A method and device for detecting a laminated lithium ion battery and an image processing apparatus

Abstract

The embodiment of the application provides a lamination type lithium ion battery detection method and device and image processing equipment, the method is applied to the image processing equipment in the detection system, the detection system further includes a ray source and a detector, the relative position of the ray source and the detector is fixed, the angle between the incident direction of the ray source and the preset edge of the lamination type lithium ion battery is a preset angle, the method comprises the following steps: obtaining the projection image of the lamination type lithium ion battery collected by the detector every time the relative position between the lamination type lithium ion battery and the ray source and the detector is changed, performing image processing on a plurality of projection images to obtain a target image, performing corner point identification on the target image to obtain positive plate corner points and negative plate corner points in the target image, and determining a detection result based on the positive plate corner points and the negative plate corner points. The embodiment of the application can solve the overlapping problem of the edge plate in the projection image, reduce the error of the calculation performance parameter, and further improve the accuracy of the lamination type lithium ion battery detection.

Description

Technical Field

[0001] This application relates to the field of battery testing technology, and in particular to a method, apparatus and image processing equipment for testing stacked lithium-ion batteries. Background Technology

[0002] Lithium-ion batteries, as an emerging energy material with advantages such as high energy density, high stability, and no pollution, are widely used in electronic products, new energy vehicles, and other fields. In the production of lithium-ion batteries, two main processes are employed: winding and stacking. Stacking involves cutting the positive and negative electrodes into small pieces and stacking them with a separator to form small battery cells. These small cells are then stacked and connected in parallel to form a large battery cell. Due to its advantages such as uniform stress distribution, resistance to deformation, and high energy density, it is widely used in the production of lithium-ion batteries.

[0003] To ensure the safety and reliability of stacked lithium-ion batteries, quality testing is required to ensure that the number of positive and negative electrode plates, the distance between the positive and negative electrode plates, and the alignment meet the testing requirements.

[0004] During the inspection of stacked lithium-ion batteries using X-ray inspection equipment, the stacked lithium-ion battery has many positive and negative electrode layers, small spacing, and the electrodes are easy to bend. In addition, the X-ray source is a point source. Therefore, the projection image of the stacked lithium-ion battery is prone to overlap of the edge electrodes. This increases the difficulty of inspecting stacked lithium-ion batteries, easily introduces errors when calculating performance parameters, and results in low accuracy of stacked lithium-ion battery inspection. Summary of the Invention

[0005] The purpose of this application is to provide a method, apparatus, system, device, and medium for testing stacked lithium-ion batteries, so as to improve the accuracy of testing stacked lithium-ion batteries. The specific technical solution is as follows:

[0006] In a first aspect, embodiments of this application provide a method for detecting stacked lithium-ion batteries, applied to an image processing device in a detection system. The detection system further includes an X-ray source and a detector, the relative positions of the X-ray source and the detector are fixed, and the angle between the incident direction of the X-ray source and a preset edge of the stacked lithium-ion battery is a preset angle. The method includes:

[0007] Each time the relative position between the stacked lithium-ion battery, the radiation source, and the detector is changed, a projection image of the stacked lithium-ion battery acquired by the detector is obtained.

[0008] Image processing is performed on multiple projected images to obtain the target image;

[0009] Corner point recognition is performed on the target image to obtain the positive electrode corner points and negative electrode corner points in the target image;

[0010] The detection result is determined based on the corner points of the positive electrode and the negative electrode.

[0011] Secondly, embodiments of this application provide a stacked lithium-ion battery detection device, applied to an image processing device in a detection system. The detection system further includes an X-ray source and a detector, the relative positions of the X-ray source and the detector are fixed, and the angle between the incident direction of the X-ray source and a preset edge of the stacked lithium-ion battery is a preset angle. The device includes:

[0012] The projection image acquisition module is used to acquire the projection image of the stacked lithium-ion battery collected by the detector every time the relative position between the stacked lithium-ion battery, the radiation source, and the detector is changed.

[0013] The target image acquisition module is used to process multiple projection images to obtain the target image.

[0014] The corner acquisition module is used to identify corners in the target image to obtain the positive electrode corners and negative electrode corners in the target image.

[0015] The detection result determination module is used to determine the detection result based on the corner points of the positive electrode and the corner points of the negative electrode.

[0016] Thirdly, embodiments of this application provide a detection system, which includes an image processing device, an X-ray source, and a detector. The relative positions of the X-ray source and the detector are fixed, and the angle between the incident direction of the X-ray source and a preset edge of the stacked lithium-ion battery is a preset angle, wherein:

[0017] The radiation source is used to emit radiation that can be transmitted through the stacked lithium-ion battery.

[0018] The detector is used to acquire projected images of the stacked lithium-ion battery;

[0019] The image processing device acquires a projected image of the stacked lithium-ion battery captured by the detector each time the user changes the relative position between the stacked lithium-ion battery, the radiation source, and the detector; performs image processing on multiple projected images to obtain a target image; identifies corner points in the target image to obtain the positive electrode corner points and negative electrode corner points in the target image; and determines the detection result based on the positive electrode corner points and negative electrode corner points.

[0020] Fourthly, embodiments of this application provide an image processing apparatus, including:

[0021] Memory, used to store computer programs;

[0022] When a processor executes a program stored in memory, it implements any of the methods described in the first aspect above.

[0023] Fifthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the methods described in the first aspect above.

[0024] Beneficial effects of the embodiments in this application:

[0025] The solution provided in this application embodiment is applied to an image processing device in a detection system. The detection system also includes an X-ray source and a detector. The relative positions of the X-ray source and the detector are fixed. The angle between the incident direction of the X-ray source and the preset edge of the stacked lithium-ion battery is a preset angle. Whenever the relative position between the stacked lithium-ion battery and the X-ray source and the detector changes, the image processing device can acquire the projected image of the stacked lithium-ion battery collected by the detector. Image processing is performed on multiple projected images to obtain a target image. Corner point recognition is performed on the target image to obtain the positive electrode corner points and negative electrode corner points in the target image. Based on the positive electrode corner points and negative electrode corner points, the detection result is determined. Because the detector can acquire a projected image of the stacked lithium-ion battery each time the relative position between the stacked lithium-ion battery, the X-ray source, and the detector is changed, and multiple projected images from different positions are acquired, image processing can be performed on multiple projected images with non-overlapping areas, or image reconstruction can be performed on multiple projected images to obtain a clear image of the stacked lithium-ion battery for calculating performance parameters. This can solve the problem of overlapping edge electrodes in the projected image, reduce the error in calculating performance parameters, and thus improve the accuracy of stacked lithium-ion battery detection. Of course, implementing any product or method of this application does not necessarily require achieving all of the above advantages simultaneously. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0027] Figure 1 This is a schematic diagram of the structure of a lithium-ion battery.

[0028] Figure 2 This is a schematic diagram of a lithium-ion battery stacking process;

[0029] Figure 3 This is a schematic diagram illustrating the principle of X-ray inspection of stacked lithium-ion batteries.

[0030] Figure 4 This is a schematic diagram illustrating how the characteristics of an X-ray point source cause image overlap on a negative electrode.

[0031] Figure 5 This is a schematic diagram of the structure of a detection system provided in an embodiment of this application;

[0032] Figure 6 A flowchart of a method for testing a stacked lithium-ion battery provided in an embodiment of this application;

[0033] Figure 7 This is a schematic diagram of an X-ray multilayer scanning stacked lithium-ion battery provided in an embodiment of this application;

[0034] Figure 8 for Figure 6 A specific flowchart of step S602 in the illustrated embodiment;

[0035] Figure 9 This is a schematic diagram of a projection image of a stacked lithium-ion battery provided in an embodiment of this application.

[0036] Figure 10 This is a schematic diagram of a target image of a stacked lithium-ion battery provided in an embodiment of this application.

[0037] Figure 11 for Figure 6 Another specific flowchart of step S602 in the illustrated embodiment;

[0038] Figure 12 This is another schematic diagram of a projected image of a stacked lithium-ion battery provided in an embodiment of this application;

[0039] Figure 13 This is another schematic diagram of a target image of a stacked lithium-ion battery provided in an embodiment of this application;

[0040] Figure 14 for Figure 11 Another specific flowchart of step S1101 in the illustrated embodiment;

[0041] Figure 15 for Figure 6 A specific flowchart of step S604 in the illustrated embodiment;

[0042] Figure 16 Based on Figure 6 The illustrated embodiment provides a schematic diagram for determining the corner points of the positive and negative electrode plates;

[0043] Figure 17Based on Figure 6 The illustrated embodiment provides a schematic diagram of the curves for determining the corner points of the positive and negative electrode plates;

[0044] Figure 18 This is a specific flowchart of a method for detecting stacked lithium-ion batteries provided in an embodiment of this application.

[0045] Figure 19 This is another specific flowchart of the stacked lithium-ion battery testing method provided in the embodiments of this application;

[0046] Figure 20 This is another specific flowchart of the stacked lithium-ion battery testing method provided in the embodiments of this application;

[0047] Figure 21 This is a schematic diagram of the structure of a stacked lithium-ion battery detection device provided in an embodiment of this application;

[0048] Figure 22 This is a schematic diagram of the structure of an image processing device provided in an embodiment of this application. Detailed Implementation

[0049] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0050] Here is a brief introduction to the performance parameters and testing methods of stacked lithium-ion batteries.

[0051] A lithium-ion battery is a rechargeable battery that works by the movement of lithium ions between the positive and negative electrodes. The structure of a lithium-ion battery is as follows: Figure 1 As shown, the device includes an aluminum-plastic film 101 and a battery cell. The battery cell has electrodes 102 and tabs 103. The battery cell can be manufactured using a lamination process, for example, as shown in... Figure 2 As shown, the positive electrode 201, the negative electrode 203 and the separator 202 are stacked to form a small battery cell, and then the small battery cells are stacked and connected in parallel to form a large battery cell, which is a stacked lithium-ion battery cell.

[0052] In the production of stacked lithium-ion battery cells, the positive electrode, separator, and negative electrode are stacked sequentially, so the number of each electrode is fixed. If there are too few negative electrodes, lithium ions continuously move from the positive electrode to the negative electrode during charging. The negative electrode cannot completely cover the positive electrode, which may result in insufficient space for the negative electrode to hold lithium ions, leading to lithium plating and posing a safety hazard to the battery.

