PCB multi-layer board slitting method based on machine vision

Abstract

The application relates to a PCB multi-layer board slitting method based on machine vision, which comprises the following steps: setting a first piece formula parameter; acquiring a PCB multi-layer board image to be recognized through a camera, processing the PCB multi-layer board image to be recognized through an image processing algorithm, obtaining a cutting parameter of the PCB multi-layer board to be recognized, comparing the cutting parameter with the first piece formula parameter, and judging whether the cutting parameter is within the tolerance range of the first piece formula parameter; if the cutting parameter is within the tolerance range of the first piece formula parameter, the PCB multi-layer board to be recognized is transferred to a cutting mechanism through a mechanical hand, the cutting mechanism cuts the PCB multi-layer board to be recognized according to the cutting parameter; and if the cutting is successful, the cutting mechanism can cut a plurality of PCB multi-layer boards of the same specification as the PCB multi-layer board to be recognized according to the first piece formula parameter. The PCB multi-layer board slitting method can automatically divide the boards, does not need to divide the boards through manual operation, has high board dividing efficiency, and does not cause damage in the board dividing process.

Description

Technical Field

[0001] The present invention belongs to the technical field of machine vision detection, and in particular relates to a PCB multilayer board slitting method based on machine vision. Background Art

[0002] In recent years, as electronic devices and electronic terminal products have become increasingly thinner and smaller, the demand for high-density and high-performance PCBs has become increasingly prominent. Improving utilization, reducing the distance between boards, and increasing production have become the general trend in the PCB industry.

[0003] However, in the PCB lamination process, the post-pressing disassembly process still relies primarily on manual separation. This involves manually trimming and recycling excess copper foil, separating multiple panels, or even skipping the trimming of excess copper foil on some production lines. This leads to long mold opening cycles, slow efficiency, and low precision. This is especially true when separating boards with soldered components, as the vibration can damage the components themselves. Therefore, automated PCB lamination and slitting is a key core technology within PCB manufacturing and a pain point that the industry urgently needs to address.

[0004] Therefore, the present invention provides a PCB multilayer board slitting method based on machine vision to solve the above technical problems. Summary of the Invention

[0005] In response to the above problems, the present invention aims to provide a PCB multilayer board slitting method based on machine vision, which has high board slitting efficiency and does not cause damage during the board slitting process.

[0006] The present invention provides a PCB multilayer board slitting method based on machine vision, comprising the following steps:

[0007] Step S1, setting first-article formula parameters, wherein the first-article formula parameters include standard values ​​of the length and width of the core of the PCB multilayer board, the cutting edge deviation value around the core of the PCB multilayer board, the number of cores of the PCB multilayer board, and the thickness of the PCB multilayer board;

[0008] Step S2: acquiring an image of the PCB multilayer board to be identified through a camera, processing the image of the PCB multilayer board to be identified through an image processing algorithm to obtain cutting parameters of the PCB multilayer board to be identified, and comparing the cutting parameters with the first-article recipe parameters to determine whether the cutting parameters are within the tolerance range of the first-article recipe parameters;

[0009] Step S3: If the cutting parameters are within the tolerance range of the first-article recipe parameters, the PCB multilayer board to be identified is transferred to a cutting mechanism by a robot arm, and the cutting mechanism cuts the PCB multilayer board to be identified according to the cutting parameters;

[0010] Step S4, if the cutting parameter is not within the tolerance range of the first-piece recipe parameter, return to step S1 to readjust the first-piece recipe parameter, and after readjusting the first-piece recipe parameter, continue the image processing and parameter comparison operation on the to-be-identified PCB multilayer board in step S2, and sequentially cycle until the cutting parameter of the to-be-identified PCB multilayer board is within the tolerance range of the readjusted first-piece recipe parameter, and then enter step S3 to perform the cutting operation;

[0011] Step S5, if the cutting mechanism successfully cuts the to-be-identified PCB multilayer board according to the cutting parameter, the cutting mechanism can cut the PCB multilayer boards of the same specification as the to-be-identified PCB multilayer board in batches according to the first-piece recipe parameter.

