Battery detection method, system and device based on combined welding and storage medium

Abstract

The invention discloses a battery detection method, system and device based on joint welding and a storage medium. The method comprises the following steps: acquiring an image of a welded battery; extracting tab size data in the battery image; inputting the battery image into a target detection model to obtain welding mark area features; inputting the welding print area features into a semantic segmentation model, and segmenting the welding print area to obtain welding spot data; and analyzing the tab size data and the welding spot data according to a preset detection logic, grading the welded battery, and determining the quality and the grade of the battery. According to the method, the battery image is processed through the target detection model and the semantic segmentation model, so that the image processing precision and accuracy are improved; meanwhile, the quality of the battery is evaluated and graded through the preset detection logic, so that the accuracy of battery detection is improved. The method can be widely applied to the technical field of intelligent manufacturing.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing technology, and in particular to a method, system, device, and storage medium for testing batteries after welding. Background Technology

[0002] Cylindrical batteries typically have two layers of tabs, one wide and one narrow. The manufacturing process requires welding the two layers of tabs together, followed by quality inspection of the welded battery.

[0003] When the welding quality is good and the tabs are clean, traditional algorithms can meet the application requirements. However, the incoming materials vary, the welding quality varies, the background texture of the tabs is different, and slight deformation of the tabs can cause the image to be too bright or too dark. It is difficult to achieve accurate detection by relying solely on traditional algorithms. When the image quality is poor, the detection accuracy is not high, and it is easy to over-detect a large number of products, resulting in a lot of waste. Summary of the Invention

[0004] The purpose of this invention is to at least partially solve one of the technical problems existing in the prior art.

[0005] Therefore, the purpose of this invention is to provide a high-precision method, system, device, and storage medium for testing batteries after welding.

[0006] To achieve the above-mentioned technical objectives, one aspect of this invention provides a battery detection method based on post-welding, comprising the following steps: acquiring a post-welding battery image; extracting tab size data from the battery image; inputting the battery image into a target detection model to obtain solder area features; inputting the solder area features into a semantic segmentation model to segment the solder area and obtain solder joint data; analyzing the tab size data and the solder joint data according to a preset detection logic, and classifying the post-welding battery to determine its quality and grade. This application improves the precision and accuracy of image processing by processing the battery image using a target detection model and a semantic segmentation model; simultaneously, the preset detection logic for battery quality assessment and grading helps improve the accuracy of battery detection.

[0007] In some embodiments, extracting the tab size data from the battery image includes:

[0008] The algorithm of finding straight lines with calipers is used to determine the upper edge line segment of the electrode tab, the edge line segment of the cell separator, the left edge line segment of the electrode tab, and the right edge line segment of the electrode tab.

[0009] The length of the electrode tab is determined by measuring the line segment from the upper edge to the edge of the cell separator.

[0010] The tab width is determined by line measurement based on the left and right edge segments.

[0011] In some embodiments, inputting the battery image into a target detection model to obtain solder joint area features includes:

[0012] The target detection model is used to determine the weld stamp data, which includes the center point coordinates, width, height, and angle of the weld stamp.

[0013] Based on the distance from the center point coordinates to the diaphragm and the height, determine the first distance of the diaphragm from the top of the solder joint and the second distance of the diaphragm from the bottom of the solder joint;

[0014] Based on the first distance, the second distance, and the solder stamp data, the characteristics of the solder stamp area are determined.

[0015] In some embodiments, inputting the solder stamp region features into a semantic segmentation model to segment the solder stamp region and obtain solder joint data includes:

[0016] The features of the solder stamp area are subjected to rectangular structural element expansion processing to extract the solder stamp features;

[0017] The solder mark features are semantically segmented to obtain each solder joint region;

[0018] The area and grayscale data of each solder joint are determined by performing an opening operation on the solder joint area using a rectangular structuring element, and the number of solder joints is counted to obtain the solder joint data.

[0019] In some embodiments, the step of analyzing the tab size data and the solder joint data according to a preset detection logic, and classifying the welded battery to determine its quality and grade, includes:

[0020] The electrode size level is determined based on the electrode size and the electrode size range threshold;

[0021] The diaphragm weld joint size level is determined based on the diaphragm weld joint size and the threshold range of diaphragm weld joint size;

[0022] The solder mark size level is determined based on the solder mark size and the solder mark size range threshold.

[0023] The solder joint area level is determined based on the solder joint area and the threshold range of solder joint area.

