Battery testing method and apparatus, model training method and apparatus, production system, and device

Abstract

The present application relates to the field of battery testing, and provided thereby are a battery testing method and apparatus, a model training method and apparatus, a production system, and a device. The battery testing method comprises: acquiring a test image containing a battery to be tested, the test image comprising a plurality of first pole pieces and a plurality of second pole pieces of said battery which are alternately arranged in a first direction, wherein the polarity of the first pole pieces and the polarity of the second pole pieces are different; on the basis of the test image, determining first pole piece fit lines used for characterizing the first pole pieces and second pole piece fit lines used for characterizing the second pole pieces, the first pole piece fit lines and the second pole piece fit lines being spaced apart in the first direction; and on the basis of adjacent first pole piece fit lines and second pole piece fit lines, determining pole piece gaps of the battery to be tested. When the contours of the pole pieces or the features of the pole pieces in the test image are discontinuous, accurate pole piece gap test data can be obtained by means of the continuous pole piece fit lines.

Description

Battery detection method, model training method, device, production system and equipment

[0001] Cross-references

[0002] This application refers to Chinese patent application No. 202310973129.X filed on August 3, 2023, entitled “Battery detection method, model training method, device, production system and equipment”, which is incorporated into this application in its entirety by reference. Technical Field

[0003] The present application relates to the field of battery detection, and in particular to a battery detection method, a battery detection device, a segmentation model training method for battery detection, a segmentation model training device for battery detection, a battery production system, an electronic device, a computer-readable storage medium, and a computer program product. Background Art

[0004] Energy conservation and emission reduction are key to the sustainable development of the automotive industry. Electric vehicles, due to their energy-saving and environmentally friendly advantages, have become an important component of the sustainable development of the automotive industry. For electric vehicles, battery technology is a key factor in their development.

[0005] To understand the internal conditions of a battery, related technologies employ a detection device to inspect the battery's interior. This device scans the battery to obtain a detection image, which is then used to understand the battery's internal conditions. Poor quality of the acquired detection image can result in discontinuities in the electrode features in the processed image, affecting the accuracy of electrode parameter measurements based on the image.

[0006] Summary of the Invention

[0007] The present application aims to solve at least one of the technical problems that exist in some situations. To this end, one object of the present application is to provide a battery detection method, a battery detection device, a segmentation model training method for battery detection, a segmentation model training device for battery detection, a battery production system, an electronic device, a computer-readable storage medium, and a computer program product, so as to improve the accuracy of electrode parameter measurement when the quality of the acquired detection image is poor.

[0008] An embodiment of the first aspect of the present application provides a battery detection method, comprising: acquiring a detection image including a battery to be tested, the detection image including a plurality of first pole pieces and a plurality of second pole pieces of the battery to be tested, the first pole pieces and the second pole pieces having different polarities, determining a first pole piece fitting line for characterizing the first pole piece and a second pole piece fitting line for characterizing the second pole piece based on the detection image, the first pole piece fitting line and the second pole piece fitting line being spaced apart along the first direction, and determining a pole piece gap of the battery to be tested based on adjacent first pole piece fitting lines and second pole piece fitting lines.

[0009] In the technical solution of the embodiment of the present application, the pole pieces in the detection image are converted into pole piece fitting lines. When the outline or pole piece features in the detection image are discontinuous, more accurate pole piece gap detection data can be obtained through the continuous pole piece fitting lines.

[0010] In some embodiments, the steps of determining a first pole piece fitting line for characterizing the first pole piece and a second pole piece fitting line for characterizing the second pole piece based on the detection image include: obtaining the first endpoint and the second endpoint of the first pole piece based on the detection image, and obtaining the third endpoint and the fourth endpoint of the second pole piece; obtaining the first pole piece fitting line corresponding to the first pole piece based on the line connecting the first endpoint and the second endpoint, and obtaining the second pole piece fitting line corresponding to the second pole piece based on the line connecting the third endpoint and the fourth endpoint. By marking the two endpoints of the pole piece and connecting the endpoints to obtain the pole piece fitting line corresponding to the pole piece, the pole piece fitting line can be obtained with fewer marking points, and since the consistency of the alternating multiple first pole pieces and the multiple second pole pieces is high, the pole piece fitting line obtained by connecting the lines can restore the characteristics of the pole piece to a certain extent, which is convenient for subsequent measurement.

[0011] In some embodiments, the battery under test includes a wound cell; the detection image is obtained by projecting radiation onto the surface of the wound cell, and the detection image includes any one of the four corners of the wound cell surface. For a wound cell, detecting the four corners of the wound cell surface can obtain a detection image including a plurality of first electrode sheets and a plurality of second electrode sheets arranged alternately, facilitating the determination of the electrode sheet spacing of the wound cell.

[0012] In some embodiments, the steps of obtaining the first and second endpoints of the first pole piece based on the detection image, and obtaining the third and fourth endpoints of the second pole piece include: inputting the detection image into a segmentation model, obtaining a pole piece mask image output by the segmentation model, the pole piece mask image including a background area and multiple mask areas, each mask area corresponding to the area where the first pole piece or the second pole piece is located; determining the first and second endpoints of the first pole piece based on the pole piece mask image, and determining the third and fourth endpoints of the second pole piece. The pole piece mask image output by the trained segmentation model can output a complete pole piece mask when the pole piece contour or pole piece feature is discontinuous in the detection image, thereby restoring the features of the pole piece to a certain extent and facilitating subsequent measurement.

[0013] In some embodiments, the steps of determining the first endpoint and the second endpoint of the first pole piece based on the pole piece mask diagram, and determining the third endpoint and the fourth endpoint of the second pole piece include: a plurality of mask areas including a plurality of first mask areas and a plurality of second mask areas, the first mask area corresponding to the area where the first pole piece is located, and the second mask area corresponding to the area where the second pole piece is located; the first mask area including a first top edge and a first bottom edge spaced apart from the first top edge along a second direction, the first endpoint being located at the first top edge, the second endpoint being located at the first bottom edge, wherein the second direction is the extension direction of the first pole piece; the second mask area including a second top edge and a second bottom edge spaced apart from the second top edge along a third direction, the third endpoint being located at the second top edge, the fourth endpoint being located at the second bottom edge, wherein the third direction is the extension direction of the first pole piece. By searching for the top edge and the bottom edge in the independent mask areas on the pole piece mask diagram, the starting point and the end point of the pole piece in the detection image can be better located, so that the obtained pole piece fitting line is closer to the original feature of the pole piece in the detection image.

[0014] In some embodiments, the first endpoint is the midpoint of the first top edge, the second endpoint is the midpoint of the first bottom edge, the third endpoint is the midpoint of the second top edge, and the fourth endpoint is the midpoint of the second bottom edge. This allows the first pole piece fitting line to be the centerline of the first mask region corresponding to the first pole piece, and the second pole piece fitting line to be the centerline of the second mask region corresponding to the second pole piece. The pole piece fitting lines are obtained based on the centerlines, so that the obtained pole piece fitting lines are closer to the original features of the pole pieces in the detection image. At the same time, adjacent pole piece fitting lines are selected based on the same standard, further improving the accuracy of subsequent measurements.

[0015] In some embodiments, the first mask area includes a first side extending along a second direction, the first endpoint is the vertex of the first side, and the second endpoint is the end point of the first side; the second mask area includes a second side extending along a third direction, the third endpoint is the vertex of the second side, and the fourth endpoint is the end point of the second side. In this case, the first pole piece fitting line is a side edge line of the first pole piece, and the second pole piece fitting line is a side edge line of the second pole piece. The pole piece fitting line is obtained based on the side edge line, so that the obtained pole piece fitting line is closer to the original features of the pole piece in the detection image, and at the same time, adjacent pole piece fitting lines are selected based on the same standard, further improving the accuracy of subsequent measurements.

[0016] In some embodiments, the step of determining the electrode gap of the battery to be tested based on adjacent first electrode piece fitting lines and second electrode piece fitting lines includes: obtaining a first distance between the adjacent first electrode piece fitting lines and the second electrode piece fitting lines along a first direction based on the adjacent first electrode piece fitting lines and the second electrode piece fitting lines, wherein the first distance is used to characterize the electrode gap of the battery to be tested; wherein the first distance is any one of a minimum distance, a maximum distance, and an average distance between the adjacent first electrode piece fitting lines and the second electrode piece fitting lines. Characterizing the electrode gap of the battery to be tested based on the distance between the first electrode piece fitting lines and the second electrode piece fitting lines can obtain relatively accurate electrode gap detection data through continuous electrode piece fitting lines.

[0017] In some embodiments, before determining a first pole piece fitting line for characterizing the first pole piece and a second pole piece fitting line for characterizing the second pole piece based on the detection image, the method further includes: performing image preprocessing on the detection image, and obtaining a plurality of first pole piece fitting lines and a plurality of second pole piece fitting lines based on the detection image after image preprocessing. By preprocessing the detection image, the interference of noise in the detection image can be reduced, making the object to be detected in the image more obvious than other areas, thereby making the pole piece features in the image more easily identifiable.

