Internal anomaly detection method and device of mesh battery and storage medium

Abstract

The invention relates to the technical field of battery detection, and provides an internal anomaly detection method and device of a mesh battery and a storage medium. The internal anomaly detection method for the mesh battery comprises the steps of obtaining, recognition, calculation and judgment, the ROI image of the battery cell is processed, image details are enhanced, the image is clear, features are obvious, image recognition is facilitated, the risk of misjudgment and missed judgment in the recognition process is reduced, then the calculation process is optimized, and the detection accuracy is improved. The coordinate difference value and the alignment difference value are obtained, whether defects exist in the battery or not and the quality of the battery are judged, and the misjudgment probability is greatly reduced; the whole process is simple and efficient, no complex algorithm needs to be added, and normal production and use of the mesh battery can be met.

Description

Technical Field

[0001] The present application relates to the field of battery detection technology, and in particular to a method, device and storage medium for detecting internal abnormalities of a mesh battery. Background Art

[0002] During the battery manufacturing process, the detection and identification of internal battery defects is critical to ensuring battery quality, improving battery performance and extending battery life.

[0003] Among them, the conventional wound laminated battery detection method is to detect the coating amount of the positive and negative electrodes inside the battery cell to determine whether there is a short circuit risk between the positive and negative electrodes. The positive and negative electrode materials of the mesh battery are arranged in a mesh-like manner, and the structures overlap with each other, making it difficult to observe their coating amount, and the conventional wound laminated battery detection method cannot be applied; and the traditional method of observing the battery surface cannot effectively detect the internal structure of the mesh battery, and cannot guarantee the alignment of the mesh battery during the production process; at the same time, due to the mutual influence and overlapping structures of the internal areas of the mesh battery cell, when using X-ray technology for detection, the image will be unclear and the features will not be obvious, resulting in a high possibility of misjudgment and missed judgment in algorithm recognition, which cannot meet the normal production and use of mesh batteries.

[0004] Therefore, the current battery detection method cannot simply and effectively detect the inside of the mesh battery to determine whether there is any abnormality inside. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the purpose of the present application is to provide a method for detecting internal abnormalities of a mesh battery, aiming to solve the problem in the prior art that it is impossible to simply and effectively detect the inside of the mesh battery and determine whether there is an abnormality therein.

[0006] The technical solution adopted by the present application to solve the technical problem is as follows: The present application provides a method for detecting internal abnormalities of a mesh battery, comprising:

[0007] Get the ROI image of the battery cell;

[0008] The ROI image of the battery cell is segmented to obtain a lower battery cell region, and the lower battery cell region is preprocessed and segmented in sequence to obtain an upper edge shell wall region; the lower battery cell region is subtracted from the upper edge shell wall region, and then enhanced and identified to obtain the top coordinates of the pole piece of each battery cell line;

[0009] Extract the highest point in the coordinates of the pole piece top of each cell line as the first highest point, correspond the coordinates of the first highest point to the upper edge straight line of the upper edge shell wall, and obtain the coordinate difference; extract the highest point and the lowest point in the coordinates of the pole piece top of each cell line, and obtain the alignment difference according to the coordinates of the highest point and the coordinates of the lowest point;

[0010] Whether the battery is defective is determined based on the coordinate difference, and the quality of the battery is determined based on the alignment difference, and the lower cell area is preprocessed and segmented in sequence.

[0011] Preferably, the ROI image segmentation of the battery cell to obtain the lower battery cell region includes:

[0012] The ROI image of the battery cell is corroded and expanded to remove the lateral portion of the battery cell groove in the ROI image of the battery cell, and the ROI image of the battery cell is segmented to obtain a lower battery cell region.

