Device and method for detecting foreign matters and defects at bottom of electromagnet suction cup

Abstract

The invention discloses a device and a method for detecting foreign matters and defects at the bottom of an electromagnet sucker. The device comprises a guide rail, a mobile platform and a control unit, the guide rail extends to the bottom of the electromagnet sucker; the moving platform comprises a translation device capable of moving along the guide rail; the synchronous belt module is mounted on the translation device and can move in the direction perpendicular to the guide rail; the visual module is mounted on the synchronous belt module and comprises a linear array camera and a laser scanner; and the control unit is connected with the translation device, the synchronous belt module and the visual module, and is configured to control the translation device and the synchronous belt module to move to drive the visual module to scan the bottom of the electromagnet suction cup so as to identify and position foreign matters and defects at the bottom of the electromagnet suction cup. According to the invention, a multi-dimensional mobile platform with a high-precision visual module is adopted to comprehensively scan the bottom of an electromagnet sucker, and intelligent identification, recording and early warning of foreign matters and defects are completed with the assistance of an intelligent AI deep learning detection algorithm and a change monitoring algorithm.

Description

Technical Field

[0001] The present invention belongs to the technical field of electromagnetic chuck detection, and in particular relates to a device and method for detecting foreign matter and defects at the bottom of an electromagnetic chuck. Background Art

[0002] With the advancement of industrialization, various industries are placing increasingly high demands on product quality. Consequently, frequent inspections of products and equipment are being conducted during the production process to ensure consistent quality. Electromagnetic suction cups in steel mills must prevent damage to workpieces when they are used to hold steel, especially for flat workpieces like plate steel. However, foreign matter and defects within the cup's base can damage the workpiece surface and exacerbate existing damage.

[0003] Currently, most inspection methods for electromagnet chucks involve: 1. Routine operator detection and elimination of abnormalities through sensory inspection; 2. Regular, detailed inspections of key electromagnet chucks by professional maintenance personnel using sensory inspections or instrumentation. However, this approach is technically demanding, time-consuming, and inefficient. Therefore, a device is needed to accurately identify foreign objects and defects on the bottom of an electromagnet chuck, scan and image them, and locate their position. Summary of the Invention

[0004] The object of the present invention is to provide a device and method for detecting foreign matter and defects at the bottom of an electromagnetic chuck, so as to realize the detection of foreign matter and defects at the bottom of an electromagnetic chuck.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: In a first aspect, the present invention provides a device for detecting foreign matter and defects at the bottom of an electromagnetic chuck, the device comprising a guide rail, a mobile platform, and a control unit; wherein: The guide rail extends to the bottom of the electromagnet suction cup; A mobile platform is configured to perform a scanning motion relative to the bottom of the electromagnetic chuck to obtain a 2D high-resolution image and three-dimensional point cloud data of the bottom of the electromagnetic chuck; the mobile platform comprises: A translation device capable of moving along the guide rail; a synchronous belt module, mounted on the translation device and movable in a direction perpendicular to the guide rail; and The vision module is installed on the synchronous belt module and includes: A linear array camera to obtain a 2D high-resolution image of the bottom of the electromagnet chuck; and a laser scanner for acquiring three-dimensional point cloud data of the bottom of the electromagnet suction cup; The control unit is connected to the translation device, synchronous belt module and vision module of the mobile platform, and is configured to control the movement of the translation device and the synchronous belt module to drive the vision module to scan the bottom of the electromagnetic suction cup, and receive the 2D high-resolution image obtained by the linear array camera and the three-dimensional point cloud data obtained by the laser scanner, and identify and locate foreign objects and defects on the bottom of the electromagnetic suction cup based on the 2D high-resolution image and the three-dimensional point cloud data.

[0006] In the above solution, the translation device includes: A second servo motor is used to drive the translation device to move along the guide rail; a translation driver connected to the control unit and configured to receive a control signal from the control unit and issue a command to the second servo motor; The second encoder is used for speed and position feedback; The second limit switch is used to provide travel limit feedback.

