Battery positive and negative electrode detection method and device based on X-ray image, equipment and medium

Through the detection method based on X-ray image, combined with the module matching algorithm and the battery boundary point detection model, the positive and negative electrode boundary recognition and correction of lithium batteries is solved, and the problems of low detection efficiency and limited accuracy in the prior art are solved. Achieve high-quality battery production.

CN118941864APending Publication Date: 2024-11-12SHENZHEN BAK POWER BATTERY CO LTD

Patent Information

Application Number
CN202410999125.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-24
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The prior art cannot accurately identify the positive and negative electrode boundaries of lithium batteries and cannot be compatible with the positive and negative electrode alignment detection of various processes, resulting in low detection efficiency and limited accuracy.

Method used

采用基于X-ray图像的检测方法,通过模块匹配算法进行图像处理,利用预先构建的电池边界点检测模型识别初始边界点集,并根据电池的预设信息进行修正,提高边界识别的准确性。

Benefits of technology

It significantly improves the detection accuracy and efficiency of the positive and negative electrode boundaries of lithium batteries, and is suitable for battery production of different processes, ensuring high-quality battery manufacturing.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941864A_ABST
    Figure CN118941864A_ABST
Patent Text Reader

Abstract

The invention provides a battery positive and negative electrode detection method and device based on an X-ray image, equipment and a medium, and the method comprises the steps: obtaining an X-ray image of a battery, carrying out the image processing according to the X-ray image based on a module matching algorithm, and obtaining a processed image; identifying the processed image through a pre-constructed battery boundary point detection model to obtain a corresponding initial positive and negative electrode boundary point set; and correcting the initial positive and negative electrode boundary point set according to preset information of the battery and the processed image to obtain positive and negative electrode boundaries of the battery. The X-ray battery image is recognized and corrected through the battery boundary point detection model, the positive and negative electrode boundaries of the battery are obtained, the accuracy and efficiency of detecting the positive and negative electrodes of the battery are improved, and high-quality production of the lithium battery is facilitated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of battery manufacturing, and specifically, to a method, device, equipment and medium for detecting positive and negative electrodes of a battery based on X-ray images. Background Art

[0002] In the production process of lithium batteries, accurate detection of the alignment of positive and negative electrodes is a key step to ensure battery quality and safety. Traditional manual microscope inspection or mechanical measurement methods are inefficient and limited in accuracy in large-scale industrial production environments, and are difficult to adapt to complex battery boundary detection. Therefore, many production lines have introduced X-ray equipment to perform internal structure detection on lithium batteries. However, the existing X-ray detection technology is affected by the battery structure, battery process and equipment performance, and often presents varying degrees of blur. Especially for batteries after winding, flattening and entering the steel shell, the image clarity is significantly reduced, and the positive and negative electrode boundaries are difficult to identify, which directly affects the accuracy of boundary recognition. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a method, device, equipment and medium for detecting positive and negative electrodes of a battery based on X-ray images, so as to solve the technical problems that the prior art cannot accurately identify the boundaries of the positive and negative electrodes and cannot be compatible with the positive and negative electrode alignment detection of various processes.

[0004] In a first aspect, the present invention provides a method for detecting positive and negative electrodes of a battery based on X-ray images, comprising:

[0005] Acquire an X-ray image of the battery, and perform image processing on the X-ray image based on a module matching algorithm to obtain a processed image;

[0006] Recognize the processed image through a pre-built battery boundary point detection model to obtain a corresponding initial positive and negative electrode boundary point set;

[0007] The initial positive and negative electrode boundary point set is corrected according to the preset information of the battery and the processed image to obtain the positive and negative electrode boundaries of the battery.

[0008] In an optional implementation, constructing the battery boundary point detection model includes:

[0009] Acquire a battery image training set, wherein the battery image training set includes multiple battery positive and negative electrode boundary images;

[0010] An initial battery boundary point detection model is constructed according to the battery image training set based on a preset deep learning model, and the initial battery boundary point detection model is trained to obtain the battery boundary point detection model.

