A high-precision correlation method for multiple cameras of a plug-in machine

Abstract

The present invention discloses a high-precision association method for multiple cameras of an insertion machine, which realizes high-precision calibration and conversion of upper and lower camera coordinate systems, thereby greatly improving the measurement, positioning and navigation performance of the visual system and meeting the strict requirements of high-end equipment production on product quality. The present invention provides a quantitative and high-precision camera association calibration method, so that machine vision technology can better meet the key requirements in high-end equipment production and achieve a leap towards automated and intelligent production.

Description

Technical Field

[0001] The present invention relates to the technical field of plug-in machines, and in particular to a high-precision correlation method for multiple cameras of a plug-in machine. Background Art

[0002] Existing plug-in machines usually use two cameras, upper and lower, to achieve visual positioning and detection, but the spatial correlation of the upper and lower cameras has always been a technical problem. The typical correlation method is to use a crosshair device to manually adjust the upper and lower cameras to align with the crosshairs to achieve correlation, but this method has the following problems: Manually moving the crosshair device and aligning the two cameras precisely requires high-precision manual operation, which is a highly difficult technical job in itself and cannot guarantee operational accuracy and repeatability.

[0003] Whether the camera is accurately aligned with the crosshairs depends entirely on the operator's subjective judgment, and quantitative evaluation and precision control cannot be achieved, which makes it difficult to accurately establish the spatial transformation relationship between the two cameras.

[0004] The focal lengths of the upper and lower cameras are different, so it is unrealistic to achieve the clearest imaging state at the same time, which will lead to differences in image quality and affect the effect of subsequent processing.

[0005] It can be seen that the existing upper and lower camera correlation method based on the cross-line device has large errors and instability, and it is difficult to achieve high-precision requirements, which seriously restricts the measurement accuracy and detection performance of the plug-in machine vision system. In order to solve this technical problem, a new method is urgently needed to achieve the spatial correlation and calibration of the upper and lower cameras with high precision and quantitativeness. Summary of the invention

[0006] In order to overcome the shortcomings of the existing technical solutions, the present invention provides a high-precision association method for multiple cameras of an insertion machine, which can effectively solve the problem that the upper and lower camera association method based on a cross-line device proposed in the background technology has large errors and instability and is difficult to achieve high precision.

[0007] The technical solution adopted by the present invention to solve the technical problem is: a method for high-precision association of multiple cameras of an insertion machine, comprising the following steps: S1: Calibrate the upper camera and the lower camera involved in the association respectively, and establish a conversion relationship from the camera coordinate system to the XY module coordinate system; S2: Use a high-precision calibration pin to identify and determine its pixel coordinates in the lower camera image, and then convert them into spatial coordinates P 下相机 ; S3: Mark the small hole on the XY module working plane and record its coordinates as P 打孔 ; S4: Move the upper camera to align with the pinhole mark, identify the pixel coordinates of the pinhole in the upper camera image, and convert them into spatial coordinates P 上相机 ; S5: Calculate the spatial coordinates P where the upper camera center coincides with the lower camera center 中心 , to achieve accurate conversion of the two camera coordinate systems.

[0008] Furthermore, the method further includes the step of repeatedly executing the above steps for different combinations of upper and lower cameras to establish a high-precision spatial coordination relationship between multiple camera combinations.

[0009] Furthermore, camera calibration is performed using a conventional calibration plate, which is achieved through feature point recognition and coordinate matching.

[0010] Furthermore, the pixel coordinates of the calibration pin in the lower camera image are obtained through an image recognition algorithm and converted into XY module coordinates.

[0011] Furthermore, the coordinates P of the pinhole mark 打孔 As a spatial positioning reference, it is used to calculate the spatial coordinate P where the center of the upper camera coincides with the center of the lower camera. 中心 .

[0012] Furthermore, P 中心 The calculation formula is: 中心 =P 上相机 - (P 打孔 -P 下相机 ).

[0013] Furthermore, the method further comprises the step of performing distortion correction on the camera calibration result to improve the accuracy of coordinate conversion.

[0014] Furthermore, the method also includes the step of ensuring the accuracy of the relative positions and postures between the cameras when establishing a multi-camera collaborative relationship.