[0007] In the above solution, the synchronous belt module includes: The first servo motor is used to drive the synchronous belt module to move in a direction perpendicular to the guide rail; a linear actuator connected to the control unit and configured to receive a control signal from the control unit and issue a command to the first servo motor; The first encoder is used for speed and position feedback; The first limit switch is used to provide travel limit feedback.

[0008] In the above scheme, the control unit outputs control signals to the translation drive and the linear drive to respectively move the position of the translation device and the synchronous belt module, and receives speed and position feedback from two encoders and travel limit feedback from two limit switches; The control unit also instructs the triggering of the line array camera and laser scanner to obtain 2D high-resolution images and 3D point cloud data of the bottom of the electromagnetic suction cup.

[0009] In the above scheme, the three-dimensional point cloud data of the bottom of the electromagnet suction cup is obtained, including: The synchronous belt module drives the laser scanner to scan the bottom of the electromagnet suction cup. The speed of the synchronous belt module during scanning is uniform and does not exceed the set speed. :

[0010] in, is the line frequency of the laser scanner, The physical size of a single line resolution of a laser scanner; The m×n dimensional matrix of the laser scanner distance measurement data sequence and the n×1 dimensional matrix of the corresponding synchronous belt module position data sequence are fused into a discrete three-dimensional space coordinate sequence:

[0011] in, i =0,1,…, m The longitudinal sampling point index of the laser linear array scan; j =0,1,…, n It is the horizontal sampling point index of the synchronous belt module; is the three-dimensional space coordinate; It is the height information of the laser scanner ranging data sequence; This results in a point cloud , and then use the trilinear interpolation algorithm to fit the uniform lattice :

[0012]

[0013]

[0014]

[0015]

[0016]

[0017]

[0018] in, is the linear interpolation formula, are the interpolation point coordinates, 、 and Point Cloud The coordinates on both sides of the interpolation point.

[0019] In the above solution, the vision module also includes a box, and the line array camera and the laser scanner are protected and fixed by the box, and the lens end of the line array camera and the laser scanner each has a window separated by a glass sheet.

[0020] In the above solution, the mobile platform further includes a cleaning device, which is a blowing device or a brushing device for cleaning the glass window of the linear array camera and / or laser scanner.

[0021] In the above solution, the device further includes an electromagnet chuck support structure, which includes: At least one auxiliary support column is used to stably support the truss at the upper end of the electromagnetic suction cup.

[0022] In the above solution, the device also includes a safety zone, and the mobile platform is docked in the safety zone when not working, and returns to the safety zone for docking after data collection is completed.

[0023] In a second aspect, the present invention further provides a method for detecting foreign matter and defects at the bottom of an electromagnetic chuck, which is applied to the device for detecting foreign matter and defects at the bottom of an electromagnetic chuck according to any one of the first aspects, and the method comprises: Obtain 2D high-resolution images and 3D point cloud data of the bottom of the electromagnetic chuck; Image enhancement is achieved through SRGAN super-resolution reconstruction and improved non-local means algorithm, and point cloud accuracy is optimized by combining multi-scale ICP registration and statistical outlier removal algorithm; Use SIFT+RANSAC feature mapping to complete image and point cloud data fusion; Based on the fused image and point cloud data, the YOLOv11-Dense network is used to detect abnormal areas. The morphological features of these abnormal areas are then extracted using the U-Net++ network integrated with the attention mechanism. Finally, a deformation detection model is established based on the 3D deformation analysis module of the PointNet++ architecture, and defect classification is achieved using the deformation detection model. At the same time, based on the fused image and point cloud data, the integrated statistical process control system performs X̄-R joint monitoring of the bottom parameters of the electromagnetic chuck to achieve deformation early warning.