[0053] In the production of stacked lithium-ion battery cells, the positive and negative electrode sheets need to be aligned and stacked separately. Poor alignment of the electrode sheets will cause capacity loss and increase the risk of short circuit.

[0054] Furthermore, in the projection image of a stacked lithium-ion battery, two types of electrode plates can be displayed. The longer electrode plate is the negative electrode plate, and the shorter electrode plate is the positive electrode plate. The portion of the negative electrode plate that is longer than the positive electrode plate is called the negative electrode redundancy, which is the distance between adjacent positive and negative electrode plates. If the distance between adjacent positive and negative electrode plates is too short, lithium dendrites that can pierce the separator can easily form, causing thermal runaway and leading to a short circuit. If the distance between adjacent positive and negative electrode plates is too long, it can easily lead to a reduction in battery capacity. Therefore, the distance between adjacent positive and negative electrode plates needs to be within a reasonable range.

[0055] Therefore, in the production process of stacked lithium-ion batteries, the number of positive and negative electrode plates, the distance between the positive and negative electrode plates, and their alignment are important performance parameters. Thus, when conducting quality inspections on stacked lithium-ion batteries, it is necessary to test these three performance parameters.

[0056] Currently, X-ray inspection equipment is used to inspect stacked lithium-ion batteries, such as... Figure 3 As shown, X-rays emitted from X-ray source 301 are transmitted through the side of stacked lithium-ion battery 302 at a certain angle. Detector 303 receives the photons that penetrate the stacked lithium-ion battery 302 and obtains a projected image to complete the corner detection of stacked lithium-ion battery 302, obtaining the corners of positive and negative electrodes. Based on the obtained corners of positive and negative electrodes, the number of positive and negative electrodes, the distance between positive and negative electrodes, and the alignment can be determined.

[0057] Because the X-rays emitted by the X-ray source 301 are cone-beam X-rays, when the X-rays are transmitted through the stacked lithium-ion battery 302, they will obliquely penetrate the multiple layers of electrodes, causing the edge electrodes to overlap in the projected image, such as... Figure 4 In the obtained projected image, there are overlapping areas 401 and non-overlapping areas 402 of the negative electrode sheets. If there are overlapping areas in the projected image, it will cause a large error when calculating performance parameters, increasing the difficulty of testing stacked lithium-ion batteries and resulting in low accuracy in the testing of stacked lithium-ion batteries.

[0058] Therefore, in order to solve the problem of overlapping edge electrodes in the projected image and improve the accuracy of stacked lithium-ion battery detection, this application provides a method, apparatus, system, image processing device, computer-readable storage medium, and computer program product for detecting stacked lithium-ion batteries. The following first describes a method for detecting stacked lithium-ion batteries provided by this application.

[0059] The stacked lithium-ion battery detection method provided in this application embodiment can be applied to, for example... Figure 5 The image processing device 503 in the detection system shown includes a radiation source 501 and a detector 502. The relative positions of the radiation source 501 and the detector 502 are fixed, and the angle between the incident direction of the radiation source 501 and the preset edge of the stacked lithium-ion battery is a preset angle.

[0060] like Figure 6 As shown, a method for testing stacked lithium-ion batteries includes:

[0061] S601, each time the relative position between the stacked lithium-ion battery and the radiation source and the detector is changed, the projected image of the stacked lithium-ion battery collected by the detector is obtained.

[0062] S602, Image processing is performed on multiple projected images to obtain the target image;

[0063] S603, perform corner point recognition on the target image to obtain the positive electrode corner point and the negative electrode corner point in the target image;

[0064] S604, determine the detection result based on the corner points of the positive electrode and the corner points of the negative electrode.

[0065] As can be seen, in the solution provided in this application embodiment, the image processing device applied to the detection system includes an X-ray source and a detector. The relative positions of the X-ray source and the detector are fixed. The angle between the incident direction of the X-ray source and the preset edge of the stacked lithium-ion battery is a preset angle. Whenever the relative position between the stacked lithium-ion battery and the X-ray source and the detector changes, the image processing device can acquire the projected image of the stacked lithium-ion battery collected by the detector, perform image processing on multiple projected images to obtain a target image, perform corner point recognition on the target image to obtain the positive electrode corner point and the negative electrode corner point in the target image, and determine the detection result based on the positive electrode corner point and the negative electrode corner point. Since the detector can acquire a projected image of a stacked lithium-ion battery each time the relative position between the stacked lithium-ion battery, the X-ray source, and the detector is changed, and multiple projected images at different positions are acquired, image processing can be performed on multiple projected images with non-overlapping areas, or image reconstruction can be performed on multiple projected images to obtain a clear image of the stacked lithium-ion battery for calculating performance parameters. This can solve the problem of overlapping edge electrodes in the projected image, reduce the error in calculating performance parameters, and thus improve the accuracy of stacked lithium-ion battery detection.

[0066] In the detection system, the relative positions of the X-ray source and the detector remain constant. To obtain projected images of the stacked lithium-ion battery at different shooting angles, the relative positions of the stacked lithium-ion battery, the X-ray source, and the detector can be changed. Therefore, in step S601, each time the relative positions of the stacked lithium-ion battery, the X-ray source, and the detector are changed, the image processing device can acquire the projected image of the stacked lithium-ion battery captured by the detector.

[0067] In one implementation, the positions of the X-ray source and detector can be fixed, while the position of the stacked lithium-ion battery can be changed. For example, the stacked lithium-ion battery can move up and down between the X-ray source and the detector. In this way, the X-ray source transmits light through the stacked lithium-ion battery at different positions, and the detector can collect the projected images of the stacked lithium-ion battery at different positions.

[0068] In another implementation, the position of the stacked lithium-ion battery can be fixed while the positions of the X-ray source and detector can be changed. For example, the X-ray source and detector can move up and down synchronously, with the X-ray source transmitting light through the stacked lithium-ion battery at different positions. In this case, the detector can acquire projected images of the stacked lithium-ion battery at different positions.

[0069] The X-ray source is a point source that can transmit light to a corner of the stacked lithium-ion battery, ensuring that the angle between the incident direction of the X-ray source and a preset edge of the stacked lithium-ion battery is a preset angle. This preset angle can be 45°, and is not specifically limited here.

[0070] For example, a detection system may include a radiation source, a detector, an image acquisition device, and a lifting and rotating platform. Figure 7 As shown, the relative positions of the X-ray source 701 and the detector 703 are fixed. The direction of the X-ray emitted by the X-ray source 701 forms a 45° angle with the edge 704 of the stacked lithium-ion battery 702, and the X-ray is parallel to the positive and negative electrodes in the stacked lithium-ion battery. With the X-ray source 701 and the detector 703 fixed in position, the stacked lithium-ion battery 702 is placed on a lifting and rotating platform that moves up and down along a line perpendicular to the line connecting the X-ray source 701 and the detector 703. In this way, each time the stacked lithium-ion battery 702 moves, the detector 703 can acquire a projected image of one stacked lithium-ion battery, and the image processing equipment can then acquire projected images of multiple stacked lithium-ion batteries.

[0071] Because the point light source characteristics of the X-ray source cause overlap of the edge electrodes in the projected image, some of the multiple projected images acquired by the image processing device contain partially overlapping electrode areas, and in some projected images, the positive and negative electrodes cannot be separated. To obtain a clear image of the stacked lithium-ion battery without overlapping areas, the image processing device can then perform image processing on the multiple projected images using a preset image processing method to obtain the target image, i.e., execute step S602.

[0072] In one embodiment, the image processing device acquires projection images of each stacked lithium-ion battery, which contain overlapping and non-overlapping areas of the electrodes. In this case, the non-overlapping areas in the multiple projection images can be extracted and fused to obtain a clear projection image as the target image.

[0073] For example, an image segmentation algorithm or a deep learning segmentation algorithm based on a network model can be used to extract the non-overlapping regions in each projected image, and then the obtained non-overlapping regions can be matched using an image registration method. Finally, the matched non-overlapping regions can be fused to obtain the target image.

[0074] In one embodiment, the projected images of each stacked lithium-ion battery acquired by the image processing device may contain non-overlapping areas of the electrodes, or there may be no non-overlapping areas at all, meaning that the positive and negative electrodes cannot be separated. In this case, image reconstruction can be performed on multiple projected images to obtain a clear image of the stacked lithium-ion battery, which can be used as the target image. The image reconstruction can be CT (Computed Tomography) reconstruction, etc., and is not specifically limited here.

[0075] Next, in step S603, the image processing device can perform corner point recognition on the target image to obtain the positive electrode corner points and negative electrode corner points in the target image. The corner points are the edge points of the electrodes and can be used as feature points.

[0076] In other words, if the target image is clear and has no overlapping areas, then each positive electrode and each negative electrode in the target image has edge points. Therefore, the corner points of the positive electrode and the corner points of the negative electrode in the target image can be obtained by the corner point detection method.

[0077] Furthermore, in step S604, based on the corner points of the positive electrode and the corner points of the negative electrode, the image processing device can determine the detection result, wherein the detection result may include the number of positive electrode and negative electrode, the distance between the positive electrode and the negative electrode, and the alignment, which are not specifically limited here.

[0078] In one implementation, since the positive and negative electrode corner points are points marked on the edges of the electrodes in the target image, the number of positive electrodes can be determined by the number of positive electrode corner points in the target image, and the number of negative electrodes can be determined by the number of negative electrode corner points in the target image.

[0079] Given the corner points of the positive and negative electrodes, the image processing device can determine the positions of the positive and negative electrode corner points, and then obtain the coordinates of any adjacent positive and negative electrode corner points. Thus, the distance between the positive and negative electrodes can be calculated using the distance formula between the two points.

[0080] Once the corner points of the positive and negative electrodes are determined, the image processing device can also perform curve fitting on the corner points of the positive and negative electrodes to obtain the corner point curves of the positive and negative electrodes. Then, based on the corner point curves of the positive and negative electrodes, the alignment degree of the positive and negative electrodes of the stacked lithium-ion battery can be calculated.

[0081] In this embodiment, since the detector can acquire a projection image of a stacked lithium-ion battery each time the relative position between the stacked lithium-ion battery, the X-ray source, and the detector is changed, and multiple projection images at different positions are acquired, image processing can be performed on multiple projection images with non-overlapping areas, or image reconstruction can be performed on multiple projection images to obtain a clear image of the stacked lithium-ion battery for calculating performance parameters. This can solve the problem of overlapping edge electrodes in the projection image, reduce the error in calculating performance parameters, and thus improve the accuracy of stacked lithium-ion battery detection.

