Pole piece defect detection method, device and storage medium

Abstract

Embodiments of the present application provide a kind of pole piece defect detection method, device and storage medium.The method comprises: obtaining the pole piece image information of battery to be detected, and multiple pole pieces are contained in battery to be detected;Key point model is input to pole piece image information, and the key point data in pole piece image information is obtained, and key point data is used to reflect the endpoint position of each pole piece in battery to be detected;Pole piece image information is input to segmentation model, and the segmentation region in pole piece image information is obtained, and segmentation region is used to reflect the region where the endpoint of each pole piece in battery to be detected is;According to key point data and / or segmentation region, the endpoint coordinates of each pole piece in battery to be detected are determined;According to endpoint coordinates, the defect of multiple pole pieces of battery to be detected is detected.The present application does not need artificial participation, and the whole detection process is less affected by artificial subjective factors, and detection result is more accurate.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of batteries, in particular to a method and device for detecting defects of electrode plates and a storage medium. BACKGROUND

[0002] Lithium batteries are batteries using lithium metal or lithium alloy as negative electrode material and non-electrolyte solution. In order to prevent the appearance of lithium branch crystals, the negative electrode of lithium batteries is required to have a certain redundancy length than the positive electrode, and the electrodes should be arranged in order and cannot have a large degree of bending and deformation. However, during the manufacturing process of lithium batteries, winding or stacking may cause a certain fluctuation in the relative position of the positive and negative electrodes of lithium batteries, thereby changing the boundary distance of the positive and negative electrodes and causing problems such as excessive redundancy of the negative electrode or no redundancy.

[0003] In the related art, the redundancy and deformation of the positive and negative electrode plates of lithium batteries are generally judged by manual experience and visual observation, which is greatly affected by subjective factors and cannot guarantee the accuracy of the defect detection of the electrode plates of lithium batteries. SUMMARY

[0004] The present application provides a method and device for detecting defects of electrode plates and a storage medium, which are less affected by subjective factors and more accurate in detection results.

[0005] The present application provides a method for detecting defects of electrode plates, which comprises the following steps:

[0006] Obtaining image information of electrode plates of a battery to be detected, wherein the battery to be detected comprises a plurality of electrode plates;

[0007] Inputting the image information of the electrode plates into a key point model to obtain key point data in the image information of the electrode plates, wherein the key point data is used to reflect the positions of the end points of the electrode plates in the battery to be detected;

[0008] Inputting the image information of the electrode plates into a segmentation model to obtain segmentation regions in the image information of the electrode plates, wherein the segmentation regions are used to reflect the regions where the end points of the electrode plates in the battery to be detected are located;

[0009] Determining the coordinates of the end points of the electrode plates in the battery to be detected according to the key point data and / or the segmentation regions;

[0010] Detecting defects of the plurality of electrode plates of the battery to be detected according to the coordinates of the end points.

[0011] In some embodiments of the present application, the step of determining the coordinates of the end points of the electrode plates in the battery to be detected according to the key point data and / or the segmentation regions comprises:

[0012] obtaining a number of end points of the plurality of pole pieces, a number of key points in the key point data, and a number of regions in the segmented regions;

[0013] determining the end point coordinates of the pole pieces based on the number of end points of the plurality of pole pieces, the number of key points in the key point data, and / or the number of regions in the segmented regions.

[0014] In some embodiments of the present application, the determining the end point coordinates of the pole pieces based on the number of end points of the plurality of pole pieces, the number of key points in the key point data, and / or the number of regions in the segmented regions comprises:

[0015] if the number of end points of the plurality of pole pieces is equal to the number of key points in the key point data and is not equal to the number of regions in the segmented regions, determining the end point coordinates of the pole pieces according to the key point data;

[0016] if the number of end points of the plurality of pole pieces is equal to the number of regions in the segmented regions and is not equal to the number of key points in the key point data, determining the end point coordinates of the pole pieces according to the segmented regions;

[0017] if the number of end points of the plurality of pole pieces is not equal to the number of key points in the key point data and is not equal to the number of regions in the segmented regions, determining the end point coordinates of the pole pieces according to the key point data and the segmented regions.

[0018] In some embodiments of the present application, the determining the end point coordinates of the pole pieces according to the key point data and the segmented regions comprises:

[0019] determining first end point coordinates of the pole pieces according to the key point data;

[0020] determining second end point coordinates of the pole pieces according to the segmented regions;

[0021] determining the end point coordinates of the pole pieces based on the first end point coordinates and the second end point coordinates.