[0013] Preferably, the lower portion of the battery cell region is pre-processed and segmented in sequence to obtain an upper edge shell wall region, including:

[0014] A filtering algorithm is used to extract the grayscale value of the lower portion of the battery cell area in the X-axis direction, and the grayscale value in the X-axis direction is stretched and filled to obtain an upper edge straight line of the upper edge shell wall;

[0015] Automatically determine a first threshold value based on the distribution of grayscale values ​​in the lower portion of the battery cell area after stretching and filling; compare the grayscale value in the lower portion of the battery cell area after stretching and filling with the first threshold value;

[0016] The portion whose gray value is greater than or equal to the first threshold is the upper edge shell wall area;

[0017] The portion whose grayscale value is smaller than the first threshold is segmented as background.

[0018] Preferably, the step of subtracting the lower cell area from the upper edge shell wall area, and then enhancing and identifying to obtain the top coordinates of the pole piece of each cell line includes:

[0019] The lower portion of the battery cell area is subjected to corrosion operation to obtain a battery cell area that does not include the inner wall; the contrast of the battery cell lines is adjusted to enhance the battery cell lines, each battery cell line in the battery cell area that does not include the inner wall is identified, and the top coordinates of the pole piece corresponding to each battery cell line are obtained by secondary identification.

[0020] Preferably, during the identification and secondary identification process, a removal of the ear interference operation is required;

[0021] The tab interference removal operation is to extract the tab area and separate the tab area from the battery cell lines.

[0022] Preferably, the step of making the coordinates of the first highest point correspond to the upper edge straight line of the upper edge shell wall to obtain the coordinate difference includes:

[0023] Using the column coordinates of the first highest point, obtaining the first row coordinates corresponding to the same column coordinates on the upper edge straight line;

[0024] Subtract the row coordinate of the first highest point from the first row coordinate to obtain a coordinate difference.

[0025] Preferably, calculating the alignment difference according to the coordinates of the highest point and the coordinates of the lowest point includes:

[0026] Taking the coordinates of the highest point and the coordinates of the lowest point as a set of data, performing straight line fitting on the set of data, and obtaining a minimum circumscribed rectangle based on the fitted straight line and the set of data;

[0027] The length of the short side of the minimum circumscribed rectangle in the vertical direction is the alignment difference.

[0028] Preferably, judging whether the battery is defective according to the coordinate difference, and then judging the quality of the battery according to the alignment difference, comprises:

[0029] Setting a first difference interval and a second difference interval;

[0030] If the coordinate difference is not within the first difference interval, it is determined that the battery is defective and the process ends;

[0031] If the coordinate difference is within the first difference range, the next process is entered; if the alignment difference is not within the second difference range, the battery is determined to be defective and the process ends;

[0032] If the coordinate difference is within the first difference range, the process proceeds to the next process; if the alignment difference is within the second difference range, the process proceeds to the next process to continue to judge the quality of the battery.

[0033] The present application also provides an internal abnormality detection device for a mesh battery, comprising:

[0034] A memory for storing a program for constructing the internal abnormality detection method of the mesh battery;

[0035] The processor is used to implement the internal abnormality detection method of the mesh battery by executing the program stored in the memory.

[0036] At the same time, the present application also provides a computer-readable storage medium, including a program, which can be executed by a processor to implement the above-mentioned internal abnormality detection method of the mesh battery.

[0037] Compared with the prior art, the present application provides a method for detecting internal abnormalities in a mesh battery. By processing the ROI image of the battery cell, the image details are enhanced, making the image clear and the features obvious, facilitating image recognition and reducing the risk of misjudgment and missed judgment during the recognition process. The calculation process is then optimized to obtain coordinate differences and alignment differences, thereby judging whether there are defects inside the battery and the quality of the battery. The entire process is simple and efficient, does not require the addition of complex algorithms, and can meet the normal production and use of mesh batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flow chart of the internal abnormality detection method of the mesh battery provided in the present application;

[0039] Figure 2 is a schematic diagram of the acquisition steps of the internal abnormality detection method of the mesh battery provided in the present application;

[0040] Figure 3 It is a schematic diagram of coordinate difference and alignment difference obtained by the internal abnormality detection method of the mesh battery provided in the present application;

[0041] Figure 4 It is a module schematic diagram of the internal abnormality detection device of the mesh battery provided in this application. DETAILED DESCRIPTION

[0042] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as limiting the present application.