[0015] Further, including: The camera calibration module is used to calibrate the upper camera and the lower camera and establish the conversion relationship from the camera coordinate system to the XY module coordinate system; A calibration pin recognition module, used to recognize the calibration pin in the lower camera image and determine its pixel coordinates; A coordinate conversion module is used to convert the pixel coordinates of the calibration pin and the pinhole into spatial coordinates, and calculate the spatial coordinates where the center of the upper camera coincides with the center of the lower camera; The multi-camera collaboration module is used to establish a high-precision spatial collaboration relationship between multiple camera combinations.

[0016] Furthermore, it also includes a distortion correction module for performing distortion correction on the camera calibration result to improve the accuracy of coordinate conversion.

[0017] Compared with the prior art, the present invention has the following beneficial effects: Easy to operate. You only need to move the module to position the camera to the approximate position, and the fine adjustment is automatically completed by the algorithm. There is no need to rely on complex mechanical equipment and human eye judgment to achieve accurate calibration, which greatly reduces the difficulty of operation.

[0018] Accuracy is guaranteed. Image recognition technology is used to obtain key data, avoid subjective judgment errors, and achieve quantitative positioning. At the same time, by moving the calibration needle up and down to the clearest object distance of each camera for imaging, obtaining a clear image helps improve recognition accuracy. These two points are crucial to obtaining high repeatability calibration results.

[0019] Low hardware requirements. No expensive mechanical equipment is required, and it can be achieved through relatively simple module movement and image acquisition equipment, which reduces equipment costs and implementation difficulties.

[0020] Digital calibration. The collected image data is converted into spatial coordinates through algorithm processing, realizing a fully digital calibration process from data acquisition, processing to calibration results. This makes the calibration process standardized and modular, and more in line with modern production requirements.

[0021] Facing the future development direction. Image recognition technology and automated algorithm calibration are two major trends in the development of intelligent machine vision. This invention conforms to these two major development directions, uses advanced technical means to achieve major innovations in calibration methods, and faces the future of machine vision technology and intelligent manufacturing. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a flow chart of the association method of the present invention; Figure 2 It is a specific step diagram of the association method of the present invention. DETAILED DESCRIPTION

[0023] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0024] The following describes the embodiments of the present disclosure through specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present disclosure. Example

[0025] like Figure 1-2 As shown, the present invention provides a method for high-precision association of multiple cameras of a plug-in machine, comprising the following steps: Step 1: Calibrate the upper and lower cameras involved in the association respectively, and establish the conversion relationship from the camera coordinate system to the XY module coordinate system. The calibration is performed using a conventional calibration plate, and is achieved through feature point recognition and coordinate matching.

[0026] Select a standard calibration plate, such as a checkerboard calibration plate. The feature points on the calibration plate, such as the intersections of the checkerboard, should be clear and easy to identify. Make sure the camera is fixed and adjust parameters such as focus and exposure to obtain a clear image. Use software tools for feature point detection and calibration, such as OpenCV.

[0027] Take images of the calibration plate from multiple different angles and positions to ensure that the calibration plate occupies a sufficient proportion in the image and that the feature points are clearly visible. Save the captured images in a folder for subsequent processing.

[0028] Use an image processing library, such as OpenCV, to load the calibration plate image; convert the image to a grayscale image to reduce the amount of calculation; use an appropriate feature point detection algorithm (such as Harris corner detection, Shi-Tomasi corner detection, or a specialized checkerboard feature point detection algorithm) to detect feature points on the calibration plate; perform sub-pixel optimization on the detected feature points to improve accuracy.

[0029] According to the design of the calibration plate, the theoretical three-dimensional coordinates of each feature point are generated. For a checkerboard calibration plate, the coordinates are usually the row and column indices of the grid points multiplied by the size of the grid; the actual image coordinates of the detected feature points are matched with the theoretical coordinates.

[0030] Solve the camera's intrinsic and extrinsic parameters. Use the matched feature point coordinates and the corresponding theoretical coordinates to solve the camera's intrinsic parameters (such as focal length, principal point, distortion coefficient) and extrinsic parameters (such as rotation matrix, translation vector) through the least squares algorithm.

[0031] After using some images for preliminary calibration, the remaining images can be used for verification. The accuracy of the calibration is evaluated by calculating the reprojection error. The average value of the reprojection error is less than 0.5 pixels. The calibration result is generally considered reliable and optimized as needed. Finally, the calculated camera parameters are saved in a file for subsequent use.

[0032] Step 2: Place a high-precision calibration pin at the center of the optical axis of the lower camera, and detect the pixel coordinates of the calibration pin in the lower camera image through image recognition.

