Battery cell detection method, device, system, equipment and storage medium

Abstract

The present disclosure provides a battery cell detection method, device, system, equipment and storage medium, which relates to the field of energy storage technology. The method includes obtaining an image to be processed, which includes a support and multiple battery cell monomers in the support; performing image recognition on the image to be processed, determining the identification angle of the support and the contour area image of multiple battery cell monomers in the image to be processed; recognizing the contour area image of multiple battery cell monomers to obtain the surface identification of each battery cell monomer; and determining the placement result of each battery cell monomer according to the surface identification of each battery cell monomer and the identification angle of the support.

Description

Background Art

[0002] With the continuous improvement of industrial automation and intelligence, remote inspection technology has been widely used in many fields such as manufacturing, energy, and transportation. Existing remote inspection technology usually relies on manual judgment and operation, which greatly limits the efficiency and accuracy of inspection. For example, in the visual inspection of reversed poles of battery cells, manual inspection is not only time-consuming, but also has a high error rate and cannot achieve real-time monitoring and automatic alarm. Summary of the invention

[0003] The present disclosure provides a battery cell detection method, device, system, equipment and storage medium, which at least to a certain extent overcome the problems of time-consuming and high error rate in manual inspection of battery cell single body visual inspection in the related art.

[0004] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0005] According to one aspect of the present disclosure, a battery cell detection method is provided, comprising: acquiring an image to be processed, wherein the image to be processed includes a support and a plurality of battery cell units in the support; performing image recognition on the image to be processed to determine the identification angle of the support and the contour area image of the plurality of battery cell units in the image to be processed; recognizing the contour area image of the plurality of battery cell units to obtain the surface identification of each battery cell unit; and determining the placement result of each battery cell unit according to the surface identification of each battery cell unit and the identification angle of the support.

[0006] According to another aspect of the present disclosure, a battery cell detection device is provided, including: an image acquisition module, used to acquire an image to be processed, wherein the image to be processed includes a support and a plurality of battery cell units in the support; a first recognition module, used to perform image recognition on the image to be processed, and determine the identification angle of the support and the contour area image of the plurality of battery cell units in the image to be processed; a second recognition module, used to recognize the contour area image of the plurality of battery cell units, and obtain the surface identification of each battery cell unit; and a result determination module, used to determine the placement result of each battery cell unit according to the surface identification of each battery cell unit and the identification angle of the support.

[0007] According to another aspect of the present disclosure, a battery cell detection system is also provided, including an image acquisition device and an intelligent detection device, wherein: the image acquisition device is used to acquire an image to be processed, the image to be processed including a support and a plurality of battery cell units in the support; the intelligent detection device is used to obtain the image to be processed; perform image recognition on the image to be processed to determine the identification angle of the support and the contour area image of the plurality of battery cell units in the image to be processed; identify the contour area image of the plurality of battery cell units to obtain the surface identification of each battery cell unit; determine the placement result of each battery cell unit according to the surface identification of each battery cell unit and the identification angle of the support.

[0008] In one embodiment of the present disclosure, the intelligent detection device is used to trigger an abnormal alarm and push abnormal information if the placement result of the battery cell is abnormal; the system also includes a quality management system, which is used to receive the abnormal information pushed by the intelligent detection device, generate an abnormal ledger according to the abnormal information, and push abnormal reminder notifications.

[0009] According to another aspect of the present disclosure, an electronic device is also provided, which includes: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-mentioned battery cell detection methods by executing the executable instructions.

[0010] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the battery cell detection method described in any one of the above is implemented.

[0011] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, any one of the above-mentioned battery cell detection methods is implemented.

[0012] In an embodiment of the present disclosure, an image to be processed is acquired, wherein the image to be processed includes a support and a plurality of battery cells in the support; image recognition is performed on the image to be processed to determine the identification angle of the support and the contour area image of the plurality of battery cells in the image to be processed; the contour area image of the plurality of battery cells is recognized to obtain the surface identification of each battery cell; and the placement result of each battery cell is determined according to the surface identification of each battery cell and the identification angle of the support. The present disclosure can determine whether a battery cell is in an abnormal state according to the identification angle of the support and the surface identification of the battery cell. The detection process is fully automated, which greatly improves the detection efficiency and accuracy, and ensures the safety and efficiency of the production process.

[0013] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0015] Figure 1 A schematic structural diagram of an energy storage system in an embodiment of the present disclosure is shown.

[0016] Figure 2 A flow chart of a battery cell detection method in an embodiment of the present disclosure is shown.

[0017] Figure 3 A flow chart of another battery cell detection method in an embodiment of the present disclosure is shown.

[0018] Figure 4 A schematic diagram of the structure of an image to be processed in an embodiment of the present disclosure is shown.

[0019] Figure 5 A flow chart of a method for determining an identification angle of a support in an embodiment of the present disclosure is shown.

[0020] Figure 6 A flow chart of another battery cell detection method in an embodiment of the present disclosure is shown.

[0021] Figure 7 A flow chart of a method for determining a placement result of a battery cell in an embodiment of the present disclosure is shown.

[0022] Figure 8 A flow chart of another battery cell detection method in an embodiment of the present disclosure is shown.

[0023] Fig. 9 A flow chart showing an example of a battery cell detection method in an embodiment of the present disclosure is shown.

[0024] Fig.10 A schematic diagram of a battery cell detection device in an embodiment of the present disclosure is shown.

[0025] Fig.11 A schematic diagram of a battery cell detection system in an embodiment of the present disclosure is shown.

[0026] Fig.12 A structural block diagram of an electronic device in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0027] Since the energy people need is highly temporal and spatial, in order to rationally use energy and improve energy utilization, it is necessary to use a medium or device to store one form of energy in the same form or convert it into another form of energy, and then release it in a specific form based on future application needs. At present, the main way to generate green electricity is to develop green energy such as photovoltaics and wind power to replace fossil energy.

[0028] At present, the generation of green electricity generally relies on photovoltaics, wind power, and hydropower, etc. However, wind and solar energy generally have problems of strong intermittency and large volatility, which will cause instability in the power grid, insufficient electricity during peak hours, and too much electricity during valley hours. Unstable voltage will also cause damage to electricity. Therefore, insufficient electricity demand or insufficient grid acceptance capacity may cause the problem of "abandoning wind and solar power". To solve these problems, we must rely on energy storage. That is, converting electrical energy into other forms of energy through physical or chemical means and storing it, and converting energy into electrical energy and releasing it when needed. In simple terms, energy storage is like a large "power bank". When photovoltaic and wind energy are sufficient, electrical energy is stored and the stored electricity is released when needed.

[0029] Taking electrochemical energy storage as an example, the present disclosure provides an energy storage device, which has a group of chemical batteries inside. The chemical elements in the batteries are mainly used as energy storage media. The charging and discharging process is accompanied by chemical reactions or changes in the energy storage media. In simple terms, the electricity generated by wind and solar energy is stored in chemical batteries. When the use of external electricity reaches a peak, the stored electricity is released for use, or transferred to places where electricity is scarce for use.