[0043] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first" and "second" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present application, unless otherwise specified, "multiple" means two or more.

[0044] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0045] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application. The drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0046] Additionally, in the following description, specific details are provided to provide a thorough understanding of the embodiments. However, one skilled in the art will appreciate that the aspects described may be practiced without these specific details.

[0047] Combined with reference Figure 1 and Figure 2 In a first embodiment of the present application, a method for detecting an internal abnormality of a mesh battery is provided, comprising:

[0048] Get the ROI image of the battery cell;

[0049] The ROI image of the battery cell is segmented to obtain a lower battery cell region, and the lower battery cell region is preprocessed and segmented in sequence to obtain an upper edge shell wall region; the lower battery cell region is subtracted from the upper edge shell wall region, and then enhanced and identified to obtain the top coordinates of the pole piece of each battery cell line;

[0050] Extract the highest point in the coordinates of the pole piece top of each cell line as the first highest point, correspond the coordinates of the first highest point to the upper edge straight line of the upper edge shell wall, and obtain the coordinate difference; extract the highest point and the lowest point in the coordinates of the pole piece top of each cell line, and obtain the alignment difference according to the coordinates of the highest point and the coordinates of the lowest point;

[0051] The defectiveness of the battery is determined based on the coordinate difference, and the quality of the battery is determined based on the alignment difference.

[0052] That is, it includes the steps of acquisition, identification, calculation and judgment, as follows:

[0053] S100 acquisition: acquiring the ROI image of the battery cell;

[0054] S200 identification: segmenting the ROI image of the battery cell to obtain a lower battery cell region, preprocessing and segmenting the lower battery cell region in sequence to obtain an upper edge shell wall region; subtracting the lower battery cell region from the upper edge shell wall region, and then enhancing and identifying to obtain the top coordinates of the pole piece of each battery cell line;

[0055] S300 calculation: extract the highest point in the coordinates of the pole piece top of each cell line as the first highest point, correspond the coordinates of the first highest point to the upper edge straight line of the upper edge shell wall, and obtain the coordinate difference; extract the highest point and the lowest point in the coordinates of the pole piece top of each cell line, and obtain the alignment difference according to the coordinates of the highest point and the coordinates of the lowest point;

[0056] S400 judgment: judging whether the battery is defective according to the coordinate difference, and then judging the quality of the battery according to the alignment difference.

[0057] The internal abnormality detection method of the mesh battery of the present embodiment processes the ROI image of the battery cell in view of the characteristics of the mesh battery that the internal areas of the battery cells influence each other and the structures overlap each other, effectively enhances the image details, makes the image clear and the features obvious, facilitates image recognition, and can accurately identify the top of the electrode of the battery cell line, thereby facilitating subsequent calculation and judgment operations; and through the coordinate difference and alignment difference, a secondary judgment is made on the defects inside the battery, which greatly reduces the probability of misjudgment.

[0058] At the same time, mesh batteries, especially mesh lithium batteries, often use excessive negative electrode design to prevent lithium dendrite precipitation from causing short circuits and thermal runaway in the battery; the coordinate difference of mesh lithium batteries is more obvious when intersecting with batteries of other structures. The use of coordinate difference can quickly and accurately determine whether the battery is defective, thereby making an initial judgment before judging the battery quality, reducing the chance of misjudgment. Therefore, the internal abnormality detection method of the mesh battery of this embodiment is a detection method for the inside of the battery cell of the mesh battery, which is simple and effective to detect the inside of the mesh battery and determine whether there is an abnormality inside it; and the internal abnormality detection method of the mesh battery of this embodiment has better detection accuracy and lower misjudgment probability for mesh lithium batteries, especially for mesh cylindrical batteries or mesh cylindrical lithium batteries.

