Intelligent detection system and detection method for press line ejector rod standby mold

Abstract

The invention provides an intelligent detection system and detection method for a press line top rod standby mold. The system comprises a projector and a visual unit which are installed on a protector. The ejector rod is arranged on the movable workbench; the ejector rod is arranged in a hole in the workbench; the projector is used for projecting an ejector rod preset arrangement diagram corresponding to the current mold and mold identification information to the table top of the movable workbench and is used for guiding the arrangement of the ejector rods; the visual unit is used for collecting images of the movable workbench after the ejector rods are arranged and judging whether the actual arrangement of the ejector rods is consistent with the standard arrangement of the mold preset by the system or not in an image recognition mode. Based on the system, the invention further provides an intelligent detection method for the press line ejector rod standby mold. According to the invention, the problems of difficult identification and low efficiency during the arrangement of the existing press line ejector rod standby mold are solved, the rapid replacement of different ejector rod arrangements of different molds is realized, the error rate of manually replacing the ejector rods is reduced, and the efficiency and accuracy of replacing the ejector rods of different molds are improved.

Description

Technical Field

[0001] The invention belongs to the technical field of punching line ejector bar standby die detection, and in particular relates to an intelligent detection system and a detection method for a punching line ejector bar standby die. Background Art

[0002] Due to the wide variety of stamping materials and dies, manual identification and placement of ejector pins on the mobile workbench is required. Each mobile workbench has multiple ejector pin placement holes, and each time the sheet metal is changed, the ejector pin arrangement for each die is different.

[0003] The existing manual replacement of ejector pins is done according to the layout diagram and the numbers on the mobile workbench. Due to the wide variety of sheet materials, the layout diagram is complex and changeable, there is a high error rate, and the replacement efficiency is low. After the ejector pins are arranged, the visual unit is used for image recognition to check whether the ejector pin arrangement is correct. However, because visual recognition is limited by light, ejector pin size, etc., there are certain errors in image recognition. Therefore, the existing ejector pin preparation method limits the speed of replacing different ejector pins in the stamping line, as well as the success rate of ejector pin arrangement inspection, affecting the efficiency of stamping processing of different sheet metal parts. Summary of the Invention

[0004] To address the above technical issues, the present invention proposes an intelligent detection system and method for press line ejector bar spare molds. This system enables rapid replacement of different ejector bar arrangements in different molds, significantly reducing the error rate of manual ejector bar replacement and improving the efficiency and accuracy of ejector bar replacement for different molds.

[0005] To achieve the above object, the present invention adopts the following technical solutions: An intelligent detection system for a punching line ejector die, comprising: a projector 2 and a visual unit 3 mounted on a guard 1; and an ejector 5 disposed on a movable workbench 4; the ejector 5 being arranged in a hole 8 on the movable workbench 4; The projector 2 is used to project the preset layout diagram of the ejector rods 5 corresponding to the current mold and the mold identification information onto the surface of the movable workbench 4 for visually guiding the layout of the ejector rods 5; The visual unit 3 is used to collect an image of the movable workbench 4 after the ejector rods 5 are arranged, and to determine whether the actual arrangement of the ejector rods is consistent with the standard arrangement of the mold preset by the system by using image recognition.

[0006] Furthermore, a first mounting bracket 6 and a second mounting bracket 7 are provided on the protection 1; The projector 2 is fixedly mounted on the first mounting bracket 6; The visual unit 3 is fixedly mounted on the second mounting bracket 7 .

[0007] Furthermore, the specific light field projected by the projector 2 includes high-contrast grid lines, uniform background light or structured light patterns, which are used to enhance the distinction between the top rod 5 and the background of the hole 8.

[0008] Furthermore, the projector 2 dynamically adjusts projection parameters according to the actual stop position of the movable workbench 4 to compensate for geometric distortion of the projected image caused by positioning error of the movable workbench 4 .

[0009] Furthermore, before performing image recognition, the visual unit 3 also includes: preprocessing the collected image of the moving workbench 4 after the top rod 5 is arranged; wherein the preprocessing process includes distortion correction, denoising, contrast enhancement and binarization processing.