[0030] At present, energy storage (i.e. energy storage) has a wide range of application scenarios, including power generation side energy storage, grid side energy storage and power consumption side energy storage. The corresponding types of energy storage equipment include:

[0031] (1) Large-scale energy storage power stations applied to wind power and photovoltaic power stations can assist renewable energy generation to meet grid connection requirements and improve the utilization rate of renewable energy. As a high-quality active / reactive power source on the power supply side, energy storage power stations can achieve load matching of electric energy in time and space, enhance the ability to absorb renewable energy, reduce instantaneous power changes, reduce the impact on the power grid, improve the absorption of new energy generation, and are of great significance in power grid system backup, relieving peak load power supply pressure, and peak and frequency regulation.

[0032] (2) Energy storage containers used on the grid side mainly perform peak load regulation, frequency regulation, and grid congestion relief. Peak load regulation can achieve peak load shaving and valley filling of electricity load, that is, charging the energy storage battery when the electricity load is low, and releasing the stored electricity during the peak load period, thereby achieving a balance between electricity production and consumption, such as energy storage power station systems;

[0033] (3) Small energy storage cabinets used on the power consumption side, mainly for self-generation and self-use of electricity, peak-valley price arbitrage, capacity cost management and improving power supply reliability. According to different application scenarios, energy storage on the power consumption side can be divided into industrial and commercial energy storage cabinets, household energy storage equipment, energy storage charging piles, etc., which are generally used in conjunction with distributed photovoltaics. Industrial and commercial users can use energy storage for valley-peak price arbitrage and capacity cost management. In the power market that implements peak-valley electricity prices, by charging the energy storage system when the electricity price is low and discharging the energy storage system when the electricity price is high, peak-valley electricity price arbitrage is achieved and electricity costs are reduced. In addition, industrial enterprises that are subject to a two-part electricity price can use energy storage systems to store energy during low electricity consumption and discharge it during peak loads, thereby reducing peak power and the maximum demand reported, and achieving the purpose of reducing capacity electricity charges. Household photovoltaic storage can improve the level of self-generation and self-use of electricity. Due to high electricity prices and poor power supply stability, the demand for household photovoltaic installations has been driven. Considering that photovoltaic power is generated during the day, and users generally have higher loads at night, energy storage can be configured to better utilize photovoltaic power, improve the level of self-generation and self-use, and reduce electricity costs. In addition, communication base stations, data centers and other fields need to configure energy storage for backup power.

[0034] See also Figure 1 , Figure 1 is a schematic diagram of the structure of an energy storage system according to an embodiment of the present disclosure, and Figure 1 The embodiment is described by taking the shared energy storage scenario on the generation / distribution side as an example, but the energy storage system disclosed in the present invention is not limited to the energy storage scenario on the generation / distribution side.

[0035] The present disclosure provides an energy storage system, which includes: an energy storage device 110, a high-voltage cable 120, a first electric energy conversion device 130, and a second electric energy conversion device 140. In the case of power generation, the first electric energy conversion device 130 and the second electric energy conversion device 140 are used to convert other forms of energy into electric energy, which is connected to the high-voltage cable 120 and supplied to the power distribution network for use. When the power load is low and the first conversion device 130 and the second electric energy conversion device 140 generate excess power, the excess power is stored in the energy storage device 110, thereby reducing the wind and solar power abandonment rates and improving the problem of new energy power generation and consumption. When the power load is high, the power grid issues an instruction to transmit the power stored in the energy storage device 110 in conjunction with the high-voltage cable 120 in a grid-connected mode to the power consumption side, thereby providing peak load regulation, frequency regulation, standby and other services for the operation of the power grid, giving full play to the peak load regulation of the power grid, promoting peak load reduction and valley filling of the power grid, and alleviating the power supply pressure of the power grid.

[0036] Optionally, the first electric energy conversion device 130 and the second electric energy conversion device 140 can convert at least one of solar energy, light energy, wind energy, thermal energy, tidal energy, biomass energy and mechanical energy into electric energy.

[0037] The number of energy storage devices 110 can be multiple, and multiple energy storage devices 110 are connected in series or in parallel. Multiple energy storage devices 110 are supported and electrically connected by isolation plates (not shown). In this embodiment, "multiple" refers to two or more. An energy storage box can also be provided outside the energy storage device 110 to accommodate the energy storage device 110.

[0038] Optionally, the energy storage device 110 may include but is not limited to battery cells, battery modules, battery packs, battery systems, etc. The actual application form of the energy storage device 110 provided in the embodiment of the present disclosure may be but is not limited to the listed products, and may also be other application forms. The embodiment of the present disclosure does not strictly limit the application form of the energy storage device 110. The embodiment of the present disclosure only takes the energy storage device 110 as a battery cell as an example for explanation. When the energy storage device 110 is a battery cell, the energy storage device 110 may be at least one of a cylindrical battery, a square battery, etc.

[0039] In the related art, for visual detection of reverse pole placement, manual inspection is time-consuming and has a high error rate, and it is impossible to achieve real-time monitoring and automatic alarm.

[0040] Although some remote inspection systems use artificial intelligence AI and cameras for image acquisition and preliminary analysis, they still have deficiencies in standard determination, abnormal alarms, and abnormal push notifications. For example, the lack of accurate determination standards may lead to false alarms or missed alarms; the response time for abnormal alarms is too long and real-time processing cannot be achieved; abnormal push notifications often rely on manual operations and lack automation and intelligence.

[0041] In order to solve at least some of the above-mentioned technical problems, the present invention provides a battery cell detection method, which obtains an image to be processed, wherein the image to be processed includes a support and multiple battery cell monomers in the support; performs image recognition on the image to be processed to determine the identification angle of the support and the contour area image of the multiple battery cell monomers in the image to be processed; identifies the contour area image of the multiple battery cell monomers to obtain the surface identification of each battery cell monomer; determines the placement result of each battery cell monomer according to the surface identification of each battery cell monomer and the identification angle of the support. The present invention can determine whether the battery cell monomer is in an abnormal state according to the identification angle of the support and the surface identification of the battery cell monomer. The detection process is fully automated, which greatly improves the detection efficiency and accuracy, and ensures the safety and efficiency of the production process.

[0042] The battery cell detection method provided by the present disclosure can be applied to multiple battery cells placed on a support surface to detect whether the poles of multiple battery cells are placed inversely. It can also be applied to multiple battery cells placed in a battery box to detect whether the poles of multiple battery cells are placed inversely.

[0043] Under the above system architecture, a method for detecting a battery cell is provided in an embodiment of the present disclosure, and the method can be executed by any electronic device with computing and processing capabilities, such as a server or a processor.

[0044] Figure 2 A flow chart of a battery cell detection method in an embodiment of the present disclosure is shown as follows: Figure 2 As shown, the battery cell detection method provided in the embodiment of the present disclosure includes the following steps:

[0045] S202, obtaining an image to be processed, where the image to be processed includes a support and a plurality of battery cells in the support.

[0046] In one embodiment, the image to be processed can be obtained by photographing multiple battery cells and supports through an image acquisition device. For example, the image acquisition device can be a camera, which can be set on the top and / or side of the battery cell to take pictures of the multiple battery cells and supports; the camera can also take a video of the multiple battery cells and supports, and obtain the image to be processed by extracting frames from the video.

[0047] The support may be a foam board having a plurality of grooves arranged in an array. The battery cells are rectangular cells, and the plurality of battery cells are placed in corresponding grooves respectively. For example, the battery cells are placed in the grooves of the foam board in 2 rows and 5 columns.