[0059] In some embodiments, in the operation of S100: acquiring the ROI image of the battery cell includes:

[0060] An X-ray image of the battery is acquired, and a binarization operation is performed on the X-ray image to obtain a ROI image of the battery cell.

[0061] It should be noted that the binarization operation is to grayscale the X-ray image of the battery, set a threshold according to the gray value difference between the battery cell and the background, extract the low gray value area, form a connected domain, and obtain the main area of ​​the battery cell, that is, the ROI image of the battery cell.

[0062] In some embodiments, in the operation of S200 identification, segmenting the ROI image of the battery cell to obtain a lower battery cell region includes:

[0063] The ROI image of the battery cell is corroded and expanded to remove the lateral portion of the battery cell groove in the ROI image of the battery cell, and the ROI image of the battery cell is segmented to obtain a lower battery cell region.

[0064] In this embodiment, the ROI image of the battery cell is first corroded and then expanded, so that the battery cell groove and its lateral part can be extracted as initial features, and then the lateral part of the battery cell groove and some small protrusions and other noise points are removed, and the contour of the object is smoothed at the same time, but the contour size of the object is not significantly changed; and it can also prevent the lateral part of the battery cell groove from blocking the top of the electrode piece, so as to prevent affecting subsequent recognition operations.

[0065] The ROI image of the battery cell is segmented to obtain the lower battery cell area, which can be directly segmented by image segmentation method; the threshold segmentation method is mainly used here, specifically, the threshold can be set according to the morphological difference or gray value difference between the upper cap area and the lower battery cell area in the ROI image of the battery cell for segmentation, thereby obtaining the lower battery cell area.

[0066] In some embodiments, in the operation of S200 identification, the preprocessing and segmenting of the lower portion of the battery cell region in sequence to obtain the upper edge shell wall region includes:

[0067] A filtering algorithm is used to extract the grayscale value of the lower portion of the battery cell area in the X-axis direction, and the grayscale value in the X-axis direction is stretched and filled to obtain an upper edge straight line of the upper edge shell wall;

[0068] Automatically determine a first threshold value based on the distribution of grayscale values ​​in the lower portion of the battery cell area after stretching and filling; compare the grayscale value in the lower portion of the battery cell area after stretching and filling with the first threshold value;

[0069] The portion whose gray value is greater than or equal to the first threshold is the upper edge shell wall area;

[0070] The portion whose grayscale value is smaller than the first threshold is segmented as background.

[0071] It should be noted that the preprocessing operation is to use a filtering algorithm to extract the grayscale value of the lower part of the battery cell area in the X-axis direction, and stretch and fill the grayscale value in the X-axis direction, so that the battery cell shell wall (including the upper edge shell wall and the lower edge shell wall) can be approximated to a straight line, which is convenient for the subsequent removal of the shell wall, making the battery cell structure simpler and clearer; at the same time, it can weaken the interference of the pole ear, eliminate the interlaced influence between the pole ear and the shell wall, and prevent the subsequent identification operation from being affected. Among them, the upper edge straight line of the upper edge shell wall and the lower edge straight line of the lower edge shell wall can be obtained according to the battery cell shell wall that is approximated to a straight line; the coordinate information of the upper edge straight line of the upper edge shell wall can be used for subsequent calculation steps.

[0072] Based on the above, the ROI image of the battery cell has been eroded and expanded, making the image contrast more obvious and less noisy. At the same time, the contrast is further improved and the noise is reduced through the filtering algorithm, which is convenient for subsequent segmentation operations; the segmentation operation, "based on the distribution of grayscale values ​​in the lower part of the battery cell area after stretching and filling, automatically determine the first threshold value; compare the grayscale value in the lower part of the battery cell area after stretching and filling with the first threshold value" can adopt Otsu's method, adaptive threshold method, and entropy-based method; Otsu's method is preferred, which automatically selects the optimal threshold by maximizing the inter-class variance, so that the process of segmenting the upper edge shell wall area is simple and efficient.

