Product defect automatic repairing method, device, equipment, storage medium and program product

Abstract

The application discloses a product defect automatic repairing method and device, equipment, storage medium and program product, relates to the technical field of laser repairing, and the product defect automatic repairing method comprises the steps of obtaining a first image of a product to be repaired, performing defect identification on the first image, determining first pixel coordinates of the defect to be repaired in the first image, converting the first pixel coordinates into first motion platform coordinates in a motion platform coordinate system, obtaining a second image of the product to be repaired when the motion platform moves to a position corresponding to the first motion platform coordinates, determining second pixel coordinates of the defect to be repaired in the second image, determining an offset of the motion platform according to the second pixel coordinates, and controlling a laser to automatically repair the product to be repaired based on the offset. Since the motion platform only needs to perform two movements, repeated trial and error in the physical space is not needed, the repairing accuracy and the repairing efficiency are considered, and the automatic repairing of the defect is realized.

Description

Technical Field

[0001] This application relates to the field of laser repair technology, and in particular to methods, apparatus, equipment, storage media and program products for automatic repair of product defects. Background Technology

[0002] Automated Optical Inspection (AOI) equipment is now widely used in the electronics manufacturing industry. While existing AOI systems can often detect product defects, most still require manual verification, location, and repair after inspection.

[0003] In existing solutions, the resistance of manufactured products is usually measured on resistance measuring equipment to determine which circuit has a short circuit or open circuit defect. However, this does not specify the exact location of the defect on the circuit. The product is then given to a human to examine the entire circuit under a microscope to find the defect and repair it manually. This method is very inefficient and involves a lot of human intervention. The circuits on the product are densely packed, and human error is possible. There is also the possibility that manual repair may damage the product. Summary of the Invention

[0004] The main objective of this application is to provide a method, apparatus, equipment, storage medium, and program product for automatic repair of product defects, aiming to solve the technical problems of low efficiency and poor accuracy of existing automatic repair methods for product defects.

[0005] To achieve the above objectives, this application proposes an automatic product defect repair method, which includes: Acquire a first image of the product to be repaired, and perform defect identification on the first image to determine the first pixel coordinates of the defect to be repaired in the first image; Convert the first pixel coordinates to the first motion platform coordinates in the motion platform coordinate system; When the motion platform moves to the position corresponding to the coordinates of the first motion platform, a second image of the product to be repaired is acquired, and the second pixel coordinates of the defect to be repaired in the second image are determined; The offset of the motion platform is determined based on the second pixel coordinates, and the laser is controlled to automatically repair the product to be repaired based on the offset.

[0006] In one embodiment, before the step of performing defect identification on the first image and determining the first pixel coordinates of the defect to be repaired in the first image, the method further includes: The first image is subjected to feature recognition to obtain the first image features; wherein, the first image is a complete product image of the product to be repaired; In the first image features, a first position of a first feature point and a second position of a second feature point are determined; wherein, the first feature point serves as a reference positioning point for the product to be repaired, used to determine the reference position of the product to be repaired in the detection image coordinate system; the second feature point serves as an angle reference point for the product to be repaired, used to determine the rotational offset of the product to be repaired relative to the detection image coordinate system; The detection position and detection angle of the product to be repaired at the detection station are determined based on the first position and the second position.

[0007] In one embodiment, the step of determining the second pixel coordinates of the defect to be repaired in the second image includes: Determine the third position of the first feature point and the fourth position of the second feature point in the second image; The repair position and repair angle of the product to be repaired at the repair station are determined based on the third position and the fourth position. The positional deviation of the product to be repaired as it moves from the detection station to the repair station is determined based on the detection location and the repair location. The angular deviation of the product to be repaired as it moves from the inspection station to the repair station is determined based on the detection angle and the repair angle. Based on the positional deviation and the angular deviation, locate the second pixel coordinates corresponding to the defect to be repaired in the second image.

[0008] In one embodiment, the step of determining the offset of the motion platform based on the second pixel coordinates includes: Obtain the center pixel coordinates of the second image, and determine the pixel difference between the second pixel coordinates and the center pixel coordinates; The pixel difference is converted into a physical distance, and the physical distance is used as the offset of the motion platform.

[0009] In one embodiment, the step of automatically repairing the product to be repaired based on the offset-controlled laser includes: The coordinates of the first motion platform are corrected based on the offset to obtain the coordinates of the second motion platform; Control the motion platform to move to the position corresponding to the coordinates of the second motion platform; When the motion platform moves to the position corresponding to the coordinates of the second motion platform, the laser is controlled to automatically repair the product to be repaired.

[0010] In one embodiment, the step of performing defect identification on the first image and determining the first pixel coordinates of the defect to be repaired in the first image includes: The first image is used to identify defects by a target detection model to determine the defect attributes of the initial defects in the first image; wherein, the defect attributes include: defect category, defect area, the first pixel coordinates corresponding to the defect to be repaired, and confidence level; The initial defects are filtered according to preset filtering rules and the defect attributes, and the filtered initial defects are taken as defects to be repaired.

[0011] Furthermore, to achieve the above objectives, this application also proposes an automatic product defect repair device, which includes: The defect identification module is used to acquire a first image of the product to be repaired, and to identify defects in the first image to determine the first pixel coordinates of the defect to be repaired in the first image. The coordinate transformation module is used to convert the coordinates of the first pixel into the coordinates of the first motion platform in the motion platform coordinate system. The position alignment module is used to acquire a second image of the product to be repaired and determine the second pixel coordinates of the defect to be repaired in the second image when the motion platform moves to the position corresponding to the coordinates of the first motion platform. An automatic repair module is used to determine the offset of the motion platform based on the second pixel coordinates, and control the laser to automatically repair the product to be repaired based on the offset.

