Electrode sheet burr detection method and apparatus, electronic device, and storage medium

Abstract

Disclosed in the present application are an electrode sheet burr detection method and apparatus, an electronic device, and a storage medium. In the present application, after an electrode sheet end face image obtained after battery electrode sheet slitting is acquired, a molten region is segmented from the electrode sheet end face image to obtain a molten region image; and burrs in the molten region are identified on the basis of a molten region height value corresponding to the molten region in the molten region image and a preset burr height threshold. By means of the method described above, the present application can prevent the problem of inaccurate burr identification caused by the relatively high similarity between features of the molten region and features of an active material, thereby improving the burr identification accuracy.

Description

Pole burr detection method, device, electronic device and storage medium

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to Chinese patent application 202311753441.4, filed on December 19, 2023, entitled “Method, device, electronic device and storage medium for detecting pole piece burrs,” and the entire contents of that application are incorporated herein by reference. Technical Field

[0003] The present application relates to the field of battery technology, and in particular to a method, device, electronic device and storage medium for detecting electrode burrs. Background Art

[0004] Laser die-cutting is widely used in the industrial manufacturing of various lithium-ion batteries. Laser die-cutting ensures that the electrode sheets achieve the desired design shape. However, during the laser die-cutting process, fluctuations in laser power or abnormalities in the incoming material can cause burr defects on the electrode sheets, which in turn affect the quality of the lithium-ion batteries. Therefore, burr detection is of great significance.

[0005] However, the burr detection method in the related art has the problem of low detection accuracy when identifying burrs on the end face of the pole piece.

[0006] Summary of the Invention

[0007] The present application provides a method, device, electronic device and storage medium for detecting electrode burrs, which can avoid the problem of inaccurate burr identification due to the high similarity between the characteristics of the molten area and the active material, thereby improving the accuracy of burr identification.

[0008] In the first aspect, the present application provides a method for detecting electrode burrs, the method comprising: obtaining an image of the electrode end face after the battery electrode is cut; segmenting the melting area from the electrode end face image to obtain a melting area image; and identifying burrs in the melting area based on the melting area height value corresponding to the melting area in the melting area image and a preset burr height threshold.

[0009] In the embodiments of the present application, by separating the melt region from the electrode end face image, the melt region is automatically identified, thereby avoiding the low melt region identification accuracy problem that exists in manual visual identification, thereby improving not only the melt region identification accuracy but also the melt region identification efficiency. Furthermore, after the melt region is identified, the burrs in the melt region can be identified by comparing the melt region height value with a preset burr height threshold, thereby avoiding the burr labeling efficiency and low accuracy caused by manual burr labeling, and improving the accuracy and efficiency of burr detection.

[0010] In some embodiments, obtaining an image of the end face of a battery pole piece after the pole piece is cut includes: obtaining an initial pole piece end face image that at least includes the pole piece end face; and cropping the initial pole piece end face image based on the position and / or proportion of the pole piece end face in the initial pole piece end face image to obtain a pole piece end face image.

[0011] By cropping the pole piece end face image, the recognition accuracy of the molten area in the pole piece end face image is improved, thereby improving the detection accuracy of burr detection.

[0012] In some embodiments, segmenting the melting region from the pole piece end face image to obtain a melting region image includes: using a region recognition model to identify the melting region from the pole piece end face image; segmenting the melting region from the pole piece end face image to obtain a melting region image, wherein the melting region image is a binary image, and the binary image includes the melting region and the background region.

[0013] The use of a region recognition model enables automatic identification of molten regions, avoiding the low accuracy of manual identification, improving both the accuracy and efficiency of molten region identification, and thus the accuracy and efficiency of burr detection. Furthermore, because molten region images contain both molten and background regions, and these two regions are highly distinguishable in molten region images, identifying burrs based on molten region images avoids inaccurate burr identification due to the high similarity between the characteristics of the molten region and the active material, thereby improving burr recognition accuracy.

[0014] In some embodiments, burrs in the melted area are identified based on a melted area height value corresponding to the melted area in the melted area image and a preset burr height threshold, including: sampling the melted area in the melted area image to obtain multiple sampling points; identifying burrs in the melted area based on a melted area height value corresponding to each sampling point and a preset burr height threshold.

[0015] By comparing the melting area height value of the melting area with the preset burr height threshold, the burrs in the melting area can be identified, thereby avoiding the problems of low burr marking efficiency and accuracy caused by manual burr marking, and improving the accuracy and efficiency of burr detection.

[0016] In some embodiments, the melting area in the melting area image is sampled to obtain a plurality of sampling points, including: performing connected domain processing on the melting area image to obtain a target melting area image; sampling the melting area in the target melting area image along a first direction to obtain a plurality of sampling points, wherein the first direction is used to characterize the extension direction of the melting area on the end face of the pole piece.

[0017] By performing connected domain processing on the melt region image, the consistency of each image region in the melt region image is ensured, reducing the risk of disconnection in the melt region during burr defect identification, thereby improving the detection accuracy of burr defects. Multi-point sampling in the lateral direction of the melt region can reduce the randomness of burr defect detection and reduce the risk of missed burr detection while ensuring a reasonable burr detection time.

[0018] In some embodiments, burrs in the melting area are identified based on the melting area height value corresponding to each sampling point and a preset burr height threshold, including: obtaining the melting area height value corresponding to each sampling point in the second direction to obtain the melting area height value of each sampling point, wherein the second direction is perpendicular to the first direction; comparing the melting area height values ​​of multiple sampling points to obtain a maximum height value; and determining that burrs exist in the melting area when the maximum height value is greater than the preset burr height threshold.

[0019] The maximum height of the molten area is compared with a preset burr height threshold to determine whether burrs are present in the molten area, improving the accuracy of burr defect detection. Furthermore, by comparing only the maximum height with the preset burr height threshold, rather than comparing the molten area height values ​​at all sampling points, the number of comparisons can be reduced, improving comparison efficiency and, consequently, enhancing the efficiency of burr detection.

[0020] In some embodiments, after determining that there are burrs in the melting region, the melting region height value corresponding to each sampling point is compared with a preset burr height threshold to obtain a comparison result; a target sampling point is determined from multiple sampling points based on the comparison result, wherein the target sampling point is a sampling point whose melting region height value is greater than the preset burr height threshold; and the position of the target sampling point in the melting region is determined as the position of the burr in the melting region.