[0048] The support may also be a battery box, in which a plurality of battery cells are arranged according to preset rules. The preset rules include that the battery cells are arranged in an array, and the positive and negative pole positions of each battery cell are determined according to actual needs.

[0049] It should be noted that the arrangement and placement of the battery cells on the support may be determined according to actual needs, and the number of battery cells may also be determined according to actual needs, and the present disclosure does not make specific limitations on this.

[0050] S204, performing image recognition on the image to be processed to determine the identification angle of the support and the contour area images of the plurality of battery cells in the image to be processed.

[0051] The support may be in a rectangular, square or other shape, and a rectangle is used as an example for explanation below. The identification angle of the support is a mark set on the support to determine the placement result of the battery cell. The identification angle of the support may be one of the corners of the rectangular support. For example, the identification angle of the support may be the missing corner of the support, or the R angle of the support, and the missing corner or the R angle may be different from the area, angle, etc. of the remaining three corners.

[0052] The contour area image of the plurality of battery cells refers to an image corresponding to the area where the contours of the plurality of battery cells distributed in an array are located, that is, an image obtained by trimming the supporting objects outside the contours of the plurality of battery cells from the image to be processed.

[0053] In one embodiment, in S204, a target detector can be used to perform image recognition on the image to be processed. The target detector can be a target detection algorithm such as R-CNN, Fast R-CNN, Faster R-CNN, YOLO (You Only Look Once, a target detection algorithm belonging to the field of deep learning, used for real-time object detection in computer vision tasks), so as to realize the recognition of multiple battery cells and the identification angle of the support.

[0054] S206, identifying the contour area images of the plurality of battery cells to obtain the surface identification of each battery cell.

[0055] In one embodiment, the surface identification of the battery cell is a QR code set on the top cover of the battery cell. The QR code is used to determine detailed information of the battery cell. For example, by scanning the QR code, the production batch of the battery cell, the battery cell code and other information can be determined.

[0056] The QR code is set between the two poles of the battery cell, and the QR code is close to one of the poles. For example, the QR code is close to the positive pole of the battery cell.

[0057] In one embodiment, the surface identification of the battery cell may also be a QR code and an explosion-proof valve disposed on the top cover of the battery cell, the explosion-proof valve being disposed at the midpoint of the two poles of the battery cell, the QR code being disposed between the explosion-proof valve and the target pole, the target pole being one of the poles of the battery cell, such as the positive pole. For example, the QR code is disposed between the explosion-proof valve and the positive pole.

[0058] S208, determining the placement result of each battery cell according to the surface mark of each battery cell and the mark angle of the support.

[0059] In one embodiment, the placement result of the battery cell can be characterized by whether the poles of the battery cell are placed upside down. The poles being placed upside down means that the positive poles and negative poles of the battery cell are placed upside down. When the poles of the battery cell are not placed upside down, the placement result of the battery cell is normal; when the poles of the battery cell are placed upside down, the placement result of the battery cell is abnormal.

[0060] When the surface identification of the battery cell includes a two-dimensional code, it can be determined whether the poles of each battery cell are placed upside down based on the relative position of the two-dimensional code and the identification angle of the support.

[0061] It should be noted that for surface markings including QR codes, the target pole of the battery cell can also be determined based on the distance between the QR code and the two poles, and then based on the relative position between the target pole and the marking angle of the support, it can be determined whether the poles of each battery cell are placed upside down.

[0062] When the surface identification of the battery cell includes a QR code and an explosion-proof valve, the position of the target pole of the battery cell can be determined according to the QR code and the explosion-proof valve, and then according to the relative position between the target pole and the identification angle of the support, it can be determined whether the poles of each battery cell are placed upside down.

[0063] In an embodiment of the present disclosure, an image to be processed is acquired, wherein the image to be processed includes a support and a plurality of battery cells in the support; image recognition is performed on the image to be processed to determine the identification angle of the support and the contour area image of the plurality of battery cells in the image to be processed; the contour area image of the plurality of battery cells is recognized to obtain the surface identification of each battery cell; and the placement result of each battery cell is determined according to the surface identification of each battery cell and the identification angle of the support. The present disclosure can determine whether a battery cell is in an abnormal state according to the identification angle of the support and the surface identification of the battery cell. The detection process is fully automated, which greatly improves the detection efficiency and accuracy, and ensures the safety and efficiency of the production process.

[0064] In one embodiment, after acquiring the image to be processed in S202, the battery cell detection method provided by the present disclosure further includes: preprocessing the image to be processed, wherein the preprocessing includes at least one of applying a mask to filter background information and performing a perspective transformation to correct geometric deformation of the image to be processed.

[0065] By establishing grinding in Python or ENVI, the background value is set to 0 to achieve the purpose of filtering background information.

[0066] OpenCN can be used to correct the image, and the correction of the image to be processed can be completed through the perspective matrix and perspective change.

[0067] It should be noted that other types of preprocessing may be performed on the image to be processed, such as adjusting the size, cropping, rotating, etc. of the image to be processed to adapt to subsequent model training and model reasoning, and the present disclosure does not make specific limitations on this.

[0068] In the embodiments of the present disclosure, by preprocessing the image to be processed, irrelevant background information can be effectively filtered, the deformation of the image to be processed can be adjusted, etc., thereby improving the effectiveness of battery cell detection.

[0069] Figure 3 Another flow chart of a battery cell detection method in an embodiment of the present disclosure is shown. Figure 2Based on the embodiment, S204 is further refined into S2042 to limit the method of determining the contour area images of the plurality of battery cells. Figure 3 As shown, in one embodiment, the battery cell detection method of the embodiment of the present disclosure includes S202, S2042, S206~S208. The method includes:

[0070] S2042: Use an object detector to perform image recognition on the battery cells in the image to be processed, determine the four corner coordinates of the plurality of battery cells, and cut out contour area images of the plurality of battery cells.

[0071] It should be noted that the specific implementation of S202 and S206 to S208 of this embodiment is the same as the specific implementation of S202 and S206 to S208 in the aforementioned embodiment, and will not be repeated here.

[0072] The target detector is used for image recognition of contour areas of multiple battery cells. The target detector can be at least one of YOLO, R-CNN, Fast R-CNN, and Faster R-CNN.

[0073] The object detector is trained based on the prediction boxes of historical images and images of contour areas of multiple battery cells.

[0074] The four-corner coordinates of the plurality of battery cells arranged in an array refer to the coordinate values ​​of the four vertices of the rectangular outline area formed by the plurality of battery cells.

[0075] like Figure 4 As shown, part of the image to be processed includes a support 210 and 10 battery cells 220, wherein the missing corner 211 at the upper left corner of the support 210 is the identification corner of the support 210, and the 10 battery cells 220 are arranged in an array of 2 rows and 5 columns, each battery cell 220 includes two poles, a QR code 222 is arranged between the two poles, and the QR code 222 is close to the target electrode 221.

[0076] In the disclosed embodiment, the target detector can quickly and accurately locate the contour areas of multiple battery cells in the image to be processed, thereby narrowing the analysis scope.