[0018] In some embodiments, the image preprocessing step includes: performing local enhancement processing on the detection image. Performing image enhancement processing on the detection image can further enhance the contrast of the image, thereby making detailed features in the image easier to identify.

[0019] In some embodiments, the local enhancement processing step includes: extracting a gradient image of the detection image; performing gradient compression on the gradient image to obtain a corresponding compressed gradient image; and performing image reconstruction based on the compressed gradient image to obtain a reconstructed detection image. By extracting the gradient image and compressing the gradient image to reduce the gradient value in the compressed gradient image, the change in the grayscale value in the reconstructed image obtained by reconstructing the compressed gradient image will also be reduced. For the reconstructed image, even if the grayscale values ​​between adjacent pixels are not much different, they can be more easily identified from the reconstructed image, making the detailed features in the reconstructed image clearer and improving the accuracy of the image recognition results. Using this method in the processing of battery detection images can improve the accuracy of battery detection results.

[0020] In some embodiments, the step of gradient compression includes: determining a contour image of the object to be detected based on a gradient image; determining a compression value of the corresponding pixel based on the gradient amplitude of each pixel in the contour image; extracting the gradient value of each pixel in the gradient image; obtaining a compressed gradient value of the corresponding pixel based on the gradient value of each pixel and the compression value of the corresponding pixel; and obtaining a corresponding compressed gradient image based on the compressed gradient value of each pixel. The contour image of the object to be detected is first determined using the gradient image, where each pixel in the contour image includes a gradient amplitude. Then, the compression value of each pixel is determined based on the gradient amplitude of each pixel. Thus, during subsequent compression processing, each pixel can be compressed. Furthermore, since the compression values ​​of each pixel are not exactly the same, different pixels can be compressed differently, with greater compression being achieved for pixels with larger gradient values ​​and less compression being achieved for pixels with smaller gradient values. Since the gradient values ​​in the first compressed gradient image and the second compressed gradient image are reduced, the difference in grayscale values ​​between pixels in a subsequent reconstructed image can be reduced, making it easier to identify detailed features in the reconstructed image.

[0021] The embodiment of the second aspect of the present application provides a battery detection device, comprising: an acquisition module configured to acquire a detection image containing a battery to be tested, the detection image comprising a plurality of first pole pieces and a plurality of second pole pieces of the battery to be tested, which are alternately arranged along a first direction, wherein the polarity of the first pole piece and the second pole piece are different; a fitting module configured to determine a first pole piece fitting line for characterizing the first pole piece and a second pole piece fitting line for characterizing the second pole piece based on the detection image, wherein the first pole piece fitting line and the second pole piece fitting line are spaced apart along the first direction; and a processing module configured to determine the pole piece gap of the battery to be tested based on adjacent first pole piece fitting lines and second pole piece fitting lines. The pole pieces in the detection image are converted into pole piece fitting lines. When the outline or pole piece features of the pole pieces in the detection image are discontinuous, more accurate pole piece gap detection data can be obtained through continuous pole piece fitting lines.

[0022] An embodiment of the third aspect of the present application provides a segmentation model training method for battery detection, including: obtaining a sample image containing a sample battery, the sample image including a plurality of first sample pole pieces and a plurality of second sample pole pieces of the sample battery arranged alternately along a first direction; obtaining a real pole piece mask image corresponding to the sample image, the real pole piece mask image including a real background area and a plurality of real mask areas, each real mask area corresponding to the area where the first sample pole piece or the second sample pole piece is located; inputting the sample image into the segmentation model to be trained, and obtaining a predicted pole piece mask image output by the segmentation model; calculating a loss value based on the real pole piece mask image and the predicted pole piece mask image; and adjusting the parameters of the segmentation model based on the loss value. The pole piece mask image output by the trained segmentation model can output a complete pole piece mask area when the pole piece contour or pole piece feature in the detection image is discontinuous. Using the real mask area as the true value of the training segmentation model can improve the accuracy of the segmentation model prediction, restore the pole piece features to a certain extent, and facilitate subsequent measurement.

[0023] In some embodiments, the step of obtaining a true pole piece mask image corresponding to a sample image includes: based on the sample image, marking a first annotation point and a second annotation point corresponding to a first sample pole piece, and marking a third annotation point and a fourth annotation point corresponding to a second sample pole piece; based on the marked first annotation point and the second annotation point of the first sample pole piece, and a preset first expansion value, obtaining a true mask area corresponding to the first pole piece; based on the marked third annotation point and the third annotation point of the second sample pole piece, and a preset second expansion value, obtaining a true mask area corresponding to the second pole piece. Marking only two annotation points for a sample pole piece can reduce the difficulty of annotation, reduce the annotation workload, and improve the annotation efficiency. At the same time, when the pole piece contour or pole piece feature is discontinuous in the sample image, the complete true mask area can be output as the true value for model training, so that the trained segmentation model can predict the complete mask area based on the discontinuous pole piece contour.

[0024] An embodiment of the fourth aspect of the present application provides a segmentation model training device for battery detection, including: a sample acquisition module, configured to acquire a sample image containing a sample battery, the sample image including a plurality of first sample pole pieces and a plurality of second sample pole pieces of the sample battery arranged alternately along a first direction; a truth module, configured to acquire a real pole piece mask image corresponding to the sample image, the real pole piece mask image including a real background area and a plurality of real mask areas, each real mask area corresponding to the area where the first sample pole piece or the second sample pole piece is located; a sample prediction module, configured to input the sample image into the segmentation model to be trained, and obtain a predicted pole piece mask image output by the segmentation model; a loss module, configured to calculate a loss value based on the real pole piece mask image and the predicted pole piece mask image; a parameter adjustment module, configured to adjust the parameters of the segmentation model based on the loss value.

[0025] The pole piece mask output by the trained segmentation model can output the complete pole piece mask area when the pole piece outline or pole piece features in the detection image are discontinuous. Using the actual mask area as the ground truth for the trained segmentation model can improve the accuracy of the segmentation model's predictions and restore the pole piece features to a certain extent, facilitating subsequent measurements.

[0026] An embodiment of the fifth aspect of the present application provides an electronic device, comprising at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the battery detection method in the above embodiment or the segmentation model training method in the above embodiment.

[0027] An embodiment of the sixth aspect of the present application provides a battery production system, including the electronic device in the above embodiment.

[0028] An embodiment of the seventh aspect of the present application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the battery detection method in the above embodiment or the segmentation model training method in the above embodiment is implemented.

[0029] An embodiment of the eighth aspect of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the battery detection method in the above embodiment or the segmentation model training method in the above embodiment.

[0030] 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

[0031] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the multiple drawings represent the same or similar components or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings only depict some embodiments disclosed in this application and should not be construed as limiting the scope of this application.

[0032] FIG1 is a flow chart of a battery detection method provided by some embodiments of the present application;

[0033] 2 is a sub-flowchart of determining a first pole piece fitting line for characterizing a first pole piece and a second pole piece fitting line for characterizing a second pole piece based on a detection image in a battery detection method provided by some embodiments of the present application;

[0034] FIG3 is a schematic diagram of a pole piece mask image output by a segmentation model in a battery detection method provided in some embodiments of the present application;

[0035] FIG4 is a structural block diagram of a battery detection device provided in some embodiments of the present application;

[0036] FIG5 is a flowchart of a segmentation model training method for battery detection provided in some embodiments of the present application;

[0037] FIG6 is a structural block diagram of a segmentation model training device for battery detection provided by some embodiments of the present application;

[0038] FIG7 illustrates an example configuration of a computer device that may be used to implement the methods described herein. DETAILED DESCRIPTION

[0039] The following embodiments of the technical solution of the present application will be described in detail with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application and are therefore only examples and are not intended to limit the scope of protection of the present application.

[0040] 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 for the purpose of describing specific embodiments only 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.

[0041] 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.

[0042] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments 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 constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0043] 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.

[0044] 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).

[0045] 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. indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings. They 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 referred device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, they should not be understood as limiting the embodiments of the present application.

[0046] In the description of the embodiments of the present application, unless otherwise expressly specified or limited, technical terms such as "installed," "connected," "connect," 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.

[0047] Currently, market developments indicate that power batteries are becoming increasingly widely used. They are not only used in energy storage systems such as hydropower, thermal, wind, and solar power plants, but are also widely used in electric vehicles like electric bicycles, electric motorcycles, and electric vehicles, as well as in military equipment and aerospace. As power battery applications continue to expand, market demand is also growing.

[0048] Batteries generally include an electrode assembly and a casing that wraps the electrode assembly. The interior of the battery is wrapped in the casing. If changes occur inside the battery, it cannot be detected from the outside. The battery needs to be scanned and inspected using a radiographic image acquisition device to obtain a test image of the battery. By analyzing the test image, the internal situation of the battery can be determined, thereby achieving battery detection.

