Laser processing equipment, control method and system thereof and readable medium

Abstract

The invention discloses laser processing equipment, a control method and system thereof and a readable medium, and relates to the field of laser processing. The laser processing equipment control method comprises the following steps: acquiring first position information of a plurality of alignment patterns of a to-be-processed area on a first area image and actual position information of the plurality of alignment patterns; and according to a position mapping relationship between the first position information and the actual position information, correcting the to-be-processed area in the first area image to obtain a second area image, and processing based on the to-be-processed area in the second area image. The first position information is corrected by using the second position information to obtain the second area image, so that the accurate position of the to-be-processed target in the to-be-processed area can be clearly reflected, and the accuracy of pattern alignment is improved; therefore, the laser processing precision can be effectively improved when the to-be-processed area in the second area image is used for processing.

Description

Technical Field

[0001] The present application relates to the field of laser processing, and in particular to a laser processing device and a control method, system and readable medium thereof. Background Art

[0002] Laser processing is a process that uses the thermal effect or photochemical reaction between the laser beam and the material to complete laser welding, laser engraving, laser cutting, laser drilling and micromachining.

[0003] When laser processing equipment creates fine patterns on an object, it typically needs to accurately determine the location of the target area and then add the pattern to that location. This operation is known as pattern alignment. Existing pattern alignment techniques rely primarily on manual measurement and positioning, a method that is not only inefficient but also susceptible to human error, leading to inaccurate positioning and thus compromising the accuracy and effectiveness of laser processing. Summary of the Invention

[0004] The main purpose of this application is to provide a laser processing device and its control method, system and readable medium, aiming to improve the accuracy of alignment.

[0005] To achieve the above objectives, the present application provides a laser processing equipment control method, the laser processing equipment control method comprising: Acquire first position information of a plurality of alignment patterns in the area to be processed on the image of the first area and actual position information of the plurality of alignment patterns; According to the position mapping relationship between the first position information and the actual position information, the area to be processed in the first area image is corrected to obtain a second area image, and processing is performed based on the area to be processed in the second area image.

[0006] Optionally, the obtaining of first position information of a plurality of alignment patterns in the area to be processed on the image of the first area is specifically as follows: Acquire a first area image of a to-be-processed area, wherein the to-be-processed area has a plurality of the alignment patterns; Acquiring the pre-positions of the plurality of alignment patterns in the first area image based on a preset alignment model; Based on the pre-positions of the plurality of alignment patterns, adding a first preset circumscribed graphic on each alignment pattern; An area of ​​interest corresponding to a first preset circumscribed graphic of each alignment pattern is determined to determine first position information corresponding to the area of ​​interest.

[0007] Optionally, after obtaining the pre-positions of the plurality of alignment patterns in the first area image based on a preset alignment model, the method further includes: sorting the first circumscribed graphics of the plurality of alignment patterns according to a preset area rule to select the first circumscribed graphics with the smallest area difference; Then, the multiple first circumscribed graphics with the smallest area difference that are screened out are sorted according to a preset orientation, so that adjacent first circumscribed graphics are connected to each other to form a preset graphic constraint, so as to filter the alignment pattern in the first area image identified according to the preset alignment model.

[0008] Optionally, determining a region of interest corresponding to a first preset circumscribed graphic of each alignment pattern to determine first position information corresponding to the region of interest includes: Acquire a region of interest according to a first preset circumscribed pattern of each alignment pattern; Acquiring contour information in the region of interest; The first position information is acquired according to the contour information.

[0009] Optionally, the contour information includes a plurality of contours corresponding to the alignment pattern, and a hierarchical relationship between the plurality of contours; Acquiring the first position information according to the contour information includes: determining an innermost contour according to the contour information; A second preset circumscribed graphic is added to the innermost contour, and the center coordinates of the second preset circumscribed graphic are obtained to determine the first position information corresponding to the center coordinates.

[0010] Optionally, obtaining the pre-positions of the plurality of alignment patterns in the first area image includes: The pre-positions of the plurality of alignment patterns in the first area image are acquired based on a preset alignment model.

[0011] Optionally, the preset alignment model is generated by: Acquire a plurality of first area images including alignment patterns; performing mask annotation on the plurality of first region images to generate an annotated alignment pattern data set; Online data enhancement technology is applied to the alignment pattern dataset to set the flipping condition of the first area image, the hyperparameter batch and the number of iterations according to a preset ratio, and iterative training is performed to generate the preset alignment model.

[0012] Optionally, the alignment pattern includes a first alignment pattern and a plurality of second alignment patterns, and the first alignment pattern and the second alignment pattern are different pattern styles.

[0013] Optionally, the first alignment pattern is a circular target including a first number of concentric circles, and the second alignment pattern is a circular target including a second number of concentric circles; wherein the first number is not equal to the second number.

[0014] Optionally, before correcting the to-be-processed area in the first region image according to the position mapping relationship between the first position information and the actual position information to obtain the second region image, and processing according to the to-be-processed area in the second region image, the method further includes: The first area image including the plurality of alignment patterns is rotated so that the direction of the first area image is consistent with a first preset direction.

[0015] Optionally, the correcting the to-be-processed area in the first area image according to the position mapping relationship between the first position information and the actual position information to obtain the second area image is specifically as follows: The plurality of alignment patterns are perspective-transformed according to the plurality of first position information and the plurality of actual position information to obtain a second area image.

[0016] In addition, to achieve the above-mentioned purpose, the present application also provides a laser processing system, which includes: a memory, a processor, and a laser processing equipment control program stored on the memory and runnable on the processor, and the laser processing equipment control program is configured to implement the laser processing equipment control method.