[0077] Figure 5 A flow chart of a method for determining the identification angle of a support in an embodiment of the present disclosure is shown. Figure 5 As shown, in one embodiment, the above S204 performs image recognition on the image to be processed to determine the identification angle of the support and the contour area images of multiple battery cells in the image to be processed, including:

[0078] S502, using a target detector to perform image recognition on a support in the image to be processed, and obtaining a corner point region image of the support, where the corner point region image is an image of a region surrounded by multiple corner points obtained by recognizing the support;

[0079] S504, extracting features from the corner point region image of the support to obtain feature values ​​of the four corners of the corner point region;

[0080] S506: Determine the identification angle of the support according to the characteristic values ​​of the four corners of the corner point area.

[0081] The corner point region image of the support is an image obtained by cutting out the background region in the image to be processed. The corner point region image is similar to a rectangle, and one of the top corners of the rectangle is set as a marker corner.

[0082] The corner points of the support are also called interest points, feature points, etc. Corner points are extreme points, points where some attributes are particularly prominent. Corner points can also be points at the intersection of two lines, or points located on two adjacent objects with different main directions. In the present disclosure, by identifying the corner points of the support, the area corresponding to the outer contour of the support can be determined by connecting each corner point in sequence, thereby obtaining a corner point area image.

[0083] It should be noted that, since the support may be deflected by a certain angle when the image acquisition device is shooting, part of the background image may exist in the corner area image.

[0084] In one embodiment, in the corner point area image, the corner point area image is binarized, and the pixel values ​​of 1 / 4 of the corners are counted as the feature values ​​of the four corners of the corner point area image, so as to determine the identification angle of the support according to the feature values ​​of the four corners of the corner point area image.

[0085] The identification angle of the support can be determined based on the relationship between the characteristic values ​​of the four corners of the corner point area image and the preset pixel threshold, wherein the preset pixel threshold can be determined according to the actual situation. For example, when the characteristic value of a corner in the corner point area image is less than the preset pixel threshold, the corner is determined to be the identification angle of the support.

[0086] The area corresponding to 1 / 4 of the pixels in the corner can also be calculated as the characteristic values ​​of the four corners of the corner point area image. The identification angle of the support can be determined based on the relationship between the characteristic values ​​of the four corners of the corner point area image and the preset face value. The above preset area threshold can be determined according to the actual situation.

[0087] In the disclosed embodiment, by extracting features from the corner area image of the support, the identification angle of the support is determined according to the feature values ​​of the four corners, thereby providing a data basis for the subsequent accurate identification of the reversed pole situation.

[0088] In one embodiment, the battery cell detection method disclosed herein also includes: determining a closed area of ​​a corner point area image; if the area of ​​the closed area is greater than a preset area threshold, re-determining the corner point area image of the support object so that the area of ​​the closed area corresponding to the re-determined corner point area image of the support object is less than or equal to the preset area threshold.

[0089] The closed area of ​​the corner area image is determined by findContour, and whether further box selection is needed is determined by comparing the relationship between the area of ​​the closed area and the preset area threshold. The above preset area threshold can be determined according to the actual situation.

[0090] When the area of ​​the closed area is larger than the preset area threshold, the corner point area image will continue to be cropped to obtain a more accurate box area, and the area of ​​the closed area corresponding to the cropped corner point area image will be re-checked to see if it is larger than the preset area threshold. Through multiple adjustments, the area of ​​the closed area corresponding to the corner point area image is made less than or equal to the preset area threshold, thereby accurately determining the contour area of ​​the support.

[0091] It should be noted that the relationship between the ratio of the area of ​​the closed region to the surface area of ​​the support and a preset ratio threshold may also be used to determine whether to make further box selections.

[0092] Figure 6 A flow chart of another battery cell detection method in an embodiment of the present disclosure is shown. Figure 2 Based on the embodiment, S206 is further defined as S2062 to define the method for determining the surface identification of each battery cell. Figure 6 As shown, the battery cell detection method of the embodiment of the present disclosure includes S202~S204, S2062 and S208. The method includes:

[0093] S2062, identifying the contour area images of the plurality of battery cells by using an image processing model to determine the surface identification of each battery cell, wherein the image processing model is trained based on the contour area images of the plurality of battery cells and corresponding surface identification labels.

[0094] In one embodiment, the image processing model may be a convolutional neural network model (CNN) and an attention mechanism Transformer. CNN may be at least one of a residual network, VGG, DenseNET, VGG16, AlexNet, etc., wherein VGG is a classic convolutional neural network architecture. Transformer may include at least one of Vision Transformer, Swin Transformer, and Detection Transformer. Vision Transformer is an image classification model.

[0095] The image processing model is a trained deep learning network model. The image processing model can process the contour area images of multiple battery cells input by the target detector and output the surface identification of each battery cell. The image processing model training process is described later.

[0096] The image processing model has powerful feature extraction capabilities and can identify the surface markings of battery cells, thereby providing basic information for subsequent abnormality judgments and improving the effectiveness and reliability of battery cell detection.

[0097] Figure 7 A flow chart of a method for determining the placement result of a battery cell in an embodiment of the present disclosure is shown. Figure 7 As shown, in one embodiment, the above S208 determines the placement result of each battery cell according to the surface mark of each battery cell and the mark angle of the support, including:

[0098] S2082. If the surface mark of the battery cell and the mark angle of the support satisfy a first preset condition, the placement result of the battery cell is determined to be normal; if the surface mark of the battery cell and the mark angle of the support do not satisfy the first preset condition, the placement result of the battery cell is determined to be abnormal.

[0099] It should be noted that the first preset condition can be pre-configured in the server or processor. It can be determined whether the relative position relationship between the identification angle of the support and the surface identification of the battery cell monomer meets the first preset condition.

[0100] The above-mentioned normal placement may be a situation where the positive electrode column and the negative electrode column of the battery cell are placed in the correct position according to regulations, and the abnormal placement refers to a situation where the positive electrode column and the negative electrode column of the battery cell are placed upside down according to regulations.

[0101] In one embodiment, the battery cell detection method of the embodiment of the present disclosure further includes: determining a placement manner of the battery cell unit on the support, and determining a first preset condition according to the placement manner of the battery cell unit on the support.

[0102] The placement of the above-mentioned multiple battery cells on the support includes horizontal placement and vertical placement, such as Figure 4 As shown, the battery cell is in a horizontal manner, that is, when the identification corner of the support is located at the upper left corner of the image to be processed, the battery cell is in the horizontal direction of the image to be processed; the vertical placement of the battery cell refers to when it is at the upper left corner of the image to be processed, the battery cell is in the vertical direction of the image to be processed.

[0103] The first preset condition may include that the surface mark of the battery cell is away from the mark angle of the support, or may be that the surface mark of the battery cell is close to the mark angle of the support.

[0104] In one embodiment, the method can also determine whether the first preset condition is met in the following manner: taking the identification angle of the support as a reference, if the surface identification of the battery cell is located within the preset identification frame, then it is determined that the surface identification of the battery cell and the identification angle of the support meet the first preset condition; if the surface identification of the battery cell is not located within the preset identification frame, then it is determined that the surface identification of the battery cell and the identification angle of the support do not meet the first preset condition.

[0105] It should be noted that corresponding identification frames are pre-configured for the multiple battery cells on the support. The specific positions of the identification frames are set corresponding to the respective battery cells, representing the target positions of the surface identification of the battery cells. The identification frames may be rectangular frames, circular frames, elliptical frames, etc.