[0022] In some embodiments of the present application, the determining the end point coordinates of the pole pieces based on the first end point coordinates and the second end point coordinates comprises:

[0023] determining third end point coordinates of the pole pieces based on a preset skeleton algorithm;

[0024] determining the end point coordinates of the pole pieces based on the first end point coordinates, the second end point coordinates, and the third end point coordinates.

[0025] In some embodiments of the present application, before the step of determining the endpoint coordinates of each tab in the battery to be detected according to the key point data and / or the segmentation region, the method further comprises:

[0026] determining whether there is an extra endpoint on each tab;

[0027] if there is, removing the extra endpoint.

[0028] In some embodiments of the present application, before the step of inputting the tab image information into the key point model, the method further comprises:

[0029] inputting the tab image information into a deformation detection model to obtain a deformation result of each tab in the battery to be detected;

[0030] determining that the deformation result is that each tab is not deformed.

[0031] In some embodiments of the present application, the battery to be detected contains a plurality of positive and negative tabs arranged in a staggered manner; and the step of performing defect detection on the plurality of tabs of the battery to be detected according to the endpoint coordinates comprises:

[0032] determining the height of each of the plurality of positive and negative tabs and the distance between each two adjacent positive or negative tabs according to the endpoint coordinates;

[0033] performing defect detection on the plurality of tabs of the battery to be detected based on the height of each of the plurality of positive and negative tabs and the distance between each two adjacent positive or negative tabs.

[0034] The embodiments of the present application also provide a tab defect detection device, which comprises:

[0035] an information acquisition module configured to acquire tab image information of a battery to be detected, wherein the battery to be detected contains a plurality of tabs;

[0036] a key point acquisition module configured to input the tab image information into a key point model to obtain key point data in the tab image information, wherein the key point data is used to reflect the position of an endpoint of each tab in the battery to be detected;

[0037] a region acquisition module configured to input the tab image information into a segmentation model to obtain a segmentation region in the tab image information, wherein the segmentation region is used to reflect the region where the endpoint of each tab in the battery to be detected is located;

[0038] a determination module configured to determine endpoint coordinates of each tab in the battery to be detected according to the key point data and / or the segmentation region;

[0039] The detection module is configured to detect defects of the tab of the battery to be detected according to the endpoint coordinates.

[0040] In some embodiments of the present application, the determination module is specifically configured to:

[0041] The determination module is configured to determine the endpoint coordinates of the tabs according to the number of endpoints of the plurality of tabs, the number of key points in the key point data, and / or the number of regions in the segmented regions.

[0042] The determination module is configured to determine the endpoint coordinates of the tabs based on the number of endpoints of the plurality of tabs, the number of key points in the key point data, and / or the number of regions in the segmented regions.

[0043] In some embodiments of the present application, the determination module is specifically configured to:

[0044] If the number of endpoints of the plurality of tabs is equal to the number of key points in the key point data and is not equal to the number of regions in the segmented regions, the determination module is configured to determine the endpoint coordinates of the tabs according to the key point data.

[0045] If the number of endpoints of the plurality of tabs is equal to the number of regions in the segmented regions and is not equal to the number of key points in the key point data, the determination module is configured to determine the endpoint coordinates of the tabs according to the segmented regions.

[0046] If the number of endpoints of the plurality of tabs is not equal to the number of key points in the key point data and is not equal to the number of regions in the segmented regions, the determination module is configured to determine the endpoint coordinates of the tabs according to the key point data and the segmented regions.

[0047] In some embodiments of the present application, the determination module is specifically configured to:

[0048] The determination module is configured to determine first endpoint coordinates of the tabs according to the key point data.

[0049] The determination module is configured to determine second endpoint coordinates of the tabs according to the segmented regions.

[0050] The determination module is configured to determine the endpoint coordinates of the tabs based on the first endpoint coordinates and the second endpoint coordinates.

[0051] In some embodiments of the present application, the determination module is specifically configured to:

[0052] The determination module is configured to determine third endpoint coordinates of the tabs based on a preset skeleton algorithm.

[0053] The determination module is configured to determine the endpoint coordinates of the tabs based on the first endpoint coordinates, the second endpoint coordinates, and the third endpoint coordinates.

[0054] In some embodiments of the present application, the device further comprises:

[0055] a determination module configured to determine whether there is an excess endpoint on each pole piece;

[0056] a removal module configured to remove the excess endpoint when the excess endpoint exists.

[0057] In some embodiments of the present application, the device further comprises:

[0058] a deformation detection module configured to input the pole piece image information into a deformation detection model to obtain a deformation result of each pole piece in the battery to be detected;

[0059] a result determination module configured to determine that the deformation result is that the pole pieces are not deformed.