[0024] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art: The present invention proposes a device and method for detecting foreign matter and defects at the bottom of an electromagnetic suction cup. A multi-dimensional mobile platform with a high-precision vision module is used to comprehensively scan the bottom of the electromagnetic suction cup. The high-precision texture information and multi-dimensional three-dimensional model information of the suction cup bottom obtained by the vision module are used, supplemented by an intelligent AI deep learning detection algorithm and a change monitoring algorithm to complete intelligent identification, recording and early warning of foreign matter and defects. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A schematic diagram of the control principle of a device for detecting foreign matter and defects at the bottom of an electromagnetic chuck provided in an embodiment of the present application; Figure 2 A top plan view of a device for detecting foreign matter and defects at the bottom of an electromagnetic chuck provided in an embodiment of the present application; Figure 3 A side plan view of a device for detecting foreign matter and defects at the bottom of an electromagnetic chuck provided in an embodiment of the present application; Figure 4 A schematic diagram of the three-dimensional structure of a device for detecting foreign matter and defects at the bottom of an electromagnetic chuck provided in an embodiment of the present application; Figure 5 A schematic diagram of the structure of a mobile platform provided in an embodiment of the present application; Figure 6 A schematic diagram of the structure of a visual module provided in an embodiment of the present application; Figure 7 A schematic diagram of the motion of a device for detecting foreign matter and defects at the bottom of an electromagnetic chuck provided in an embodiment of the present application; Figure 8 A detection flow chart of a device for detecting foreign matter and defects at the bottom of an electromagnetic chuck provided in an embodiment of the present application; Figure 9 A flow chart of a method for detecting foreign matter and defects at the bottom of an electromagnetic chuck provided in an embodiment of the present application.

[0026] In the figure: 1-electromagnetic suction cup, 2-truss, 3-auxiliary support column, 4-safety zone, 5-guide rail, 6-mobile platform, 7-translation device, 8-synchronous belt module, 9-vision module, 10-line array camera, 11-laser scanner, 12-glass window, 13-blowing device, 14-drag chain, 15-box. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Based on the embodiments provided in this application, all other embodiments obtained by those of ordinary skill in the art without making creative work are within the scope of protection of the present invention.

[0028] Obviously, the drawings described below are merely examples or embodiments of the present application. Those skilled in the art can, without inventive effort, apply the present application to other similar scenarios based on these drawings. Furthermore, it is also understood that, although the effort involved in such a development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, changes in design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as an insufficiency of the content disclosed in this application.

[0029] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it refer to independent or alternative embodiments that are mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments unless there is a conflict.

[0030] Unless otherwise defined, technical or scientific terms used herein shall have the ordinary meaning as understood by persons of ordinary skill in the art to which this application belongs. The terms "a," "an," "an," "the," and similar expressions used herein do not denote quantitative limitations and may refer to either the singular or the plural. The terms "comprise," "include," "have," and any variations thereof, used herein, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or modules (units) is not limited to the listed steps or units but may also include steps or units not listed, or may include other steps or units inherent to the process, method, product, or apparatus. The terms "connected," "connected," "coupled," and similar expressions used herein are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. As used herein, "plurality" means two or more. "And / or" describes an association between associated objects, indicating that three possible relationships exist. For example, "A and / or B" may mean: A exists alone; A and B exist simultaneously; or B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific order for the objects.

[0031] This application provides a high-precision device for detecting foreign matter and defects on the bottom of an electromagnetic chuck. Specifically, it is a device for the steel industry that automatically scans, detects, identifies, and locates foreign matter and defects on the bottom of an electromagnetic chuck. The device uses a multi-dimensional motion platform with a high-precision vision module to perform a comprehensive visual scan of the bottom of the electromagnetic chuck. The device uses the high-precision texture information and multi-dimensional three-dimensional model information of the chuck bottom acquired by the vision module, supplemented by intelligent AI deep learning detection algorithms and change monitoring algorithms, to intelligently identify, record, and issue early warnings for foreign matter and defects on the bottom of the electromagnetic chuck.

[0032] The structure of the device for detecting foreign matter and defects at the bottom of the electromagnet chuck according to the embodiment of the present application is as follows: Figure 2 、 Figure 3 and Figure 4 As shown, it consists of four major modules: auxiliary support 3, mobile platform 6, exclusive safety zone 4, and guide rail 5; the auxiliary support column 3 is used to stably support the truss 2 at the upper end of the electromagnet suction cup 1, so that the electromagnet basically tends to a stationary state; the mobile platform 6 moves along the guide rail 5 for scanning and detection; an exclusive safety zone 4 is set on the leftmost side of the guide rail 5 for storing equipment after the detection is completed.