[0017] In addition, to achieve the above-mentioned purpose, the present application also provides a laser processing device, which includes: A workbench, wherein the workbench has an area to be processed; a processing head, the processing head being slidably disposed on the workbench, and the processing head being used to process the workpiece on the area to be processed; A plurality of alignment patterns, wherein the plurality of alignment patterns are arranged at edge positions of the area to be processed, and lines connecting any two of the plurality of alignment patterns form a closed area; A camera device is provided on the workbench, and is used to capture a first area image of the area to be processed and obtain first position information of a plurality of alignment patterns on the first area image; an alignment device, the alignment device being mounted on the processing head and being used to obtain actual position information of a plurality of alignment patterns as the processing head moves; and A laser processing system is used to implement the laser processing equipment control method.

[0018] In addition, to achieve the above-mentioned purpose, the present application also provides a computer-readable medium on which a computer program is stored. When the computer program is executed by a processor, the control method of the laser processing equipment is implemented.

[0019] In the present application scheme, the first position information of multiple alignment patterns in the area to be processed on the first area image and the actual position information of multiple alignment patterns are first obtained, that is, the position information of multiple alignment patterns in the area to be processed on the image and the position coordinates in the actual laser processing equipment, and then according to the position mapping relationship between the first position information and the actual position information, the positions of the alignment patterns in the first area image and the actual alignment patterns are mapped to each other, so as to judge the deviation of the area to be processed in the first area image, so as to correct the area to be processed and obtain the corrected image, which is also the second area image. In this way, the second area image can clearly reflect the precise position of the target to be processed in the area to be processed, so as to improve the accuracy of pattern alignment, thereby reducing the inaccuracy caused by human factors and effectively improving the laser processing accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, without paying any creative labor, obtaining other drawings based on these drawings is also within the scope of protection of the present application.

[0021] Figure 1 This is a flow chart of a laser processing equipment control method according to an embodiment of the present application; Figure 2 This is a flow chart of a laser processing equipment control method according to another embodiment of the present application; Figure 3 This is a flow chart of a laser processing equipment control method according to another embodiment of the present application; Figure 4 This is a flow chart of a laser processing equipment control method according to another embodiment of the present application; Figure 5 This is a flow chart of a laser processing equipment control method according to another embodiment of the present application; Figure 6 This is a flow chart of a laser processing equipment control method according to another embodiment of the present application; Figure 7 This is a flow chart of a laser processing equipment control method according to another embodiment of the present application; Figure 8 This is a flow chart of a laser processing equipment control method according to another embodiment of the present application; Figure 9 This is a flow chart of a laser processing equipment control method according to another embodiment of the present application; Figure 10 is a schematic diagram of a first alignment pattern according to an embodiment of the present application; Figure 11 is a schematic diagram of a second alignment pattern according to an embodiment of the present application; Figure 12 A schematic diagram of a first alignment pattern according to another embodiment of the present application; Figure 13 is a schematic diagram of a second alignment pattern according to another embodiment of the present application; Figure 14 A schematic diagram of a first region image according to an embodiment of the present application; Figure 15 FIG. 1 is a schematic diagram of a second region image according to an embodiment of the present application.

[0022] The realization of the objectives, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments, and well-known modules, units and their connections, links, communications or operations are not shown or described in detail. In addition, the described features, architectures or functions can be combined in any way in one or more embodiments. It should be understood by those skilled in the art that the various embodiments described below are only for illustration and are not intended to limit the scope of protection of this application. It can also be easily understood that the modules or units or processing methods in each embodiment described herein and shown in the drawings can be combined and designed according to various different configurations. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0024] The limitations of various nouns or methods referred to in the following embodiments, except for those that are logically untenable, are generally based on the broad concepts that can be implemented under the premise of the disclosure in the embodiments. Under such an understanding, the various specific subordinate specific limitations of the nouns or methods should be regarded as the invention content of this application, and should not be understood narrowly or interpreted in a biased manner on the grounds that the specification does not disclose the specific limitations. Similarly, under the premise that it can be logically implemented, the order of the steps in the method is flexible and changeable, and the specific subordinate specific limitations in the broad concepts of various nouns or methods all fall within the scope of protection of this application.

[0025] Laser processing is a process that uses the thermal or photochemical reaction between the laser beam and the material to perform processes such as laser welding, laser engraving, laser cutting, laser drilling, and micromachining. When laser processing equipment creates fine patterns on an object, it typically requires accurate positioning of the object and then adding the pattern to that location. This process is known as alignment. Existing techniques typically use visual algorithms for alignment, but this method suffers from low accuracy.

[0026] To solve the above problems, this application proposes a laser processing equipment control method, referring to Figure 1 In one embodiment of the present application, the laser processing equipment control method includes steps S100-S200, wherein: S100, obtaining first position information of a plurality of alignment patterns in an area to be processed on an image of a first area and actual position information of the plurality of alignment patterns; S200 , correcting the area to be processed in the first area image according to a position mapping relationship between the plurality of first position information and the actual position information to obtain a second area image, and performing processing based on the area to be processed in the second area image.

[0027] It is understood that the laser processing equipment includes a laser processing head and a workbench, and the laser processing head can emit laser light to process the workpiece to be processed on the workbench. In this embodiment, the area to be processed is included in the area where the workbench is located, which can be a partial area or the entire area of ​​the workbench. The workpiece to be processed can be any material that requires laser processing, such as stone, wood, or metal. The shape of the workpiece to be processed can be a flat surface, a curved surface, or an irregular shape. The specific material and shape of the workpiece to be processed are not limited here. The alignment pattern can be a geometric shape or an irregular shape such as a circle, triangle, or rectangle. Multiple alignment patterns can be set at different positions in the area to be processed by the user by pasting, spraying, or other methods. Optionally, the alignment pattern is generally set at the edge of the area to be processed, and adjacent alignment patterns are connected to each other to form a closed area, generally in the form of a matrix, triangle, irregular quadrilateral, or other polygon. The first position information is the center coordinates of the alignment pattern in the area to be processed in a photo taken by a camera device, such as a mobile phone or an industrial camera, and the center coordinates include multiple alignment patterns.