[0012] Preferably, the to-be-identified PCB multilayer board is a PCB multilayer board with four board cores.

[0013] Preferably, in step S2, the image of the to-be-identified PCB multilayer board is processed by an image processing algorithm to obtain the cutting parameter of the to-be-identified PCB multilayer board, specifically including:

[0014] Step S21, performing a smoothing preprocessing operation on the image of the to-be-identified PCB multilayer board to remove noise points on the image of the to-be-identified PCB multilayer board;

[0015] Step S22, using an adaptive segmentation algorithm to segment the image of the to-be-identified PCB multilayer board after the smoothing preprocessing operation, to segment the board cores on the image of the to-be-identified PCB multilayer board and perform a binaryzation processing on the segmented board cores to form a segmented image;

[0016] Step S23, performing a minimum bounding rectangle on the board cores on the segmented image, and finding the center point of the minimum bounding rectangle of the board cores and the angle of the board cores, and correcting the segmented image as a whole according to the angle of the board cores to obtain a cutting image;

[0017] Step S24, comparing the minimum bounding rectangles of the left and right two board cores on the cutting image to obtain the left and right edge lines of the vertical middle cutting region. Then, the position of the vertical middle cutting line is found according to the left and right edge lines. A minimum bounding rectangle is further performed on the minimum bounding rectangles of the left and right two board cores on the cutting image to obtain a first bounding rectangle of the left and right two board cores. The four vertex coordinates of the first bounding rectangle are compared to obtain the edge lines of the left and right side cutting regions. The edge lines of the left and right side cutting regions are respectively moved a fixed distance to the left and right sides to obtain a left cutting line and a right cutting line;

[0018] Step S25: Compare the minimum circumscribed rectangles of the upper and lower cores on the cropped image to obtain the upper and lower edges of the horizontal middle cropping area; then calculate the position of the horizontal middle cropping line based on the upper and lower edges; create another minimum circumscribed rectangle based on the minimum circumscribed rectangles of the upper and lower cores on the cropped image to obtain a second circumscribed rectangle of the upper and lower cores; compare the coordinates of the four vertices of the second circumscribed rectangle to obtain the edges of the upper and lower cropping areas; move the edges of the upper and lower cropping areas to the upper and lower sides by a fixed distance respectively to obtain the upper cropping line and the lower cropping line.

[0019] Step S26, perform nine-point calibration on the cut image through a manipulator and a camera to obtain the cutting parameters of the PCB multilayer board to be identified, wherein the cutting parameters include the length and width values ​​of the board core on the cut image, the number of board cores on the cut image, the thickness of the board core on the cut image, the coordinate value of the center point of the board core on the cut image, and the coordinate values ​​of the vertical middle cutting line, the horizontal middle cutting line, the upper cutting line, the lower cutting line, the left cutting line and the right cutting line on the cut image.

[0020] Preferably, the cutting mechanism consists of a rolling cutter and / or a rolling cutter.

[0021] Compared with the related art, the present invention provides a PCB multilayer board cutting method based on machine vision, including: setting the first-piece formula parameters; obtaining the image of the PCB multilayer board to be identified by a camera, processing the image of the PCB multilayer board to be identified by an image processing algorithm, obtaining the cutting parameters of the PCB multilayer board to be identified, comparing the cutting parameters with the first-piece formula parameters, and judging whether the cutting parameters are within the tolerance range of the first-piece formula parameters; if the cutting parameters are within the tolerance range of the first-piece formula parameters, transporting the PCB multilayer board to be identified to a cutting mechanism by a manipulator, and the cutting mechanism cutting the PCB multilayer board to be identified according to the cutting parameters; if the cutting is successful, the cutting mechanism can batch cut PCB multilayer boards of the same specification as the PCB multilayer board to be identified according to the first-piece formula parameters. The present invention accurately finds the edge position of the PCB multilayer board and automatically cuts it to form a board-cutting action through visual operation analysis and mechanism cutting principles, which can realize automatic board cutting without manual board cutting. Its board cutting efficiency is high, and the board cutting process will not cause damage. In addition, the use of the present invention can meet the beat requirements of the entire line and realize the automation, intelligence, unmanned and online operation of the board separation action. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of a method for slitting a multilayer PCB based on machine vision according to the present invention;

