3D laser polishing equipment based on structured light reverse design and intelligent detection method

Abstract

The invention discloses 3D laser polishing equipment based on structured light reverse design and an intelligent detection method. The 3D laser polishing equipment comprises a base, a working platform, a mounting bracket and a laser generator, the mounting bracket and the laser generator are arranged on the working platform; the laser generator is arranged above the mounting bracket, and a vibration lens and a first driving assembly are further arranged on the mounting bracket; the laser generator is connected with the vibration lens, and an objective table is further arranged on the working platform. A collecting device is further arranged on the side face of the working platform and used for collecting machining data of the workpiece on the objective table. According to the method, reverse design is conducted on the workpiece based on the structured light imaging technology, the three-dimensional structure model is obtained, the surface quality is evaluated, the more uniform and consistent processing effect on the complex structure surface is achieved by analyzing surface structure information and planning the laser polishing path, and the ideal surface is obtained.

Description

Technical Field

[0001] The present invention belongs to the field of laser processing technology, and specifically relates to 3D laser polishing equipment and an intelligent detection method based on structured light reverse design. Background Art

[0002] Laser polishing is a high-precision, high-efficiency and highly flexible surface treatment method, which is of great significance in the field of manufacturing and processing. This technology has brought important impacts by using the high energy and focusing ability of lasers to finely process the surface of materials. First of all, laser polishing can achieve highly precise surface processing, eliminate or reduce the defects, unevenness and layering left in the manufacturing process, and improve the surface finish and overall appearance of the product. This is crucial for some industries that require high-quality appearance, such as automotive, aerospace, medical devices and jewelry manufacturing.

[0003] In recent years, with the development of structured light scanning technology, reverse engineering and defect detection technology have made significant breakthroughs. In the field of industrial manufacturing, the support of structured light scanning technology has become particularly critical. Structured light imaging can perform three-dimensional modeling of the initial state of the workpiece before laser polishing. By capturing the coded pattern on the surface of the workpiece, an accurate three-dimensional model can be generated to record the shape and surface features of the workpiece before polishing. By analyzing the structural information of the workpiece surface, the areas that need special polishing are determined, and the laser polishing path is planned according to the three-dimensional model to automatically adjust the focal length to achieve a more uniform and consistent surface treatment effect and obtain an ideal surface.

[0004] In the prior art, for example, the technical content recorded in the patent with publication number: CN112102268A: an intelligent laser polishing module based on machine learning for polishing ultra-thin masks, which combines machine learning algorithms with laser polishing, and is used to polish ultra-thin masks to improve their surface finish. The module mainly includes a laser polishing image data acquisition and classification module, a laser polishing flatness, surface roughness, and defect recognition module, a polishing area precise recognition and division module, and a mask laser polishing online precise continuity splicing module.

[0005] In the prior art, during the laser processing, the three-dimensional information of the workpiece is not obtained before laser polishing, resulting in the inability to automatically plan the path; at the same time, due to the undulating shape of the workpiece, the laser cannot automatically focus according to the shape, resulting in unsatisfactory processing results and difficulty in achieving high-precision quality control and optimization of surface treatment; after laser polishing, there are defects, flaws or irregular pits on the workpiece that cannot be found in local positions, making it impossible to fully evaluate the processing effect as a whole. Summary of the invention

[0006] The purpose of the present invention is to provide 3D laser polishing equipment and an intelligent detection method based on structured light reverse design, so as to solve the problems in the prior art proposed in the background technology, that in the laser processing process, the operator relies on experience or blindly adjusts the laser polishing parameters, resulting in unstable processing effect; before laser polishing, the three-dimensional information of the workpiece is not obtained, and due to the ups and downs of the workpiece shape, the processing effect is uneven, and it is difficult to achieve high-precision quality control and optimization of surface treatment; after laser polishing, there are defects, flaws or irregular pits on the workpiece that cannot be found in local positions, and the processing effect cannot be fully evaluated as a whole.

[0007] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0008] The 3D laser polishing equipment based on structured light reverse design includes a base, a working platform, a mounting bracket and a laser generator; wherein the working platform is fixedly arranged on the base, and the mounting bracket and the laser generator are arranged on the working platform;

[0009] The laser generator is arranged above the mounting bracket, and a galvanometer lens and a first driving assembly are also arranged on the mounting bracket; the laser generator is connected to the galvanometer lens, and the galvanometer lens is fixedly arranged on the first driving assembly, and the first driving assembly is used to drive the galvanometer lens to move up and down; a stage is also arranged on the working platform, and the stage is arranged below the galvanometer lens;

[0010] A collection device is also provided on the side of the work platform. The collection device is arranged toward the direction of the stage and is used to collect processing data of the workpiece on the stage. An adjustment component is provided below the collection device and is used to adjust the height of the collection device.

[0011] According to the above technical solution, the first drive component includes a fixed base, a drive motor and a transmission screw; wherein the fixed base is fixedly connected to the mounting bracket, the transmission screw is rotatably arranged inside the fixed base, and the drive motor is arranged on the fixed base and connected to the transmission screw.