[0028] It is understandable that the laser processing equipment includes a laser processing head, an alignment device and a workbench. The laser processing head can emit laser to process the consumables on the area to be processed. The alignment device is installed on the laser processing head so that it can move with the movement of the laser processing head. The alignment device is used to project a light spot for alignment. The shape of the light spot can be a cross-shaped light spot, a circular light spot, etc. Here, the alignment light spot is a cross-shaped light spot as an example for explanation. The alignment device is controlled to project the light spot, and then the laser processing head is controlled to move so that the intersection center point of the cross-shaped light spot projected by the alignment device coincides with the center point of the alignment pattern of the area to be processed. At this time, a command to record coordinates is sent to the laser processing equipment, so that the laser processing equipment can record the actual position information of the corresponding alignment pattern. The laser processing equipment is controlled to record the actual position information of the remaining alignment patterns in the same way until the actual position information of all alignment patterns is recorded.

[0029] In laser processing equipment, users can set up a shooting device, such as a mobile phone camera, an industrial camera, or a machine camera, to shoot the first area image of the area to be processed. Since the camera's shooting angle of view is a conical angle of view, the edge position often appears unbalanced stretching during the shooting process, just like someone is pulling the edge position of the area to be processed with force. This can easily lead to the problem of deviation in the first position information of multiple alignment patterns in the area to be processed on the first area image. Therefore, it is necessary to compare the coordinates of the alignment patterns in the real position, that is, the actual position information, and use the mapping relationship between the two to correct the first area image. Please refer to Figure 14 and Figure 15 As shown, after the image of the first area is generally taken out by the camera, there is a certain angle deviation in the alignment patterns of the four corners. Even if the camera is placed directly above the area to be processed, this problem will still exist. Therefore, with the help of the actual position information of multiple alignment patterns, the actual position information can be located by the laser positioning device or other positioning device on the laser processing equipment and input into the control device for one-to-one mapping. For example, Figure 14 The lower right corner is a solid circle target or two concentric circle targets, and the corresponding alignment pattern in the actual position information is also a solid circle target or two concentric circle targets. The other three positions are four concentric circle targets or three concentric circle targets. By using the circle targets for alignment mapping, the image of the processing area can be corrected, thereby ensuring the accuracy of laser processing. Figure 14In the first area of ​​the image, each alignment pattern appears in the form of different concentric circle targets. The precise position information of these targets in reality is captured by the laser processing equipment and matched with the position information in the image. Through this one-to-one mapping relationship, the deformation in the image can be accurately calculated, and then the image can be generated through perspective transformation, flipping, stretching and other means. Figure 15 The second area image shown here accurately corrects the position of the area to be processed. This process significantly reduces positioning inaccuracies caused by factors such as camera angle, equipment errors, or human operation, ensuring stable alignment in a variety of complex environments.

[0030] Optionally, in one implementation of the present embodiment, the alignment pattern includes a first alignment pattern and a plurality of second alignment patterns. The first alignment pattern may be a geometric figure or an irregular shape such as a circle, triangle, rectangle, etc., and the second alignment pattern may also be a geometric figure or an irregular shape such as a circle, triangle, rectangle, etc., but the shape and / or size of the second alignment pattern must be different from the first alignment pattern. Among them, the first alignment pattern and the second alignment pattern may both be circles, the first alignment pattern may be a solid circle, a single layer of concentric circles or multiple layers of concentric circles, and the second alignment pattern may be a solid circle, a single layer of concentric circles or multiple layers of concentric circles; the first alignment pattern and the second alignment pattern are different pattern styles. Among them, if Figure 12 As shown, a single layer of concentric circles refers to a solid circle with a ring outside, such as Figure 11 and Figure 13 As shown, the combination of a central solid circle and multiple rings constitutes a multi-layer concentric circle. Thus, by placing the first alignment pattern and multiple second patterns in a predetermined direction, sequence, or pattern on the processing area, the captured image of the first area, including the first and second alignment patterns, can provide not only positional information but also directional information, thereby facilitating recognition of the first area image and enhancing the robustness and accuracy of pattern alignment.

[0031] Optional, reference Figures 10 to 13 In another implementation of this embodiment, the first alignment pattern is a circular target comprising a first number of concentric circles, and the second alignment pattern is a circular target comprising a second number of concentric circles; wherein the first number and the second number are not equal. Optionally, the first alignment pattern may be a circular target comprising one concentric circle, or may be a circular target comprising two, three, four, or another number of concentric circles; the second alignment pattern may also comprise one, two, three, four, or another number of circular targets. The first and second alignment patterns are circular targets, and the specific number of concentric circles they each comprise is not limited herein; it is sufficient that the numbers of concentric circles they each comprise are unequal.

[0032] In this embodiment, the first alignment pattern and the second alignment pattern include circular targets with different numbers of concentric circles, so that by arranging the first alignment pattern and multiple second patterns in the area to be processed in a preset direction, sequence or graphic, the first area image including the first alignment pattern and the second alignment pattern captured by the camera device can not only provide position information and direction information, which is helpful for the recognition of the first area image, but also optimize the corresponding image recognition algorithm, which is helpful for quickly and accurately obtaining the first position information of the alignment pattern.

[0033] This embodiment obtains the first position information of multiple alignment patterns in the area to be processed on the first area image and the actual position information of multiple alignment patterns, that is, the position information of multiple alignment patterns in the area to be processed on the image and the position coordinates in the actual laser processing equipment, and then maps the positions of the alignment patterns in the first area image and the actual alignment patterns according to the position mapping relationship between the first position information and the actual position information, so as to judge the deviation of the area to be processed in the first area image, thereby correcting the area to be processed and obtaining the corrected image, which is also the second area image. In this way, the second area image can clearly reflect the precise position of the target to be processed in the area to be processed, so as to improve the accuracy of pattern alignment, thereby reducing the inaccuracy caused by human factors and effectively improving the laser processing accuracy.