[0023] Figure 2A schematic diagram of the process of obtaining cutting parameters according to the present invention;

[0024] Figure 3 This is a schematic diagram of cutting the multi-layer PCB to be identified according to the present invention. DETAILED DESCRIPTION

[0025] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0026] Please also refer to the attached Figure 1-2 The present invention provides a PCB multilayer board cutting method based on machine vision, comprising the following steps:

[0027] Step S1, setting first-article formula parameters, wherein the first-article formula parameters include standard values ​​of the length and width of the core of the PCB multilayer board, the cutting edge deviation value around the core of the PCB multilayer board, the number of cores of the PCB multilayer board, and the thickness of the PCB multilayer board;

[0028] Step S2: acquiring an image of the PCB multilayer board to be identified through a camera, processing the image of the PCB multilayer board to be identified through an image processing algorithm to obtain cutting parameters of the PCB multilayer board to be identified, and comparing the cutting parameters with the first-article recipe parameters to determine whether the cutting parameters are within the tolerance range of the first-article recipe parameters;

[0029] Step S3: If the cutting parameters are within the tolerance range of the first-article recipe parameters, the PCB multilayer board to be identified is transferred to a cutting mechanism by a robot arm, and the cutting mechanism cuts the PCB multilayer board to be identified according to the cutting parameters;

[0030] In step S4, if the cutting parameters are not within the tolerance range of the first-article formula parameters, the process returns to step S1 to readjust the first-article formula parameters. After readjusting the first-article formula parameters, the process proceeds to step S2 to continue performing image processing and parameter comparison operations on the multilayer PCB to be identified, and the process repeats until the cutting parameters of the multilayer PCB to be identified are within the tolerance range of the adjusted first-article formula parameters, and then the process proceeds to step S3 to perform the cutting operation.

[0031] In step S5, if the cutting mechanism successfully cuts the PCB multilayer board to be identified according to the cutting parameters, the cutting mechanism can batch cut PCB multilayer boards of the same specification as the PCB multilayer board to be identified according to the first piece recipe parameters.

[0032] In this embodiment, the PCB multilayer board to be identified is a PCB multilayer board with four board cores. Of course, it is not limited to this. The PCB multilayer board to be identified can also be a PCB multilayer board with 2 board cores, or a PCB multilayer board with 6 board cores, or a PCB multilayer board with 8 board cores, etc. The board separation principle is consistent with the principle of this method and is within the scope of protection of the present invention.

[0033] Specifically, in step S2, the image of the PCB multilayer board to be identified is processed by an image processing algorithm to obtain the cutting parameters of the PCB multilayer board to be identified, which specifically includes:

[0034] Step S21, performing a smoothing preprocessing operation on the PCB multilayer board image to be identified to remove noise on the PCB multilayer board image to be identified;

[0035] Step S22, using an adaptive segmentation algorithm to segment the PCB multilayer board image to be identified after the smoothing preprocessing operation, so as to segment the core of the PCB multilayer board image to be identified and binarize the PCB multilayer board image to be identified with the core segmented, to form a segmented image;

[0036] Step S23, making a minimum circumscribed rectangle for the board core on the segmented image, calculating the center point of the minimum circumscribed rectangle of the board core and the angle of the board core, and performing overall correction on the segmented image according to the angle of the board core to obtain a cropped image;

[0037] Step S24: Compare the minimum bounding rectangles of the left and right cores on the cropped image to determine the left and right edges of the vertical middle cropping area. Then, determine the position of the vertical middle cropping line 1 based on the left and right edges. Create another minimum bounding rectangle based on the minimum bounding rectangles of the left and right cores on the cropped image to obtain the first bounding rectangles of the left and right cores. Compare the coordinates of the four vertices of the first bounding rectangle to determine the edges of the left and right cropping areas. Move the edges of the left and right cropping areas to the left and right by a fixed distance, respectively, to obtain left cropping line 2 and right cropping line 3.