[0049] Taking lithium-ion batteries as an example, the gap between the cathode and anode sheets and the overhang (OH) between the cathode and anode sheets are important factors affecting the safety performance of wound batteries. Regarding the gap between the cathode and anode sheets: During the charge and discharge process of the battery, when the gap between the cathode and anode sheets is large, the lithium ions cannot be fully embedded in the anode channel, resulting in the accumulation of some lithium ions and the occurrence of lithium plating, which poses a greater safety risk. Therefore, it is required that the gap between the cathode and anode sheets in the battery should not be too large. Regarding the overhang: During the charge and discharge process of the lithium battery, lithium ions will repeatedly deintercalate between the cathode and anode sheets. However, when lithium ions are released from the cathode but there are not enough anode channels to embed them, it is easy to cause lithium ion accumulation and lithium plating, causing certain product safety risks. Therefore, it is necessary to ensure that the cathode sheet is within the coating range of the anode as a whole, that is, to ensure that the anode exceeds the cathode, that is, the gap between the cathode and anode sheets is greater than 0. Therefore, it is necessary to detect the internal spacing and the gap between the cathode and anode sheets in order to promptly detect internal defects in the battery and reduce the outflow of defective batteries. For wound batteries, the inner ring is much thicker than the outer ring. The detection radiation from the radiographic image acquisition device penetrates the inner and outer rings at different thicknesses, resulting in significant grayscale differences in the detection image corresponding to the inner and outer rings. However, the thickness difference between adjacent rings is relatively small, resulting in smaller grayscale variations in the detection image. The grayscale difference between adjacent rings is very small compared to the maximum grayscale difference, making it difficult to distinguish adjacent rings in the detection image. This results in unclear features in the detection image, affecting battery inspection results. Furthermore, if the grayscale values ​​of pixels in the detection image are similar, the variation between pixels with similar grayscale values ​​is relatively small, making it difficult to distinguish these pixels. This results in discontinuity in the detection image, affecting the accuracy of internal battery spacing and misalignment of the cathode and anode electrodes.

[0050] An embodiment of the present application provides a battery detection method, the method comprising: acquiring a detection image comprising a battery to be tested, the detection image comprising a plurality of first pole pieces and a plurality of second pole pieces alternately arranged along a first direction of the battery to be tested, wherein the polarity of the first pole piece and the second pole piece are different. Based on the detection image, a first pole piece fitting line for characterizing the first pole piece and a second pole piece fitting line for characterizing the second pole piece are determined, the first pole piece fitting line and the second pole piece fitting line being spaced apart along the first direction. Based on the adjacent first pole piece fitting lines and the second pole piece fitting lines, the pole piece gap of the battery to be tested is determined. The plurality of first pole pieces and the plurality of second pole pieces alternately arranged along the first direction are characterized by the first pole piece fitting line and the second pole piece fitting line. When the outline or pole piece feature of the pole piece in the detection image is discontinuous, more accurate pole piece gap detection data can also be obtained through the continuous pole piece fitting line.

[0051] The battery detection method disclosed in the embodiment of the present application can be used to detect the gap between pole pieces. For detection images of different qualities, it can provide relatively accurate pole piece gap detection data, thereby improving the accuracy of pole piece gap detection to a certain extent.

[0052] Referring to FIG1 , FIG1 is a flow chart of a battery detection method provided in some embodiments of the present application. The battery detection method 100 includes:

[0053] S110 , acquiring a detection image of a battery to be tested, where the detection image includes a plurality of first pole pieces and a plurality of second pole pieces alternately arranged along a first direction of the battery to be tested, wherein the first pole pieces and the second pole pieces have different polarities.

[0054] S120, determining a first pole piece fitting line for characterizing the first pole piece and a second pole piece fitting line for characterizing the second pole piece based on the detection image, wherein the first pole piece fitting line and the second pole piece fitting line are spaced apart along a first direction.

[0055] S130 , determining a gap between electrode sheets of the battery to be tested based on adjacent first electrode sheet fitting lines and second electrode sheet fitting lines.

[0056] In an embodiment of the present application, the object to be inspected may be a battery, a battery cell, or any other object that can be inspected by a radiographic image acquisition device.

[0057] In an embodiment of the present application, the detection image is an image acquired by an image acquisition device. Exemplarily, the detection image may be a processed image.

[0058] Exemplarily, the image acquisition device may be a radiographic image acquisition device, a CT scanner, or a PET scanner, and the detection image may be a detection image acquired by the radiographic image acquisition device. Exemplarily, the radiographic image acquisition device may include a radiation source and a flat-panel detector, wherein the detection image may be acquired by the radiation source and the flat-panel detector. Radiation emitted by the radiation source is projected onto the flat-panel detector through the battery under test, and the flat-panel detector acquires the detection image based on the received radiation. The radiation emitted by the radiation source may be X-rays, gamma rays, beta rays, or the like.

[0059] In the embodiment of the present application, the first pole piece may be a cathode pole piece, and the second pole piece may be an anode pole piece. Of course, it is understandable that in some cases, the first pole piece may also be an anode pole piece, and the second pole piece may be a cathode pole piece.

[0060] Exemplarily, the first pole piece fitting line and the second pole piece fitting line can be obtained by algorithm prediction or by manual annotation. When the first pole piece fitting line and the second pole piece fitting line are obtained by algorithm prediction, multiple annotation points representing a single pole piece can be obtained by the pole piece prediction result of the trained model, and the multiple annotation points in the single pole piece are connected to form a first pole piece fitting line corresponding to the first pole piece or a second pole piece fitting line corresponding to the second pole piece. When manual annotation is used to obtain the first pole piece fitting line and the second pole piece fitting line, the annotator can obtain multiple annotation points representing a single pole piece by punctuation on the detection image, and connect the multiple annotation points in the single pole piece to form a first pole piece fitting line corresponding to the first pole piece or a second pole piece fitting line corresponding to the second pole piece.

[0061] When the number of marked points on a single pole piece is greater than two, the first pole piece fitting line and the second pole piece fitting line can be a broken line connecting the multiple marked points in sequence, or a fitted straight line of the multiple marked points obtained by linear regression. When the number of marked points on a single pole piece is equal to two, the first pole piece fitting line or the second pole piece fitting line is obtained based on the line connecting the two marked points on the single pole piece.

[0062] In an embodiment of the present application, a first pole piece fitting line corresponding to each first pole piece in the detection image and a second pole piece fitting line corresponding to each second pole piece can be determined. It is also possible to determine only the first pole piece fitting lines corresponding to some of the first pole pieces in the detection image and the second pole piece fitting lines corresponding to some of the second pole pieces. For example, the edge area of ​​the detection image may be distorted due to shooting limitations, so the first pole piece fitting line and the second pole piece fitting line can be determined by selecting the first pole piece and the second pole piece in a partial area (non-distorted area).

[0063] In an embodiment of the present application, the first electrode in the detection image can be characterized by a continuous first electrode fitting line, and the second electrode in the detection image can be characterized by a continuous second electrode fitting line. Since the first electrode and the second electrode are arranged alternately along the first direction and have a high consistency, the first electrode fitting line and the second electrode fitting line should also be fitted based on the same standard, so that the obtained first electrode fitting line and the second electrode fitting line are more consistent with the corresponding electrode features on the basis of conforming to the arrangement rules of the first electrode and the second electrode, so as to facilitate obtaining the electrode gap of the battery to be tested by measuring the relevant detection data of the adjacent first electrode fitting line and the second electrode fitting line. Wherein, the adjacent first electrode fitting line and the second electrode fitting line means that there is no other electrode fitting line between the two electrode fitting lines. Since the first electrode and the second electrode are arranged alternately along the first direction, the electrode gap of the battery to be tested is the misalignment of the cathode and anode electrodes in the first direction. The first electrode fitting line can extend along the second direction, wherein the second direction intersects with the first direction, and the second electrode fitting line can extend along the third direction, wherein the first direction intersects with the third direction. For example, the second direction may be parallel to the third direction. For example, the first direction may be perpendicular to the second direction or the first direction may be perpendicular to the third direction.

[0064] The first pole piece fitting line and the second pole piece fitting line are used to represent a plurality of first pole pieces and a plurality of second pole pieces arranged alternately along a first direction. Even when the outline or pole piece features in the detection image are discontinuous, more accurate pole piece gap detection data can be obtained by using the continuous pole piece fitting line.

[0065] Referring to FIG. 2 , FIG. 2 is a sub-flowchart of step S120 in the battery detection method provided in some embodiments of the present application. According to some embodiments of the present application, step S120 of determining a first electrode piece fitting line for characterizing the first electrode piece and a second electrode piece fitting line for characterizing the second electrode piece based on the detection image includes:

[0066] S121, obtaining a first endpoint and a second endpoint of the first electrode piece, and obtaining a third endpoint and a fourth endpoint of the second electrode piece based on the detection image;

[0067] S122, obtaining a first pole piece fitting line corresponding to the first pole piece based on a line connecting the first endpoint and the second endpoint, and obtaining a second pole piece fitting line corresponding to the second pole piece based on a line connecting the third endpoint and the fourth endpoint.