[0060] In some embodiments of the present application, the battery to be detected comprises a plurality of positive pole pieces and negative pole pieces arranged in a staggered manner; and the detection module is specifically configured to:

[0061] determine, according to the endpoint coordinates, a height of each of the positive pole pieces and the negative pole pieces and a distance between every two adjacent positive pole pieces or negative pole pieces;

[0062] perform defect detection on the plurality of pole pieces of the battery to be detected based on the height of each of the positive pole pieces and the negative pole pieces and the distance between every two adjacent positive pole pieces or negative pole pieces.

[0063] Embodiments of the present application also provide a non-transitory machine-readable storage medium having executable code stored thereon, which, when executed by a processor of an electronic device, causes the processor to perform the pole piece defect detection method described above.

[0064] In embodiments of the present application, by obtaining pole piece image information of a battery to be detected and inputting the pole piece image information into a key point model and a segmentation model, key point data and segmentation regions in the pole piece image information are obtained respectively. Based on the key point data and / or the segmentation regions, endpoint coordinates of each pole piece in the battery to be detected can be determined, and then defect detection on a plurality of pole pieces of the battery to be detected can be completed according to the endpoint coordinates. In summary, the present application can automatically obtain key point data reflecting endpoint positions of each pole piece and segmentation regions reflecting endpoint regions of each pole piece according to the key point model and the segmentation model, without human intervention. The endpoint coordinates of each pole piece can be quickly obtained according to the key point data and / or the segmentation regions, and then defect detection on a plurality of pole pieces of the battery to be detected can be completed according to the endpoint coordinates. The entire detection process is less affected by human subjective factors, and the detection result is more accurate. BRIEF DESCRIPTION OF DRAWINGS

[0065] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:

[0066] Figure 1 A flowchart of a pole piece defect detection method provided for an exemplary embodiment of the application;

[0067] Figure 2 A specific example diagram of pole piece image information at different angles provided for an exemplary embodiment of the application;

[0068] Figure 3 A specific example diagram of pole piece image information provided for an exemplary embodiment of the application;

[0069] Figure 4 A flowchart of a method for determining end point coordinates of each pole piece provided for an exemplary embodiment of the application;

[0070] Figure 5 A structural schematic diagram of a pole piece defect detection device provided for an exemplary embodiment of the application. DETAILED DESCRIPTION

[0071] In order to make the objects, technical solutions and advantages of the application clearer, the technical solutions of the application will be described below in detail with reference to the specific embodiments of the application and the corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the application, but not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0072] A lithium battery is a battery using a lithium metal or lithium alloy as a negative electrode material and a non-electrolyte solution. In order to prevent the occurrence of lithium branch crystals, the negative electrode of the lithium battery is required to have a certain redundancy length than the positive electrode, and the electrodes should be arranged in order and cannot have a large degree of bending and deformation. However, during the manufacturing process of the lithium battery, the winding or stacking may cause a certain fluctuation in the relative positions of the positive and negative electrodes of the lithium battery, thereby changing the boundary distance of the positive and negative electrodes and causing problems such as excessive redundancy of the negative electrode or no redundancy. In the related art, the redundancy and deformation of the positive and negative pole pieces of the lithium battery are generally judged by artificial experience and naked eye observation, which is greatly affected by subjective factors and cannot guarantee the accuracy of the pole piece defect detection of the lithium battery. In view of this, an embodiment of the application provides a pole piece defect detection method.

[0073] Figure 1 A flowchart of a pole piece defect detection method provided for an embodiment of the application is shown in FIG. 1, which includes the following steps. Figure 1

[0074] ​Step 101, obtaining tab image information of a battery to be detected, the battery to be detected containing a plurality of tabs.

[0075] It should be understood that one battery to be detected contains a plurality of positive tabs and a plurality of negative tabs, and the positive tab and the negative tab form a pair two by two, and are distributed in cross-over and stacking. For example, the number of tabs can be 59, of which 29 are positive tabs and 30 are negative tabs.

[0076] Step 102, inputting the tab image information into a key point model to obtain key point data in the tab image information, the key point data being used to reflect the end point positions of the tabs in the battery to be detected.

[0077] Step 103, inputting the tab image information into a segmentation model to obtain a segmentation region in the tab image information, the segmentation region being used to reflect the regions where the end points of the tabs in the battery to be detected are located.

[0078] Step 104, determining the end point coordinates of the tabs in the battery to be detected according to the key point data and / or the segmentation region.

[0079] Step 105, performing defect detection on the plurality of tabs of the battery to be detected according to the end point coordinates.