[0012] According to the above technical solution, the transmission screw is also connected to the connecting seat, and the transmission screw is connected to the galvanometer lens through the connecting seat.

[0013] According to the above technical solution, a mounting plate is also provided on the mounting bracket, one end of the mounting plate is fixedly connected to the mounting bracket, the galvanometer lens is slidably connected to the mounting plate, and a distance measuring device is provided below the mounting plate, which is used to measure the distance between the galvanometer lens and the stage.

[0014] According to the above technical solution, a second drive assembly is also arranged under the loading platform, and the second drive assembly includes a turntable, a transverse drive assembly and a longitudinal drive assembly, wherein the turntable is arranged on the longitudinal drive assembly, the longitudinal drive assembly is arranged on the transverse drive assembly, and the loading platform is arranged on the turntable.

[0015] According to the above technical solution, the adjustment component includes a mounting seat, an adjustment base and an adjustment bracket; wherein the collection device is arranged on the mounting seat, the mounting seat is fixedly arranged on the adjustment bracket, the adjustment bracket is arranged on the adjustment base, and the adjustment base is fixedly connected to the working platform; the adjustment bracket includes a first adjustment plate, a second adjustment plate, a first connecting rod and a second connecting rod; the first connecting rod and the second connecting rod are both arranged at a position between the first adjustment plate and the second adjustment plate, and the first connecting rod and the second connecting rod are rotatably connected, and the adjustment bracket is used to adjust the height of the collection device.

[0016] According to the above technical solution, the two ends of the first connecting rod are rotatably connected to the first adjustment plate and the second adjustment plate respectively, and one end of the second connecting rod is rotatably connected to the second adjustment plate; a connecting column is arranged below the first adjustment plate, and a threaded rod is rotatably arranged in the middle of the connecting column, and the other end of the second connecting rod is connected to the threaded rod.

[0017] According to the above technical solution, the acquisition device includes a projector and a acquisition camera; wherein the projector and the acquisition camera are arranged toward the direction of the stage, and the acquisition camera is arranged on the side of the projector.

[0018] An intelligent detection method for 3D laser polishing equipment based on structured light reverse design includes the following steps:

[0019] Step S1, placing a calibration plate on the stage, controlling the stage to rotate to different angles and collecting images of the calibration plate through a collection device, performing binocular calibration, and after completion, removing the calibration plate and placing the workpiece to be polished;

[0020] Step S2, using a Gray code and phase shift coded light generation algorithm to obtain a Gray code and a phase shift image, write the image into a projector, and adjust the length and width of the projection format according to the size of the workpiece to be polished;

[0021] Step S3, setting hard triggering of the two cameras and the projector, the projector projects coded light, and the two cameras synchronously collect multiple groups of coded light modulated by the workpiece;

[0022] Step S4, using the internal and external parameters obtained by binocular positioning in step S1 to perform stereo correction on the image collected in step S3; using a decoding algorithm to expand the phase of the corrected image to obtain an absolute phase;

[0023] Step S5, using the absolute phase information to perform binocular stereo matching, calculate the offset of corresponding pixels in the left and right images, establish a triangulation model, and obtain the three-dimensional information of the workpiece;

[0024] Step S6, rotating the stage to a certain angle; repeating steps S3 to S5, stitching the three-dimensional images of the workpiece collected at different angles to obtain complete three-dimensional information of the workpiece;

[0025] Step S7, denoising and filtering the complete three-dimensional information of the workpiece; exporting it into STL format using Geomagic point cloud and transferring it to the laser polishing system;

[0026] Step S8, distinguishing the shape features of the workpiece to be polished according to the three-dimensional information of the workpiece to be polished obtained in step S6, planning the laser polishing path according to the three-dimensional contour of the workpiece to be polished, and ensuring that the polishing effect on the workpiece surface is evenly distributed;

[0027] Step S9, controlling the polishing preview frame of the galvanometer lens to align with the workpiece surface and start polishing;

[0028] Step S10, after polishing is completed, repeating steps S3 to S6, performing three-dimensional reconstruction on the polished workpiece, and using computer vision and image processing algorithms to analyze surface defects of the workpiece, and detecting surface defects of the workpiece caused by incompleteness in the manufacturing process or insufficient polishing;

[0029] Step S11, judging the detection result, if the workpiece has no obvious defects, it indicates that the polishing quality of the workpiece is good and has met expectations; if there are defects on the surface of the workpiece, the product quality is further improved by adjusting the polishing parameters.