[0082] Furthermore, the number of positive and negative electrode plates, the distance between the positive and negative electrode plates, and the alignment of the stacked lithium-ion batteries obtained through the embodiments of this application do not rely on any other predictive models. Therefore, the performance parameters are calculated based on the inherent properties of the stacked lithium-ion battery itself, resulting in high accuracy. Moreover, the detection system has a simple structure and achieves the desired effect without the expensive slip ring structure of conventional CT, solving practical problems and reducing costs.

[0083] As one implementation method of this application, such as Figure 8 As shown, the steps described above for processing multiple projected images using a preset image processing method to obtain a target image may include:

[0084] S801, a preset segmentation algorithm is used to segment the pole piece region in each projection image to obtain the non-overlapping region of the pole piece in each projection image;

[0085] In the multiple projected images of stacked lithium-ion batteries acquired by the image processing device, some projected images contain overlapping areas of the electrodes, while others contain non-overlapping areas of the electrodes. The overlapping areas are those where the edges of the electrodes in the projected images overlap, while the non-overlapping areas are those where the edges of the electrodes in the projected images do not overlap.

[0086] Therefore, image segmentation can be performed on each projected image to extract the non-overlapping regions of the electrodes in each projected image for registration and fusion, resulting in a clear stacked lithium-ion battery. Image segmentation is the technique and process of dividing an image into several specific regions with unique properties and extracting the target of interest.

[0087] In one embodiment, the image processing device can use a preset segmentation algorithm to segment the polaribe regions in each projected image, thereby separating the non-overlapping regions of the polaribes in each projected image and obtaining the non-overlapping regions of the polaribes in each projected image. The preset segmentation algorithm can be an image segmentation algorithm, including threshold-based segmentation algorithms, region-based segmentation algorithms, edge-based segmentation algorithms, etc., or it can be a deep learning segmentation algorithm based on various network models; no specific limitation is made here.

[0088] For example, an image processing device acquires projected images of n stacked lithium-ion batteries, such as... Figure 9 As shown, five exemplary projected images are presented, each containing overlapping and non-overlapping areas of the pole pieces. Among them, [the following is a list of images, likely from a different source]... Figure 9As can be seen, the central ray emitted by the X-ray source corresponds to the non-overlapping area of ​​the pole pieces in the projected image, as shown in non-overlapping areas 901 and 902. The positive and negative pole pieces are clearly visible and easily distinguishable within the non-overlapping area, with distinct edges. The dashed box is only used to indicate the non-overlapping area and does not imply that the projected image includes the area within the dashed box.

[0089] The region corresponding to the rays far from the center of the radiation source in the projected image is the overlapping region of the pole pieces. In this overlapping region, it is difficult to distinguish between positive and negative pole pieces. Therefore, a region-based segmentation method can be used to segment the pole piece regions in n projected images, obtaining the non-overlapping regions of the pole pieces in each projected image, thus obtaining the non-overlapping regions of the n pole pieces.

[0090] S802, Based on the feature points of the non-overlapping region, determine the fusion position relationship of the non-overlapping region in the target image;

[0091] Since the image processing device acquires projection images of stacked lithium-ion batteries at different positions, the non-overlapping areas in different projection images are located in different positions within the electrode area included in the projection image. They may be located at the leftmost position in the electrode area, a position closer to the left in the electrode area, or a position in the middle of the electrode area.

[0092] Therefore, when two non-overlapping regions are similarly positioned in their respective projected images, the feature points within these two non-overlapping regions have matching feature points. Thus, when these matching feature points are fused to obtain the target image, their positions in the target image are identical. Therefore, the fusion positional relationship of the non-overlapping regions in the target image can be determined based on their feature points. Here, the feature points are the points on the edges of the pole pieces within the non-overlapping regions.

[0093] For example, such as Figure 9 As shown, in the first and second projected images, the non-overlapping regions 901 and 902 are positioned similarly in their respective projected images. Since the feature points included in non-overlapping regions 901 and 902 have matching feature points, these matching feature points will be located in the same position in the target image when fused together. Therefore, the image processing device can find the feature points in non-overlapping regions 901 and 902, and then determine the corresponding fusion position relationship of non-overlapping regions 901 and 902 in the target image based on the found feature points.

[0094] As one implementation, the step of determining the fusion position relationship of the non-overlapping regions in the target image based on the feature points of the non-overlapping regions may include:

[0095] Feature points are extracted from non-overlapping regions in each projected image; the feature points in each non-overlapping region are matched to obtain matching results; based on the matching results and the positions of the feature points in the corresponding non-overlapping regions, the fusion position relationship of each non-overlapping region in the target image is determined. The matching results indicate whether the feature points in each non-overlapping region correspond to the same point on the edge of the electrode.

[0096] Image processing equipment can extract feature points from non-overlapping regions and use image registration algorithms to match these feature points to obtain a matching result. Image registration is the process of matching and overlaying two or more images acquired at different times, using different sensors, or under different conditions.

[0097] Specifically, the image processing device can determine whether feature points in each non-overlapping region correspond to the same point on the edge of the electrode in the target image. If the matching result indicates that multiple feature points are mutually matched, then these feature points correspond to the same point on the edge of the electrode in the target image. If the matching result indicates that multiple feature points are not mutually matched, then these feature points correspond to different points on the edge of the electrode in the target image.

[0098] Furthermore, the image processing device can determine the fusion position relationship between each non-overlapping region and its corresponding location in the target image based on the matching result and the position of the feature point in the corresponding non-overlapping region.

[0099] The relationship between the positions of matching feature points in each non-overlapping region can represent the coordinate mapping relationship between non-overlapping regions with matching feature points. This can also represent the coordinate mapping relationship when merging non-overlapping regions with matching feature points into a target image. Therefore, the image processing device can determine the fusion position relationship of each non-overlapping region in the target image based on the matching result and the position of the feature points in the corresponding non-overlapping regions.

[0100] For example, such as Figure 9 As shown, the feature points included in the non-overlapping regions 901 and 902 have matching feature points. The image processing device can determine the fusion position relationship between the non-overlapping regions 901 and 902 in the target image based on the relationship between the positions of these matching feature points in the non-overlapping regions 901 and 902. Furthermore, when fusing to obtain the target image, the non-overlapping regions 901 and 902 can be fused according to this fusion position relationship.

[0101] S803, based on the fusion positional relationship, perform image fusion processing on the non-overlapping regions to obtain the target image.

[0102] Next, the image processing device can perform image fusion processing on non-overlapping areas based on the fusion positional relationship to obtain the target image. Image fusion involves processing image data of the same target to extract valuable information and form a high-quality image.

[0103] For example, following the example in step S801 above, after obtaining the non-overlapping regions of n pole pieces, the image processing device can use an image feature registration algorithm to find the feature points of the n non-overlapping regions. Based on the found feature points of the non-overlapping regions, the fusion position relationship of the n non-overlapping regions in the target image is determined, and then image fusion processing is performed on the n non-overlapping regions to obtain a projected image of a pole piece without overlapping regions, i.e., the target image. Figure 10 As shown, the positive and negative electrodes are clearly visible in the target image, solving the problem of overlapping edge electrodes in the projected image.

[0104] As can be seen, in this embodiment, the image processing device can use a preset segmentation algorithm to segment the electrode regions in each projected image, obtaining non-overlapping regions of the electrodes in each projected image. Based on the feature points of the non-overlapping regions, the fusion position relationship of the non-overlapping regions in the target image is determined. Based on the fusion position relationship, image fusion processing is performed on the non-overlapping regions to obtain the target image. Since multiple projected images from different locations can be acquired, image segmentation, image registration, and image fusion processing can be performed on multiple projected images with non-overlapping regions to obtain a clear image of the stacked lithium-ion battery for calculating performance parameters. Therefore, the problem of overlapping edge electrodes in the projected image can be solved, the error in calculating performance parameters can be reduced, and the accuracy of stacked lithium-ion battery detection can be improved.

[0105] As one implementation method of this application, such as Figure 11 As shown, the steps described above for processing multiple projected images to obtain a target image may include:

[0106] S1101, For each projected image, calculate the projection error of each pixel in the projected image;

[0107] The projection error refers to the difference between the pixel value of a pixel in the projected image and the theoretical pixel value.

[0108] S1102, Based on the projection error corresponding to each pixel in each projected image, calculate the projection error corresponding to the projected image.

[0109] When a X-ray source transmits light through a stacked lithium-ion battery, the projected image may not only contain overlapping areas of the electrodes, but also, due to environmental factors or noise, may appear unclear or blurry. This can result in the positive and negative electrodes being indistinguishable in the projected images of multiple stacked lithium-ion batteries acquired by the image processing equipment. Figure 12 As shown, in the projected images of the six stacked lithium-ion batteries acquired by the image processing device, the positive and negative electrode sheets cannot be separated. Therefore, the projected images can be reconstructed based on the projection errors of the pixels in each image.

[0110] Image processing equipment can calculate the projection error of each pixel in a projected image. After obtaining the projection errors of all pixels in the projected image, it can calculate the overall projection error of the projected image. The projection error is the difference between the actual pixel value of a pixel in the projected image and the calculated theoretical pixel value.

[0111] For example, such as Figure 12 As shown, the detector acquires projection images of six stacked lithium-ion batteries, each of which is a 256x256 pixel image. For each projection image, the image processing device can calculate the projection error of 256x256 pixels. Based on the obtained projection error of 256x256 pixels, the corresponding projection error of that projection image can be obtained. Therefore, the projection errors corresponding to the six projection images can be obtained.

[0112] S1103, Based on the projection error corresponding to each projection image, the projection image is corrected to obtain the corrected projection image;

[0113] S1104, Reconstruct the target image based on each corrected projection image.

[0114] After obtaining the projection error corresponding to each projected image, the image processing device can use the projection error of each projected image as a correction value to correct the projected image, thereby obtaining a corrected projected image. Furthermore, the image processing device can perform image reconstruction based on the obtained corrected projected images, which can sharpen blurred areas in the projected images to obtain a clear image of the stacked lithium-ion battery, i.e., the target image.

