Battery detection method, battery detection device, apparatus, storage medium, and product

Abstract

The application relates to a battery detection method, a battery detection device, a device, a storage medium and a product. The method comprises the following steps: acquiring an image of a semi-finished battery obtained in a battery assembly process, acquiring feature information of a detection item in a region of interest in the image, and obtaining a defect detection result of the semi-finished battery based on the feature information of the detection item in the region of interest and a reference quantity. Since the detection item comprises at least one of an assembly part and an assembly area before the assembly part is installed, and the feature information of the detection item in the region of interest in the image is acquired, the feature information of the detection item in the region of interest and the reference quantity are used to detect the assembly part installation condition in the image, the defect detection result of the assembly part installation condition of the semi-finished battery is obtained, and the accuracy of the obtained defect detection result is improved.

Description

Technical Field

[0001] This application relates to the field of battery technology, and in particular to a battery testing method, battery testing equipment, apparatus, storage medium, and product. Background Technology

[0002] During battery production and assembly, various components are required. Due to the diverse shapes of these components, misassembly, omissions, and mixing of components can easily occur. These defects can lead to short circuits or irreparable problems in subsequent production processes, posing significant risks. Therefore, it is necessary to inspect the semi-finished batteries for misassembly, omissions, and mixing of components during the battery production and assembly process.

[0003] Currently, the method of manually visually inspecting semi-finished batteries for defects such as incorrect, missing, or mixed assembly of parts has the problem of low accuracy in defect detection. Summary of the Invention

[0004] Therefore, it is necessary to provide a battery testing method, battery testing equipment, apparatus, storage medium, and product that can improve the accuracy of defect detection results obtained by detecting defects such as misassembly, omission, and mixing of components in semi-finished batteries.

[0005] Firstly, this application provides a battery detection method. The method includes:

[0006] Acquire images of semi-finished batteries obtained during the battery assembly process;

[0007] Acquire feature information of detected items in the region of interest in the image; the detected items include at least one of the assembly parts and the assembly area before the assembly parts are installed.

[0008] In the region of interest, determine the number of detection items that are consistent with the feature information corresponding to the region of interest;

[0009] Based on the number of detections and references corresponding to each region of interest in the image, the defect detection results of the semi-finished battery are obtained;

[0010] Among these, obtaining feature information of detected items in the region of interest in the image includes:

[0011] Obtain the target battery type for the semi-finished battery;

[0012] Based on the correspondence between battery type and the feature information of the detected items in the labeled region of interest samples, the feature information of the detected items in the region of interest in the image is obtained.

[0013] Among them, the region of interest sample is the image region of the detected item in the image sample of the semi-finished battery of the battery type.

[0014] In this embodiment, by acquiring an image of a semi-finished battery obtained during battery assembly, feature information of detection items in the region of interest (ROI) of the image is obtained. The number of detection items in the ROI that match the feature information corresponding to the ROI is determined. Based on the number of detection items corresponding to each ROI and a reference number in the image, the defect detection result of the semi-finished battery is obtained. Since the detection items include at least one from the assembly area before assembly, and the feature information of the detection items in the ROI of the image is obtained, the assembly installation status in the image can be detected based on the feature information and reference number of the detection items in the ROI, thus obtaining the defect detection result of the assembly installation status of the semi-finished battery, improving the accuracy of the obtained defect detection result. Furthermore, by acquiring the target battery type of the semi-finished battery, and based on the correspondence between the battery type and the feature information of the detection items in the labeled ROI samples, the feature information of the detection items in the ROI of the image is obtained. This eliminates the need for battery testing equipment to identify the feature information of the detection items in the ROI, thereby improving the efficiency of obtaining the feature information of the detection items in the ROI of the image, and thus improving the efficiency of image detection to obtain the defect detection result of the semi-finished battery.

[0015] In one embodiment, the defect detection results of the semi-finished battery are obtained based on the number of detections and the number of references corresponding to each region of interest in the image, including:

[0016] When the number of tests corresponding to the region of interest is inconsistent with the reference number, the defect detection result is determined to be that the semi-finished battery has a defect;

[0017] When the number of tests corresponding to each region of interest is consistent with the reference number, the defect detection result is determined to be that the semi-finished battery has no defects.

[0018] In this embodiment, when the number of detections corresponding to the region of interest is inconsistent with the reference number, the defect detection result is determined to be that the semi-finished battery has a defect; when the number of detections corresponding to each region of interest is consistent with the reference number, the defect detection result is determined to be that the semi-finished battery does not have a defect. The defect detection result can be obtained by directly comparing the number of detections with the reference number, which improves the efficiency of obtaining the defect detection result.

[0019] In one embodiment, obtaining feature information of detected items in a region of interest in an image includes:

[0020] Mark the regions of interest in the image;

[0021] Based on the type, feature information, and identifier code of the detected items in the region of interest, determine the target information identifier of the feature information;

[0022] The comparison result between the target information identifier and the information identifier of the feature information in the first feature information set is determined; the first feature information set is the set of feature information corresponding to the region of interest in the acquired image;

[0023] Based on the comparison results and the first feature information set, a second feature information set is obtained; the feature information corresponding to the region of interest in the image includes the feature information in the second feature information set.

[0024] In this embodiment, the target information identifier of the feature information is determined according to the type, feature information and identification code of the detected item in the region of interest. The feature information of the detected item is identified by the target information identifier, which makes it easier to determine whether the target information identifier exists in the first feature information set based on the target information identifier. If the target information identifier does not exist, the first feature information set is updated, thereby enriching the feature information in the first feature information set.

[0025] In one embodiment, a second feature information set is obtained based on the comparison result and the first feature information set, including:

[0026] If the comparison result shows that there is a target information identifier in the information identifier corresponding to the first feature information set, and there is a labelable next region of interest in the image, determine the comparison result corresponding to the next region of interest, and obtain the second feature information set based on the comparison result corresponding to the next region of interest and the first feature information set.

[0027] If the comparison result shows that the target information identifier exists in the information identifier and there is no next region of interest in the image, the first feature information set is determined as the second feature information set.

[0028] In this embodiment, a rich set of second feature information can be obtained, which facilitates the subsequent use of the rich set of second feature information to conduct a more comprehensive inspection of the assembly defects of the semi-finished battery with the same battery type as the semi-finished battery, thereby improving the accuracy of the defect detection results of the obtained semi-finished battery.

[0029] In one embodiment, the method further includes:

[0030] If the comparison result indicates the existence of a target information identifier, establish a correspondence between the feature information corresponding to the same information identifier as the region of interest and the target information identifier.

[0031] In this embodiment, when the comparison result indicates the presence of target information, a correspondence is established between the labeled region of interest and the feature information in the first feature information set. This establishes an association between the labeled region of interest and the feature information in the first feature information set, thereby facilitating the subsequent detection of the image and quickly determining the feature information corresponding to the region of interest in the image based on the correspondence.

[0032] In one embodiment, the method further includes:

[0033] If the comparison result indicates that no target information identifier exists, the feature information is determined to be the feature information of the detected item in the region of interest, and the feature information is added to the first feature information set.