[0011] In an optional embodiment, performing image processing based on the X-ray image based on the module matching algorithm to obtain a processed image includes:

[0012] Performing block matching on the X-ray image by using the module matching algorithm to obtain a plurality of key areas, wherein the key areas include positive and negative electrode boundaries;

[0013] Performing image processing on each of the key areas to obtain a processed image, wherein the image processing includes denoising, contrast adjustment, and edge enhancement.

[0014] In an optional implementation, the initial positive and negative electrode boundary point set is corrected according to the preset information of the battery and the processed image to obtain the positive and negative electrode boundaries of the corresponding battery, including:

[0015] Optimizing the positive and negative electrode boundary point sets according to preset information of the battery to obtain optimized positive and negative electrode boundaries, wherein the preset information includes battery characteristics and battery geometry information;

[0016] The optimized positive and negative electrode boundaries are corrected based on the X-ray image and the processed image to obtain the positive and negative electrode boundaries of the corresponding battery.

[0017] In an optional implementation, optimizing the positive and negative electrode boundary point set according to preset information of the battery to obtain optimized positive and negative electrode boundaries includes:

[0018] The positive and negative electrode boundary point sets are screened by using the battery characteristics and the battery geometry information to obtain screened positive and negative electrode boundary point sets.

[0019] The screened positive and negative boundary point sets are optimized according to morphological operations to obtain the optimized positive and negative boundary points.

[0020] In an optional embodiment, the step of correcting the optimized positive and negative electrode boundaries based on the X-ray image and the processed image to obtain the positive and negative electrode boundaries of the battery includes:

[0021] Fusing the optimized positive and negative electrode boundaries with the X-ray image and the processed image to obtain fused positive and negative electrode boundaries;

[0022] The boundaries of the fused positive and negative electrodes are corrected to obtain the boundaries of the positive and negative electrodes of the battery.

[0023] In a second aspect, the present invention provides a battery positive and negative electrode detection device based on X-ray images, the device comprising:

[0024] An acquisition module, used for acquiring an X-ray image of the battery, and performing image processing on the X-ray image based on a module matching algorithm to obtain a processed image;

[0025] A recognition module, used to recognize the processed image through a deep learning model to obtain a corresponding initial positive and negative boundary point set;

[0026] An output module is used to correct the initial positive and negative electrode boundary point set according to the preset information of the battery and the processed image, obtain the positive and negative electrode boundaries of the corresponding battery, and output the positive and negative electrode boundaries.

[0027] In an optional implementation, the acquisition module includes:

[0028] A matching unit, used for performing block matching on the X-ray image by using the module matching algorithm to obtain a plurality of key areas including positive and negative electrode boundaries;

[0029] A processing unit is used to perform image processing on the key area to obtain a processed image, wherein the image processing includes denoising, contrast adjustment and edge enhancement.

[0030] In a third aspect, the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor is running, the computer program executes the battery positive and negative pole detection method based on X-ray images described in the aforementioned embodiment.

[0031] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, which, when running on a processor, executes the battery positive and negative electrode detection method based on X-ray images as described in the aforementioned embodiment.

[0032] The above technical solution adopted in this application has the following effects:

[0033] The present application provides a method, device, equipment and medium for detecting positive and negative electrodes of a battery based on X-ray images. The method comprises: obtaining an X-ray image of a battery, performing image processing on the X-ray image based on a module matching algorithm to obtain a processed image; recognizing the processed image through a pre-constructed battery boundary point detection model to obtain a corresponding initial positive and negative boundary point set; and correcting the initial positive and negative boundary point set according to preset information of the battery and the processed image to obtain the positive and negative boundaries of the battery. The present application identifies and corrects X-ray battery images through a battery boundary point detection model to obtain the positive and negative boundaries of the battery, thereby improving the accuracy and efficiency of detecting the positive and negative electrodes of the battery and facilitating the high-quality production of lithium batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 A schematic diagram of a process of a battery positive and negative electrode detection method based on X-ray images provided by an embodiment of the present invention;

[0036] Figure 2 It is another flow chart of a method for detecting positive and negative electrodes of a battery based on X-ray images provided by an embodiment of the present invention;

