Feeding table position registration method based on 3D vision

Abstract

The invention relates to a feeding table position registration method based on 3D vision, belongs to the technical field of position registration, and solves the problem of low precision of positioning compensation of a feeding table in the prior art. The method comprises the steps that at least three circular marks are arranged on a workpiece; after the feeding table is controlled to convey the workpiece to a designated position, an image of the workpiece at the current position is collected through the 3D camera to serve as a target image; acquiring a workpiece point cloud from the target image as a target point cloud based on ellipse fitting processing on the circular mark; matching the target point cloud with the reference point cloud to obtain translation deviation and rotation deviation between the current position and the specified position; the reference point cloud is a workpiece point cloud of a specified position collected under a template attitude; and based on the translation deviation and the rotation deviation, the feeding table is controlled to move to a designated position. By means of the method, the precision of positioning compensation of the feeding table can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of position registration, and in particular to a loading platform registration method based on 3D vision. Background Art

[0002] The position registration technology of the loading table is mainly used in industrial production to achieve accurate positioning of the parts to be processed. At present, there is a certain error in the positioning of the servo loading table in industrial production. In order to ensure that the positioning accuracy of the parts to be processed meets the requirements of subsequent processes such as spraying after the loading table is in place, the positioning error of the loading table needs to be compensated.

[0003] 3D vision error positioning compensation is a method that uses three-dimensional visual information to detect and correct system errors. Its basic principle is to obtain the position information of the target object through a 3D camera and compare it with the expected position, thereby detecting the system error and compensating it by adjusting the system parameters so that the target object reaches the expected position. Using 3D vision for error positioning compensation is of great value in industrial automation, robot control, aerospace and other fields.

[0004] However, the existing 3D vision error positioning compensation technology has low positioning compensation accuracy, which cannot meet the accuracy requirements of the loading table position alignment, and is not conducive to the stability and adaptability of positioning. Summary of the invention

[0005] In view of the above analysis, an embodiment of the present invention aims to provide a loading table registration method based on 3D vision, so as to solve the problem of low accuracy of positioning compensation of the loading table in the prior art.

[0006] On the one hand, an embodiment of the present invention provides a method for aligning a loading platform position based on 3D vision, the method comprising:

[0007] Setting at least three circular marks on the workpiece;

[0008] After controlling the loading platform to transport the workpiece to a designated position, a 3D camera is used to capture an image of the workpiece at the current position as a target image;

[0009] Based on the ellipse fitting processing of the circular mark, a workpiece point cloud is acquired from the target image as a target point cloud;

[0010] Matching the target point cloud with the reference point cloud to obtain the translation deviation and rotation deviation between the current position and the specified position; the reference point cloud is the workpiece point cloud at the specified position collected under the template posture;

[0011] Based on the translation deviation and the rotation deviation, the loading platform is controlled to move to a specified position.

[0012] Based on the further improvement of the above method, the ellipse fitting process based on the circular mark is used to obtain the workpiece point cloud from the target image as the target point cloud, including the following steps:

[0013] Performing binarization processing on the target image to obtain a binarized image;

[0014] Based on performing ellipse fitting processing on the circular marks on the binary image, obtaining the three-dimensional coordinates of each circular mark in the image coordinate system;

[0015] Convert the coordinates of the circular mark in the image coordinate system into the coordinates of the loading platform coordinate system;

[0016] The acquired three-dimensional coordinates of the circular mark in the loading platform coordinate system are recorded to form target point cloud data.

[0017] Based on the further improvement of the above method, the target image is binarized to obtain a binarized image, including:

[0018] For each pixel in the target image, a neighborhood is defined with the pixel as the center;

[0019] Calculate the local threshold corresponding to the pixel point according to the pixel value of the neighborhood;

[0020] The current pixel value of each pixel is compared with its local threshold. If the current pixel value is greater than the threshold, the pixel value of the pixel is set to the maximum value; if the current pixel value is less than the threshold, the pixel value of the pixel is set to zero, thereby obtaining a binary image.