[0021] After detecting the presence of a burr defect in the molten area, further detecting the specific location of the burr in the molten area can improve the efficiency and accuracy of burr detection. Based on the identification of whether there is a burr defect in the molten area, by determining the specific defect location when a burr defect is identified, the cause of the burr defect can be analyzed, and then the laser die-cutting equipment or the manufacturing process of the pole piece can be improved to improve the production quality of the pole piece, and thus improve the production quality of the battery.

[0022] In some embodiments, after comparing the melting area height values ​​of multiple sampling points to obtain the maximum height value, if there is no target sampling point among the multiple sampling points, the average of the melting area height values ​​of the multiple sampling points is calculated to obtain the average height value of the melting area; based on the average height value of the melting area and the preset melting area height range, it is determined whether the quality of the battery electrode meets the target requirements.

[0023] In the case that there are no burr defects in the molten area, further product compliance inspection is carried out on the molten area to improve the quality of the battery pole pieces and thus improve the production quality of the battery.

[0024] In some embodiments, after obtaining the height value of the melting area corresponding to each sampling point in the second direction and obtaining the height value of the melting area of ​​each sampling point, a statistical analysis is performed on the height values ​​of the melting areas corresponding to multiple sampling points to determine the abnormal height range corresponding to the abnormal sampling point; based on the abnormal height range, an abnormal sampling point is identified from the multiple sampling points, wherein the height value of the melting area of ​​the abnormal sampling point is within the abnormal height range; and the abnormal sampling point is filtered out from the multiple sampling points to obtain multiple filtered sampling points.

[0025] By identifying and filtering out abnormal sampling points in the molten area image, the influence of abnormal sampling points on burr detection is avoided, thereby improving the accuracy of burr detection.

[0026] In some embodiments, the region recognition model includes at least a U-Net model.

[0027] By using the U-Net model to identify the molten area in the electrode end face image, the recognition cost of the molten area is reduced.

[0028] In the second aspect, the present application also provides a device for detecting burrs on a pole piece, which includes: an image acquisition module, configured to acquire an image of the pole piece end face after the battery pole piece is cut; a melt identification module, configured to segment the melt area from the pole piece end face image to obtain a melt area image; a burr identification module, configured to identify burrs in the melt area based on the melt area height value corresponding to the melt area in the melt area image and a preset burr height threshold.

[0029] In a third aspect, the present application further provides an electronic device comprising: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the method for detecting pole piece burrs as described in the first aspect is implemented.

[0030] In a fourth aspect, the present application further provides a readable storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, the method for detecting pole piece burrs as described in the first aspect is implemented.

[0031] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The features, advantages and technical effects of exemplary embodiments of the present application will be described below with reference to the accompanying drawings.

[0033] FIG1 is a schematic diagram of a pole piece end face in the related art;

[0034] FIG2 is a schematic diagram of a pole piece end face in the related art;

[0035] FIG3 is a schematic diagram of an application environment of a method for detecting electrode burrs according to an embodiment of the present application;

[0036] FIG4 is a flow chart of a method for detecting electrode burrs according to an embodiment of the present application;

[0037] FIG5 is a schematic diagram of an image of a molten area according to an embodiment of the present application;

[0038] FIG6 is a schematic diagram of the structure of a U-Net model according to an embodiment of the present application;

[0039] FIG7 is a schematic diagram of an image of a melted area after sampling according to an embodiment of the present application;

[0040] FIG8 is a schematic diagram of an image of a melted area after sampling and filtering out abnormal sampling points according to an embodiment of the present application;

[0041] FIG9 is a schematic diagram of a device for detecting electrode burrs according to another embodiment of the present application;

[0042] FIG10 is a schematic diagram of the hardware structure of an electronic device according to another embodiment of the present application.

[0043] In the accompanying drawings, the drawings are not necessarily drawn to scale. DETAILED DESCRIPTION

[0044] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned figure descriptions are intended to cover non-exclusive inclusions.

[0046] In the description of the embodiments of this application, the technical terms "first" and "second" are used only to distinguish different objects and should not be understood to indicate or imply relative importance or implicitly specify the quantity, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, the meaning of "plurality" is more than two, unless otherwise clearly and specifically defined.

[0047] Reference to an "embodiment" in the embodiments of the present application means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present 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 with other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in the embodiments of the present application may be combined with other embodiments.

[0048] In the description of the embodiments of this application, the term "and / or" is simply a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent the following three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0049] In the description of the embodiments of the present application, the term "multiple" refers to more than two (including two). Similarly, "multiple groups" refers to more than two groups (including two groups), and "multiple pieces" refers to more than two pieces (including two pieces).

[0050] In the description of the embodiments of the present application, the technical terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the embodiments of the present application.

[0051] In the description of the embodiments of the present application, unless otherwise expressly specified or limited, technical terms such as "installed," "connected," "connected," and "fixed" should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integration; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; internal connections between two components or interactions between two components. Those skilled in the art can understand the specific meanings of the above terms in the embodiments of the present application based on specific circumstances.

[0052] Laser die-cutting is widely used in the industrial manufacturing of various lithium-ion batteries. Laser die-cutting of lithium-ion battery pole pieces can achieve the desired shape after die-cutting. However, during the laser die-cutting process, laser power fluctuations or material anomalies can cause burrs on the end faces of the pole pieces, thus affecting the quality of the lithium-ion batteries.

[0053] Traditional burr detection schemes usually have a good detection rate for protruding planar burrs. For example, pulse cameras and gray point cameras can be used to capture images, and linear fitting technology can be used to judge and measure burrs. However, the above method is only applicable to the detection of thin film burrs and cannot be applied to the detection of burrs on the end faces of multi-layer structures. Moreover, the end face burr scene is complex, and most of them require manual visual inspection through a microscope. In addition, the characteristics of the molten area where the burr is located are highly similar to the characteristics of the active materials covered above and below it. For example, in the end face image of the pole piece shown in Figure 1, the molten area is highly similar to the active material area, and the way the human eye identifies the molten area is prone to misidentification. In related technologies, manual annotation is usually used to mark the burrs on the end face of the pole piece. For example, in the end face image of the pole piece shown in Figure 2, the inspector marked the height of the burr at position [1] and position [2]. For example, the burr height at position [1] is 35.0μm, and the burr height at position [2] is 31.1μm. The above manual burr annotation method relies on manual experience, and the annotation is unstable, resulting in a low burr detection accuracy. At the same time, the manual burr detection solution consumes a lot of manpower, has low efficiency, and increases the cost of burr detection.