[0068] Obtaining the first endpoint and the second endpoint of the first pole piece based on the detection image can be understood as obtaining two feature points corresponding to the first pole piece on the detection image. These two feature points can be feature points representing the head of the first pole piece, and feature points representing the tail of the first pole piece. It can be understood that these feature points do not actually represent the actual starting point and the actual tail of the first pole piece, but are used to represent the starting point of the first pole piece that can be observed on the detection image (that is, the head feature point) and the end point of the first pole piece that can be observed on the detection image (that is, the tail feature point).

[0069] Obtaining the third endpoint and the fourth endpoint of the second pole piece based on the detection image can be understood as obtaining two feature points corresponding to the second pole piece on the detection image. These two feature points can be feature points representing the head of the second pole piece, and feature points representing the tail of the second pole piece. It can be understood that these feature points do not actually represent the actual starting point and the actual tail of the second pole piece, but are used to represent the starting point of the second pole piece that can be observed on the detection image (that is, the head feature point) and the end point of the second pole piece that can be observed on the detection image (that is, the tail feature point).

[0070] The first endpoint, the second endpoint, the third endpoint, and the fourth endpoint can be obtained by manual punctuation or by machine punctuation.

[0071] In the embodiment of the present application, the first endpoint and the third endpoint should be selected based on the same marked reference, and the second endpoint and the fourth endpoint should be selected based on the same marked reference, so that the interval between the adjacent first electrode piece fitting line and the second electrode piece fitting line is equal to the interval between the adjacent first electrode piece and the second electrode piece in the detection image.

[0072] For the detection image, the pole piece detail features are represented by pixels of different grayscale values. Since the quality of the detection image may be good or bad, the corresponding pole piece detail features may be complete or missing. By marking the head and tail of the pole piece and connecting them, the missing part of the pole piece features in the detection image can be ignored. The pole piece fitting line obtained by connecting the lines can restore the pole piece features to a certain extent, which is convenient for subsequent measurement.

[0073] According to some embodiments of the present application, the battery to be tested includes a wound battery cell; the detection image is obtained by projecting rays onto the surface of the wound battery cell for shooting, and the detection image includes any one of the four corners of the surface of the wound battery cell.

[0074] For wound batteries, in order to obtain multiple first pole pieces and multiple second pole pieces arranged alternately along the first direction, it is necessary to project rays onto the surface of the wound battery cell for shooting. The surface of the wound battery cell can be the large surface direction of the wound battery cell. For multiple first pole pieces and multiple second pole pieces arranged alternately along the first direction, it is necessary to shoot the four corners of the surface of the wound battery cell. When shooting other areas of the surface of the wound battery cell (except the four corners), due to the characteristics of the winding, it is impossible to obtain the detection image of the multiple first pole pieces and multiple second pole pieces arranged alternately along the first direction.

[0075] For a wound battery cell, detecting the four corners of the surface of the wound battery cell can obtain a detection image including a plurality of first pole pieces and a plurality of second pole pieces arranged alternately, which facilitates obtaining the pole piece spacing of the wound battery cell.

[0076] According to some embodiments of the present application, the steps of obtaining the first endpoint and the second endpoint of the first pole piece based on the detection image, and obtaining the third endpoint and the fourth endpoint of the second pole piece include: inputting the detection image into the segmentation model, obtaining the pole piece mask map output by the segmentation model, the pole piece mask map including a background area and multiple mask areas, each mask area corresponding to the area where the first pole piece or the second pole piece is located; determining the first endpoint and the second endpoint of the first pole piece based on the pole piece mask map, and determining the third endpoint and the fourth endpoint of the second pole piece.

[0077] The segmentation model can compare the grayscale value of each pixel in the detection image with the grayscale value of adjacent pixels based on the grayscale features of adjacent pixels, and classify the pixels into appropriate categories according to the comparison results to achieve the segmentation of various targets on the image.

[0078] In an embodiment of the present application, for the detection of the electrode gap of the battery to be tested, it is necessary to distinguish the areas where the first electrode, the second electrode and the background area are located. Therefore, the use of a segmentation model can better classify each pixel based on different grayscale ranges, output a electrode mask image, separate the electrode and the background, and use multiple mask areas to represent the area where the first electrode or the second electrode is located.

[0079] The pole piece mask image output by the trained segmentation model can output a complete pole piece mask when the pole piece contour or pole piece features in the detection image are discontinuous, thereby restoring the pole piece features to a certain extent and facilitating subsequent measurements.

[0080] According to some embodiments of the present application, the steps of determining the first endpoint and the second endpoint of the first pole piece based on the pole piece mask diagram, and determining the third endpoint and the fourth endpoint of the second pole piece include: multiple mask areas include multiple first mask areas and multiple second mask areas, the first mask area corresponds to the area where the first pole piece is located, and the second mask area corresponds to the area where the second pole piece is located; the first mask area includes a first top edge and a first bottom edge spaced from the first top edge along a second direction, the first endpoint is located at the first top edge, and the second endpoint is located at the first bottom edge, wherein the second direction is the extension direction of the first pole piece; the second mask area includes a second top edge and a second bottom edge spaced from the second top edge along a third direction, the third endpoint is located at the second top edge, and the fourth endpoint is located at the second bottom edge, wherein the third direction is the extension direction of the second pole piece.

[0081] In an embodiment of the present application, since it is necessary to measure the spacing between the first pole piece and the second pole piece in the first direction, it is necessary to further distinguish the first pole piece and the second pole piece, and divide the multiple mask areas into multiple independent pole piece masks, wherein each independent pole piece mask may include a first mask area representing the first pole piece, and also include a second mask area representing the second pole piece.

[0082] For each independent mask area, it includes a top edge and a bottom edge spaced from the top edge in the second direction or the third direction, wherein the top edge can be used as a set of starting points of the corresponding pole piece (in some embodiments, the set is the first endpoint or the third endpoint), and the bottom edge can be used as a set of cutoff points of the corresponding pole piece (in some embodiments, the set is the second endpoint or the fourth endpoint). It can be understood that, on the output mask image, along the second direction, there is no pixel representing the first pole piece above the first top edge, and there is no pixel representing the first pole piece below the first bottom edge. Or along the third direction, there is no pixel representing the second pole piece above the second top edge, and there is no pixel representing the second pole piece below the second bottom edge. Exemplarily, the second direction can be parallel to the third direction.

[0083] Referring to FIG3 , FIG3 is a schematic diagram of a pole piece mask image 300 output by a segmentation model in a battery detection method provided in some embodiments of the present application. The mask image includes a background region 303, a first mask region 301, and a second mask region 302. The first mask region 301 and the second mask region 302 are arranged alternately along a first direction X, and are used to represent a plurality of first pole pieces and a plurality of second pole pieces arranged alternately along the first direction. The first mask region 301 includes a first top edge 3011 and a first bottom edge 3012 spaced apart from the first top edge 3011 along a second direction Y. The first endpoint 3013 is located on the first top edge 3011, and the second endpoint 3014 is located on the first bottom edge 3012. The line connecting the first endpoint 3013 and the second endpoint 3014 is a first pole piece fitting line 3015. The second mask region 302 includes a second top edge 3021 and a second bottom edge 3022 spaced apart from the second top edge 3021 along a third direction (when the second direction is parallel to the third direction, both the third direction and the second direction are in the Y direction). A third endpoint 3023 is located on the second top edge 3021, and a fourth endpoint 3024 is located on the second bottom edge 3022. The line connecting the third endpoint 3023 and the fourth endpoint 3024 is a second electrode sheet fitting line 3025. By measuring the distance d between the first electrode sheet fitting line 3015 and the second electrode sheet fitting line 3025, the electrode sheet gap of the battery under test can be obtained.

[0084] By searching for the top and bottom edges in the independent mask area on the pole piece mask image, the starting point and the end point of the pole piece in the detection image can be better located, so that the obtained pole piece fitting line is closer to the original features of the pole piece in the detection image.

[0085] According to some embodiments of the present application, the first endpoint is the midpoint of the first top side, the second endpoint is the midpoint of the first bottom side; the third endpoint is the midpoint of the second top side, and the fourth endpoint is the midpoint of the second bottom side.

[0086] The first pole piece fitting line 3015 is the center line of the first mask area 301 corresponding to the first pole piece, and the second pole piece fitting line 3025 is the center line of the second mask area 302 corresponding to the second pole piece.

[0087] The pole piece fitting line is obtained based on the center line, so that the obtained pole piece fitting line is closer to the original features of the pole piece in the detection image. At the same time, adjacent pole piece fitting lines are selected based on the same standard, further improving the accuracy of subsequent measurements.