[0021] Based on the further improvement of the above method, the formula for calculating the local threshold corresponding to the pixel point according to the pixel value of the neighborhood is:

[0022]

[0023]

[0024] Where T(x,y) is the local threshold corresponding to the pixel (x,y), W(x,y) represents the neighborhood centered on the pixel (x,y), I(i,j) is the pixel value corresponding to the pixel (i,j), ω(i,j) is the weight corresponding to the pixel (i,j), and σ is the standard deviation of the Gaussian distribution.

[0025] Based on the further improvement of the above method, the method of performing ellipse fitting processing on the circular marks on the binary image to obtain the three-dimensional coordinates of each circular mark in the image coordinate system includes the following steps:

[0026] Performing ellipse fitting on the contours of the circular marks on the binary image to obtain an ellipse equation of each circular mark contour and the coordinates of the center of each ellipse;

[0027] Perform plane fitting according to the coordinates of the center of each ellipse to obtain the fitting plane equation and the normal vector of the fitting plane;

[0028] Calculating an affine transformation matrix according to the ellipse equation of the circular mark and the normal vector of the fitting plane;

[0029] The coordinates of the center of each ellipse are converted into corresponding circle center coordinates through the affine transformation matrix, and the circle center coordinates are used as the three-dimensional coordinates of the circular mark in the image coordinate system.

[0030] Based on the further improvement of the above method, the plane fitting is performed according to the coordinates of each circular mark in the image coordinate system to obtain the fitting plane equation and the normal vector of the fitting plane, including:

[0031] Take the two circular markers as a group and calculate the direction vector of the line connecting the center points of the two circular markers;

[0032] Based on the cross product of the direction vectors of the two connecting lines, the least squares method is used to solve the normal vector of the fitting plane and obtain the fitting plane equation.

[0033] Based on the further improvement of the above method, the affine transformation matrix between the image coordinate system and the loading platform coordinate system is calculated according to the ellipse equation of the circular mark and the normal vector of the fitting plane, including the steps of:

[0034] Determine the major axis a and the minor axis b of the ellipse according to the ellipse equation of the circular mark;

[0035] Calculate the angle θ between the normal vector of the fitting plane and the normal vector of the loading platform plane;

[0036] The affine transformation matrix θ between the image coordinate system and the loading platform coordinate system is calculated according to the major axis a and minor axis b of the ellipse and the angle θ. The formula is T s :

[0037]

[0038] r = 0.5 × (a + b);

[0039] Where r is the average radius of the ellipse.

[0040] Based on the further improvement of the above method, the target point cloud is matched with the reference point cloud based on an iterative closest point algorithm to obtain the translation deviation and rotation deviation between the current position and the specified position.

[0041] Based on the further improvement of the above method, the target point cloud is matched with the reference point cloud based on the iterative closest point algorithm to obtain the translation deviation and rotation deviation between the current position and the specified position, including the following steps:

[0042] Set the initial translation deviation and rotation deviation;

[0043] For each point in the target point cloud, a nearest neighbor search algorithm is used to find the closest point in the reference point cloud, and the corresponding two points are used as a matching point pair;

[0044] Calculating the error between the target point cloud and the reference point cloud according to the matching point pairs;

[0045] With the goal of minimizing the error between the target point cloud and the reference point cloud, the translation deviation and the rotation deviation are optimized and adjusted to obtain the optimal translation deviation and rotation deviation as the translation deviation and rotation deviation between the current position and the specified position.

[0046] Based on the further improvement of the above method, the formula for the error between the target point cloud and the reference point cloud is:

[0047]

[0048] q i =Rp i +t;

[0049] Where E(R,t) is the error, p i is the coordinate of the i-th point of the target point cloud, q i is the coordinate of the i-th point of the reference point cloud, R is the rotation deviation, and t is the translation deviation.

[0050] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0051] 1. The present invention provides a loading table position alignment method based on 3D vision, which can effectively improve the accuracy, stability and adaptability of positioning compensation, realize the automatic positioning compensation of the workpiece to be processed, and can effectively reduce the labor intensity of manual labor and improve production efficiency.

