Method and system for correcting laser emission direction adjustment

Abstract

The invention discloses a correction method and system for laser emission direction adjustment, and relates to the technical field of image processing, and the method comprises the steps: capturing laser spot real-time image data through a CCD camera, extracting a spot region through a U-Net model, and setting a standardized reference region through calculating the side length of a square frame; selecting light spots by using a square frame to establish a relative coordinate system, carrying out gridding processing and positioning the centers of the light spots, and finally converting the light spots into actual coordinates of a world coordinate system; acquiring reflected and transmitted light spot coordinates and light spot position data from the position sensitive detector and the optical fiber sensor, and calculating deviation by using structural projection to obtain a corrected cursor; and a processing plane network model is constructed in the laser correction server, a galvanometer rotation angle set is calculated, laser scanning is executed, and feedback and correction are performed according to the deviation between the actual light spot position and the theoretical positioning point. According to the invention, the light spot region is extracted through the U-Net model, high-precision light spot segmentation is realized under the complex background and noise interference, and the accuracy of light spot positioning is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a correction method and system for adjusting laser direction. Background Art

[0002] In recent years, with the widespread application of laser technology in industrial processing, medical treatment, scientific research and other fields, the accuracy and stability of laser direction adjustment have become key technologies; traditional laser direction adjustment methods mainly rely on mechanical adjustment and simple sensor feedback. Although these methods can achieve the positioning of laser spots to a certain extent, they still have many shortcomings in applications with high precision requirements; for example, mechanical adjustment methods are easily affected by environmental factors, and the adjustment process is cumbersome; simple sensor feedback methods are difficult to capture changes in the spot position in real time and accurately, resulting in low positioning accuracy.

[0003] In the prior art, although some systems use high-speed CCD cameras for image acquisition and combine them with basic image processing algorithms to extract the light spot area, these methods still perform poorly in complex backgrounds and cannot provide sufficient robustness and accuracy; in addition, the existing coordinate system establishment and transformation methods are usually based on simple geometric relationships and lack a detailed analysis of the light spot shape and distribution, which affects the final positioning accuracy; these shortcomings are particularly obvious in applications that require high-precision positioning. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a correction method and system for laser direction adjustment to solve the problem of light spot positioning accuracy.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, an embodiment of the present invention provides a correction method for laser direction adjustment, which includes installing a laser, installing a position-sensitive detector through a pentaprism, receiving reflected and transmitted light beams, and selecting a high-speed CCD camera; equipping a laser correction server, connecting the CCD camera to the laser correction server through a cable, and collecting data through an optical fiber sensor; capturing real-time image data of the laser spot through the CCD camera, extracting the spot area using a U-Net model, and setting a standardized reference area by calculating the side length of a square box; using a square box to select the spot to establish a relative coordinate system, gridding and locating the center of the spot, and finally converting it into actual coordinates of the world coordinate system; obtaining the reflected and transmitted spot coordinates and spot position data from the position-sensitive detector and the optical fiber sensor, and using structural projection to calculate the deviation to obtain a correction cursor; constructing a processing plane network model in the laser correction server, calculating the galvanometer angle set, performing laser scanning, and providing feedback and correction based on the deviation between the actual spot position and the theoretical positioning point.

[0008] As a preferred solution of the correction method for adjusting the laser direction of the present invention, the laser is installed, a position sensitive detector is installed through a pentaprism, the reflected and transmitted light beams are received, and a high-speed CCD camera is selected. The specific steps are:

[0009] Select a stable continuous wave laser and purchase a pentaprism;

[0010] Fix the laser on the optical platform and adjust the height and angle so that the horizontal beam emitted by the laser points to the pentaprism;

[0011] A pentaprism is placed in the path of the beam emitted by the laser to split the laser beam into a reflected beam and a transmitted beam;

[0012] A PSD1 position sensitive detector 1 is installed on one side of the pentaprism to receive the laser beam reflected by the pentaprism;

[0013] A PSD2 position sensitive detector 2 is installed in the direction of the light beam directly transmitted through the pentaprism to receive and measure the laser beam transmitted through the pentaprism.

[0014] As a preferred solution of the correction method for adjusting the laser direction of the present invention, wherein: the laser correction server is equipped, a CCD camera is connected to the laser correction server via a cable, and data is collected via an optical fiber sensor;

[0015] Select Basler acA2040-180kmNIR as the CCD camera;

[0016] Aim the CCD camera at the SLM printer and the target processing plane to capture the real-time image of the laser spot;

[0017] Adjust the height and angle of the CCD camera;

[0018] Use the Thorlabs PT3 / M adjustment bracket to fix the CCD camera on the optical table with screws;

[0019] Select a computer equipped with an Intel i7 processor, 16GB RAM and a high-speed SSD as the laser correction server;

[0020] Install Windows 10 on the laser correction server;

[0021] Install the Basler pylon SDK driver for CCD cameras;

[0022] Install OpenCV image processing software and write control software;

[0023] Configure the exposure time and gain parameters of the CCD camera;

[0024] Write image processing algorithms and data processing algorithms to calculate the light spot position deviation;

[0025] Connect the SLM printer to the laser correction server via an Ethernet cable;

[0026] Connect the CCD camera to the laser correction server using a Thorlabs PGC102 cable;

[0027] Configure the fiber optic sensor, write the data acquisition software, collect the fiber optic data and upload it to the laser correction server via USB.

[0028] As a preferred solution of the correction method for adjusting the laser direction of the present invention, the real-time image data of the laser spot is captured by a CCD camera, the spot area is extracted using a U-Net model, and the standardized reference area is set by calculating the side length of the square box. The specific steps are:

[0029] The CCD camera captures the real-time image data of the laser spot on the processing plane and transmits the data to the laser correction server;

[0030] Extract pixels from real-time image data of laser spot;

[0031] Load the pre-trained U-Net model on the laser correction server;

[0032] Make sure the model input size is consistent with the CCD camera output image size;

[0033] Input the real-time image data of the laser spot into the U-Net model;

[0034] The U-Net model outputs a binary mask;

[0035] In the binary mask, white represents the spot area and black represents the background;

[0036] Record the pixel position and identify the spot area boundary on the binary mask output by the U-Net model using the Canny edge detection algorithm;

[0037] According to the preset laser spot diameter, the side length of the square box is calculated, and the expression is:

[0038]

[0039] Among them, L is the side length of the square, n is the safety factor, and φ is the laser spot diameter.