[0012] In addition, to achieve the above objectives, this application also proposes an automatic product defect repair device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the automatic product defect repair method as described above.

[0013] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the product defect automatic repair method described above.

[0014] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the product defect automatic repair method described above.

[0015] One or more technical solutions proposed in this application have at least the following technical effects: This application acquires a first image of the product to be repaired, identifies defects in the first image, and determines the first pixel coordinates of the defect in the first image. The first pixel coordinates are then converted to first motion platform coordinates in the motion platform coordinate system. With the motion platform moving to the position corresponding to the first motion platform coordinates, a second image of the product to be repaired is acquired, and the second pixel coordinates of the defect in the second image are determined. The offset of the motion platform is determined based on the second pixel coordinates, and the laser is controlled to automatically repair the product based on the offset. This solution employs a two-level positioning architecture. The large-field-of-view camera at the inspection station provides coarse positioning coordinates covering the entire product area, enabling the platform to move to the macroscopic region where the defect is located. The small-field-of-view camera at the repair station performs secondary imaging within this region, calculating the offset using the center of the second image as a reference. Since the second image is taken after the platform has moved to the coarse positioning position, the image center corresponds to the platform's current actual stopping position. The pixel difference of the defect deviating from the center in the image, multiplied by the pixel equivalent, directly equals the offset in physical space. This offset does not need to be converted back to the motion platform coordinate system and can be directly superimposed onto the first motion platform coordinates for correction, avoiding the additional errors introduced by the secondary coordinate conversion. The motion platform only needs to perform two movements to accurately align the defect with the laser focus without repeated attempts in physical space, thus balancing repair accuracy and efficiency and achieving automatic defect repair. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating an embodiment of the automatic product defect repair method of this application. Figure 2 This is a flowchart illustrating Embodiment 2 of the automatic product defect repair method of this application. Figure 3 This is a flowchart illustrating Embodiment 3 of the automatic product defect repair method of this application; Figure 4 This is a schematic diagram of the focusing process for automatic repair in one implementation of the automatic product defect repair method of this application; Figure 5 This is a schematic diagram of the module structure of the automatic product defect repair device according to an embodiment of this application; Figure 6 This is a schematic diagram of the device structure of the hardware operating environment involved in the automatic product defect repair method in the embodiments of this application.

[0019] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0021] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0022] The main solution of this application embodiment is as follows: acquire a first image of the product to be repaired, identify defects in the first image, and determine the first pixel coordinates of the defect to be repaired in the first image; convert the first pixel coordinates into first motion platform coordinates in the motion platform coordinate system; acquire a second image of the product to be repaired when the motion platform moves to the position corresponding to the first motion platform coordinates, and determine the second pixel coordinates of the defect to be repaired in the second image; determine the offset of the motion platform according to the second pixel coordinates, and control the laser to automatically repair the product to be repaired based on the offset.

[0023] This application provides a solution that first performs coarse positioning based on the first pixel coordinates obtained from the first image, and then moves the motion platform to the corresponding position. After the motion platform is in position, a second image is acquired, and the second pixel coordinates of the defect in the second image are determined. Based on the second pixel coordinates, the current offset of the platform is determined, and the laser is controlled to perform repair based on this offset. The accurate measurement results of the secondary fine positioning compensate for the coordinate transformation error that exists during the first positioning, thereby realizing automatic repair of product defects and avoiding the problems of inaccurate platform positioning and laser failure to accurately hit the defect due to coordinate transformation errors.

[0024] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a computer or server, or an electronic device or virtual device capable of performing the above functions. The following description uses an automatic product defect repair device (hereinafter referred to as the repair device) as an example to illustrate this embodiment and the subsequent embodiments.

[0025] Based on this, embodiments of this application provide an automatic product defect repair method, referring to... Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the automatic product defect repair method of this application.

[0026] In this embodiment, the automatic product defect repair method includes steps S10 to S40: Step S10: Obtain a first image of the product to be repaired, and perform defect identification on the first image to determine the first pixel coordinates of the defect to be repaired in the first image.

[0027] It should be noted that the automatic product defect repair method of this application embodiment can be applied to an automatic product defect repair system, which may include two independent workstations: an inspection station and a repair station. The inspection station can be a workstation used to perform initial optical inspection on the product to be repaired and record the defect location. To realize the function of the inspection station, this station may be equipped with an inspection camera (image acquisition device) and a motion platform. The repair station can be a workstation used to perform laser repair processing on the product to be repaired based on the inspection results of the inspection station; this station may be equipped with a repair camera and a laser.

[0028] In some embodiments of this application, before the product to be repaired enters the inspection station, the product to be repaired can be taken out of the material box by an automatic loading and unloading device (such as a robot or conveyor belt) and placed on the motion platform of the inspection station, so that the entire surface of the product to be repaired can be scanned or imaged by the inspection camera to obtain the first image of the product to be repaired.

[0029] It is understood that the aforementioned product to be repaired may be a product that needs to be tested for defects, such as printed circuit boards, ceramic thick film boards, or display panels, and this application embodiment does not limit this.

[0030] It should be noted that the first image can be a complete product image obtained by imaging the entire product to be repaired at the inspection station using an image acquisition device (such as an inspection camera).

[0031] It should be understood that the surface of the product to be repaired may typically have conductive lines or functional patterns. These lines or patterns may develop defects such as short circuits, open circuits, foreign objects, or scratches due to process variations during manufacturing. By performing defect identification on the first image, the abnormal area can be located from the image, and the pixel coordinates of the abnormal area can be determined.