[0034] Optional, reference Figure 2 In another embodiment of the present application, based on Figure 1 In the embodiment shown, first position information of a plurality of alignment patterns of the area to be processed on the image of the first area is obtained, specifically in steps S110-S140, wherein: S110, acquiring a first area image of an area to be processed; the area to be processed has a plurality of alignment patterns; S120, obtaining the pre-positions of the plurality of alignment patterns in the first area image; S130, adding a first preset circumscribed graphic to each alignment pattern based on the pre-positions of the plurality of alignment patterns; S140 : Determine a region of interest corresponding to a first preset circumscribed graphic of each alignment pattern, to determine first position information corresponding to the region of interest.

[0035] Optional, reference Figure 15 The first area image includes a first alignment pattern and a second alignment pattern; wherein the first alignment pattern is a circular target including two concentric circles, and the second alignment pattern is a circular target including four concentric circles, and the first alignment pattern is set to one, and the second alignment pattern is set to three, and the first alignment pattern and the second alignment pattern are set in a rectangular shape as a whole.

[0036] In this embodiment, the first area image can be captured by a mobile device with a camera function (such as a mobile phone, smartwatch, iPad, or industrial camera), or by a camera built into the laser processing equipment. After the first area image is captured, it can be processed directly on the mobile device or transmitted to a computer or cloud server via the mobile device for processing.

[0037] In this embodiment, a registration model can be deployed to identify the registration patterns in the first region image using the registration model to obtain the pre-positions of multiple registration patterns in the first region image. Alternatively, the first region image can be processed using an image processing algorithm to identify the approximate positions of the registration patterns as pre-positions. In this embodiment, the first registration pattern and the second registration pattern are circular targets with different numbers of concentric circles. Through the above steps, the approximate positions of the center coordinates of the circular targets in the first and second registration patterns can be well identified, i.e., the pre-positions of multiple registration patterns, which also constitutes a coarse positioning process for the multiple registration patterns.

[0038] In this embodiment, after the pre-position is determined, a first preset circumscribed graphic is added to each alignment pattern. This first circumscribed graphic can be a circumscribed circle, circumscribed rectangle, circumscribed triangle, or other graphic that differs from the alignment pattern. The first circumscribed graphic can mark the alignment pattern's border, making it easier to identify, track, or measure the alignment pattern. Furthermore, processing the first circumscribed graphic helps reduce the amount of data processed, thereby improving algorithm efficiency. Optionally, the first and second alignment patterns in this embodiment utilize circular targets with different numbers of concentric circles. Therefore, the first circumscribed graphic definition is added to each alignment pattern using a corresponding circumscribed rectangle.

[0039] In this embodiment, the alignment patterns within each alignment pattern are cropped according to the first preset circumscribed pattern, so as to obtain the region of interest.

[0040] It's understood that in the field of image processing, a region of interest (ROI) is a region selected from an image that serves as the focus of image analysis. This region is then delineated for further processing. Using the ROI to delineate a target can reduce processing time and increase accuracy. In this embodiment, the alignment pattern is delineated using a first preset external graphic, so that the ROI selects an area within the alignment pattern for image analysis.

[0041] In this embodiment, during the processing of the region of interest, the Otsu adaptive threshold algorithm can be used to automatically select the optimal threshold value and perform binarization processing on the region of interest, converting the region of interest into a binary image to better identify the contour information of the alignment pattern. A morphological closing operation can also be performed first to connect the component parts of the object and fill the object's holes, and then a morphological opening operation can be performed to open up the object's areas, separate the object, and maintain the general shape. Through morphological closing and opening operations, holes in the contour can be filled and small noise can be removed, resulting in a smoother and more accurate contour, thereby obtaining more accurate first position information.

[0042] It should be understood that the contour information includes multiple contours corresponding to the alignment pattern and the hierarchical relationship between the multiple contours. The contour information in the region of interest is obtained using the OpenCV-based findContours method to determine the hierarchical relationship between the multiple contours. Furthermore, when the selected alignment pattern is a circular target consisting of two concentric circles, the contour information refers to the two concentric circles in the alignment pattern and the inclusion relationship between the two concentric circles.

[0043] Optionally, refer to Figure 4 In another embodiment of the present application, based on Figure 2 In the embodiment shown, determining a region of interest corresponding to a first preset circumscribed graphic of each alignment pattern to determine first position information corresponding to the region of interest includes steps S141-S143, wherein: S141, obtaining a region of interest according to a first preset circumscribed pattern of each alignment pattern; S142, obtaining contour information in the region of interest; S143. Acquire the first position information according to the contour information.

[0044] In this embodiment, the alignment pattern is circled by using the first preset circumscribed pattern, so that a region of interest (ROI) selects an area of ​​the alignment pattern from the first preset circumscribed pattern for image analysis.

[0045] In this embodiment, the contour information in the region of interest can be obtained based on the findContours method of OpenCV. The contour information includes multiple contours corresponding to the alignment pattern. When the selected alignment pattern is a circular target including two concentric circles, the contour information is the two concentric circles in the alignment pattern, and the center coordinate position of the contour information is obtained based on the contour information. The coordinate position can be used as the first position information.

[0046] In addition, the region of interest can be subjected to image enhancement processing, converted into a grayscale image, and Gaussian filtering. Through the above processing, the quality of the region of interest can be improved, noise and interference can be reduced, and the contour information of the alignment pattern can be identified more accurately.

[0047] In addition, the acquired contours can be screened to improve the accuracy of identifying the alignment pattern. The screening process can exclude contours with a circularity less than 0.8, exclude contours with an area less than 0.2 times the area of ​​the region of interest, and exclude contours whose area to the area of ​​the contour's circumscribed rectangle does not satisfy contour_area / contour_bun_area>(π / 4)+0.1, where contour_area is the area of ​​the contour and contour_bun_area is the area of ​​the contour's circumscribed rectangle.