[0030] According to the above technical solution, in step S2, the Gray code and phase shift coded light generation algorithm is used to obtain the Gray code and phase shift diagram as follows:

[0031] Step S201, first create a folder for storing the generated pattern and define the parameters;

[0032] Step S202, then call two functions to generate patterns: use m_make_phase_shift_patterns function to generate a phase shift image and store the result in patterns_phaseshift; use m_make_gray_code_patterns function to generate a Gray code image and store the result in patterns_graycode; then, generate a complementary Gray code image and store it in patterns_com; finally, write the generated image to the specified folder and write the pixel values ​​of the image to a CSV file for subsequent processing and analysis;

[0033] The m_make_phase_shift_patterns function generates phase shift patterns based on the following formula:

[0034]

[0035] The function accepts input parameters A, B, T, N, W, and H, which represent background light intensity, modulation intensity, phase shift cycle width, number of patterns, image width, and height, respectively;

[0036] Step S203, calculate the phase shift frequency f according to the input parameters; for each pattern k, calculate the corresponding gray value array I according to the above formula k ;

[0037] The grayscale value array of each pattern is normalized and stored in the Is array, which is used to generate a grayscale image; finally, the generated phase shift pattern array Is and image representation Is_img are returned;

[0038] Step S204, function m_make_gray_code_patterns(n,W,H) implements the function of generating Gray code patterns; first, a set of Gray code strings are generated by calling the m_gray_code function and stored in codes; then, according to the generated Gray code strings, each Gray code is converted into a binary representation and stored in the codes_mat matrix; then, according to the width W of the image and the number of Gray codes, the width W_per of each grid is calculated; then, by looping through each Gray code, the binary value corresponding to each Gray code is assigned to the corresponding pixel point in the pattern, thereby generating a Gray code pattern; finally, two all-white patterns are created to determine the threshold of each pixel, and they are stored in the second to last row and the last row of the patterns array.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] In the present invention, the workpiece is reverse designed based on structured light imaging technology, a three-dimensional structural model is obtained and the surface quality is evaluated. By analyzing the surface structure information, the laser polishing path is planned and the focus is automatically adjusted to achieve a more uniform and consistent processing effect on the surface of the complex structure and obtain an ideal surface.

[0041] The polished workpiece is reconstructed in three dimensions, and the surface defects are analyzed using computer vision and image processing algorithms to detect surface defects such as cracks and uneven surfaces caused by incompleteness or insufficient polishing during the manufacturing process. According to the test results, the polishing process is improved in a timely manner to improve product quality and production efficiency. At the same time, the equipment combines the imaging principle of Gray code + phase shift method to further improve the precision and accuracy of detection, making the entire process more intelligent and reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the overall structure of the present invention;

[0043] Figure 2 It is a front view structural schematic diagram of the present invention;

[0044] Figure 3 It is a schematic diagram of the structure of the acquisition device of the present invention;

[0045] Figure 4 This is a schematic diagram of the structure of the first driving component of the present invention;

[0046] Figure 5 A point cloud data diagram of the workpiece processed by the present invention;

[0047] Figure 6 This is a schematic diagram of the workpiece structure after polishing according to the present invention.

[0048] Markings in the figure: 100-base, 200-working platform, 300-mounting bracket, 400-laser generator, 500-vibrating lens, 600-stage, 700-fixed base, 800-driving motor, 900-transmission screw, 110-connecting seat, 111-mounting plate, 112-distance measuring device, 113-lateral drive assembly, 114-longitudinal drive assembly, 115-mounting seat, 116-adjustment base, 117-first adjustment plate, 118-second adjustment plate, 119-first connecting rod, 120-second connecting rod, 121-connecting column, 122-threaded rod, 123-handle. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] Embodiment 1

[0051] like Figure 1As shown, the 3D laser polishing equipment based on structured light reverse design includes a base 100, a working platform 200, a mounting bracket 300 and a laser generator 400; wherein the working platform 200 is fixedly arranged on the base 100, and the mounting bracket 300 and the laser generator 400 are arranged on the working platform 200;

[0052] The laser generator 400 is arranged above the mounting bracket 300, and a galvanometer lens 500 and a first driving assembly are also arranged on the mounting bracket 300; the laser generator 400 is connected to the galvanometer lens 500, and the galvanometer lens 500 is fixedly arranged on the first driving assembly, and the first driving assembly is used to drive the galvanometer lens 500 to move up and down; a stage 600 is also arranged on the working platform 200, and the stage 600 is arranged below the galvanometer lens 500;

[0053] like Figure 3 As shown, a collection device is also provided on the side of the working platform 200, and the collection device is arranged toward the direction of the stage 600, and the collection device is used to collect processing data of the workpiece on the stage 600; an adjustment component is provided below the collection device, and the adjustment component is used to adjust the height of the collection device.

[0054] like Figure 4 As shown, the first driving assembly includes a fixed base 700, a driving motor 800 and a transmission screw 900; wherein the fixed base 700 is fixedly connected to the mounting bracket 300, the transmission screw 900 is rotatably arranged inside the fixed base 700, and the driving motor 800 is arranged on the fixed base 700 and connected to the transmission screw 900. The transmission screw 900 is also connected to the connecting base 110, and the transmission screw 900 is connected to the galvanometer lens 500 through the connecting base 110.

[0055] Specifically, the driving motor 800 drives the transmission screw 900 to rotate and connect, thereby driving the connection seat 110 connected to the transmission screw 900 to move on the transmission screw 900, and then driving the vibrating lens 500 to move up and down. This allows the height of the vibrating lens 500 to be adjusted, which facilitates the focusing of the vibrating lens 500 and the workpiece to be polished.

[0056] A mounting plate 111 is also provided on the mounting bracket 300, one end of the mounting plate 111 is fixedly connected to the mounting bracket 300, the galvanometer lens 500 is slidably connected to the mounting plate 111, and a distance measuring device 112 is provided below the mounting plate 111, and the distance measuring device 112 is used to measure the distance between the galvanometer lens 500 and the stage 600.