[0034] If there is a labelable next region of interest in the image, determine the comparison result corresponding to the next region of interest, and obtain the second feature information set based on the comparison result corresponding to the next region of interest and the first feature information set with added feature information;

[0035] If no next region of interest exists in the image, the first set of feature information with added feature information is determined as the second set of feature information.

[0036] In this embodiment, by updating the first feature information set, a rich second feature information set can be obtained, which facilitates the subsequent use of the rich second feature information set to conduct a more comprehensive detection of the image and improves the accuracy of the defect detection results of the obtained semi-finished battery.

[0037] In one embodiment, the method further includes:

[0038] When feature information is added to the first feature information set, a correspondence is established between the region of interest and the feature information in the second feature information set.

[0039] In this embodiment, by adding feature information to the first feature information set, a correspondence is established between the region of interest and the feature information in the second feature information set, which facilitates the subsequent detection of the image by quickly determining the target feature information corresponding to the region of interest in the image based on index connection.

[0040] In one embodiment, the detection item also includes foreign objects.

[0041] Since the inspection items can also include foreign objects, it is possible to not only inspect the assemblies, but also detect foreign objects in the images, thereby improving the reliability of the defect inspection results of the obtained semi-finished batteries.

[0042] Secondly, this application also provides a battery testing device, which includes:

[0043] Image acquisition component, used to acquire images of semi-finished batteries obtained during battery assembly;

[0044] The processor is used to acquire feature information of detection items in regions of interest (ROIs) of an image; determine the number of detection items in the ROI that are consistent with the feature information corresponding to the ROI; and obtain the defect detection results of the semi-finished battery based on the number of detection items and the reference number corresponding to each ROI in the image; the detection items include at least one in the assembly area before the assembly is installed.

[0045] Among these, obtaining feature information of detected items in the region of interest in the image includes:

[0046] Obtain the target battery type for the semi-finished battery;

[0047] Based on the correspondence between battery type and the feature information of the detected items in the labeled region of interest samples, the feature information of the detected items in the region of interest in the image is obtained.

[0048] Among them, the region of interest sample is the image region of the detected item in the image sample of the semi-finished battery of the battery type.

[0049] Thirdly, this application also provides a battery testing device. The device includes:

[0050] The first acquisition module is used to acquire images of semi-finished batteries obtained during the battery assembly process;

[0051] The second acquisition module is used to detect feature information of items in the region of interest in the image; the detected items include at least one of the assembly parts and the assembly area before the assembly parts are installed.

[0052] The detection module is used to determine the number of detection items in the region of interest that are consistent with the feature information corresponding to the region of interest; based on the number of detection items and the reference number corresponding to each region of interest in the image, the defect detection results of the semi-finished battery are obtained.

[0053] Among these, obtaining feature information of detected items in the region of interest in the image includes:

[0054] Obtain the target battery type for the semi-finished battery;

[0055] Based on the correspondence between battery type and the feature information of the detected items in the labeled region of interest samples, the feature information of the detected items in the region of interest in the image is obtained.

[0056] Among them, the region of interest sample is the image region of the detected item in the image sample of the semi-finished battery of the battery type.

[0057] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, the computer program being executed by a processor using the steps of any of the methods described above.

[0058] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of any of the methods described above.

[0059] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0060] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiments below. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0061] Figure 1 This is a diagram illustrating the application environment of the battery detection method in one embodiment;

[0062] Figure 2 This is a schematic flowchart of a battery detection method provided in an embodiment of this application;

[0063] Figure 3 This is a flowchart illustrating a method for obtaining defect detection results provided in an embodiment of this application;

[0064] Figure 4 This is a flowchart illustrating a method for obtaining a set of feature information provided in an embodiment of this application;

[0065] Figure 5 This is a schematic diagram of a region of interest and its corresponding feature information provided in an embodiment of this application;

[0066] Figure 6 This is a flowchart illustrating another method for obtaining a set of feature information provided in an embodiment of this application;

[0067] Figure 7 This is a schematic flowchart of another battery detection method provided in an embodiment of this application;

[0068] Figure 8This is a schematic diagram of a battery detection system provided in an embodiment of this application;

[0069] Figure 9 This is a schematic diagram of the structure of a battery detection device provided in an embodiment of this application. Detailed Implementation

[0070] The embodiments of the technical solution of this application will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of this application and are therefore merely examples, and should not be used to limit the scope of protection of this application.

[0071] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0072] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0073] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0074] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0075] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0076] In the description of the embodiments of this application, the technical terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the embodiments of this application and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of this application.

[0077] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0078] During the battery production and assembly process, various assembly parts are required for assembly. Due to the diverse shapes of these parts, it is easy to cause problems such as incorrect, missing, or mixed assembly of battery parts, which can lead to significant risks. Therefore, it is necessary to detect incorrect, missing, or mixed assembly of battery parts.

[0079] Currently, manually visually inspecting batteries for defects such as incorrect, missing, or mixed assembly of components has the problem of low accuracy in defect detection.

[0080] To address the aforementioned technical problems, this application provides a battery testing method that can be applied to a battery testing device. This battery testing device can be a server, and its internal structure diagram can be as follows: Figure 1As shown, the battery testing device includes a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a battery testing method. The display screen can be an LCD screen or an e-ink display screen. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the device's casing, or an external keyboard, touchpad, or mouse.

[0081] Those skilled in the art will understand that Figure 1 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the battery testing equipment to which the present application is applied. A specific battery testing equipment may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0082] In one embodiment, such as Figure 2 As shown, Figure 2 This is a flowchart illustrating a battery detection method provided in an embodiment of this application, applied to... Figure 1 Taking the battery testing equipment in the example, the following steps are included:

[0083] S201, Obtain an image of the semi-finished battery obtained during the battery assembly process.

[0084] In the battery manufacturing and assembly process, semi-finished batteries refer to batteries that have completed partial assembly but have not yet been packaged or entered the next process stage. Due to the diverse types of components, the varying assembly positions and quantities depending on the product model, and the involvement of manual operation in the assembly process, problems such as incorrect assembly, omissions, and mixed assembly are prone to occur. To achieve automated quality monitoring of the assembly process, this step first uses image acquisition equipment to photograph the semi-finished batteries, and, in conjunction with appropriately arranged light sources, automatically photographs the assembly area where the components are located, obtaining high-definition, highly consistent images. The image acquisition equipment includes, for example, industrial cameras and vision sensors.

[0085] It can reduce the color difference and grayscale value of the detected items in the image to less than 40, and transform a three-dimensional, detailed object into a two-dimensional image with clear outlines and simple features, which makes it easier to identify the region of interest of the detected items in the image.

[0086] S202, acquire feature information of the detected items in the region of interest in the image; the detected items include at least one of the assembly parts and the assembly area before the assembly parts are installed.

[0087] A deep learning-based model can be used to annotate regions of interest (ROIs) in an image, and the features of the detected items within these ROIs can be analyzed to obtain corresponding feature information. Detected items may include assemblies and / or the assembly area before assembly. Alternatively, detected items may include at least one of assemblies and the assembly area before assembly, as well as foreign objects.