[0115] In one implementation, since the attenuation of the radiation emitted by the radiation source is related to the length it passes through the object, the projected image will differ depending on the length the radiation passes through the stacked lithium-ion battery. For a clear target image, it is mapped according to different passing lengths, and the resulting projected image is the corrected projected image corresponding to that passing length. That is, the passing length, the target image, and the corrected projected image satisfy AF = P, where A represents the length the radiation passes through the stacked lithium-ion battery, F represents the target image to be reconstructed, and P represents the corrected projected image. Therefore, after calculating each passing length and the corresponding corrected projected image, the target image to be reconstructed can be calculated according to the above formula. This reconstructed target image is the image obtained by backprojecting the corrected projected image.

[0116] For example, an image processing device acquires projected images of n stacked lithium-ion batteries, in which the positive and negative electrode plates cannot be separated. Figure 12 The image shows projected images of six stacked lithium-ion batteries. The image processing device can obtain the lengths of the rays passing through the stacked lithium-ion batteries, A1, A2…An, and based on the correction values ​​corresponding to each projected image, it can obtain corrected projected images, P1, P2…Pn. Then, according to the formula AF = P, substituting the known A1, A2…An and P1, P2…Pn into this formula, a corresponding system of equations can be established, and the target image can be calculated, as shown below. Figure 13 As shown, the positive and negative electrodes are clearly visible in the target image, resolving the issues of overlapping and blurring of edge electrodes in the projected image.

[0117] As can be seen, in this embodiment, the image processing device can calculate the projection error of each pixel in each projected image. The projection error represents the difference between the pixel value and the theoretical pixel value of a pixel in the projected image. Based on the projection errors corresponding to each pixel in each projected image, the projection error corresponding to that projected image is calculated. The projected image is then corrected based on the projection errors corresponding to each projected image to obtain a corrected projected image. Image reconstruction is then performed based on each corrected projected image to obtain the target image. Since multiple projected images from different locations can be acquired, image processing can be performed on these multiple projected images to obtain a clear image of the stacked lithium-ion battery for calculating performance parameters. This solves the problem of overlapping edge electrodes in the projected image, reduces the error in calculating performance parameters, and thus improves the accuracy of stacked lithium-ion battery detection.

[0118] As one implementation method of this application, such as Figure 14As shown, the steps described above for calculating the projection error of each pixel in a projected image can include:

[0119] S1401, calculate the through length corresponding to each projected image based on the distance between the radiation source and the detector, the distance between the radiation source and the stacked lithium-ion battery, and the distance the stacked lithium-ion battery moves relative to the radiation source each time;

[0120] Wherein, the passing length is the length through which the rays emitted by the ray source travel in the stacked lithium-ion battery.

[0121] When a radiation source transmits light through a stacked lithium-ion battery, the intensity of the radiation decreases. Because the distance the radiation travels through the stacked lithium-ion battery varies—that is, the length it travels within the battery is different—the degree of intensity attenuation during transmission varies. Consequently, the pixel values ​​of each pixel in the projected image acquired by the detector are different.

[0122] Based on the geometric relationships existing in the detection system, the passing length corresponding to each projected image can be calculated, that is, the length that the rays emitted by the X-ray source travel in the stacked lithium-ion battery.

[0123] In one embodiment, the image processing device can calculate the through length corresponding to each projected image based on the distance between the X-ray source and the detector, the distance between the X-ray source and the stacked lithium-ion battery, and the distance the stacked lithium-ion battery moves relative to the X-ray source each time.

[0124] For example, such as Figure 7 As shown, the relative positions of the X-ray source 701 and the detector 703 in the detection system are fixed. The distance between the X-ray source and the detector is denoted as SDD (Source Detector Distance). The stacked lithium-ion battery can be placed on the worktable, and the distance between the X-ray source and the stacked lithium-ion battery is denoted as SOD (Source Object Distance). The worktable can move up and down along the line perpendicular to the connection between the X-ray source and the detector at a preset distance Δs. That is, the distance that the stacked lithium-ion battery moves relative to the X-ray source each time is Δs. Then, based on the known geometric relationship in the detection system, the length of the X-ray emitted by the X-ray source passing through the stacked lithium-ion battery can be calculated, that is, the passing length corresponding to each projected image can be obtained.

[0125] For example, the rays emitted by the X-ray source 701 pass through the stacked lithium-ion battery 702 and reach the detector 703. The stacked lithium-ion battery 702 is discretized into n*n finite pixels, each pixel being a square with a side length of d. The pixel grid is numbered sequentially. Each ray from the X-ray source 701 to the detector 703 is a straight line. It is determined whether the equation of the straight line intersects with the discretized pixel grid. If they do not intersect, the length of the line is 0. If they intersect, the distance to each grid is calculated, and the grid number and grid length are recorded, thus obtaining the length of the line passing through each projected image.

[0126] S1402, For each projected image, based on the passing length of each ray corresponding to the projected image, calculate the theoretical pixel value of the pixel point corresponding to each ray in the detector;

[0127] S1403, for each projected image, calculate the difference between the pixel value of each pixel and the corresponding theoretical pixel value to obtain the projection error corresponding to each pixel in the projected image.

[0128] For each projected image, after calculating the passing length of each ray corresponding to the projected image, the image processing device can calculate the theoretical pixel value of the corresponding pixel in the detector based on the passing distance. In this way, with the pixel value and theoretical pixel value of the corresponding pixel in the detector for each ray, the difference between the pixel value of each pixel collected by the detector and the corresponding theoretical pixel value can be calculated, thereby obtaining the projection error corresponding to each pixel in the projected image.

[0129] For example, if a ray S emitted by a radiation source transmits through a stacked lithium-ion battery, the length of the ray S through the stacked lithium-ion battery can be calculated based on the geometric relationships existing in the detection system, denoted as L. Based on this length L, the theoretical pixel value P of the pixel corresponding to the ray S in the projected image can be calculated. 理论 The pixel value of the pixel corresponding to the ray S collected by the detector is P. 实际 Then, the pixel value P of that pixel can be calculated. 实际 With the corresponding theoretical pixel value P 理论 The difference between the two values ​​is used to obtain the projection error corresponding to that pixel. Similarly, the difference between the pixel value of each pixel in the projected image and the corresponding theoretical pixel value can be calculated to obtain the projection error corresponding to each pixel in the projected image.

[0130] As can be seen, in this embodiment, the image processing device can calculate the penetration length corresponding to each projected image based on the distance between the X-ray source and the detector, the distance between the X-ray source and the stacked lithium-ion battery, and the distance the stacked lithium-ion battery moves relative to the X-ray source each time. Based on the penetration length corresponding to each X-ray in each projected image, the theoretical pixel value of the pixel corresponding to each X-ray in the detector is calculated, and then the difference between the pixel value of each pixel and the corresponding theoretical pixel value is calculated to obtain the projection error corresponding to each pixel in the projected image. Therefore, the projection error corresponding to the projected image can be calculated based on the projection error corresponding to each pixel in each projected image, thereby making the correction value of the obtained projected image more accurate.

[0131] As one embodiment of this application, the step of calculating the projection error corresponding to the projection image based on the projection error corresponding to each pixel in each projection image may include:

[0132] For each projected image, the projection errors corresponding to each pixel in the projected image are summed to obtain the projection error corresponding to that projected image.

[0133] Since the radiation emitted by the radiation source is affected by noise or environmental factors during the transmission of radiation to the stacked lithium-ion battery, and the impact on each radiation is different, the image processing device can calculate the projection error of each projected image based on the projection error of all the pixels in the projected image after obtaining the projection error of all the pixels in the projected image.

[0134] In one embodiment, for each projected image, the image processing device can sum the projection errors of all pixels in the obtained projected image to obtain the projection error corresponding to the projected image.

[0135] In another implementation, for each projected image, the image processing device can perform a weighted summation of the projection errors of all pixels in the obtained projected image to obtain the projection error corresponding to the projected image.

[0136] For example, if each projected image is a 256x256 image, then for each projected image, the image processing device can calculate the projection error of 256x256 pixels, sum the projection errors of the 256x256 pixels to obtain the projection error corresponding to the projected image, or it can perform a weighted summation of the projection errors of the 256x256 pixels according to the influence of each pixel to obtain the projection error corresponding to the projected image.

[0137] As can be seen, in this embodiment, for each projected image, the projection errors corresponding to each pixel in the projected image are summed to obtain the projection error corresponding to that projected image. In this way, using the projection error corresponding to the projected image as a correction value to correct the projected image can reduce the influence of noise or environmental factors, making the corrected projected image clearer.

[0138] As one embodiment of this application, the step of determining the detection result based on the corner points of the positive electrode and the corner points of the negative electrode may include:

[0139] Based on the corner points of the positive electrode and the corner points of the negative electrode, the detection parameters of the stacked lithium-ion battery are determined. The detection parameters include at least one of the following: the number of positive and negative electrodes of the stacked lithium-ion battery, the distance between the positive and negative electrodes, and the alignment of the target electrode. The target electrode includes both positive and negative electrodes.

[0140] The alignment of the target electrode is calculated based on the corner point curve of the target electrode, which is obtained by curve fitting of the corner points of the target electrode.

[0141] For cases where the detection parameters include the number of positive and negative electrodes, the distance between the positive and negative electrodes, and the alignment of the target electrode, such as... Figure 15 As shown, the methods for determining the number of positive and negative electrodes, the distance between the positive and negative electrodes, and the alignment of the target electrodes based on the corner points of the positive and negative electrodes can include:

[0142] S1501, Based on the corner points of the positive electrode and the corner points of the negative electrode, determine the number of positive electrode and negative electrode of the stacked lithium-ion battery;

[0143] If the image processing device uses a preset corner recognition algorithm to identify the corners of the positive and negative electrodes in the target image, and marks each corner in the target image, then the number of positive and negative electrodes in the stacked lithium-ion battery can be determined based on the number of positive and negative electrode corners. The preset corner recognition algorithm can be a corner detection algorithm or a deep learning algorithm based on neural networks; no specific limitation is made here.

[0144] For example, such as Figure 16 As shown, the image processing device uses a corner detection algorithm to mark the corners in the target image, which can obtain a target image marked with the corners of the positive electrode and the corners of the negative electrode. Then, based on the number of corners of the positive electrode and the number of corners of the negative electrode, the number of positive electrode and the number of negative electrode in the stacked lithium-ion battery can be determined, that is, the number of positive electrode is 30 and the number of negative electrode is 31.

[0145] S1502, determine the distance between adjacent positive and negative electrodes in the stacked lithium-ion battery based on the positions of the corner points of the positive electrode and the negative electrode.