[0040] As a preferred solution of the correction method for adjusting the laser direction of the present invention, the steps of using a square frame to select the light spot to establish a relative coordinate system, gridding and locating the center of the light spot, and finally converting it into the actual coordinates of the world coordinate system are as follows:

[0041] Move the square frame to frame the boundary of the spot area and establish a relative coordinate system where the square is located;

[0042] Find the center position of the spot area, and draw a square box obtained by calculating the side length of the square box with the center position as the center;

[0043] Draw the lower left corner of the square box as the origin of the new relative coordinate system;

[0044] Using the side length of the square box as the interval distance, divide the square box into multiple small cells, establish a grid GA, and number all the unit grids in the grid GA;

[0045] Calculate the relative coordinate system of the center of each unit grid, and correspond the relative coordinates to the numbers of the unit grids to form a mapping list;

[0046] Traverse all unit grids, check each unit grid, and mark the grid containing the spot pixel as the spot unit grid;

[0047] In the direction of the horizontal axis of the relative coordinate system, the number of unit grids containing the spot pixels is counted to obtain the horizontal coordinate;

[0048] In the ordinate axis direction of the relative coordinate system, the number of unit grids containing the spot pixels is counted to obtain the ordinate;

[0049] Combine the obtained horizontal and vertical coordinates to obtain the coordinates of the center of the light spot in the relative coordinate system;

[0050] Use the established coordinate mapping list to convert the coordinates of the center of the light spot in the relative coordinate system into the actual coordinates of the center of the light spot in the world coordinate system.

[0051] As a preferred solution of the correction method for adjusting the laser direction of the present invention, the coordinates of the reflected and transmitted light spots and the light spot position data are obtained from the position sensitive detector and the optical fiber sensor, and the correction cursor is obtained by calculating the deviation using the structural projection. The specific steps are as follows:

[0052] Obtain the spot coordinates of the reflected light beam from the PSD1 position sensitive detector 1;

[0053] Obtain the spot coordinates of the projected light beam from the PSD2 position sensitive detector 2;

[0054] Read the data of the optical fiber sensor and perform preprocessing to obtain the light spot position data of the optical fiber sensor;

[0055] An infrared structured light projector is installed near the laser;

[0056] The CCD camera records the infrared structured light image, and the Fourier transform algorithm is used to extract the deformation information of the infrared structured light image;

[0057] According to the pattern deformation information, the precise position of the light spot is calculated, and the expression is:

[0058]

[0059] Among them, x s Represents the coordinate of the spot position on the horizontal axis, y s represents the light spot position coordinate of the ordinate, λ is the wavelength value of the light, I0 represents the image intensity value taken at a 0° phase offset, I1 represents the image intensity value taken at a 90° phase offset, I2 represents the image intensity value taken at a 180° phase offset, and I3 represents the image intensity value taken at a 270° phase offset;

[0060] Fusion of data from PSD1, PSD2, fiber optic sensors and structured light projection;

[0061] Based on the fused data, the half-width is calculated using the expression:

[0062]

[0063] Where ω is the half-angle width of the beam at the focus of the lens, d is the diameter of the laser spot, and f is the focal length of the lens;

[0064] Use geometric relations to transform the coordinates on PSD1 and PSD2 into coordinates in the world coordinate system relative to the laser. The expression is:

[0065] x″=x3+(x5-x3)·cos(ω);

[0066] y″=y3+(y5-y3)·sin(ω);

[0067] Wherein, x″ is the coordinate of the corrected light spot on the X-axis, y″ is the coordinate of the corrected light spot on the Y-axis, x3 is the X-coordinate of the reflected light beam measured by PSD1 position sensitive detector 1, y3 is the Y-coordinate of the reflected light beam measured by PSD1 position sensitive detector 1, x5 is the X-coordinate of the transmitted light beam measured by PSD2 position sensitive detector, y5 is the Y-coordinate of the transmitted light beam measured by PSD2 position sensitive detector, cos(ω) is the cosine value of the angle ω, and sin(ω) is the sine value of the angle ω;

[0068] After calculation, the coordinates after space angle correction are obtained.

[0069] As a preferred solution of the correction method for adjusting the laser direction of the present invention, the processing plane network model is constructed in the laser correction server, the galvanometer angle set is calculated, the laser scanning is performed, and feedback and correction are performed according to the deviation between the actual spot position and the theoretical positioning point. The specific steps are as follows:

[0070] Constructing a mesh model of the processing plane in the laser correction server;

[0071] The grid nodes of each unit grid are used as theoretical positioning points to form a two-dimensional coordinate system and define the ideal laser spot position;

[0072] Based on the constructed processing plane network model, the galvanometer rotation angle corresponding to each grid node is calculated to obtain the galvanometer rotation angle set, which is expressed as:

[0073]

[0074] Among them, α i is the galvanometer rotation angle along the X-axis, α j is the galvanometer rotation angle along the Y axis, i is the index of the grid node in the X direction, j is the index of the grid node in the Y direction, and d x is the spacing of the grid nodes in the X direction, d y is the spacing of the grid nodes in the Y direction, and f is the focal length of the lens;

[0075] The calculated galvanometer angle set is sent to the SLM printer to perform laser scanning;

[0076] The SLM printer adjusts the position of the laser beam according to the galvanometer angle set;

[0077] The CCD camera captures the actual light spot position and compares it with the theoretical positioning point;

[0078] Record the light spot position deviation and generate correction points;

[0079] According to the recorded light spot position deviation, the galvanometer angle set is recalculated to further correct the light spot position.

[0080] In a second aspect, the present invention provides a correction system for adjusting the laser direction, including a hardware installation module, a connection module, an image acquisition module, a coordinate conversion module, a deviation calculation module and a feedback correction module; the hardware installation module is used to install a laser, install a position-sensitive detector through a pentaprism, receive reflected and transmitted light beams, and select a high-speed CCD camera; the connection module is used to equip a laser correction server, connect the CCD camera to the laser correction server through a cable, and collect data through an optical fiber sensor; the image acquisition module is used to capture real-time image data of the laser spot through a CCD camera, and use a U-Net model to extract light The coordinate conversion module is used to select the light spot by the square box to establish a relative coordinate system, grid the light spot and locate the center of the light spot, and finally convert it into the actual coordinate of the world coordinate system; the deviation calculation module is used to obtain the reflected and transmitted light spot coordinates and the light spot position data from the position sensitive detector and the optical fiber sensor, and calculate the deviation by structural projection to obtain the correction cursor; the feedback correction module is used to build a processing plane network model in the laser correction server, calculate the galvanometer angle set, perform laser scanning, and provide feedback and correction according to the deviation between the actual light spot position and the theoretical positioning point.