[0088] Among them, the components may include, but are not limited to, fireproof components, insulating components, sealing components, fasteners, foam, and protective covers. In other words, the types of components may include the aforementioned fireproof components, insulating components, sealing components, fasteners, foam, and protective covers.

[0089] In this embodiment of the application, the feature information of the detected items in the region of interest in the image can be obtained in the following manner:

[0090] In one possible implementation, the features of the detection items in the region of interest in the image can be observed manually. For example, the color, appearance, and shape of the assembly area can be observed. The observed features are then input as feature information into the battery detection device, enabling the battery detection device to acquire the feature information of the detection items in the region of interest.

[0091] In another possible implementation, the target battery type of the semi-finished battery can be obtained. Based on the correspondence between the battery type and the feature information of the labeled region of interest, feature information can be acquired. That is, this correspondence stores the feature information of the detection items in the region of interest corresponding to the battery type. The feature information corresponding to the battery type that is the same as the target battery type in this correspondence can be directly used as the feature information of the detection items in the region of interest in the image.

[0092] S203. Based on the feature information and reference quantity of the detection items in the region of interest, the defect detection results of the semi-finished battery are obtained.

[0093] The number of detection items that match the feature information corresponding to the region of interest can be determined in the region of interest. Based on the number of detection items and the reference number corresponding to each region of interest in the image, the defect detection results of the semi-finished battery can be obtained.

[0094] For example, when the detection items include assemblies, i.e., when the regions of interest (ROIs) of the assemblies in the image are directly labeled using a forward annotation method, if the reference quantity of an assembly in a certain ROI is 2, but the detection quantity of that ROI is 1, it means that the actual detection quantity of the assembly is less than the reference quantity, indicating that the assembly is missing, and a defect detection result indicating a defect can be output. If the reference quantity of an assembly in a certain ROI is 2, but the actual detection quantity of that ROI is 3, it means that the actual quantity of the assembly is greater than the reference quantity, indicating that the assembly is overloaded, and a defect detection result indicating a defect can be output.

[0095] Another example is when the detection item includes the assembly area before the installation of the components. That is, the region of interest of the assembly area before the installation of the components in the semi-finished battery image is marked by reverse annotation. The installed components can be marked as the foreground first, and then the region of interest of the assembly area without components can be obtained by reverse selection or model calculation. That is, the region of interest of the assembly area before the installation of components is obtained.

[0096] The assembly area before component installation refers to the area where the component is located after installation, excluding the component. For example, if fireproof components are installed in the first area and insulating components are installed in the second area, then the first area is the assembly area before component installation, and the second area is also the assembly area before component installation.

[0097] For ease of description, the assembly area before the assembly component is installed is referred to as the assembly area. This assembly area is marked in the image using a reverse annotation method. When this assembly area is considered the region of interest (ROI), it is the area where the assembly component should be installed, but where no assembly component exists. This assembly area is used as a detection item. The correct installation of the assembly component in this area is defined as the presence of the assembly component, and the corresponding reference quantity, i.e., the "background quantity," can be set to 0, meaning the assembly component covers the background. Therefore, the "background quantity" is 0. If the feature information corresponding to this detection item is used to detect the ROI, and the resulting detection quantity is 0, which is the same as the reference quantity of 0, it means no background was detected, and the assembly component is actually installed in the assembly area. Thus, the ROI detection is considered successful. If the resulting detection quantity is 1, which is different from the reference quantity of 0, it means the background was detected, but the assembly component did not cover the background. Thus, the ROI detection is considered unsuccessful.

[0098] Another example is when the inspection items include assemblies and the assembly area before assembly installation, that is, combining reverse annotation and forward annotation methods for inspection to further improve inspection accuracy. For example, when reverse annotation is used first, the number of inspections is the same as the reference number, confirming that assemblies are actually installed in the assembly area. Then, forward annotation is used for inspection. Assuming that the assembly area requires bolts, forward annotation can be used to further determine whether the assembly in the assembly area is indeed a bolt. For example, the number of bolts with the same feature information as the bolts can be determined in the region of interest. If the number of inspections is inconsistent with the reference number of bolts, defects such as missing or extra bolts can be identified.

[0099] In this embodiment, by acquiring an image of a semi-finished battery obtained during battery assembly, feature information of detection items in the region of interest (ROI) of the image is obtained. Based on the feature information and reference quantity of the detection items in the ROI, the defect detection result of the semi-finished battery is obtained. Since the detection items include at least one of the assembly parts and the assembly area before assembly parts installation, and by acquiring the feature information of the detection items in the ROI of the image, the assembly installation status in the image can be detected based on the feature information and reference quantity of the detection items in the ROI, thereby obtaining the defect detection result of the assembly installation status of the semi-finished battery, improving the accuracy of the obtained defect detection result.

[0100] In one embodiment, S202 above, obtaining the feature information of the detected items in the region of interest in the image, can be achieved in the following way:

[0101] The target battery type of the semi-finished battery is obtained. Based on the correspondence between the battery type and the feature information of the detection items in the labeled region of interest (ROI) samples, the feature information of the detection items in the ROI in the image is obtained. The ROI samples are the image regions containing the detection items in the image samples of the semi-finished battery of the battery type.

[0102] The target battery type of a semi-finished battery can be obtained based on its product identification code. This can be achieved by scanning the product identification code and calling the MES interface based on a pre-defined code segment within the code. The target battery type determines the types, quantities, locations, and appearance characteristics of the parts that should be assembled with that battery model. Based on the correspondence between the target battery type and the feature information of the detection items in the marked region of interest sample, the feature information of the detection items in the region of interest in the image can be obtained. This feature information is then used for subsequent defect detection.

[0103] The number of region-of-interest (ROI) samples corresponding to each battery type can be multiple, and the feature information of the detection items in each RIO sample can be obtained in advance. For ease of description, the feature information corresponding to all RIO samples is denoted as the feature information set. That is, based on the correspondence between battery type and feature information set, the feature information set corresponding to the target battery type can be determined, and the feature information in the feature information set corresponding to the target battery type is the feature information of the detection items in the RIO in the image.

[0104] Furthermore, a region of interest (ROI) sample under a specific battery type can correspond to a region of interest (ROI) under the target battery type, and the feature information corresponding to this ROI sample is the feature information corresponding to that ROI. The ROI sample corresponding to the target ROI can be found through position matching. Specifically, based on the coordinate position of the ROI in the image, combined with image registration or coordinate transformation methods, it is mapped to the coordinate system of the image sample of the semi-finished battery type, and the ROI sample with the highest positional overlap or closest distance is found and used as the ROI sample corresponding to the target ROI.

[0105] The correspondence is established in advance. The region of interest (ROI) samples of the semi-finished battery pattern samples of the battery type can be manually marked. The color, shape and other feature information of the test items in the marked ROI samples can be manually identified and determined as the feature information corresponding to the ROI sample.