[0033] Select a thin calibration pin whose diameter and length should be suitable for the camera's resolution and field of view. Make sure the lower camera is stably mounted and its optical axis is perpendicular to the plane where the calibration pin is located. Use a precision calibration tool to adjust the position of the camera to ensure that the center of the optical axis is aligned with the center of the calibration pin.

[0034] Use a precision translation stage or robotic arm to precisely place the calibration pin at the center of the camera's optical axis, ensuring that the calibration pin is perpendicular to the camera's imaging plane.

[0035] Within the detected edge, the centroid of the calibration pin is calculated by weighted averaging the edge pixels, and the calculated centroid coordinates are used as the pixel coordinates of the calibration pin in the image.

[0036] To improve the accuracy, the pixel coordinates of the calibration pin can be measured multiple times under different lighting conditions and angles, and the average value can be calculated. The camera can be calibrated using the extracted pixel coordinates and the actual length of the calibration pin and camera parameters to further improve the measurement accuracy.

[0037] Step 3: According to the calibration result of the lower camera, convert the pixel coordinates of the calibration pin in the image into XY module coordinates to obtain the spatial coordinates P of the calibration pin at the center of the lower camera. 下相机 .

[0038] Make sure the camera has been accurately calibrated to obtain the camera's intrinsic parameter matrix (including focal length, principal point, distortion coefficient, etc.) and extrinsic parameter matrix (including rotation matrix and translation vector).

[0039] The conversion from pixel coordinates to normalized coordinates uses the camera's intrinsic parameter matrix to convert the pixel coordinates (u, v) of the calibration pin in the image to normalized image coordinates (x, y).

[0040] The conversion from normalized coordinates to world coordinates uses the camera's external parameter matrix to convert the normalized coordinates (x, y, 1) to world coordinates (X W , Y W , Z W ).

[0041] Once we have the position of the calibration needle in the world coordinate system (X W , Y W , Z W ), and knowing the relationship between the XY module coordinate system and the world coordinate system, we can calculate the position of the calibration needle in the XY module coordinate system (X 模组 ,Y 模组） However, the question requires the spatial coordinates P of the calibration needle at the center of the lower camera. 下相机 , which usually means we need to convert the world coordinates (X W , Y W, Z W ) is converted to coordinates in a coordinate system with the camera center as its origin.

[0042] The optical axis of the camera is perpendicular to the XY plane, and the center of the camera coincides with the center of the XY coordinate system. We can simply calculate the position of the camera center in the world coordinate system by subtracting the position of the camera center in the world coordinate system (X C , Y C , Z C ) to get P 下相机 .Right now: P 下相机 = (X W -X C , Y W -Y C , Z W -Z C ) If the position and orientation of the camera do not correspond exactly to the XY module coordinate system, then we need to use a more complex transformation matrix to transform the world coordinates into coordinates in a coordinate system with the camera center as the origin.

[0043] When the position and direction of the camera do not completely correspond to the XY module coordinate system, we need to use a conversion matrix (also called a transformation matrix) to convert the world coordinates into coordinates in a coordinate system with the center of the camera as the origin (usually called the camera coordinate system). This conversion matrix usually includes two parts, rotation and translation, and can be expressed as a 4x4 homogeneous transformation matrix.

[0044] The specific form of the transformation matrix is ​​as follows:

[0045] in: Where R is a 3x3 rotation matrix, which is used to describe the rotation relationship from the world coordinate system to the camera coordinate system.

[0046] t is a 3x1 translation vector that describes the translation relationship from the world coordinate system to the camera coordinate system.

[0047] 0 is a 1x3 zero vector used to extend homogeneous coordinates.

[0048] 1 is a constant term used to keep the scale of homogeneous coordinates unchanged.

[0049] Using this transformation matrix, we can transform the world coordinates (Xw, Yw, Zw, 1) to the camera coordinates (Xc, Yc, Zc, 1). The specific calculation is as follows: First, the world coordinates are rotated by the rotation matrix R, and then the translation vector t is added, that is,

[0050] Step 4: Move the calibration needle to the working plane of the XY module and make a small hole mark on the surface. Record the coordinate as P 打孔 , as a spatial positioning reference.

[0051] Move the calibration pin to the working plane of the XY module and make a small hole mark on the surface. Record the coordinates as P punch hole as a spatial positioning reference. This can be achieved by following the steps below: Use the XY module's control system to move the calibration pin to the preset starting position on the work plane.