[0032] It is understandable that the aforementioned first pixel coordinates are the positional information represented in pixels in the image coordinate system corresponding to the first image.

[0033] In one example, before entering the inspection station, the product to be repaired can be removed from the hopper by an automatic loading and unloading device and placed on a motion platform at the inspection station. The motion platform carries the product to be repaired and moves it to a position below the inspection camera. The inspection camera scans and images the product, generating a complete product image covering its entire surface. The repair equipment can use this complete product image as the first image. Next, the repair equipment performs defect identification processing on the first image, locating all abnormal areas from it using image analysis algorithms. For each located defect area, the position of a reference point within that defect area in the first image can be determined and recorded as pixel coordinates. The repair equipment can then identify the defect requiring repair from all detected defects and obtain the corresponding first pixel coordinates. For example, the reference point of the defect area could be the center point of the defect area or a corner point of the defect area's bounding box; the repair equipment can use the pixel coordinates of this reference point as the first pixel coordinates.

[0034] For ease of understanding, the following explanation uses examples, but does not limit the scope of this embodiment. For instance, suppose the product to be repaired is a printed circuit board (PCB). The suction nozzle of the automatic loading and unloading device picks up a PCB from the material box and places it on the motion platform of the inspection station. The motion platform carries the PCB and moves at a constant speed horizontally. Simultaneously, a line scan camera scans the surface of the PCB line by line, generating a complete grayscale image of the PCB (first image). The repair equipment uses this grayscale image as the first image. The repair equipment performs defect identification processing on the first image, locating multiple abnormal areas marked as defects. The repair equipment selects one defect area as the defect to be repaired and extracts the pixel coordinates of the center point of this defect area in the first image as (1240, 860). These pixel coordinates are the first pixel coordinates.

[0035] In some embodiments of this application, to avoid misidentification of defects, the step of performing defect identification on the first image and determining the first pixel coordinates of the defect to be repaired in the first image includes: performing defect identification on the first image using a target detection model to determine the defect attributes of the initial defect in the first image; wherein, the defect attributes include: defect category, defect area, the first pixel coordinates corresponding to the defect to be repaired, and confidence level; filtering the initial defect according to the defect attributes using preset filtering rules, and using the filtered initial defect as the defect to be repaired.

[0036] It is understood that the above-mentioned object detection model can be a model built based on a deep learning-based object detection algorithm, such as the YOLO series model, the SSD model, etc., and the embodiments of this application do not limit it.

[0037] It should be noted that the aforementioned initial defects can be all targets identified as defects by the target detection model after inference on the first image. These initial defects have not yet been screened and confirmed, including targets with actual defects as well as false targets generated due to image noise, background interference, or model misjudgment. If these initial defects are directly used as defects to be repaired for laser repair, the repair equipment will perform laser irradiation at locations where defects do not actually exist, causing unnecessary laser energy damage to the surface of the product to be repaired, resulting in product scrap or reduced yield. In addition, the target detection model may output multiple overlapping detection boxes for the same real defect. If these duplicate detection targets are not deduplicated, the repair equipment will perform laser irradiation multiple times on the same defect location, resulting in over-repair and further damage to the product substrate. Therefore, before using the detection results for subsequent coordinate transformation and laser repair, the aforementioned initial defects need to be screened to remove false detection targets and duplicate detection targets, retaining only the truly reliable defects as defects to be repaired.

[0038] It is understood that the aforementioned defect category can be information characterizing the type of defect, such as short circuit, open circuit, foreign object, or scratch type, etc., and this application embodiment does not limit this. The aforementioned defect area can be the size of the defect region expressed in pixels or physical area. The aforementioned confidence level can be a parameter used to characterize the reliability of the detection result, and this parameter can be output by the target detection model.

[0039] It should be understood that the aforementioned preset filtering rules can be a pre-defined set of screening conditions used to remove defects that do not meet the repair conditions or have low detection confidence from all initial defects output by the target detection model. Pre-defined filtering rules can be used to filter initial defects based on defect attributes, and the filtered initial defects can be used as defects to be repaired.

[0040] For example, the preset filtering rules include: an area filtering rule, which determines that when the area of ​​the initial defect is less than a first preset threshold, the initial defect is an unrepairable micro-defect and does not belong to the defects to be repaired; a confidence filtering rule, which determines that when the confidence of the initial defect is lower than a second preset threshold, the initial defect is a false detection result of the model or a detection result with low reliability and does not belong to the defects to be repaired; a duplicate target filtering rule, which determines that when the intersection-union ratio among multiple initial defects is greater than a third preset threshold, the multiple initial defects are duplicate detection targets corresponding to the same real defect, and only one of them is retained, while the rest do not belong to the defects to be repaired; and an edge filtering rule, which determines that when the initial defect is located within a fourth preset distance range of the edge of the first image, the initial defect may be a false defect caused by image stitching artifacts or incomplete cropping and does not belong to the defects to be repaired.

[0041] In some embodiments of this application, during the automated laser repair process, the defect coordinates obtained from the detection station and the repair station cannot be directly applied to the repair station due to differences in camera mounting positions, motion platform coordinate systems, and equipment structural positions. Therefore, it is necessary to establish a spatial correspondence between the detection image coordinate system and the repair image coordinate system so that the defect position output by the detection station can be accurately transferred to the repair station for processing.

[0042] Step S20: Convert the first pixel coordinates into the first motion platform coordinates in the motion platform coordinate system; Step S30: When the motion platform moves to the position corresponding to the coordinates of the first motion platform, acquire the second image of the product to be repaired, and determine the second pixel coordinates of the defect to be repaired in the second image.