[0081] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the correction method for laser direction adjustment as described in the first aspect of the present invention is implemented.

[0082] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the method for correcting the laser direction adjustment as described in the first aspect of the present invention is implemented.

[0083] The beneficial effects of the present invention are as follows: by extracting the light spot area through the U-Net model, the system can achieve high-precision light spot segmentation under complex background and noise interference; combining with Canny edge detection further improves the accuracy of boundary recognition; setting a standardized reference area unifies the processing standard and simplifies the subsequent coordinate system establishment and conversion process, thereby significantly improving the accuracy and robustness of light spot positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0085] Figure 1 This is a flow chart of the correction method for adjusting the laser direction in Example 1.

[0086] Figure 2 Flow chart for setting a standardized reference area for capturing real-time images and extracting spot areas in Example 1. DETAILED DESCRIPTION

[0087] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0088] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0089] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0090] Example 1, reference Figure 1 and Figure 2, which is the first embodiment of the present invention, and provides a correction method for adjusting the laser direction, comprising the following steps: installing a laser, installing a position-sensitive detector through a pentaprism, receiving reflected and transmitted light beams, and selecting a high-speed CCD camera; equipping a laser correction server, connecting the CCD camera to the laser correction server through a cable, and collecting data through an optical fiber sensor; capturing real-time image data of the laser spot through the CCD camera, extracting the spot area using the U-Net model, and setting a standardized reference area by calculating the side length of a square box; selecting the spot using a square box to establish a relative coordinate system, gridding and locating the center of the spot, and finally converting it into the actual coordinates of the world coordinate system; obtaining the reflected and transmitted spot coordinates and the spot position data from the position-sensitive detector and the optical fiber sensor, and obtaining a correction cursor by calculating the deviation using structural projection; constructing a processing plane network model in the laser correction server, calculating the galvanometer angle set, performing laser scanning, and providing feedback and correction based on the deviation between the actual spot position and the theoretical positioning point.

[0091] S1. Install the laser, install the position sensitive detector through the pentaprism, receive the reflected and transmitted light beams, select the high-speed CCD camera, equip the laser correction server, and connect the CCD camera to the laser correction server through a cable.

[0092] Going a step further, choose a stable continuous wave laser and purchase a pentaprism;

[0093] Fix the laser on the optical platform and adjust the height and angle so that the horizontal beam emitted by the laser points to the pentaprism;

[0094] A pentaprism is placed in the path of a beam emitted by a laser to split the laser beam into a reflected beam and a transmitted beam;

[0095] A PSD1 position sensitive detector 1 is installed on one side of the pentaprism to receive the laser beam reflected by the pentaprism;

[0096] A PSD2 position sensitive detector 2 is installed in the direction of the light beam directly transmitted through the pentaprism to receive and measure the laser beam transmitted through the pentaprism.

[0097] It should be noted that the pentaprism is used as a beam splitter to split the incident laser beam into two parts, one part is reflected to PSD1, and the other part is transmitted to PSD2; this helps to obtain information from multiple light paths at the same time and improve measurement accuracy;

[0098] It should be noted that a high-precision laser beam detection system is constructed by installing a laser, a pentaprism and a position-sensitive detector; its function is to accurately measure the position of the reflected and transmitted light beams, providing an accurate data basis for subsequent spot positioning and error correction.

[0099] S2, equipped with a laser correction server, connects the CCD camera to the laser correction server via a cable, and collects data through an optical fiber sensor.

[0100] Going a step further, the Basler acA2040-180kmNIR was selected as the CCD camera;

[0101] Aim the CCD camera at the SLM printer and the target processing plane to capture the real-time image of the laser spot;

[0102] Adjust the height and angle of the CCD camera so that its field of view covers the entire processing area;

[0103] Use the Thorlabs PT3 / M adjustment bracket to fix the CCD camera on the optical table with screws;

[0104] Select a computer equipped with an Intel i7 processor, 16GB RAM and a high-speed SSD as the laser correction server;

[0105] Install Windows 10 on the laser correction server;

[0106] Install the Basler pylon SDK driver for CCD cameras;

[0107] Install OpenCV image processing software and write control software;

[0108] Configure the exposure time and gain parameters of the CCD camera to obtain the best image quality;

[0109] Write image processing algorithms and data processing algorithms to calculate the light spot position deviation;

[0110] Connect the SLM printer to the laser correction server via an Ethernet cable;

[0111] Connect the CCD camera to the laser correction server using a Thorlabs PGC102 cable;

[0112] Configure the fiber optic sensor, write the data acquisition software, collect the fiber optic data and upload it to the laser correction server via USB.

[0113] It should be noted that the SLM printer adjusts the position of the internal galvanometer according to the received galvanometer angle command, thereby changing the direction of the laser beam so that it scans on the processing plane according to the expected trajectory; SLM printers usually include components such as a microprocessor, a galvanometer control system, and a laser, which can quickly respond to commands sent by the server and perform complex graphics printing tasks.

[0114] It should be noted that the Basler acA2040-180kmNIRzuow was chosen as the CCD camera because it has high resolution (2048x1088 pixels), high frame rate (up to 180fps) and good near-infrared sensitivity, which is suitable for capturing rapidly changing and low-contrast laser spot images.

[0115] It should be noted that the laser correction server, as the control center of the entire system, is responsible for receiving image data from the CCD camera, processing this data to calculate the position of the laser spot, generating the corresponding galvanometer angle set based on the calculation results, and sending it to the SLM printer for execution; in addition, it is also responsible for storing and analyzing historical data in order to continuously optimize system performance.

[0116] It should be noted that the method for adjusting the height and angle of the laser is specifically implemented as follows:

[0117] First, place the laser on an adjustable height stand. Then, turn on the laser and observe whether the output beam is level. If the beam is tilted, adjust the height of the stand or the leveling feet of the base until the beam is completely level.