[0037] Figure 3 It is another flow chart of a method for detecting positive and negative electrodes of a battery based on X-ray images provided by an embodiment of the present invention;

[0038] Figure 4 It is another flow chart of a method for detecting positive and negative electrodes of a battery based on X-ray images provided by an embodiment of the present invention;

[0039] Figure 5 It is another flow chart of a method for detecting positive and negative electrodes of a battery based on X-ray images provided by an embodiment of the present invention;

[0040] Figure 6 It is a structural schematic diagram of a battery positive and negative electrode detection device based on X-ray images provided by an embodiment of the present invention;

[0041] Figure 7 It is another structural schematic diagram of a battery positive and negative electrode detection device based on X-ray images provided in an embodiment of the present invention.

[0042] Icon: 600-battery positive and negative pole detection device based on X-ray image; 601-acquisition module; 602-identification module; 603-output module; 6011-matching unit; 6012-processing unit. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

[0044] The components of the embodiments of the present invention generally described and shown in the drawings herein may be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0045] Hereinafter, the terms "including", "having" and their cognates, which may be used in various embodiments of the present invention, are intended only to indicate specific features, numbers, steps, operations, elements, components or combinations of the foregoing items, and should not be understood as first excluding the existence of one or more other features, numbers, steps, operations, elements, components or combinations of the foregoing items or adding the possibility of one or more features, numbers, steps, operations, elements, components or combinations of the foregoing items.

[0046] Furthermore, the terms “first”, “second”, “third”, etc. are merely used for distinguishing descriptions and are not to be understood as indicating or implying relative importance.

[0047] Unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meanings as those generally understood by those skilled in the art to which the various embodiments of the present invention belong. The terms (such as those defined in generally used dictionaries) will be interpreted as having the same meanings as the contextual meanings in the relevant technical field and will not be interpreted as having idealized meanings or overly formal meanings unless clearly defined in the various embodiments of the present invention.

[0048] Example 1

[0049] Since the traditional battery positive and negative electrode detection method is mainly achieved through manual microscopes or mechanical measuring tools, there are many shortcomings, especially when X-ray equipment is introduced to perform internal structure detection on lithium batteries. The collected X-ray images are usually blurred, making it impossible to accurately identify the positive and negative electrodes of the battery and even more incompatible with the positive and negative electrode alignment detection of various processes.

[0050] In response to the above technical problems, an embodiment of the present application provides a battery positive and negative electrode detection method based on X-ray images, which aims to solve the technical problems that the prior art cannot accurately identify the boundaries of the positive and negative electrodes and is not compatible with the positive and negative electrode alignment detection of various processes.

[0051] Reference Figure 1 , Figure 1 It is a flow chart of the battery positive and negative electrode detection method based on X-ray images provided in this application.

[0052] The method for detecting positive and negative electrodes of a battery based on X-ray images provided in an embodiment of the present application includes steps S100 to S300, and each step is described below.

[0053] Step S100: Acquire an X-ray image of the battery, and perform image processing on the X-ray image based on a module matching algorithm to obtain a processed image.

[0054] Specifically, the X-ray image of the battery is preferably obtained by scanning the battery to be inspected with an X-ray device, mainly for lithium batteries. By scanning the battery with an X-ray device, the internal structure and potential defects of the battery can be observed non-destructively to ensure battery performance and safety, thereby obtaining a fuzzy image of the internal structure of the battery.

[0055] In this embodiment, a fuzzy image of the internal structure of the battery to be tested is obtained by an X-ray device, and then a region of interest (ROI) is selected using a module matching algorithm. According to the structural characteristics of the lithium battery, the entire image is matched in blocks to select the key area containing the positive and negative electrode boundaries. Among them, the module matching algorithm includes but is not limited to: a simple correlation matching algorithm, a normalized cross-correlation algorithm, a square difference sum algorithm, a pyramid matching algorithm, and a feature matching algorithm. Finally, a variety of image enhancement algorithms are used to process the selected ROI area, including but not limited to using filtering to denoise the image and remove random noise in the image; using methods such as histogram equalization and logarithmic transformation to enhance the contrast of important structures in the image; using methods such as Laplace transform and Sobel operator to further highlight the edge information in the image, thereby obtaining a processed image, significantly improving the quality and readability of the image, and making subsequent image analysis and processing more accurate and effective.