[0043] It should be noted that the aforementioned motion platform coordinate system can be a three-dimensional spatial coordinate system corresponding to the motion platform, used to describe the position of the object in the real world. The aforementioned first motion platform coordinates can be position information expressed in physical length units, with the mechanical origin of the motion platform as the reference.

[0044] It is understood that the aforementioned motion platform can be a mechanical motion device used to carry the product to be repaired and drive the product to be repaired to move along the X-axis and Y-axis in the horizontal plane.

[0045] It should be noted that the second image mentioned above may be a local image obtained by imaging a local area containing the defect to be repaired on the product to be repaired using a repair camera at the repair station. The field of view covered by the second image is smaller than that covered by the first image.

[0046] It should be explained that the aforementioned second pixel coordinate information can be positional information expressed in pixels within the image coordinate system of the second image.

[0047] In some implementations, after obtaining the first pixel coordinates of the defect to be repaired, the repair equipment can first map the first pixel coordinates from the image coordinate system to the motion platform coordinate system according to a pre-established mapping relationship between the pixel coordinate system and the motion platform coordinate system, thus obtaining the first motion platform coordinates in physical length units under the motion platform coordinate system. This mapping relationship can be determined in advance through the calibration process of a calibration plate. Next, the repair equipment can send the first motion platform coordinates to the motion platform's control system, which drives the motion platform to move horizontally, moving the product to be repaired to the position corresponding to the first motion platform coordinates. After the motion platform is in position, the repair equipment controls the repair camera at the repair station to take a picture of the product to be repaired, acquiring a second image containing the local area where the defect to be repaired is located. After acquiring the second image, the repair equipment performs defect recognition processing on the second image, locating the position of the defect to be repaired from the second image and recording this position in pixel coordinate form, thus obtaining the second pixel coordinates.

[0048] For ease of understanding, the following explanation uses examples, but does not limit the scope of this embodiment. For example, the repair device obtains the first pixel coordinates of the defect to be repaired as (1240, 860). The repair device reads a pre-stored affine transformation matrix, substitutes the pixel coordinates into the affine transformation matrix, performs multiplication and addition operations, and obtains the motion platform coordinates in millimeters (150.5, 200.3) in the motion platform coordinate system. This coordinate is used as the first motion platform coordinate. The repair device sends the motion platform coordinates (150.5, 200.3) to the motion platform controller. The controller drives the X-axis and Y-axis motors to move the platform to the specified coordinate position. After the platform is in position, the repair device triggers an area scan camera to take a picture of a local area of ​​the circuit board at that position, obtaining a magnified image with a field of view of 0.72 mm by 0.86 mm. This magnified image is used as the second image. The repair device then performs defect recognition processing on the second image, locates the defect to be repaired in the second image, and outputs the pixel coordinates of the defect in the second image as (320, 240), which is used as the second pixel coordinates.

[0049] In some embodiments of this application, the embodiments of this application Step S40: Determine the offset of the motion platform based on the second pixel coordinates, and control the laser to automatically repair the product to be repaired based on the offset.

[0050] It should be noted that this offset can be used to characterize the physical distance between the defect to be repaired and the current actual stopping position of the motion platform in the second image. The aforementioned laser can be a device for emitting a laser beam to repair product defects.

[0051] In practical applications, after obtaining the second pixel coordinates of the defect to be repaired, the repair equipment can first determine the offset between the current actual stopping position of the motion platform and the true position of the defect. Then, the repair equipment can control the laser to perform automatic repair operations based on the offset. Specifically, the repair equipment corrects the first motion platform coordinates based on the offset, superimposing the offset onto the first motion platform coordinates to obtain the corrected second motion platform coordinates. These second motion platform coordinates are then sent to the motion platform's control system, which drives the motion platform to move to the position corresponding to the second motion platform coordinates. After the motion platform reaches the position corresponding to the second motion platform coordinates, the repair equipment controls the laser to turn on, causing the laser beam to irradiate the location of the defect to be repaired, completing the repair process. After the repair is completed, the repair equipment also controls the laser to turn off and records the current defect repair result.

[0052] This application embodiment acquires a first image of the product to be repaired, identifies defects in the first image, and determines the first pixel coordinates of the defect in the first image. The first pixel coordinates are then converted to first motion platform coordinates in the motion platform coordinate system. With the motion platform moving to the position corresponding to the first motion platform coordinates, a second image of the product to be repaired is acquired, and the second pixel coordinates of the defect in the second image are determined. The offset of the motion platform is determined based on the second pixel coordinates, and the laser is controlled to automatically repair the product based on the offset. This solution employs a two-level positioning architecture. The large-field-of-view camera at the inspection station provides coarse positioning coordinates covering the entire product area, enabling the platform to move to the macroscopic region where the defect is located. The small-field-of-view camera at the repair station performs secondary imaging within this region, calculating the offset using the center of the second image as a reference. Since the second image is taken after the platform has moved to the coarse positioning position, the image center corresponds to the platform's current actual stopping position. The pixel difference of the defect deviating from the center in the image, multiplied by the pixel equivalent, directly equals the offset in physical space. This offset does not need to be converted back to the motion platform coordinate system and can be directly superimposed onto the first motion platform coordinates for correction, avoiding the additional errors introduced by the secondary coordinate conversion. The motion platform only needs to perform two movements to accurately align the defect with the laser focus without repeated attempts in physical space, thus balancing repair accuracy and efficiency and achieving automatic defect repair.

[0053] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in Embodiment 1 above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 , Figure 2 This is a flowchart illustrating the second embodiment of the automatic product defect repair method of this application.