[0048] Based on the contours corresponding to the above alignment patterns and the hierarchical relationship between the contours, refer to Figure 5 In another embodiment of the present application, based on Figure 4 In the illustrated embodiment, obtaining the first position information according to the contour information includes steps S1431-S1432, wherein: S1431, determining the innermost contour according to the contour information; S1432: Add a second preset circumscribed graphic to the innermost contour, and obtain the center coordinates of the second preset circumscribed graphic to determine the first position information corresponding to the center coordinates.

[0049] In this embodiment, when the selected alignment pattern is a circular target including two concentric circles, the contour information is the two concentric circles in the alignment pattern and the inclusion relationship between the two concentric circles. Therefore, the smaller circle is selected as the innermost contour.

[0050] In this embodiment, the second preset circumscribed image can be a circumscribed circle, rectangle, triangle, or other figure that differs from the alignment pattern. This second circumscribed image can mark the alignment pattern's border, making it easier to identify, track, or measure it. Furthermore, processing the second preset circumscribed image helps reduce the amount of data processed, thereby improving algorithm efficiency. In this embodiment, since the alignment pattern uses a concentric circle target, the second preset circumscribed image for the innermost contour is a rectangle circumscribing the innermost contour of the concentric circle target.

[0051] In this embodiment, the coordinates of the center of the second preset circumscribed figure are used as the center coordinates of the alignment pattern to obtain the first position. By using the coordinates of the center of the second preset circumscribed figure as the center coordinates of the alignment pattern, the deviation between the center of mass of the innermost contour and the actual center can be effectively reduced, thereby improving alignment accuracy. Similarly, when the selected alignment pattern is a circular target consisting of four concentric circles, a circular target consisting of one concentric circle, or a circular target consisting of any other number of concentric circles, the first position information of the alignment pattern can be obtained using the above scheme.

[0052] It should be understood that the first area image captured by a mobile device with a camera function (such as a mobile phone, smart watch, iPad, etc.) or by the camera built into the laser processing equipment may contain other patterns in addition to the alignment pattern, which may affect the accuracy of the alignment pattern recognition.

[0053] To solve the above problems, refer to Figure 3 In another embodiment of the present application, based on Figure 2 In the embodiment shown, after obtaining the pre-positions of the plurality of alignment patterns in the first region image based on the preset alignment model, the process further includes steps S150-S160, wherein: S150, sorting the first circumscribed graphics of the plurality of alignment patterns according to a preset area rule to select the first circumscribed graphics with the smallest area difference; S160 , sorting the plurality of first circumscribed graphics with the smallest area difference according to a preset orientation, so that adjacent first circumscribed graphics are connected to each other to form a preset graphic constraint, so as to filter the alignment pattern in the first region image identified according to the preset alignment model.

[0054] In this embodiment, filtering the alignment pattern in the first area image identified according to the preset alignment model can exclude other patterns in the first area image except the alignment pattern, thereby avoiding interference with the recognition of the alignment pattern, thereby improving the accuracy of alignment pattern recognition.

[0055] The first area image includes a first alignment pattern and a second alignment pattern; the first alignment pattern is a circular target including two concentric circles, and the second alignment pattern is a circular target including four concentric circles, and the first alignment pattern is set as one and the second alignment pattern is set as three, and the first alignment pattern and the second alignment pattern are arranged in a rectangular shape as a whole. Therefore, the order is first performed according to the area rule, and then the circumscribed rectangles with an aspect ratio less than the first preset aspect ratio and the circumscribed rectangles with an aspect ratio greater than the second preset aspect ratio are excluded. The specific values ​​of the first preset aspect ratio and the second preset aspect ratio can be flexibly selected by the user. In this embodiment, the value of the first aspect ratio is 0.8, and the value of the second aspect ratio is 1.2.

[0056] The circumscribed rectangles are arranged according to their areas to select four circumscribed rectangles with the smallest area difference, and one of the circumscribed rectangles has a circular target including two concentric circles, and the other three circumscribed rectangles have circular targets including four concentric circles.

[0057] After screening according to the area rule, the center coordinates of the circumscribed rectangle of the circular target and the coordinates of the four vertices of the machinable area are obtained respectively. The four vertex coordinates correspond to the coordinates of the upper left end, the upper right end, the lower right end and the lower left end of the area to be processed respectively. According to the sorting requirements of the preset orientation, the center coordinates of the circumscribed rectangle are compared with the coordinates of the four vertices of the machinable area, and they are arranged according to the position of the circumscribed rectangle. The circular targets closest to the upper left end, upper right end, lower right end and lower left end of the area to be processed are selected respectively.

[0058] Then, the centers of adjacent circumscribed rectangles are connected to obtain four vectors. Based on the preset graphic constraints, the circumscribed rectangles corresponding to adjacent vectors with an angle less than a first preset angle and those corresponding to adjacent vectors with an angle greater than a second preset angle are excluded. The first and second preset angles can be flexibly selected by the user. In this embodiment, the first preset angle is 60 degrees, and the second preset angle is 120 degrees.

[0059] In addition, when the first alignment pattern and / or the second alignment pattern cannot be normally acquired after filtering by the third preset rule, and a circular target consisting of two concentric circles and / or three circular targets consisting of four concentric circles cannot be acquired, the user is reminded to reacquire the first area image.

[0060] Optionally, refer to Figure 6 In another embodiment of the present application, based on Figure 2 In the embodiment shown, the step of obtaining the pre-positions of the plurality of alignment patterns in the first region image includes step S121, which is executed after step S110 and before step S130, wherein: S121 : Acquire preliminary positions of the plurality of alignment patterns in the first region image based on a preset alignment model.

[0061] In this embodiment, the preset alignment model can be a Pytorch model capable of recognizing alignment patterns, obtained by training a plurality of first region images including alignment patterns using a convolutional neural network. The preset alignment model can be pre-saved by the user in the laser processing equipment, or can be stored in a remote cloud, where the user calls the preset alignment model to recognize the alignment pattern in the first region image. The preset alignment model can also be temporarily saved on a mobile device, where the user selects the corresponding preset alignment model to process the first region image as needed. The preset alignment model can be used to identify the pre-positions of the plurality of alignment patterns in the first region image, where the pre-positions are the approximate positions of the alignment patterns in the first region image.