[0146] After the image processing device identifies the corner points of the positive and negative electrodes, it can determine the positions of the positive and negative electrode corner points. That is, it can obtain the position coordinates of any adjacent positive and negative electrode corner points. Then, the distance between adjacent positive and negative electrodes can be calculated according to the distance formula between the two points.

[0147] In one embodiment, the image processing device can calculate the distance between multiple adjacent positive and negative electrodes, and then calculate the average value of the distance as the distance between the positive and negative electrodes.

[0148] For example, such as Figure 16 As shown, the coordinates of the first negative electrode corner point from the left are (x1, y1), and the coordinates of the first positive electrode corner point from the left are (x2, y2). Therefore, the distance between the positive and negative electrodes can be calculated using the formula for the distance between two points.

[0149] S1503, Perform curve fitting on the corner points of the positive electrode and the corner points of the negative electrode respectively to obtain the corner point curves of the positive electrode and the corner point curves of the negative electrode.

[0150] S1504, calculate the alignment degree of the positive electrode of the stacked lithium-ion battery based on the positive electrode corner curve, and calculate the alignment degree of the negative electrode of the stacked lithium-ion battery based on the negative electrode corner curve.

[0151] Once the image processing equipment identifies the corner points of the positive and negative electrodes, curve fitting can be performed on these corner points to obtain the corner point curves for the positive and negative electrodes, which in turn yield the positive and negative electrode edge curves of the stacked lithium-ion battery. Since the corner point curves of the positive and negative electrodes characterize the distribution of the positive electrode edge contours, respectively, the alignment of the stacked lithium-ion battery can be determined based on these curves. Specifically, the alignment of the positive electrode is calculated using the corner point curves, and the alignment of the negative electrode is calculated using the corner point curves.

[0152] In one embodiment, the image processing device can calculate the difference between the maximum and minimum values ​​of the positive electrode corner curve, which can characterize the degree of fluctuation of the positive electrode corner curve as the positive electrode alignment. Similarly, the image processing device can calculate the difference between the maximum and minimum values ​​of the negative electrode corner curve, which can characterize the degree of fluctuation of the negative electrode corner curve as the negative electrode alignment.

[0153] For example, such as Figure 17 As shown, the image processing device performs curve fitting on the corner points of the positive and negative electrodes in the target image, obtaining curve 1701 for the positive electrode corner and curve 1702 for the negative electrode corner. Then, the maximum value X of curve 1701 for the positive electrode corner can be calculated. max and minimum value X min The difference can characterize the fluctuation of the positive electrode corner curve 1701, serving as the alignment degree of the positive electrode in a stacked lithium-ion battery. Similarly, the maximum value Y of the negative electrode corner curve 1702 can be calculated. max and minimum value Y min The difference can characterize the fluctuation of the negative electrode corner curve 1702, and serve as the alignment of the negative electrode in a stacked lithium-ion battery.

[0154] As can be seen, in this embodiment, the image processing device can determine the number of positive and negative electrodes in the stacked lithium-ion battery based on the corner points of the positive and negative electrodes. Based on the positions of the corner points of the positive and negative electrodes, it determines the distance between adjacent positive and negative electrodes in the stacked lithium-ion battery. Curve fitting is then performed on the corner points of the positive and negative electrodes to obtain the corner point curves for the positive and negative electrodes, respectively. The alignment of the positive electrodes in the stacked lithium-ion battery is calculated based on the positive electrode corner point curve, and the alignment of the negative electrodes is calculated based on the negative electrode corner point curve. By calculating performance parameters based on a clear image of the stacked lithium-ion battery, the problem of overlapping edge electrodes in the projected image can be solved, reducing the error in calculating performance parameters and thus improving the accuracy of stacked lithium-ion battery detection.

[0155] As one embodiment of this application, after determining the number of positive and negative electrodes, the distance between the positive and negative electrodes, and the alignment based on the corner points of the positive and negative electrodes, the above method may further include:

[0156] The number of positive and negative electrodes, the distance between positive and negative electrodes, and the alignment are compared with preset numbers of positive and negative electrodes, distance between positive and negative electrodes, and alignment, respectively, to obtain comparison results;

[0157] If the comparison results meet the preset qualified product conditions, the stacked lithium-ion battery is determined to be a qualified product.

[0158] Since the image processing device can determine the number of positive and negative electrodes, the distance between them, and the alignment of the stacked lithium-ion battery based on the corner points of the positive and negative electrodes, the image processing device can compare the determined number of positive and negative electrodes, the distance between them, and the alignment with standard values. That is, it can compare them with preset numbers of positive and negative electrodes, distances between them, and alignment respectively, and then obtain the comparison results.

[0159] If the comparison results meet the preset qualified product conditions, the stacked lithium-ion battery can be determined to be a qualified product; if the comparison results do not meet the preset qualified product conditions, the stacked lithium-ion battery can be determined to be a non-qualified product.

[0160] For example, an image processing device determines that the number of positive electrode plates in a stacked lithium-ion battery is 31 and the number of negative electrode plates is 26, the distance between the positive and negative electrode plates is L, and the alignment degree is Q. The preset number of positive electrode plates is 31 and the number of negative electrode plates is 30, and the distance between the positive and negative electrode plates is L. 标 And alignment Q 标 Among them, the number of positive electrode plates, the distance L between the positive and negative electrode plates, and the preset distance are L. 标 Alignment Q and preset alignment Q 标 The product meets the preset qualified product conditions. However, the difference between the number of negative electrode sheets and the preset number of negative electrode sheets is 4, which does not meet the preset qualified product conditions. Therefore, the stacked lithium-ion battery is a defective product.

[0161] As can be seen, in this embodiment, the image processing device can compare the number of positive and negative electrodes, the distance between the positive and negative electrodes, and their alignment with preset values ​​for the same parameters, to obtain comparison results. If the comparison results meet preset qualified product conditions, the stacked lithium-ion battery can be determined to be a qualified product. Since the performance parameters are calculated based on a clear image of the stacked lithium-ion battery, the problem of overlapping edge electrodes in the projected image can be solved, reducing the error in calculating performance parameters and thus improving the accuracy of stacked lithium-ion battery detection.

[0162] As one embodiment of this application, the above-described detection system may further include a workbench;

[0163] The positions of the radiation source and the detector are fixed. The stacked lithium-ion battery is placed on the worktable, which moves up and down along a line perpendicular to the line connecting the radiation source and the detector at preset intervals; or...

[0164] The stacked lithium-ion battery is placed on the workbench, the position of the workbench is fixed, and the radiation source and the detector move up and down along a line perpendicular to the line connecting the radiation source and the detector at a preset distance.

[0165] In the detection system provided in this application embodiment, the positions of the X-ray source and the detector are relatively fixed. Therefore, in order to obtain the projected images of multiple stacked lithium-ion batteries, the positions of the X-ray source and the detector, or the positions of the stacked lithium-ion batteries, can be changed synchronously.

[0166] In one embodiment, the positions of the radiation source and detector can be fixed, and a stacked lithium-ion battery can be placed on a worktable. The stacked lithium-ion battery then moves up and down along a predetermined distance perpendicular to the line connecting the radiation source and detector, moving with the worktable. This reduces the complexity of the mechanical structure and increases system stability. The worktable can be a lifting and rotating platform, and is not specifically limited thereto.

[0167] In other words, by keeping the positions of the X-ray source and the detector unchanged, the X-ray source transmits light through the stacked lithium-ion battery as it moves up and down. Thus, each time the stacked lithium-ion battery moves a preset distance with the worktable, the detector can capture a projected image of the battery at that position.

[0168] In one embodiment, a stacked lithium-ion battery can be placed on a worktable to fix the position of the worktable, and the X-ray source and detector can move up and down along a line perpendicular to the connection between the X-ray source and detector at a preset distance.

[0169] In other words, the position of the stacked lithium-ion battery remains unchanged, while the X-ray source and detector move synchronously up and down. In this way, each time the X-ray source and detector move at a preset distance, the image processing equipment can acquire a projected image of the stacked lithium-ion battery.

[0170] As can be seen, in this embodiment, the detection system only requires one set of X-ray source and detector. The positions of the X-ray source and detector can be fixed, and the stacked lithium-ion battery can be placed on the worktable and move up and down along the line perpendicular to the connection between the X-ray source and detector. Alternatively, the position of the worktable on which the stacked lithium-ion battery is placed can be fixed, and the X-ray source and detector can move up and down along the line perpendicular to the connection between the X-ray source and detector at a preset distance, thereby completing the acquisition of the projected image. The mechanical structure is simple, the detection system is easy to operate, has high accuracy, and low cost, and can be extended to other similar battery detection.

[0171] Figure 18 This is a specific flowchart of a stacked lithium-ion battery testing method provided in an embodiment of this application. The following is in conjunction with... Figure 18 The method for testing stacked lithium-ion batteries provided in this application is illustrated with examples. For instance... Figure 18 As shown, the stacked lithium-ion battery testing method provided in this application embodiment may include the following steps:

[0172] S1801, sequentially acquire several projected images along a preset direction;

[0173] In the process of inspecting stacked lithium-ion batteries using a detection system, the positions of the X-ray source and detector are fixed. The stacked lithium-ion battery can be placed on a worktable (such as a lifting and rotating stage). The X-ray source transmits light to a specific corner of the stacked lithium-ion battery; that is, the incident direction of the X-ray emitted by the source forms a certain angle with the preset edge of the stacked lithium-ion battery, typically set to 45°, and the rays are parallel to the positive and negative electrodes. The stacked lithium-ion battery moves up and down with the worktable. In this way, the detector can sequentially acquire several projected images along the preset direction, and the image processing equipment can then acquire these projected images.

[0174] S1802, segment, register, and fuse the projected image;

[0175] If the image processing device acquires multiple non-overlapping projection images, these multiple projection images can be segmented, registered, and fused to obtain a clear projection image.

[0176] S1803, perform linear CT reconstruction on the projection image;

[0177] If some of the projected images acquired by the image processing device have overlapping areas, and the positive and negative electrodes of some projected images are difficult to distinguish, and there are no non-overlapping areas, then linear CT reconstruction can be performed on the acquired projected images to obtain clear reconstructed images.

[0178] S1804, Process the fused projection image or the reconstructed image to obtain the number of positive and negative electrode plates, the distance between adjacent positive and negative electrode plates, and the alignment.