[0054] like Figure 2As shown in the embodiment of this application, before the step of performing defect identification on the first image and determining the first pixel coordinates of the defect to be repaired in the first image, the method further includes: Step S110: Perform feature recognition on the first image to obtain the first image features; Wherein, the first image is a complete product image of the product to be repaired; Step S120: Determine the first position of the first feature point and the second position of the second feature point in the first image features; Wherein, the first feature point serves as the reference positioning point of the product to be repaired, used to determine the reference position of the product to be repaired in the detection image coordinate system; the second feature point serves as the angle reference point of the product to be repaired, used to determine the rotational offset of the product to be repaired relative to the detection image coordinate system; Step S130: Determine the detection position and detection angle of the product to be repaired at the detection station based on the first position and the second position.

[0055] It is understood that the aforementioned first image features may be a set of visual information extracted from the first image, such as contours, corners, edges, or specific patterns.

[0056] It should be noted that the aforementioned first feature point can be a pre-defined position mark on the product to be repaired. This position mark can serve as a reference positioning point for identifying the position, such as a positioning hole, copper foil cross mark, or solder pad corner point on the product. This first feature point can be used to determine the reference position of the product. The aforementioned second feature point can be another position mark with a fixed geometric relationship to the first feature point. This second feature point can serve as an angular reference point, used to determine the rotational offset of the product to be repaired relative to the coordinate system by the direction of the line connecting the second feature point and the first feature point.

[0057] In some embodiments of this application, the positional relationship between two feature points can be used to determine the positional status of the product to be repaired in the inspection station: ; in, The coordinates of the first feature point are: The coordinates of the second feature point are... The rotation angle of the product to be repaired relative to the detection coordinate system.

[0058] It should be noted that for any defect to be repaired, its corresponding pixel coordinates in the detection image coordinate system can be expressed as: The repair device can obtain the center pixel coordinates of the first image. And based on this, determine the CCD coordinates in the physical coordinate system of the image: ; ; in, Represents pixel equivalent, which indicates the specific size (unit can be millimeters) of a pixel in physical space.

[0059] Furthermore, based on the pre-calibrated mapping matrix M, the CCD coordinates can be transformed into the first motion platform coordinates in the motion platform coordinate system. This matrix can be an affine transformation matrix, containing transformation parameters such as rotation, translation, scaling, and shearing. The transformation process can be as follows: ; in, That is, CCD coordinates , Coordinates of the first motion platform .

[0060] Furthermore, by superimposing the platform compensation value and offset correction onto the coordinates of the first motion platform, the final laser repair coordinates can be obtained.

[0061] It should be understood that the above-mentioned detection position can be the translation of the product to be repaired at the detection station relative to the origin of the detection image coordinate system, and the detection angle can be the rotation of the product to be repaired at the detection station relative to the horizontal axis of the detection image coordinate system.

[0062] It should be noted that the aforementioned detection image coordinate system can be a coordinate system associated with the first image, which can be used to describe the spatial relationships of various positions in the first image. For example, the origin of the detection image coordinate system can be located at the top left corner of the first image, with the horizontal direction to the right being the positive x-axis and the vertical direction downwards being the positive y-axis, and the coordinate values ​​are in pixels.

[0063] In its implementation, after acquiring a first image of the product to be repaired, the repair equipment performs feature recognition processing on the first image to extract all identifiable visual features of the product, thus obtaining the first image features. Subsequently, it searches for pre-defined first and second feature points within the first image features, obtaining the first position of the first feature point and the second position of the second feature point in the first image, respectively. The first feature point serves as the reference positioning point for the product to be repaired, and its first position represents the reference position of the product in the detection image coordinate system. The second feature point serves as the angular reference point for the product to be repaired, and the repair equipment can calculate the rotation angle of the product relative to the detection image coordinate system based on the coordinate difference between the first and second positions. Specifically, the repair equipment can calculate the difference in the horizontal and vertical coordinates of the second feature point relative to the first feature point, and calculate the angle value using the ratio of these differences. After completing the above calculations, the repair equipment uses the first position as the detection position of the product and the calculated angle as the detection angle of the product, thereby determining the position of the product to be repaired at the detection station.

[0064] In some embodiments of this application, the step of determining the second pixel coordinates of the defect to be repaired in the second image includes: determining a third position of the first feature point and a fourth position of the second feature point in the second image; determining the repair position and repair angle of the product to be repaired at the repair station based on the third position and the fourth position; determining the positional deviation of the product to be repaired as it moves from the detection station to the repair station based on the detection position and the repair position; determining the angular deviation of the product to be repaired as it moves from the detection station to the repair station based on the detection angle and the repair angle; and locating the second pixel coordinates corresponding to the defect to be repaired in the second image based on the positional deviation and the angular deviation.

[0065] It should be noted that the third position mentioned above can be the pixel coordinates of the first feature point in the second image, and the fourth position mentioned above can be the pixel coordinates of the second feature point in the second image.

[0066] It is understandable that the above-mentioned repair position is the physical position of the first feature point on the repair station, and the above-mentioned repair angle is the angle value determined according to the direction of the line connecting the third position and the fourth position, which is used to characterize the degree of rotation of the product to be repaired relative to the horizontal axis of the repair image coordinate system on the repair station.

[0067] It should be understood that the above-mentioned repaired image coordinate system, which is the coordinate system associated with the second image, can be used to describe the spatial relationships of various locations in the second image.

[0068] It should be noted that after the product to be repaired completes inspection at the inspection station, it can be moved from the inspection station to the repair station for repair. During the movement, the automatic loading and unloading device can remove the product to be repaired from the motion platform at the inspection station and place it on the motion platform at the repair station. Due to robotic arm gripping errors, conveyor belt positioning errors, or slight slippage between the product and the motion platform during the two placements, the actual position and angle of the product to be repaired at the repair station are not completely consistent with its position and angle at the inspection station. That is, the product undergoes translational and rotational offsets during the handling process. The aforementioned positional and angular deviations are quantitative descriptions of the translational and rotational amounts generated during this handling process.