[0073] In some embodiments, in the operation of S200 identification, the subtraction of the lower portion of the battery cell area and the upper edge shell wall area is performed, and then the top coordinates of the pole piece of each battery cell line are obtained by enhancement and identification, including:

[0074] The lower portion of the battery cell area is subjected to corrosion operation to obtain a battery cell area that does not include the inner wall; the contrast of the battery cell lines is adjusted to enhance the battery cell lines, each battery cell line in the battery cell area that does not include the inner wall is identified, and the top coordinates of the pole piece corresponding to each battery cell line are obtained by secondary identification.

[0075] The purpose of subtracting the lower cell region from the upper edge shell wall region is to perform an etching operation on the lower cell region to obtain a cell region that does not include the inner wall, thereby making the image simpler and clearer and preventing the shell wall from affecting subsequent operations.

[0076] To adjust the contrast and enhance the cell lines, a local contrast enhancement method, a nonlinear mapping method and a linear contrast stretching method may be specifically used. Among them, the local contrast enhancement method is preferred. The method can adjust the position of the cell lines in the image based on the image that has undergone pre-processing operations such as stretching, filling, corrosion and expansion, thereby enhancing the contrast between the lines and the background without affecting other parts of the entire image, and can improve the accuracy of subsequent recognition operations.

[0077] Specifically, adjusting the contrast to enhance the cell lines and identifying each cell line in the cell area not including the inner wall can be performed simultaneously, that is, each cell line can be identified while enhancing the contrast between the line and the background, thereby improving processing efficiency.

[0078] Then, secondary identification is performed through the identification information of each battery cell line to obtain the top coordinates of the pole piece, making the obtained top coordinate information of the pole piece more accurate.

[0079] In some embodiments, during the identification and secondary identification process, a de-electrode interference operation is required;

[0080] The tab interference removal operation is to extract the tab area and separate the tab area from the battery cell lines.

[0081] The tab interference removal operation can remove the tabs that are in the same direction and in the same line as the cell pole piece, thereby preventing the top of the pole piece from being identified at the tab position and affecting the identification of the top coordinates of the pole piece.

[0082] The lug area can be extracted by first corroding and then dilating, and then the lug part can be removed as the background by automatically determining the threshold. Specifically, the threshold can be automatically determined by Otsu's method, adaptive threshold method, or entropy-based method, preferably Otsu's method.

[0083] In some embodiments, in the calculation operation of S300, the coordinates of the first highest point are matched with the upper edge straight line of the upper edge shell wall to obtain the coordinate difference, including:

[0084] Using the column coordinates of the first highest point, obtaining the first row coordinates corresponding to the same column coordinates on the upper edge straight line;

[0085] Subtract the row coordinate of the first highest point from the first row coordinate to obtain a coordinate difference.

[0086] That is, the coordinate difference is the difference between the row coordinate of the first highest point and the first row coordinate corresponding to the same column coordinate of the first highest point on the upper edge straight line. The first row coordinate is obtained based on the information of the first highest point and the upper edge straight line. It is necessary to construct the same coordinate system of the first highest point and the upper edge straight line, and connect the operation of S200 recognition and the operation of S300 calculation to obtain the coordinate difference later.

[0087] In some implementations, in the operation of calculating in S300, calculating the alignment difference according to the coordinates of the highest point and the coordinates of the lowest point includes:

[0088] Taking the coordinates of the highest point and the coordinates of the lowest point as a set of data, performing straight line fitting on the set of data, and obtaining a minimum circumscribed rectangle based on the fitted straight line and the set of data;

[0089] The length of the short side of the minimum circumscribed rectangle in the vertical direction is the alignment difference.