[0179] After the projected image is fused or reconstructed, a clear image of the stacked lithium-ion battery is obtained. The image processing device can process the image of the stacked lithium-ion battery to calculate the number of positive and negative electrode plates, the distance between adjacent positive and negative electrode plates, and the alignment.

[0180] S1805 compares the number of positive and negative electrode plates, the distance between adjacent positive and negative electrode plates, and the alignment of the stacked lithium-ion battery with the specified number of positive and negative electrode plates to determine whether the stacked lithium-ion battery is a qualified product.

[0181] After determining the number of positive and negative electrode plates, the distance between adjacent positive and negative electrode plates, and their alignment, the image processing equipment can compare these values ​​with those of a predefined stacked lithium-ion battery to determine whether the stacked lithium-ion battery is a qualified product. If the comparison results meet the preset qualified product conditions, the stacked lithium-ion battery is determined to be a qualified product; otherwise, it is determined to be a non-qualified product.

[0182] As can be seen, in the solution provided in this application embodiment, the image processing device applied to the detection system includes an X-ray source and a detector. The relative positions of the X-ray source and the detector are fixed. The angle between the incident direction of the X-ray source and the preset edge of the stacked lithium-ion battery is a preset angle. Whenever the relative position between the stacked lithium-ion battery and the X-ray source and the detector changes, the image processing device can acquire the projected image of the stacked lithium-ion battery collected by the detector, perform image processing on multiple projected images to obtain a target image, perform corner point recognition on the target image to obtain the positive electrode corner point and the negative electrode corner point in the target image, and determine the detection result based on the positive electrode corner point and the negative electrode corner point. Since the detector can acquire a projected image of a stacked lithium-ion battery each time the relative position between the stacked lithium-ion battery, the X-ray source, and the detector is changed, and multiple projected images at different positions are acquired, image processing can be performed on multiple projected images with non-overlapping areas, or image reconstruction can be performed on multiple projected images to obtain a clear projected image or reconstructed image for calculating performance parameters. This can solve the problem of overlapping edge electrodes in the projected image, reduce the error in calculating performance parameters, and thus improve the accuracy of stacked lithium-ion battery detection.

[0183] Figure 19 This is another specific flowchart of the stacked lithium-ion battery testing method provided in the embodiments of this application. The following is in conjunction with... Figure 19The method for testing stacked lithium-ion batteries provided in this application is illustrated with examples. For instance... Figure 19 As shown, the stacked lithium-ion battery testing method provided in this application embodiment may include the following steps:

[0184] S1901, through the detection system, acquires projection images of stacked lithium-ion batteries at n different locations;

[0185] In the process of testing stacked lithium-ion batteries using a detection system, the positions of the X-ray source and detector are fixed. The stacked lithium-ion battery can be placed on a worktable (such as a lifting and rotating table), and the stacked lithium-ion battery moves up and down with the worktable. In this way, the detector can collect n projected images of the stacked lithium-ion battery at different positions, and then the image processing equipment can acquire n projected images.

[0186] S1902, the clear positive and negative polar regions are segmented and extracted using image segmentation algorithms or deep learning segmentation algorithms based on various network models to obtain n non-overlapping regions;

[0187] If the image processing device acquires multiple projection images with non-overlapping regions, the image processing device can use image segmentation algorithms or deep learning segmentation algorithms based on various network models to segment and extract the clear positive and negative pole regions in each projection image, thereby obtaining n non-overlapping pole regions.

[0188] S1903, feature matching is performed on feature points of the n non-overlapping regions obtained by segmentation using an image registration algorithm;

[0189] Image processing equipment can extract feature points from n non-overlapping regions, and then use image registration algorithms and other methods to match the feature points of the n non-overlapping regions to obtain the matching result.

[0190] S1904, the registered n non-overlapping regions are fused into a single image of a stacked lithium-ion battery with clear positive and negative electrodes.

[0191] Image processing equipment can perform image fusion processing on the registered n non-overlapping region images to obtain a clear image of a stacked lithium-ion battery with positive and negative electrode plates.

[0192] S1905 uses image recognition algorithms or deep learning algorithms based on neural networks to identify and count the corner points of the positive and negative electrode sheets in images of stacked lithium-ion batteries.

[0193] Image processing equipment can use image recognition algorithms or deep learning algorithms based on neural networks to identify corner points in the fused image of stacked lithium-ion batteries, obtain the corner points of the positive electrode and the negative electrode, and determine the number of positive and negative electrode plates in the stacked lithium-ion battery based on the number of positive and negative electrode corner points.

[0194] S1906, based on the distance formula between two points, the distance between adjacent positive and negative electrode plates is obtained, and curve fitting is performed on the corner points of the positive and negative electrodes to obtain the corner point curves of the positive and negative electrodes of the stacked lithium-ion battery, and the alignment is calculated.

[0195] The image processing device can determine the positions of the positive and negative electrode corners based on the positive and negative electrode corners. Then, it can determine the position coordinates of any adjacent positive and negative electrode corners, and then calculate the distance between adjacent positive and negative electrodes according to the distance formula between the two points.

[0196] The image processing device performs curve fitting on the corner points of the positive electrode and the corner points of the negative electrode respectively, which can obtain the corner point curves of the positive electrode and the corner point curves of the negative electrode. The alignment degree of the positive electrode of the stacked lithium-ion battery is calculated based on the corner point curve of the positive electrode, and the alignment degree of the negative electrode of the stacked lithium-ion battery is calculated based on the corner point curve of the negative electrode.

[0197] S1907 compares the calculated number of positive and negative electrode plates, the distance between adjacent positive and negative electrode plates, and the alignment of the positive and negative electrode plates with the standard values ​​to determine whether the parameters fluctuate within the specified range, thereby determining whether the stacked lithium-ion battery is a qualified product.

[0198] If the performance parameters fluctuate within the specified range, the stacked lithium-ion battery can be judged as a qualified product; otherwise, it can be judged as a substandard product.

[0199] As can be seen, in the solution provided in this application embodiment, the image processing device applied to the detection system includes an X-ray source and a detector. The relative positions of the X-ray source and the detector are fixed. The angle between the incident direction of the X-ray source and the preset edge of the stacked lithium-ion battery is a preset angle. Whenever the relative position between the stacked lithium-ion battery and the X-ray source and the detector changes, the image processing device can acquire the projected image of the stacked lithium-ion battery collected by the detector, perform image processing on multiple projected images to obtain a target image, perform corner point recognition on the target image to obtain the positive electrode corner point and the negative electrode corner point in the target image, and determine the detection result based on the positive electrode corner point and the negative electrode corner point. Since the detector can acquire a projected image of a stacked lithium-ion battery each time the relative position between the stacked lithium-ion battery, the X-ray source, and the detector is changed, and multiple projected images with non-overlapping areas can be acquired, image processing can be performed on these multiple projected images to obtain a clear projected image for calculating performance parameters. This can solve the problem of overlapping edge electrodes in the projected image, reduce the error in calculating performance parameters, and thus improve the accuracy of stacked lithium-ion battery detection.

[0200] Figure 20 This is another specific flowchart of the stacked lithium-ion battery testing method provided in the embodiments of this application. The following is in conjunction with... Figure 20 The method for testing stacked lithium-ion batteries provided in this application is illustrated with examples. For instance... Figure 20 As shown, the stacked lithium-ion battery testing method provided in this application embodiment may include the following steps:

[0201] S2001, acquire projection images of stacked lithium-ion batteries at n different locations;

[0202] In the process of testing stacked lithium-ion batteries using a detection system, the positions of the X-ray source and detector are fixed. The stacked lithium-ion battery can be placed on a worktable (such as a lifting and rotating table), and the stacked lithium-ion battery moves up and down with the worktable. In this way, the detector can collect n projected images of the stacked lithium-ion battery at different positions, and then the image processing equipment can acquire n projected images.

[0203] S2002, Determine the geometric relationship of the detection system;

[0204] S2003, calculate the distance of each ray passing through the stacked lithium-ion battery using geometric relationships, and calculate the estimated value of the projection in the projection direction based on the distance;

[0205] Given the distance between the X-ray source and the detector, the distance between the X-ray source and the stacked lithium-ion battery, and the distance the stacked lithium-ion battery moves relative to the X-ray source each time, the image processing device can use the determined geometric relationships to calculate the passing length corresponding to each projected image, that is, it can calculate the length that each X-ray travels through the stacked lithium-ion battery.

[0206] Based on the passing length of each ray corresponding to the projected image, the image processing device can calculate the theoretical pixel value of the corresponding pixel in the detector for each ray, which is used as an estimate of the projection in that projection direction.

[0207] S2004, calculate the error between the actual pixel value of the collected pixel and the calculated theoretical pixel value of the pixel;

[0208] The detector collects the pixel value of the corresponding pixel point for each ray in the detector. The image processing device can calculate the difference between the pixel value of the pixel point and the corresponding theoretical pixel value, and then obtain the projection error corresponding to the pixel point.

[0209] S2005: Sum the errors to calculate the correction value of the image;

[0210] Image processing equipment can calculate the projection error corresponding to all pixels in each projected image. Then, the projection errors corresponding to each pixel in each projected image can be summed to obtain the correction value of each projected image.

[0211] S2006, Based on the correction value, the image to be reconstructed is corrected to obtain the reconstructed image of the stacked lithium-ion battery.

[0212] The image processing device can correct the projected image based on each correction value to obtain a corrected projected image, and then perform image reconstruction based on each corrected projected image to obtain a clear reconstructed image of the stacked lithium-ion battery.

[0213] S2007 uses image recognition algorithms or deep learning algorithms based on neural networks to identify and count the corner points of the positive and negative electrode sheets in the reconstructed image of the stacked lithium-ion battery.

[0214] S2008: The distance between adjacent positive and negative electrode plates is calculated based on the distance formula between two points, and curve fitting is performed on the corner points of the positive and negative electrodes to obtain the corner point curves of the positive and negative electrodes of the stacked lithium-ion battery, and the alignment is calculated.

[0215] S2009 compares the calculated number of positive and negative electrode plates, the distance between adjacent positive and negative electrode plates, and the alignment of the positive and negative electrode plates with the standard values ​​to determine whether the parameters fluctuate within the specified range, thereby determining whether the stacked lithium-ion battery is a qualified product.

[0216] The specific implementation methods of steps S2007-S2009 have been described in detail in the above embodiments, and therefore will not be repeated here.