[0069] Understandably, since the defect to be repaired is fixed to the product surface, its position relative to the product remains unchanged when the product as a whole is translated or rotated. However, the absolute pixel coordinates of the defect in the image coordinate system change with the translation and rotation of the product. Directly using the first pixel coordinates of the defect in the first image as the defect position in the second image would lead to positioning errors. Therefore, we can first determine the positional and angular deviations of the product during its movement. By using these deviations as coordinate corrections for the defect, we can perform rotational and translational corrections on the defect's coordinates in the second image, matching the defect coordinates with the actual posture of the product at the repair station, thereby accurately locating the defect in the second image.

[0070] This embodiment of the application obtains first image features by performing feature recognition on a first image; wherein, the first image is a complete product image of the product to be repaired; a first position of a first feature point and a second position of a second feature point are determined in the first image features; wherein, the first feature point serves as a reference positioning point for the product to be repaired, used to determine the reference position of the product to be repaired in the detection image coordinate system; the second feature point serves as an angle reference point for the product to be repaired, used to determine the rotational offset of the product to be repaired relative to the detection image coordinate system; the detection position and detection angle of the product to be repaired at the detection station are determined based on the first position and the second position. Since feature recognition is performed on the first image at the detection station, the preset first feature point and second feature point on the product to be repaired are identified. The first feature point serves as a reference positioning point, providing the reference position of the product in the detection image coordinate system, and the second feature point serves as an angle reference point, with the rotational angle of the product relative to the detection image coordinate system determined by the direction of the line connecting this point and the first feature point. Based on these two feature points, a product local coordinate system can be established with the first feature point as the origin. No matter how the product is translated or rotated at the inspection station, the coordinate value of the defect in the product local coordinate system remains unchanged, providing a position and angle reference for the construction of the mapping map. This makes the defect coordinates no longer dependent on the absolute position of the product in the image, reducing the requirements for loading accuracy.

[0071] Based on the first and / or second embodiments of this application, in the third embodiment of this application, the content that is the same as or similar to the first and / or second embodiments described above can be referred to the above description and will not be repeated hereafter. Based on this, please refer to... Figure 3 , Figure 3 This is a flowchart illustrating Embodiment 3 of the automatic product defect repair method of this application.

[0072] like Figure 3 As shown in the embodiment of this application, the step of determining the offset of the motion platform based on the second pixel coordinates includes: Step S41: Obtain the center pixel coordinates of the second image and determine the pixel difference between the second pixel coordinates and the center pixel coordinates; Step S42: Convert the pixel difference into a physical distance and use the physical distance as the offset of the motion platform.

[0073] It should be noted that the aforementioned center pixel coordinates are the pixel coordinates of the center position of the second image in the coordinate system of the repaired image. Since the second image is captured by the repair camera after the motion platform moves to the position corresponding to the coordinates of the first motion platform, the center pixel coordinates of the second image correspond to the physical position of the center of the optical axis of the repair camera pointing to the surface of the product to be repaired, that is, the mapping point of the current actual stopping position of the motion platform in the second image.

[0074] It is understandable that the aforementioned pixel difference can be the horizontal pixel difference obtained by subtracting the horizontal coordinate of the second pixel coordinate from the horizontal coordinate of the center pixel coordinate, and the vertical pixel difference obtained by subtracting the vertical coordinate of the second pixel coordinate from the vertical coordinate of the center pixel coordinate. This pixel difference, expressed in pixels, characterizes the degree to which the defect to be repaired deviates from the image center in the second image. The physical meaning of the pixel difference between the second pixel coordinate and the center pixel coordinate is that the center of the second image corresponds to the optical axis center of the repair camera, which is the imaging point corresponding to the current actual stopping position of the motion platform. Therefore, this pixel difference directly reflects the deviation between the true location of the defect and the current stopping position of the platform on the image plane.

[0075] It should be understood that the physical distance mentioned above is the distance value obtained by multiplying the pixel difference by the pixel equivalent.

[0076] In its implementation, after acquiring the center pixel coordinates of the second image, the repair device subtracts the center pixel coordinates from the second pixel coordinates of the defect to be repaired, obtaining the horizontal and vertical pixel differences between the second pixel coordinates and the center pixel coordinates. Subsequently, the repair device multiplies the horizontal pixel difference by a preset pixel equivalent to obtain the horizontal physical distance in millimeters; it also multiplies the vertical pixel difference by the pixel equivalent to obtain the vertical physical distance in millimeters. The repair device uses these horizontal and vertical physical distances together as the offsets of the motion platform along the X and Y axes in the horizontal plane. These offsets can be directly used to correct the position of the motion platform; that is, starting from its current actual stopping position, the motion platform moves along the X and Y axes by the aforementioned horizontal and vertical physical distances respectively, aligning the optical axis center of the repair camera with the actual position of the defect to be repaired.

[0077] In some embodiments of this application, the laser focusing process can be referred to Figure 4 , Figure 4 This is a schematic diagram of the focusing process for automatic repair in one implementation of the automatic product defect repair method of this application. In this embodiment, the step of automatically repairing the product to be repaired by controlling the laser based on the offset includes: correcting the coordinates of the first motion platform according to the offset to obtain the coordinates of the second motion platform; controlling the motion platform to move to the position corresponding to the second motion platform coordinates; and controlling the laser to automatically repair the product to be repaired when the motion platform moves to the position corresponding to the second motion platform coordinates.