[0052] 2. In the present invention, by setting a plurality of circular marks distributed along the circumference on the workpiece, and using a 3D camera to capture images, the current workpiece point cloud, i.e., the target point cloud, is obtained by combining ellipse fitting processing, so as to achieve precise positioning of the workpiece, and then the target point cloud is matched with the reference point cloud to obtain the translation deviation and rotation deviation between the current position and the specified position, and the loading table is controlled to move to the specified position based on the translation deviation and rotation deviation. Such processing can automatically correct the position and posture of the workpiece, improve the accuracy of assembly or processing, and at the same time reduce manual intervention, thereby improving work efficiency and automation level.

[0053] In the present invention, the above-mentioned technical solutions can also be combined with each other to achieve more preferred combination solutions. Other features and advantages of the present invention will be described in the subsequent description, and some advantages can become obvious from the description, or can be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained through the contents particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The drawings are only for the purpose of illustrating particular embodiments and are not to be considered limiting of the present invention. Like reference symbols denote like components throughout the drawings.

[0055] Figure 1 This is a flow chart of a method for aligning the position of a loading platform based on 3D vision according to an embodiment of the present invention;

[0056] Figure 2 A schematic diagram of the distribution positions of circular marks on a workpiece according to an embodiment of the present invention;

[0057] Reference numerals:

[0058] 1- Workpiece; 2- Circular mark. DETAILED DESCRIPTION

[0059] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0060] The embodiment of the present invention provides a method for aligning the position of a loading platform based on 3D vision, such as Figure 1 As shown in , the method includes:

[0061] Step 1, setting at least three circular marks 2 on the workpiece 1;

[0062] Step 2, after controlling the loading platform to transport the workpiece 1 to the designated position, the image of the workpiece 1 at the current position is captured by a 3D camera as a target image;

[0063] Step 3, based on the ellipse fitting processing of the circular mark 2, obtaining the workpiece point cloud from the target image as the target point cloud;

[0064] Step 4, matching the target point cloud with the reference point cloud to obtain the translation deviation and rotation deviation between the current position and the specified position; the reference point cloud is the workpiece point cloud at the specified position collected under the template posture;

[0065] Based on the translation deviation and the rotation deviation, the loading platform is controlled to move to a specified position.

[0066] Compared with the prior art, the embodiment of the present invention provides a loading table position alignment method based on 3D vision, which can effectively improve the accuracy, stability and adaptability of positioning compensation, realize automatic positioning compensation of the workpiece 1 to be processed, and can effectively reduce manual labor intensity and improve production efficiency.

[0067] Specifically, in an embodiment of the present invention, by setting a plurality of circular marks 2 distributed along the circumference on the workpiece 1, and using a 3D camera to capture images, and combining ellipse fitting processing to obtain the current workpiece point cloud, that is, the target point cloud, it is possible to accurately position the workpiece 1, and then match the target point cloud with the reference point cloud to obtain the coordinate transformation matrix, and decompose it into translation deviation and rotation deviation. The loading table is controlled to move to the specified position based on the translation deviation and rotation deviation. Such processing can automatically correct the position and posture of the workpiece 1, improve the accuracy of assembly or processing, and at the same time reduce manual intervention, thereby improving work efficiency and automation level.

[0068] Step 1: Set at least three circular marks 2 on the workpiece 1.

[0069] During implementation, at least three circular marks 2 distributed along the circumference are set on the workpiece 1, such as Figure 2 as shown in .

[0070] Step 2: After controlling the loading platform to transport the workpiece 1 to the designated position, a 3D camera is used to capture an image of the workpiece 1 at the current position as a target image.

[0071] It should be noted that the loading platform in the embodiment of the present invention transports the workpiece 1 by translational motion and rotational motion. The loading platform receives the control instruction and transports the workpiece 1 to the specified position. However, due to the positioning deviation, the loading platform cannot stop accurately at the specified position, which causes the positioning of the workpiece 1 to deviate, that is, there is a deviation between the current position of the workpiece 1 and the specified position.

[0072] Among them, the 3D camera can obtain a planar image and depth information of the photographed object, that is, three-dimensional position information.