[0056] Reference Figure 2 Specifically, in one implementation of this embodiment, step S100 includes:

[0057] Step S101: performing block matching on the X-ray image by using the module matching algorithm to obtain a plurality of key areas, wherein the key areas include positive and negative electrode boundaries.

[0058] Specifically, the module matching algorithm includes but is not limited to: simple correlation matching algorithm, normalized cross-correlation algorithm, square difference sum algorithm, pyramid matching algorithm and feature matching algorithm. Block matching refers to dividing the image into multiple smaller blocks or areas, and then applying the module matching algorithm to the divided blocks or areas respectively to identify and locate specific features or structures in the image.

[0059] In this embodiment, first, one or more template images are created based on the typical X-ray responses of the known positive and negative boundaries of the battery or a previously manually marked example image, and the X-ray image is matched block by block using a module matching algorithm to identify and locate the positive and negative boundaries in the X-ray image, thereby accurately identifying key areas such as the positive and negative boundaries from complex X-ray images.

[0060] Step S102: performing image processing on each of the key areas to obtain a processed image, wherein the image processing includes denoising, contrast adjustment and edge enhancement.

[0061] Specifically, image processing includes but is not limited to image denoising, histogram equalization and logarithmic transformation, as well as Laplace transform and Sobel operator and other methods.

[0062] In this embodiment, an image enhancement algorithm is applied to the selected ROI area, including denoising, contrast adjustment and edge enhancement, to improve the clarity of the image. Specifically, the image is denoised by filtering, including but not limited to replacing the central pixel by calculating the average pixel value in the neighborhood, reducing salt and pepper noise, or / and replacing the central pixel with the median of the pixel value in the neighborhood, removing salt and pepper noise, or / and convolving the image with a Gaussian kernel, smoothing the image, removing Gaussian noise, thereby removing random noise in the image, and then using histogram equalization to adjust the grayscale distribution of the image so that the probability of each grayscale in the image is roughly equal, thereby enhancing the overall contrast. Alternatively, a logarithmic transformation can be used to apply a logarithmic function to the pixel values ​​to expand the dark areas of the image, improve the contrast of low grayscale areas, and enhance the contrast of specific features or structures in the image. Finally, edge enhancement methods such as Laplace transform and Sobel operator can be used to further highlight the edge information in the image, thereby significantly improving the clarity and analyzability of the image, especially for complex or low-contrast images such as X-ray images, to ensure that the enhanced image is clear and not distorted, so as to achieve the best image enhancement effect.

[0063] In one implementation of this embodiment, the lithium battery is scanned by an X-ray device, the voltage is set to 55kV, and the current is set to 7mA to obtain a fuzzy image of the internal structure of the lithium battery, and then the module matching algorithm is used to block each image and select the key area containing the positive and negative pole boundaries. Among them, the template size is set to 960x160 pixels. Then, a Gaussian filter (kernel size is 3x3) is applied to the selected ROI area to remove noise, histogram equalization is used to enhance the image contrast, and Laplace transform is used to highlight the edge information in the image. The experimental results show that after image processing of the X-ray image, the denoising, contrast and edge clarity are significantly improved, the positive and negative pole boundaries become more obvious and recognizable, and the detection accuracy is improved by about 30%.

[0064] Step S200: recognizing the processed image through a pre-built battery boundary point detection model to obtain a corresponding initial positive and negative electrode boundary point set.

[0065] Specifically, constructing a battery boundary point detection model includes: obtaining a battery image training set, wherein the battery image training set includes a variety of battery positive and negative electrode boundary images, specifically refers to collecting a large number of annotated image data of the positive and negative electrode boundaries of lithium batteries, including battery images under various processes and forms, for training deep learning models to improve the generalization ability of the model. Based on a preset deep learning model, an initial battery boundary point detection model is constructed according to the battery image training set, and the initial battery boundary point detection model is trained to obtain a battery boundary point detection model, wherein the preset deep learning model is a pre-selected deep learning architecture suitable for image classification or object detection, such as a convolutional neural network (CNN), U-Net (for semantic segmentation), MaskR-CNN (for instance segmentation), etc. Training the initial battery boundary point detection model includes but is not limited to using deep learning technology to train a boundary recognition model, and then optimizing the model performance through methods such as hyperparameter tuning, learning rate adjustment, and data enhancement.