[0024] The quality and grade of the battery are determined based on the tab size grade, the separator solder joint size grade, the solder mark size grade, and the solder joint area grade.

[0025] In some embodiments, the method further includes:

[0026] Displays the battery detection interface;

[0027] Upon receiving a trigger operation on the first control on the battery detection interface, the grading parameter setting interface is displayed;

[0028] The editing or selection operation received on the second control group in the graded parameter setting interface determines the graded interval threshold; the second control group includes controls corresponding to each interval threshold in the graded interval threshold.

[0029] Based on the threshold values ​​of the graded intervals, a preset detection logic is determined.

[0030] In some embodiments, determining the grading interval threshold based on the editing or selection operation on the second control group received at the grading parameter setting interface includes:

[0031] Based on the received editing operations for the interval thresholds corresponding to each control in the second control group, and based on the input interval endpoint thresholds, the hierarchical interval thresholds are determined.

[0032] Alternatively, based on the received selection operation for the interval threshold corresponding to each control in the second control group, the current interval endpoint threshold is updated according to the selected threshold adjustment degree; and the updated interval endpoint threshold is used as the new hierarchical interval threshold.

[0033] On the other hand, embodiments of the present invention propose a battery testing system based on post-welding testing, comprising:

[0034] The first module is used to acquire images of the battery after welding.

[0035] The second module is used to extract the tab size data from the battery image;

[0036] The third module is used to input the battery image into the target detection model to obtain the solder area features; and input the solder area features into the semantic segmentation model to segment the solder area and obtain the solder joint data.

[0037] The fourth module is used to analyze the tab size data and the solder joint data according to the preset detection logic, and to classify the battery after welding to determine the quality and grade of the battery.

[0038] On the other hand, embodiments of the present invention provide a battery testing device based on the welded battery, comprising:

[0039] At least one processor;

[0040] At least one memory for storing at least one program;

[0041] When the at least one program is executed by the at least one processor, the at least one processor implements the above-described battery detection method based on post-welding.

[0042] On the other hand, embodiments of the present invention provide a storage medium storing a processor-executable program, which, when executed by a processor, is used to implement the above-described battery detection method based on post-welding.

[0043] The embodiments of this application include at least the following beneficial effects: The method provided by the embodiments of this invention includes: acquiring a battery image after welding; extracting tab size data from the battery image; inputting the battery image into a target detection model to obtain solder area features; inputting the solder area features into a semantic segmentation model to segment the solder area and obtain solder joint data; analyzing the tab size data and the solder joint data according to a preset detection logic, and grading the welded battery to determine the quality and grade of the battery. This application processes the battery image using a target detection model and a semantic segmentation model, improving the precision and accuracy of image processing; simultaneously, the preset detection logic for battery quality assessment and grading helps improve the accuracy of battery detection. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following description is provided with accompanying drawings of the relevant technical solutions in the embodiments of the present invention or the prior art. It should be understood that the accompanying drawings described below are only for the purpose of clearly illustrating some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0045] Figure 1 This is a schematic diagram of the electrode tabs before welding.

[0046] Figure 2 This is a schematic diagram of the actual electrode tabs after welding.

[0047] Figure 3 A good image of the battery after welding;

[0048] Figure 4 An image of a battery after welding with a complex background texture;

[0049] Figure 5 An image of the battery after welding, showing a mixed welding area and background;

[0050] Figure 6 Image of a battery after welding, showing uneven brightness and darkness due to electrode deformation;

[0051] Figure 7 This is a flowchart illustrating an embodiment of the battery testing method based on post-welding provided by the present invention.

[0052] Figure 8 This is a schematic flowchart of another embodiment of the battery testing method based on post-welding provided by the present invention;

[0053] Figure 9 A schematic diagram of the dimension markings for one embodiment of the electrode size data provided by the present invention;

[0054] Figure 10 A schematic diagram illustrating one embodiment of the welding stamp area provided by the present invention;

[0055] Figure 11 A schematic diagram illustrating one embodiment of the expansion of the welding stamp area provided by the present invention;

[0056] Figure 12 A schematic diagram illustrating one embodiment of the welding stamp area cutout provided by the present invention;

[0057] Figure 13 A schematic diagram illustrating the effect of one embodiment of the solder joint semantic segmentation provided by the present invention;

[0058] Figure 14 A schematic diagram illustrating the effect of an embodiment of the solder joint area opening operation provided by the present invention;

[0059] Figure 15 A schematic diagram illustrating one embodiment of the detection results provided by the present invention;

[0060] Figure 16 A schematic diagram of an embodiment of the battery testing system based on post-welding provided by the present invention;

[0061] Figure 17 This is a schematic diagram of one embodiment of the battery testing device based on welding provided by the present invention. Detailed Implementation

[0062] The embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. The step numbers in the following embodiments are set only for ease of explanation, and there is no limitation on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0063] This invention relates to the field of visual inspection technology for intelligent manufacturing equipment of lithium batteries, and specifically to a visual inspection method and device after welding.