[0057] Specifically, one end of the mounting plate 111 is fixedly connected to the mounting bracket 300, and a slide groove is provided on the mounting plate 111. The mounting plate 111 is slidably connected to the galvanometer lens 500 through the slide groove. When the galvanometer lens 500 moves up and down, it will slide relatively on the mounting plate 111, thereby changing the distance between the galvanometer lens 500 and the distance measuring device 112. When the galvanometer lens 500 exceeds the threshold set by the distance measuring device 112, or the galvanometer lens 500 contacts the distance measuring device 112, the distance measuring device 112 will cause an alarm, thereby preventing the distance between the galvanometer lens 500 and the workpiece to be polished from being too close or too far, affecting the processing of the workpiece to be polished.

[0058] like Figure 2 As shown, a second drive assembly is also arranged below the worktable 600, and the second drive assembly includes a turntable, a transverse drive assembly 113 and a longitudinal drive assembly 114, wherein the turntable is arranged on the longitudinal drive assembly 114, the longitudinal drive assembly 114 is arranged on the transverse drive assembly 113, and the worktable 600 is arranged on the turntable.

[0059] Specifically, by setting the stage 600 on a turntable, the turntable drives the stage 600 to rotate, thereby adjusting the angle of the workpiece to be polished on the stage 600, so as to facilitate the acquisition device to collect images of the workpiece to be polished for subsequent processing.

[0060] Furthermore, the transverse driving assembly 113 and the longitudinal driving assembly 114 are used to move the stage 600 , wherein the transverse driving assembly 113 drives the stage 600 to move left and right, and the longitudinal driving assembly 114 drives the stage 600 to move forward and backward.

[0061] Furthermore, the turntable, the transverse drive assembly 113 and the longitudinal drive assembly 114 all adopt existing devices, wherein the turntable adopts an electric turntable.

[0062] like Figure 3 As shown, the adjustment assembly includes a mounting seat 115, an adjustment base 116 and an adjustment bracket; wherein the collection device is arranged on the mounting seat 115, the mounting seat 115 is fixedly arranged on the adjustment bracket, the adjustment bracket is arranged on the adjustment base 116, and the adjustment base 116 is fixedly connected to the working platform 200; the adjustment bracket includes a first adjustment plate 117, a second adjustment plate 118, a first connecting rod 119 and a second connecting rod 120; the first connecting rod 119 and the second connecting rod 120 are both arranged at a position between the first adjustment plate 117 and the second adjustment plate 118, and the first connecting rod 119 and the second connecting rod 120 are rotatably connected, and the adjustment bracket is used to adjust the height of the collection device.

[0063] The two ends of the first connecting rod 119 are rotatably connected to the first adjustment plate 117 and the second adjustment plate 118 respectively, and one end of the second connecting rod 120 is rotatably connected to the second adjustment plate 118; a connecting column 121 is arranged below the first adjustment plate 117, and a threaded rod 122 is rotatably arranged in the middle of the connecting column 121, the other end of the second connecting rod 120 is connected to the threaded rod 122, and a handle 123 is arranged at the end of the threaded rod 122.

[0064] The acquisition device includes a projector and an acquisition camera; wherein the projector and the acquisition camera are arranged toward the direction of the stage 600, and the acquisition camera is arranged on the side of the projector.

[0065] Specifically, the adjustment component also includes an angle adjustment module; the angle adjustment module includes a first adjustment seat and a second adjustment seat; wherein the second adjustment seat is fixedly connected to the mounting seat 115, the first adjustment seat is rotatably connected to the second adjustment seat, an arc groove is also provided on the first adjustment seat, and a limiting column corresponding to the second arc groove is provided on the second adjustment seat, the limiting column is provided inside the arc groove, and is used for rotation limiting of the first adjustment seat.

[0066] Furthermore, the galvanometer lens adopts an existing device, such as the Jinhai Chuang FL8220-3D-500 fiber optic dynamic focusing galvanometer, which includes an autofocus function, an autofocus distance of 200nm-600nm, and is equipped with a Taiwan Xingcheng MM3D board.

[0067] Furthermore, the laser distance sensor adopts an existing device, such as: OD Precision OD5-85T20 displacement measurement sensor of SICK.

[0068] Furthermore, the lateral drive assembly and the longitudinal drive assembly adopt existing devices, such as the THK-SSR15XV1UUC1 driver of the THKSSR series.

[0069] Embodiment 2

[0070] This embodiment is a further refinement of the first embodiment.