[0066] In this embodiment, the processed image is input into a trained battery boundary point detection model to identify the positive and negative pole boundaries. The battery boundary point detection model outputs a series of boundary points or boundary line segments, representing the detected positive and negative pole boundaries, thereby improving the efficiency and accuracy of detection.

[0067] Step S300: correcting the initial positive and negative electrode boundary point set according to the preset information of the battery and the processed image to obtain the positive and negative electrode boundaries of the battery.

[0068] Specifically, the preset information includes battery characteristics and battery geometry information. For example, battery characteristics include the type of battery chemical composition, the working principle of the battery, the battery performance parameters, and the positive and negative electrode boundaries of the battery are regularly fluctuating, and the grayscale values ​​of adjacent areas have similarities and other prior knowledge. Battery geometry information includes the physical size (length, width, height) and shape (cylindrical, square, soft pack, etc.) of the battery, the internal structure of the battery, the packaging method of the battery, the battery pack layout, and the Y coordinate values ​​of adjacent points will not deviate too much, the distance has a threshold, and the slope will not be 90° and other geometric rules. Correction specifically refers to correcting and improving the identified positive and negative electrode boundaries according to geometric constraints and actual application requirements.

[0069] In this embodiment, the battery characteristics and battery geometry information of the battery to be detected are used to further process the recognition results, including removing abnormal points or noise that do not conform to the actual product characteristics through threshold screening and morphological operations. Specifically, the identified boundaries are corrected and improved according to geometric constraints and actual application requirements. First, morphological operations are applied to adjust the continuity and integrity of the boundaries formed by the initial positive and negative boundary point sets using closing and opening operations. Then, fusion processing operations are applied to combine the initial positive and negative boundary point sets with the X-ray image and the processed image, and the boundaries are iteratively corrected multiple times to ensure the accuracy and continuity of the boundary segments, thereby obtaining the corrected positive and negative boundaries for subsequent detection and analysis.

[0070] Reference Figure 3 Specifically, in one implementation of this embodiment, step S300 includes step S301-step S302, and each step is described below.

[0071] Step S301: Optimizing the positive and negative electrode boundary point sets according to preset information of the battery to obtain optimized positive and negative electrode boundaries, wherein the preset information includes battery characteristics and battery geometry information.

[0072] Specifically, the preset information includes battery characteristics and battery geometric information. For example, battery characteristics include the type of battery chemical composition, the working principle of the battery, the battery performance parameters, and the positive and negative boundaries of the battery fluctuate regularly, the grayscale values ​​of adjacent areas have similarities and other prior knowledge. The battery geometric information includes the physical size (length, width, height) and shape (cylindrical, square, soft pack, etc.) of the battery, the internal structure of the battery, the packaging method of the battery, the battery pack layout, and geometric rules such as the Y coordinate values ​​of adjacent points will not deviate too much, the distance has a threshold, and the slope will not be 90°.

[0073] In this embodiment, the preset information of the battery to be detected is used to optimize the positive and negative boundary point sets. Since the area segmented by the battery boundary point detection model is similar to a rectangular area, and the present application requires the upper and lower sides of the rectangular area, the two sides are obtained according to the slope constraint between the points and the Y coordinate clustering, and then the distance is calculated based on two points of the vertical line selected from the two sides. The distance is determined according to the user's preset. The specific operations include removing abnormal points or noise that do not conform to the actual product characteristics through threshold screening, and adjusting the continuity and integrity of the boundary through morphological operations, using closing operations and opening operations. Among them, the threshold screening is based on histogram analysis, and the outliers are removed through the clustering algorithm to obtain two complete boundary line segments, i.e., the optimized positive and negative boundaries.