[0064] Some cylindrical batteries typically have two layers of tabs, one wide and one narrow, as shown in the image before welding. Figure 1 As shown. The process requires welding the two layers of tabs together, and the image after welding is shown below. Figure 2 As shown.

[0065] For visual inspection after welding, one embodiment mentions performing binarization processing to extract the number of weld points, welding area parameters, and overlapping area positions. Parameters such as the area, size, and grayscale of the weld points characterize their quality. If any one of the weld point count, welding area parameters, or overlapping area parameters fails to meet the requirements, the weld is considered unqualified. This solution only mentions simple methods like binarization processing and does not demonstrate the application of AI algorithms. Relying solely on traditional algorithms makes it difficult to adapt to complex background textures. However, this application combines traditional algorithms with AI algorithms, leveraging the advantages of AI algorithms to accurately calculate relevant parameters such as weld points and welding areas even with complex background textures and poor welding quality.

[0066] When the welding quality is good and the tabs are clean, traditional algorithms can meet the application requirements. However, the incoming materials vary, the welding quality is inconsistent, the background texture of the tabs is different, and slight deformation of the tabs can cause the image to be too bright or too dark. It is difficult to achieve accurate detection by relying solely on traditional algorithms. When the image quality is poor, it is easy to over-detect a large number of products, resulting in a lot of waste.

[0067] By combining deep learning with traditional algorithms, the advantages of each are leveraged, resulting in more accurate detection not only in cases of good imaging but also in cases of poor imaging quality. Then, parameters are set according to customer needs to classify the welded products, reducing product waste and improving the final product yield for the customer.

[0068] Good weld images such as Figure 3 As shown, a typical complex imaging weld image is as follows: Figure 3 , Figure 4 , Figure 5 , Figure 6 As shown. Figure 6 (a) and (b) are welding images showing uneven brightness and darkness of the electrode deformation under two different embodiments.

[0069] The following describes in detail, with reference to the accompanying drawings, the battery testing method and system based on the soldered battery according to the embodiments of the present invention. First, the battery testing method based on the soldered battery according to the embodiments of the present invention will be described with reference to the accompanying drawings.

[0070] Reference Figure 7 and Figure 8 This invention provides a method for detecting batteries after soldering. This method can be applied to terminals, servers, or software running on either terminal or server. Terminals can be tablets, laptops, desktop computers, etc., but are not limited to these. Servers can be independent physical servers, server clusters or distributed systems composed of multiple physical servers, or cloud servers providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The battery detection method based on soldering in this invention mainly includes the following steps:

[0071] S100: Obtain an image of the battery after welding;

[0072] S200: Extract the tab size data from the battery image;

[0073] S300: Input the battery image into the target detection model to obtain solder area features; input the solder area features into the semantic segmentation model to segment the solder area and obtain solder joint data;

[0074] S400: Based on the preset detection logic, analyze the tab size data and the solder joint data, and classify the welded battery to determine the quality and grade of the battery.

[0075] In some possible implementations, the preset detection logic in this application can determine the pass / fail status of multiple parameter measurement results in the tab size data, and simultaneously determine the pass / fail status of multiple parameter measurement results in the solder joint data. The preset detection logic can adjust various thresholds in the pass / fail judgment, and can also set the grading method in the grading process. It is understood that the acquisition of battery images in this application follows industry standards or is performed according to user-defined standards. The imaging scheme involved in this application mainly includes an area array camera, an FA lens, a dome light source, a light source controller, and an industrial control computer. The camera is triggered by the hardware to capture the image, obtaining the welded image to be processed, i.e., the battery image.

[0076] In some embodiments, extracting the tab size data from the battery image includes:

[0077] The algorithm of finding straight lines with calipers is used to determine the upper edge line segment of the electrode tab, the edge line segment of the cell separator, the left edge line segment of the electrode tab, and the right edge line segment of the electrode tab.