[0073] Step 3, based on the ellipse fitting processing of the circular mark 2, obtaining the workpiece point cloud from the target image as the target point cloud, includes the following steps:

[0074] Performing binarization processing on the target image to obtain a binarized image;

[0075] Based on performing ellipse fitting processing on the circular marks 2 on the binary image, obtaining the three-dimensional coordinates of each circular mark 2 in the image coordinate system;

[0076] Convert the coordinates of the circular mark 2 in the image coordinate system into the coordinates of the loading platform coordinate system;

[0077] The acquired three-dimensional coordinates of the circular mark 2 in the loading platform coordinate system are recorded to form target point cloud data.

[0078] In the embodiment of the present invention, high-precision positioning of the circular mark 2 can be achieved through binarization processing and ellipse fitting. The binarization processing simplifies the gray value of the image into two values, usually 0 and 255. The binarization processing facilitates subsequent ellipse fitting and edge detection by simplifying the image, enhancing contrast, reducing noise and other effects.

[0079] Specifically, performing binarization processing on the target image to obtain a binarized image includes:

[0080] For each pixel in the target image, a neighborhood is defined with the pixel as the center;

[0081] Calculate the local threshold corresponding to the pixel point according to the pixel value of the neighborhood;

[0082] The current pixel value of each pixel is compared with its local threshold. If the current pixel value is greater than the threshold, the pixel value of the pixel is set to the maximum value; if the current pixel value is less than the threshold, the pixel value of the pixel is set to zero, thereby obtaining a binary image.

[0083] In the embodiment of the present invention, when performing binarization processing, an adaptive threshold processing technology is used to complete the threshold processing of the image using a variable threshold. The adaptive threshold processing adaptively determines the threshold for different local areas, and the result is more accurate, which improves the image segmentation accuracy and can better retain the image detail information.

[0084] Specifically, the formula for calculating the local threshold corresponding to the pixel point according to the pixel value of the neighborhood is:

[0085]

[0086]

[0087] Where T(x,y) is the local threshold corresponding to the pixel (x,y), W(x,y) represents the neighborhood centered on the pixel (x,y), I(i,j) is the pixel value corresponding to the pixel (i,j), ω(i,j) is the weight corresponding to the pixel (i,j), and σ is the standard deviation of the Gaussian distribution.

[0088] In the embodiment of the present invention, a local threshold is obtained by calculating a weighted average value of adjacent areas around each pixel, and the threshold is used to process the current pixel.

[0089] After binarization, the image only contains two pixel values, which reduces the complexity of the image and highlights the contour and shape features in the image, which can improve the accuracy of subsequent ellipse fitting.

[0090] Specifically, the method of obtaining the three-dimensional coordinates of each circular mark 2 in the image coordinate system based on performing ellipse fitting processing on the circular mark 2 on the binary image includes the following steps:

[0091] Performing ellipse fitting on the contours of the circular marks 2 on the binary image to obtain an ellipse equation of the contour of each circular mark 2 and the coordinates of the center of each ellipse;

[0092] Perform plane fitting according to the coordinates of the center of each ellipse to obtain the fitting plane equation and the normal vector of the fitting plane;

[0093] Calculating an affine transformation matrix according to the ellipse equation of the circular mark 2 and the normal vector of the fitting plane;

[0094] The coordinates of the center of each ellipse are converted into corresponding circle center coordinates through the affine transformation matrix, and the circle center coordinates are used as the three-dimensional coordinates of the circular mark 2 in the image coordinate system.

[0095] More specifically, performing ellipse fitting on the contours of the circular marks 2 on the binary image to obtain the ellipse equation of each contour of the circular mark 2 and the coordinates of the center of each ellipse includes the following steps:

[0096] The edge detection algorithm is used to extract the edge point coordinates of the circular mark 2 on the binary image, and the least square method is used to fit the edge point coordinates to obtain the ellipse equation of the contour of the circular mark 2;

[0097] The coordinates of the center of the ellipse are obtained based on the ellipse equation of the circular marker 2 outline.