[0074] Reference Figure 4 Specifically, in one implementation of this embodiment, step S301 includes:

[0075] Step S3011: Utilizing the battery knowledge and the battery geometry information, the positive and negative electrode boundary point sets are screened to obtain screened positive and negative electrode boundary point sets.

[0076] In this embodiment, removing abnormal points or noise that do not conform to the battery characteristics by threshold screening specifically includes determining one or more thresholds according to the battery characteristics and the normal operating range. For example, it is set based on the density, thickness or contrast of the electrode material with the surrounding material. Then, using the pre-set threshold, the intensity value of the pixel in the image is compared with the threshold, and pixels below or above the threshold are regarded as abnormal points or noise, which will be marked or deleted. For points far away from normal values, statistical methods or rule-based methods are used to identify them, aiming to improve data quality and accuracy of results.

[0077] Step S3012: improving the screened positive and negative boundary point sets according to morphological operations to obtain the improved positive and negative boundary points.

[0078] In this embodiment, the screened positive and negative boundary point sets are improved according to morphological operations, including but not limited to removing small noise points on the boundary through opening operations without changing the basic shape of the boundary, effectively removing isolated points in the boundary point set caused by image noise, and closing small gaps or holes in the boundary through closing operations while keeping the general shape of the boundary unchanged, effectively filling small gaps missed by the detection algorithm, thereby obtaining improved positive and negative boundaries.

[0079] Step S302: correcting the optimized positive and negative electrode boundaries based on the X-ray image and the processed image to obtain the positive and negative electrode boundaries of the corresponding battery.

[0080] In this embodiment, the optimized positive and negative electrode boundaries are fused with the X-ray image and the processed image to evaluate whether the optimized boundaries are consistent with the actual structure of the battery. The optimized boundaries are automatically adjusted by moving boundary points, adjusting the curvature of the boundary line, or correcting the boundary shape to more accurately reflect the positive and negative electrode distribution inside the battery, ensure that the corrected boundaries are logically consistent with the battery geometric information and battery characteristics, and verify whether the corrected boundaries are reasonable, for example, the relative position and shape should comply with the standards for battery design and manufacturing. Finally, multiple iterative corrections are performed to ensure the accuracy and continuity of the boundary segments, thereby obtaining the positive and negative electrode boundaries of the battery, fully considering the internal structure information of the battery provided by the X-ray image, thereby improving the accuracy and reliability of boundary recognition.

[0081] Reference Figure 5 Specifically, in one implementation of this embodiment, step S302 includes:

[0082] Step S3021: Fusing the improved positive and negative electrode boundaries with the X-ray image and the processed image to obtain fused positive and negative electrode boundaries.

[0083] In this embodiment, the perfected positive and negative electrode boundaries are fused with the X-ray image and the processed image, and an image fusion algorithm (such as weighted average, multi-resolution fusion, etc.) is used to optimize the fusion effect of the boundary and the background to ensure that the visual effect of the image is both clear and natural, so as to obtain the fused positive and negative electrode boundaries, which not only shows the internal structure of the battery, but also highlights the key information of the positive and negative electrode boundaries.

[0084] Step S3022: correcting the fused positive and negative electrode boundaries to obtain the positive and negative electrode boundaries of the battery.

[0085] In this embodiment, the fused positive and negative electrode boundaries are corrected in combination with the battery geometry information and battery characteristics. For example, key feature points are verified, such as the junction of the positive and negative electrodes, the center point or edge of the battery, to ensure that these points are accurately positioned, and a consistency check is performed to ensure that the width, shape and position of the positive and negative electrode boundaries are consistent throughout the battery area to avoid overall errors caused by local anomalies, so as to ultimately determine the positive and negative electrode boundaries of the battery.

[0086] In one implementation of this embodiment, the battery is tested under various processes:

[0087] The first process: 50 images of batteries made by the winding process.

[0088] The second process: 50 battery images of the flattening process.

[0089] The third process: 50 images of batteries in the steel shell process.

[0090] Unified detection process: Under each process, the same detection operation steps of the battery positive and negative electrode detection method based on X-ray images provided in this embodiment are adopted.