[0106] In one embodiment, a plurality of battery cells are arranged in an array on a support, and an identification corner of the support is located at one of the top corners of the support. The identification corner may include a notched corner, an R corner, or a chamfer.

[0107] There is no specific limitation on the size of the missing corner, R angle or chamfer, as long as the identification angle can be distinguished from other top angles of the support.

[0108] Taking the identification corner of the support as a reference means that the identification corner of the support is located at the upper left corner of the image to determine the position of the preset identification frame of each battery cell in the image.

[0109] It should be noted that the identification corner of the support may also be located at the upper right corner of the image, the position of the preset identification frame in the image may be determined, and adaptive adjustment may be made to determine whether the first preset condition is met. The present disclosure does not make any specific limitation on this.

[0110] It should be noted that the surface identification includes a QR code of the battery cell, and the QR code is close to one of the poles of the battery cell. For example, the QR code can be set close to the positive pole of the battery cell, or it can be set close to the negative pole. By placing the QR code close to one of the poles, the two poles of the battery cell can be accurately distinguished, providing basic data for subsequent judgment of the placement of the battery cell, thereby improving detection efficiency and accuracy.

[0111] In a feasible implementation, based on the first preset condition, the set position of the surface identification of each battery cell in the contour area image can be determined, the contour area images of multiple battery cell cells can be identified, and it is determined whether each battery cell has a surface identification at the set position. If so, the placement result of the battery cell is determined to be normal; if not, the placement result of the battery cell is determined to be abnormal.

[0112] The surface identification may also include an explosion-proof valve and a QR code. The QR code is set near one of the poles of the battery cell, and the explosion-proof valve is set between the two poles of the battery cell. For example, the QR code can be set near the positive pole of the battery cell, or it can be set near the negative pole, and the explosion-proof valve is set at the midpoint of the line connecting the two poles. The placement of the battery cell can be comprehensively judged through the QR code and the identification angle of the explosion-proof valve and the support.

[0113] For example, continue to refer to Figure 4 , the single-cell monomer is placed horizontally, and the QR code is set between the explosion-proof valve and the target pole (one of the above-mentioned poles). If the angle between the identification angle of the support, the explosion-proof valve and the QR code is greater than 90°, it means that the directions of the QR code and the identification angle are opposite, the surface identification of the battery cell monomer is far away from the identification angle of the support, and the battery cell monomer is placed normally; if the angle between the identification angle of the support, the explosion-proof valve and the QR code is less than 90°, the directions of the surface QR code and the identification angle are the same, the surface identification of the battery cell monomer is close to the identification angle of the support, and the battery cell monomer is placed abnormally.

[0114] The case where the battery cell is placed vertically is similar to the case where it is placed horizontally, and will not be described in detail here.

[0115] In the disclosed embodiment, by combining the identification angle of the support and the surface identification of each battery cell to detect the pole installation condition of the battery cell, abnormal conditions such as reverse pole placement can be automatically identified without human intervention, greatly improving the accuracy and efficiency of detection.

[0116] In one embodiment, the above S208 determines the placement result of each battery cell according to the surface marking of each battery cell and the marking angle of the support, including: determining the target pole of each battery cell according to the surface marking of each battery cell; determining the placement result of each battery cell according to the marking angle of the support and the target pole of each battery cell.

[0117] The target pole of the battery cell is used to detect whether the pole of the battery cell is installed normally in combination with the identification angle of the support.

[0118] The target pole may be any one of the two poles of the battery cell. It is only necessary to place the surface mark of the battery cell close to the target pole. For example, the target pole may be the positive pole or the negative pole.

[0119] It should be noted that the pole far away from the surface mark of the battery cell may also be used as the target pole, and the second preset condition and other detection standards may be adaptively adjusted.

[0120] In the disclosed embodiment, by combining the identification angle of the support and the target pole of each battery cell to detect the pole installation condition of the battery cell, abnormal conditions such as reverse pole placement can be automatically identified without human intervention, greatly improving the accuracy and efficiency of detection.

[0121] In one embodiment, the above-mentioned determination of the placement result of each battery cell based on the identification angle of the support and the target pole of each battery cell includes: when the identification angle of the support and the target pole of the battery cell meet the second preset condition, determining that the placement result of the battery cell is a normal installation; when the identification angle of the support and the target pole of the battery cell do not meet the second preset condition, determining that the placement result of the battery cell is an abnormal installation.

[0122] It should be noted that the second preset condition can be pre-configured in the server or processor. It can be determined whether the relative position relationship between the identification angle of the support and the target pole of the battery cell meets the second preset condition. For example, when the target pole of the battery cell is closer to the identification angle of the support than the other pole of the battery cell, it is determined that the identification angle of the support and the target pole of the battery cell do not meet the second preset condition, and the placement result of the battery cell is abnormal installation, that is, the pole of the battery cell is placed upside down. If the target pole of the battery cell is far away from the identification angle of the support, the placement result of the battery cell is determined to be normal installation.

[0123] Specifically, the identification angle of the support and the distance between the target pole and another pole of the battery cell can be calculated respectively. When the distance corresponding to the target pole of the battery cell is greater than the distance corresponding to the other pole, it is determined that the identification angle of the support and the target pole of the battery cell meet the second preset condition, and the placement result of the battery cell is normal.

[0124] In one embodiment, the battery cell detection method in the embodiment of the present disclosure also includes: determining the placement of multiple battery cell units on a support; determining a second preset condition based on the placement of multiple battery cell units on the support, the second preset condition including an identification angle of a target pole of the battery cell unit away from the support.

[0125] It should be noted that when determining whether the target pole is far away from the identification angle of the support by calculating the distance between the pole and the identification angle of the support, the distance between the target pole of each battery cell and the identification angle of the support can also be calculated separately, and by comparing the relationship between the above distance and the preset distance threshold, it is determined whether the target pole of the battery cell is far away from the identification angle of the support. In this case, it is necessary to set a corresponding preset distance threshold for each battery cell.

[0126] The surface identification includes a QR code of the battery cell, or a QR code and an explosion-proof valve, wherein the QR code and the explosion-proof valve are arranged between two poles of the battery cell, and the QR code is arranged between the explosion-proof valve and a target pole of the two poles.

[0127] In the disclosed embodiment, through comprehensive analysis of the identification angle of the support and the target pole of the battery cell, the pole installation status of the battery cell can be accurately determined, which is fully automated, greatly improves the detection efficiency and accuracy, and ensures the safety and efficiency of the production process.

[0128] Figure 8 A flow chart of another battery cell detection method according to an embodiment of the present disclosure is shown. Figure 2 On the basis of the embodiment, after S208, S210 is added to limit the situations in which abnormal warning is performed. Figure 8 As shown, the battery cell detection method of the embodiment of the present disclosure includes S202 to S210. The method includes:

[0129] S210: If the placement result of the battery cell is abnormal, an abnormal alarm is triggered and abnormal information is pushed.

[0130] In one embodiment, when it is determined that the battery cell is placed abnormally, an alarm mechanism is triggered, for example, a visual alarm is triggered, a red warning box is displayed on the screen, or an audio alarm is triggered, a voice alarm is issued through a speaker.