[0080] In actual application, an X-ray generator (also referred to as an Xray generator) can be used to emit X-rays to penetrate the interiors of the plurality of tabs of the battery, and a flat panel detector can be used to receive the X-rays to perform imaging, that is, to obtain the above-mentioned tab image information. In specific implementation, four flat panel detectors can be arranged at four corners of the plurality of tabs to respectively collect tab image information. For details, refer to Figure 2 For ease of description, one of the four tab image information is taken as an example for subsequent description:

[0081] After obtaining the tab image information, in order to improve the accuracy of subsequent tab end point finding, the obtained tab image information can be preprocessed to eliminate background interference. Then, the preprocessed tab image information is respectively input into a key point model and a segmentation model. Through the key point model, key point data in the tab image information is obtained. Through the segmentation model, a segmentation region in the tab image information is obtained. The key point data is used to reflect the end point positions of the tabs in the battery to be detected, and the segmentation region is used to reflect the regions where the end points of the tabs in the battery to be detected are located.

[0082] It should be noted that the key point model and the segmentation model are trained neural network models. Specifically, in training, the key point model can take the collected multiple pole piece images as training samples, and the key points marked on the pole piece images as labels for model training. And in training, the segmentation model can take the collected multiple pole piece images as training samples, and the segmented regions marked on the pole piece images as labels for model training.

[0083] After obtaining the key point data and the segmentation region through the key point model and the segmentation model, the end point coordinates of each pole piece in the battery to be detected can be determined according to the key point data and / or the segmentation region, and then the multiple pole pieces of the battery to be detected can be defectively detected according to the end point coordinates. The specific method of determining the end point coordinates can be seen in the subsequent embodiments, which will not be described here in detail.

[0084] It should be noted that before determining the end point coordinates of each pole piece in the battery to be detected according to the key point data and / or the segmentation region, the method further comprises: determining whether there are redundant end points on each pole piece; if there are, the redundant end points are removed.

[0085] It should be understood that in the process of finding points through the key point model and the segmentation model, there may be misidentification of the end points, which may result in multiple end points on one pole piece. At this time, the present application removes the redundant end points to ensure the accuracy of finding the end points of each pole piece.

[0086] In this embodiment, the battery to be detected contains multiple positive pole pieces and negative pole pieces arranged in a staggered manner; the defect detection of the multiple pole pieces of the battery to be detected according to the end point coordinates comprises:

[0087] According to the end point coordinates, the height of the multiple positive pole pieces and negative pole pieces, and the distance between every two adjacent positive pole pieces or negative pole pieces are determined.

[0088] Based on the height of the multiple positive pole pieces and negative pole pieces, and the distance between every two adjacent positive pole pieces or negative pole pieces, the defect detection of the multiple pole pieces of the battery to be detected is performed.

[0089] In order to facilitate understanding, the specific method of defect detection will be described below in combination with Figure 2 and Figure 3

[0090] In Figure 3 , the part enclosed by the box is the end point of the found pole piece, and Figure 2 , the four small pictures are the pole piece image information at four angles, and Figure 2 , there are 29 positive pole pieces and 30 negative pole pieces, Figure 2 ​The labels “L1, L2...L29” indicate the distance from each positive electrode to the negative electrode. Under normal circumstances, the negative electrode will be a certain distance higher than the positive electrode (here called the standard distance). By comparing whether “L1, L2...L29” is greater than or less than this standard distance within a certain range, it is proven that there is a problem with the redundancy of the electrode and the electrode has a defect.

[0091] and Figure 2 The "upper limit difference" in the equation represents the phase difference between the highest points of all positive electrode plates. If this phase difference is within the set range, it indicates that there are no particularly high or low endpoints among the positive electrode plates, meeting the requirements. Otherwise, it proves that the electrode plate has a defect. Similarly, the "lower limit difference" represents the phase difference between the highest points of all negative electrode plates. If this phase difference is within the set range, it indicates that there are no particularly high or low endpoints among the positive electrode plates, meeting the requirements. Otherwise, it proves that the electrode plate has a defect.

[0092] Figure 2 The "negative electrode drop" refers to the drop between two adjacent negative electrode points. If this drop is within the set range, it proves that the electrode meets the requirements; otherwise, it proves that the electrode has a defect. Similarly, the "positive electrode drop" refers to the drop between two adjacent positive electrode points. If this drop is within the set range, it proves that the electrode meets the requirements; otherwise, it proves that the electrode has a defect.