[0088] According to some embodiments of the present application, the first mask area includes a first side extending along a second direction, the first endpoint is the vertex of the first side, and the second endpoint is the end point of the first side; the second mask area includes a second side extending along a third direction, the third endpoint is the vertex of the second side, and the fourth endpoint is the end point of the second side.

[0089] At this time, the first pole piece fitting line is the edge line of one side of the first pole piece, and the second pole piece fitting line is the edge line of one side of the second pole piece. It can be understood that in order to obtain an accurate pole piece gap measurement value, when the first pole piece fitting line is the left edge line of the first pole piece, the second pole piece fitting line should also be the left edge line of the second pole piece, and when the first pole piece fitting line is the right edge line of the first pole piece, the second pole piece fitting line should also be the right edge line of the second pole piece.

[0090] The pole piece fitting line is obtained based on the edge line on one side, so that the obtained pole piece fitting line is closer to the original features of the pole piece in the detection image. At the same time, the adjacent pole piece fitting lines are selected based on the same standard, further improving the accuracy of subsequent measurements.

[0091] The step of determining the electrode gap of the battery to be tested based on the adjacent first electrode piece fitting line and the second electrode piece fitting line includes: along the first direction, based on the adjacent first electrode piece fitting line and the second electrode piece fitting line, obtaining the first distance between the adjacent first electrode piece fitting line and the second electrode piece fitting line, the first distance being used to characterize the electrode gap of the battery to be tested; wherein the first distance is any one of the minimum distance, maximum distance and average distance between the adjacent first electrode piece fitting line and the second electrode piece fitting line.

[0092] In an embodiment of the present application, the first pole piece in the detection image can be characterized by a continuous first pole piece fitting line, and the second pole piece in the detection image can be characterized by a continuous second pole piece fitting line. Since the first pole piece and the second pole piece are arranged alternately along the first direction and have high consistency, the first pole piece fitting line and the second pole piece fitting line should also be fitted based on the same standard, so that the obtained first pole piece fitting line and the second pole piece fitting line are more in line with the corresponding pole piece features on the basis of conforming to the arrangement rules of the first pole piece and the second pole piece, so that the pole piece gap of the battery to be tested can be characterized by the first distance by measuring the first distance between the adjacent first pole piece fitting line and the second pole piece fitting line. Wherein, the adjacent first pole piece fitting line and the second pole piece fitting line mean that there is no other pole piece fitting line between the two pole piece fitting lines.

[0093] Since the first pole piece and the second pole piece are arranged alternately along the first direction, the pole piece gap of the battery to be tested is the misalignment of the positive and negative pole pieces in the first direction, that is, the misalignment (first distance) of the first pole piece fitting line and the second pole piece fitting line in the first direction.

[0094] The first pole piece fitting line may extend along a second direction, wherein the second direction intersects the first direction, and the second pole piece fitting line may extend along a third direction, wherein the first direction intersects the third direction. Therefore, along the first direction, the first pole piece fitting line and the second pole piece fitting line include a minimum spacing, a maximum spacing, and an average spacing obtained by measuring the spacing multiple times, wherein the first distance may be any one of the minimum spacing, the maximum spacing, and the average spacing obtained by measuring the spacing multiple times. It is understandable that, along the first direction, the detection image may include multiple first distances, and the multiple first distances are used to characterize the spacing between different first pole piece fitting lines and second pole piece fitting lines, and the multiple first distances should be calculated based on a unified standard. That is, when the minimum spacing is used as the first spacing to characterize the pole piece gap of the battery to be tested, each first distance in the detection image is the minimum distance between adjacent first pole piece fitting lines and second pole piece fitting lines; when the average spacing is used as the first spacing to characterize the pole piece gap of the battery to be tested, each first distance in the detection image is the average distance between adjacent first pole piece fitting lines and second pole piece fitting lines.

[0095] The electrode gap of the battery to be tested is represented based on the distance between the first electrode fitting line and the second fitting line. When the outline or electrode feature of the electrode in the detection image is discontinuous, more accurate electrode gap detection data can also be obtained through the continuous electrode fitting line.

[0096] According to some embodiments of the present application, before determining a first pole piece fitting line for characterizing the first pole piece and a second pole piece fitting line for characterizing the second pole piece based on the detection image, the method further includes: performing image preprocessing on the detection image, and obtaining multiple first pole piece fitting lines and multiple second pole piece fitting lines based on the detection image after image preprocessing.

[0097] In an embodiment of the present application, image preprocessing may include an image enhancement step and an image denoising step, wherein the image enhancement step may include increasing the contrast of the image to make details in the image more easily recognizable. As a feasible implementation, local histogram processing may be performed on the detection image to enhance the contrast in the image.

[0098] The image denoising step may include at least one of mean filtering denoising, Gaussian filtering denoising, and three-dimensional block matching denoising.

[0099] Denoising multiple initial images can remove random noise, such as quantum noise, aliasing noise, and electronic noise, significantly improving the signal-to-noise ratio of the initial images. Enhancing image contrast can avoid amplifying any noise that may exist in the image, making the object being detected more distinct compared to other areas of the image.

[0100] Therefore, by preprocessing the detection image, the interference of noise in the detection image can be reduced, making the object to be detected in the image more obvious than other areas, thereby making the electrode features in the image easier to identify.

[0101] According to some embodiments of the present application, the image preprocessing step includes: performing local enhancement processing on the detection image.

[0102] The image enhancement step may include increasing the contrast of the image so that details in the image are more easily recognized. As a feasible implementation, local histogram processing may be performed on the detection image to enhance the contrast in the image.

[0103] By processing the local histogram, the contrast in the image can be enhanced while avoiding amplifying the noise that may exist in the image, making the object to be detected in the image more obvious and easier to identify than other areas.

[0104] According to some embodiments of the present application, the step of local enhancement processing includes: extracting a gradient image of the detection image; performing gradient compression on the gradient image to obtain a corresponding compressed gradient image; and performing image reconstruction based on the compressed gradient image to obtain a reconstructed detection image.

[0105] The pixels in the detection image are arranged in an array. For a pixel point in the detection image, there are at least two adjacent pixels in two directions. Therefore, when extracting the gradient image, it is necessary to extract the first gradient image of the detection image along the first direction and the second gradient image of the detection image along the second direction.

[0106] For the first gradient image, the gradient value of a pixel therein can be understood as the difference between the grayscale value of the corresponding pixel on the detection image and the grayscale value of the previous pixel on the detection image, where the corresponding pixel on the detection image and the previous pixel on the detection image are adjacently arranged along a first direction, and the previous pixel on the detection image is located in front of the corresponding pixel on the detection image along the first direction. The magnitude of the gradient value indicates the magnitude of the grayscale value change between the detection pixel and the adjacent pixel in the detection image. A larger gradient value indicates a greater grayscale value change between the detection pixel and the adjacent pixel in the detection image.

[0107] The compression processing compresses the gradient values ​​so that the gradient values ​​in the first compressed gradient image are smaller than those in the first gradient image, and the gradient values ​​in the second compressed gradient image are smaller than those in the second gradient image. Image reconstruction is then performed based on the first compressed gradient image and the second compressed gradient image to obtain a reconstructed image of the object to be detected. Compared to the detected image, the grayscale value variation in the reconstructed image is reduced.

[0108] By extracting a gradient image (in some embodiments, the gradient image may include a first gradient image and a second gradient image) and compressing the gradient image to reduce the gradient value in the resulting compressed gradient image, the variation in grayscale values ​​in the reconstructed image obtained by reconstructing the compressed gradient image is also reduced. For the reconstructed image, even if the grayscale values ​​between adjacent pixels are not significantly different, they can be more easily identified from the reconstructed image, making the detailed features in the reconstructed image clearer and improving the accuracy of the image recognition results. Applying this method to the processing of battery test images can improve the accuracy of battery test results.

[0109] According to some embodiments of the present application, the step of gradient compression includes: determining a contour image of an object to be detected based on a gradient image; determining a compression value of a corresponding pixel based on the gradient amplitude of each pixel in the contour image; extracting the gradient value of each pixel in the gradient image; obtaining a compressed gradient value of the corresponding pixel based on the gradient value of each pixel and the compression value of the corresponding pixel; and obtaining a corresponding compressed gradient image based on the compressed gradient value of each pixel point.

[0110] The contour image of the object to be detected, determined based on the gradient image, has a gradient magnitude at each pixel in the contour image that includes information from both the first and second gradient images. Therefore, a compression value is determined based on the gradient magnitude of each pixel in the contour image. This can be used to compress both the first and second gradient images. The compression value indicates the degree of gradient value compression.

[0111] The magnitude of the gradient amplitude indicates the direction and rate of change of the gradient value of the contour image at that pixel point. The larger the gradient amplitude, the more dramatic the change of the contour at that pixel point, that is, the steeper the gradient of the contour image at that point, and the faster the change.