[0071] An intelligent detection method for 3D laser polishing equipment based on structured light reverse design includes the following steps:

[0072] Step S1, placing a calibration plate on the stage 600, controlling the stage 600 to rotate to different angles and collecting the calibration plate image through the acquisition device, performing binocular calibration, and after completion, removing the calibration plate and placing the workpiece to be polished;

[0073] First, multiple images are acquired to cover various angles and positions that the camera may be used in. When acquiring images, ensure that the images cover different angles and distances. This ensures that the calibration results are accurate for various use cases. Then corner detection is performed, using a specific checkerboard template to detect the corners of the checkerboard in the image. This step uses the detectCheckerboardPoints function in Matlab. When detecting corners, the noise and occlusion that may exist in the image, as well as the difference in image quality, are considered. If a sufficient number of corners cannot be detected, the image is skipped or the user is prompted to re-acquire. After the corners are detected, these corners are matched with known 3D world coordinates. Assuming that the checkerboard is on a plane, the 3D coordinates of the corners are assumed to be arranged on the plane at fixed intervals. The camera is calibrated using the matching of the corners and the 3D coordinates. In this way, the camera's internal parameters (such as focal length, principal point) and external parameters (such as rotation, translation), as well as the distortion coefficients, are found. During the calibration process, the distortion of the camera is considered and corrected to improve the accuracy of the calibration. Finally, after the calibration is completed, the accuracy of the calibration results is evaluated. Use metrics such as reprojection error to evaluate the quality of the calibration and perform further optimization if necessary. For example, you can recalibrate parts of the image or adjust the calibration parameters to get better results.

[0074] Step S2, using a Gray code and phase shift coded light generation algorithm to obtain a Gray code and a phase shift image, write the image into a projector, and adjust the length and width of the projection format according to the size of the workpiece to be polished;

[0075] Step S3, setting hard triggering of the two cameras and the projector, the projector projects coded light, and the two cameras synchronously collect multiple groups of coded light modulated by the workpiece;

[0076] Step S4, using the internal and external parameters obtained by binocular positioning in step S1 to perform stereo correction on the image collected in step S3; using a decoding algorithm to expand the phase of the corrected image to obtain an absolute phase;

[0077] Specifically, intrinsic parameters are parameters related to the inherent characteristics of the camera, defining how the camera converts 3D world coordinates into 2D image plane coordinates. They include focal length, principal point position, pixel size, and distortion coefficient. Extrinsic parameters describe the position and orientation of the camera in the world coordinate system, and determine the relative position of two cameras in the world coordinate system. They include rotation matrices and translation vectors.

[0078] Specifically:

[0079]

[0080] First, use formula (1) to calculate the value of each pixel I in the Gray code plus phase shift method k ,in:

[0081] A: Background light intensity, which means the light intensity not affected by modulation.

[0082] B: Modulation intensity, which indicates the amplitude of the modulated light intensity.

[0083] Phase value, indicating the phase information at each pixel.

[0084] Phase shift value, indicating the amount of phase shift at each pixel.

[0085] f: Phase shift frequency, a constant that indicates the frequency of phase change per unit length, also called fringe period or pitch, usually refers to the number of pixels in a single period. The number of periods in the entire field of view is called the number of periods.

[0086] x: represents the pixel coordinate in the encoding direction, that is, the position along the encoding direction.

[0087] Then, according to formula (2), multiple equations are solved simultaneously to obtain the phase at each pixel:

[0088]

[0089] k(x,y)=(V(x,y)) (4)

[0090]

[0091] Next, according to formula (3), the four-bit Gray code GC i (x, y) is used to calculate the code value V(x, y) of the pixel point (x, y). Each Gray code corresponds to a binary bit, and the code value of the pixel point is obtained by weighted summation according to its position in the binary code.

[0092] Then, formula (4) describes how to calculate the Gray code k(x, y) according to the code value V(x, y), that is, rounding the code value to the nearest integer as the value of the Gray code.

[0093] Finally, formula (5) describes how to convert the Gray code k(x, y) into a phase value in represents the wrapped phase value obtained by formula (2). Add the binary code phase 2πk(x, y) corresponding to the Gray code to the wrapped phase The final absolute phase value is obtained

[0094] These formulas together constitute the phase calculation and encoding process in the phase encoding method. The following is a partial algorithm implementation form for establishing the mapping relationship in Gray code decoding.

[0095]

[0096]

[0097] Step S5, using the phase information to perform binocular stereo matching, calculate the offset of corresponding pixels in the left and right images, establish a triangulation model, and obtain the three-dimensional information of the workpiece;

[0098] Step S6, rotating the stage 600 to 90°, 180°, and 270°; repeating steps S3 to S5, stitching the three-dimensional images of the workpiece collected at different angles to obtain complete three-dimensional information of the workpiece;

[0099] For the specific stereo matching steps, use C++ to create the phaseMatch function, and search from the upper left corner to the lower right corner by traversing each pixel in the left image. To control the search method, you can choose global search or local search. If the previous global search finds a matching point pair, the search range of the next pixel will become a local search, otherwise continue the global search. If a matching pixel pair is found, sub-pixel interpolation is performed, and then the matching pixel pair is added to cps, which is the matching point pair container. Finally, a Boolean value is returned to indicate whether the pixel pair is successfully matched. Some of the code is as follows:

[0100] bool Matcher::phaseMatch(cv::Mat&phaL,cv::Mat&phaR,vector <pair<cv::Point2f,cv::Point2f> >&cps)

[0101] for(int Y_L = wh; Y_L <phaL.rows-wh;Y_L++){

[0102] bool flag_glob_search=true;