[0098] The standard equation of the ellipse is:

[0099] Ax 2 +bxy+Cy 2 +Dx+Ey+F=0;

[0100] The quadratic form of the standard equation of an ellipse is:

[0101] x T Qx=0;

[0102]

[0103]

[0104] The objective function of fitting the edge point coordinates using the least squares method is:

[0105] X opt =argmin(E ellipse );

[0106]

[0107] In the formula, X opt To optimize the goal, E ellipse is the error; u i and v i are the coordinates of the i-th edge point of the extracted circle center mark; u projected and v projected are the coordinates of the i-th edge point to be found.

[0108] More specifically, performing plane fitting according to the coordinates of each circular mark 2 in the image coordinate system to obtain a fitting plane equation and a normal vector of the fitting plane includes:

[0109] Take the two circular markers 2 as a group and calculate the direction vector of the line connecting the center points of the two circular markers 2;

[0110] Based on the cross product of the direction vectors of the two connecting lines, the least squares method is used to solve the normal vector of the fitting plane and obtain the fitting plane equation.

[0111] The fitting plane equation is: n x X+n y Y+n z Z = 0;

[0112] The plane normal vector is:

[0113] Among them, X, Y, and Z are the coordinates of the X-axis, Y-axis, and Z-axis in the image coordinate system, respectively. x 、n y 、n z are the normal vectors of the fitted plane Components on the X-axis, Y-axis, and Z-axis.

[0114] The normal vector of the fitted plane can be obtained by the direction vector of any two connecting lines. The cross product of is:

[0115] Since the direction vectors of any two connecting lines can be calculated, the least squares method is used for fitting to obtain the optimal value of the normal vector of the fitting plane.

[0116] More specifically, the affine transformation matrix between the image coordinate system and the loading platform coordinate system is calculated according to the ellipse equation of the circular mark 2 and the normal vector of the fitting plane, comprising the steps of:

[0117] Determine the major axis a and the minor axis b of the ellipse according to the ellipse equation of the circular mark 2;

[0118] Calculate the angle θ between the normal vector of the fitting plane and the normal vector of the loading platform plane;

[0119] The affine transformation matrix θ between the image coordinate system and the loading platform coordinate system is calculated according to the major axis a and minor axis b of the ellipse and the angle θ. The formula is T s :

[0120]

[0121] r = 0.5 × (a + b);

[0122] Where r is the average radius of the ellipse.

[0123] It should be noted that the normal vector of the loading platform plane is Known,

[0124] Specifically, converting the coordinates of the circular mark 2 in the image coordinate system into the coordinates of the loading platform coordinate system includes:

[0125] The coordinates of the circular mark 2 in the image coordinate system are converted into the coordinates of the camera coordinate system according to the intrinsic parameter matrix of the 3D camera;

[0126] According to the external parameter matrix of the 3D camera, the coordinates of the circular mark 2 in the image coordinate system are converted into the coordinates of the loading platform coordinate system.

[0127] The coordinates of the circular mark 2 in the image coordinate system are converted into the coordinates of the camera coordinate system according to the intrinsic parameter matrix of the 3D camera. The formula is as follows:

[0128]

[0129] Where [uv 1] T is the coordinate expression of the circular marker 2 in the image coordinate system, [x c y c z c ]T is the coordinate expression of the circular marker 2 in the camera coordinate system, and K is the intrinsic parameter matrix.

[0130] The companies in the internal reference matrix are:

[0131] According to the external parameter matrix of the 3D camera, the coordinates of the circular mark 2 in the image coordinate system are converted into the coordinates of the loading platform coordinate system. The formula is:

[0132]

[0133] In the formula, left [x w y w z w 1] T is the coordinate point expression of circular mark 2 in the loading platform coordinate system, [x c y c z c 1] T It is the expression of the ellipse coordinate point in the camera coordinate system, and T is the external parameter matrix.

[0134] The formula of the external parameter matrix is:

[0135]

[0136] Where R is a 3×3 rotation matrix and t is a 3×1 translation matrix.