[0052] The position of the calibration pin is precisely adjusted by the control system to ensure that it remains perpendicular to the working plane.

[0053] After confirming that the calibration pin is in the correct position, start the punching device and punch a hole at the location of the calibration pin.

[0054] The coordinates of the hole are recorded as Pdrill as a reference for spatial positioning. High-precision measuring equipment such as a laser rangefinder or a three-dimensional coordinate measuring machine can be used to measure and record the coordinates.

[0055] Step 5: Move the upper camera to align the center of the image with the pinhole mark, and obtain the pixel coordinates of the pinhole in the upper camera image through image recognition.

[0056] The image containing the pinhole mark is captured by the camera to ensure that the image is clear and has moderate contrast for subsequent processing. The captured image is pre-processed, such as grayscale, filtering, binarization, etc. to improve recognizability, so as to improve image quality and simplify subsequent processing steps.

[0057] Use image processing algorithms to extract features of the pinhole mark, including edges, contours or specific shapes. Use a pre-set template of a pinhole mark, and then perform template matching in the actual image to identify the position and features of the pinhole mark. Hough transform and Canny edge detection methods can also be used to identify the pinhole mark.

[0058] Coordinate calculation,Based on the extracted features, the pixel coordinates of the pinhole mark in the image are calculated, and the coordinates of the,center of mass of the image are used as the pixel coordinates of the pinhole mark.

[0059] Calculate the offset. According to the pixel coordinates of the pinhole mark and the pixel coordinates of the center of the image, calculate the offset between the two. The offset is the pixel difference in the horizontal direction (X axis) and the vertical direction (Y axis).

[0060] Control the camera movement, convert the calculated offset into the control command of the XY module, and use the PLC controller to drive the camera movement to align the small hole mark.

[0061] Step 6: According to the calibration results of the upper camera, convert the pixel coordinates of the pinhole in the image into XY module coordinates to obtain the spatial coordinates P of the pinhole at the center of the upper camera. 上相机 .

[0062] Camera calibration,First, the camera needs to be calibrated to obtain the camera's intrinsic parameter matrix (including focal length, principal point coordinates, distortion coefficient, etc.) and extrinsic parameter matrix (including rotation matrix and translation vector). After the calibration is completed, the camera's intrinsic parameter matrix is ​​obtained.

[0063] Dedistortion,The image is dedistorted using the distortion coefficients in the intrinsic parameter matrix to eliminate the influence of lens distortion on pixel coordinates.

[0064] Dedistort the points in the camera coordinate system according to the distortion coefficients.

[0065] Radial distortion: mainly occurs at the edge of the image, causing the image to shrink toward the center or expand outward. Use radial distortion coefficients (k1, k2, k3) to perform radial correction on the image.

[0066] Tangential distortion: It is caused by the fact that the lens and the imaging plane are not completely parallel. The tangential distortion coefficients (p1, p2) are used to perform tangential correction on the image.

[0067] Normalized coordinates, convert the pixel coordinates of the pinhole in the image into normalized coordinates. Normalized coordinates are relative to the size of the image sensor, with the origin located at the center of the image, and the x and y axes are parallel to the horizontal and vertical directions of the image sensor, respectively.

[0068] Depth information,Pixel coordinates contain only two-dimensional information (x and y), while camera coordinates are three-dimensional (X, Y, Z).,The depth information is the distance between the pinhole and the plane,which is parallel to the camera plane. This distance is directly used as the,Z coordinate.

[0069] Coordinate conversion: Use the intrinsic matrix and depth information to convert the normalized coordinates to camera coordinates. This step usually involves matrix multiplication and inverse operations.

[0070] Conversion of camera coordinates to XY module coordinates: The extrinsic matrix uses the extrinsic matrix (rotation matrix and translation vector) obtained during camera calibration to convert the camera coordinates into world coordinates (i.e. XY module coordinates). The extrinsic matrix describes the position and orientation of the camera relative to the world coordinate system (here, the XY module coordinate system).

[0071] Calculate the spatial coordinates, and convert the camera coordinates into XY module coordinates through matrix multiplication and addition operations to obtain the spatial coordinates P of the pinhole at the center of the upper camera. 上相机 .

[0072] Step 7: Calculate the spatial coordinates where the upper camera center coincides with the lower camera center: P 中心 =P 上相机 -(P 打孔 -P 下相机 ) .

[0073] Step 8: Solve for P 中心 After that, the accurate conversion from the lower camera coordinate system to the upper camera coordinate system can be realized, and high-precision positioning coordination of the two cameras can be achieved.