[0093] As can be seen from the above examples, the parameters for defect detection ("upper pole difference", "lower pole difference", "negative pole drop", "positive pole drop", etc.) can all be determined based on the height of multiple positive and negative electrode plates, as well as the distance between any two adjacent positive or negative electrode plates. Therefore, based on the height of multiple positive and negative electrode plates, as well as the distance between any two adjacent positive or negative electrode plates, the defect detection of multiple electrodes in the battery to be tested can be completed.

[0094] The electrode defect detection method provided in this application acquires electrode image information of the battery to be tested and inputs the electrode image information into a key point model and a segmentation model to obtain key point data and segmented regions in the electrode image information, respectively. Based on the key point data and / or segmented regions, the endpoint coordinates of each electrode in the battery to be tested can be determined, and then the defect detection of multiple electrodes of the battery to be tested can be completed according to the endpoint coordinates. In summary, this application can automatically acquire key point data reflecting the endpoint position of each electrode and segmented regions reflecting the endpoint regions of each electrode according to the key point model and the segmentation model, without manual intervention. The endpoint coordinates of each electrode can be quickly obtained according to the key point data and / or segmented regions, and then the defect detection of multiple electrodes of the battery to be tested can be completed according to the endpoint coordinates. The entire detection process is less affected by human subjective factors, and the detection results are more accurate.

[0095] In the light of the above, the specific method for determining the end point coordinates of each pole piece is introduced as follows, Figure 4 The flow chart for determining the end point coordinates of each pole piece is shown in FIG. 4, which includes the following steps: Figure 4

[0096] In step 401, if the number of end points of the plurality of pole pieces is equal to the number of key points in the key point data, but not equal to the number of regions in the segmentation region, then the end point coordinates of each pole piece are determined according to the key point data.

[0097] In step 402, if the number of end points of the plurality of pole pieces is equal to the number of regions in the segmentation region, but not equal to the number of key points in the key point data, then the end point coordinates of each pole piece are determined according to the segmentation region.

[0098] In step 403, if the number of end points of the plurality of pole pieces is not equal to the number of key points in the key point data, nor equal to the number of regions in the segmentation region, then the end point coordinates of each pole piece are determined according to the key point data and the segmentation region.

[0099] In a specific implementation, for example, it is assumed that the number of end points corresponding to the plurality of pole pieces is 59. Then:

[0100] As an implementation manner, if the number of key points is equal to 59, and the number of regions in the segmentation region is less than 59, then the key point data is taken as the final output end point coordinates.

[0101] As another implementation manner, if the number of regions in the segmentation region is equal to 59, and the number of key points is less than 59, then the coordinates of the specified points in the segmentation region are taken as the final output end point coordinates. It should be noted that there are multiple methods for determining the end point coordinates in the segmentation region, for example, the midpoint coordinates of the segmentation region can be taken as the end point coordinates, or a point at a certain position in the segmentation region can be set as the end point coordinates according to an experience value.

[0102] As still another implementation manner, the first end point coordinates of each pole piece can be determined according to the key point data, the second end point coordinates of each pole piece can be determined according to the segmentation region, and the end point coordinates of each pole piece can be determined based on the first end point coordinates and the second end point coordinates.

[0103] ​In a specific implementation, for example, if the number of key points is 57 and the number of regions in the segmented region is 58, i.e., the number of both is less than the number of end points 59 corresponding to the pole piece, the key point data and the segmented region are complemented by coordinate comparison, and 59 coordinate points are completed. Specifically, the coordinates corresponding to the 57 key points (first end point coordinates) can be determined first, and then the coordinates of the specified points corresponding to the 58 segmented regions (second end point coordinates) are determined. The coordinates of the two are compared one by one. If the coordinates corresponding to the key points and the coordinates of the specified points corresponding to the segmented regions are the same, it is determined that the same point is the final end point. If the coordinates corresponding to the key points are not found in the segmented region, the coordinates corresponding to the key points are supplemented into the segmented region. If 59 coordinate points can be completed at this time, the completed image is output as the final end point finding result.

[0104] In the embodiment, by comparing the number of end points of the plurality of pole pieces, the number of key points in the key point data, and the number of regions in the segmented region, the complete end point coordinates of each pole piece can be obtained without human intervention, and the accuracy is high.

[0105] Of course, after the key point data and the segmented region are complemented by coordinate comparison, there may be a few points. In order to further improve the accuracy of point finding, in the embodiment of the application, the end point coordinates of each pole piece are determined based on the first end point coordinates and the second end point coordinates, including:

[0106] The third end point coordinates of each pole piece are determined based on the set skeleton algorithm.