[0010] Furthermore, the process of performing image recognition by the visual unit 3 includes: Identify the reference points on the workbench surface, establish an accurate grid coordinate system of holes 8, and calculate the theoretical center coordinates of each hole 8; Perform connected domain analysis on the pre-processed image, extract the connected domain that meets the preset top rod size and shape characteristics, and calculate its centroid coordinates; Traverse the theoretical center coordinates of each hole 8, search for the center of mass of the connected domain of the push rod within the preset search radius, determine whether the state of the hole 8 is "with push rod" or "without push rod", and generate the actual push rod arrangement bitmap; The actual ejector pin arrangement bitmap is compared bit by bit with the preset current mold standard ejector pin arrangement bitmap.

[0011] Furthermore, the preset search radius is set to be smaller than half of the distance between adjacent holes 8 .

[0012] Furthermore, the system also includes an alarm module; The alarm module is used to receive the determination signal output by the visual unit 3, trigger an audible and visual alarm when it is determined that the arrangement is wrong, and prevent the mobile workbench 4 from returning to the production line or subsequent mold change operations.

[0013] The present invention also proposes an intelligent detection method for a punching line ejector die, comprising the following steps: The projector 2 is used to project the preset layout diagram of the ejector rods 5 corresponding to the current mold and the mold identification information onto the surface of the movable workbench 4 for visual guidance of the layout of the ejector rods 5; The visual unit 3 is used to collect the image of the movable workbench 4 after the ejector rods 5 are arranged, and the image recognition method is used to determine whether the actual arrangement of the ejector rods is consistent with the standard arrangement of the mold preset by the system.

[0014] Furthermore, the image recognition process includes: Identify the reference points on the workbench surface, establish an accurate grid coordinate system of holes 8, and calculate the theoretical center coordinates of each hole 8; Perform connected domain analysis on the pre-processed image, extract the connected domain that meets the preset top rod size and shape characteristics, and calculate its centroid coordinates; Traverse the theoretical center coordinates of each hole 8, search for the center of mass of the connected domain of the push rod within the preset search radius, determine whether the state of the hole 8 is "with push rod" or "without push rod", and generate the actual push rod arrangement bitmap; The actual ejector pin arrangement bitmap is compared bit by bit with the preset current mold standard ejector pin arrangement bitmap.

[0015] The effects provided in the summary of the invention are only the effects of the embodiments, not all the effects of the invention. One of the above technical solutions has the following advantages or beneficial effects: The present invention proposes an intelligent detection system and method for ejector pin preparation molds on a press line. The system includes a projector and a visual unit mounted on a guard; an ejector pin disposed on a movable workbench; the ejector pin is arranged in a hole on the movable workbench; the projector is used to project a preset ejector pin layout diagram and mold identification information corresponding to the current mold onto the surface of the movable workbench, which is used to visually guide the placement of the ejector pins; The visual unit is used to collect images of the mobile workbench after the ejector pins are arranged, and to use image recognition to determine whether the actual arrangement of the ejector pins is consistent with the standard arrangement of the mold preset by the system. Based on an intelligent detection system for the ejector pin standby mold of a press line, the present invention also proposes an intelligent detection method for the ejector pin standby mold of a press line. The present invention can solve the technical problems of difficulty in identifying and low efficiency in the arrangement of the ejector pin standby mold of the existing press line, realize the rapid replacement of different ejector pin arrangements for different molds, greatly reduce the error rate of manual replacement of ejector pins, and improve the efficiency and accuracy of replacement of ejector pins for different molds.

[0016] The projector in the present invention has a certain ability to automatically adjust and adapt to projection distortion to ensure that the projection image is not distorted, thereby guiding manual placement of ejector rods in holes on the mobile workbench according to the projection layout diagram, reducing the error rate of manual ejector rod mold layout and improving replacement efficiency.

[0017] In this invention, the vision unit takes a picture of the push rod arrangement on the mobile workbench, transmits the image back for image recognition, and performs intelligent inspection of the push rod arrangement. The projector casts light and shadows that sharpen the light and shadows of the push rods, improving the success rate of image recognition and ensuring the correct push rod arrangement, thus promptly avoiding losses caused by incorrect push rod placement. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a schematic diagram of an intelligent detection system for a punching line ejector bar standby die proposed in Example 1 of the present invention; Figure 2 A top view of the mobile workbench proposed in Example 1 of the present invention; Figure 3 This is a schematic diagram showing the effect of the projector proposed in Example 1 of the present invention projecting the ejector rod layout diagram and mold information onto the surface of the movable workbench; Figure 4 This is a flow chart of an intelligent detection method for a press line ejector bar standby die proposed in Example 2 of the present invention.