[0078] It should be noted that when the motion platform moves to the position corresponding to the coordinates of the second motion platform, the laser can be controlled to automatically repair the product to be repaired. Specifically, when the motion platform moves to the position corresponding to the coordinates of the second motion platform, the repair device can initiate an automatic focusing process. Automatic focusing can be achieved using a Z-axis scanning method: the repair device controls the motion platform to move along the Z-axis from the starting position, gradually moving the Z-axis to the set position. During the movement, for each preset step (e.g., 0.002 mm), the repair camera is controlled to acquire one focused image (i.e., continuously moving and taking pictures within a certain range along the Z-axis). The repair device calculates the image sharpness value for each acquired focused image. The sharpness value can be calculated by performing a convolution operation on the image using the Laplacian operator. The repair device records the Z-axis coordinate value corresponding to the position with the highest sharpness value among all acquired positions, uses this Z-axis coordinate value as the target focusing position, and controls the motion platform to move to this target focusing position (i.e., find and record the Z-axis position with the highest sharpness) to complete coarse focusing. This target focusing position is the coarse focus point position.

[0079] Furthermore, based on the aforementioned coarse focusing, the restoration equipment can control the Z-axis to move three times in one direction from the coarse focus point position to capture images and determine if the sharpness continuously decreases. If so, it reverts and readjusts the direction to achieve reverse focusing. If not, the coarse focus point position can be taken as the precise focus position, i.e., the focus position has been found, and the Z-axis is moved to the optimal focus position (i.e., the precise focus position). After the motion platform moves to the optimal focus position, the restoration equipment can send an activation command to the laser. The laser, according to the received activation command, emits a laser beam according to preset restoration parameters.

[0080] For example, the preset repair parameters include laser power, pulse width, frequency, and duration: laser power controls the power of the laser output beam, in watts; pulse width controls the duration of a single laser pulse, in microseconds; frequency controls the number of laser pulses emitted by the laser per unit time, in kilohertz; and duration controls the total duration for which the laser continuously emits light and irradiates the defect to be repaired, in milliseconds.

[0081] It should be noted that the above-mentioned preset repair parameters are read from the pre-stored formula according to the defect category of the defect to be repaired. After the laser beam irradiates the location of the defect to be repaired, the defective material is melted, vaporized, or ablated through the laser thermal effect, thereby completing the physical repair of the defect.

[0082] In some embodiments of this application, after the laser emits light for a preset duration, the repair device sends a shutdown command to the laser, and the laser stops emitting light. After the repair is completed, the repair device records the repair time of the current defect and stores the repair result (e.g., "repair successful" or "repair failed") in a local storage medium. The repair device also records the repair location of the current defect in a mapping graph for subsequent data traceability and quality analysis.

[0083] This embodiment of the application obtains the center pixel coordinates of the second image and determines the pixel difference between the second pixel coordinates and the center pixel coordinates; the pixel difference is converted into a physical distance, and the physical distance is used as the offset of the motion platform. Since the offset is calculated with the center of the second image as a reference, the additional error caused by converting the second pixel coordinates back to the absolute position in the world coordinate system and comparing it with the coordinates of the first motion platform is avoided.

[0084] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the automatic product defect repair method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0085] This application also provides an automatic product defect repair device; please refer to... Figure 5 , Figure 5 This is a schematic diagram of the module structure of the automatic product defect repair device according to an embodiment of this application. The automatic product defect repair device includes: The defect identification module 10 is used to acquire a first image of the product to be repaired, and to identify defects in the first image to determine the first pixel coordinates of the defect to be repaired in the first image. Coordinate transformation module 20 is used to convert the first pixel coordinates into first motion platform coordinates in the motion platform coordinate system; The position alignment module 30 is used to acquire a second image of the product to be repaired and determine the second pixel coordinates of the defect to be repaired in the second image when the motion platform moves to the position corresponding to the coordinates of the first motion platform. The automatic repair module 40 is used to determine the offset of the motion platform based on the second pixel coordinates, and control the laser to automatically repair the product to be repaired based on the offset.

[0086] The automatic product defect repair device provided in this application, employing the automatic product defect repair method described in the above embodiments, can solve the technical problems of low efficiency and poor accuracy in existing automatic product defect repair methods. Compared with the prior art, the beneficial effects of the automatic product defect repair device provided in this application are the same as those of the automatic product defect repair method described in the above embodiments, and other technical features in the automatic product defect repair device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0087] This application provides an automatic product defect repair device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the automatic product defect repair method in Embodiment 1 above.

[0088] The following is for reference. Figure 5 The diagram illustrates a structural schematic suitable for implementing an automatic product defect repair device according to embodiments of this application. The automatic product defect repair device in embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), tablet computers, PMPs (Portable Media Players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 5 The product defect automatic repair device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0089] like Figure 5As shown, the automatic product defect repair device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the automatic product defect repair device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows the automated product defect repair device to communicate wirelessly or wiredly with other devices to exchange data. While the figure shows automated product defect repair devices with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0090] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0091] The automatic product defect repair device provided in this application, employing the automatic product defect repair method described in the above embodiments, can solve the technical problems of low efficiency and poor accuracy in existing automatic product defect repair methods. Compared with the prior art, the beneficial effects of the automatic product defect repair device provided in this application are the same as those of the automatic product defect repair method described in the above embodiments, and other technical features of this automatic product defect repair device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0092] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0093] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0094] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the product defect automatic repair method in the above embodiments.

[0095] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or flash memory, optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0096] The aforementioned computer-readable storage medium may be included in the automatic product defect repair device; or it may exist independently and not assembled into the automatic product defect repair device.