[0137] During implementation, the intrinsic parameter matrix and extrinsic parameter matrix of the 3D camera can be obtained through calibration.

[0138] Specifically, the target point cloud is matched with the reference point cloud based on an iterative closest point algorithm to obtain a translation deviation and a rotation deviation between a current position and a specified position.

[0139] More specifically, matching the target point cloud with the reference point cloud based on an iterative closest point algorithm to obtain a translation deviation and a rotation deviation between a current position and a specified position includes the following steps:

[0140] Set the initial translation deviation and rotation deviation;

[0141] For each point in the target point cloud, a nearest neighbor search algorithm is used to find the closest point in the reference point cloud, and the corresponding two points are used as a matching point pair;

[0142] Calculating the error between the target point cloud and the reference point cloud according to the matching point pairs;

[0143] With the goal of minimizing the error between the target point cloud and the reference point cloud, the translation deviation and the rotation deviation are optimized and adjusted to obtain the optimal translation deviation and rotation deviation as the translation deviation and rotation deviation between the current position and the specified position.

[0144] Define the reference point cloud as P = (p1, p2, p3, ..., p N ); the coordinates of each point in the reference point cloud are p i =(x i ,y i ,z i ) T ; The target point cloud is Q = (q1, q2, q3, ..., q N );The coordinates of each point in the target point cloud are q i =(x i ′,y i ′,z i ′) T .

[0145] The formula for the error between the target point cloud and the reference point cloud is:

[0146]

[0147] q i =Rp i +t;

[0148] Where E(R,t) is the error, p i is the coordinate of the i-th point of the reference point cloud, q i is the coordinate of the i-th point of the target point cloud, R is the rotation deviation, and t is the translation deviation.

[0149] During implementation, the least square method is used to optimize and adjust the translation deviation and the rotation deviation.

[0150] The formulas for the translation deviation and the rotation deviation are:

[0151] R=VU T ;

[0152]

[0153] Among them, U and V T are the left singular vector matrix and the right singular vector matrix of the covariance matrix H, respectively. and are the centroids of the reference point cloud and the target point cloud respectively. They are calculated by the following steps:

[0154] Calculate the centroid of the reference point cloud and the target point cloud and

[0155]

[0156]

[0157] Centralized Data:

[0158]

[0159]

[0160] Calculate the covariance matrix H:

[0161]

[0162] Singularity decomposition:

[0163] H=U∑V T ;

[0164] Perform singular value decomposition (SVD) on the covariance matrix H to obtain the left singular vector matrix U, the singular value matrix Σ, and the right singular vector matrix V T .

[0165] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0166] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for registering the position of a loading platform based on 3D vision, characterized in that: The method comprises: Setting at least three circular marks on the workpiece; After controlling the loading platform to transport the workpiece to a designated position, an image of the workpiece at the current position is captured by a 3D camera as a target image; Based on the ellipse fitting processing of the circular mark, acquiring a workpiece point cloud from the target image as a target point cloud; Matching the target point cloud with the reference point cloud to obtain the translation deviation and rotation deviation between the current position and the specified position; the reference point cloud is the workpiece point cloud at the specified position collected under the template posture; Based on the translation deviation and the rotation deviation, the loading platform is controlled to move to a specified position.

2. The method for aligning the position of a loading platform according to claim 1, characterized in that: The step of obtaining a workpiece point cloud as a target point cloud from the target image based on an ellipse fitting process on a circular mark comprises the following steps: Performing binarization processing on the target image to obtain a binarized image; Based on performing ellipse fitting processing on the circular marks on the binary image, obtaining the three-dimensional coordinates of each circular mark in the image coordinate system; Convert the coordinates of the circular mark in the image coordinate system into the coordinates of the loading platform coordinate system; The acquired three-dimensional coordinates of the circular mark in the loading platform coordinate system are recorded to form target point cloud data.