[0033] like Figure 5As shown, the mobile platform 6 is mainly composed of a translation device 7, a synchronous belt module 8, a visual module 9, a blower device 13 and a drag chain 14. The translation device 7 has its own motor and is located at the bottom of the mobile platform 6. It moves along the guide rail 5 and can adjust the walking speed; the synchronous belt module 8 is arranged along the length of the electromagnet suction cup 1 and carries the visual module 9 for scanning and detection; Figure 6 As shown, the vision module 9 mainly includes a line array camera 10 and a laser scanner 11, both of which are protected and fixed by a box 15. There is a window separated by a glass sheet at each lens end to prevent impurities from blocking the lens line of sight; a blower device 13 is set on the side of the vision module 9 to blow and clean impurities on the glass windows 12 of the line array camera 10 and the laser scanner 11, thereby reducing the amount of iron filings and impurities staying on the glass windows 12 to avoid affecting the detection results; the drag chain 14 is used for wiring layout of the entire equipment.

[0034] like Figure 1 As shown, the control unit is the main control device, which outputs control signals to the translation drive and the linear drive, respectively moving the position of the translation device 7 and the synchronous belt module 8, and will receive position signal feedback from the two drives; after receiving the signal, the linear drive sends an instruction to the first servo motor, and receives the speed position feedback of the first encoder and the stroke limit feedback of the first limit switch; the translation drive sends an instruction to the second servo motor after receiving the signal, and receives the speed position feedback of the second encoder and the stroke limit feedback of the second limit switch; the control unit instructs to trigger the linear array camera 10 and the laser scanner 11, and collects data from them to obtain feedback; the control unit sends instructions to control the start and stop of the solenoid valve, so that the blower 13 cleans the debris in front of the glass lens in the visual module 9.

[0035] Furthermore, the air blower can be replaced by a brushing mechanism. For example, a brushing mechanism can be used to remove impurities by brushing the glass windows of a line scan camera and / or laser scanner. The track-based motion of the translation mechanism can be replaced by a screw-track module or a synchronous belt module.

[0036] The visual module is based on multimodal data fusion and deep learning technology, and has built an intelligent detection system covering the entire process of "data collection-feature analysis-defect determination-process monitoring". Figure 9As shown, the system uses a high-dynamic-range line scan camera and a laser scanner to simultaneously capture 2D high-resolution images and dense point clouds. Image enhancement is achieved through SRGAN super-resolution reconstruction and an improved non-local means algorithm. Multi-scale ICP registration and statistical outlier removal are combined to optimize point cloud accuracy. After RGB-D data fusion using SIFT+RANSAC feature mapping, a YOLOv11-Dense network is used for rapid foreign object identification. A U-Net++ network incorporating an attention mechanism is then used to extract pixel-level morphological features. Subsequently, a submillimeter deformation detection model is established using a 3D deformation analysis module based on the PointNet++ architecture. A statistical process control (SPC) system is integrated to perform joint X̄-R monitoring of parameters such as flatness and roughness, ultimately achieving closed-loop control for defect classification, deformation warning, and quality traceability.

[0037] Abnormal regions are first detected using the YOLOv11-Dense network, followed by a more refined segmentation of the detected regions (U-Net++). This obtains the 2D shape of the abnormal regions, extracts their depth, and then constructs a 3D model for classification using PointNet++. In PointNet++, the image undergoes upsampling processing, such as super-resolution, to create a more refined model of the abnormal regions. The statistical process control (SPC) system performs XR (mean-range) joint monitoring based on information such as surface roughness. By analyzing XR (mean-range) plots, it can identify any sudden data changes in the region, thereby providing early warnings.

[0038] Quality traceability refers to the preservation of previous defect data, such as appearance, depth characteristics, and other information, so that the data can be analyzed later when the problem needs to be traced.