[0131] When an abnormal placement is detected, an automatic push mechanism can be activated to push detailed abnormal information to the quality management system (QMS). The abnormal information may include but is not limited to the specific event, location, image acquisition device name, inspection items, and real-time images of the site, thereby providing rich data support for subsequent analysis and processing.

[0132] In the embodiments of the present disclosure, real-time monitoring and alarm can be achieved. By collecting images in real time and performing an abnormality determination periodically, once an abnormality is detected, an alarm mechanism is immediately triggered, including sound and light alarms and remote notifications, to ensure that the abnormality can be handled in a timely manner. The real-time monitoring and rapid response capabilities are strong, which greatly improves the safety and reliability of the production process.

[0133] In order to deepen the understanding of the battery cell detection method provided by the present disclosure, the following Fig. 9 Provide detailed explanation.

[0134] like Fig. 9 As shown, the battery cell detection method disclosed in the present invention includes a VGG model training process, a YOLO model training process and a model reasoning process.

[0135] Among them, the VGG model training process includes:

[0136] S901, creation of a data set: collecting images containing multiple battery cells to form an image data set, ensuring the diversity and representativeness of the data set.

[0137] S902, data set preprocessing: preprocess the images in the data set, such as adjusting the image size, cropping, rotating, etc., to adapt to subsequent VGG model training and model inference.

[0138] S903. Pre-train VGG model.

[0139] S904: Obtain a batch size of data set.

[0140] S905: Random data enhancement.

[0141] S906: Train and record the loss function value.

[0142] S907. Determine whether the loss function value after multiple rounds of training has not decreased. If so, execute S908. If not, return to S904 and retrain the VGG model.

[0143] S908. Save the optimal VGG model.

[0144] It should be noted that in S901, the data set for training the VGG model can use the contour area images of multiple battery cells cropped in the above embodiment, and use a labeling box to mark the surface identification of the battery cell in the contour area image as the corresponding surface identification label.

[0145] The loss function of the VGG model can be a cross entropy loss function, a mean square error loss function, an L1 loss function, a Smooth L1 loss function, etc. The loss function value is determined by the surface identification label and the surface identification predicted by the VGG model.

[0146] The training rounds of the VGG model can be determined according to actual needs, and this disclosure does not make any specific limitations on this.

[0147] The YOLO model training process includes:

[0148] S909: Creation of data set.

[0149] S910, Labelme marked prediction box.

[0150] S911. Train the YOLO model.

[0151] S912: Save the optimal YOLO model.

[0152] It should be noted that in S909, the images in the data set may be the images to be processed in the aforementioned embodiment and the images after image recognition is performed on the images to be processed.

[0153] The model inference process includes:

[0154] S913: Obtain an image to be processed.

[0155] S914: Image preprocessing.

[0156] S915, using YOLO to detect the pre-processed image to be processed, and obtain contour area images of multiple battery cells and missing corners of the support.

[0157] S916: Use the VGG model to process the contour area images of the plurality of battery cells to determine the QR code of each battery cell.

[0158] In addition, the method also includes determining the placement result of the battery cell based on the missing corner of the support and the QR code of the battery cell, which is not shown in the drawings.

[0159] When the battery cell is placed abnormally, an abnormal alarm is triggered and abnormal information can be pushed.

[0160] Based on the same inventive concept, the present disclosure also provides a battery cell detection device in the following embodiments. Since the principle of solving the problem in the device embodiment is similar to that in the above method embodiment, the implementation of the device embodiment can refer to the implementation of the above method embodiment, and the repeated parts will not be repeated.

[0161] Fig.10 A schematic diagram of a battery cell detection device in an embodiment of the present disclosure is shown. Fig.10As shown, the device includes an image acquisition module 1010, a first recognition module 1020, a second recognition module 1030 and a result determination module 1040. The image acquisition module 1010 is used to acquire an image to be processed, which includes a support and a plurality of battery cells in the support; the first recognition module 1020 is used to perform image recognition on the image to be processed, and determine the identification angle of the support and the contour area image of the plurality of battery cells in the image to be processed; the second recognition module 1030 is used to recognize the contour area image of the plurality of battery cells and obtain the surface identification of each battery cell; the result determination module 1040 is used to determine the placement result of each battery cell according to the surface identification of each battery cell and the identification angle of the support.

[0162] It should be noted that the image acquisition module 1010, the first recognition module 1020, the second recognition module 1030 and the result determination module 1040 correspond to S202 to S208 in the method embodiment, and the examples and application scenarios implemented by the modules and the corresponding steps are the same, but are not limited to the contents disclosed in the method embodiment. It should be noted that the modules as part of the device can be executed in a computer system such as a set of computer executable instructions.

[0163] In one embodiment, the result determination module 1040 is used to determine that the placement result of the battery cell is normal if the surface mark of the battery cell and the mark angle of the support meet the first preset condition; if the surface mark of the battery cell and the mark angle of the support do not meet the first preset condition, then determine that the placement result of the battery cell is abnormal.

[0164] In one embodiment, the device further includes a first condition determination module not shown in the drawings, the first condition determination module is used to determine the placement of the battery cell on the support, and determine the first preset condition according to the placement of the battery cell on the support.

[0165] In one embodiment, the result determination module 1040 is also used to use the identification angle of the support as a reference. If the surface identification of the battery cell is located within a preset identification frame, it is determined that the surface identification of the battery cell and the identification angle of the support meet a first preset condition; if the surface identification of the battery cell is not located within the preset identification frame, it is determined that the surface identification of the battery cell and the identification angle of the support do not meet the first preset condition.

[0166] It should be noted that multiple battery cells are arranged in an array on a support; the identification corner of the support is located at one of the top corners of the support, and the identification corner includes a notched corner, an R angle or a chamfered angle; the surface marking includes a QR code of the battery cell, or a QR code and an explosion-proof valve, the QR code is close to one of the poles of the battery cell, and the explosion-proof valve is arranged between the two poles of the battery cell.

[0167] In one embodiment, the device further includes an early warning module which is not shown in the drawings. The early warning module is used to trigger an abnormal alarm and push abnormal information if the placement result of the battery cell monomer is abnormal.

[0168] In one embodiment, the first recognition module 1020 is used to use an object detector to perform image recognition on the battery cells in the image to be processed, determine the four corner coordinates of the multiple battery cells, and cut out contour area images of the multiple battery cells.

[0169] In one embodiment, the first recognition module 1020 is used to use a target detector to perform image recognition on the support in the image to be processed to obtain a corner point area image of the support, which is an image of the area surrounded by multiple corner points obtained by identifying the support; perform feature extraction on the corner point area image of the support to obtain feature values ​​of the four corners of the corner point area; and determine the identification angle of the support based on the feature values ​​of the four corners of the corner point area.

[0170] In one embodiment, the first recognition module 1020 is also used to determine the closed area of ​​the corner point area image; if the area of ​​the closed area is greater than a preset area threshold, the corner point area image of the support is re-determined so that the area of ​​the closed area corresponding to the re-determined corner point area image of the support is less than or equal to the preset area threshold.

[0171] In one embodiment, the second recognition module 1030 is used to recognize contour area images of multiple battery cells through an image processing model to determine the surface identification of each battery cell. The image processing model is trained based on contour area images of multiple battery cells and corresponding surface identification labels.