[0019] Legend: 1-Protection; 2-Projector; 3-Visual unit; 4-Moving workbench; 5-Ejector rod; 6-First mounting bracket; 7-Second mounting bracket; 8-Hole. DETAILED DESCRIPTION

[0020] In order to clearly illustrate the technical features of this solution, the present invention is described in detail below through specific implementation methods and in conjunction with the accompanying drawings. The disclosure below provides many different embodiments or examples for realizing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. In addition, the present invention may repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplicity and clarity and does not itself indicate the relationship between the various embodiments and / or settings discussed. It should be noted that the components illustrated in the accompanying drawings are not necessarily drawn to scale. The present invention omits descriptions of well-known components and processing technologies and processes to avoid unnecessary limitations on the present invention.

[0021] Example 1 Embodiment 1 of the present invention proposes an intelligent detection system for press line ejector bar standby dies, which is used to solve the technical problems of difficulty in identifying and low efficiency in the arrangement of existing press line ejector bar standby dies. Figure 1 This is a schematic diagram of an intelligent detection system for a punching line ejector bar standby die proposed in Example 1 of the present invention; The system includes a projector 2 and a visual unit 3 mounted on a guard 1; and a top rod 5 provided on a mobile workbench 4; the top rod 5 is arranged in a hole 8 on the mobile workbench 4; The projector 2 is used to project the preset layout diagram of the ejector rods 5 corresponding to the current mold and the mold identification information onto the surface of the movable workbench 4, which is used to visually guide the layout of the ejector rods 5; The visual unit 3 is used to collect the image of the movable workbench 4 after the ejector rods 5 are arranged, and to determine whether the actual arrangement of the ejector rods is consistent with the standard arrangement of the mold preset by the system by using image recognition.

[0022] A first mounting bracket 6 and a second mounting bracket 7 are provided on the protection 1 ; the projector 2 is fixedly mounted on the first mounting bracket 6 ; and the visual unit 3 is fixedly mounted on the second mounting bracket 7 .

[0023] Embodiment 1 of the present invention proposes an intelligent detection system for the mold preparation of the ejector bar of the stamping line. The projector 2 projects the layout diagram of the ejector bar 5 and the mold number and other information onto the surface of the movable workbench 4. Since there is a certain error each time the movable workbench 4 is opened to the mold changing position, the projector 2 is required to have a certain ability to automatically adjust and adapt to the projection distortion to ensure that the projection image is not distorted, thereby guiding the manual placement of the ejector bar 5 in the hole on the movable workbench 4 according to the projection layout diagram, reducing the error rate of manual ejector bar 5 mold preparation arrangement and improving the replacement efficiency; the visual system 3 takes a picture of the ejector bar 5 arrangement on the movable workbench 4, transmits the image back for image recognition, and performs intelligent inspection on the ejector bar 5 arrangement. The light and shadow projected by the projector 2 sharpens the light and shadow of the ejector bar 5, improves the success rate of image recognition, ensures the correctness of the ejector bar 5 arrangement, and thus timely avoids the loss caused by the error in the arrangement of the ejector bar 5.

[0024] Figure 2 This is a top view of the mobile workbench proposed in Example 1 of the present invention; in this application, since the mobile workbench 4 is too heavy, there will be an error of more than ten millimeters each time it is opened to the theoretical mold change position. The projector 2 has the function of automatically adjusting and adapting to the projection distortion to ensure that the image projected on the surface of the mobile workbench 4 is not distorted. In addition, the projected light and shadow have the effect of sharpening the light and shadow of the top rod 5, thereby improving the accuracy of the subsequent visual unit 3 in checking the arrangement of the top rod 5.

[0025] The specific light field projected by the projector visual unit 2 includes high-contrast grid lines, uniform background light or structured light pattern, which is used to enhance the distinction between the top rod visual unit 5 and the background of the hole visual unit 8.

[0026] In the present application, before the visual unit 3 performs image recognition, it first pre-processes the collected image of the moving workbench 4 after the top rod 5 is arranged; the pre-processing process includes distortion correction, denoising, contrast enhancement and binarization processing.

[0027] Although the projector has corrected its own projection distortion, the camera lens itself may have distortion (such as barrel distortion and pincushion distortion). Use the camera calibration parameters to perform geometric correction on the image to ensure that straight lines in the image are also straight lines in the real world.

[0028] Apply filters (such as Gaussian filtering and median filtering) to reduce image sensor noise, illumination noise, etc.

[0029] Use histogram equalization or other methods to stretch the image's grayscale range to make the top bars and holes more distinct. The projector's sharpening effect is particularly effective at this stage.