[0118] Use two perpendicular mirrors to assist with alignment; place one mirror in front of the laser and the other near the pentaprism; adjust the angle of the laser until the light spots reflected from the two mirrors overlap; this indicates that the laser is correctly aligned with the pentaprism;

[0119] The pentaprism should be perpendicular to the incident light beam. Its correct position can be judged by observing whether the direct light spot passing through the pentaprism is evenly distributed. If necessary, the position of the pentaprism can be fine-tuned using a precision adjustment stand.

[0120] It should be noted that an efficient data acquisition and processing system is built to achieve high-precision laser spot detection and correction; by selecting a high-performance camera and fixing it on the optical platform, it is ensured that the real-time image of the laser spot on the processing plane can be captured; the laser correction server is equipped with powerful computing resources, runs the Windows 10 operating system, and has the necessary drivers and image processing software installed; the optimal exposure time and gain parameters are configured to obtain high-quality image data; in addition, the SLM printer is connected via an Ethernet cable, and additional data is collected using a fiber optic sensor to ensure the comprehensiveness and accuracy of the system. Ultimately, these data are used to calculate the spot position deviation, providing accurate information for subsequent laser correction.

[0121] S3, capture the real-time image data of the laser spot through the CCD camera, use the U-Net model to extract the spot area, and set the standardized reference area by calculating the side length of the square box.

[0122] S301, the CCD camera captures the real-time image data of the laser spot on the processing plane, and transmits the data to the laser correction server;

[0123] S302, extracting pixels from the real-time image data of the laser spot;

[0124] S303, loading a pre-trained U-Net model on a laser correction server;

[0125] Make sure the model input size is consistent with the CCD camera output image size;

[0126] Input the real-time image data of the laser spot into the U-Net model;

[0127] S304, the U-Net model outputs a binary mask;

[0128] In the binary mask, white represents the spot area and black represents the background;

[0129] S305, recording the pixel position, and identifying the spot area boundary on the binary mask output by the U-Net model through the Canny edge detection algorithm;

[0130] S306. Calculate the side length of the square frame according to the preset laser spot diameter. The expression is:

[0131]

[0132] Wherein, L is the side length of the square, n is the safety factor, which is 5, and φ is the laser spot diameter.

[0133] It should be noted that the specific parameter settings and result processing of Canny edge detection are as follows:

[0134] Set the high threshold parameter to a certain percentage of the maximum gradient value of the image, such as 70%;

[0135] Set the low threshold parameter to half of the high threshold;

[0136] Use a 3x3 or 5x5 kernel size for preprocessing to reduce the impact of noise;

[0137] The output of the Canny algorithm is a binary image with white lines representing detected edges; these edges can be further refined, for example by using connected domain analysis to identify individual object outlines, or by combining other morphological operations to clean up the edge map.

[0138] It should be noted that the calculation of the side length of the square box For example,

[0139] Set the laser spot diameter φ = 0.5 mm and the safety factor n = 5;

[0140] Substituting the known parameters into the formula, we get:

[0141]

[0142] L = 0.5 mm + 0.1 mm = 0.6 mm;

[0143] The side length of the square frame is 0.6 mm.

[0144] It should be noted that the real-time image of the laser spot is captured by a CCD camera, and the U-Net model is used to extract the spot area; the image data is transmitted to the laser correction server, the pre-trained U-Net model is loaded, the input size is ensured to be consistent, and a binary mask is generated to distinguish the spot from the background; the Canny edge detection algorithm is used to identify the spot boundary, and the side length of the square box is calculated to set the standardized reference area; a series of operations ensure the accurate extraction of the spot position and the effective removal of noise, providing a reliable data basis for subsequent spatial angle correction and error correction.

[0145] S4. Use a square frame to select the light spot to establish a relative coordinate system, grid it and locate the center of the light spot, and finally convert it into the actual coordinates of the world coordinate system.

[0146] Furthermore, the square frame is moved to frame the boundary of the spot area and establish a relative coordinate system where the square is located;

[0147] Find the center position of the spot area, and draw a square box obtained by calculating the side length of the square box with the center position as the center;

[0148] Draw the lower left corner of the square box as the origin of the new relative coordinate system;

[0149] Using the side length of the square box as the interval distance, divide the square box into multiple small cells, establish a grid GA, and number all the unit grids in the grid GA;

[0150] Calculate the relative coordinate system of the center of each unit grid, and correspond the relative coordinates to the numbers of the unit grids to form a mapping list;

[0151] Traverse all unit grids, check each unit grid, and mark the grid containing the spot pixel as the spot unit grid;

[0152] In the direction of the abscissa axis of the relative coordinate system, count the number of unit grids containing the spot pixels to obtain the abscissa x′;

[0153] In the ordinate axis direction of the relative coordinate system, count the number of unit grids containing the spot pixels to obtain the ordinate y′;

[0154] Combine the obtained horizontal and vertical coordinates to obtain the coordinates of the center of the light spot in the relative coordinate system;

[0155] Use the established coordinate mapping list to convert the coordinates of the center of the light spot in the relative coordinate system into the actual coordinates of the center of the light spot in the world coordinate system.

[0156] It should be noted that when the number of unit grids is an even number, the abscissa x′ takes the midpoint of the line connecting the centers of the two centered unit grids as the target point; when the number is an odd number, the abscissa x′ takes the center of the centered grid as the target point;

[0157] When the number of unit grids is an even number, the midpoint of the line connecting the centers of the two unit grids in the middle is taken as the target point of the ordinate y′; when the ordinate y′ is an odd number, the center of the unit grid in the middle is taken as the target point of the ordinate y′.

[0158] It should be noted that the square box with a side length of 0.6 mm was moved to the position of the laser spot in the image to ensure that it completely covered the spot;

[0159] Take the lower left corner of the square box as the origin of the new relative coordinate system;

[0160] The side length of the square box is 0.6 mm as the spacing distance, and the square box is divided into multiple small cells to establish the grid GA.