[0106] Because battery testing equipment can store multiple battery types and their corresponding feature information sets, it can be applied to defect detection of semi-finished batteries of various types. For example, if the equipment is currently using the feature information set corresponding to one battery type for testing, and subsequently needs to test images of semi-finished batteries corresponding to another battery type, it can automatically switch to using the feature information set corresponding to that other battery type for online defect detection based on the correspondence between different battery types and different feature information sets.

[0107] In this embodiment, since the target battery type of the semi-finished battery is obtained, and the feature information of the detection items in the region of interest in the image is obtained according to the correspondence between the battery type and the feature information of the detection items in the labeled region of interest sample, the feature information of the detection items in the region of interest in the image is obtained. This eliminates the need for battery detection equipment to identify the feature information of the detection items in the region of interest, thereby improving the efficiency of obtaining the feature information of the detection items in the region of interest in the image, and thus improving the efficiency of detecting the image and obtaining the defect detection results of the semi-finished battery.

[0108] In one embodiment, such as Figure 3 As shown, Figure 3 This is a flowchart illustrating a method for obtaining defect detection results according to an embodiment of this application. This embodiment relates to a possible implementation of how to obtain defect detection results for a semi-finished battery based on the first feature information and reference quantity of detection items in the region of interest. Based on the above embodiment, S203 may include the following steps:

[0109] S301, determine the number of detection items in the region of interest that are consistent with the feature information corresponding to the region of interest.

[0110] Using the feature information corresponding to the region of interest, search for the number of detection items that match the feature information in the region of interest, and record the total number of all detection items that match the feature information as the detection count.

[0111] If a detection item matching the feature information is found within the region of interest, the detection count is incremented by one. The initial value of the detection count before the search can be 0.

[0112] The similarity between the feature information of each detection item in the region of interest and the specified feature information can be determined. If the similarity is greater than a preset similarity, the feature information of the detection item in the region of interest is determined to be consistent with the specified feature information, and the detection count can be increased by one. For example, if there are three assemblies in the region of interest, and the feature information of two of the assemblies is consistent with the specified feature information, then the detection count is equal to 2.

[0113] S302, based on the number of detections and references corresponding to each region of interest in the image, the defect detection results of the semi-finished battery are obtained.

[0114] For each region of interest, the number of tests corresponding to that region of interest can be compared with a preset reference number. The reference number is derived from the preset value of that type of test item corresponding to the battery type of the semi-finished battery. For example, if the number of assemblies that should be assembled in a certain region of interest is 3, then the reference number for that region of interest is 3.

[0115] When the number of tests corresponding to each region of interest is inconsistent with the reference number, the defect detection result is determined to be that the semi-finished battery has a defect; when the number of tests corresponding to each region of interest is consistent with the reference number, the defect detection result is determined to be that the semi-finished battery does not have a defect.

[0116] In this embodiment, by determining the number of detection items in the region of interest that match the feature information corresponding to the region of interest, and based on the number of detection items and the reference number corresponding to each region of interest in the image, the defect detection result of the semi-finished battery is obtained. Since the number of detection items matching the feature information is determined in the region of interest, the accuracy of the obtained number of detections can be improved, thereby improving the accuracy of the obtained defect detection result.

[0117] In one embodiment, S302 above, obtaining the defect detection results of the semi-finished battery based on the number of detections and references corresponding to each region of interest in the image, can be achieved in the following way:

[0118] When the number of tests corresponding to the region of interest is inconsistent with the reference number, the defect detection result is determined to be that the semi-finished battery has a defect;

[0119] When the number of tests corresponding to each region of interest is consistent with the reference number, the defect detection result is determined to be that the semi-finished battery has no defects.

[0120] For example, if the detection item includes assemblies, the region of interest (ROI) of the assemblies is marked using positive annotation. The reference number of assemblies is, for example, 3. If the detection number corresponding to this ROI is 2, that is, two assemblies with the same feature information as the assemblies are found in the ROI, since the detection number is inconsistent with the reference number, it means that the area that should have 3 assemblies installed only has 2 assemblies installed. Therefore, it can be determined that the ROI detection fails.

[0121] If there is a discrepancy between the number of tests and the reference number, the defect detection result can be determined to indicate that the semi-finished battery has a defect. In other words, as long as there is a region of interest that fails the test, the defect detection result can be determined to indicate that the semi-finished battery has a defect.

[0122] If the number of tests is consistent with the reference number in each of the first comparison results, the defect detection result is determined to be that the semi-finished battery has no defects. That is, if all regions of interest in the image pass the detection, the defect detection result can be determined to be that the semi-finished battery has no defects.

[0123] In this embodiment, when the number of detections corresponding to the region of interest is inconsistent with the reference number, the defect detection result is determined to be that the semi-finished battery has a defect; when the number of detections corresponding to each region of interest is consistent with the reference number, the defect detection result is determined to be that the semi-finished battery does not have a defect. The defect detection result can be obtained by directly comparing the number of detections with the reference number, which improves the efficiency of obtaining the defect detection result.

[0124] In one embodiment, such as Figure 4 As shown, Figure 4 This is a flowchart illustrating a method for obtaining a feature information set according to an embodiment of this application. The method includes the following steps:

[0125] S401, mark the region of interest in the image.

[0126] Multiple regions of interest can be obtained by labeling components, foreign objects, etc. in the image of a semi-finished battery. For example, labeling the first component in the image yields one region of interest, and labeling the first foreign object in the image yields another region of interest.

[0127] S402, determine the target information identifier of the feature information based on the type, feature information and identification code of the detected item in the region of interest.

[0128] The detection items in the region of interest can also include foreign objects. The types of detection items can include, for example, both assemblies and foreign objects. The target information identifier of the feature information can be determined based on the type of detection item in the marked region of interest, the feature information of the detection item, and the identification code of the detection item.

[0129] For example, taking the first assembly as the detected item in the marked region of interest, the first assembly belongs to a certain assembly type. Based on its color, shape, and grayscale value, the subclass of that assembly type to which the first assembly belongs can be determined. Finally, based on the assembly type, the subclass of that assembly type to which the first assembly belongs, and the identification code of the first assembly, the target information identifier of the first assembly's feature information is determined. The target information identifier can include the assembly type of the first assembly, the subclass of that assembly type to which the first assembly belongs, and the identification code of the first assembly.

[0130] S403, determine the comparison result between the target information identifier and the information identifier of the feature information in the first feature information set.

[0131] The first feature information set is a set of feature information corresponding to the currently acquired target battery type. Each feature information in the first feature information set has an information identifier. The target information identifier can be compared with the information identifiers of each feature information in the first feature information set to determine whether there is an information identifier in the first feature information set that is the same as the target information identifier, that is, to determine whether the target information identifier exists in the first feature information set.

[0132] S404, Based on the comparison results and the first feature information set, obtain the second feature information set.

[0133] The feature information corresponding to the region of interest in the image includes the feature information in the second feature information set.

[0134] If the comparison result shows that the target information identifier exists in the information identifier of the first feature information set, and there is no next region of interest in the image, then the first feature information set is determined to be the second feature information set.

[0135] If the comparison result shows that the target information identifier exists in the information identifier of the first feature information set, and there is a labelable next region of interest in the image, the comparison result corresponding to the next region of interest is determined, and the second feature information set is obtained based on the comparison result corresponding to the next region of interest and the first feature information set.