[0054] In response to the defects in the related art, the present application provides a method for detecting electrode burrs. The method uses a neural network model to identify the molten area in the end face of the electrode to solve the problems of low recognition accuracy and low recognition efficiency caused by manual identification of the molten area in the related art, thereby improving the accuracy of molten area identification, and further improving the accuracy and efficiency of burr detection. After accurately identifying the molten area, multi-point sampling is performed on the molten area, and the height value of the molten area corresponding to each sampling point is detected to detect the burr defects in the molten area. This process does not rely on manual experience, thereby improving the accuracy and stability of burr marking and reducing the manpower consumption and detection cost of burr detection.

[0055] It should be noted that the electrode burr detection method provided in this application can be applied to the application environment shown in Figure 3. In Figure 3, user 301 operates on terminal device 302 through human-computer interaction. For example, user 301 can input the end face image of the battery electrode after laser die-cutting on the display interface of terminal device 302. Terminal device 302 uses the solution provided in the embodiment of this application to detect the acquired end face image to identify burr defects in the end face image. Among them, terminal device 302 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, etc. In the embodiment of this application, the battery can be a lithium-ion battery, and the battery electrode can be an electrode of a lithium-ion battery.

[0056] In addition, it should be noted that, in actual applications, the end face image after laser die-cutting can also be directly input into the terminal device 302 in FIG. 3 by the laser die-cutting device. This method can simplify the steps of manual operation and improve work efficiency.

[0057] In one embodiment, FIG4 shows a flow chart of a method for detecting a burr on a pole piece. The method may be applied to the terminal device 302 in FIG3 . As shown in FIG4 , the method may include the following steps S401 to S403 :

[0058] Step S401, obtaining an end face image of the battery electrode after the electrode is cut.

[0059] As an example, a laser die-cutting device performs laser die-cutting on a battery electrode to obtain an electrode end face. Simultaneously, the laser die-cutting device can capture an image of the laser end face, thereby obtaining an image of the electrode end face and storing the image in a terminal device. To inspect the quality of the electrode, the inspector can input the image of the electrode end face to be inspected into a burr defect detection application on the terminal device, which processes the image and identifies burr defects in the electrode end face.

[0060] In another example, a laser die-cutting device performs laser die-cutting on a battery pole piece to obtain a pole piece end face. At the same time, the laser die-cutting device can capture an image of the laser end face to obtain a pole piece end face image, and input the pole piece end face image into an application for detecting burr defects in a terminal device to process the pole piece end face image and identify burr defects in the pole piece end face.

[0061] Step S402 : Segment the molten area from the electrode end face image to obtain a molten area image.

[0062] In an embodiment of the present application, the terminal device can automatically identify the molten area in the electrode end face image without manual identification, which not only improves the identification efficiency of the molten area, but also avoids the problem of low recognition accuracy in manual identification.

[0063] As an example, the terminal device may segment the molten area image from the pole piece end face image by image processing, or may identify the molten area image from the pole piece end face image by a neural network model.

[0064] Step S403 : identifying burrs in the molten area according to the molten area height value corresponding to the molten area in the molten area image and a preset burr height threshold.

[0065] In step S403, the preset burr height threshold may be a height value determined by a tester based on actual experience, or an upper limit of a melting area height range determined by a terminal device performing statistics on the height values ​​of melting areas without burr defects.

[0066] As an example, the terminal device can sample the melting area and compare the melting area height value corresponding to each sampling point with the burr height threshold. When a sampling point is detected where the melting area height value is greater than the burr height threshold, it can be determined that there is a burr on the end face of the pole piece.

[0067] As another example, the terminal device can compare the melting area height values ​​of multiple sampling points and compare the maximum melting area height value with the burr height threshold. If the maximum melting area height value is greater than the burr height threshold, it can be determined that there are burrs on the end face of the pole piece.

[0068] Based on the scheme defined in steps S401 to S403 above, it can be seen that the embodiment of the present application, by separating the molten area from the electrode end face image, automatically identifies the molten area, avoiding the problem of low molten area identification accuracy existing in manual visual identification, not only improving the identification accuracy of the molten area, but also improving the identification efficiency of the molten area. Furthermore, after identifying the molten area, the burrs in the molten area can be identified by comparing the molten area height value of the molten area with a preset burr height threshold, thereby avoiding the problem of low burr labeling efficiency and accuracy caused by manual burr labeling, and improving the accuracy and detection efficiency of burr detection.

[0069] In some embodiments of the present application, optionally, obtaining an image of the end face of the battery pole piece after the pole piece is cut includes: obtaining an initial pole piece end face image that at least includes the pole piece end face; and cropping the initial pole piece end face image based on the position and / or proportion of the pole piece end face in the initial pole piece end face image to obtain a pole piece end face image.

[0070] In the above embodiment, the initial pole piece end face image may contain other content besides the battery pole piece, which will reduce the accuracy of the molten area recognition. Moreover, the battery pole piece usually accounts for a small proportion of the initial pole piece end face image. Therefore, in order to improve the recognition accuracy of the molten area, in the embodiment of the present application, before identifying the molten area in the pole piece end face image, the initial pole piece end face image is first cropped according to the position and / or proportion of the pole piece end face in the initial pole piece end face image to remove other interference factors in the pole piece end face image except the active material. At the same time, the proportion of the pole piece end face in the image is increased, which provides a basis for the subsequent identification of the molten area and the detection of burr defects.

[0071] In some embodiments of the present application, the melting area is segmented from the pole piece end face image to obtain the melting area image, including: using a region recognition model to identify the melting area from the pole piece end face image; segmenting the melting area from the pole piece end face image to obtain the melting area image.