[0161] It should be noted that in this step, a square box is used to select the spot area, a new coordinate origin is set, and the box is divided into small cells with a side length of 0.6 mm to form a grid GA; the center coordinates of each unit grid are calculated, and the center position is found according to the spot pixel distribution; finally, the relative coordinates are converted into actual world coordinates using coordinate mapping. By establishing a relative coordinate system and gridding it, the center of the laser spot is accurately located and converted into the actual coordinates of the world coordinate system, which helps to accurately measure the spot position and provide data support for subsequent precise control.

[0162] S5. Obtain the reflected and transmitted light spot coordinates and light spot position data from the position sensitive detector and the optical fiber sensor, and calculate the deviation using the structural projection to obtain a correction cursor.

[0163] Furthermore, the spot coordinates of the reflected light beam are obtained from the PSD1 position sensitive detector 1;

[0164] Obtain the spot coordinates of the projected light beam from the PSD2 position sensitive detector 2;

[0165] Read the data of the optical fiber sensor and perform preprocessing to obtain the light spot position data of the optical fiber sensor;

[0166] An infrared structured light projector is installed near the laser;

[0167] The CCD camera records the infrared structured light image, and the Fourier transform algorithm is used to extract the deformation information of the infrared structured light image;

[0168] According to the pattern deformation information, the precise position of the light spot is calculated, and the expression is:

[0169]

[0170] Among them, x s Represents the coordinate of the spot position on the horizontal axis, y s represents the light spot position coordinate of the ordinate, λ is the wavelength value of the light, I0 represents the image intensity value taken at a 0° phase offset, I1 represents the image intensity value taken at a 90° phase offset, I2 represents the image intensity value taken at a 180° phase offset, and I3 represents the image intensity value taken at a 270° phase offset;

[0171] It should be noted that the data of PSD1, PSD2, fiber optic sensor and structured light projection are fused to obtain more accurate position measurement information, which is expressed as:

[0172] x'=w1·x3+w2·x5+w f ·xf+w s ·x s ;

[0173] y'=w1·y3+w2·y5+w f ·yf+w s ·y s ;

[0174] Among them, x' is the abscissa of the integrated spot position, y' is the ordinate of the integrated spot position, w1 is the PSD1 data weight, w2 is the PSD2 data weight, and w f is the fiber optic sensor data weight, w s is the structured light projection data weight, x3 is the X-axis coordinate of the reflected light beam detected by PSD1, y3 is the Y-axis coordinate of the reflected light beam detected by PSD1, x5 is the X-axis coordinate of the transmitted light beam detected by PSD2, and y5 is the Y-axis coordinate of the transmitted light beam detected by PSD2;

[0175] Based on the fused data, the half-width is calculated using the expression:

[0176]

[0177] Where ω is the half-angle width of the beam at the focus of the lens, d is the diameter of the laser spot, and f is the focal length of the lens;

[0178] Use geometric relations to transform the coordinates on PSD1 and PSD2 into coordinates in the world coordinate system relative to the laser. The expression is:

[0179] x″=x3+(x5-x3)·cos(ω);

[0180] y″=y3+(y5-y3)·sin(ω);

[0181] Wherein, x″ is the coordinate of the corrected light spot on the X-axis, y″ is the coordinate of the corrected light spot on the Y-axis, x3 is the X-coordinate of the reflected light beam measured by PSD1 position sensitive detector 1, y3 is the Y-coordinate of the reflected light beam measured by PSD1 position sensitive detector 1, x5 is the X-coordinate of the transmitted light beam measured by PSD2 position sensitive detector, y5 is the Y-coordinate of the transmitted light beam measured by PSD2 position sensitive detector, cos(ω) is the cosine value of the angle ω, and sin(ω) is the sine value of the angle ω;

[0182] After calculation, the coordinates after space angle correction are obtained.

[0183] It should be noted that structural projection is a three-dimensional imaging technology that projects known patterns of light (usually infrared or visible light) onto the surface of a target object, and then uses a camera to capture the deformation of these patterns on the surface of the object; by analyzing the deformed patterns, the three-dimensional geometric information of the object's surface can be calculated, including depth, shape and position.

[0184] It should be noted that in the optical system, ω represents the half-angle width of the light beam at the focus of the lens; it is a parameter to measure the degree of focusing of the laser beam after passing through the lens; specifically, when the laser beam passes through a lens with a focal length of f, the light beam will converge into a point at the focus of the lens; the beam angle ω during this convergence process can be used to describe the relationship between the spot size and the focal length.

[0185] It should be noted that the use of cos(ω) and sin(ω) is to correct the coordinate offset caused by the different beam propagation paths; in the dual PSD system, the reflected and transmitted beams will have different position offsets due to the different propagation path lengths; by calculating ω and applying trigonometric functions, this deviation caused by the optical path difference can be corrected.

[0186] The physical meaning is that by considering the propagation characteristics of the light beam on different paths (such as refraction), the true position of the light spot can be located more accurately; here, cos(ω) is used to adjust the position in the X-axis direction, while sin(ω) is used to adjust the position in the Y-axis direction, thereby achieving precise spatial angle correction.

[0187] It should be noted that the coordinates and position data of the reflected and transmitted light spots are obtained through position-sensitive detectors and fiber optic sensors, and high-precision positioning is performed using an infrared structured light projector. Specifically, the data of each sensor is read and preprocessed, and the deformation information of the structured light image is extracted using a Fourier transform algorithm to calculate the precise position of the light spot. More accurate position information is obtained through data fusion, the half-angle width ω is calculated, and the coordinates on PSD1 and PSD2 are converted into corrected coordinates in the world coordinate system using geometric relationships. A series of steps ensure high-precision measurement and correction of the light spot position and improve the overall positioning accuracy of the system.

[0188] S6. Construct a processing plane network model in the laser correction server, calculate the galvanometer angle set, perform laser scanning, and provide feedback and correction based on the deviation between the actual spot position and the theoretical positioning point.

[0189] Furthermore, a mesh model of the processing plane is constructed in the laser correction server;

[0190] The grid nodes of each unit grid are used as theoretical positioning points to form a two-dimensional coordinate system and define the ideal laser spot position;

[0191] Based on the constructed processing plane network model, the galvanometer rotation angle corresponding to each grid node is calculated to obtain the galvanometer rotation angle set, which is expressed as:

[0192]

[0193] Among them, α i is the galvanometer rotation angle along the X-axis, α j is the galvanometer rotation angle along the Y axis, i is the index of the grid node in the X direction, j is the index of the grid node in the Y direction, and d x is the spacing of the grid nodes in the X direction, d y is the spacing of the grid nodes in the Y direction, and f is the focal length of the lens;

[0194] The calculated galvanometer angle set is sent to the SLM printer to perform laser scanning;

[0195] The SLM printer adjusts the position of the laser beam according to the galvanometer angle set;

[0196] The CCD camera captures the actual light spot position and compares it with the theoretical positioning point;

[0197] Record the light spot position deviation and generate correction points;

[0198] According to the recorded light spot position deviation, the galvanometer angle set is recalculated to further correct the light spot position.