[0107] The end point coordinates of each pole piece are determined based on the first end point coordinates, the second end point coordinates, and the third end point coordinates.

[0108] In a specific implementation, based on the above example, for example, if 59 points cannot be completed by the key point data and the segmented region, the set skeleton algorithm is still used for point finding, i.e., the third end point coordinates of each pole piece are obtained, and the first end point coordinates, the second end point coordinates, and the third end point coordinates are complemented by coordinate comparison (in simple terms, if there is a skeleton between two points far apart, the search area can be determined according to the position of the skeleton and the left and right points). If 59 coordinate points can be completed smoothly, the completed image is output as the final end point finding result. If it still cannot be completed, the result of "error" or "end" is output.

[0109] Specific steps of finding points by using the skeleton algorithm can be: a Gaussian smoothing kernel second-order partial derivative is convolved with an image to establish a Hessian matrix to obtain a skeleton line normal direction, and a Taylor polynomial expansion is applied in the normal direction to calculate a skeleton line sub-pixel position.

[0110] By using the key point model, the segmentation region model and the skeleton algorithm, three point finding results (first end point coordinates, second end point coordinates and third end point coordinates) are obtained, and comprehensive analysis is performed on the three point finding results, so that the point finding efficiency and accuracy of each pole piece end point can be improved.

[0111] In addition, it should be understood that the pole piece will be severely deformed after being extruded, and it is meaningless to find the end points of each pole piece by using the key point model, the segmentation model and the skeleton algorithm after deformation, because it is definitely inaccurate. Therefore, before the pole piece image information is input into the key point model, the method further includes

[0112] inputting the pole piece image information into the deformation detection model to obtain the deformation result of each pole piece in the battery to be detected;

[0113] determining that the deformation result is that each pole piece is not deformed.

[0114] In actual application, based on a plurality of training samples (a plurality of images of pole pieces) and labels (deformed pole pieces) obtained in advance, the neural network model is trained to obtain a trained deformation detection model. Then, the pole piece image information is input into the trained deformation detection model, and the deformation result of each pole piece in the battery to be detected is obtained. Only when it is determined that the deformation result is that each pole piece is not deformed, the pole piece image information is input into the key point model and the segmentation model, and the subsequent steps are continued.

[0115] In summary, the pole piece image information of the battery to be detected is obtained, and the pole piece image information is input into the key point model and the segmentation model, and then the key point data in the pole piece image information and the segmentation region are obtained. Based on the key point data and / or the segmentation region, the end point coordinates of each pole piece in the battery to be detected are determined, and then the defect detection of the plurality of pole pieces of the battery to be detected is completed according to the end point coordinates. In summary, the key point data reflecting the end point position of each pole piece and the segmentation region reflecting the end point region of each pole piece can be automatically obtained according to the key point model and the segmentation model without human intervention. The end point coordinates of each pole piece can be quickly obtained according to the key point data and / or the segmentation region, and then the defect detection of the plurality of pole pieces of the battery to be detected is completed according to the end point coordinates. The entire detection process is less affected by human subjective factors, and the detection result is more accurate.

[0116] The advantages of the present application are summarized as follows:

[0117] 1. Higher efficiency: manual detection is low in efficiency, while the X-ray detection speed of the present application is much faster.

[0118] 2. Lower overall cost: machines are more efficient than manual detection, and in the long run, the cost of machine X-ray detection is lower.

[0119] 3. Information integration: X-ray detection equipment can measure multiple technical parameters at one time through multi-station detection method (four corners are detected at the same time), such as alignment of the product to be detected, positive electrode length, negative electrode bending angle, etc.

[0120] 4. Digital statistical management: SPC data statistics. In simple terms, a mark is given to each battery for subsequent calling and inquiry.

[0121] 5. More objective and stable: In the manual detection process, the detection results are affected by factors such as personal standards, emotions, energy, etc. The machine strictly follows the set standard, and the detection result is more objective, reliable and stable.

[0122] 6. Avoid secondary pollution: manual operation sometimes brings uncertain pollution sources, and polluted workpieces.

[0123] 7. Simple maintenance: low technical requirements for operators, long service life and other advantages.

[0124] 8. High adaptability to different forms of incoming materials, compatible with images with poor imaging effect, and accurate point finding.

[0125] Figure 5 A structure diagram of an electrode sheet defect detection device provided by an embodiment of the present application is shown in FIG. 1, which includes an information acquisition module 51, a key point acquisition module 52, a region acquisition module 53, a determination module 54, and a detection module 55. Figure 5

[0126] The information acquisition module 51 is configured to acquire electrode sheet image information of a battery to be detected, and the battery to be detected includes a plurality of electrode sheets.