[0072] In the above embodiment, the region recognition model is a neural network model for identifying the melt region. For example, the region recognition model may be a U-Net model. The melt region image is a binary image including the melt region and the background region.

[0073] In the embodiment of the present application, the region recognition model is used to realize automatic recognition of the molten area, avoiding the problem of low accuracy of molten area recognition in manual recognition, improving the recognition accuracy and efficiency of the molten area, and thereby improving the accuracy and efficiency of burr detection.

[0074] It should be noted that in actual applications, products with burr defects are usually rare, and collecting a large number of defect samples is difficult. In addition, too many parameters in the neural network model will lead to overfitting. However, due to the smaller scale of the U-Net model, there is no need to use a large number of defect samples. Moreover, the U-Net model has fewer parameters and does not cause overfitting. Therefore, in the embodiment of the present application, the U-Net model is used to identify the molten area in the electrode end face image, which reduces the cost of identifying the molten area.

[0075] In the above embodiment, the melt region image is a binary image, i.e., a binary image consisting of 0s and 1s, where the 1-valued area represents the image area that needs to be processed, and the 0-valued area represents the image area that is shielded and does not need to be processed. In the embodiment of the present application, 1 represents the melt region in the electrode end face image, and 0 represents the background region in the electrode end face image. For example, in the schematic diagram of the melt region image shown in Figure 5, the black area represents the background area, and the white area represents the melt region.

[0076] It should be noted that the electrode end face image is converted into a binary image (i.e., a molten region image), and then the burr defects in the molten region of the electrode end face are identified based on the molten region image. Since the molten region image is a binary image, that is, all information in the electrode end face image is converted into 0 and 1, it can avoid interference of other information in the electrode end face image on burr detection, thereby improving the accuracy of burr detection. Moreover, in the molten region image, the molten region and the background region have a high degree of distinction. Therefore, identifying burrs based on the molten region image can avoid the problem of inaccurate burr identification due to the high similarity between the characteristics of the molten region and the active material, thereby improving the accuracy of burr identification.

[0077] In some embodiments of the present application, the region recognition model includes at least a downsampling module, an upsampling module, a convolution module and a feature stitching module, wherein the melting area is segmented from the pole piece end face image to obtain the melting area image, including: performing multiple downsampling on the pole piece end face image through the downsampling module to obtain a downsampling feature map generated by each downsampling; performing convolution processing on the downsampling feature map generated by each downsampling through the convolution module to obtain a convolution feature map; performing upsampling and feature stitching processing on the convolution feature map respectively through the upsampling module and the feature stitching module to obtain the melting area features corresponding to the melting area in the pole piece end face image; identifying the melting area and the background area from the pole piece end face image based on the melting area features; performing binarization processing on the melting area and the background area in the pole piece end face image, and segmenting the melting area from the binarized pole piece end face image to obtain the melting area image.

[0078] As an example, a schematic diagram of the structure of the U-Net model is shown in Figure 6. In Figure 6, the input image of the U-Net model is the electrode end face image, and the output image is the molten area image.

[0079] As can be seen from Figure 6, in the embodiment of the present application, the U-Net model adopts a codec structure that combines upsampling and downsampling, wherein downsampling can increase the robustness to disturbances of the input image, such as image translation, rotation, etc.; in the upsampling process, the features are spliced, that is, in the embodiment of the present application, a multi-scale fusion method is used to upsample the downsampled feature map obtained by downsampling. This method improves the problem of insufficient upsampling information, improves the accuracy of small target recognition, and improves the effect of target edge segmentation.

[0080] It should be noted that compared with the manual identification of molten areas in related technologies, the use of the U-Net model shown in Figure 6 to identify molten areas in the electrode end face image can improve the recognition speed and accuracy of the molten areas. In the embodiment of the present application, the U-Net model for identifying molten areas can increase the manual recognition rate from 90s / sample to 0.2s / sample, an efficiency increase of 450 times. The measurement repeatability and reproducibility increased from 61.33%, which did not meet the measurement system evaluation index, to less than 1%, meeting the measurement system MSA (Measurement System Analysis) evaluation index and providing an accurate input source for the subsequent calculation of the burr location and the average height of the molten area.

[0081] In some embodiments of the present application, optionally, the convolution feature map is upsampled and feature stitched respectively by the upsampling module and the feature stitching module to obtain the melting area feature corresponding to the melting area in the electrode end face image, including: performing the Mth upsampling process on the convolution feature map corresponding to the Nth downsampling by the upsampling module to obtain the upsampling feature map corresponding to the Mth upsampling; performing the Mth feature stitching on the upsampling feature map corresponding to the Mth upsampling and the convolution feature map corresponding to the N-1th downsampling by the feature stitching module to obtain the initial defect segmentation map; performing the Mth feature stitching on the initial defect segmentation The graph iteratively performs the following operations until M=P to obtain a target defect segmentation map, and determines that the feature corresponding to the target defect segmentation map is a melting area feature: the initial defect segmentation map obtained by the M-th feature splicing is upsampled by the upsampling module to obtain an upsampled feature map corresponding to the M+1-th upsampling; the upsampled feature map corresponding to the M+1-th upsampling and the feature map generated by the N-2-th downsampling are feature spliced ​​by the feature splicing module to obtain the target defect segmentation map; M is added by one, N is subtracted by one, and the initial defect segmentation map is updated to the target defect segmentation map.

[0082] In the above embodiment, M+N=P, N is a positive integer greater than or equal to 1, M is a positive integer greater than or equal to 1, P is a positive integer greater than 1, and P-1 is the number of downsampling times of the region recognition model. For example, in the U-Net model shown in Figure 6, the number of downsampling times is 4, then P=5.

[0083] The above embodiment is explained below with reference to FIG6 .