[0199] It should be noted that the network model of the processing plane is constructed. This model is composed of a series of grid nodes, and each node represents a theoretical positioning point;

[0200] The mesh nodes form a two-dimensional coordinate system that is used to define the ideal laser spot position.

[0201] It should be noted that the steps of laser scanning performed by the SLM printer are:

[0202] The laser correction server encapsulates the galvanometer angle set data into a TCP / IP data packet;

[0203] Use TCP / IP communication protocol for data transmission;

[0204] Send the data packet to the SLM printer via the Ethernet line;

[0205] The controller inside the SLM printer parses the data packets received from the laser correction server, extracts the galvanometer angle information, and converts the information into the corresponding motor drive signal;

[0206] According to the extracted galvanometer angle information, the motor control of the SLM printer adjusts the galvanometer to move to the specified angle;

[0207] When the galvanometer is adjusted into place, the laser emits a laser beam, which is reflected by the galvanometer onto the processing plane.

[0208] It should be noted that α i and α j are the galvanometer rotation angles along the X-axis and Y-axis directions, respectively. The formula describes the angle that the galvanometer needs to rotate at a given grid node (i, j) in order to accurately illuminate the laser beam at that point; where d x and d y is the grid node spacing, f is the focal length of the lens;

[0209] The derivation is based on the principle of geometric optics, considering the refraction of light emitted from the laser source after passing through the lens. By calculating the relationship between the incident angle and the exit angle of the light, the required galvanometer rotation angle can be obtained; in a physical sense, these formulas can help determine the angle that the galvanometer must rotate to ensure that the laser can accurately point to each theoretical positioning point on the processing plane.

[0210] It should be noted that the laser correction server sends data packets containing the galvanometer angle to the SLM printer via the Ethernet interface; these data packets are usually encapsulated using the TCP / IP communication protocol and may contain additional control information, such as serial numbers, timestamps, etc.

[0211] The SLM printer has a dedicated controller inside, which parses the data packets from the server, extracts the galvanometer angle information, and converts it into the corresponding motor drive signal; then, the motor drives the galvanometer to move to the specified angle, thereby changing the direction of the laser beam. During this process, the SLM printer also monitors the error between the actual angle and the instruction, and compensates in real time to ensure high-precision positioning.

[0212] It should be noted that by constructing a grid model of the processing plane, the galvanometer angle corresponding to each grid node is calculated to form a galvanometer angle set, which is sent to the SLM printer for laser scanning; the CCD camera captures the actual spot position and compares it with the theoretical positioning point, records the deviation and generates a correction point. The galvanometer angle is recalculated based on the deviation to further correct the spot position; this step ensures that the laser beam can accurately irradiate each theoretical positioning point on the processing plane, achieving high-precision laser positioning and processing.

[0213] This embodiment also provides a correction system for adjusting the laser direction, including: a hardware installation module, a connection module, an image acquisition module, a coordinate conversion module, a deviation calculation module and a feedback correction module; the hardware installation module is used to install the laser, install a position-sensitive detector through a pentaprism, receive reflected and transmitted light beams, and select a high-speed CCD camera; the connection module is used to equip a laser correction server, connect the CCD camera to the laser correction server through a cable, and collect data through an optical fiber sensor; the image acquisition module is used to capture real-time image data of the laser spot through a CCD camera, extract the spot area using a U-Net model, and set a standardized reference area by calculating the side length of a square box; the coordinate conversion module is used to use a square box to select the spot to establish a relative coordinate system, grid processing and locate the center of the spot, and finally convert it into the actual coordinates of the world coordinate system; the deviation calculation module is used to obtain the reflected and transmitted spot coordinates and spot position data from the position-sensitive detector and the optical fiber sensor, and use structural projection to calculate the deviation to obtain a correction cursor; the feedback correction module is used to construct a processing plane network model in the laser correction server, calculate the galvanometer angle set, perform laser scanning, and provide feedback and correction according to the deviation between the actual spot position and the theoretical positioning point.

[0214] This embodiment also provides a computer device, which is suitable for the correction method of laser direction adjustment, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the correction method of laser direction adjustment proposed in the above embodiment.

[0215] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0216] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the correction method for adjusting the laser direction proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination of them, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0217] In summary, the present invention extracts the light spot area through the U-Net model, and the system can achieve high-precision light spot segmentation under complex background and noise interference; combined with Canny edge detection, the accuracy of boundary recognition is further improved; setting a standardized reference area unifies the processing standard and simplifies the subsequent coordinate system establishment and conversion process, thereby significantly improving the accuracy and robustness of light spot positioning.

[0218] Example 2, referring to Table 1, is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of the correction method for laser direction adjustment are provided.

[0219] First, the system was built and the hardware was installed. A stable continuous wave laser was selected and fixed on the optical platform. The height and angle were adjusted to ensure that the horizontal beam emitted by the laser pointed to the pentaprism. The pentaprism was placed in the beam path emitted by the laser to split the laser beam into a reflected beam and a transmitted beam. A PSD1 position sensitive detector 1 was installed on one side of the pentaprism to receive the laser beam reflected by the pentaprism. A PSD2 position sensitive detector 2 was installed in the direction of the beam directly through the pentaprism to receive and measure the laser beam transmitted through the pentaprism. A Baslerac A2040-180 km NIR was selected as the CCD camera and it was facing the SLM printer and the target processing plane. Its height and angle were adjusted so that its field of view covered the entire processing area. It was fixed on the optical platform using a Thorlabs PT3 / M adjustment stand. A laser correction server was configured, the necessary software and drivers were installed, and the CCD camera was connected to the laser correction server via a cable.