[0136] If the comparison result obtained in S403 is that there is no target information identifier in the first feature information set, then the feature information corresponding to the target information identifier can be added to the first feature information set. If there is a labelable next region of interest in the image, the comparison result corresponding to the next region of interest is determined. Based on the comparison result corresponding to the next region of interest and the first feature information set with added feature information, the second feature information set is obtained.

[0137] It should be noted that if the battery testing equipment does not store the second feature information set obtained in this step, it can store the second feature information set obtained in this step in the battery testing equipment. That is, it stores the second feature information set corresponding to the target battery type of the semi-finished battery in S201. Since each feature information in the second feature information set has a corresponding information identifier, it is convenient for the battery testing equipment to use the feature information corresponding to each information identifier in the second feature information set to detect the detection items of the region of interest in other semi-finished batteries of the target battery type. There is no need for the battery testing equipment to repeat the steps of S401 to S404 to determine the feature information set of the region of interest in other semi-finished batteries, thereby improving the defect detection efficiency of other semi-finished batteries.

[0138] In this embodiment, the target information identifier of the feature information is determined according to the type, feature information and identification code of the detected item in the region of interest. The feature information of the detected item is identified by the target information identifier, which facilitates the subsequent determination of whether the target information identifier exists in the first feature information set based on the target information identifier. If the target information identifier does not exist, the first feature information set is updated to enrich the feature information in the first feature information set, thereby obtaining the feature information corresponding to the target battery type.

[0139] In one embodiment, S404 above, which obtains the second feature information set based on the comparison result and the first feature information set, can be implemented in the following way:

[0140] If the comparison result shows that there is a target information identifier in the information identifier corresponding to the first feature information set, and there is a labelable next region of interest in the image, determine the comparison result corresponding to the next region of interest, and obtain the second feature information set based on the comparison result corresponding to the next region of interest and the first feature information set.

[0141] If the comparison result shows that the target information identifier exists in the information identifier and there is no next region of interest in the image, the first feature information set is determined as the second feature information set.

[0142] like Figure 5 As shown, Figure 5 This is a schematic diagram illustrating a region of interest and its corresponding feature information provided in an embodiment of this application. The types of detection items may include, for example, two types: assemblies and foreign objects. Figure 5 The description is based on the type of assembly (Class A) and the type of foreign object (Class B). Class A includes two subclasses: Aa and Ab.

[0143] For example, assuming the first feature information set contains only feature information Aa1, the region of interest marked in the image of the semi-finished battery in S401 is... Figure 5 In the image of the semi-finished battery on the left, the area enclosed by the dashed box in the lower left corner has a target information identifier of Aa1 for the feature information of the detected item within the dashed box. In this case, the first feature information set contains an identifier identical to the target information identifier Aa1. This means that the first feature information set already contains the feature information of the detected item within the dashed box in the lower left corner. Therefore, it is unnecessary to repeatedly add the feature information corresponding to the target information identifier Aa1 to the first feature information set, thus avoiding duplicate feature information in the first feature information set.

[0144] If, subsequently, there is a labelable next region of interest in the image of the semi-finished battery, for example, if the next region of interest is... Figure 5The region image selected by the dashed box in the lower right corner of the image of the semi-finished battery on the left can be further compared using the methods described in S401 to S403 above. Based on this comparison result and the first feature information set, a second feature information set is obtained. For example, referring to the above example, it can be seen that the comparison result corresponding to the region image selected by the dashed box in the lower right corner does not contain the target information identifier of the region image selected by the dashed box in the lower right corner in the information identifier of the first feature information set. Therefore, the feature information corresponding to the region image selected by the dashed box in the lower right corner can be added to the first feature information set. For ease of explanation, the target information identifier of the feature information corresponding to the region image selected by the dashed box in the lower right corner is denoted as Ab2. That is, after the feature information corresponding to the target information identifier Ab2 is added to the first feature information set, the current first feature information set includes feature information Aa1 and feature information Ab2, where feature information Ab2 refers to the feature information with target information identifier Ab2. In this way, the first feature information set can be updated based on the comparison result corresponding to the next region of interest to obtain a richer second feature information set.

[0145] Another example is to assume that the obtained first set of feature information is as follows: Figure 5 As shown on the right, the first set of feature information includes feature information Aa1, Aa2, Aa3, Ab1, Ab2, Ab3, Ba0, Bb0, and Bc0. The region of interest marked in the image of the semi-finished battery in S401 is... Figure 5 In the image of the semi-finished battery on the left, if there is no next region of interest to be labeled in the image of the semi-finished battery, since feature information Aa1 already exists in the first feature information set, in this case, the first feature information set can be directly used as the second feature information set.

[0146] It should be noted that feature information Aa2 refers to the feature information of target information identified as Aa2, feature information Aa3 refers to the feature information of target information identified as Aa3, and so on. These will not be elaborated on here.

[0147] The method for determining the target information identifier of the feature information of the detected item when the detected item in the marked region of interest is an assembly is the same as the method for determining the target information identifier of the feature information of the foreign object when the detected item in the marked region of interest is a foreign object. For example, the target information identifier of the first foreign object is Ba0, the target information identifier of the second foreign object is Bb0, and the target information identifier of the third foreign object is Bc0.

[0148] In this embodiment, a rich set of second feature information can be obtained, which facilitates the subsequent use of the rich set of second feature information to conduct a more comprehensive inspection of the assembly defects of the semi-finished battery with the same battery type as the semi-finished battery, thereby improving the accuracy of the defect detection results of the obtained semi-finished battery.

[0149] In one embodiment, the method further includes:

[0150] If the comparison result indicates the existence of a target information identifier, establish a correspondence between the feature information corresponding to the same information identifier as the region of interest and the target information identifier.

[0151] like Figure 5 As shown, when there is an information identifier in the first feature information set that is identical to the target information identifier Aa1, a system is established. Figure 5 The correspondence between the region of interest in the lower left corner of the left side of the image and the feature information Aa1 can be established through methods such as... Figure 5 The indexed connection is shown below. Figure 5 The above is merely an illustrative example, showing only four index joins, with a dashed line representing one index join.

[0152] In this embodiment, when the comparison result indicates the existence of a target information identifier, a correspondence is established between the region of interest and the feature information corresponding to the same information identifier as the target information identifier. This establishes an association between the labeled region of interest and the feature information in the first feature information set, thereby storing the correspondence between the target battery type and the feature information of the detected item in the region of interest. This facilitates the subsequent use of this correspondence to quickly determine the feature information corresponding to the region of interest in other images to be detected.

[0153] In one embodiment, the second set of feature information can also be obtained by the following method:

[0154] If the comparison result indicates that no target information identifier exists, the feature information is determined to be the feature information of the detected item in the region of interest, and the feature information is added to the first feature information set; if there is a labelable next region of interest in the image, the comparison result corresponding to the next region of interest is determined, and a second feature information set is obtained based on the comparison result corresponding to the next region of interest and the first feature information set with added feature information; if there is no next region of interest in the image, the first feature information set with added feature information is determined as the second feature information set.