[0062] In this embodiment, an alignment model is deployed to identify alignment patterns in the first region image, thereby obtaining the pre-positions of multiple alignment patterns in the first region image. This allows for the coarse positions of the center coordinates of the circular targets in the first and second alignment patterns to be accurately identified, i.e., the pre-positions of the multiple alignment patterns, thus achieving coarse positioning of the multiple alignment patterns.

[0063] Optionally, refer to Figure 7 In another embodiment of the present application, based on Figure 6 In the embodiment shown, the generation method of obtaining the preset alignment model is specifically steps S1211-S1213, wherein: S1211, acquiring a plurality of first area images including alignment patterns; S1212, performing mask annotation on the plurality of first region images to generate an annotated alignment pattern data set; S1213. Apply online data enhancement technology to the alignment pattern dataset to set the flipping condition of the first area image, hyperparameter batch and number of iterations according to a preset ratio, and perform iterative training to generate the preset alignment model.

[0064] In this embodiment, in order to better identify the alignment pattern in the first area image, a preset alignment model is used. The preset alignment model can be one that has been generated and stored in the laser processing equipment, stored in a computer, or stored in a cloud server, and is run in response to user instructions to identify the alignment pattern of the first area image.

[0065] Alternatively, training and generation can be performed in a control device. During this generation process, a registration pattern dataset is first acquired. This registration pattern dataset is obtained by masking a predetermined number of images of the first region. Masking is a technique for marking specific regions within an image, which helps identify the target object within the image and distinguish it from other parts.

[0066] In this embodiment, the first area image can be captured by a mobile device with a camera function (such as a mobile phone, smart watch, iPad, etc.), or by a camera built into the laser processing equipment. Optionally, the first area image can also be directly imported.

[0067] In this embodiment, the greater the number of first-region images, the more accurate the alignment model obtained through subsequent convolutional neural network training. The user can flexibly set the preset number based on the desired accuracy of the preset alignment model. Furthermore, the higher the resolution of the first-region images, the more accurate the alignment model obtained through subsequent convolutional neural network training. The resolution of the first-region images can also be flexibly selected based on user needs.

[0068] Optionally, in this embodiment, 500 first area images with a resolution of 1600x900 can be obtained by taking pictures with a mobile phone, and then the 500 first area images with a resolution of 1600x900 can be masked and annotated to obtain the alignment pattern dataset, and then the alignment pattern dataset can be divided into a training set, a verification set and a test set.

[0069] In this embodiment, after obtaining the alignment pattern data set, the data set is trained using a convolutional neural network to obtain a corresponding preset alignment model.

[0070] During the training process, this example sets the training batch size to 16, the number of training iterations to 250, and the confidence level to a hyperparameter of 0.5. The ratio of images flipped upside down and left-right is set to 50%, and the random rotation angle range is set to 0-360 degrees. Iterative training is then performed based on the alignment pattern dataset, i.e., the divided training, validation, and test sets, to obtain the preset alignment model.

[0071] In this example, a convolutional neural network is used for training to obtain a PyTorch model capable of recognizing alignment patterns. This example can employ a variety of convolutional neural network architectures for training, such as YOLOv8, Faster R-CNN, SSD, and Mask R-CNN. This example uses YOLOv8 as an example for training; other convolutional neural network architectures are trained in a similar manner. During training, the model parameters are continuously optimized to achieve greater accuracy and stability in recognizing alignment patterns.

[0072] In this embodiment, training a dataset using a convolutional neural network can improve the accuracy and robustness of the alignment model. During training, hyperparameters such as batch size and number of iterations can be adjusted based on actual conditions to achieve better model performance. Furthermore, various optimization algorithms, such as Adam and RMSProp, can be used to accelerate model convergence and improve model accuracy.

[0073] Optional, reference Figure 8 In another embodiment of the present application, based on Figures 1 to 7 In any of the illustrated embodiments, before correcting the to-be-processed area in the first area image based on the position mapping relationship between the first position information and the actual position information to obtain a second area image, and performing processing according to the to-be-processed area in the second area image, after obtaining first position information of a plurality of alignment patterns of the to-be-processed area on the first area image and actual position information of the plurality of alignment patterns, the step further includes step S300, which is executed after step S100 and before step S200, wherein: S300: Rotate a first region image including a plurality of alignment patterns so that a direction of the first region image is consistent with a first preset direction.

[0074] Optionally, in this embodiment, the first area image includes a first alignment pattern and a second alignment pattern; wherein the first alignment pattern is a circular target including a solid circle (such as Figure 10 As shown in FIG), the second alignment pattern is a circular target including three concentric circles (as shown in FIG). Figure 11 As shown), the first alignment pattern is set as one, the second alignment pattern is set as three, and the arrangement of the first alignment pattern and the second alignment pattern is arranged in a rectangular shape as a whole.

[0075] Optionally, in this embodiment, the first area image includes a first alignment pattern and a second alignment pattern; wherein the first alignment pattern is a circular target including two concentric circles (such as Figure 12 As shown in FIG), the second alignment pattern is a circular target including four concentric circles (as shown in FIG). Figure 13As shown), the first alignment pattern is set as one, the second alignment pattern is set as three, and the arrangement of the first alignment pattern and the second alignment pattern is arranged in a rectangular shape as a whole. Figure 10 and Figure 11 The circular target, Figure 12 and Figure 13 The center of the circular target is marked, and the area of ​​the framed center area is smaller, which can make the determination process of the first position more accurate and reliable. In this way, in steps S161 to S162, the center coordinates of the second preset circumscribed figure obtained are more accurate.