[0084] In Figure 6, the pole piece end face image is input into the U-Net model. The convolution module performs a 3×3 convolution on the pole piece end face image. The convolved image is then downsampled 2×2 by the downsampling module to obtain a downsampled feature map, completing a downsampling process. This downsampled feature map serves as the input image for the next downsampling. By repeating the above operation, a downsampled feature map generated by multiple downsampling operations can be obtained. After the last downsampling, the downsampled feature map generated by the last downsampling is input into the convolution module and a 3×3 convolution operation is performed. The convolved downsampled feature map (i.e., the convolution feature map) serves as the input image for the upsampling module. The upsampling module performs a 2×2 upsampling process on the convolution feature map to obtain an upsampled feature map. This upsampled feature map is feature-joined with the convolution feature map corresponding to the penultimate downsampling to obtain the initial defect segmentation map. The above operation is then repeated on the initial defect segmentation map until all upsampling is complete. The resulting defect segmentation map is then subjected to a 3×3 convolution followed by a 1×1 convolution to obtain the target defect segmentation map. Feature extraction is then performed on the target defect segmentation map to obtain the melt region features. Finally, the electrode end face image is segmented based on the melt region features to obtain the background region and the melt region. The electrode end face image is then binarized to obtain the melt region image.

[0085] As can be seen from the above, the electrode end face image is used as the input image of the U-Net model. The U-Net model performs four downsampling operations on the electrode end face image, reducing the image resolution and obtaining four intermediate feature maps: layer 1, layer 2, layer 3, and layer 4. The intermediate feature map layer 4 is then convolved and upsampled, and then concatenated with the convolved intermediate feature map layer 3. This process is repeated until the last upsampling is completed. The feature map obtained from the last upsampling is then subjected to a 1x1 convolution to obtain a low-resolution defect segmentation image. This process continues in this way, ultimately resulting in a melt region image with the same resolution as the original electrode end face image.

[0086] It should be noted that by adopting the multi-scale fusion method to upsample the downsampled feature map, the problem of insufficient upsampling information is improved, the accuracy of small target recognition is improved, the effect of target edge segmentation is improved, and the accuracy of molten area recognition is improved, thereby providing a basis for the accurate identification of burr defects.

[0087] In some embodiments of the present application, optionally, burrs in the melted area are identified based on a melted area height value corresponding to the melted area in the melted area image and a preset burr height threshold, including: sampling the melted area in the melted area image to obtain multiple sampling points; identifying burrs in the melted area based on a melted area height value corresponding to each sampling point and a preset burr height threshold.

[0088] As an example, the terminal device may sample the melted area in the melted area image shown in FIG5 along the horizontal direction, thereby obtaining multiple sampling points in the horizontal area. The height value of the white area corresponding to each sampling point is the height value of the melted area corresponding to each sampling point.

[0089] By comparing the melting area height value of the melting area with the preset burr height threshold, the burrs in the melting area can be identified, thereby avoiding the problems of low burr marking efficiency and accuracy caused by manual burr marking, and improving the accuracy and efficiency of burr detection.

[0090] In some embodiments of the present application, optionally, sampling the melted area in the melted area image to obtain multiple sampling points includes: performing connected domain processing on the melted area image to obtain a target melted area image; sampling the melted area in the target melted area image along a first direction to obtain multiple sampling points.

[0091] In the above embodiment, the first direction is used to represent the extension direction of the molten region on the end face of the electrode. That is, the first direction is the lateral direction of the molten region. For example, in Figure 5, the first direction is the horizontal direction. In other words, in the embodiment of the present application, the molten region is sampled in the horizontal direction to obtain multiple sampling points. The sampling frequency of the multiple sampling points can be set by the tester according to actual needs.

[0092] In addition, in the above embodiment, a connected domain is a connected area in a binary image consisting of adjacent pixels with the same pixel value. Adjacent pixels can be adjacent pixels in the vertical, horizontal, or diagonal directions in the binary image. In the embodiment of the present application, a connected domain algorithm in related art can be used to perform maximum connected domain processing on the melted region image.

[0093] It should be noted that, in the embodiments of the present application, connected domain processing is performed on the melt region image to ensure the consistency of each image region in the melt region image, reducing the risk of disconnection in the melt region during burr defect identification, thereby improving the detection accuracy of burr defects. In addition, in the embodiments of the present application, multi-point sampling is performed in the lateral direction of the melt region to reduce the randomness of burr defect detection and reduce the risk of missed burr detection while ensuring a reasonable burr detection time.

[0094] In some embodiments of the present application, optionally, burrs in the melting area are identified based on the melting area height value corresponding to each sampling point and a preset burr height threshold, including: obtaining the melting area height value corresponding to each sampling point in the second direction to obtain the melting area height value of each sampling point; comparing the melting area height values ​​of multiple sampling points to obtain a maximum height value; when the maximum height value is greater than the preset burr height threshold, determining that burrs exist in the melting area.

[0095] In the above embodiment, the second direction is perpendicular to the first direction. For example, in FIG5 , the second direction is perpendicular to the melt region. The melt region height value at each sampling point is the height value of each sampling point in the second direction. For example, in the sampled melt region image shown in FIG7 , each white line within the melt region represents a sampling point, and accordingly, the length of each white line is the melt region height value at each sampling point.

[0096] It should be noted that in the embodiments of the present application, the maximum height value of the molten region is compared with a preset burr height threshold to determine whether a burr is present in the molten region, thereby improving the accuracy of burr defect detection. Furthermore, by comparing only the maximum height value with the preset burr height threshold, rather than comparing the molten region height values ​​at all sampling points with the preset burr height threshold, the number of comparisons can be reduced, improving comparison efficiency and, consequently, improving the efficiency of burr detection.

[0097] In some embodiments of the present application, optionally, after determining that there are burrs in the melting area, the terminal device compares the melting area height value corresponding to each sampling point with a preset burr height threshold to obtain a comparison result; determines a target sampling point from multiple sampling points based on the comparison result; and determines the position of the target sampling point in the melting area as the position of the burr in the melting area.

[0098] In the above embodiment, the target sampling point is the sampling point where the height of the melt region is greater than the preset burr height threshold, that is, the location of the target sampling point is the location of the burr. In other words, in the embodiment of the present application, if a burr defect is determined to exist in the melt region, the specific location of the burr in the melt region is further detected.

[0099] It should be noted that after detecting the presence of a burr defect in the molten area, further detecting the specific location of the burr in the molten area can improve the efficiency and accuracy of burr detection. In addition, from the above content, it can be seen that the solution provided by the embodiment of the present application can not only identify whether there is a burr defect in the molten area, but also determine the specific defect location when a burr defect is identified, so as to facilitate the subsequent analysis of the cause of the burr defect, and then improve the laser die-cutting equipment or the manufacturing process of the pole piece to improve the production quality of the pole piece, and then improve the production quality of the battery.