[0155] For example, assuming the first feature information set contains only feature information Aa1, the region of interest marked in the semi-finished battery image in S401 is... Figure 5In the image of the semi-finished battery on the left, the region selected by the dashed box in the lower right corner has no corresponding feature information in the first feature information set, as only feature information Aa1 exists. In this case, the feature information corresponding to the dashed box in the lower right corner can be directly added to the first feature information set. It can then be determined whether a next labelable region of interest exists. If a labelable next region of interest exists in the image of the semi-finished battery, the comparison result corresponding to the next region of interest is determined. Based on the comparison result corresponding to the next region of interest and the first feature information set with added feature information, a second feature information set is obtained. If no next region of interest exists in the image of the semi-finished battery, the first feature information set with added feature information is used as the second feature information set.

[0156] In this embodiment, by updating the first feature information set, a rich second feature information set can be obtained, which facilitates the subsequent use of the rich second feature information set to conduct a more comprehensive detection of the image and improves the accuracy of the defect detection results of the obtained semi-finished battery.

[0157] In one embodiment, when feature information is added to a first feature information set, a correspondence is established between the region of interest and the feature information in a second feature information set.

[0158] If the comparison result indicates that no target information identifier exists, the feature information corresponding to the target information identifier can be added to the first feature information set. With the feature information corresponding to the target information identifier added to the first feature information set, a correspondence is established between the labeled region of interest and the feature information corresponding to the target information identifier in the first feature information set. For example... Figure 5 As shown, in conjunction with the above examples, the marked region of interest is... Figure 5 In the image of the semi-finished battery on the left, if the target information identifier for the marked region of interest is not found in the information identifier of the first feature information set, the feature information corresponding to the target information identifier can be added to the first feature information set to update the first feature information set. The updated first feature information set includes feature information Aa1 and feature information Ab2, and a system is established as follows: Figure 5 The illustrated region of interest is connected by an index to feature information Ab2 in the updated first feature information set.

[0159] In this embodiment, by adding feature information to the first feature information set, a correspondence is established between the region of interest and the feature information in the second feature information set, which facilitates the subsequent detection of the image and allows for the rapid determination of the feature information corresponding to the region of interest in the image based on the correspondence.

[0160] In one embodiment, the detection item may also include foreign objects.

[0161] For regions of interest (ROIs) marked with positive annotations, the image can be detected based on these ROIs to determine whether foreign objects exist in the image.

[0162] The number of detection items (i.e., the reference number of foreign objects) in the region of interest (ROI) of a foreign object can be set to 0. For ease of description, the ROI of the foreign object is denoted as the foreign object ROI. If the foreign object ROI corresponds to a region of interest in the image, and a detection item is detected in that ROI that matches the feature information of the foreign object in the foreign object ROI, then the detection item is considered a foreign object, and the detection count of the detected foreign object is incremented by 1. For example, if the initial value of the detection count of the detected foreign object is 0, and the detection count of the detected foreign object is 1, which is not equal to the reference number of 0, then the ROI detection has failed, and the defect detection result of the semi-finished battery can be determined to be that the semi-finished battery has a defect.

[0163] In this embodiment, since the detection items can also include foreign objects, it is possible to not only detect the assembly, but also detect foreign objects in the image, thereby improving the reliability of the defect detection results of the obtained semi-finished battery.

[0164] In one embodiment, to provide a clearer explanation of the aforementioned feature information set, the following is combined with... Figure 6 To explain, Figure 6 This is a flowchart illustrating another method for obtaining a set of feature information provided in an embodiment of this application.

[0165] The region of interest in the image of the semi-finished battery of the battery type is marked. Based on the type, feature information and identification code of the detection items in the marked region of interest, the target information identifier of the feature information is determined, and it is determined whether the target information identifier exists in the information identifier of the first feature information set.

[0166] If a target information identifier exists, a correspondence is directly established between the labeled region of interest and the feature information corresponding to the target information identifier in the first feature information set. Furthermore, it can be determined whether there is a labelable next region of interest in the semi-finished battery image of the battery type. If a next region of interest exists, the comparison result corresponding to the next region of interest is determined. Based on the comparison result corresponding to the next region of interest and the first feature information set, a second feature information set is obtained. If there is no next region of interest, the first feature information set is determined as the second feature information set.

[0167] If no target information identifier exists, the feature information corresponding to the target information identifier is added to the first feature information set; if there is a labelable next region of interest in the image of the semi-finished battery of the battery type, the comparison result corresponding to the next region of interest is determined, and a second feature information set is obtained based on the comparison result corresponding to the next region of interest and the first feature information set with added feature information; if there is no next region of interest in the image of the semi-finished battery of the battery type, the first feature information set with added feature information is determined as the second feature information set.

[0168] In one embodiment, such as Figure 7 As shown, Figure 7 This is a schematic flowchart of another battery detection method provided in an embodiment of this application. The method includes the following steps:

[0169] S701, acquire an image of the semi-finished battery obtained during the battery assembly process and the target battery type of the semi-finished battery.

[0170] S702: Based on the correspondence between battery type and the feature information of the detected items in the labeled region of interest samples, obtain the feature information of the detected items in the region of interest in the image.

[0171] S703, Obtain the region of interest in the image.

[0172] S704, determine the number of detection items in the region of interest that are consistent with the feature information corresponding to the region of interest.

[0173] S705, when the number of tests corresponding to the region of interest is inconsistent with the reference number, the defect detection result is determined to be that the semi-finished battery has a defect.

[0174] If the initial defect detection result indicates a defect in the semi-finished battery, a prompt message can be output to indicate the defect and initiate rework. After rework, the process can return to step S701 for re-inspection of the reworked battery. If the defect detection result differs from the initial result, it means the semi-finished battery has already undergone rework and still has defects. In this case, the semi-finished battery can be sent to the rework workshop for further repair, and the defect detection result can be sent to a display device for display.

[0175] S706, when the number of tests corresponding to each region of interest is consistent with the reference number, the defect detection result is determined to be that the semi-finished battery has no defects.

[0176] In one embodiment, combined Figure 8 The overall process of the battery detection method according to the embodiments of this application will be described. Figure 8 This is a schematic diagram of a battery detection system provided in an embodiment of this application.

[0177] The imaging device can capture images of semi-finished batteries during the assembly process. If the image quality of the obtained images is unsatisfactory, the device can be adjusted and images reacquired until the reacquired images meet the requirements. Figure 8 The industrial control computer or server shown sends images, enabling it to acquire images of semi-finished batteries during the battery assembly process. The image quality must meet requirements such as clarity, color difference of targets, and grayscale values ​​less than 40.

[0178] Battery testing equipment can be smart cameras, industrial PCs, or servers. Uninterruptible power supplies (UPS) can power the industrial PCs, ensuring uninterrupted operation and preventing data loss or equipment damage. The number of correspondences between battery types, labeled images, and feature information sets stored in a smart camera is less than the number stored in an industrial PC or server. For example, a smart camera might store labeled images and feature information sets corresponding to three battery types, while an industrial PC or server might store labeled images and feature information sets for more than three battery types. When the battery testing equipment is a smart camera, the smart camera itself can capture images of the semi-finished battery.