[0220] Secondly, after starting the system, the CCD camera captures the laser spot image on the processing plane in real time and transmits the data to the laser correction server. On the server, the pre-trained U-Net model is loaded to ensure that the model input size is consistent with the CCD camera output image size; the real-time image data of the laser spot is input into the U-Net model to generate a binary mask to distinguish the spot from the background; the Canny edge detection algorithm is used to identify the spot boundary, and the side length of the square box is calculated to set the standardized reference area.

[0221] Next, move the square box to select the spot area, establish a relative coordinate system, and divide the box into multiple small cells to form a grid GA; calculate the relative coordinates of the center of each unit grid, and find the center position according to the pixel distribution of the spot; finally, use coordinate mapping to convert the relative coordinates into actual world coordinates; obtain the coordinates of the reflected and transmitted spots from PSD1 and PSD2, and read the data of the fiber optic sensor; install an infrared structured light projector near the laser, record the infrared structured light image, use the Fourier transform algorithm to extract deformation information, and calculate the precise position of the spot; fuse the multi-sensor data to obtain more accurate position measurement information. Based on the fused data, calculate the half-angle width ω, and convert the coordinates on PSD1 and PSD2 into coordinates in the world coordinate system relative to the laser to obtain the coordinates after spatial angle correction.

[0222] Finally, a grid model of the processing plane is constructed in the laser correction server, and the galvanometer angle corresponding to each grid node is calculated to form a galvanometer angle set, which is sent to the SLM printer for laser scanning; the CCD camera captures the actual spot position and compares it with the theoretical positioning point, records the deviation and generates a correction point. The galvanometer angle is recalculated based on the deviation, and the spot position is further corrected; through this series of steps, high-precision laser direction adjustment and correction are achieved, ensuring that the laser spot can accurately irradiate every theoretical positioning point on the processing plane.

[0223] The details are shown in Table 1 below:

[0224] Table 1 Performance comparison of laser direction adjustment methods

[0225]

[0226] It can be seen from the above experimental data that the present invention significantly improves the light spot positioning accuracy and detection robustness by using the U-Net model and the Canny edge detection algorithm; secondly, the system response time is greatly shortened, and the real-time processing capability is improved; then, multi-sensor data fusion and geometric relationship conversion further enhance the stability and accuracy of the light spot position; finally, the experimental results show that the present invention is significantly superior to the prior art in terms of light spot positioning accuracy, robustness, response speed and stability.

[0227] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A correction method for laser direction adjustment, characterized in that: include, Install the laser, install the position sensitive detector through the pentaprism, receive the reflected and transmitted beams, and select a high-speed CCD camera; Equipped with a laser correction server, the CCD camera is connected to the laser correction server through a cable, and the data is collected through a fiber optic sensor; The real-time image data of the laser spot is captured by a CCD camera, the spot area is extracted using the U-Net model, and the standardized reference area is set by calculating the side length of the square box; Use a square frame to select the light spot to establish a relative coordinate system, grid it and locate the center of the light spot, and finally convert it into the actual coordinates of the world coordinate system; Obtain the reflected and transmitted light spot coordinates and light spot position data from the position sensitive detector and the optical fiber sensor, and calculate the deviation using the structural projection to obtain the correction cursor; A processing plane network model is constructed in the laser correction server, the galvanometer angle set is calculated, laser scanning is performed, and feedback and correction are performed based on the deviation between the actual spot position and the theoretical positioning point.

2. The method for correcting the laser direction adjustment according to claim 1, characterized in that: The laser is installed, a position sensitive detector is installed through a pentaprism, reflected and transmitted light beams are received, and a high-speed CCD camera is selected. The specific steps are: Select a stable continuous wave laser and purchase a pentaprism; Fix the laser on the optical platform and adjust the height and angle so that the horizontal beam emitted by the laser points to the pentaprism; A pentaprism is placed in the path of the beam emitted by the laser to split the laser beam into a reflected beam and a transmitted beam; A PSD1 position sensitive detector 1 is installed on one side of the pentaprism to receive the laser beam reflected by the pentaprism; A PSD2 position sensitive detector 2 is installed in the direction of the light beam directly transmitted through the pentaprism to receive and measure the laser beam transmitted through the pentaprism.

3. The method for correcting laser direction adjustment according to claim 2, characterized in that: The laser correction server is equipped, the CCD camera is connected to the laser correction server via a cable, and data is collected via an optical fiber sensor; Select Basler acA2040-180kmNIR as the CCD camera; Aim the CCD camera at the SLM printer and the target processing plane to capture the real-time image of the laser spot; Adjust the height and angle of the CCD camera; Use the Thorlabs PT3 / M adjustment bracket to fix the CCD camera on the optical table with screws; Select a computer equipped with an Intel i7 processor, 16 GB RAM and a high-speed SSD as the laser correction server; Install Windows 10 on the laser correction server; Install the Basler pylon SDK driver for CCD cameras; Install OpenCV image processing software and write control software; Configure the exposure time and gain parameters of the CCD camera; Write image processing algorithms and data processing algorithms to calculate the spot position deviation; Connect the SLM printer to the laser correction server via an Ethernet cable; Connect the CCD camera to the laser correction server using a Thorlabs PGC102 cable; Configure the fiber optic sensor, write the data acquisition software, collect the fiber optic data and upload it to the laser correction server via USB.

4. The method for correcting laser direction adjustment according to claim 3, characterized in that: The laser spot real-time image data is captured by a CCD camera, the spot area is extracted using a U-Net model, and a standardized reference area is set by calculating the side length of a square frame. The specific steps are: The CCD camera captures the real-time image data of the laser spot on the processing plane and transmits the data to the laser correction server; Extract pixels from real-time image data of laser spot; Load the pre-trained U-Net model on the laser correction server; Make sure the model input size is consistent with the CCD camera output image size; Input the real-time image data of the laser spot into the U-Net model; The U-Net model outputs a binary mask; In the binary mask, white represents the spot area and black represents the background; Record the pixel position and identify the spot area boundary on the binary mask output by the U-Net model using the Canny edge detection algorithm; According to the preset laser spot diameter, the side length of the square box is calculated, and the expression is: Among them, L is the side length of the square, n is the safety factor, and φ is the laser spot diameter.