[0217] As can be seen, in the solution provided in this application embodiment, the image processing device applied to the detection system includes an X-ray source and a detector. The relative positions of the X-ray source and the detector are fixed, and the angle between the incident direction of the X-ray source and the preset edge of the stacked lithium-ion battery is a preset angle. Each time the relative position between the stacked lithium-ion battery and the X-ray source and detector changes, the image processing device can acquire a projected image of the stacked lithium-ion battery collected by the detector. Image processing is performed on multiple projected images to obtain a target image. Corner point recognition is performed on the target image to obtain the positive electrode corner points and negative electrode corner points in the target image. Based on the positive electrode corner points and negative electrode corner points, the detection result is determined. Since the detector can acquire a projected image of one stacked lithium-ion battery each time the relative position between the stacked lithium-ion battery and the X-ray source and detector changes, and multiple projected images at different positions are acquired, image reconstruction can be performed on multiple projected images to obtain a clear reconstructed image for calculating performance parameters. This solves the problem of overlapping edge electrodes in the projected image, reduces the error in calculating performance parameters, and thus improves the accuracy of stacked lithium-ion battery detection.

[0218] Corresponding to the above-described stacked lithium-ion battery method, this application also provides a stacked lithium-ion battery device. The following describes the stacked lithium-ion battery device provided by this application.

[0219] like Figure 21 As shown, a stacked lithium-ion battery detection device is used in an image processing device within a detection system. The detection system further includes an X-ray source and a detector. The relative positions of the X-ray source and the detector are fixed. The angle between the incident direction of the X-ray source and a preset edge of the stacked lithium-ion battery is a preset angle. The device includes:

[0220] The projection image acquisition module 2110 is used to acquire the projection image of the stacked lithium-ion battery collected by the detector every time the relative position between the stacked lithium-ion battery and the radiation source and the detector is changed.

[0221] The target image acquisition module 2120 is used to perform image processing on multiple projection images to obtain a target image;

[0222] The corner acquisition module 2130 is used to perform corner recognition on the target image to obtain the positive electrode corner and the negative electrode corner in the target image;

[0223] The detection result determination module 2140 is used to determine the detection result based on the corner points of the positive electrode and the corner points of the negative electrode.

[0224] As can be seen, in the solution provided in this application embodiment, the image processing device applied to the detection system includes an X-ray source and a detector. The relative positions of the X-ray source and the detector are fixed. The angle between the incident direction of the X-ray source and the preset edge of the stacked lithium-ion battery is a preset angle. Whenever the relative position between the stacked lithium-ion battery and the X-ray source and the detector changes, the image processing device can acquire the projected image of the stacked lithium-ion battery collected by the detector, perform image processing on multiple projected images to obtain a target image, perform corner point recognition on the target image to obtain the positive electrode corner point and the negative electrode corner point in the target image, and determine the detection result based on the positive electrode corner point and the negative electrode corner point. Since the detector can acquire a projected image of a stacked lithium-ion battery each time the relative position between the stacked lithium-ion battery, the X-ray source, and the detector is changed, and multiple projected images at different positions are acquired, image processing can be performed on multiple projected images with non-overlapping areas, or image reconstruction can be performed on multiple projected images to obtain a clear image of the stacked lithium-ion battery for calculating performance parameters. This can solve the problem of overlapping edge electrodes in the projected image, reduce the error in calculating performance parameters, and thus improve the accuracy of stacked lithium-ion battery detection.

[0225] As one embodiment of this application, the target image acquisition module 2120 described above may include:

[0226] The first acquisition submodule is used to segment the pole region in each projection image using a preset segmentation algorithm to obtain the non-overlapping region of the pole in each projection image, wherein the non-overlapping region is the region in the projection image where the edges of the poles do not overlap.

[0227] The second acquisition submodule is used to determine the fusion position relationship of the non-overlapping region in the target image based on the feature points of the non-overlapping region, wherein the feature points are points on the edge of the pole piece in the non-overlapping region;

[0228] The third acquisition submodule is used to perform image fusion processing on the non-overlapping regions based on the fusion positional relationship to obtain the target image.

[0229] As one embodiment of this application, the second acquisition submodule described above may include:

[0230] A feature point extraction unit is used to extract feature points in non-overlapping regions of each projection image;

[0231] The matching result acquisition unit is used to match the feature points of each non-overlapping region to obtain the matching result, wherein the matching result indicates whether the feature points of each non-overlapping region correspond to the same point on the edge of the electrode.

[0232] The fusion position relationship determination unit is used to determine the fusion position relationship of each non-overlapping region in the target image based on the matching result and the position of the feature point in the corresponding non-overlapping region.

[0233] As one embodiment of this application, the target image acquisition module 2120 described above may include:

[0234] The first calculation submodule is used to calculate the projection error of each pixel in the projection image for each projection image, wherein the projection error represents the difference between the pixel value of the pixel in the projection image and the theoretical pixel value;

[0235] The second calculation submodule is used to calculate the projection error corresponding to the projection image based on the projection error corresponding to each pixel point included in each projection image.

[0236] The fourth acquisition submodule is used to correct the projected image based on the projection error corresponding to each image projection, so as to obtain the corrected projected image.

[0237] The fifth acquisition submodule is used to reconstruct the target image based on each corrected projection image.

[0238] As one embodiment of this application, the first computing submodule may include:

[0239] The first calculation unit is used to calculate the passage length corresponding to each projected image based on the distance between the radiation source and the detector, the distance between the radiation source and the stacked lithium-ion battery, and the distance the stacked lithium-ion battery moves relative to the radiation source each time. The passage length is the length that the radiation emitted by the radiation source travels through the stacked lithium-ion battery.

[0240] The second calculation unit is used to calculate the theoretical pixel value of the pixel point corresponding to each ray in the detector for each projected image, based on the passing length corresponding to each ray in the projected image.

[0241] The first acquisition unit is used to calculate the difference between the pixel value of each pixel and the corresponding theoretical pixel value for each projected image, so as to obtain the projection error corresponding to each pixel in the projected image.

[0242] As one embodiment of this application, the second calculation submodule described above may include:

[0243] The second acquisition unit is used to sum the projection errors corresponding to each pixel in each projected image to obtain the projection error corresponding to the projected image.

[0244] As one embodiment of this application, the detection result determination module 2140 may include:

[0245] The detection parameter determination submodule is used to determine the detection parameters of the stacked lithium-ion battery based on the corner points of the positive electrode and the negative electrode. The detection parameters include at least one of the following: the number of positive and negative electrodes in the stacked lithium-ion battery, the distance between the positive and negative electrodes, and the alignment of the target electrode. The target electrode includes both positive and negative electrodes. The alignment of the target electrode is calculated based on a corner point curve, which is obtained by curve fitting of the corner points of the target electrode.

[0246] As one embodiment of this application, the above-described detection system may further include a workbench;

[0247] The positions of the radiation source and the detector are fixed. The stacked lithium-ion battery is placed on the worktable, which moves up and down along a line perpendicular to the line connecting the radiation source and the detector at preset intervals; or...

[0248] The stacked lithium-ion battery is placed on the workbench, the position of the workbench is fixed, and the radiation source and the detector move up and down along a line perpendicular to the line connecting the radiation source and the detector at a preset distance.

[0249] This application also provides a detection system, such as... Figure 5 As shown, the detection system includes an image processing device 503, an X-ray source 501, and a detector 502. The relative positions of the X-ray source 501 and the detector 502 are fixed. The angle between the incident direction of the X-ray source 501 and a preset edge of the stacked lithium-ion battery is a preset angle, wherein:

[0250] The radiation source 501 is used to emit radiation that transmits through the stacked lithium-ion battery.

[0251] The detector 502 is used to acquire images of the stacked lithium-ion battery;

[0252] The image processing device 503 is used to acquire a projected image of the stacked lithium-ion battery collected by the detector each time the relative position between the stacked lithium-ion battery, the radiation source, and the detector is changed; to process multiple projected images to obtain a target image; to identify corner points in the target image to obtain the positive electrode corner points and negative electrode corner points in the target image; and to determine the detection result based on the positive electrode corner points and negative electrode corner points.

[0253] As can be seen, in the solution provided in this application embodiment, the image processing device applied to the detection system includes an X-ray source and a detector. The relative positions of the X-ray source and the detector are fixed. The angle between the incident direction of the X-ray source and the preset edge of the stacked lithium-ion battery is a preset angle. Whenever the relative position between the stacked lithium-ion battery and the X-ray source and the detector changes, the image processing device can acquire the projected image of the stacked lithium-ion battery collected by the detector, perform image processing on multiple projected images to obtain a target image, perform corner point recognition on the target image to obtain the positive electrode corner point and the negative electrode corner point in the target image, and determine the detection result based on the positive electrode corner point and the negative electrode corner point. Since the detector can acquire a projected image of a stacked lithium-ion battery each time the relative position between the stacked lithium-ion battery, the X-ray source, and the detector is changed, and multiple projected images at different positions are acquired, image processing can be performed on multiple projected images with non-overlapping areas, or image reconstruction can be performed on multiple projected images to obtain a clear image of the stacked lithium-ion battery for calculating performance parameters. This can solve the problem of overlapping edge electrodes in the projected image, reduce the error in calculating performance parameters, and thus improve the accuracy of stacked lithium-ion battery detection.

[0254] As one embodiment of this application, the image processing device 503 can be specifically used to segment the pole region in each projected image using a preset segmentation algorithm to obtain the non-overlapping region of the pole in each projected image, wherein the non-overlapping region is the region in the projected image where the pole edges do not overlap, and to determine the fusion position relationship of the non-overlapping region in the target image based on the feature points of the non-overlapping region, wherein the feature points are the points on the pole edges in the non-overlapping region, and to perform image fusion processing on the non-overlapping region based on the fusion position relationship to obtain the target image.

[0255] As one embodiment of this application, the image processing device 503 can be specifically used to extract feature points of non-overlapping regions in each projected image, match the feature points of each non-overlapping region to obtain a matching result, and determine the fusion position relationship of each non-overlapping region in the target image based on the matching result and the position of the feature points in the corresponding non-overlapping region; wherein, the matching result indicates whether the feature points of each non-overlapping region correspond to the same point on the edge of the electrode.