[0103] for(int X_L=wh;X_L <phaL.cols-wh;X_L++){

[0104] cv::Rect roi_L(cv::Point(X_L-wh,Y_L-wh),winSize);

[0105] cv::Mat BOX_L=phaL(roi_L);

[0106] flag_find = false;

[0107] if (is_valid_box(BOX_L)) {

[0108] if(flag_glob_search){

[0109] XR=phase_search(BOX_L,phaR,Y_L,wh);

[0110] XR==-1? flag_find=false:flag_find=true;

[0111] bool flag_glob_search=true;

[0112] On this basis, the search_sub_pixel function is used to perform sub-pixel interpolation on the matching points to obtain more accurate matching results. First, check whether the search boundary is out of bounds. If it is out of bounds, return the original value directly. Then, through spline interpolation, interpolate the five pixel points near the matching point to obtain a more accurate matching point. If the interpolation result is within the specified range, return the interpolation result, otherwise return the result after other processing. Part of the code is as follows:

[0113] double Matcher::search_sub_pixel(cv::Mat&BOX_L,cv::Mat&phaR,int XR,int YR){

[0114] if(XR-2<0||XR+2>phaR.cols){

[0115] return double(XR);

[0116] }

[0117] double xs[5],ys[5];

[0118] int idx = 0;

[0119] for(int i=-2;i<=2;++i){

[0120] double pha_v=phaR.at <double>(YR,XR+i);

[0121] if (pha_v == 0) {

[0122] return double(XR);

[0123] }

[0124] xs[idx] = pha_v;

[0125] ys[idx] = double(XR+i);

[0126] idx+=1;

[0127] }.

[0128] Step S7, denoising and filtering the complete three-dimensional information of the workpiece; exporting it into STL format using Geomagic point cloud and transferring it to the laser polishing system;

[0129] Step S8, distinguishing the shape features of the workpiece to be polished according to the three-dimensional information of the workpiece to be polished obtained in step S6, planning the laser polishing path according to the three-dimensional contour of the workpiece to be polished, and ensuring that the polishing effect on the workpiece surface is evenly distributed;

[0130] Step S9, controlling the polishing preview frame of the galvanometer lens 500 to align with the workpiece surface and starting polishing;

[0131] Step S10, after polishing is completed, repeating steps S3 to S6, performing three-dimensional reconstruction on the polished workpiece, and using computer vision and image processing algorithms to analyze surface defects of the workpiece, and detecting surface defects of the workpiece caused by incompleteness in the manufacturing process or insufficient polishing;

[0132] Step S11, judging the detection result, if the workpiece has no obvious defects, it indicates that the polishing quality of the workpiece is good and has met expectations; if there are defects on the surface of the workpiece, the product quality is further improved by adjusting the polishing parameters.

[0133] In step S2, the Gray code and phase shift coded light generation algorithm is used to obtain the Gray code and phase shift diagram as follows:

[0134] Step S201, first create a folder for storing the generated pattern and define the parameters;

[0135] Step S202, then call two functions to generate patterns: use m_make_phase_shift_patterns function to generate a phase shift image and store the result in patterns_phaseshift; use m_make_gray_code_patterns function to generate a Gray code image and store the result in patterns_graycode; then, generate a complementary Gray code image and store it in `patterns_com`; finally, write the generated image to the specified folder and write the pixel values ​​of the image to a CSV file for subsequent processing and analysis;

[0136] The m_make_phase_shift_patterns function generates phase shift patterns based on the following formula:

[0137]

[0138] The function accepts input parameters A, B, T, N, W, and H, which represent background light intensity, modulation intensity, phase shift cycle width, number of patterns, image width, and height, respectively;

[0139] Step S203, calculate the phase shift frequency f according to the input parameters; for each pattern k, calculate the corresponding gray value array I according to the above formula k ;

[0140] The grayscale value array of each pattern is normalized and stored in the Is array, which is used to generate a grayscale image; finally, the generated phase shift pattern array Is and image representation Is_img are returned;

[0141] Some of the code is as follows:

[0142] function[Is,Is_img]=m_make_phase_shift_patterns(A,B,T,N,W,H)

[0143] Is = cell (N, 1);

[0144] Is_img = zeros(N,H,W);

[0145] xs=0:W-1;

[0146] f = W / T;

[0147] fork=0:N-1

[0148] I=A+B*cos(2.*f*xs / W*pi-2.*k / N*pi);

[0149] Is{k+1}=I;

[0150] Is_img(k+1,:,:)=repmat(Is{k+1} / 255.,H,1);

[0151] end

[0152] end.

[0153] Step S204, function m_make_gray_code_patterns(n,W,H) implements the function of generating Gray code patterns; first, a set of Gray code strings are generated by calling the m_gray_code function and stored in codes; then, according to the generated Gray code strings, each Gray code is converted into a binary representation and stored in the codes_mat matrix; then, according to the width W of the image and the number of Gray codes, the width W_per of each grid is calculated; then, by looping through each Gray code, the binary value corresponding to each Gray code is assigned to the corresponding pixel point in the pattern, thereby generating a Gray code pattern; finally, two all-white patterns are created to determine the threshold of each pixel, and they are stored in the second to last row and the last row of the patterns array.