[0078] The length of the electrode tab is determined by measuring the line segment from the upper edge to the edge of the cell separator.

[0079] The tab width is determined by line measurement based on the left and right edge segments.

[0080] The electrode size data extracted in this application includes electrode width and electrode length. Based on the original image, traditional algorithms are used to extract the edges of each target, and then the electrode width and length are obtained. Of course, the parameter type of the electrode size data can be adjusted according to customer needs; this application does not impose specific limitations. (See reference...) Figure 9 As shown, 910 represents the tab width, 920 represents the tab length, 930 represents the solder mark width, 940 represents the solder mark height, 950 represents the distance between the diaphragm and the upper edge of the solder joint, and 960 represents the distance between the diaphragm and the lower edge of the solder joint.

[0081] In some embodiments, inputting the battery image into a target detection model to obtain solder joint area features includes:

[0082] The target detection model is used to determine the weld stamp data, which includes the center point coordinates, width, height, and angle of the weld stamp.

[0083] Based on the distance from the center point coordinates to the diaphragm and the height, determine the first distance of the diaphragm from the top of the solder joint and the second distance of the diaphragm from the bottom of the solder joint;

[0084] Based on the first distance, the second distance, and the solder stamp data, the characteristics of the solder stamp area are determined.

[0085] This application extracts solder stamp data, including the location, angle, length, and width of the solder stamp.

[0086] The original image was processed using AI Model 1 (i.e., the target detection model) to extract the location, angle, length, and width of the solder joint. AI Model 1 is a directional target detection model, which, compared to conventional target detection models, can adapt to situations where the target is tilted to varying degrees, making the detection and positioning more accurate. The position of the cell end face was extracted using a traditional algorithm, and then the distances from the separator to the top edge of the solder joint and from the separator to the bottom edge of the solder joint (i.e., the first distance and the second distance) were calculated.

[0087] In some embodiments, inputting the solder stamp region features into a semantic segmentation model to segment the solder stamp region and obtain solder joint data includes:

[0088] The features of the solder stamp area are subjected to rectangular structural element expansion processing to extract the solder stamp features;

[0089] The solder mark features are semantically segmented to obtain each solder joint region;

[0090] The area and grayscale data of each solder joint are determined by performing an opening operation on the solder joint area using a rectangular structuring element, and the number of solder joints is counted to obtain the solder joint data.

[0091] This application extracts solder joint quality data, including the number of solder joints and the quality of each solder joint. The solder joint area is dilated, and a local image is cropped. AI Model 2 is used in this local image to extract each solder joint. The number of solder joints, the area of ​​each solder joint, the mean, and the variance are calculated. Model 2 is a semantic segmentation model. Logically, Model 1 in this application is a special case of an object detection model, namely a directional object detection model. Model 2 is a semantic segmentation model. The difference lies in the annotation content. Furthermore, directional object detection models are primarily designed for object detection scenarios, typically outputting the bounding box of the object, orientation, category information, and confidence score. Speech segmentation models, on the other hand, generally output a mask for segmented objects, category information, and confidence score. Object detection is generally used for localization, while speech segmentation is mainly used to segment objects and obtain their contour regions.

[0092] In some embodiments, the step of analyzing the tab size data and the solder joint data according to a preset detection logic, and classifying the welded battery to determine its quality and grade, includes:

[0093] The electrode size level is determined based on the electrode size and the electrode size range threshold;

[0094] The diaphragm weld joint size level is determined based on the diaphragm weld joint size and the threshold range of diaphragm weld joint size;

[0095] The solder mark size level is determined based on the solder mark size and the solder mark size range threshold.

[0096] The solder joint area level is determined based on the solder joint area and the threshold range of solder joint area.

[0097] The quality and grade of the battery are determined based on the tab size grade, the separator solder joint size grade, the solder mark size grade, and the solder joint area grade.

[0098] This application's data analysis and judgment comprehensively assesses the tab size, solder joint specifications, and solder joint quality to determine whether the welded cell is OK or NG. The tab size includes tab length and tab width; the separator solder joint size includes the distance between the separator and the top edge of the solder joint and the distance between the separator and the bottom edge of the solder joint; the solder joint size includes solder joint height and solder joint width. Of course, those skilled in the art can set parameters for quality and grade judgment according to actual needs. The specific testing logic can be such that if one parameter fails, the test result is unqualified; or a weight can be assigned to each parameter, and the final result is determined based on the judgment result of each parameter and its weight.