[0038] Step S25, comparing the minimum circumscribed rectangles of the upper and lower cores on the cutting image to obtain the upper and lower edge lines of the horizontal middle cutting region; then calculating the position of the horizontal middle cutting line 4 according to the upper and lower edge lines; making a minimum circumscribed rectangle based on the minimum circumscribed rectangles of the upper and lower cores on the cutting image to obtain the second circumscribed rectangle of the upper and lower cores; comparing the coordinates of the four vertices of the second circumscribed rectangle to obtain the edge lines of the upper and lower side cutting regions; moving the edge lines of the upper and lower side cutting regions upward and downward by a fixed distance respectively to obtain the upper side cutting line 5 and the lower side cutting line 6.

[0039] Step S26, nine-point calibration of the cutting image by the mechanical hand and the camera to obtain the cutting parameters of the multi-layer PCB to be identified, including the length and width values of the cores on the cutting image, the number of cores on the cutting image, the thickness of the cores on the cutting image, the center point coordinate values of the cores on the cutting image, and the coordinate values of the vertical middle cutting line 1, the horizontal middle cutting line 4, the upper side cutting line 5, the lower side cutting line 6, the left side cutting line 2 and the right side cutting line 3 on the cutting image.

[0040] In addition, it should be noted that the nine-point calibration of the cutting image by the mechanical hand and the camera obtains the conversion relationship between the cutting image pixel coordinates and the mechanical hand coordinates. According to the results of the nine-point calibration, the length and width values of the cores on the cutting image, the number of cores on the cutting image, the thickness of the cores on the cutting image, the center point coordinate values of the cores on the cutting image, and the coordinate values of the vertical middle cutting line 1, the horizontal middle cutting line 4, the upper side cutting line 5, the lower side cutting line 6, the left side cutting line 2 and the right side cutting line 3 on the cutting image are obtained. Finally, the above parameters are sent to the mechanical hand and the cutting mechanism for grabbing and cutting.

[0041] In the present embodiment, the cutting mechanism is composed of a milling cutter and / or a hob.

[0042] Compared with the related art, the application provides a PCB multi-layer board slitting method based on machine vision, which comprises the following steps: setting a first piece formula parameter; acquiring an image of a PCB multi-layer board to be identified by a camera, processing the image of the PCB multi-layer board to be identified by an image processing algorithm, obtaining a cutting parameter of the PCB multi-layer board to be identified, comparing the cutting parameter with the first piece formula parameter, and judging whether the cutting parameter is within a tolerance range of the first piece formula parameter; if the cutting parameter is within the tolerance range of the first piece formula parameter, transferring the PCB multi-layer board to be identified to a cutting mechanism by a mechanical hand, and cutting the PCB multi-layer board to be identified according to the cutting parameter by the cutting mechanism; and if the cutting is successful, cutting a plurality of PCB multi-layer boards of the same specification as the PCB multi-layer board to be identified according to the first piece formula parameter by the cutting mechanism. According to the application, the edge position of the PCB multi-layer board is accurately found and the board cutting action is automatically formed by visual operation analysis and mechanism cutting principle, so that the automatic board cutting can be realized, the manual board cutting is not needed, the board cutting efficiency is high, and the board cutting process will not cause damage. In addition, the application can meet the whole line beat requirement, realize the automatic, intelligent, unmanned and connected board cutting action.

[0043] It should be noted that the above-mentioned embodiments are understood as illustrative rather than limiting the protection scope of the application, and the protection scope of the application is subject to the claims. Some non-essential improvements and adjustments made by those skilled in the art without departing from the essence and scope of the application still belong to the protection scope of the application.