[0256] As one embodiment of this application, the image processing device 503 can be specifically used to calculate the projection error of each pixel in each projected image, wherein the projection error represents the difference between the pixel value and the theoretical pixel value of the pixel in the projected image. Based on the projection error corresponding to each pixel in each projected image, the projection error corresponding to the projected image is calculated. Based on the projection error corresponding to each projected image, the projected image is corrected to obtain a corrected projected image. Based on each corrected projected image, image reconstruction is performed to obtain a target image.

[0257] As one embodiment of this application, the image processing device 503 can be specifically used to calculate the passing length corresponding to each projected image based on the distance between the radiation source and the detector, the distance between the radiation source and the stacked lithium-ion battery, and the distance the stacked lithium-ion battery moves relative to the radiation source each time. The passing length is the length that the radiation emitted by the radiation source travels through the stacked lithium-ion battery. For each projected image, based on the passing length corresponding to each radiation in the projected image, the theoretical pixel value of the pixel corresponding to each radiation in the detector is calculated. For each projected image, the difference between the pixel value of each pixel and the corresponding theoretical pixel value is calculated to obtain the projection error corresponding to each pixel in the projected image.

[0258] As one embodiment of this application, the image processing device 503 can be specifically used to sum the projection errors corresponding to each pixel in the projected image for each projected image, so as to obtain the projection error corresponding to the projected image.

[0259] As one embodiment of this application, the image processing device 503 can specifically be used to determine the detection parameters of the stacked lithium-ion battery based on the corner points of the positive electrode and the corner points of the negative electrode. The detection parameters include at least one of the following: the number of positive and negative electrodes of the stacked lithium-ion battery, the distance between the positive and negative electrodes, and the alignment of the target electrode. The target electrode includes both positive and negative electrodes. The alignment of the target electrode is calculated based on the corner point curve of the target electrode, which is obtained by curve fitting of the corner points of the target electrode.

[0260] As one embodiment of this application, the detection system further includes a workbench;

[0261] The positions of the radiation source and the detector are fixed. The stacked lithium-ion battery is placed on the worktable, which moves up and down along a line perpendicular to the line connecting the radiation source and the detector at preset intervals; or...

[0262] The stacked lithium-ion battery is placed on the workbench, the position of the workbench is fixed, and the radiation source and the detector move up and down along a line perpendicular to the line connecting the radiation source and the detector at a preset distance.

[0263] This application also provides an image processing device, such as... Figure 22 As shown, it includes:

[0264] Memory 2201 is used to store computer programs;

[0265] The processor 2202, when executing the program stored in the memory 2201, implements the steps of the stacked lithium-ion battery detection method described in any of the above embodiments.

[0266] As can be seen, in the solution provided in this application embodiment, the image processing device applied to the detection system includes an X-ray source and a detector. The relative positions of the X-ray source and the detector are fixed. The angle between the incident direction of the X-ray source and the preset edge of the stacked lithium-ion battery is a preset angle. Whenever the relative position between the stacked lithium-ion battery and the X-ray source and the detector is changed, the image processing device can acquire the projected image of the stacked lithium-ion battery collected by the detector. Multiple projected images are processed to obtain a target image. Corner point recognition is performed on the target image to obtain the positive electrode corner point and the negative electrode corner point in the target image. Based on the positive electrode corner point and the negative electrode corner point, the detection result is determined. Since the detector can acquire a projected image of a stacked lithium-ion battery each time the relative position between the stacked lithium-ion battery, the X-ray source, and the detector is changed, and multiple projected images at different positions are acquired, image processing can be performed on multiple projected images with non-overlapping areas, or image reconstruction can be performed on multiple projected images to obtain a clear image of the stacked lithium-ion battery for calculating performance parameters. This can solve the problem of overlapping edge electrodes in the projected image, reduce the error in calculating performance parameters, and thus improve the accuracy of stacked lithium-ion battery detection.

[0267] Furthermore, the aforementioned electronic device may also include a communication bus and / or a communication interface, with the processor 2202, the communication interface, and the memory 2201 communicating with each other via the communication bus.

[0268] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0269] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0270] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0271] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0272] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the steps of any of the above-described stacked lithium-ion battery detection methods.

[0273] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the stacked lithium-ion battery detection methods described in the above embodiments.

[0274] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a solid-state drive (SSD), etc.

[0275] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0276] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, systems, image processing devices, computer-readable storage media, and computer program products are basically similar to the method embodiments, and therefore the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0277] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A method for testing stacked lithium-ion batteries, characterized in that, An image processing device is applied to a detection system, the detection system further comprising an X-ray source and a detector, the relative positions of the X-ray source and the detector being fixed, and the angle between the incident direction of the X-ray source and a preset edge of the stacked lithium-ion battery being a preset angle, the method comprising: Each time the relative position between the stacked lithium-ion battery, the radiation source, and the detector is changed, a projection image of the stacked lithium-ion battery acquired by the detector is obtained. Image processing is performed on multiple projected images to obtain the target image; Corner point recognition is performed on the target image to obtain the positive electrode corner points and negative electrode corner points in the target image; The detection result is determined based on the corner points of the positive electrode and the negative electrode. The step of processing multiple projected images to obtain a target image includes: A preset segmentation algorithm is used to segment the pole piece regions in each projected image, obtaining non-overlapping regions of the pole pieces in each projected image. These non-overlapping regions are areas where the pole piece edges do not overlap. Based on feature points of these non-overlapping regions, the corresponding fusion position relationships in the target image are determined. These feature points are points on the pole piece edges within the non-overlapping regions. Based on these fusion position relationships, image fusion processing is performed on the non-overlapping regions to obtain the target image; or... Based on the distance between the radiation source and the detector, the distance between the radiation source and the stacked lithium-ion battery, and the distance the stacked lithium-ion battery moves relative to the radiation source each time, the passing length corresponding to each projected image is calculated, where the passing length is the length that the radiation emitted by the radiation source travels through the stacked lithium-ion battery. For each projected image, based on the passing length corresponding to each ray in the projected image, the theoretical pixel value of the pixel corresponding to each ray in the detector is calculated. For each projected image, the difference between the pixel value of each pixel and the corresponding theoretical pixel value is calculated to obtain the projection error corresponding to each pixel in the projected image. Based on the projection errors corresponding to each pixel in each projected image, the projection error corresponding to the projected image is calculated. Based on the projection error corresponding to each projected image, the projected image is corrected to obtain a corrected projected image. Based on each corrected projected image, image reconstruction is performed to obtain the target image.

2. The method according to claim 1, characterized in that, The step of determining the fusion position relationship of the non-overlapping regions in the target image based on the feature points of the non-overlapping regions includes: Extract feature points from the non-overlapping regions in each of the projected images; The feature points of each non-overlapping region are matched to obtain the matching result, wherein the matching result indicates whether the feature points of each non-overlapping region correspond to the same point on the edge of the electrode. Based on the matching results and the positions of the feature points in the corresponding non-overlapping regions, the fusion position relationship of each non-overlapping region in the target image is determined.

3. The method according to claim 1, characterized in that, The step of calculating the projection error corresponding to each projection image based on the projection error corresponding to each pixel in each projection image includes: For each projected image, the projection errors corresponding to each pixel in the projected image are summed to obtain the projection error corresponding to that projected image.

4. The method according to claim 1, characterized in that, The step of determining the detection result based on the corner points of the positive electrode and the corner points of the negative electrode includes: Based on the corner points of the positive electrode and the corner points of the negative electrode, the detection parameters of the stacked lithium-ion battery are determined. The detection parameters include at least one of the following: the number of positive and negative electrodes of the stacked lithium-ion battery, the distance between the positive and negative electrodes, and the alignment of the target electrode. The target electrode includes both positive and negative electrodes. The alignment of the target electrode is calculated based on the corner point curve of the target electrode, which is obtained by curve fitting of the corner points of the target electrode.

5. The method according to any one of claims 1-4, characterized in that, The detection system also includes a workbench; The positions of the radiation source and the detector are fixed. The stacked lithium-ion battery is placed on the worktable, which moves up and down along a line perpendicular to the line connecting the radiation source and the detector at preset intervals; or... The stacked lithium-ion battery is placed on the workbench, the position of the workbench is fixed, and the radiation source and the detector move up and down along a line perpendicular to the line connecting the radiation source and the detector at a preset distance.

6. A stacked lithium-ion battery testing device, characterized in that, An image processing device is applied to a detection system, the detection system further comprising an X-ray source and a detector, the relative positions of the X-ray source and the detector being fixed, and the angle between the incident direction of the X-ray source and a preset edge of the stacked lithium-ion battery being a preset angle, the device comprising: The projection image acquisition module is used to acquire the projection image of the stacked lithium-ion battery collected by the detector every time the relative position between the stacked lithium-ion battery, the radiation source, and the detector is changed. The target image acquisition module is used to process multiple projection images to obtain the target image. The corner acquisition module is used to identify corners in the target image to obtain the positive electrode corners and negative electrode corners in the target image. The detection result determination module is used to determine the detection result based on the corner points of the positive electrode and the corner points of the negative electrode. Specifically, the target image acquisition module is used to segment the pole piece regions in each projected image using a preset segmentation algorithm to obtain non-overlapping regions of the pole pieces in each projected image, wherein the non-overlapping regions are regions in the projected images where the pole piece edges do not overlap; based on the feature points of the non-overlapping regions, determine the corresponding fusion position relationship of the non-overlapping regions in the target image, wherein the feature points are points on the pole piece edges in the non-overlapping regions; based on the fusion position relationship, perform image fusion processing on the non-overlapping regions to obtain the target image; or, Based on the distance between the radiation source and the detector, the distance between the radiation source and the stacked lithium-ion battery, and the distance the stacked lithium-ion battery moves relative to the radiation source each time, the passing length corresponding to each projected image is calculated, where the passing length is the length that the radiation emitted by the radiation source travels through the stacked lithium-ion battery. For each projected image, based on the passing length corresponding to each ray in the projected image, the theoretical pixel value of the pixel corresponding to each ray in the detector is calculated. For each projected image, the difference between the pixel value of each pixel and the corresponding theoretical pixel value is calculated to obtain the projection error corresponding to each pixel in the projected image. Based on the projection errors corresponding to each pixel in each projected image, the projection error corresponding to the projected image is calculated. Based on the projection error corresponding to each projected image, the projected image is corrected to obtain a corrected projected image. Based on each corrected projected image, image reconstruction is performed to obtain the target image.

7. An image processing device, characterized in that, include: Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the method described in any one of claims 1-5.

Images