[0091] Experimental results: Under three different processes, this embodiment shows excellent adaptability and high precision. The recognition accuracy rate under the winding process is 100%, the recognition accuracy rate under the flattening process is 100%, and the recognition accuracy rate under the steel shell process is 99%, which shows the wide applicability of the technical solution, significantly improves the accuracy, efficiency and wide applicability of detection, and provides solid technical support for the manufacturing and quality control of lithium batteries.

[0092] The application adopts the above technical solution to achieve the following effects:

[0093] The present application provides a method for detecting positive and negative electrodes of a battery based on an X-ray image, the method comprising: acquiring an X-ray image of a battery, performing image processing on the X-ray image based on a module matching algorithm to obtain a processed image; recognizing the processed image through a pre-constructed battery boundary point detection model to obtain a corresponding initial positive and negative boundary point set; and correcting the initial positive and negative boundary point set according to preset information of the battery and the processed image to obtain the positive and negative boundaries of the battery. The present application identifies and corrects X-ray battery images through a battery boundary point detection model to obtain the positive and negative boundaries of the battery, thereby improving the accuracy and efficiency of detecting the positive and negative electrodes of the battery and facilitating the high-quality production of lithium batteries.

[0094] Example 2

[0095] In addition, the present application provides a battery positive and negative electrode detection device 600 based on X-ray images, and its functional module schematic diagram is as follows: Figure 6 As shown, a battery positive and negative electrode detection device 600 based on X-ray images includes:

[0096] An acquisition module 601 is used to acquire an X-ray image of a battery, and perform image processing on the X-ray image based on a module matching algorithm to obtain a processed image;

[0097] An identification module 602 is used to identify the processed image through a deep learning model to obtain a corresponding initial positive and negative boundary point set;

[0098] The output module 603 is used to modify the initial positive and negative electrode boundary point set according to the preset information of the battery and the processed image, obtain the positive and negative electrode boundaries of the corresponding battery, and output the positive and negative electrode boundaries.

[0099] like Figure 7As shown, in one implementation, the acquisition module 601 includes:

[0100] A matching unit 6011 is used to perform block matching on the X-ray image by using the module matching algorithm to obtain a plurality of key areas including positive and negative electrode boundaries;

[0101] The processing unit 6012 is used to perform image processing on the key area to obtain a processed image, wherein the image processing includes denoising, contrast adjustment and edge enhancement.

[0102] The battery positive and negative electrode detection device 600 based on X-ray images provided in this embodiment can implement the battery positive and negative electrode detection method based on X-ray images provided in Example 1, and will not be described again here to avoid repetition.

[0103] Example 3

[0104] In addition, an embodiment of the present application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program runs on the processor, the battery positive and negative pole detection method based on X-ray images provided in Example 1 is executed.

[0105] Example 4

[0106] The present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the method for detecting positive and negative electrodes of a battery based on X-ray images provided in Example 1 is implemented.

[0107] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0108] In addition, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0109] Furthermore, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0110] It should be noted that if the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can essentially be embodied in the form of a software product, or the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM) random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0111] In this document, relational terms such as first and second, etc. are used merely to distinguish one entity or operation from another entity or operation, but do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0112] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting positive and negative electrodes of batteries based on X-ray images. It is characterized in that include: Acquire an X-ray image of the battery, and perform image processing on the X-ray image based on a module matching algorithm to obtain a processed image; Recognize the processed image through a pre-built battery boundary point detection model to obtain a corresponding initial positive and negative electrode boundary point set; The initial positive and negative electrode boundary point set is corrected according to the preset information of the battery and the processed image to obtain the positive and negative electrode boundaries of the battery.

2. The method for detecting positive and negative electrodes of a battery based on X-ray images according to claim 1, It is characterized in that Constructing the battery boundary point detection model includes: Acquire a battery image training set, wherein the battery image training set includes multiple battery positive and negative electrode boundary images; An initial battery boundary point detection model is constructed according to the battery image training set based on a preset deep learning model, and the initial battery boundary point detection model is trained to obtain the battery boundary point detection model.