[0112] Each pixel in the contour image has a gradient magnitude. The compression value of each pixel is then determined based on the gradient magnitude. This allows compression of each pixel during subsequent compression processing. Furthermore, since the compression values ​​for each pixel are not exactly the same, different pixels can be compressed differently. Pixels with larger gradient values ​​are compressed more, while pixels with smaller gradient values ​​are compressed less.

[0113] Since the gradient value in the compressed gradient image is reduced, the difference in grayscale values ​​between pixels in the subsequent reconstructed image can be reduced, making it easier to identify detailed features in the reconstructed image.

[0114] Refer to Figure 4, which is a structural block diagram of a battery detection device provided in some embodiments of the present application. The battery detection device 400 includes: an acquisition module 401, configured to acquire a detection image containing a battery to be tested, the detection image including a plurality of first pole pieces and a plurality of second pole pieces of the battery to be tested, which are alternately arranged along a first direction, wherein the polarity of the first pole piece and the second pole piece are different. A fitting module 402, configured to determine a first pole piece fitting line for characterizing the first pole piece and a second pole piece fitting line for characterizing the second pole piece based on the detection image, wherein the first pole piece fitting line and the second pole piece fitting line are spaced apart along the first direction. A processing module 403, configured to determine the pole piece gap of the battery to be tested based on adjacent first pole piece fitting lines and second pole piece fitting lines.

[0115] In an embodiment of the present application, the object to be inspected may be a battery, a battery cell, or any other object that can be inspected by a radiographic image acquisition device.

[0116] In an embodiment of the present application, the detection image is an image acquired by an image acquisition device. Exemplarily, the detection image may be a processed image.

[0117] Exemplarily, the image acquisition device may be a radiographic image acquisition device, a CT scanning device, or a PET scanning device, and the detection image may be a detection image acquired by the radiographic image acquisition device. Exemplarily, the radiographic image acquisition device may include a radiation source and a flat-panel detector, wherein the detection image may be acquired by the radiation source and the flat-panel detector, the radiation emitted by the radiation source is projected onto the flat-panel detector through the battery to be tested, and the flat-panel detector acquires the detection image based on the received radiation. The radiation emitted by the radiation source may be X-rays, gamma rays, beta rays, etc. In an embodiment of the present application, the first electrode may be a cathode electrode, and the second electrode may be an anode electrode. Of course, it is understandable that in some cases, the first electrode may also be an anode electrode, and the second electrode may be a cathode electrode.

[0118] In an embodiment of the present application, the first pole piece in the detection image can be characterized by a continuous first pole piece fitting line, and the second pole piece in the detection image can be characterized by a continuous second pole piece fitting line. Since the first pole piece and the second pole piece are arranged alternately along the first direction and have high consistency, the first pole piece fitting line and the second pole piece fitting line should also be fitted based on the same standard, so that the obtained first pole piece fitting line and the second pole piece fitting line are more in line with the corresponding pole piece characteristics on the basis of conforming to the arrangement rules of the first pole piece and the second pole piece, so as to facilitate obtaining the pole piece gap of the battery to be tested by measuring the relevant detection data of the adjacent first pole piece fitting lines and the second pole piece fitting lines.

[0119] Since the first and second pole pieces are arranged alternately along the first direction, the pole piece gap of the battery under test is the misalignment of the cathode and anode pole pieces in the first direction. The first pole piece fitting line can extend along a second direction, where the second direction intersects the first direction, and the second pole piece fitting line can extend along a third direction, where the first direction intersects the third direction. Exemplarily, the second direction can be parallel to the third direction. Exemplarily, the first direction can be perpendicular to the second direction, or the first direction can be perpendicular to the third direction.

[0120] The pole pieces in the detection image are converted into pole piece fitting lines. When the outline or pole piece features in the detection image are discontinuous, more accurate pole piece gap detection data can be obtained through the continuous pole piece fitting lines.

[0121] Referring to FIG5 , FIG5 is a flowchart of a segmentation model training method for battery detection provided in some embodiments of the present application. The segmentation model training method 500 for battery detection includes:

[0122] S501 : Acquire a sample image containing a sample battery, where the sample image includes a plurality of first sample pole pieces and a plurality of second sample pole pieces of the sample battery alternately arranged along a first direction.

[0123] S502 , obtaining a real pole piece mask image corresponding to the sample image, where the real pole piece mask image includes a real background area and multiple real mask areas, each real mask area corresponding to an area where the first sample pole piece or the second sample pole piece is located.

[0124] S503: Input the sample image into the segmentation model to be trained, and obtain the predicted pole piece mask image output by the segmentation model.

[0125] S504: Calculate the loss value based on the actual pole piece mask diagram and the predicted pole piece mask diagram.

[0126] S505: Adjust the parameters of the segmentation model based on the loss value.

[0127] In an embodiment of the present application, a plurality of sample images containing sample batteries may be obtained as a dataset for training a segmentation model. The sample images are images acquired by a radiographic image acquisition device. For example, the plurality of sample images may be images that have undergone simple processing.

[0128] The pole piece area in the sample image is marked to obtain the true pole piece mask corresponding to the sample image. The true pole piece mask is used as the ground truth of the segmentation model for supervised learning.

[0129] After the segmentation model outputs a predicted pole mask, the actual pole mask is compared with the predicted one, and the cross-entropy loss function is used to calculate the loss value. If the predicted pole mask differs significantly from the actual one, the loss value will also be large, requiring model parameter adjustment and iteration. If the predicted pole mask is highly consistent with the actual one, the loss value converges and approaches the optimal value, or an iteration is completed when a local optimal value is found. At this point, backpropagation can be used to update the internal parameters of the segmentation model to complete training.

[0130] The pole piece mask output by the trained segmentation model can output the complete pole piece mask area when the pole piece outline or pole piece features in the detection image are discontinuous. Using the actual mask area as the ground truth for the trained segmentation model can improve the accuracy of the segmentation model's predictions and restore the pole piece features to a certain extent, facilitating subsequent measurements.

[0131] According to some embodiments of the present application, the step of obtaining a real pole piece mask image corresponding to the sample image includes:

[0132] Based on the sample image, mark the first and second marked points corresponding to the first sample pole piece, and mark the third and fourth marked points corresponding to the second sample pole piece; based on the marked first and second marked points of the first sample pole piece, and the preset first expansion value, obtain the real mask area corresponding to the first pole piece; based on the marked third and fourth marked points of the second sample pole piece, and the preset second expansion value, obtain the real mask area corresponding to the second pole piece.

[0133] In an embodiment of the present application, based on the sample image, the steps of marking the first and second marking points corresponding to the first sample pole piece, and marking the third and fourth marking points corresponding to the second sample pole piece can be completed by manual marking.

[0134] The preset first expansion value can be obtained through the empirical value of the first sample electrode. After obtaining the first and second annotation points, the first and second annotation points are connected to form a first annotation line corresponding to the first sample electrode. The first annotation line is expanded along the first direction. Based on the preset first expansion value, the first annotation line is expanded into a first annotation area (i.e., the actual mask area corresponding to the first electrode). In this case, the first annotation area can be a parallelogram, with the long side of the parallelogram being equal to the length of the first annotation line and the short side of the parallelogram being equal to the width of the first expansion value. In some embodiments, the first annotation area can be a rectangle.

[0135] The preset second extension value can be obtained through the empirical value of the second sample electrode. After obtaining the third and fourth annotation points, the third and fourth annotation points are connected to form a second annotation line corresponding to the second sample electrode. The second annotation line is extended along the first direction. Based on the preset second extension value, the second annotation line is expanded into a second annotation area (i.e., the actual mask area corresponding to the second electrode). In this case, the second annotation area can be a parallelogram, with the long side of the parallelogram being equal to the length of the second annotation line and the short side of the parallelogram being equal to the width of the second extension value. In some embodiments, the second annotation area can be a rectangle.

[0136] Marking a sample electrode with only two annotation points can reduce the difficulty and workload of annotation and improve annotation efficiency. At the same time, when the electrode contour or electrode features in the sample image are discontinuous, the complete true mask area can be output as the true value for model training, allowing the trained segmentation model to predict the complete mask area based on the discontinuous electrode contour.

[0137] Referring to Figure 6, Figure 6 is a structural block diagram of a segmentation model training device for battery detection provided in some embodiments of the present application. The segmentation model training device 600 for battery detection includes: a sample acquisition module 601, configured to acquire a sample image containing a sample battery, the sample image including a plurality of first sample pole pieces and a plurality of second sample pole pieces of the sample battery arranged alternately along a first direction. A truth module 602, configured to acquire a true pole piece mask image corresponding to the sample image, the true pole piece mask image including a true background area and a plurality of true mask areas, each true mask area corresponding to an area where the first sample pole piece or the second sample pole piece is located; a sample prediction module 603, configured to input the sample image into the segmentation model to be trained, and acquire a predicted pole piece mask image output by the segmentation model; a loss module 604, configured to calculate a loss value based on the true pole piece mask image and the predicted pole piece mask image; a parameter adjustment module 605, configured to adjust the parameters of the segmentation model based on the loss value.