[0099] In some embodiments, the method further includes:

[0100] Displays the battery detection interface;

[0101] Upon receiving a trigger operation on the first control on the battery detection interface, the grading parameter setting interface is displayed;

[0102] The editing or selection operation received on the second control group in the graded parameter setting interface determines the graded interval threshold; the second control group includes controls corresponding to each interval threshold in the graded interval threshold.

[0103] Based on the threshold values ​​of the graded intervals, a preset detection logic is determined.

[0104] This application provides a threshold adjustment method in the preset detection logic, which can adjust the number of levels in the grading and the interval endpoint values ​​of each interval threshold for level determination according to customer needs. The first control is the control that triggers the grading parameter settings, and the position and shape of the first control can be adjusted. The triggering operation can be a click operation, a selection operation, a touch operation, etc. The second control group is used to design the number of grading levels and the interval endpoint values ​​corresponding to each level.

[0105] In some embodiments, determining the grading interval threshold based on the editing or selection operation on the second control group received at the grading parameter setting interface includes:

[0106] Based on the received editing operations for the interval thresholds corresponding to each control in the second control group, and based on the input interval endpoint thresholds, the hierarchical interval thresholds are determined.

[0107] Alternatively, based on the received selection operation for the interval threshold corresponding to each control in the second control group, the current interval endpoint threshold is updated according to the selected threshold adjustment degree; and the updated interval endpoint threshold is used as the new hierarchical interval threshold.

[0108] Logically, the threshold values ​​at the two endpoints of each graded interval can be determined directly by inputting them into the second control group, or the threshold values ​​of the graded intervals can be updated by setting the threshold adjustment degree.

[0109] The detection method in this application will now be described in detail with reference to a specific embodiment:

[0110] The specific process for extracting electrode size data is as follows:

[0111] The algorithm of finding straight lines with calipers is used to obtain the upper edge line segment of the tab, the edge line segment of the cell separator, the left edge line segment of the tab, and the right edge line segment of the tab.

[0112] The length of the tab is obtained by measuring the line segment from the upper edge of the tab to the edge of the cell separator. The width of the tab is obtained by measuring the line segments at the left and right edges of the tab.

[0113] The calculation method for line measurement is as follows: the perpendicular distance from the midpoint of line segment L1 to L2 is denoted as d1, and the perpendicular distance from the midpoint of line segment L2 to L1 is denoted as d2. Then the distance between line segment L1 and line segment L2 is (d1+d2)*0.5.

[0114] The specific process for extracting solder stamp data is as follows:

[0115] The model for directional target detection is called to obtain the center point coordinates (x, y) of the weld, the width w of the weld, the height h of the weld, and the angle θ of the circumscribed rectangle of the weld. The distance from the center point (x, y) to the straight line of the diaphragm edge is denoted as d3. The distance from the diaphragm to the top edge of the weld is d3 + h * 0.5, and the distance from the diaphragm to the bottom edge of the weld is d3 - h * 0.5. θ is constrained to [-1°, 1°].

[0116] The specific process for extracting solder joint data is as follows:

[0117] First, obtain the area of ​​the solder mark, such as Figure 10 As shown.

[0118] Then, use a rectangular structuring element to expand the area. The size of the rectangular structuring element is 20*20. The expanded area is as follows: Figure 11 As shown.

[0119] Then, the image is cut out based on the original image and the expanded area of ​​the solder joint, such as... Figure 12 As shown.

[0120] Then, on the cut-out image, the regions of each solder joint are obtained through semantic segmentation, such as... Figure 13 As shown.

[0121] Then, an opening operation is performed using a rectangular structuring element with dimensions of 7x7, such as... Figure 14 As shown.

[0122] Then, the area and average grayscale value of each solder joint are calculated, and the number of solder joints is counted. This yields the solder joint data.

[0123] The data analysis and judgment process is as follows:

[0124] The dimensions of each electrode tab are checked for compliance based on its size specifications. The dimensions and location of each solder mark are checked for compliance. The number of solder joints, their area, and the average value are used to determine their compliance. Finally, the overall quality of the finished product is assessed. Figure 15 This illustrates a determination process.