[0074] Step 9: Repeat steps 2 to 8 for different upper and lower cameras to establish a high-precision spatial coordination relationship between multiple camera combinations to achieve precise positioning and calibration of the entire visual system.

[0075] When using the embodiment of the present application: Code for camera calibration using OpenCV: import cv2 import numpy as np import glob # Set the size of the checkerboard calibration board chessboard_size = (9, 6) square_size = 1.0 # The side length of each small grid, in meters # Prepare a list for storing image points and corresponding 3D points objpoints = [] # The position of the 3D point in the world coordinate system imgpoints = [] # Pixel coordinates of 2D points on the image plane # Generate the 3D coordinates of the corner points of the chessboard objp = np.zeros((chessboard_size[0] * chessboard_size[1], 3),np.float32) objp[:,:2]=np.mgrid[0:chessboard_size[0],0:chessboard_size[1]].T.reshape(-1, 2) objp *= square_size # Read the calibration plate image images = glob.glob('path_to_images / *.jpg') for fname in images: img = cv2.imread(fname) gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) # Find the corner points of the chessboard ret, corners = cv2.findChessboardCorners(gray, chessboard_size,None) # If corner points are found, add object points and image points if ret: objpoints.append(objp) # Improve the accuracy of corner points corners2 = cv2.cornerSubPix(gray, corners, (11, 11), (-1, -1),criteria) imgpoints.append(corners2) # Draw and display corner points img = cv2.drawChessboardCorners(img, chessboard_size,corners2, ret) cv2.imshow('img', img) cv2.waitKey(500) cv2.destroyAllWindows() # Calibrate the camera ret, mtx, dist, rvecs, tvecs = cv2.calibrateCamera(objpoints,imgpoints, gray.shape[::-1], None, None) # Print calibration results print("Camera intrinsic parameter matrix:\n", mtx) print("Distortion coefficient:\n", dist) print("Rotation vector:\n", rvecs) print("Translation vector:\n", tvecs) # Save the calibration results np.savez('camera_parameters.npz', mtx=mtx, dist=dist, rvecs=rvecs,tvecs=tvecs) The present invention discloses a high-precision association method for multiple cameras of an insertion machine, so as to realize high-precision calibration and conversion of upper and lower camera coordinate systems, thereby greatly improving the measurement, positioning and navigation performance of the visual system and meeting the strict requirements of high-end equipment production on product quality.

[0076] A new method is used to achieve high-precision spatial correlation and calibration of the upper and lower cameras, and to establish an accurate mapping relationship between the two camera coordinate systems, thereby ensuring that various measurements, positioning and detection based on the vision system meet high-precision requirements.

[0077] The new method avoids reliance on the subjective judgment of the operator and adopts quantitative algorithms and methods to achieve automatic association between the upper and lower cameras, which reduces the errors and instability caused by human factors and can achieve high repeatability.

[0078] High-precision camera calibration results enable precise measurement and analysis in image data, greatly improving the accuracy of vision-based detection, positioning, and navigation, while also simplifying and optimizing related algorithms and processes to increase efficiency.

[0079] Precise and quantitative automatic multi-camera calibration and association are the cornerstones of machine vision positioning, navigation and control in intelligent production equipment. The implementation of this invention provides key technical guarantees for the intelligent upgrade of vision systems and production equipment.

[0080] In summary, the purpose of the present invention is to provide a quantitative and high-precision camera correlation calibration method, so that machine vision technology can better meet the key requirements in the production of high-end equipment and achieve a leap towards automated and intelligent production.

[0081] The present invention uses a high-precision calibration target to achieve a spatial reference, places the calibration target in the workspace, obtains its spatial coordinates through high-precision measurement, and uses it as a spatial reference to achieve multi-camera correlation calibration. This avoids the error caused by subjective judgment of the overlap of the field of view of two cameras in the prior art.

[0082] The present invention uses image recognition to achieve high-precision positioning, detects the pixel coordinates of the calibration target in each camera image, and converts them into spatial coordinates according to the camera calibration results to achieve high-precision camera positioning. This overcomes the shortcomings of the existing technology that the operation is difficult and the repeatability is difficult to ensure.

[0083] The present invention uses an automated algorithm to achieve precise calibration. According to the coordinates of the calibration target in space and the spatial coordinates calculated by the recognition results of the two cameras, the spatial coordinates of the coincidence of the centers of the two cameras are automatically solved to establish the coordinate relationship. This avoids the errors caused by manual calibration and greatly improves the accuracy.