[0090] The coordinates of the highest point and the coordinates of the lowest point refer to the highest point and the lowest point of the point coordinates of each battery cell line. If there are n battery cell lines, the coordinates of the top of the pole piece of the battery cell line are the highest point and the lowest point among the n coordinates.

[0091] The conventional algorithm directly calculates the height difference based on the highest and lowest points of each battery cell line. When the battery cell line is continuously tilted, the height difference between the highest and lowest points of the battery cell line may be large, but it does not mean that there is a large height difference between each layer of pole pieces. It may just be that the tilt direction tends to be a straight line, resulting in no obvious misalignment or unevenness overall, which in turn makes the battery defect judgment incorrect and the abnormality cannot be identified.

[0092] This embodiment performs straight line fitting based on a set of data including the coordinates of the highest point and the coordinates of the lowest point, and can obtain a fitted straight line with a lower deviation value, thereby reducing the error. The fitted straight line is combined with the coordinates of the set of data to obtain the minimum circumscribed rectangle, which can reflect the actual distribution of the top of the pole piece in the inclined direction, so that the actual condition inside the battery can be judged, making the judgment more accurate and reducing the probability of misjudgment. The length of the short side of the minimum circumscribed rectangle in the vertical direction is the alignment difference. The smaller the value, the better the neatness of the pole piece arrangement.

[0093] In some embodiments, in the operation of S400, judging whether the battery is defective according to the coordinate difference, and judging the quality of the battery according to the alignment difference, includes:

[0094] Setting a first difference interval and a second difference interval;

[0095] If the coordinate difference is not within the first difference interval, it is determined that the battery is defective and the process ends;

[0096] If the coordinate difference is within the first difference range, the next process is entered; if the alignment difference is not within the second difference range, the battery is determined to be defective and the process ends;

[0097] If the coordinate difference is within the first difference range, the process proceeds to the next process; if the alignment difference is within the second difference range, the process proceeds to the next process to continue to judge the quality of the battery.

[0098] Specifically, the first difference interval and the second difference interval can be set according to actual conditions. At the same time, the third difference interval and the fourth difference interval can be further set within the second difference interval. For example, the second difference interval is 1-10n, the third difference interval is 1-3n, and the fourth difference interval is 4-6n. If the alignment difference enters the third difference interval, it means that the battery has excellent quality. If the alignment difference enters the fourth difference interval, it means that the battery has good quality. If the alignment difference enters the second difference interval but cannot enter the third difference interval and the fourth difference interval, it means that the battery has qualified quality. Specifically, when the obtained alignment difference is 8n, it is determined that the battery has qualified quality.

[0099] The mesh battery is tested according to the method of the first embodiment of the present application, and the obtained schematic diagram is as follows: Figure 3 As shown, the alignment degree (ie, the alignment difference) of the battery is 7.99159, and the spacing (ie, the coordinate difference) is 11.7179.

[0100] Combined with reference Figure 4 In a second embodiment of the present application, a device 100 for detecting an internal abnormality of a mesh battery is provided, comprising:

[0101] A memory 102, used to store a program for constructing the internal abnormality detection method of the mesh battery according to any one of the above embodiments;

[0102] The processor 101 is configured to implement the internal abnormality detection method of the mesh battery of any one of the above embodiments by executing the program stored in the memory 102 .

[0103] It should be noted that the specific execution process and technical effects of the internal abnormality detection device of the mesh battery of the present application are basically the same as the specific execution process and technical effects of the internal abnormality detection method of the mesh battery of the present application, so they will not be repeated here.

[0104] Meanwhile, a third embodiment of the present application provides a computer-readable storage medium, including a program, which can be executed by a processor to implement the internal abnormality detection method of the mesh battery of any one of the above embodiments.