[0125] For each measurement item / parameter, the customer will provide a specific acceptable range. For example, the acceptable range for tab length is [17, 20]. Therefore, if the currently measured tab length is 17.56mm, it is acceptable; if it is 16mm, it is unacceptable. The above example illustrates the process of grading into two levels (acceptable and unacceptable). This specific range can be set according to the manufacturer's process specifications. Generally, over-standard conditions are also considered, and the specifications are appropriately relaxed.

[0126] The grading specifications also expose parameters to the software interface for easy configuration, which can reduce over-killing, improve equipment yield, and increase output. For example, for tab width, another level can be set, with acceptable ranges of 3.8 to 4 and 6.5 to 6.7. Tab widths falling within this range are considered acceptable level one; those falling within the 4 to 6.5 range are qualified level one. This is equivalent to setting three grading ranges. The threshold values ​​for the two grading ranges are written through the second control group. Specifically, control 1 sets the threshold values ​​for the qualified range (4 and 6.5), and control 2 sets the threshold values ​​for the acceptable range (3.8, 4 and 6.5, 6.7). It is understood that control 1 and control 2 can determine the threshold values ​​by inputting specific values; alternatively, they can input / select the threshold adjustment degree, such as increasing by 0.2 or decreasing by 0.1. Furthermore, those skilled in the art can set more grading ranges as needed.

[0127] This application enables the classification of different levels of electrode data, solder stamp data, and solder joint data by the manufacturer, which can be entered into the vision software. This allows for the grading of post-weld quality, thereby improving the yield of equipment output and reducing waste.

[0128] This invention comprehensively utilizes directional target detection AI algorithms, semantic segmentation AI algorithms, caliper-based edge finding algorithms, morphological algorithms, and other technologies, combining traditional and AI algorithms to achieve accurate measurement of electrode tabs, accurate positioning and calculation of solder marks, and accurate segmentation of solder joints. This results in more complete and accurate electrode tab size data, solder mark data, and solder joint data, providing a better data foundation for accurate post-weld inspection and product grading based on the data, reducing over-killing, and improving equipment yield.

[0129] In summary, the method provided in this application includes: acquiring an image of the battery after welding; extracting tab size data from the battery image; inputting the battery image into a target detection model to obtain solder area features; inputting the solder area features into a semantic segmentation model to segment the solder area and obtain solder joint data; analyzing the tab size data and the solder joint data according to preset detection logic, and grading the welded battery to determine its quality and grade. This application processes the battery image using a target detection model and a semantic segmentation model, improving the precision and accuracy of image processing; simultaneously, the preset detection logic for battery quality assessment and grading helps improve the accuracy of battery detection.

[0130] Secondly, refer to the appendix Figure 16 A battery testing system based on post-welding is described according to an embodiment of the present invention. The system specifically includes:

[0131] The first module 610 is used to acquire images of the battery after welding.

[0132] The second module 620 is used to extract the tab size data from the battery image;

[0133] The third module 630 is used to input the battery image into the target detection model to obtain solder area features; and input the solder area features into the semantic segmentation model to segment the solder area to obtain solder joint data.

[0134] The fourth module 640 is used to analyze the tab size data and the solder joint data according to the preset detection logic, and to classify the battery after welding to determine the quality and grade of the battery.

[0135] It is evident that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0136] Reference Figure 17 This invention provides a battery testing device based on the welded battery, comprising:

[0137] At least one processor 710;

[0138] At least one memory 720 is used to store at least one program;

[0139] When the at least one program is executed by the at least one processor 710, the at least one processor 710 implements the battery detection method based on the soldered battery.

[0140] Similarly, the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0141] This invention also provides a computer-readable storage medium storing a processor-executable program, which, when executed by a processor, performs the aforementioned battery detection method based on post-welding.

[0142] Similarly, the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0143] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this invention are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is altered and sub-operations described as part of a larger operation are executed independently.

[0144] Furthermore, although the invention has been described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the invention. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional skill of an engineer. Therefore, those skilled in the art can implement the invention as set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the specific concepts disclosed are merely illustrative and not intended to limit the scope of the invention, which is determined by the full scope of the appended claims and their equivalents.

[0145] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several programs to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0146] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequential list of executable programs for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, a program execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can retrieve and execute a program from or in conjunction with such a program execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain, store, communicate, propagate, or transmit a program for use by or in conjunction with a program execution system, apparatus, or device.