[0172] In one embodiment, the device also includes a preprocessing module not shown in the accompanying drawings, which is used to preprocess the image to be processed after acquiring the image to be processed. The preprocessing includes at least one of applying a mask to filter background information and performing a perspective transformation to correct the geometric deformation of the image to be processed.

[0173] In an embodiment of the present disclosure, an image to be processed is acquired, wherein the image to be processed includes a support and a plurality of battery cells in the support; image recognition is performed on the image to be processed to determine the identification angle of the support and the contour area image of the plurality of battery cells in the image to be processed; the contour area image of the plurality of battery cells is recognized to obtain the surface identification of each battery cell; and the placement result of each battery cell is determined according to the surface identification of each battery cell and the identification angle of the support. The present disclosure can determine whether a battery cell is in an abnormal state according to the identification angle of the support and the surface identification of the battery cell. The detection process is fully automated, which greatly improves the detection efficiency and accuracy, and ensures the safety and efficiency of the production process.

[0174] Fig.11 FIG. 1 is a schematic diagram showing a structure of a battery cell detection system provided by an embodiment of the present disclosure. Fig.11 As shown, in one embodiment, the battery cell detection system provided by the present disclosure includes an image acquisition device 1110 and an intelligent detection device 1120, wherein:

[0175] An image acquisition device 1110 is used to acquire an image to be processed, wherein the image to be processed includes a support and a plurality of battery cells in the support;

[0176] The intelligent detection device 1120 is used to obtain an image to be processed; perform image recognition on the image to be processed to determine the identification angle of the support and the contour area image of multiple battery cells in the image to be processed; identify the contour area image of multiple battery cells to obtain the surface identification of each battery cell; and determine the placement result of each battery cell according to the surface identification of each battery cell and the identification angle of the support.

[0177] In one embodiment, the image acquisition device 1110 may be a camera or a video camera that collects images in real time. In order to ensure that the camera can clearly capture the position of the pole of the battery cell in real time and provide an accurate data basis for subsequent AI analysis and judgment, the installation position and angle of the camera are standardized.

[0178] Camera deployment: At the inspection site, high-definition cameras are installed according to the location and range of the poles to be inspected. The high-definition cameras have sufficient resolution and shooting coverage to ensure that all key pole installation pictures or images can be captured.

[0179] Camera positioning: The camera can be installed in a suitable position to capture the installation of the pole from multiple angles and heights. For example, it can include vertical installation of the camera to obtain the top view of the battery cell, horizontal installation of the camera to obtain the side view of the battery cell, and tilted installation (for example, 45° to the vertical direction) to obtain the side image of the battery cell.

[0180] Image acquisition: The camera collects the images of the pole installation in real time. The camera runs continuously and constantly captures the scene to provide continuous monitoring and surveillance data.

[0181] Image quality control: Ensure that the captured images are clear, without blur or distortion. This can be achieved by adjusting the camera's focal length, aperture, and exposure settings. Ensure that the camera covers an area with no blind spots and can fully capture the installation of the pole.

[0182] Image storage: The camera can store the captured images in real time inside the system or in an external storage device for subsequent analysis and processing by intelligent detection equipment.

[0183] The implementation method of the intelligent detection device can refer to the battery cell detection method in the aforementioned embodiment, and the similarities will not be repeated here.

[0184] In one embodiment, the intelligent detection device 1120 is used to trigger an abnormal alarm and push abnormal information if the placement result of the battery cell is abnormal; the battery cell detection system disclosed in the present invention also includes a quality management system 1130, which is used to receive the abnormal information pushed by the intelligent detection device 1120, generate an abnormal ledger according to the abnormal information, and push abnormal reminder notifications.

[0185] When the intelligent detection device 1120 detects abnormal placement of a battery cell, the intelligent detection device 1120 immediately starts the automatic push mechanism and transmits detailed abnormal information to the quality management system 1130. The abnormal information can record the specific time, specific location, camera name, detection item, actual scene, abnormal battery cell location, etc. of the abnormality, providing rich data support for subsequent analysis and processing.

[0186] After receiving the abnormal information pushed by the intelligent detection equipment, the quality management system 1130 can automatically generate a complete abnormal ledger and intelligently associate information such as factory, monitoring area, battery cell code, and processing measures to ensure that each abnormal event can be recorded and tracked in detail.

[0187] The quality management system 1130 can also push abnormal reminder notifications to relevant responsible personnel to ensure that the relevant responsible personnel can receive abnormal alarms in time and take prompt actions.

[0188] When the abnormal battery cell is properly handled, a manual contact alarm can be executed in the quality management system 1130, and the relevant battery cell code and handling measures can be filled in to ensure the integrity of the operation and the traceability of the abnormal record.

[0189] In the disclosed embodiments, through the above-mentioned series of efficient and automated exception handling, it is not only possible to respond quickly to abnormal situations, but also possible to perform detailed ledger records and in-depth statistical analysis through the quality management system, thereby helping to improve and optimize the production process and enhance the overall production efficiency and quality control level.

[0190] By combining artificial intelligence (AI) technology and quality management systems, remote intelligent monitoring and automatic detection of key components in the industrial production process are realized. On the basis of real-time monitoring, abnormal situations are automatically determined and corresponding treatment measures are taken quickly, such as triggering alarms, generating abnormal ledgers, and pushing abnormal reminder messages. This model not only improves production efficiency and quality control levels, but also significantly reduces labor costs, providing an efficient and accurate remote inspection solution for related industries such as battery cell manufacturing.

[0191] Automated ledger management enables the effective recording and tracking of abnormal situations in the production process, which helps improve production efficiency and quality control. At the same time, manual operation release and recording are realized. When the abnormality is handled, the relevant personnel can manually release the alarm in the quality management system and record the handling measures. The integrity and traceability of the operation records help improve the transparency and manageability of the production process.

[0192] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0193] Refer to the following Fig.12 1200 according to this embodiment of the present disclosure is described. Fig.12 The electronic device 1200 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0194] like Fig.12 As shown, the electronic device 1200 is in the form of a general computing device. The components of the electronic device 1200 may include but are not limited to: at least one processing unit 1210, at least one storage unit 1220, and a bus 1230 connecting different system components (including the storage unit 1220 and the processing unit 1210).

[0195] The storage unit stores a program code, and the program code can be executed by the processing unit 1210, so that the processing unit 1210 executes the steps described in the above "Exemplary Method" section of this specification according to various exemplary embodiments of the present disclosure. For example, the processing unit 1210 can execute the following steps of the above method embodiment: obtaining an image to be processed, the image to be processed includes a support and multiple battery cells in the support; performing image recognition on the image to be processed, determining the identification angle of the support and the contour area image of multiple battery cells in the image to be processed; recognizing the contour area image of multiple battery cells to obtain the surface identification of each battery cell; determining the placement result of each battery cell according to the surface identification of each battery cell and the identification angle of the support.

[0196] The storage unit 1220 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 12201 and / or a cache storage unit 12202 , and may further include a read-only storage unit (ROM) 12203 .