[0076] In this embodiment, Figure 15 As shown, the first preset direction is the direction of the first region image when the first alignment pattern is at the lower right end. If the direction of the first region image is inconsistent with the first preset direction, the first region image can be rotated 90 degrees, 180 degrees, or 270 degrees to align the direction of the first region image with the first preset direction. By adjusting the direction of the first region image, the position and direction of the alignment pattern in the image can be ensured to be consistent, thereby improving alignment accuracy and stability.

[0077] In this embodiment, first, the first position information of multiple alignment patterns in the area to be processed is obtained. Then, based on the multiple first position information and the actual position information corresponding to the multiple alignment patterns, a corresponding position mapping relationship is generated or determined. The actual position information is the position of the multiple alignment patterns within the laser processing equipment. Based on this position mapping relationship, the target area on the workpiece to be processed is then determined. This allows the precise position of the target area within the laser processing equipment to be determined, thereby improving the accuracy of pattern alignment. Adding the pattern to be processed to the target area for laser processing can effectively improve laser processing accuracy.

[0078] Optional, reference Figure 9 In another embodiment of the present application, based on Figure 1 In the embodiment shown, the to-be-processed area in the first area image is corrected according to the position mapping relationship between the first position information and the actual position information to obtain the second area image, specifically in step S210, wherein: S210: Perform perspective transformation according to the plurality of first position information and the plurality of actual position information to obtain a second area image.

[0079] Perspective transformation is a method for converting an image from one view to another based on the geometric relationship between the two views. In this embodiment, perspective transformation is used to transform the representation of the area to be processed in the first region image to its representation in the second region image based on the relationship between the first position information and the actual position information. This transformation enables a more accurate representation of the area to be processed within the coordinate system of the laser processing equipment. This allows the precise position of the alignment pattern within the laser processing equipment to be determined, thereby improving alignment accuracy. Furthermore, the application of perspective transformation eliminates errors caused by shooting angle, significantly improving alignment precision.

[0080] In addition, to achieve the above-mentioned purpose, the present application also provides a laser processing system, which includes: a memory, a processor, and a laser processing equipment control program stored on the memory and runnable on the processor, and the laser processing equipment control program is configured to implement the laser processing equipment control method.

[0081] It is worth noting that since the laser processing system of the present application is based on the above-mentioned laser processing equipment control method, the embodiments of the control device of the present application include all technical solutions of all embodiments of the above-mentioned laser processing equipment control method, and the technical effects achieved are also exactly the same, which will not be repeated here.

[0082] It is understood that a laser processing system may include a laser processing device and a control terminal. The laser processing device may be provided with the control terminal, and the laser processing device may directly run a control program and perform laser processing. Alternatively, the laser processing device may be connected to the control terminal via a communication link, with the control terminal processing and sending control commands to the laser processing device, which then performs laser processing.

[0083] In addition, to achieve the above-mentioned purpose, the present application further provides a laser processing device, which includes a workbench, a processing head, a plurality of alignment patterns, a camera device, an alignment device, and a laser processing system, wherein: The workbench has an area to be processed; the processing head is slidably arranged on the workbench, and the processing head is used to process the workpiece to be processed on the area to be processed; multiple alignment patterns are arranged at the edge position of the area to be processed, and the lines between multiple alignment patterns constitute a closed area; the camera device is arranged on the workbench, and the camera device is used to shoot the first area image of the area to be processed and obtain the first position information of multiple alignment patterns on the first area image; the alignment device is installed on the processing head, and as the processing head moves, the alignment device is used to obtain the actual position information of multiple alignment patterns; the laser processing system is used to implement the laser processing equipment control method described in all the above embodiments, and the laser processing equipment control method corrects the area to be processed in the first area image according to the position mapping relationship between the first position information and the actual position information to obtain the second area image; the control device is also used to control the processing head to process the workpiece to be processed based on the area to be processed in the second area image.

[0084] In this embodiment, the processing head can be a laser processing head, and the laser processing head can emit laser to process the workpiece to be processed on the workbench. In this embodiment, the area to be processed can be a partial area or the entire area of ​​the workbench. The workpiece to be processed can be any material that requires laser processing, such as stone, wood or metal. The shape of the workpiece to be processed can be a plane, a curved geometric shape or an irregular shape, etc. The specific material and shape of the workpiece to be processed are not limited here. The alignment pattern can be a geometric shape or an irregular shape such as a circle, a triangle, a rectangle, etc., and multiple alignment patterns can be set at different positions in the area to be processed by the user by pasting, spraying or other methods. Optionally, the alignment pattern is generally set at the edge position of the area to be processed, and adjacent alignment patterns are connected to each other in pairs to form a closed area, generally a matrix, a triangle or an irregular quadrilateral, or other polygons. The first position information is the center coordinates of the alignment pattern in the area to be processed in a photo taken by a shooting device, such as a mobile phone or an industrial camera, and the center coordinates include multiple alignment patterns.

[0085] It is understood that the alignment device is mounted on the laser processing head so that it can move with the movement of the laser processing head. The alignment device is used to project a light spot for alignment, and the light spot shape can be a cross-shaped light spot, a circular light spot, etc. Here, the alignment light spot is a cross-shaped light spot as an example for explanation. The alignment device is controlled to project the light spot, and then the laser processing head is controlled to move so that the intersection center point of the cross-shaped light spot projected by the alignment device coincides with the center point of the alignment pattern in the area to be processed. At this time, a command to record coordinates is sent to the laser processing equipment, which can cause the laser processing equipment to record the actual position information of the corresponding alignment pattern. The laser processing equipment is controlled to record the actual position information of the remaining alignment patterns in the same way until the actual position information of all alignment patterns is recorded.

[0086] In the laser processing equipment, the user can set up a shooting device, such as a mobile phone camera, an industrial camera, or a machine camera, to shoot an image of a first area of ​​the area to be processed.

[0087] It is worth noting that since the laser processing equipment of the present application is based on the above-mentioned laser processing system, the embodiments of the laser processing equipment of the present application include all technical solutions of all embodiments of the above-mentioned laser processing system, and the technical effects achieved are also exactly the same. Other identical contents will not be repeated here.