[0100] In some embodiments of the present application, optionally, after comparing the melting area height values ​​of multiple sampling points to obtain the maximum height value, if there is no target sampling point among the multiple sampling points, the terminal device calculates the average of the melting area height values ​​of the multiple sampling points to obtain the average height value of the melting area; and then determines whether the quality of the battery electrode meets the target requirements based on the average height value of the melting area and a preset melting area height range.

[0101] As an example, in an embodiment of the present application, the height value of the melting area of ​​each sampling point can be counted, and the height value of the melting area of ​​each sampling point can be compared with the preset burr height threshold. When it is detected that the height values ​​of the melting areas of all sampling points are less than the preset burr height threshold, it can be determined that there are no burr defects in the melting area. At this time, the terminal device can use the box plot statistical method to calculate the average height value of the melting area. If the average height value is within the preset melting area height range, it is determined that the battery electrode meets the requirements of electrode production (i.e., the target demand); if the average height value is outside the preset melting area height range, it can be determined that the battery electrode does not meet the requirements of electrode production. In actual application, the battery electrode can be discarded.

[0102] It should be noted that the preset melting area height range is the height range of the melting area in qualified battery electrode sheets. This preset melting area height range can be set by the tester based on actual experience, or it can be determined by the terminal device through statistical analysis of historical melting area height values ​​of battery electrode sheets.

[0103] In addition, it should be noted that when there are no burr defects in the molten area, further product compliance inspection is carried out on the molten area to improve the quality of the battery pole pieces and thus improve the production quality of the battery.

[0104] In some embodiments of the present application, optionally, after obtaining the height value of the melting area corresponding to each sampling point in the second direction and obtaining the height value of the melting area of ​​each sampling point, the terminal device further performs statistical analysis on the height values ​​of the melting areas corresponding to multiple sampling points to determine the abnormal height range corresponding to the abnormal sampling point; then, based on the abnormal height range, the abnormal sampling point is identified from the multiple sampling points, and the abnormal sampling point is filtered out from the multiple sampling points to obtain multiple filtered sampling points.

[0105] In the above embodiment, the melting region height value of the abnormal sampling point is within the abnormal height range. The abnormal sampling point is a sampling point where the melting region height value is much smaller than the average melting region height value. For example, the abnormal height range is 0-h, where h can be determined by multiplying the proportional coefficient by the average melting region height value. The proportional coefficient can be a positive number greater than 0 and less than 1. The proportional coefficient can be set by the tester based on actual experience.

[0106] As an example, in the sampled melt region image shown in FIG7 , the melt region height values ​​of some sampling points are much smaller than the average melt region height value. These sampling points are abnormal sampling points (such as the sampling points within the area marked by the black box in FIG7 ). To avoid the impact of abnormal sampling points on burr defect identification, in an embodiment of the present application, when abnormal sampling points are detected in the melt region image, the abnormal sampling points in the melt region image are filtered out, thereby obtaining the melt region image after filtering out the abnormal sampling points shown in FIG8 . The terminal device then detects the melt region image after filtering out the abnormal sampling points to identify burrs in the melt region.

[0107] It should be noted that by identifying and filtering out abnormal sampling points in the molten area image, the influence of abnormal sampling points on burr detection is avoided, thereby improving the accuracy of burr detection.

[0108] Based on the above content, it can be seen that the solution provided in the embodiment of the present application uses a neural network learning model to segment the melting area image from the electrode end face image, and detects the melting area height value of the melting area in the melting area image to determine whether there are burrs in the melting area. In addition, the embodiment of the present application also detects whether the battery electrode is compliant based on the melting area height value of the melting area in the melting area image, so as to improve the quality of battery products, provide the safety of battery products, and reduce the production cost of batteries.

[0109] The solution provided in the embodiment of the present application identifies the melting area based on U-Net. This method reduces the manpower consumption of burr detection, makes burr detection stably reproducible, reduces the risk of battery cell puncture, improves the safety of battery products, and reduces production costs.

[0110] In addition, the solution provided in the embodiment of the present application can not only detect whether there are burr defects in the melting area by performing statistical analysis on the melting area height values ​​of the sampling points in the melting area image, but also determine the position of the burr when the burr defects are detected in the melting area, thereby improving the efficiency and accuracy of burr detection and enhancing the stability and reproducibility of burr detection.

[0111] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0112] In one embodiment, as shown in FIG9 , a device for detecting electrode burrs is provided. The device 900 includes an image acquisition module 901 , a melting identification module 902 , and a burr identification module 903 .

[0113] The image acquisition module 901 is configured to acquire an image of the end face of the battery electrode after the electrode is cut;

[0114] The melt identification module 902 is configured to segment the melt area from the electrode end face image to obtain a melt area image;

[0115] The burr identification module 903 is configured to identify burrs in the melt region according to a melt region height value corresponding to the melt region in the melt region image and a preset burr height threshold.

[0116] In one example, the image acquisition module 901 is specifically configured to acquire an initial pole piece end face image including at least the pole piece end face; based on the position and / or proportion of the pole piece end face in the initial pole piece end face image, the initial pole piece end face image is cropped to obtain the pole piece end face image.

[0117] In one example, the melt identification module 902 is specifically configured to use a region recognition model to identify the melt area from the pole piece end face image; segment the melt area from the pole piece end face image to obtain a melt area image, wherein the melt area image is a binary image, and the binary image includes the melt area and the background area.

[0118] In one example, the burr identification module 903 includes a sampling module and a first identification module. The sampling module is configured to sample the melted area in the melted area image to obtain a plurality of sampling points; the first identification module is configured to identify burrs in the melted area based on the melted area height value corresponding to each sampling point and a preset burr height threshold.

[0119] In one example, the sampling module is specifically configured to perform connected domain processing on the melting area image to obtain a target melting area image; sample the melting area in the target melting area image along a first direction to obtain multiple sampling points, wherein the first direction is used to characterize the extension direction of the melting area on the end face of the pole piece.