[0197] The storage unit 1220 may also include a program / utility 12204 having a set (at least one) of program modules 12205, such program modules 12205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0198] Bus 1230 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0199] The electronic device 1200 may also communicate with one or more external devices 1240 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 1200, and / or may communicate with any device that enables the electronic device 1200 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 1250. In addition, the electronic device 1200 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 1260. As shown, the network adapter 1260 communicates with other modules of the electronic device 1200 via a bus 1230. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 1200, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0200] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes a number of instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0201] In particular, according to an embodiment of the present disclosure, the process described in the above reference flow chart may be implemented as a computer program product, which includes: a computer program, which implements the above-mentioned battery cell capacity calibration method when executed by a processor.

[0202] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, which may be a readable signal medium or a readable storage medium. In some possible implementations, various aspects of the present disclosure may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary implementations of the present disclosure described in the above “Exemplary Method” section of this specification.

[0203] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0204] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0205] Through the description of the above implementation modes, it is easy for those skilled in the art to understand that the example implementation modes described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation mode of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the implementation mode of the present disclosure.

[0206] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the appended claims.

Claims

1. A battery cell detection method, characterized in that: include: Acquire an image to be processed, wherein the image to be processed includes a support and a plurality of battery cells in the support; Performing image recognition on the image to be processed to determine the identification angle of the support and the contour area images of the plurality of battery cells in the image to be processed; Recognize the contour area images of the plurality of battery cells to obtain the surface identification of each battery cell; Determining the placement result of each battery cell according to the surface identification of each battery cell and the identification angle of the support; Wherein, the step of determining the placement result of each battery cell according to the surface identification of each battery cell and the identification angle of the support includes: If the surface mark of the battery cell and the mark angle of the support meet the first preset condition, it is determined that the placement result of the battery cell is normally placed; taking the mark angle of the support as a reference, if the surface mark of the battery cell is located within the preset mark frame, it is determined that the surface mark of the battery cell and the mark angle of the support meet the first preset condition; If the surface mark of the battery cell and the mark angle of the support do not meet the first preset condition, it is determined that the placement result of the battery cell is abnormal.

2. The battery cell detection method according to claim 1, characterized in that: The method further comprises: The placement mode of the battery cell on the support is determined, and the first preset condition is determined according to the placement mode of the battery cell on the support.

3. The battery cell detection method according to claim 1 or 2, characterized in that: The method further comprises: If the surface mark of the battery cell is not located within the preset mark frame, it is determined that the surface mark of the battery cell and the mark angle of the support do not meet the first preset condition.

4. The battery cell detection method according to claim 1, characterized in that: The plurality of battery cells are arranged in an array on the support; The identification angle of the support is located at one of the top angles of the support, and the identification angle includes a notched angle, an R angle or a chamfered angle; The surface identification includes a two-dimensional code of the battery cell, or an explosion-proof valve and a two-dimensional code, wherein the two-dimensional code is close to one of the poles of the battery cell, and the explosion-proof valve is arranged between two poles of the battery cell.

5. The battery cell detection method according to claim 1, characterized in that: The method further comprises: If the placement result of the battery cell monomer is abnormal, an abnormal alarm is triggered and abnormal information is pushed.

6. The battery cell detection method according to claim 1, characterized in that: The performing image recognition on the image to be processed to determine the identification angle of the support and the contour area images of the plurality of battery cells in the image to be processed includes: The target detector is used to perform image recognition on the battery cells in the image to be processed, the four corner coordinates of the multiple battery cells are determined, and the contour area images of the multiple battery cells are cut out.

7. The battery cell detection method according to claim 1, characterized in that: The performing image recognition on the image to be processed to determine the identification angle of the support and the contour area images of the plurality of battery cells in the image to be processed includes: Using a target detector to perform image recognition on the support in the image to be processed, to obtain a corner point area image of the support, wherein the corner point area image is an image of an area surrounded by multiple corner points obtained by recognizing the support; Performing feature extraction on the corner point area image of the support to obtain feature values ​​of the four corners of the corner point area; The identification angle of the support is determined according to the characteristic values ​​of the four corners of the corner point area.

8. The battery cell detection method according to claim 7, characterized in that: The method further comprises: Determine a closed area of ​​the corner point area image; If the area of ​​the closed region is greater than a preset area threshold, the corner point region image of the support is re-determined so that the area of ​​the closed region corresponding to the re-determined corner point region image of the support is less than or equal to the preset area threshold.

9. The battery cell detection method according to claim 1, characterized in that: The identifying the contour area images of the plurality of battery cells to obtain the surface identification of each battery cell includes: The contour area images of the plurality of battery cells are identified by an image processing model to determine the surface identification of each battery cell. The image processing model is trained based on the contour area images of the plurality of battery cells and the corresponding surface identification labels.

10. The battery cell detection method according to claim 1, characterized in that: After acquiring the image to be processed, the method further includes: The image to be processed is preprocessed, and the preprocessing includes at least one of applying a mask to filter background information and performing a perspective transformation to correct geometric deformation of the image to be processed.

11. A battery cell detection device, characterized in that: include: An image acquisition module, used for acquiring an image to be processed, wherein the image to be processed includes a support and a plurality of battery cells in the support; A first recognition module is used to perform image recognition on the image to be processed, and determine the identification angle of the support and the contour area image of the plurality of battery cells in the image to be processed; A second recognition module is used to recognize the contour area images of the plurality of battery cells to obtain the surface identification of each battery cell; A result determination module is used to determine the placement result of each battery cell according to the surface mark of each battery cell and the mark angle of the support; and is also used to determine that the placement result of the battery cell is normally placed if the surface mark of the battery cell and the mark angle of the support meet a first preset condition, and to determine that the surface mark of the battery cell and the mark angle of the support meet the first preset condition if the surface mark of the battery cell is located within a preset mark frame, taking the mark angle of the support as a reference; If the surface mark of the battery cell and the mark angle of the support do not meet the first preset condition, it is determined that the placement result of the battery cell is abnormal.

12. A battery cell detection system, characterized in that: It includes image acquisition equipment and intelligent detection equipment, including: The image acquisition device is used to acquire an image to be processed, wherein the image to be processed includes a support and a plurality of battery cells in the support; The intelligent detection device is used to obtain an image to be processed; perform image recognition on the image to be processed to determine the identification angle of the support and the contour area image of the multiple battery cells in the image to be processed; recognize the contour area image of the multiple battery cells to obtain the surface identification of each battery cell; determine the placement result of each battery cell according to the surface identification of each battery cell and the identification angle of the support; if the surface identification of the battery cell and the identification angle of the support meet a first preset condition, the placement result of the battery cell is determined to be normal, and with the identification angle of the support as a reference, if the surface identification of the battery cell is located within a preset identification frame, it is determined that the surface identification of the battery cell and the identification angle of the support meet the first preset condition; if the surface identification of the battery cell and the identification angle of the support do not meet the first preset condition, the placement result of the battery cell is determined to be abnormal.

13. The battery cell detection system according to claim 12, characterized in that: The intelligent detection device is used to trigger an abnormal alarm and push abnormal information if the placement result of the battery cell monomer is abnormal; The system also includes a quality management system for receiving the abnormal information pushed by the intelligent detection device, generating an abnormality ledger according to the abnormal information, and pushing abnormality reminder notifications.

14. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; Wherein, the processor is configured to execute the battery cell detection method according to any one of claims 1 to 10 by executing the executable instructions.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the battery cell detection method according to any one of claims 1 to 10 is implemented.

Images