[0138] In an embodiment of the present application, a plurality of sample images containing sample batteries may be obtained as a dataset for training a segmentation model. The sample images are images acquired by a radiographic image acquisition device. For example, the plurality of sample images may be images that have undergone simple processing.

[0139] The pole piece area in the sample image is marked to obtain the true pole piece mask corresponding to the sample image. The true pole piece mask is used as the ground truth of the segmentation model for supervised learning.

[0140] After the segmentation model outputs a predicted pole mask, the actual pole mask is compared with the predicted one, and the cross-entropy loss function is used to calculate the loss value. If the predicted pole mask differs significantly from the actual one, the loss value will also be large, requiring model parameter adjustment and iteration. If the predicted pole mask is highly consistent with the actual one, the loss value converges and approaches the optimal value, or an iteration is completed when a local optimal value is found. At this point, backpropagation can be used to update the internal parameters of the segmentation model to complete training.

[0141] The pole piece mask output by the trained segmentation model can output the complete pole piece mask area when the pole piece outline or pole piece features in the detection image are discontinuous. Using the actual mask area as the ground truth for the trained segmentation model can improve the accuracy of the segmentation model's predictions and restore the pole piece features to a certain extent, facilitating subsequent measurements.

[0142] According to some embodiments of the present application, an electronic device is provided, comprising at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the battery detection method in the above-mentioned embodiment or the segmentation model training method in the above-mentioned embodiment.

[0143] According to some embodiments of the present application, a battery production system is provided, comprising the electronic device in the above embodiments.

[0144] According to some embodiments of the present application, a computer-readable storage medium is provided, storing a computer program, which, when executed by a processor, implements the battery detection method in the above embodiment or the segmentation model training method in the above embodiment.

[0145] According to some embodiments of the present application, a computer program product is provided, including a computer program, which, when executed by a processor, implements the battery detection method in the above embodiment or the segmentation model training method in the above embodiment.

[0146] Illustrative examples of such a computer device, non-transitory computer-readable storage medium, and computer program product are described below in conjunction with FIG. 7 .

[0147] Computer device 700 can be a variety of different types of devices. Examples of computer device 700 include, but are not limited to, desktop computers, server computers, laptop or netbook computers, mobile devices (e.g., tablet computers, cellular or other wireless phones (e.g., smartphones), notepad computers, mobile stations), wearable devices (e.g., eyeglasses, watches), entertainment devices (e.g., entertainment appliances, set-top boxes communicatively coupled to a display device, game consoles), televisions or other display devices, automotive computers, and the like.

[0148] The computer device 700 may include at least one processor 702, memory 704, communication interface(s) 706, a display device 708, other input / output (I / O) devices 710, and one or more mass storage devices 712, all capable of communicating with one another, such as via a system bus 714 or other appropriate connection.

[0149] The processor 702 may be a single processing unit or multiple processing units, all of which may include a single or multiple computing units or multiple cores. The processor 702 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuits, and / or any device that manipulates signals based on operational instructions. Among other capabilities, the processor 702 may be configured to retrieve and execute computer-readable instructions stored in the memory 704, mass storage device 712, or other computer-readable media, such as program code for an operating system 716, program code for application programs 718, program code for other programs 720, and the like.

[0150] The memory 704 and the mass storage device 712 are examples of computer-readable storage media for storing instructions that are executed by the processor 702 to implement the various functions described above. For example, the memory 704 may generally include both volatile memory and non-volatile memory (e.g., RAM, ROM, etc.). In addition, the mass storage device 712 may generally include a hard drive, a solid-state drive, removable media, including external and removable drives, memory cards, flash memory, floppy disks, optical disks (e.g., CDs, DVDs), storage arrays, network attached storage, storage area networks, etc. The memory 704 and the mass storage device 712 may all be collectively referred to herein as memory or computer-readable storage media, and may be non-transitory media capable of storing computer-readable, processor-executable program instructions as computer program code, which may be executed by the processor 702 as a specific machine configured to implement the operations and functions described in the examples herein.

[0151] A number of programs may be stored on the mass storage device 712. These programs include an operating system 716, one or more application programs 718, other programs 720, and program data 722, and they may be loaded into the memory 704 for execution. Examples of such applications or program modules may include, for example, computer program logic (e.g., computer program code or instructions) for implementing the client application, method 200, method 400, and / or other embodiments described herein.

[0152] 7 as being stored in memory 704 of computer device 700, modules 716, 718, 720, and 722, or portions thereof, may be implemented using any form of computer-readable media accessible by computer device 700. As used herein, "computer-readable media" includes at least two types of computer-readable media, namely, computer-readable storage media and communication media.

[0153] Computer-readable storage media include volatile and non-volatile, removable and non-removable media implemented by any method or technology for storing information, such as computer-readable instructions, data structures, program modules or other data. Computer-readable storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD), or other optical storage devices, magnetic cassettes, magnetic tape, magnetic disk storage devices or other magnetic storage devices, or any other non-transmission media that can be used to store information for access by a computer device. In contrast, communication media can embody computer-readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transmission mechanism. Computer-readable storage media as defined herein does not include communication media.

[0154] One or more communication interfaces 706 are used to exchange data with other devices, such as via a network, direct connection, and the like. Such communication interfaces can be one or more of the following: any type of network interface (e.g., a network interface card (NIC)), a wired or wireless (such as an IEEE 802.11 wireless LAN (WLAN)) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth™ interface, a Near Field Communication (NFC) interface, and the like. The communication interface 706 can facilitate communication within a variety of network and protocol types, including wired networks (e.g., LAN, cable, and the like) and wireless networks (e.g., WLAN, cellular, satellite, and the like), the Internet, and the like. The communication interface 706 can also provide communication with external storage devices (not shown), such as storage arrays, network attached storage, storage area networks, and the like.

[0155] In some examples, a display device 708 such as a monitor may be included for displaying information and images to the user. Other I / O devices 710 may be devices that receive various inputs from the user and provide various outputs to the user, and may include a touch input device, a gesture input device, a camera, a keyboard, a remote control, a mouse, a printer, an audio input / output device, and the like.

[0156] The technology described herein can be supported by these various configurations of the computer device 700 and is not limited to the specific examples of the technology described herein. For example, the functionality can also be implemented in whole or in part on a "cloud" by using a distributed system. The cloud includes and / or represents a platform for resources. The platform abstracts the underlying functionality of the hardware (e.g., servers) and software resources of the cloud. Resources can include applications and / or data that can be used when performing computing processing on a server away from the computer device 700. Resources can also include services provided over the Internet and / or through a subscriber network such as a cellular or Wi-Fi network. The platform can abstract resources and functions to connect the computer device 700 to other computer devices. Therefore, the implementation of the functions described herein can be distributed throughout the cloud. For example, functions can be implemented partially on the computer device 700 and partially through a platform that abstracts the functions of the cloud.

[0157] An embodiment of the present application provides a battery detection method, which inputs a detection image into a segmentation model to obtain a pole piece mask image output by the segmentation model, wherein the pole piece mask image includes a background area and multiple mask areas, each mask area corresponding to the area where the first pole piece or the second pole piece is located; based on the pole piece mask image, the first endpoint and the second endpoint of the first pole piece are determined, and the third endpoint and the fourth endpoint of the second pole piece are determined.

[0158] The multiple mask regions include multiple first mask regions and multiple second mask regions, wherein the first mask region corresponds to the region where the first pole piece is located, and the second mask region corresponds to the region where the second pole piece is located; the first mask region includes a first top edge and a first bottom edge spaced apart from the first top edge along a second direction, wherein the first endpoint is located at the first top edge, and the second endpoint is located at the first bottom edge; the second mask region includes a second top edge and a second bottom edge spaced apart from the second top edge along a second direction, wherein the third endpoint is located at the second top edge, and the fourth endpoint is located at the second bottom edge. The first endpoint is the midpoint of the first top edge, the second endpoint is the midpoint of the first bottom edge, the third endpoint is the midpoint of the second top edge, and the fourth endpoint is the midpoint of the second bottom edge.

[0159] Referring to Figure 3, the pole piece mask diagram 300 output by the segmentation model includes a background area 303, a first mask area 301 and a second mask area 302, wherein the first mask area 301 and the second mask area 302 are arranged alternately along the first direction X, and are used to characterize multiple first pole pieces and multiple second pole pieces arranged alternately along the first direction, the first mask area 301 includes a first top edge 3011 and a first bottom edge 3012 spaced from the first top edge 3011 along the second direction Y, wherein the first endpoint 3013 is located on the first top edge 3011, and the second endpoint 3014 is located on the first bottom edge 3012, and the line connecting the first endpoint 3013 and the second endpoint 3014 is the first pole piece fitting line 3015. The second mask region 302 includes a second top edge 3021 and a second bottom edge 3022 spaced apart from the second top edge 3021 along a second direction Y. A third endpoint 3023 is located on the second top edge 3021, and a fourth endpoint 3024 is located on the second bottom edge 3022. The line connecting the third endpoint 3023 and the fourth endpoint 3024 is a second electrode sheet fitting line 3025. By measuring the distance d between the first electrode sheet fitting line 3015 and the second electrode sheet fitting line 3025, the electrode sheet gap of the battery under test can be obtained.