Claims

1. A PCB multilayer board cutting method based on machine vision, characterized in that: The method comprises the following steps: Step S1, setting first-article formula parameters, wherein the first-article formula parameters include standard values ​​of the length and width of the core of the PCB multilayer board, the cutting edge deviation value around the core of the PCB multilayer board, the number of cores of the PCB multilayer board, and the thickness of the PCB multilayer board; Step S2: acquiring an image of the PCB multilayer board to be identified through a camera, processing the image of the PCB multilayer board to be identified through an image processing algorithm to obtain cutting parameters of the PCB multilayer board to be identified, and comparing the cutting parameters with the first-article recipe parameters to determine whether the cutting parameters are within the tolerance range of the first-article recipe parameters; Step S3: If the cutting parameters are within the tolerance range of the first-article recipe parameters, the PCB multilayer board to be identified is transferred to a cutting mechanism by a robot arm, and the cutting mechanism cuts the PCB multilayer board to be identified according to the cutting parameters; In step S4, if the cutting parameters are not within the tolerance range of the first-article formula parameters, the process returns to step S1 to readjust the first-article formula parameters. After readjusting the first-article formula parameters, the process proceeds to step S2 to continue performing image processing and parameter comparison operations on the multilayer PCB to be identified, and the process repeats until the cutting parameters of the multilayer PCB to be identified are within the tolerance range of the adjusted first-article formula parameters, and then the process proceeds to step S3 to perform the cutting operation. In step S5, if the cutting mechanism successfully cuts the PCB multilayer board to be identified according to the cutting parameters, the cutting mechanism can batch cut PCB multilayer boards of the same specification as the PCB multilayer board to be identified according to the first piece recipe parameters.

2. A PCB multilayer board slitting method based on machine vision according to claim 1, characterized in that: The PCB multilayer board to be identified is a PCB multilayer board with four board cores.

3. The method for slitting a multilayer PCB based on machine vision according to claim 2, characterized in that: In step S2, the image of the multilayer PCB to be identified is processed by an image processing algorithm to obtain cutting parameters of the multilayer PCB to be identified, specifically including: Step S21, performing a smoothing preprocessing operation on the PCB multilayer board image to be identified to remove noise on the PCB multilayer board image to be identified; Step S22, using an adaptive segmentation algorithm to segment the PCB multilayer board image to be identified after the smoothing preprocessing operation, so as to segment the core of the PCB multilayer board image to be identified and binarize the PCB multilayer board image to be identified with the core segmented, to form a segmented image; Step S23, making a minimum circumscribed rectangle for the board core on the segmented image, calculating the center point of the minimum circumscribed rectangle of the board core and the angle of the board core, and performing overall correction on the segmented image according to the angle of the board core to obtain a cropped image; Step S24: Compare the minimum circumscribed rectangles of the left and right cores on the cropped image to obtain the left and right sidelines of the vertical middle cropping area, and then calculate the position of the vertical middle cropping line based on the left and right sidelines; create another minimum circumscribed rectangle based on the minimum circumscribed rectangles of the left and right cores on the cropped image to obtain a first circumscribed rectangle of the left and right cores; compare the coordinates of the four vertices of the first circumscribed rectangle to obtain the sidelines of the left and right cropping areas; and move the sidelines of the left and right cropping areas to the left and right by a fixed distance, respectively, to obtain a left cropping line and a right cropping line. Step S25: Compare the minimum circumscribed rectangles of the upper and lower cores on the cropped image to obtain the upper and lower edges of the horizontal middle cropping area; then calculate the position of the horizontal middle cropping line based on the upper and lower edges; create another minimum circumscribed rectangle based on the minimum circumscribed rectangles of the upper and lower cores on the cropped image to obtain a second circumscribed rectangle of the upper and lower cores; compare the coordinates of the four vertices of the second circumscribed rectangle to obtain the edges of the upper and lower cropping areas; and move the edges of the upper and lower cropping areas upward and downward by a fixed distance, respectively, to obtain an upper cropping line and a lower cropping line. Step S26, perform nine-point calibration on the cut image through a manipulator and a camera to obtain the cutting parameters of the PCB multilayer board to be identified, wherein the cutting parameters include the length and width values ​​of the board core on the cut image, the number of board cores on the cut image, the thickness of the board core on the cut image, the coordinate value of the center point of the board core on the cut image, and the coordinate values ​​of the vertical middle cutting line, the horizontal middle cutting line, the upper cutting line, the lower cutting line, the left cutting line and the right cutting line on the cut image.

4. The method for slitting a multilayer PCB based on machine vision according to claim 1, characterized in that: The cutting mechanism is composed of a rolling cutter and / or a rolling cutter.

Images