[0030] Use histogram equalization or other methods to stretch the image's grayscale range to make the top bars and holes more distinct. The projector's sharpening effect is particularly effective at this stage.

[0031] Because the projector's light and shadow sharpening provides a stable high contrast, selecting the threshold becomes relatively easy and robust. The top bar area (especially the top or illuminated surface) will be significantly brighter than the bottom of the hole or the shadow area, thus clearly appearing as a white blob in the binary image.

[0032] In this application, the purpose of image recognition performed by the visual unit 3 is to locate and identify all hole positions and push rod positions in the pre-processed image (usually a binary image). Specifically, it includes: Since the holes of the workbench are arranged in a grid with a fixed pattern, the system usually knows the general structure of the grid (number of rows, columns, and hole spacing) in advance; Identify the reference points on the workbench surface, establish an accurate grid coordinate system of holes 8, and calculate the theoretical center coordinates of each hole 8; Perform connected domain analysis on the pre-processed image, extract the connected domain that meets the preset top rod size and shape characteristics, and calculate its centroid coordinates; The theoretical center coordinates of each hole 8 are traversed, and the center of mass of the connected domain of the push rod is searched within the preset search radius. The status of the hole 8 is determined to be "with push rod" or "without push rod", and a bitmap of the actual push rod layout is generated. If a blob center of mass that meets the push rod characteristics exists within the search radius, the hole is considered to have a push rod. If no blob center of mass that meets the push rod characteristics exists within the search radius, the hole is considered to have no push rod. The status of each hole on the entire workbench grid is recorded (1 for push rod, 0 for no push rod), forming a bitmap of the actual detected push rod layout.

[0033] The actual ejector pin layout bitmap is compared bit by bit with the preset standard ejector pin layout bitmap for the current mold. If the status values ​​(1 / 0) of all holes in the two bitmaps are exactly the same, the ejector pin layout is determined to be correct.

[0034] There is a discrepancy: If the status value of any hole does not match (for example, the standard requires a rod but there is actually no rod - missed placement; the standard requires no rod but there is actually a rod - over-placed; or placed in the wrong position - wrong placement will essentially cause the status of a specific hole to not match), it is determined to be an incorrect top rod arrangement.

[0035] The preset search radius is set to be less than half of the distance between adjacent holes 8 .

[0036] The system also includes an alarm module; the alarm module is used to receive the judgment signal output by the visual unit 3, trigger an audible and visual alarm when it is determined that the layout is wrong, and prevent the mobile workbench 4 from returning to the production line or subsequent mold change operations.

[0037] Embodiment 1 of the present invention proposes an intelligent detection system for press line ejector rod standby molds, which enables rapid replacement of different ejector rod arrangements for different molds, greatly reduces the error rate of manual ejector rod replacement, and improves the efficiency and accuracy of ejector rod replacement for different molds.

[0038] Embodiment 1 of the present invention proposes an intelligent detection system for ejector bar preparation molds of a stamping line, in which the projector has a certain ability to automatically adjust and adapt to projection distortion to ensure that the projection image is not distorted, thereby guiding manual placement of the ejector bar in the hole on the mobile workbench according to the projection layout diagram, reducing the error rate of manual ejector bar preparation mold layout and improving replacement efficiency.

[0039] In Example 1, a smart inspection system for ejector pin preparation on a press line is proposed. A visual unit photographs the ejector pin layout on a mobile worktable, transmits the image back for image recognition, and performs intelligent inspection of the ejector pin layout. A projector projects light and shadows that sharpen the ejector pins, improving the success rate of image recognition and ensuring the correct ejector pin layout, thereby preventing losses caused by incorrect ejector pin placement.

[0040] Example 2 Based on the intelligent detection system for the punching line ejector die proposed in Example 1 of the present invention, Example 2 of the present invention further proposes an intelligent detection method for the punching line ejector die. Figure 4 This is a flow chart of an intelligent detection method for a press line ejector bar standby die proposed in Example 2 of the present invention.

[0041] In step S400, the projector 2 is used to project the preset layout diagram of the ejector rods 5 corresponding to the current mold and the mold identification information onto the surface of the movable workbench 4 for visual guidance of the layout of the ejector rods 5; In step S410, the visual unit 3 is used to collect an image of the movable workbench 4 after the ejector rods 5 are arranged, and image recognition is used to determine whether the actual arrangement of the ejector rods is consistent with the standard mold arrangement preset by the system.