[0127] The key point acquisition module 52 is configured to input the electrode sheet image information into a key point model to obtain key point data in the electrode sheet image information, and the key point data is used to reflect the end point positions of the electrode sheets in the battery to be detected.

[0128] The region acquisition module 53 is configured to input the electrode sheet image information into a segmentation model to obtain a segmentation region in the electrode sheet image information, and the segmentation region is used to reflect the regions where the end points of the electrode sheets in the battery to be detected are located.

[0129] The determination module 54 is configured to determine end point coordinates of the electrode sheets in the battery to be detected according to the key point data and / or the segmentation region.

[0130] The detection module 55 is configured to perform defect detection on the electrode sheets of the battery to be detected according to the end point coordinates.

[0131] In some embodiments of the present application, the determination module 54 is specifically configured to:

[0132] ​obtain a number of end points of the plurality of pole pieces, a number of key points in the key point data, and a number of regions in the segmented regions;

[0133] determine the end point coordinates of the pole pieces based on the number of end points of the plurality of pole pieces, the number of key points in the key point data, and / or the number of regions in the segmented regions.

[0134] In some embodiments of the present application, the determining module 54 is specifically configured to:

[0135] if the number of end points of the plurality of pole pieces is equal to the number of key points in the key point data and is not equal to the number of regions in the segmented regions, determine the end point coordinates of the pole pieces according to the key point data;

[0136] if the number of end points of the plurality of pole pieces is equal to the number of regions in the segmented regions and is not equal to the number of key points in the key point data, determine the end point coordinates of the pole pieces according to the segmented regions;

[0137] if the number of end points of the plurality of pole pieces is not equal to the number of key points in the key point data and is not equal to the number of regions in the segmented regions, determine the end point coordinates of the pole pieces according to the key point data and the segmented regions.

[0138] In some embodiments of the present application, the determining module 54 is specifically configured to:

[0139] determine first end point coordinates of the pole pieces according to the key point data;

[0140] determine second end point coordinates of the pole pieces according to the segmented regions;

[0141] determine the end point coordinates of the pole pieces based on the first end point coordinates and the second end point coordinates.

[0142] In some embodiments of the present application, the determining module 54 is specifically configured to:

[0143] determine third end point coordinates of the pole pieces based on a set skeleton algorithm;

[0144] determine the end point coordinates of the pole pieces based on the first end point coordinates, the second end point coordinates, and the third end point coordinates.

[0145] In some embodiments of the present application, the apparatus further comprises:

[0146] a determining module configured to determine whether there are redundant end points on each pole piece;

[0147] a removing module configured to remove the redundant end points when the redundant end points exist.

[0148] In some embodiments of the present application, the device further comprises:

[0149] a deformation detection module, configured to input the pole piece image information into a deformation detection model to obtain a deformation result of each pole piece in the battery to be detected;

[0150] a result determination module, configured to determine that the deformation result is that the each pole piece is not deformed.

[0151] In some embodiments of the present application, the battery to be detected comprises a plurality of positive pole pieces and negative pole pieces arranged in a staggered manner; and the detection module 55 is specifically configured to:

[0152] determine, according to the end point coordinates, a height of each of the plurality of positive pole pieces and negative pole pieces, and a distance between every two adjacent positive pole pieces or negative pole pieces;

[0153] perform defect detection on the plurality of pole pieces of the battery to be detected based on the height of each of the plurality of positive pole pieces and negative pole pieces, and the distance between every two adjacent positive pole pieces or negative pole pieces.

[0154] The embodiments of the present application also provide a non-transitory machine readable storage medium, which stores executable code, and when the executable code is executed by a processor of an electronic device, the processor performs the above pole piece defect detection method.

[0155] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0156] The present application is described with reference to flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in a flow or multiple flows and / or blocks Figure 1 The functions specified in a flow or multiple flows and / or blocks

[0157] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0158] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions that are executed on the computer or other programmable apparatus provide steps for implementing the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0159] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0160] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) and / or cache memory, non-volatile memory, such as read-only memory (ROM), EPROM, and / or flash memory. The memory is an example of computer readable media.

[0161] Computer readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology for storing information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer readable media does not include transitory media, such as modulated data signals and carrier waves.

[0162] It should also be noted that the terms "comprising", "comprises" or other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0163] The above embodiments are only used to illustrate the present application, but not to limit it. Instead of the above, various modifications and changes can be made to the application by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application shall fall into the scope of the claims of the application.