[0154] Some of the code is as follows

[0155] function[patterns]=m_make_gray_code_patterns(n,W,H)

[0156] codes = m_gray_code(n);

[0157] [~,num]=size(codes);

[0158] codes_mat=zeros(n,num,'int8');

[0159] for col=1:num

[0160] code_col=str2mat(codes(col));

[0161] for row=1:n

[0162] codes_mat(row,col)=str2num(code_col(row));

[0163] end

[0164] end.

[0165] Embodiment 3

[0166] This embodiment provides a specific implementation method.

[0167] The sample selected in this embodiment is AlSi10Mg, which is a common 3D printing aluminum alloy. AlSi10Mg material is widely used in the field of additive manufacturing, has excellent mechanical properties and corrosion resistance, and is particularly suitable for manufacturing high-strength, lightweight complex parts. Using existing additive manufacturing technology, the metal powder is melted layer by layer by a laser beam to print out a complex structure with multiple planes and curved surfaces at different heights. However, due to problems such as step effect, spheroidization effect, and powder adhesion, the processed surface often has a variety of inherent defects, which makes it difficult to meet the requirements of long-life service and now requires further processing.

[0168] According to the 3D laser polishing equipment and intelligent detection method based on structured light reverse design provided by the present invention, firstly, the sample is reversely designed using the structured light binocular imaging technology to obtain the point cloud data of the AlSi10Mg material additive manufacturing part (such as Figure 5 As shown in the figure), the point cloud data is accurately captured in the STL format and imported into the laser polishing system. In this process, the polishing trajectory is planned according to the obtained 3D model, and the path and energy distribution of the laser polishing are precisely controlled to ensure the best polishing effect. Then, the polishing is started and the laser is manipulated to act on the surface of the workpiece. Laser polishing technology uses a high-energy laser beam to melt the surface of the workpiece and remove tiny defects and roughness. By importing the point cloud data into the system and precisely controlling it, laser polishing can improve the surface quality more evenly and effectively. Finally, Figure 6 The workpiece after polishing is shown. After laser polishing, the surface of the workpiece becomes significantly smoother, the roughness is greatly reduced, and the ideal surface quality is achieved. Through the above steps, the accuracy of point cloud data and the efficiency of laser polishing are used to significantly improve the surface quality of the workpiece and meet the high-standard industrial needs.

[0169] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0170] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.< / double>

Claims

1. 3D laser polishing equipment based on structured light reverse design, characterized by: It includes a base, a working platform, a mounting bracket and a laser generator; wherein the working platform is fixedly arranged on the base, and the mounting bracket and the laser generator are arranged on the working platform; The laser generator is arranged above the mounting bracket, and a galvanometer lens and a first driving assembly are also arranged on the mounting bracket; the laser generator is connected to the galvanometer lens, and the galvanometer lens is fixedly arranged on the first driving assembly, and the first driving assembly is used to drive the galvanometer lens to move up and down; a stage is also arranged on the working platform, and the stage is arranged below the galvanometer lens; A collection device is also provided on the side of the work platform. The collection device is arranged toward the direction of the stage and is used to collect processing data of the workpiece on the stage. An adjustment component is provided below the collection device and is used to adjust the height of the collection device.

2. The 3D laser polishing equipment based on structured light reverse design according to claim 1, characterized in that: The first driving assembly includes a fixed base, a driving motor and a transmission screw; The fixed base is fixedly connected to the mounting bracket, the transmission screw is rotatably arranged inside the fixed base, and the driving motor is arranged on the fixed base and connected to the transmission screw.

3. The 3D laser polishing equipment based on structured light reverse design according to claim 2, characterized in that: The transmission lead screw is also connected to the connecting seat, and the transmission lead screw is connected to the galvanometer lens through the connecting seat.

4. The 3D laser polishing equipment based on structured light reverse design according to claim 3, characterized in that: A mounting plate is also provided on the mounting bracket, one end of the mounting plate is fixedly connected to the mounting bracket, the galvanometer lens is slidably connected to the mounting plate, and a distance measuring device is provided below the mounting plate for measuring the distance between the galvanometer lens and the stage.

5. The 3D laser polishing equipment based on structured light reverse design according to claim 4, characterized in that: A second drive assembly is also arranged below the loading platform, and the second drive assembly includes a turntable, a transverse drive assembly and a longitudinal drive assembly, wherein the turntable is arranged on the longitudinal drive assembly, the longitudinal drive assembly is arranged on the transverse drive assembly, and the loading platform is arranged on the turntable.

6. The 3D laser polishing equipment based on structured light reverse design according to claim 1, characterized in that: The adjustment assembly includes a mounting seat, an adjustment base and an adjustment bracket; wherein the collection device is arranged on the mounting seat, the mounting seat is fixedly arranged on the adjustment bracket, the adjustment bracket is arranged on the adjustment base, and the adjustment base is fixedly connected to the working platform; the adjustment bracket includes a first adjustment plate, a second adjustment plate, a first connecting rod and a second connecting rod; the first connecting rod and the second connecting rod are both arranged at a position between the first adjustment plate and the second adjustment plate, and the first connecting rod and the second connecting rod are rotatably connected, and the adjustment bracket is used to adjust the height of the collection device.