[0042] Before the visual unit 3 performs image recognition, it first pre-processes the collected image of the moving workbench 4 after the top rod 5 is arranged; the pre-processing process includes distortion correction, denoising, contrast enhancement and binarization processing.

[0043] Although the projector has corrected its own projection distortion, the camera lens itself may have distortion (such as barrel distortion and pincushion distortion). Use the camera calibration parameters to perform geometric correction on the image to ensure that straight lines in the image are also straight lines in the real world.

[0044] Apply filters (such as Gaussian filtering and median filtering) to reduce image sensor noise, illumination noise, etc.

[0045] Use histogram equalization or other methods to stretch the image's grayscale range to make the top bars and holes more distinct. The projector's sharpening effect is particularly effective at this stage.

[0046] Use histogram equalization or other methods to stretch the image's grayscale range to make the top bars and holes more distinct. The projector's sharpening effect is particularly effective at this stage.

[0047] Because the projector's light and shadow sharpening provides a stable high contrast, selecting the threshold becomes relatively easy and robust. The top bar area (especially the top or illuminated surface) will be significantly brighter than the bottom of the hole or the shadow area, thus clearly appearing as a white blob in the binary image.

[0048] In this application, the purpose of image recognition performed by the visual unit 3 is to locate and identify all hole positions and push rod positions in the pre-processed image (usually a binary image). Specifically, it includes: Since the holes of the workbench are arranged in a grid according to a fixed pattern, the system usually knows the general structure of the grid (number of rows, number of columns, and hole spacing) in advance.

[0049] Identify the reference points on the workbench surface, establish an accurate grid coordinate system of holes 8, and calculate the theoretical center coordinates of each hole 8; Perform connected domain analysis on the pre-processed image, extract the connected domain that meets the preset top rod size and shape characteristics, and calculate its centroid coordinates; The theoretical center coordinates of each hole 8 are traversed, and the center of mass of the connected domain of the push rod is searched within the preset search radius. The status of the hole 8 is determined to be "with push rod" or "without push rod", and a bitmap of the actual push rod layout is generated. If a blob center of mass that meets the push rod characteristics exists within the search radius, the hole is considered to have a push rod. If no blob center of mass that meets the push rod characteristics exists within the search radius, the hole is considered to have no push rod. The status of each hole on the entire workbench grid is recorded (1 for push rod, 0 for no push rod), forming a bitmap of the actual detected push rod layout.

[0050] The actual ejector pin layout bitmap is compared bit by bit with the preset standard ejector pin layout bitmap for the current mold. If the status values ​​(1 / 0) of all holes in the two bitmaps are exactly the same, the ejector pin layout is determined to be correct.

[0051] There is a discrepancy: If the status value of any hole does not match (for example, the standard requires a rod but there is actually no rod - missed placement; the standard requires no rod but there is actually a rod - over-placed; or placed in the wrong position - wrong placement will essentially cause the status of a specific hole to not match), it is determined to be an incorrect top rod arrangement.

[0052] The preset search radius is set to be less than half of the distance between adjacent holes 8 .

[0053] According to the determination signal output by the receiving visual unit 3, when it is determined that the arrangement is wrong, an audible and visual alarm is triggered, and the movable workbench 4 is prevented from returning to the production line or the subsequent mold change operation.

[0054] The description of the relevant parts of the intelligent detection method for the top bar standby mold of a stamping line proposed in Example 2 of the present invention can be found in the detailed description of the corresponding parts of the intelligent detection system for the top bar standby mold of a stamping line provided in Example 1 of the present application, and will not be repeated here.

[0055] Example 2 of the present invention proposes an intelligent detection method for the spare mold of the ejector bar of the stamping line, which realizes the rapid replacement of different ejector bar arrangements for different molds, greatly reduces the error rate of manual replacement of ejector bars, and improves the efficiency and accuracy of replacing ejector bars for different molds.

[0056] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device comprising a series of elements are inherent to the elements. In the absence of further restrictions, the elements limited by the statement "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device comprising the elements. In addition, the above-mentioned technical solutions provided in the embodiments of the present application are not described in detail in accordance with the corresponding technical solutions in the prior art to achieve the same principle, so as to avoid excessive elaboration.

[0057] Although the above description is of specific embodiments of the present invention in conjunction with the accompanying drawings, it does not limit the scope of protection of the present invention. For those skilled in the art, other different forms of modifications or variations can be made based on the above description. It is not necessary and impossible to list all embodiments here. Based on the technical solution of the present invention, various modifications or variations that can be made by those skilled in the art without expending creative effort are still within the scope of protection of the present invention.