Claims

1. A method for detecting electrode defects, characterized in that, include: Acquire electrode image information of the battery to be tested, wherein the battery to be tested contains multiple electrodes; The electrode image information is input into the key point model to obtain key point data in the electrode image information. The key point data is used to reflect the endpoint positions of each electrode in the battery to be tested. The electrode image information is input into the segmentation model to obtain the segmentation region in the electrode image information. The segmentation region is used to reflect the region where the endpoints of each electrode in the battery to be detected are located. The method involves obtaining the number of endpoints of the plurality of electrodes, the number of key points in the key point data, and the number of regions in the segmented region. If the number of endpoints of the plurality of electrodes is equal to the number of key points in the key point data but not equal to the number of regions in the segmented region, then the endpoint coordinates of each electrode are determined based on the key point data. If the number of endpoints of the plurality of electrodes is equal to the number of regions in the segmented region but not equal to the number of key points in the key point data, then the endpoint coordinates of each electrode are determined based on the segmented region. If the number of endpoints of the plurality of electrodes is not equal to the number of key points in the key point data or the number of regions in the segmented region, then the endpoint coordinates of each electrode are determined based on the key point data and the segmented region. Based on the endpoint coordinates, defect detection is performed on multiple electrodes of the battery to be tested.

2. The method according to claim 1, characterized in that, Determining the endpoint coordinates of each electrode based on the key point data and the segmented region includes: Based on the key point data, determine the coordinates of the first endpoint of each electrode. Based on the segmented region, determine the coordinates of the second endpoint of each electrode; The endpoint coordinates of each electrode are determined based on the first endpoint coordinates and the second endpoint coordinates.

3. The method according to claim 2, characterized in that, Determining the endpoint coordinates of each electrode based on the first endpoint coordinates and the second endpoint coordinates includes: Based on the established skeleton algorithm, the coordinates of the third endpoint of each electrode are determined; The endpoint coordinates of each electrode are determined based on the first endpoint coordinates, the second endpoint coordinates, and the third endpoint coordinates.

4. The method according to claim 1, characterized in that, Before obtaining the number of endpoints of the plurality of pole pieces, the number of key points in the key point data, and the number of regions in the segmented region, the method further includes: Determine if there are any redundant endpoints on each electrode; If they exist, the redundant endpoints are removed.

5. The method according to claim 1, characterized in that, Before inputting the polarimetric image information into the key point model, the method further includes: The electrode image information is input into the deformation detection model to obtain the deformation results of each electrode in the battery to be tested; The deformation result is determined to be that none of the electrodes are deformed.

6. The method according to claim 1, characterized in that, The battery under test includes multiple staggered positive and negative electrode plates; the defect detection of the multiple electrode plates of the battery under test based on the endpoint coordinates includes: Based on the endpoint coordinates, determine the height of the plurality of positive electrode plates and the negative electrode plates, as well as the distance between any two adjacent positive electrode plates or negative electrode plates; Based on the height of the multiple positive and negative electrode plates, and the distance between any two adjacent positive or negative electrode plates, defect detection is performed on the multiple electrode plates of the battery under test.

7. A device for detecting electrode defects, characterized in that, include: The information acquisition module is used to acquire electrode image information of the battery to be tested, wherein the battery to be tested contains multiple electrodes; The key point acquisition module is used to input the electrode image information into the key point model to obtain key point data in the electrode image information. The key point data is used to reflect the endpoint positions of each electrode in the battery to be tested. The region acquisition module is used to input the electrode image information into the segmentation model to obtain the segmented region in the electrode image information. The segmented region is used to reflect the region where the endpoints of each electrode in the battery to be detected are located. A determining module is used to obtain the number of endpoints of the plurality of electrodes, the number of key points in the key point data, and the number of regions in the segmented region; if the number of endpoints of the plurality of electrodes is equal to the number of key points in the key point data, but not equal to the number of regions in the segmented region, then the endpoint coordinates of each electrode are determined based on the key point data; if the number of endpoints of the plurality of electrodes is equal to the number of regions in the segmented region, but not equal to the number of key points in the key point data, then the endpoint coordinates of each electrode are determined based on the segmented region; if the number of endpoints of the plurality of electrodes is not equal to the number of key points in the key point data and the number of regions in the segmented region, then the endpoint coordinates of each electrode are determined based on the key point data and the segmented region. The detection module is used to perform defect detection on the electrode of the battery to be tested based on the endpoint coordinates.

8. A non-transitory machine-readable storage medium, characterized in that, The non-transitory machine-readable storage medium stores executable code that, when executed by a processor of an electronic device, causes the processor to perform the method as described in any one of claims 1 to 6.

Images