[0120] In one example, the first identification module is specifically configured to obtain the melting area height value corresponding to each sampling point in the second direction, and obtain the melting area height value of each sampling point, wherein the second direction is perpendicular to the first direction; compare the melting area height values ​​of multiple sampling points to obtain the maximum height value; when the maximum height value is greater than a preset burr height threshold, it is determined that there are burrs in the melting area.

[0121] In one example, the device for detecting burrs on a pole piece further includes: a burr position detection module, configured to, after determining that a burr exists in the melting area, compare the melting area height value corresponding to each sampling point with a preset burr height threshold to obtain a comparison result; determine a target sampling point from a plurality of sampling points based on the comparison result, wherein the target sampling point is a sampling point whose melting area height value is greater than the preset burr height threshold; and determine the position of the target sampling point in the melting area as the position of the burr in the melting area.

[0122] In one example, the electrode burr detection device also includes: an abnormal data filtering module, which is configured to obtain the height value of the melting area corresponding to each sampling point in the second direction, obtain the melting area height value of each sampling point, and then perform statistical analysis on the melting area height values ​​corresponding to multiple sampling points to determine the abnormal height range corresponding to the abnormal sampling point; identify abnormal sampling points from multiple sampling points based on the abnormal height range, wherein the melting area height value of the abnormal sampling point is within the abnormal height range; and filter out abnormal sampling points from the multiple sampling points to obtain multiple filtered sampling points.

[0123] In one example, the device for detecting burrs on a pole piece further includes: a compliance detection module, which is configured to compare the melting area height values ​​of a plurality of sampling points to obtain a maximum height value, and then, when there is no target sampling point among the plurality of sampling points, calculate the average value of the melting area height values ​​of the plurality of sampling points to obtain an average height value of the melting area; and determine whether the quality of the battery pole piece meets the target requirements based on the average height value of the melting area and a preset melting area height range.

[0124] In one example, the region recognition model includes at least a U-Net model.

[0125] In one embodiment, the present application further provides an electronic device, which includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the above-mentioned method for detecting pole piece burrs is implemented.

[0126] FIG10 shows a schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application.

[0127] The electronic device may include a processor 1001 and a memory 1002 storing computer program instructions.

[0128] Specifically, the processor 1001 may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0129] The memory 1002 may include a large capacity memory configured as data or instructions. By way of example and not limitation, the memory 1002 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 1002 may include removable or non-removable (or fixed) media. Where appropriate, the memory 1002 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 1002 is a non-volatile solid-state memory.

[0130] The memory may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical or other physical / tangible memory storage devices. Thus, generally, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.

[0131] The processor 1001 reads and executes computer program instructions stored in the memory 1002 to implement any one of the pole piece burr detection methods in the above embodiments.

[0132] In one example, the electronic device may further include a communication interface 1003 and a bus 1010. As shown in FIG10 , the processor 1001, the memory 1002, and the communication interface 1003 are connected via the bus 1010 and communicate with each other.

[0133] The communication interface 1003 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0134] Bus 1010 comprises hardware, software or both, couples the parts of electronic equipment to each other.For example, and not limitation, bus can comprise accelerated graphics port (AGP) or other graphics bus, enhanced industry standard architecture (EISA) bus, front side bus (FSB), hypertransport (HT) interconnection, industry standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations.In suitable cases, bus 1010 can comprise one or more buses.Although the present application embodiment describes and shows specific bus, the application considers any suitable bus or interconnection.

[0135] In one embodiment, the present application further provides a readable storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, the above-mentioned method for detecting pole piece burrs is implemented.

[0136] It should be understood that the present application is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and illustrated as examples. However, the method process of the present application is not limited to the specific steps described and illustrated. Those skilled in the art can make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present application.

[0137] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier wave. "Machine-readable medium" can include any medium that can store or transmit information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROMs, flash memories, erasable ROMs (EROMs), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0138] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps. In other words, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0139] The above reference is according to the method, apparatus, equipment and the flowchart and / or block diagram of the computer program product of the embodiment of the present application.It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be realized by computer program instructions.These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a kind of machine, so that these instructions executed by the processor of the computer or other programmable data processing device enable the realization of the function / action specified in one or more boxes of the flowchart and / or block diagram.Such a processor can be but is not limited to a general-purpose processor, a special-purpose processor, a special application processor or a field programmable logic circuit.It is also understood that each box in the block diagram and / or the flowchart and the combination of the boxes in the block diagram and / or the flowchart can also be realized by the dedicated hardware that performs the specified function or action, or can be realized by the combination of dedicated hardware and computer instructions.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and they should all be included in the scope of the claims and specification of the present application. In particular, as long as there is no structural conflict, the various technical features mentioned in the various embodiments can be combined in any way. The present application is not limited to the specific embodiments disclosed herein, but includes all technical solutions that fall within the scope of the claims.

Claims

1. A method for detecting a burr on a pole piece, comprising: Acquire the end face image of the battery electrode after the electrode is cut; Segmenting the molten area from the pole piece end face image to obtain a molten area image; The burrs in the molten area are identified according to the molten area height value corresponding to the molten area in the molten area image and a preset burr height threshold.

2. The method according to claim 1, wherein: Obtain the end face image of the battery electrode after cutting, including: Acquire an initial pole piece end surface image including at least the pole piece end surface; Based on the position and / or proportion of the pole piece end face in the initial pole piece end face image, the initial pole piece end face image is cropped to obtain the pole piece end face image.

3. The method according to claim 1, wherein: Segmenting the molten area from the pole piece end face image to obtain the molten area image includes: Using a region recognition model to identify the molten region from the pole piece end face image; The melting area is segmented from the pole piece end face image to obtain the melting area image, wherein the melting area image is a binary image including the melting area and a background area.

4. The method according to claim 3, wherein: Identifying burrs in the molten area according to a molten area height value corresponding to the molten area in the molten area image and a preset burr height threshold comprises: Sampling the molten area in the molten area image to obtain a plurality of sampling points; The burrs in the molten area are identified according to the molten area height value corresponding to each sampling point and the preset burr height threshold.

5. The method according to claim 4, wherein: Sampling the molten area in the molten area image to obtain a plurality of sampling points, including: Performing connected domain processing on the molten area image to obtain a target molten area image; The molten area in the target molten area image is sampled along a first direction to obtain the plurality of sampling points, wherein the first direction is used to characterize the molten area on the end face of the pole piece. Extension direction.