[0088] In addition, to achieve the above-mentioned purpose, the present application also provides a computer-readable medium on which a computer program is stored. When the computer program is executed by a processor, the control method of the laser processing equipment described in all the above embodiments is implemented.

[0089] It is worth noting that since the computer-readable medium of the present application is based on the above-mentioned laser processing system, the embodiments of the computer-readable medium of the present application include all technical solutions of all embodiments of the above-mentioned laser processing system, and the technical effects achieved are also exactly the same, which will not be repeated here.

[0090] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0091] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0092] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present application, or the part that contributes to the existing technology, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0093] The above are only preferred embodiments of the present application and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A laser processing equipment control method, characterized in that: The laser processing equipment control method includes: Acquire first position information of a plurality of alignment patterns in the area to be processed on the image of the first area and actual position information of the plurality of alignment patterns; According to the position mapping relationship between the first position information and the actual position information, the area to be processed in the first area image is corrected to obtain a second area image, and processing is performed based on the area to be processed in the second area image.

2. The laser processing equipment control method according to claim 1, wherein: The step of obtaining first position information of a plurality of alignment patterns in the area to be processed on the image of the first area is specifically as follows: Acquire a first area image of a to-be-processed area, wherein the to-be-processed area has a plurality of the alignment patterns; Acquiring the pre-positions of the plurality of alignment patterns in the first area image; Based on the pre-positions of the plurality of alignment patterns, adding a first preset circumscribed graphic on each alignment pattern; An area of ​​interest corresponding to a first preset circumscribed graphic of each alignment pattern is determined to determine first position information corresponding to the area of ​​interest.

3. The laser processing equipment control method according to claim 2, wherein: After obtaining the pre-positions of the plurality of alignment patterns in the first area image based on the preset alignment model, the method further includes: sorting the first circumscribed graphics of the plurality of alignment patterns according to a preset area rule to select the first circumscribed graphics with the smallest area difference; Then, the multiple first circumscribed graphics with the smallest area difference that are screened out are sorted according to a preset orientation, so that adjacent first circumscribed graphics are connected to each other to form a preset graphic constraint, so as to filter the alignment pattern in the first area image identified according to the preset alignment model.

4. The laser processing equipment control method according to claim 2, wherein: Determining a region of interest corresponding to a first preset circumscribed graphic of each alignment pattern to determine first position information corresponding to the region of interest includes: Acquire a region of interest according to a first preset circumscribed pattern of each alignment pattern; Acquiring contour information in the region of interest; The first position information is acquired according to the contour information.

5. The laser processing equipment control method according to claim 4, wherein: The contour information includes a plurality of contours corresponding to the alignment pattern, and a hierarchical relationship between the plurality of contours; Acquiring the first position information according to the contour information includes: determining an innermost contour according to the contour information; A second preset circumscribed graphic is added to the innermost contour, and the center coordinates of the second preset circumscribed graphic are obtained to determine the first position information corresponding to the center coordinates.

6. The laser processing equipment control method according to claim 2, wherein: The obtaining of the pre-positions of the plurality of alignment patterns in the first area image includes: The pre-positions of the plurality of alignment patterns in the first area image are acquired based on a preset alignment model.

7. The laser processing equipment control method according to claim 6, wherein: The generation method of obtaining the preset alignment model is specifically as follows: Acquire a plurality of first area images including alignment patterns; performing mask annotation on the plurality of first region images to generate an annotated alignment pattern data set; Online data enhancement technology is applied to the alignment pattern dataset to set the flipping condition of the first area image, the hyperparameter batch and the number of iterations according to a preset ratio, and iterative training is performed to generate the preset alignment model.

8. The laser processing equipment control method according to claim 1, wherein: The alignment pattern includes a first alignment pattern and a plurality of second alignment patterns, and the first alignment pattern and the second alignment pattern are different pattern styles.

9. The laser processing equipment control method according to claim 8, wherein: The first alignment pattern is a circular target including a first number of concentric circles, and the second alignment pattern is a circular target including a second number of concentric circles; wherein the first number is not equal to the second number.

10. The laser processing equipment control method according to any one of claims 1 to 9, characterized in that: Before correcting the to-be-processed area in the first area image according to the position mapping relationship between the first position information and the actual position information to obtain the second area image and processing according to the to-be-processed area in the second area image, the method further includes: The first area image including the plurality of alignment patterns is rotated so that the direction of the first area image is consistent with a first preset direction.

11. The laser processing equipment control method according to claim 1, wherein: The correcting the to-be-processed area in the first area image according to the position mapping relationship between the first position information and the actual position information to obtain the second area image is specifically as follows: The plurality of alignment patterns are perspective-transformed according to the plurality of first position information and the plurality of actual position information to obtain a second area image.

12. A laser processing system, characterized in that: The laser processing system includes: a memory, a processor, and a laser processing equipment control program stored in the memory and executable on the processor, wherein the laser processing equipment control program is configured to implement the laser processing equipment control method according to any one of claims 1 to 11.

13. A laser processing device, characterized in that: The laser processing equipment includes: A workbench, wherein the workbench has an area to be processed; a processing head, the processing head being slidably disposed on the workbench, and the processing head being used to process the workpiece on the area to be processed; A plurality of alignment patterns, wherein the plurality of alignment patterns are arranged at edge positions of the area to be processed, and lines connecting any two of the plurality of alignment patterns form a closed area; A camera device is provided on the workbench, and is used to capture a first area image of the area to be processed and obtain first position information of a plurality of alignment patterns on the first area image; an alignment device, the alignment device being mounted on the processing head and being used to obtain actual position information of a plurality of alignment patterns as the processing head moves; and A laser processing system, wherein the laser processing system is used to implement the laser processing equipment control method according to any one of claims 1 to 11.

14. A computer-readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the control method of the laser processing equipment according to any one of claims 1 to 11 is implemented.

Images