[0160] 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 battery detection method, the method comprising: Acquire a detection image including a battery to be tested, wherein the detection image includes a plurality of first pole pieces and a plurality of second pole pieces of the battery to be tested that are alternately arranged along a first direction, wherein the first pole pieces and the second pole pieces have different polarities; Determine, based on the detection image, a first pole piece fitting line for characterizing the first pole piece and a second pole piece fitting line for characterizing the second pole piece, wherein the first pole piece fitting line and the second pole piece fitting line are spaced apart along a first direction; Based on the adjacent first pole piece fitting line and the second pole piece fitting line, the pole piece gap of the battery to be tested is determined.

2. The method according to claim 1, wherein: The step of determining a first pole piece fitting line for characterizing the first pole piece and a second pole piece fitting line for characterizing the second pole piece based on the detection image comprises: Obtaining a first endpoint and a second endpoint of the first pole piece based on the detection image, and obtaining a third endpoint and a fourth endpoint of the second pole piece; A first pole piece fitting line corresponding to the first pole piece is obtained based on a line connecting the first endpoint and the second endpoint, and a second pole piece fitting line corresponding to the second pole piece is obtained based on a line connecting the third endpoint and the fourth endpoint.

3. The method according to claim 1 or 2, wherein: The battery to be tested includes a wound battery cell; The detection image is obtained by projecting rays onto the surface of the wound battery cell for shooting, and the detection image includes any one of the four corners of the surface of the wound battery cell.

4. The method according to claim 2, wherein: The steps of obtaining the first endpoint and the second endpoint of the first pole piece based on the detection image, and obtaining the third endpoint and the fourth endpoint of the second pole piece include: Input the detection image into a segmentation model to obtain a pole piece mask image output by the segmentation model, wherein the pole piece mask image includes a background area and a plurality of mask areas, each of the mask areas corresponding to an area where the first pole piece or the second pole piece is located; A first endpoint and a second endpoint of the first pole piece are determined based on the pole piece mask image, and a third endpoint and a fourth endpoint of the second pole piece are determined.

5. The method according to claim 4, wherein: The steps of determining the first endpoint and the second endpoint of the first pole piece based on the pole piece mask image, and determining the third endpoint and the fourth endpoint of the second pole piece include: The plurality of mask regions include a plurality of first mask regions and a plurality of second mask regions, the first mask regions correspond to the regions where the first pole piece is located, and the second mask regions correspond to the regions where the second pole piece is located; The first mask region includes a first top edge and a first bottom edge spaced from the first top edge along a second direction, the first endpoint is located at the first top edge, and the second endpoint is located at the first bottom edge, wherein the second direction is an extension direction of the first pole piece; The second mask area includes a second top edge and a second bottom edge spaced apart from the second top edge along the third direction, the third endpoint is located at the second top edge, and the fourth endpoint is located at the second bottom edge, wherein the third direction is the extension direction of the second pole piece.

6. The method according to claim 5, wherein: The first endpoint is the midpoint of the first top side, and the second endpoint is the midpoint of the first bottom side; The third endpoint is the midpoint of the second top side, and the fourth endpoint is the midpoint of the second bottom side.

7. The method according to claim 5, wherein: The first mask region includes a first side extending along the second direction, the first endpoint is a vertex of the first side, and the second endpoint is an end point of the first side; The second mask region includes a second side extending along the third direction, the third endpoint is a vertex of the second side, and the fourth endpoint is an end point of the second side.

8. The method according to any one of claims 1 to 7, wherein: The step of determining the electrode sheet gap of the battery to be tested based on the adjacent first electrode sheet fitting line and the second electrode sheet fitting line includes: Along the first direction, based on the adjacent first pole piece fitting line and the second pole piece fitting line, a first distance between the adjacent first pole piece fitting line and the second pole piece fitting line is obtained, and the first distance is used to characterize the pole piece gap of the battery to be tested; wherein, The first distance is any one of a minimum distance, a maximum distance and an average distance between the adjacent first pole piece fitting lines and the second pole piece fitting lines.

9. The method according to any one of claims 1 to 8, wherein: Before determining a first pole piece fitting line for characterizing the first pole piece and a second pole piece fitting line for characterizing the second pole piece based on the detection image, the method further includes: The detection image is subjected to image preprocessing, and a plurality of first pole piece fitting lines and a plurality of second pole piece fitting lines are obtained based on the detection image after image preprocessing.

10. The method according to claim 9, wherein: The image preprocessing step comprises: Performing local enhancement processing on the detection image.

11. The method according to claim 10, wherein: The local enhancement processing step comprises: Extracting a gradient image of the detection image; Performing gradient compression on the gradient image to obtain a corresponding compressed gradient image; Image reconstruction is performed based on the compressed gradient image to obtain a reconstructed detection image.

12. The method according to claim 11, wherein: The step of gradient compression comprises: Determine a contour image of the object to be detected based on the gradient image; Determining a compression value of a corresponding pixel based on a gradient magnitude of each pixel in the contour image; Extracting the gradient value of each pixel in the gradient image; Obtaining a compressed gradient value of the corresponding pixel based on the gradient value of each pixel and the compression value of the corresponding pixel; A corresponding compressed gradient image is obtained based on the compressed gradient value of each pixel.

13. A battery detection device, comprising: an acquisition module configured to acquire a detection image containing a battery to be tested, wherein the detection image includes a plurality of first pole pieces and a plurality of second pole pieces of the battery to be tested that are alternately arranged along a first direction, wherein the first pole pieces and the second pole pieces have different polarities; A fitting module, configured to determine a first pole piece fitting line for characterizing the first pole piece and a second pole piece fitting line for characterizing the second pole piece based on the detection image, wherein the first pole piece fitting line and the second pole piece fitting line are spaced apart along a first direction; The processing module is configured to determine the pole piece gap of the battery to be tested based on the adjacent first pole piece fitting line and the second pole piece fitting line.

14. A segmentation model training method for battery detection, the training method comprising: Acquire a sample image containing a sample battery, wherein the sample image includes a plurality of first sample pole pieces and a plurality of second sample pole pieces of the sample battery alternately arranged along a first direction; Acquire a real pole piece mask image corresponding to the sample image, wherein the real pole piece mask image includes a real background area and a plurality of real mask areas, each of the real mask areas corresponds to an area where the first sample pole piece or the second sample pole piece is located; Inputting the sample image into the segmentation model to be trained, and obtaining a predicted pole piece mask image output by the segmentation model; Calculating a loss value based on the actual pole piece mask diagram and the predicted pole piece mask diagram; Parameters of the segmentation model are adjusted based on the loss value.

15. The training method according to claim 14, wherein: The step of obtaining a real pole piece mask image corresponding to the sample image comprises: Based on the sample image, mark a first marking point and a second marking point corresponding to the first sample pole piece, and mark a third marking point and a fourth marking point corresponding to the second sample pole piece; Based on the marked first and second marked points of the first sample pole piece, and the preset first expansion value, the real mask area corresponding to the first pole piece is obtained; based on the marked third and third marked points of the second sample pole piece, and the preset second expansion value, the real mask area corresponding to the second pole piece is obtained.

16. A segmentation model training device for battery detection, the training device comprising: A sample acquisition module is configured to acquire a sample image containing a sample battery, wherein the sample image includes a plurality of first sample pole pieces and a plurality of second sample pole pieces of the sample battery alternately arranged along a first direction; A truth value module is configured to obtain a real pole piece mask image corresponding to the sample image, wherein the real pole piece mask image includes a real background area and a plurality of real mask areas, each of which corresponds to an area where the first sample pole piece or the second sample pole piece is located; A sample prediction module is configured to input the sample image into a segmentation model to be trained, and obtain a predicted polar mask image output by the segmentation model; A loss module, configured to calculate a loss value based on the actual pole piece mask diagram and the predicted pole piece mask diagram; A parameter adjustment module is configured to adjust the parameters of the segmentation model based on the loss value.

17. An electronic device comprising: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the battery detection method described in any one of claims 1 to 12 or the segmentation model training method described in claim 14 or 15.

18. A battery production system comprising the electronic device according to claim 17.

19. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the battery detection method according to any one of claims 1 to 12 or the segmentation model training method according to claim 14 or 15 is implemented.

20. A computer program product, comprising a computer program, which, when executed by a processor, implements the battery detection method according to any one of claims 1 to 12 or the segmentation model training method according to claim 14 or 15.

Images