6. The method according to claim 5, wherein: Identifying burrs in the molten area according to the molten area height value corresponding to each sampling point and a preset burr height threshold includes: Acquire a melting area height value corresponding to each sampling point in a second direction to obtain a melting area height value of each sampling point, wherein the second direction is perpendicular to the first direction; Comparing the melting area height values ​​of the plurality of sampling points to obtain a maximum height value; When the maximum height value is greater than the preset burr height threshold, it is determined that there are burrs in the melting area.

7. The method according to claim 6, after determining that there are burrs in the molten area, the method further comprises: Comparing the melting area height value corresponding to each sampling point with the preset burr height threshold to obtain a comparison result; Determine a target sampling point from the plurality of sampling points according to the comparison result, wherein the target sampling point is a sampling point whose melting area height value is greater than the preset burr height threshold; The position of the target sampling point in the melting region is determined as the position of the burr in the melting region.

8. The method according to claim 7, after comparing the melting area height values ​​of the plurality of sampling points to obtain the maximum height value, the method further comprises: When the target sampling point does not exist among the multiple sampling points, calculating an average value of the melting area height values ​​of the multiple sampling points to obtain an average height value of the melting area; Whether the quality of the battery electrode sheet meets the target requirement is determined based on the average height value of the melting area and the preset melting area height range.

9. The method according to claim 6, after obtaining the height value of the melting area corresponding to each sampling point in the second direction and obtaining the height value of the melting area of ​​each sampling point, the method further comprises: Performing statistical analysis on the melting area height values ​​corresponding to the plurality of sampling points to determine the abnormal height range corresponding to the abnormal sampling points; Identifying the abnormal sampling point from the plurality of sampling points based on the abnormal height range, wherein the height value of the melting area of ​​the abnormal sampling point is within the abnormal height range; The abnormal sampling points are filtered out from the multiple sampling points to obtain multiple filtered sampling points.

10. The method according to claim 3, wherein: The region recognition model is a neural network model for identifying the molten region, and the region recognition model at least includes a downsampling module, an upsampling module, a convolution module, and a feature splicing module.

11. The method according to claim 10, wherein: The step of segmenting the molten area from the pole piece end face image to obtain the molten area image comprises: Downsampling the pole piece end face image multiple times by the downsampling module to obtain a downsampling feature map generated by each downsampling; The convolution module performs convolution processing on the down-sampled feature map generated by each down-sampling to obtain a convolution feature map; The convolution feature map is upsampled and feature stitched by the upsampling module and the feature stitching module respectively, so as to obtain the melting area feature corresponding to the melting area in the pole piece end face image; Identifying a molten area and a background area from the pole piece end face image based on the molten area features; The molten area and the background area in the pole piece end face image are binarized, and the molten area is segmented from the pole piece end face image after the binarization process to obtain the molten area image.

12. The method according to claim 11, wherein: The convolution feature map is upsampled and feature stitched by the upsampling module and the feature stitching module respectively, so as to obtain the melting area feature corresponding to the melting area in the pole piece end face image, including: Performing an M-th upsampling process on the convolution feature map corresponding to the N-th downsampling through the upsampling module to obtain an upsampling feature map corresponding to the M-th upsampling, wherein M+N=P, N is a positive integer greater than or equal to 1, M is a positive integer greater than or equal to 1, P is a positive integer greater than 1, and P-1 is the number of downsampling times of the region recognition model; The feature splicing module performs the Mth feature splicing on the up-sampled feature map corresponding to the Mth up-sampling and the convolution feature map corresponding to the N-1th down-sampling to obtain the initial defect Segmentation map; An iterative operation is performed on the initial defect segmentation map until M=P, a target defect segmentation map is obtained, and a feature corresponding to the target defect segmentation map is determined to be the molten area feature.

13. The method according to claim 12, wherein: The convolution feature map is upsampled and feature stitched by the upsampling module and the feature stitching module respectively, so as to obtain the melting area feature corresponding to the melting area in the pole piece end face image, including: The upsampling module performs upsampling processing on the initial defect segmentation map obtained by the M-th feature splicing to obtain an upsampled feature map corresponding to the M+1-th upsampling; The feature splicing module performs feature splicing on the up-sampled feature map corresponding to the M+1th up-sampling and the feature map generated by the N-2th down-sampling to obtain the target defect segmentation map; An addition operation is performed on M, and an subtraction operation is performed on N, and the initial defect segmentation map is updated to the target defect segmentation map.

14. The method according to any one of claims 10 to 13, wherein: The region recognition model at least includes a U-Net model.

15. The method according to claim 5, wherein: The first direction is a lateral direction of the melting region, wherein the melting region in the target melting region image is sampled along the first direction to obtain the multiple sampling points, including: The melting area is sampled along the lateral direction of the melting area according to a preset sampling frequency to obtain the plurality of sampling points.

16. The method according to claim 8, wherein: Calculating the average value of the melting area height values ​​of the plurality of sampling points to obtain the average height value of the melting area includes: The height values ​​of the melting areas at the plurality of sampling points are statistically analyzed by box plots to obtain an average height value of the melting areas.

17. The method according to claim 8 or 16, wherein: Determining whether the quality of the battery electrode sheet meets the target requirement based on the average height value of the molten area and the preset molten area height range includes: If the average height value is within the preset melting area height range, it is determined that the quality of the battery electrode sheet meets the target requirement; If the average height value is outside the preset melting area height range, it can be determined that the quality of the battery electrode sheet does not meet the target requirement.

18. A device for detecting burrs on a pole piece, comprising: An image acquisition module is configured to acquire an end face image of the battery electrode after the electrode is cut; A melting identification module is configured to segment the melting area from the pole piece end face image to obtain a melting area image; The burr recognition module is configured to recognize the burrs in the molten area according to the molten area height value corresponding to the molten area in the molten area image and a preset burr height threshold.

19. An electronic device comprising: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, the method for detecting pole piece burrs as described in any one of claims 1-9 is implemented.

20. A readable storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the method for detecting pole piece burrs according to any one of claims 1 to 9.

Images