[0039] like Figure 7 and Figure 8 As shown, the detection process of the detection device is as follows: 1. When not in operation, the robot is initially parked in a dedicated safety zone to avoid human-machine injuries. 2. When in working state, the mobile platform quickly moves horizontally along the guide rail to the bottom of the electromagnet suction cup, and the linear array camera and laser scanner are ready for operation; 3. The linear array camera and laser scanner receive the signal and start responding. The synchronous belt module moves from one end toward the other end of the electromagnet, scanning, identifying, and modeling the bottom of the electromagnet suction cup along the way. 4. Data collection and transmission are carried out simultaneously. The industrial computer compares and summarizes the information from the linear array camera and laser scanner to determine the location of defects and foreign objects and mark them; 5. If there are still tasks to be done, the bottom of the next electromagnetic suction cup can be scanned and inspected continuously; if the work task is completed, it will return to the exclusive safety zone for protection.

[0040] The device uses a high dynamic range line array camera and a laser scanner to synchronously capture 2D high-resolution images and dense point clouds. The detection method is as follows: 1. The fixed encoder value of the synchronous belt module (set in advance according to the fixed placement of the electromagnet suction cup) is used as the trigger for the scanner point cloud modeling. To ensure the scanning uniformity and scanning accuracy of the synchronous belt module movement, the speed of the synchronous belt module must be uniform and not greater than:

[0041] in, is the line frequency of the scanner, The physical size of the scanner's single-line resolution (such as 0.05mm).

[0042] 2. After entering the scan, the laser scanner ranging data sequence (m×n dimensional matrix) 'And the corresponding synchronous belt module position data sequence n×1 dimensional matrix is ​​fused into a discrete three-dimensional space coordinate sequence:

[0043] in, i =0,1,…, m The longitudinal sampling point index of the laser linear array scan; j =0,1,…, n It is the horizontal sampling point index of the synchronous belt module; point It is the pure height information in the scanner ranging data sequence.

[0044] 3. Use trilinear interpolation algorithm to fit uniform lattice , interpolation process:

[0045]

[0046]

[0047]

[0048]

[0049]

[0050]

[0051] in, is the linear interpolation formula.

[0052] In summary, the present application provides a device and method for detecting foreign matter and defects at the bottom of an electromagnetic chuck, which comprehensively identifies foreign matter and defects at the bottom of the electromagnetic chuck by photographing with a linear array camera and high-precision scanning modeling with a laser scanner.

[0053] It should be pointed out that, according to the needs of implementation, the various steps / components described in this application can be split into more steps / components, or two or more steps / components or partial operations of steps / components can be combined into new steps / components to achieve the purpose of the present invention.

[0054] It will be easily understood by those skilled in the art that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A device for detecting foreign matter and defects at the bottom of an electromagnetic chuck, characterized in that: The device includes a guide rail, a mobile platform and a control unit; wherein: The guide rail extends to the bottom of the electromagnet suction cup; A mobile platform is configured to perform a scanning motion relative to the bottom of the electromagnetic chuck to obtain a 2D high-resolution image and three-dimensional point cloud data of the bottom of the electromagnetic chuck; the mobile platform comprises: A translation device, movable along the guide rail; a synchronous belt module, mounted on the translation device and movable in a direction perpendicular to the guide rail; and The vision module is installed on the synchronous belt module and includes: A linear array camera to obtain a 2D high-resolution image of the bottom of the electromagnet chuck; and a laser scanner for acquiring three-dimensional point cloud data of the bottom of the electromagnet suction cup; The control unit is connected to the translation device, synchronous belt module and vision module of the mobile platform, and is configured to control the movement of the translation device and the synchronous belt module to drive the vision module to scan the bottom of the electromagnetic suction cup, and receive the 2D high-resolution image obtained by the linear array camera and the three-dimensional point cloud data obtained by the laser scanner, and identify and locate foreign objects and defects on the bottom of the electromagnetic suction cup based on the 2D high-resolution image and the three-dimensional point cloud data.

2. The device for detecting foreign matter and defects at the bottom of an electromagnetic chuck according to claim 1, characterized in that: The translation device includes: A second servo motor is used to drive the translation device to move along the guide rail; a translation driver connected to the control unit and configured to receive a control signal from the control unit and issue a command to the second servo motor; The second encoder is used for speed and position feedback; The second limit switch is used to provide travel limit feedback.