3. The method for aligning the position of a loading platform according to claim 2, characterized in that: Perform binarization on the target image to obtain a binarized image, including: For each pixel in the target image, a neighborhood is defined with the pixel as the center; Calculate the local threshold corresponding to the pixel point according to the pixel value of the neighborhood; The current pixel value of each pixel is compared with its local threshold. If the current pixel value is greater than the threshold, the pixel value of the pixel is set to the maximum value; if the current pixel value is less than the threshold, the pixel value of the pixel is set to zero, thereby obtaining a binary image.

4. The method for aligning the position of a loading platform according to claim 3, characterized in that: The formula for calculating the local threshold corresponding to the pixel point according to the pixel value of the neighborhood is: Where T(x,y) is the local threshold corresponding to the pixel (x,y), W(x,y) represents the neighborhood centered on the pixel (x,y), I(i,j) is the pixel value corresponding to the pixel (i,j), ω(i,j) is the weight corresponding to the pixel (i,j), and σ is the standard deviation of the Gaussian distribution.

5. The method for aligning the position of a loading platform according to claim 2, characterized in that: The method of performing ellipse fitting processing on the circular marks on the binary image to obtain the three-dimensional coordinates of each circular mark in the image coordinate system comprises the following steps: Performing ellipse fitting on the contours of the circular marks on the binary image to obtain an ellipse equation of each circular mark contour and the coordinates of the center of each ellipse; Perform plane fitting according to the coordinates of the center of each ellipse to obtain the fitting plane equation and the normal vector of the fitting plane; Calculating an affine transformation matrix according to the ellipse equation of the circular mark and the normal vector of the fitting plane; The coordinates of the center of each ellipse are converted into corresponding circle center coordinates through the affine transformation matrix, and the circle center coordinates are used as the three-dimensional coordinates of the circular mark in the image coordinate system.

6. The method for aligning the position of a loading platform according to claim 5, characterized in that: The step of performing plane fitting according to the coordinates of each circular mark in the image coordinate system to obtain a fitting plane equation and a normal vector of the fitting plane includes: Take the two circular markers as a group and calculate the direction vector of the line connecting the center points of the two circular markers; Based on the cross product of the direction vectors of the two connecting lines, the least squares method is used to solve the normal vector of the fitting plane and obtain the equation of the fitting plane.

7. The method for aligning the position of a loading platform according to claim 5, characterized in that: The step of calculating the affine transformation matrix between the image coordinate system and the loading platform coordinate system according to the ellipse equation of the circular mark and the normal vector of the fitting plane comprises the following steps: Determine the major axis a and the minor axis b of the ellipse according to the ellipse equation of the circular mark; Calculate the angle θ between the normal vector of the fitting plane and the normal vector of the loading platform plane; The affine transformation matrix θ between the image coordinate system and the loading platform coordinate system is calculated according to the major axis a and minor axis b of the ellipse and the angle θ. The formula is T s : r = 0.5 × (a + b); Where r is the average radius of the ellipse.

8. The method for aligning the position of a loading platform according to any one of claims 1 to 7, characterized in that: The target point cloud is matched with the reference point cloud based on an iterative closest point algorithm to obtain a translation deviation and a rotation deviation between a current position and a specified position.

9. The method for aligning the position of a loading platform according to claim 8, characterized in that: The method of matching the target point cloud with the reference point cloud based on an iterative closest point algorithm to obtain a translation deviation and a rotation deviation between a current position and a specified position includes the following steps: Set the initial translation deviation and rotation deviation; For each point in the target point cloud, a nearest neighbor search algorithm is used to find the closest point in the reference point cloud, and the corresponding two points are used as a matching point pair; Calculating the error between the target point cloud and the reference point cloud according to the matching point pairs; With the goal of minimizing the error between the target point cloud and the reference point cloud, the translation deviation and the rotation deviation are optimized and adjusted to obtain the optimal translation deviation and rotation deviation as the translation deviation and rotation deviation between the current position and the specified position.

10. The method for aligning the position of a loading platform according to claim 9, characterized in that: The formula for the error between the target point cloud and the reference point cloud is: q i =Rp i +t; Where E(R,t) is the error, p i is the coordinate of the i-th point of the target point cloud, q i is the coordinate of the i-th point of the reference point cloud, R is the rotation deviation, and t is the translation deviation.

Images