[0179] The battery testing equipment can obtain the target battery type of the semi-finished battery through MES (Manufacturing Execution System). Based on the correspondence between the battery type and the feature information of the detection items in the labeled region of interest (ROI) samples, it obtains the feature information of the detection items in the ROI of the image. Having obtained the feature information of the detection items in the ROI of the image, the battery testing method provided in this application is used to test the image, obtaining the defect detection results of the semi-finished battery. The battery testing equipment can then send the obtained defect detection results to a display screen for display.

[0180] If the battery testing equipment fails to acquire feature information of the detection item in the region of interest of the image, an image with satisfactory image quality is sent to the server. The server then uses the battery testing method provided in this application embodiment to test the image and obtain the defect detection result of the semi-finished battery.

[0181] For example, if the battery detection device is a smart camera, and the smart camera has not acquired feature information of the detected items in the region of interest (ROI) of the image, then an image with acceptable quality can be sent to the server. If the server does not store feature information of the detected items in the ROI of the image, then the above-mentioned... Figure 4 The corresponding embodiment provides a method for obtaining feature information of detected items in the region of interest (ROI) of the image. The server can send the feature information of the detected items in the ROI of the image to the smart camera for storage, so that the smart camera can use the stored feature information of the detected items in the ROI of the image for defect detection in subsequent inspections.

[0182] If the feature information set sent to the smart camera does not meet the Measurement System Analysis (MSA) standard, which is a core method for evaluating the accuracy and reliability of a measurement system, then the MSA test fails. This means that the feature information set must be examined to determine if any anomalies exist. If anomalies are found, the anomaly-laden features are updated, and the test is repeated until the updated features meet the MSA standard.

[0183] In one embodiment, a battery testing device is also provided, the battery testing device comprising:

[0184] Image acquisition component, used to acquire images of semi-finished batteries obtained during battery assembly;

[0185] The processor is used to obtain the target battery type of the semi-finished battery, obtain the first feature information of the detection items in the region of interest corresponding to the target battery type, and perform image detection based on the first feature information and reference quantity of the detection items in the region of interest to obtain the defect detection result of the semi-finished battery; the detection items include at least one in the assembly area before the assembly is installed.

[0186] The processor can be Figure 8 The processors in the smart camera, industrial computer, and server shown can be either the processor in the smart camera or the image acquisition component in the smart camera. Figure 8 The filming equipment used in the filming.

[0187] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0188] Based on the same inventive concept, this application also provides a battery testing apparatus for implementing the battery testing method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more battery testing apparatus embodiments provided below can be found in the limitations of the battery testing method described above, and will not be repeated here.

[0189] In one embodiment, such as Figure 9 As shown, Figure 9 This is a schematic diagram of a battery testing device provided in an embodiment of this application. The battery testing device 900 includes:

[0190] The first acquisition module 901 is used to acquire images of semi-finished batteries obtained during the battery assembly process;

[0191] The second acquisition module 902 is used to acquire feature information of detected items in the region of interest in the image; the detected items include at least one of the assembly parts and the assembly area before the assembly parts are installed.

[0192] The detection module 903 is used to detect images based on the feature information and reference quantity of the detection items in the region of interest, and obtain the defect detection results of the semi-finished battery.

[0193] In one embodiment, the second acquisition module 902 is specifically used to acquire the target battery type of the semi-finished battery; and to acquire the feature information of the detection items in the region of interest in the image based on the correspondence between the battery type and the feature information of the detection items in the labeled region of interest sample; wherein, the region of interest sample is the image region of the detection items in the image sample of the semi-finished battery of the battery type.

[0194] In one embodiment, the detection module 903 is specifically used to determine the number of detection items in the region of interest that are consistent with the feature information corresponding to the region of interest; and to obtain the defect detection result of the semi-finished battery based on the number of detection items and the reference number corresponding to each region of interest in the image.

[0195] In one embodiment, the detection module 903 is specifically used to determine that the defect detection result is that the semi-finished battery has a defect when the number of detections corresponding to the region of interest is inconsistent with the reference number; and to determine that the defect detection result is that the semi-finished battery does not have a defect when the number of detections corresponding to each region of interest is consistent with the reference number.

[0196] In one embodiment, the battery detection device 900 may further include:

[0197] The annotation module is used to annotate regions of interest in an image.

[0198] The first determining module is used to determine the target information identifier of the feature information based on the type, feature information and identifier code of the detected item in the region of interest;

[0199] The second determining module is used to determine the comparison result between the target information identifier and the information identifier of the feature information in the first feature information set; the first feature information set is the set of feature information corresponding to the region of interest in the acquired image;

[0200] The third acquisition module is used to acquire the second feature information set based on the comparison result and the first feature information set; the feature information corresponding to the region of interest in the image includes the feature information in the second feature information set.

[0201] In one embodiment, the third acquisition module is specifically used to determine the comparison result corresponding to the next region of interest when the comparison result is that a target information identifier exists in the information identifier corresponding to the first feature information set and there is a labelable next region of interest in the image; and to acquire a second feature information set based on the comparison result corresponding to the next region of interest and the first feature information set; and to determine the first feature information set as the second feature information set when the comparison result is that a target information identifier exists in the information identifier and there is no next region of interest in the image.

[0202] In one embodiment, the battery detection device 900 may further include:

[0203] A module is established to establish a correspondence between the feature information corresponding to the same information identifier as the target information identifier when the comparison result indicates the existence of the target information identifier.

[0204] In one embodiment, the third acquisition module is specifically configured to: when the comparison result indicates that no target information identifier exists, determine that the feature information is the feature information of the detected item in the region of interest, and add the feature information to the first feature information set; when there is a labelable next region of interest in the image, determine the comparison result corresponding to the next region of interest, and obtain a second feature information set based on the comparison result corresponding to the next region of interest and the first feature information set with added feature information; when there is no next region of interest in the image, determine the first feature information set with added feature information as the second feature information set.

[0205] In one embodiment, the module is further configured to establish a correspondence between the region of interest and the feature information in the second feature information set when feature information is added to the first feature information set.

[0206] In one embodiment, the detection item also includes foreign objects.

[0207] Each module in the aforementioned battery testing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the battery testing device in hardware form or independent of it, or stored in the memory of the battery testing device in software form, so that the processor can call and execute the corresponding operations of each module.

[0208] In one embodiment, a battery testing device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of any of the above method embodiments. The technical principles and effects are similar and will not be described again here.

[0209] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of any of the above method embodiments. The technical principles and effects are similar and will not be repeated here.

[0210] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of any of the above method embodiments. The technical principles and effects are similar and will not be repeated here.

