model is as follows:

[0091] (a) A spatial rectangular coordinate system is established with the focal point of the line laser as the origin of the coordinate system, the direction of the normal to the viewport plane outward as the Z-axis, the intersection of the laser plane in the air and the viewport plane parallel to the X-axis, and the cross product of the Z-axis and the X-axis as the Y-axis.

[0092] (b) The intersection of the laser plane in the air and the viewport plane is taken as the directrix of the laser surface in the water, and the refracted laser ray is taken as the generatrix of the laser surface in the water. The ruled surface established by the above directrix and generatrix is ​​the laser surface in the water.

[0093] (c) Take a point (x, a, b) on the intersection line of the laser plane in air and the viewport plane. T Then the intersection line can be represented as:

[0094]

[0095] The line connecting a point on the intersection line to the origin is a laser ray in the laser plane, and its direction vector v0 can be expressed as:

[0096]

[0097] (d) According to the law of refraction, the direction vector v1 of the laser beam in the water after refraction is:

[0098]

[0099] (e) The equation of the line laser ruled surface constructed based on the equation of the directrix (14) and the equation of the generatrix (16) of the ruled surface can be expressed as:

[0100]

[0101] in, The highest degree of the x, y, z terms in the equation is 4, therefore the laser surface in the water in this coordinate system is a quartic surface.

[0102] Furthermore, points on a quartic surface L Point P(x,y,z) after rigid body transformation T It can be represented as:

[0103]

[0104] Right now:

[0105]

[0106] Substituting equation (19) into equation (17) does not change the degree of the highest degree term, so the underwater laser surface is a quartic surface in any coordinate system.

[0107] Furthermore, the general equation for a quartic surface has 35 coefficients (including constant terms), as follows:

[0108]

[0109] Among them, c ijk x is the equation of the quartic surface i y j z k The coefficient of the term.

[0110] Furthermore, in step S3, the quartic surface fitting error function can be defined according to the general equation of a quartic surface as follows:

[0111]

[0112] Where, {P1,P2,P3,...,P i ,...,P n Let} be the set of points of the surface to be fitted, i.e., P i =[x i y i z i ] T Let c be the coordinates of point i on the linear laser surface in a Cartesian coordinate system, and n be the total number of points on the surface to be fitted; pqr x is the error function for fitting a quartic surface. p y q z r The coefficient of the term.

[0113] Based on the surface points and the quartic surface fitting error function, the coefficients c of each term are solved with the objective of minimizing the quartic surface fitting error. pqr The precise calibration of the underwater laser surface was completed.

[0114] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for calibrating line laser surfaces for underwater three-dimensional visual measurement, characterized in that, The scanner is sealed in a housing and includes a camera, galvanometer, and laser. The calibration process includes the following steps: S1. Underwater, acquire images of the checkerboard calibration board in multiple poses using a camera, i.e., image A. Determine the sealed viewport parameters on the camera imaging path based on the checkerboard corner positions in image A. The sealed viewport parameters include the distance from the camera optical center to the viewport plane. and the normal of the viewport plane ; S2. Underwater, adjust the angle of the laser beam emitted by the laser using a galvanometer, and acquire images of the checkerboard calibration plate with line laser stripes in multiple poses using a camera, i.e., image B; determine the intersection point of the line laser surface and the checkerboard calibration plate, i.e., the surface point, based on the sealed viewport parameters and image B. S3. Construct a quartic surface fitting error function for the line laser surface, and obtain the coefficients in the quartic surface fitting error function by fitting the surface points to complete the underwater line laser surface calibration. The aforementioned linear laser surface refers to a ruled surface formed by using the intersection of the laser plane in air and the viewport plane as the guideline of the underwater linear laser surface, and the refracted laser ray as the generatrix of the underwater linear laser surface; this linear laser surface is a quartic surface.

2. The line laser surface calibration method for underwater three-dimensional vision measurement as described in claim 1, characterized in that, In step S2, the surface points are determined based on the sealed viewport parameters and image B, including the following steps: S21. Extract the coordinates of each chessboard corner point from image B, and then calculate the coordinate system of the underwater chessboard calibration plate. To the scanner coordinate system pose transformation matrix ; S22. Extract the center position of the laser stripes on each checkerboard calibration board from image B. And based on the camera intrinsic parameter matrix Convert it into camera rays in the air : ; S23, Based on sealed viewport parameters and camera rays in the air Determine camera rays in water ; S24. Based on the camera rays in the water and pose transformation matrix The points on the surface are obtained by solving for the intersection points of the spatial ray and the plane.

3. The line laser surface calibration method for underwater three-dimensional vision measurement as described in claim 2, characterized in that, Step S23, based on the sealed viewport parameters and the camera rays in the air Determine camera rays in water : in, Indicates camera rays in water The point on and The distance between them Indicates camera rays in the air Intersection with the viewport plane, The direction vector corresponding to the camera ray in the water is specifically: in, Indicates the refractive index of air. It represents the refractive index of water.

4. The line laser surface calibration method for underwater three-dimensional vision measurement as described in claim 1, characterized in that, In step S1, the sealed viewport parameters on the camera imaging path are determined based on the positions of the checkerboard corner points on image A, including the following steps: S11. Extract the coordinates of each chessboard corner point from image A. And based on the camera intrinsic parameter matrix Convert it into camera rays in the air : S12. Based on the camera ray in the air along the camera imaging path. Normal of the viewport plane The line connecting the camera's optical center and the corner points of the checkerboard grid. The three lines are coplanar, and constraint equations are constructed as follows: in, and Representing the coordinate system of the chessboard grid To the scanner coordinate system The rotation matrix and translation vector in the pose transformation matrix; S13. Based on the collinearity constraint of refraction in the camera imaging path, construct the constraint equations: in, For camera rays in water, The refraction point at the viewport plane; S14. Solve the constraint equations in S12 and S13 simultaneously to obtain the sealed viewport parameters. and .

5. The line laser surface calibration method for underwater three-dimensional vision measurement as described in claim 4, characterized in that, Step S1 also includes the following steps: S15. Using the sealed viewport parameters calculated in S14 as initial values, and with the goal of minimizing the reprojection error of the checkerboard corner points, the final values ​​of the sealed viewport parameters are obtained through optimization.

6. The line laser surface calibration method for underwater three-dimensional vision measurement as described in any one of claims 2-5, characterized in that, The camera intrinsic parameter matrix The method for determining it is as follows: Images of the checkerboard calibration board in multiple poses, i.e., image C, are pre-captured in the air using a camera; the camera intrinsic parameter matrix is ​​then calculated based on image C and the camera calibration algorithm. .

7. The line laser surface calibration method for underwater three-dimensional vision measurement as described in claim 1, characterized in that, In step S3, the constructed quartic surface fitting error function for the line laser surface is... as follows: in, Points on the laser surface i coordinates n This represents the total number of points on the surface to be fitted. In the fourth-order surface fitting error function The coefficient of the term.

8. A line laser surface calibration system for underwater three-dimensional visual measurement, characterized in that, Includes a processor for executing the line laser surface calibration method for underwater three-dimensional vision measurement as described in any one of claims 1-7.

Images