[0147] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0148] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable program execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0149] In the foregoing description of this specification, references to terms such as "one embodiment," "another embodiment," or "some embodiments" indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of the present invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0150] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0151] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. A method for testing batteries after welding, characterized in that, Includes the following steps: Obtain images of the battery after welding; Extract the tab size data from the battery image; The battery image is input into a target detection model to obtain solder area features; the solder area features are then input into a semantic segmentation model to segment the solder area and obtain solder joint data. According to the preset detection logic, the electrode size data and the solder joint data are analyzed, and the battery after welding is graded to determine the quality and grade of the battery.

2. The battery testing method based on post-welding as described in claim 1, characterized in that, The extraction of tab size data from the battery image includes: The algorithm of finding straight lines with calipers is used to determine the upper edge line segment of the electrode tab, the edge line segment of the cell separator, the left edge line segment of the electrode tab, and the right edge line segment of the electrode tab. The length of the electrode tab is determined by measuring the line segment from the upper edge to the edge of the cell separator. The tab width is determined by line measurement based on the left and right edge segments.

3. The battery testing method based on post-welding as described in claim 1, characterized in that, The step of inputting the battery image into the target detection model to obtain the solder area features includes: The target detection model is used to determine the weld stamp data, which includes the center point coordinates, width, height, and angle of the weld stamp. Based on the distance from the center point coordinates to the diaphragm and the height, determine the first distance of the diaphragm from the top of the solder joint and the second distance of the diaphragm from the bottom of the solder joint; Based on the first distance, the second distance, and the solder stamp data, the characteristics of the solder stamp area are determined.

4. The battery testing method based on post-welding as described in claim 1, characterized in that, The step of inputting the features of the solder stamp area into a semantic segmentation model to segment the solder stamp area and obtain solder joint data includes: The features of the solder stamp area are subjected to rectangular structural element expansion processing to extract the solder stamp features; The solder mark features are semantically segmented to obtain each solder joint region; The area and grayscale data of each solder joint are determined by performing an opening operation on the solder joint area using a rectangular structuring element, and the number of solder joints is counted to obtain the solder joint data.

5. The battery testing method based on post-welding as described in claim 1, characterized in that, The step involves analyzing the tab size data and the solder joint data according to a preset detection logic, and classifying the welded battery to determine its quality and grade. This includes: The electrode size level is determined based on the electrode size and the electrode size range threshold; The diaphragm weld joint size level is determined based on the diaphragm weld joint size and the threshold range of diaphragm weld joint size; The solder mark size level is determined based on the solder mark size and the solder mark size range threshold. The solder joint area level is determined based on the solder joint area and the threshold range of solder joint area. The quality and grade of the battery are determined based on the tab size grade, the separator solder joint size grade, the solder mark size grade, and the solder joint area grade.

6. The battery testing method based on post-welding as described in claim 1, characterized in that, The method further includes: Displays the battery detection interface; Upon receiving a trigger operation on the first control on the battery detection interface, the grading parameter setting interface is displayed; The editing or selection operation received on the second control group in the graded parameter setting interface determines the graded interval threshold; the second control group includes controls corresponding to each interval threshold in the graded interval threshold. Based on the threshold values ​​of the graded intervals, a preset detection logic is determined.

7. The battery testing method based on post-welding as described in claim 6, characterized in that, The process of determining the grading interval threshold based on the editing or selection operation of the second control group received at the grading parameter setting interface includes: Based on the received editing operations for the interval thresholds corresponding to each control in the second control group, and based on the input interval endpoint thresholds, the hierarchical interval thresholds are determined. Alternatively, based on the received selection operation for the interval threshold corresponding to each control in the second control group, the current interval endpoint threshold is updated according to the selected threshold adjustment degree; and the updated interval endpoint threshold is used as the new hierarchical interval threshold.

8. A battery testing system based on post-welding testing, characterized in that, include: The first module is used to acquire images of the battery after welding. The second module is used to extract the tab size data from the battery image; The third module is used to input the battery image into the target detection model to obtain the solder area features; and input the solder area features into the semantic segmentation model to segment the solder area and obtain the solder joint data. The fourth module is used to analyze the tab size data and the solder joint data according to the preset detection logic, and to classify the battery after welding to determine the quality and grade of the battery.

9. A battery testing device based on post-welding testing, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the battery detection method based on the soldered battery as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a processor-executable program, characterized in that, The processor-executable program, when executed by the processor, is used to implement the battery detection method based on any one of claims 1 to 7.

Images