[0211] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0212] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0213] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0214] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and not to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and they should all be covered within the scope of the claims and specification of this application. In particular, as long as there is no structural conflict, the various technical features mentioned in the embodiments can be combined in any way. This application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A battery testing method, characterized in that, The method includes: Acquire images of semi-finished batteries obtained during the battery assembly process; Obtain feature information of detected items in the region of interest of the image; the detected items include assemblies and the assembly area before the assemblies are installed. In the region of interest, determine the number of detection items that are consistent with the feature information corresponding to the region of interest; Based on the number of detections and the number of references corresponding to each region of interest in the image, the defect detection results of the semi-finished battery are obtained; The step of obtaining feature information of detected items in the region of interest in the image includes: The region of interest (ROI) in the image is marked; the subclass of the detection item is determined based on the feature information of the detection item in the ROI; the target information identifier of the feature information is determined based on the type of the detection item in the ROI, the subclass of the type, and the identifier code; the comparison result between the target information identifier and the information identifier of the feature information in the first feature information set is determined; the first feature information set is the set of feature information corresponding to the ROI in the acquired image; based on the comparison result and the first feature information set, a second feature information set is obtained; the feature information corresponding to the ROI in the image includes the feature information in the second feature information set, and the target information identifier includes the type of the detection item in the ROI, the subclass of the type, and the identifier code; The step of obtaining a second feature information set based on the comparison result and the first feature information set includes: If the comparison result shows that the target information identifier exists in the information identifier corresponding to the first feature information set, and there is a labelable next region of interest in the image, the comparison result corresponding to the next region of interest is determined, and the second feature information set is obtained based on the comparison result corresponding to the next region of interest and the first feature information set. If the comparison result indicates that the target information identifier exists in the information identifier and the next region of interest does not exist in the image, then the first feature information set is determined to be the second feature information set. If the comparison result indicates that the target information identifier does not exist, the feature information is determined to be the feature information of the detection item in the region of interest, and the feature information is added to the first feature information set; if there is a labelable next region of interest in the image, the comparison result corresponding to the next region of interest is determined, and the second feature information set is obtained based on the comparison result corresponding to the next region of interest and the first feature information set to which the feature information has been added; if there is no next region of interest in the image, the first feature information set to which the feature information has been added is determined to be the second feature information set.

2. The method according to claim 1, characterized in that, The step of obtaining the defect detection result of the semi-finished battery based on the detection quantity and reference quantity corresponding to each region of interest in the image includes: When the number of detections corresponding to the region of interest is inconsistent with the reference number, the defect detection result is determined to be that the semi-finished battery has a defect; When the number of detections corresponding to each region of interest is consistent with the reference number, the defect detection result is determined to be that the semi-finished battery has no defects.

3. The method according to claim 1, characterized in that, The method further includes: If the comparison result indicates the existence of the target information identifier, a correspondence is established between the feature information corresponding to the same information identifier as the region of interest and the target information identifier.

4. The method according to claim 1, characterized in that, The method further includes: When the feature information is added to the first feature information set, a correspondence is established between the region of interest and the feature information in the second feature information set.

5. The method according to claim 1, characterized in that, The detection items also include foreign objects.

6. A battery testing device, characterized in that, The battery testing equipment includes: Image acquisition component, used to acquire images of semi-finished batteries obtained during battery assembly; The processor is configured to acquire feature information of detection items in regions of interest (ROIs) of the image; determine the number of detection items in the ROIs that are consistent with the feature information corresponding to the ROIs; and obtain the defect detection results of the semi-finished battery based on the number of detection items and the reference number corresponding to each ROI in the image; the detection items include assemblies and assembly areas before assembly. The step of obtaining feature information of detected items in the region of interest in the image includes: The region of interest (ROI) in the image is marked; the subclass of the detection item is determined based on the feature information of the detection item in the ROI; the target information identifier of the feature information is determined based on the type of the detection item in the ROI, the subclass of the type, and the identifier code; the comparison result between the target information identifier and the information identifier of the feature information in the first feature information set is determined; the first feature information set is the set of feature information corresponding to the ROI in the acquired image; based on the comparison result and the first feature information set, a second feature information set is obtained; the feature information corresponding to the ROI in the image includes the feature information in the second feature information set, and the target information identifier includes the type of the detection item in the ROI, the subclass of the type, and the identifier code; The step of obtaining a second feature information set based on the comparison result and the first feature information set includes: If the comparison result shows that the target information identifier exists in the information identifier corresponding to the first feature information set, and there is a labelable next region of interest in the image, the comparison result corresponding to the next region of interest is determined, and the second feature information set is obtained based on the comparison result corresponding to the next region of interest and the first feature information set. If the comparison result indicates that the target information identifier exists in the information identifier and the next region of interest does not exist in the image, then the first feature information set is determined to be the second feature information set. If the comparison result indicates that the target information identifier does not exist, the feature information is determined to be the feature information of the detection item in the region of interest, and the feature information is added to the first feature information set; if there is a labelable next region of interest in the image, the comparison result corresponding to the next region of interest is determined, and the second feature information set is obtained based on the comparison result corresponding to the next region of interest and the first feature information set to which the feature information has been added; if there is no next region of interest in the image, the first feature information set to which the feature information has been added is determined to be the second feature information set.

7. A battery testing device, characterized in that, The device includes: The first acquisition module is used to acquire images of semi-finished batteries obtained during the battery assembly process; The second acquisition module is used to acquire feature information of detection items in the region of interest in the image; the detection items include the assembly and the assembly area before the assembly is installed. The detection module is used to determine the number of detection items in the region of interest that are consistent with the feature information corresponding to the region of interest; and to obtain the defect detection result of the semi-finished battery based on the number of detection items and the reference number corresponding to each region of interest in the image. The step of obtaining feature information of detected items in the region of interest in the image includes: The region of interest (ROI) in the image is marked; the subclass of the detection item is determined based on the feature information of the detection item in the ROI; the target information identifier of the feature information is determined based on the type of the detection item in the ROI, the subclass of the type, and the identifier code; the comparison result between the target information identifier and the information identifier of the feature information in the first feature information set is determined; the first feature information set is the set of feature information corresponding to the ROI in the acquired image; based on the comparison result and the first feature information set, a second feature information set is obtained; the feature information corresponding to the ROI in the image includes the feature information in the second feature information set, and the target information identifier includes the type of the detection item in the ROI, the subclass of the type, and the identifier code; The step of obtaining a second feature information set based on the comparison result and the first feature information set includes: If the comparison result shows that the target information identifier exists in the information identifier corresponding to the first feature information set, and there is a labelable next region of interest in the image, the comparison result corresponding to the next region of interest is determined, and the second feature information set is obtained based on the comparison result corresponding to the next region of interest and the first feature information set. If the comparison result indicates that the target information identifier exists in the information identifier and the next region of interest does not exist in the image, then the first feature information set is determined to be the second feature information set. If the comparison result indicates that the target information identifier does not exist, the feature information is determined to be the feature information of the detection item in the region of interest, and the feature information is added to the first feature information set; if there is a labelable next region of interest in the image, the comparison result corresponding to the next region of interest is determined, and the second feature information set is obtained based on the comparison result corresponding to the next region of interest and the first feature information set to which the feature information has been added; if there is no next region of interest in the image, the first feature information set to which the feature information has been added is determined to be the second feature information set.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images