[0097] The aforementioned computer-readable storage medium carries one or more programs that, when executed by the automatic product defect repair device, cause the automatic product defect repair device to: Acquire a first image of the product to be repaired, and perform defect identification on the first image to determine the first pixel coordinates of the defect to be repaired in the first image; Convert the first pixel coordinates to the first motion platform coordinates in the motion platform coordinate system; When the motion platform moves to the position corresponding to the coordinates of the first motion platform, a second image of the product to be repaired is acquired, and the second pixel coordinates of the defect to be repaired in the second image are determined; The offset of the motion platform is determined based on the second pixel coordinates, and the laser is controlled to automatically repair the product to be repaired based on the offset.

[0098] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0099] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation that may be implemented in systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0100] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0101] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described automatic product defect repair method, thereby solving the technical problems of low efficiency and poor accuracy in existing automatic product defect repair methods. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the automatic product defect repair method provided in the above embodiments, and will not be repeated here.

[0102] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the product defect automatic repair method described above.

[0103] The computer program product provided in this application can solve the technical problems of low efficiency and poor accuracy of existing automatic product defect repair methods. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the automatic product defect repair methods provided in the above embodiments, and will not be repeated here.

[0104] The above description is only a part of the embodiments of this application and does not limit the scope of protection of this application. All equivalent structural transformations made under the technical concept of this application and using the content of this application specification and drawings, or direct / indirect applications in other related technical fields, are included in the scope of protection of this application.

Claims

1. A method for automatic repair of product defects, characterized in that, The method includes: Acquire a first image of the product to be repaired, and perform defect identification on the first image to determine the first pixel coordinates of the defect to be repaired in the first image; Convert the first pixel coordinates to the first motion platform coordinates in the motion platform coordinate system; When the motion platform moves to the position corresponding to the coordinates of the first motion platform, a second image of the product to be repaired is acquired, and the second pixel coordinates of the defect to be repaired in the second image are determined; The offset of the motion platform is determined based on the second pixel coordinates, and the laser is controlled to automatically repair the product to be repaired based on the offset.

2. The automatic product defect repair method as described in claim 1, characterized in that, Before the step of performing defect identification on the first image and determining the first pixel coordinates of the defect to be repaired in the first image, the method further includes: The first image is subjected to feature recognition to obtain the first image features; wherein, the first image is a complete product image of the product to be repaired; In the first image features, a first position of a first feature point and a second position of a second feature point are determined; wherein, the first feature point serves as a reference positioning point for the product to be repaired, used to determine the reference position of the product to be repaired in the detection image coordinate system; the second feature point serves as an angle reference point for the product to be repaired, used to determine the rotational offset of the product to be repaired relative to the detection image coordinate system; The detection position and detection angle of the product to be repaired at the detection station are determined based on the first position and the second position.

3. The automatic product defect repair method as described in claim 2, characterized in that, The step of determining the second pixel coordinates of the defect to be repaired in the second image includes: Determine the third position of the first feature point and the fourth position of the second feature point in the second image; The repair position and repair angle of the product to be repaired at the repair station are determined based on the third position and the fourth position. The positional deviation of the product to be repaired as it moves from the detection station to the repair station is determined based on the detection location and the repair location. The angular deviation of the product to be repaired as it moves from the inspection station to the repair station is determined based on the detection angle and the repair angle. Based on the positional deviation and the angular deviation, locate the second pixel coordinates corresponding to the defect to be repaired in the second image.

4. The automatic product defect repair method as described in claim 1, characterized in that, The step of determining the offset of the motion platform based on the second pixel coordinates includes: Obtain the center pixel coordinates of the second image, and determine the pixel difference between the second pixel coordinates and the center pixel coordinates; The pixel difference is converted into a physical distance, and the physical distance is used as the offset of the motion platform.

5. The automatic product defect repair method as described in claim 1, characterized in that, The step of automatically repairing the product to be repaired based on the offset-controlled laser includes: The coordinates of the first motion platform are corrected based on the offset to obtain the coordinates of the second motion platform; Control the motion platform to move to the position corresponding to the coordinates of the second motion platform; When the motion platform moves to the position corresponding to the coordinates of the second motion platform, the laser is controlled to automatically repair the product to be repaired.

6. The automatic product defect repair method as described in claim 1, characterized in that, The step of performing defect identification on the first image and determining the first pixel coordinates of the defect to be repaired in the first image includes: The first image is used to identify defects by a target detection model to determine the defect attributes of the initial defects in the first image; wherein, the defect attributes include: defect category, defect area, the first pixel coordinates corresponding to the defect to be repaired, and confidence level; The initial defects are filtered according to preset filtering rules and the defect attributes, and the filtered initial defects are taken as defects to be repaired.

7. An automatic product defect repair device, characterized in that, The automatic product defect repair device includes: The defect identification module is used to acquire a first image of the product to be repaired, and to identify defects in the first image to determine the first pixel coordinates of the defect to be repaired in the first image. The coordinate transformation module is used to convert the coordinates of the first pixel into the coordinates of the first motion platform in the motion platform coordinate system. The position alignment module is used to acquire a second image of the product to be repaired and determine the second pixel coordinates of the defect to be repaired in the second image when the motion platform moves to the position corresponding to the coordinates of the first motion platform. An automatic repair module is used to determine the offset of the motion platform based on the second pixel coordinates, and control the laser to automatically repair the product to be repaired based on the offset.

8. An automatic product defect repair device, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the product defect automatic repair method as described in any one of claims 1 to 6.

9. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the automatic product defect repair method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the automatic product defect repair method as described in any one of claims 1 to 6.

Images