[0084] The present invention extends to multi-camera collaboration, repeatedly performs calibration target recognition and spatial positioning, and finally establishes a high-precision coordinate conversion relationship between multiple cameras to achieve system-level visual positioning and control. This expands the limitation of the prior art that can only perform two-camera calibration.

[0085] The present invention has high precision, high automation and strong intelligence. The method realizes the automation, digitization and precise quantification of multi-camera positioning calibration, and meets the requirements of high-end equipment for high precision, high efficiency and intelligence of visual systems. This makes it surpass the limited effect of existing technologies.

[0086] In summary, the present invention uses calibration targets and image recognition technology to achieve automated algorithm calibration, and builds a new method for high-precision positioning and collaborative work of multiple cameras. This method avoids the problems of subjective judgment, difficulty in manual operation, and difficulty in ensuring positioning accuracy in the prior art, and achieves intelligent upgrading of the multi-camera vision system, which is its main different feature from the prior art.

[0087] The above is only for explaining the implementation mode of the present invention and is not intended to limit the present invention. For those skilled in the art, any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention without creative work should be included in the protection scope of the present invention.

Claims

1. A high-precision association method for multiple cameras of an insertion machine, characterized in that: The following steps are involved: S1: Calibrate the upper camera and the lower camera involved in the association respectively, and establish a conversion relationship from the camera coordinate system to the XY module coordinate system; S2: Use a high-precision calibration pin to identify and determine its pixel coordinates in the lower camera image, and then convert them into spatial coordinates P 下相机 ; S3: Mark the small hole on the XY module working plane and record its coordinates as P 打孔 ; S4: Move the upper camera to align with the pinhole mark, identify the pixel coordinates of the pinhole in the upper camera image, and convert them into spatial coordinates P 上相机 ; S5: Calculate the spatial coordinates P where the upper camera center coincides with the lower camera center 中心 , to achieve accurate conversion of the two camera coordinate systems.

2. According to claim 1, a method for high-precision association of multiple cameras of an insertion machine is characterized in that: The method also includes the steps of: repeatedly executing the above steps for different combinations of upper and lower cameras to establish a high-precision spatial coordination relationship between multiple camera combinations.

3. A method for high-precision association of multiple cameras of an insertion machine according to claim 1 or 2, characterized in that: Camera calibration is performed using a conventional calibration plate through feature point recognition and coordinate matching.

4. According to claim 1, a method for high-precision association of multiple cameras of an insertion machine is characterized in that: The pixel coordinates of the calibration pin in the lower camera image are obtained through the image recognition algorithm and converted into XY module coordinates.

5. The method for high-precision association of multiple cameras of an insertion machine according to claim 1 is characterized in that: The coordinates of the hole mark P 打孔 As a spatial positioning reference, it is used to calculate the spatial coordinate P where the center of the upper camera coincides with the center of the lower camera. 中心 .

6. A method for high-precision association of multiple cameras of an insertion machine according to claim 1, characterized in that: P 中心 The calculation formula is: 中心 =P 上相机 - (P 打孔 -P 下相机 ).

7. The method for high-precision association of multiple cameras of an inserter according to claim 1 is characterized in that: The method also includes the steps of: performing distortion correction on the camera calibration result to improve the accuracy of coordinate transformation.

8. The method for high-precision association of multiple cameras of an insertion machine according to claim 1, characterized in that: The method also includes the steps of: when establishing a multi-camera collaborative relationship, ensuring the accuracy of the relative positions and postures between the cameras.

9. A high-precision association system for multiple cameras of an insertion machine, characterized in that: include: The camera calibration module is used to calibrate the upper camera and the lower camera and establish the conversion relationship from the camera coordinate system to the XY module coordinate system; A calibration pin recognition module, used to recognize the calibration pin in the lower camera image and determine its pixel coordinates; A coordinate conversion module is used to convert the pixel coordinates of the calibration pin and the pinhole into spatial coordinates, and calculate the spatial coordinates where the center of the upper camera coincides with the center of the lower camera; The multi-camera collaboration module is used to establish a high-precision spatial collaboration relationship between multiple camera combinations.

10. The high-precision association system for multiple cameras of an insertion machine according to claim 9, characterized in that: It also includes a distortion correction module for performing distortion correction on the camera calibration result to improve the accuracy of coordinate transformation.

Images