Claims

1. An intelligent detection system for punching line ejector bar preparation, characterized in that: include: A projector (2) and a visual unit (3) mounted on the guard (1); and a top rod (5) disposed on the movable workbench (4); the top rod (5) is arranged in a hole (8) on the movable workbench (4); The projector (2) is used to project a preset layout diagram of the ejector rods (5) corresponding to the current mold and mold identification information onto the surface of the movable workbench (4) for visually guiding the layout of the ejector rods (5); The visual unit (3) is used to collect an image of the movable workbench (4) after the ejector rods (5) are arranged, and to use image recognition to determine whether the actual arrangement of the ejector rods is consistent with the standard arrangement of the mold preset by the system.

2. The intelligent detection system for punching line ejector rod preparation mold according to claim 1 is characterized in that: A first mounting bracket (6) and a second mounting bracket (7) are provided on the protection (1); The projector (2) is fixedly mounted on a first mounting bracket (6); The visual unit (3) is fixedly mounted on the second mounting bracket (7).

3. The intelligent detection system for punching line ejector die according to claim 1, characterized in that: The specific light field projected by the projector (2) includes high-contrast grid lines, uniform background light or structured light patterns, which are used to enhance the distinction between the top rod (5) and the hole (8) background.

4. The intelligent detection system for punching line ejector rod preparation mold according to claim 1 is characterized in that: The projector (2) also dynamically adjusts projection parameters according to the actual stop position of the movable workbench (4), compensating for geometric distortion of the projection image caused by positioning errors of the movable workbench (4).

5. The intelligent detection system for punching line ejector rod preparation die according to claim 1 is characterized in that: Before performing image recognition, the visual unit (3) further includes: pre-processing the collected image of the mobile workbench (4) after the top rod (5) is arranged; wherein the pre-processing process includes distortion correction, denoising, contrast enhancement and binarization processing.

6. The intelligent detection system for punching line ejector die according to claim 5, characterized in that: The process of image recognition performed by the vision unit (3) includes: Identify the workbench table reference point, establish an accurate hole (8) grid coordinate system, and calculate the theoretical center coordinates of each hole (8); Perform connected domain analysis on the pre-processed image, extract the connected domain that meets the preset top rod size and shape characteristics, and calculate its centroid coordinates; Traversing the theoretical center coordinates of each hole (8), searching for the center of mass of the connected domain of the top rod within the preset search radius, determining whether the state of the hole (8) is "with top rod" or "without top rod", and generating an actual top rod arrangement bitmap; The actual ejector pin arrangement bitmap is compared bit by bit with the preset current mold standard ejector pin arrangement bitmap.

7. The intelligent detection system for punching line ejector rod preparation die according to claim 6, characterized in that: The preset search radius is set to be less than half the distance between adjacent holes (8).

8. The intelligent detection system for punching line ejector rod preparation mold according to claim 1 is characterized in that: The system also includes an alarm module; The alarm module is used to receive the determination signal output by the visual unit (3), trigger an audible and visual alarm when it is determined that the arrangement is wrong, and prevent the mobile workbench (4) from returning to the production line or subsequent mold change operations.

9. An intelligent detection method for punching line ejector die, characterized in that: The following steps are involved: Using a projector (2), a preset layout diagram of the ejector rods (5) corresponding to the current mold and mold identification information are projected onto the surface of the movable workbench (4) for visually guiding the layout of the ejector rods (5); The visual unit (3) is used to collect an image of the movable workbench (4) after the ejector rods (5) are arranged, and an image recognition method is used to determine whether the actual arrangement of the ejector rods is consistent with the standard arrangement of the mold preset by the system.

10. The intelligent detection method for the punching line ejector die according to claim 9, characterized in that: The image recognition process includes: Identify the workbench table reference point, establish an accurate hole (8) grid coordinate system, and calculate the theoretical center coordinates of each hole (8); Perform connected domain analysis on the pre-processed image, extract the connected domain that meets the preset top rod size and shape characteristics, and calculate its centroid coordinates; Traversing the theoretical center coordinates of each hole (8), searching for the center of mass of the connected domain of the top rod within the preset search radius, determining whether the state of the hole (8) is "with top rod" or "without top rod", and generating an actual top rod arrangement bitmap; The actual ejector pin arrangement bitmap is compared bit by bit with the preset current mold standard ejector pin arrangement bitmap.

Images