3. The device for detecting foreign matter and defects at the bottom of the electromagnetic chuck according to claim 2, characterized in that: The synchronous belt module includes: The first servo motor is used to drive the synchronous belt module to move in a direction perpendicular to the guide rail; a linear actuator connected to the control unit and configured to receive a control signal from the control unit and issue a command to the first servo motor; The first encoder is used for speed and position feedback; The first limit switch is used to provide travel limit feedback.

4. The device for detecting foreign matter and defects at the bottom of the electromagnetic chuck according to claim 3, characterized in that: The control unit outputs control signals to the translation drive and linear drive to respectively move the position of the translation device and the synchronous belt module, and receives speed and position feedback from two encoders and travel limit feedback from two limit switches; The control unit also instructs the triggering of the line array camera and laser scanner to obtain 2D high-resolution images and 3D point cloud data of the bottom of the electromagnetic suction cup.

5. The device for detecting foreign matter and defects at the bottom of an electromagnetic chuck according to any one of claims 1 to 4, characterized in that: Obtain 3D point cloud data of the bottom of the electromagnetic chuck, including: The synchronous belt module drives the laser scanner to scan the bottom of the electromagnet suction cup. The speed of the synchronous belt module during scanning is uniform and does not exceed the set speed. : in, is the line frequency of the laser scanner, It is the physical size of single line resolution of laser scanner; The m×n dimensional matrix of the laser scanner distance measurement data sequence and the n×1 dimensional matrix of the corresponding synchronous belt module position data sequence are fused into a discrete three-dimensional space coordinate sequence: in, i =0,1,…, m The longitudinal sampling point index of the laser linear array scan; j =0,1,…, n It is the horizontal sampling point index of the synchronous belt module; is the three-dimensional space coordinate; It is the height information of the laser scanner ranging data sequence; This results in a point cloud , and then use the trilinear interpolation algorithm to fit the uniform lattice : in, is the linear interpolation formula, are the interpolation point coordinates, 、 and Point Cloud The coordinates on both sides of the interpolation point.

6. The device for detecting foreign matter and defects at the bottom of an electromagnetic chuck according to claim 1, characterized in that: The vision module also includes a box, and the line array camera and the laser scanner are protected and fixed by the box, and the lens end of the line array camera and the laser scanner each has a window separated by a glass sheet.

7. The device for detecting foreign matter and defects at the bottom of an electromagnetic chuck according to claim 6, characterized in that: The mobile platform further comprises a cleaning device, which is a blowing device or a brushing device for cleaning the glass window of the linear array camera and / or the laser scanner.

8. The device for detecting foreign matter and defects at the bottom of an electromagnetic chuck according to claim 1, characterized in that: The device also includes an electromagnet chuck support structure, which includes: At least one auxiliary support column is used to stably support the truss at the upper end of the electromagnetic suction cup.

9. The device for detecting foreign matter and defects at the bottom of an electromagnetic chuck according to claim 1, characterized in that: The device also includes a safety zone, where the mobile platform is docked when not in operation, and returns to the safety zone for docking after data collection is completed.

10. A method for detecting foreign matter and defects at the bottom of an electromagnetic chuck, characterized in that: The device for detecting foreign matter and defects at the bottom of an electromagnetic chuck as claimed in any one of claims 1 to 9 is characterized in that the method comprises: Obtain 2D high-resolution images and 3D point cloud data of the bottom of the electromagnetic chuck; Image enhancement is achieved through SRGAN super-resolution reconstruction and improved non-local means algorithm, and point cloud accuracy is optimized by combining multi-scale ICP registration and statistical outlier removal algorithm; Use SIFT+RANSAC feature mapping to complete image and point cloud data fusion; Based on the fused image and point cloud data, the YOLOv11-Dense network is used to detect abnormal areas. The morphological features of these abnormal areas are then extracted using the U-Net++ network integrated with the attention mechanism. Finally, a deformation detection model is established based on the 3D deformation analysis module of the PointNet++ architecture, and defect classification is achieved using the deformation detection model. At the same time, based on the fused image and point cloud data, the integrated statistical process control system performs X̄-R joint monitoring of the bottom parameters of the electromagnetic chuck to achieve deformation early warning.

Images