7. The 3D laser polishing equipment based on structured light reverse design according to claim 6, characterized in that: The two ends of the first connecting rod are rotatably connected to the first adjustment plate and the second adjustment plate respectively, and one end of the second connecting rod is rotatably connected to the second adjustment plate; a connecting column is arranged below the first adjustment plate, a threaded rod is rotatably arranged in the middle of the connecting column, the other end of the second connecting rod is connected to the threaded rod, and a handle is arranged at the end of the threaded rod.

8. The 3D laser polishing equipment based on structured light reverse design according to claim 1, characterized in that: The acquisition device includes a projector and an acquisition camera; wherein the projector and the acquisition camera are arranged toward the direction of the stage, and the acquisition camera is arranged on the side of the projector.

9. Intelligent detection method for 3D laser polishing equipment based on structured light reverse design, characterized by: The detection method uses any one of the laser polishing equipment described in claims 1 to 8, and the detection method comprises the following steps: Step S1, placing a calibration plate on the stage, controlling the stage to rotate to different angles and collecting images of the calibration plate through a collection device, performing binocular calibration, and after completion, removing the calibration plate and placing the workpiece to be polished; Step S2, using a Gray code and phase shift coded light generation algorithm to obtain a Gray code and a phase shift image, write the image into a projector, and adjust the length and width of the projection format according to the size of the workpiece to be polished; Step S3, setting hard triggering of the two cameras and the projector, the projector projects coded light, and the two cameras synchronously collect multiple groups of coded light modulated by the workpiece; Step S4, using the internal and external parameters obtained by binocular positioning in step S1 to perform stereo correction on the image collected in step S3; using a decoding algorithm to expand the phase of the corrected image to obtain an absolute phase; Step S5, using the phase information of the absolute phase to perform binocular stereo matching, calculate the offset of corresponding pixels in the left and right images, establish a triangulation model, and obtain the three-dimensional information of the workpiece; Step S6, rotating the stage to a certain angle by means of a turntable; repeating steps S3 to S5, stitching the three-dimensional images of the workpiece collected at different angles, and obtaining complete three-dimensional information of the workpiece; Step S7, denoising and filtering the complete three-dimensional information of the workpiece; exporting it using Geomagic point cloud and transferring it to the laser polishing system; Step S8, distinguishing the shape features of the workpiece to be polished according to the three-dimensional information of the workpiece to be polished obtained in step S6, planning the laser polishing path according to the three-dimensional contour of the workpiece to be polished, and ensuring that the polishing effect on the workpiece surface is evenly distributed; Step S9, controlling the polishing preview frame of the galvanometer lens to align with the workpiece surface and start polishing; Step S10, after polishing is completed, repeating steps S3 to S6, performing three-dimensional reconstruction on the polished workpiece, and using computer vision and image processing algorithms to analyze surface defects of the workpiece, and detecting surface defects of the workpiece caused by incompleteness in the manufacturing process or insufficient polishing; Step S11, judging the detection result, if the workpiece has no obvious defects, it indicates that the polishing quality of the workpiece is good and has met expectations; if there are defects on the surface of the workpiece, the product quality is further improved by adjusting the polishing parameters.

10. The intelligent detection method of 3D laser polishing equipment based on structured light reverse design according to claim 9, characterized in that: In step S2, the Gray code and phase shift coded light generation algorithm is used to obtain the Gray code and phase shift diagram as follows: Step S201, first create a folder for storing the generated pattern and define the parameters; Step S202, then call two functions to generate patterns: use m_make_phase_shift_patterns function to generate a phase shift image and store the result in patterns_phaseshift; use m_make_gray_code_patterns function to generate a Gray code image and store the result in patterns_gray code; then, generate a complementary Gray code image and store it in patterns_com; finally, write the generated image to the specified folder and write the pixel values ​​of the image to a CSV file for subsequent processing and analysis; The m_make_phase_shift_patterns function generates phase shift patterns based on the following formula: The function accepts input parameters A, B, T, N, W, and H, which represent background light intensity, modulation intensity, phase shift cycle width, number of patterns, image width, and height, respectively; Step S203, calculate the phase shift frequency f according to the input parameters; for each pattern k, calculate the corresponding gray value array I according to the above formula k ; The grayscale value array of each pattern is normalized and stored in the Is array, which is used to generate a grayscale image; finally, the generated phase shift pattern array Is and image representation Is_img are returned; Step S204, function m_make_gray_code_patterns(n,W,H) implements the function of generating Gray code patterns; first, a set of Gray code strings are generated by calling the m_gray_code function and stored in codes; then, according to the generated Gray code string, each Gray code is converted into a binary representation and stored in the codes_mat matrix; then, according to the width W of the image and the number of Gray codes, the width W_per of each grid is calculated; then, by looping through each Gray code, the binary value corresponding to each Gray code is assigned to the corresponding pixel point in the pattern, thereby generating a Gray code pattern; finally, two all-white patterns are created to determine the threshold of each pixel, and they are stored in the second to last row and the last row of the patterns array.

Images