5. The method for correcting laser direction adjustment according to claim 4, characterized in that: The method uses a square frame to select the light spot to establish a relative coordinate system, performs grid processing and locates the center of the light spot, and finally converts it into the actual coordinates of the world coordinate system. The specific steps are: Move the square frame to frame the boundary of the spot area and establish a relative coordinate system where the square is located; Find the center position of the spot area, and draw a square box obtained by calculating the side length of the square box with the center position as the center; Draw the lower left corner of the square box as the origin of the new relative coordinate system; Using the side length of the square box as the interval distance, divide the square box into multiple small cells, establish a grid GA, and number all the unit grids in the grid GA; Calculate the relative coordinate system of the center of each unit grid, and correspond the relative coordinates to the numbers of the unit grids to form a mapping list; Traverse all unit grids, check each unit grid, and mark the grid containing the spot pixel as the spot unit grid; In the direction of the horizontal axis of the relative coordinate system, the number of unit grids containing the spot pixels is counted to obtain the horizontal coordinate; In the ordinate axis direction of the relative coordinate system, the number of unit grids containing the spot pixels is counted to obtain the ordinate; Combine the obtained horizontal and vertical coordinates to obtain the coordinates of the center of the light spot in the relative coordinate system; Use the established coordinate mapping list to convert the coordinates of the center of the light spot in the relative coordinate system into the actual coordinates of the center of the light spot in the world coordinate system.

6. The method for correcting the laser direction adjustment according to claim 5, characterized in that: The steps of obtaining the reflected and transmitted light spot coordinates and light spot position data from the position sensitive detector and the optical fiber sensor, and obtaining the correction cursor by calculating the deviation using the structural projection are as follows: Obtain the spot coordinates of the reflected light beam from the PSD1 position sensitive detector 1; Obtain the spot coordinates of the projected light beam from the PSD2 position sensitive detector 2; Read the data of the optical fiber sensor and perform preprocessing to obtain the light spot position data of the optical fiber sensor; An infrared structured light projector is installed near the laser; The CCD camera records the infrared structured light image, and the Fourier transform algorithm is used to extract the deformation information of the infrared structured light image; According to the pattern deformation information, the precise position of the light spot is calculated, and the expression is: Among them, x s Represents the coordinate of the spot position on the horizontal axis, y s represents the light spot position coordinate of the ordinate, λ is the wavelength value of the light, I0 represents the image intensity value taken at a 0° phase offset, I1 represents the image intensity value taken at a 90° phase offset, I2 represents the image intensity value taken at a 180° phase offset, and I3 represents the image intensity value taken at a 270° phase offset; Fusion of data from PSD1, PSD2, fiber optic sensors and structured light projection; Based on the fused data, the half-width is calculated using the expression: Where ω is the half-angle width of the beam at the focus of the lens, d is the diameter of the laser spot, and f is the focal length of the lens; Use geometric relations to transform the coordinates on PSD1 and PSD2 into coordinates in the world coordinate system relative to the laser. The expression is: x″=x3+(x5-x3)·cos(ω); y″=y3+(y5-y3)·sin(ω); Wherein, x″ is the coordinate of the corrected light spot on the X-axis, y″ is the coordinate of the corrected light spot on the Y-axis, x3 is the X-coordinate of the reflected light beam measured by PSD1 position sensitive detector 1, y3 is the Y-coordinate of the reflected light beam measured by PSD1 position sensitive detector 1, x5 is the X-coordinate of the transmitted light beam measured by PSD2 position sensitive detector, y5 is the Y-coordinate of the transmitted light beam measured by PSD2 position sensitive detector, cos(ω) is the cosine value of the angle ω, and sin(ω) is the sine value of the angle ω; After calculation, the coordinates after space angle correction are obtained.

7. The method for correcting laser direction adjustment according to claim 6, characterized in that: The processing plane network model is constructed in the laser correction server, the galvanometer angle set is calculated, laser scanning is performed, and feedback and correction are performed according to the deviation between the actual spot position and the theoretical positioning point. The specific steps are: Constructing a mesh model of the processing plane in the laser correction server; The grid nodes of each unit grid are used as theoretical positioning points to form a two-dimensional coordinate system and define the ideal laser spot position; Based on the constructed processing plane network model, the galvanometer rotation angle corresponding to each grid node is calculated to obtain the galvanometer rotation angle set, which is expressed as: Among them, α i is the galvanometer rotation angle along the X-axis, α j is the galvanometer rotation angle along the Y axis, i is the index of the grid node in the X direction, j is the index of the grid node in the Y direction, and d x is the spacing of the grid nodes in the X direction, d y is the spacing of the grid nodes in the Y direction, and f is the focal length of the lens; The calculated galvanometer angle set is sent to the SLM printer to perform laser scanning; The SLM printer adjusts the position of the laser beam according to the galvanometer angle set; The CCD camera captures the actual light spot position and compares it with the theoretical positioning point; Record the light spot position deviation and generate correction points; According to the recorded light spot position deviation, the galvanometer angle set is recalculated to further correct the light spot position.

8. A laser direction adjustment correction system, based on the laser direction adjustment correction method according to any one of claims 1 to 7, characterized in that: Including hardware installation module, connection module, image acquisition module, coordinate conversion module, deviation calculation module and feedback correction module; The hardware installation module is used to install the laser, install the position sensitive detector through the pentaprism, receive the reflected and transmitted light beams, and select the high-speed CCD camera; The connection module is used to equip the laser correction server, connect the CCD camera to the laser correction server via a cable, and collect data via an optical fiber sensor; The image acquisition module is used to capture the real-time image data of the laser spot through a CCD camera, extract the spot area using a U-Net model, and set a standardized reference area by calculating the side length of a square frame; The coordinate conversion module is used to select the light spot by a square frame to establish a relative coordinate system, perform grid processing and locate the center of the light spot, and finally convert it into the actual coordinate of the world coordinate system; The deviation calculation module is used to obtain the reflected and transmitted light spot coordinates and light spot position data from the position sensitive detector and the optical fiber sensor, and calculate the deviation using the structural projection to obtain the correction cursor; The feedback correction module is used to construct a processing plane network model in the laser correction server, calculate the galvanometer angle set, perform laser scanning, and provide feedback and correction based on the deviation between the actual spot position and the theoretical positioning point.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the processor implements the steps of the laser direction adjustment correction method described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for correcting the laser direction adjustment according to any one of claims 1 to 7 are implemented.

Images