3. The method for detecting positive and negative electrodes of a battery based on X-ray images according to claim 2, It is characterized in that The method of performing image processing based on the X-ray image based on the module matching algorithm to obtain a processed image includes: Performing block matching on the X-ray image by using the module matching algorithm to obtain a plurality of key areas, wherein the key areas include positive and negative electrode boundaries; Performing image processing on each of the key areas to obtain a processed image, wherein the image processing includes denoising, contrast adjustment, and edge enhancement.

4. The method for detecting positive and negative electrodes of a battery based on X-ray images according to claim 3, It is characterized in that The step of correcting the initial positive and negative electrode boundary point set according to the preset information of the battery and the processed image to obtain the positive and negative electrode boundaries of the corresponding battery includes: Optimizing the positive and negative electrode boundary point sets according to preset information of the battery to obtain optimized positive and negative electrode boundaries, wherein the preset information includes battery characteristics and battery geometry information; The optimized positive and negative electrode boundaries are corrected based on the X-ray image and the processed image to obtain the positive and negative electrode boundaries of the corresponding battery.

5. The method for detecting positive and negative electrodes of a battery based on X-ray images according to claim 4, It is characterized in that The method of optimizing the positive and negative electrode boundary point sets according to preset information of the battery to obtain optimized positive and negative electrode boundaries, wherein the preset information includes battery characteristics and battery geometry information, includes: Using the battery characteristics and the battery geometry information, the positive and negative electrode boundary point sets are screened to obtain screened positive and negative electrode boundary point sets; The screened positive and negative boundary point sets are optimized according to morphological operations to obtain the optimized positive and negative boundary points.

6. The method for detecting positive and negative electrodes of a battery based on X-ray images according to claim 4, It is characterized in that The step of correcting the optimized positive and negative electrode boundaries based on the X-ray image and the processed image to obtain the positive and negative electrode boundaries of the battery includes: Fusing the optimized positive and negative electrode boundaries with the X-ray image and the processed image to obtain fused positive and negative electrode boundaries; The boundaries of the fused positive and negative electrodes are corrected to obtain the boundaries of the positive and negative electrodes of the battery.

7. A battery positive and negative electrode detection device based on X-ray images, It is characterized in that The device comprises: An acquisition module, used for acquiring an X-ray image of the battery, and performing image processing on the X-ray image based on a module matching algorithm to obtain a processed image; A recognition module, used to recognize the processed image through a deep learning model to obtain a corresponding initial positive and negative boundary point set; An output module is used to correct the initial positive and negative electrode boundary point set according to the preset information of the battery and the processed image, obtain the positive and negative electrode boundaries of the corresponding battery, and output the positive and negative electrode boundaries.

8. The battery positive and negative electrode detection device based on X-ray images according to claim 7, It is characterized in that The acquisition module comprises: A matching unit, used for performing block matching on the X-ray image by using the module matching algorithm to obtain a plurality of key areas including positive and negative electrode boundaries; A processing unit is used to perform image processing on the key area to obtain a processed image, wherein the image processing includes denoising, contrast adjustment and edge enhancement.

9. An electronic device, It is characterized in that It comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is run by the processor, the method for detecting positive and negative electrodes of a battery based on X-ray images as claimed in any one of claims 1 to 6 is executed.

10. A computer-readable storage medium, It is characterized in that It stores a computer program, which, when running on a processor, executes the battery positive and negative electrode detection method based on X-ray images as described in any one of claims 1 to 6.

Citation Information

Patent Citations

  • Deep learning-based packaged chip defect detection method

    CN113362306A

  • Lithium battery X-ray image enhancement method and device and storage medium

    CN114494098A

  • Lithium battery core cladding structure X-Ray detection system and detection method

    CN115131583A

  • Battery polarity detection method, device and equipment and storage medium

    CN117372414A

  • Lithium battery pole piece detection method and device

    CN117557565A

Cited By

  • Intelligent recognition method for catching pteroptychus leopardus eggs underwater

    CN120071111A

  • Lithium battery production early warning and optimization method based on X-ray detection and neural network

    CN120706991A