[0105] In summary, the present application provides a method for detecting internal abnormalities in a mesh battery. By processing the ROI image of the battery cell, the image details are enhanced, making the image clear and the features obvious, facilitating image recognition, and reducing the risk of misjudgment and missed judgment during the recognition process. The calculation process is then optimized to obtain coordinate differences and alignment differences, thereby judging whether there are defects inside the battery and the quality of the battery, greatly reducing the chance of misjudgment. The entire process is simple and efficient, does not require the addition of complex algorithms, and can meet the normal production and use of mesh batteries.

[0106] Finally, it should be noted that the above examples are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above examples, a person of ordinary skill in the art should understand that he or she can still modify the technical solutions described in the above examples, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the examples of the present application.

Claims

1. A method for detecting internal abnormality of a mesh battery, characterized in that: include: Get the ROI image of the battery cell; The ROI image of the battery cell is segmented to obtain a lower battery cell region, and the lower battery cell region is preprocessed and segmented in sequence to obtain an upper edge shell wall region; the lower battery cell region is subtracted from the upper edge shell wall region, and then enhanced and identified to obtain the top coordinates of the pole piece of each battery cell line; Extract the highest point in the coordinates of the pole piece top of each cell line as the first highest point, correspond the coordinates of the first highest point to the upper edge straight line of the upper edge shell wall, and obtain the coordinate difference; extract the highest point and the lowest point in the coordinates of the pole piece top of each cell line, and obtain the alignment difference according to the coordinates of the highest point and the coordinates of the lowest point; The defectiveness of the battery is determined based on the coordinate difference, and the quality of the battery is determined based on the alignment difference.

2. The internal abnormality detection method of the mesh battery according to claim 1, characterized in that: The ROI image of the battery cell is segmented to obtain a lower battery cell area, including: The ROI image of the battery cell is corroded and expanded to remove the lateral portion of the battery cell groove in the ROI image of the battery cell, and the ROI image of the battery cell is segmented to obtain a lower battery cell region.

3. The internal abnormality detection method of the mesh battery according to claim 1, characterized in that: The pre-processing and segmenting of the lower portion of the battery cell region in sequence to obtain the upper edge shell wall region includes: A filtering algorithm is used to extract the grayscale value of the lower portion of the battery cell area in the X-axis direction, and the grayscale value in the X-axis direction is stretched and filled to obtain an upper edge straight line of the upper edge shell wall; Automatically determine a first threshold value based on the distribution of grayscale values ​​in the lower portion of the battery cell area after stretching and filling; compare the grayscale value in the lower portion of the battery cell area after stretching and filling with the first threshold value; The portion whose gray value is greater than or equal to the first threshold is the upper edge shell wall area; The portion whose grayscale value is smaller than the first threshold is segmented as background.

4. The internal abnormality detection method of a mesh battery according to claim 1, characterized in that: After subtracting the lower cell area from the upper edge shell wall area, the top coordinates of the pole piece of each cell line are obtained through enhancement and identification, including: The lower portion of the battery cell area is subjected to corrosion operation to obtain a battery cell area that does not include the inner wall; the contrast of the battery cell lines is adjusted to enhance the battery cell lines, each battery cell line in the battery cell area that does not include the inner wall is identified, and the top coordinates of the pole piece corresponding to each battery cell line are obtained by secondary identification.

5. The internal abnormality detection method of a mesh battery according to claim 1, characterized in that: During the identification and secondary identification process, it is necessary to perform a pole ear interference removal operation; The tab interference removal operation is to extract the tab area and separate the tab area from the battery cell lines.

6. The internal abnormality detection method of a mesh battery according to claim 1, characterized in that: The step of making the coordinate of the first highest point correspond to the upper edge straight line of the upper edge shell wall to obtain the coordinate difference includes: Using the column coordinates of the first highest point, obtaining the first row coordinates corresponding to the same column coordinates on the upper edge straight line; Subtract the row coordinate of the first highest point from the first row coordinate to obtain a coordinate difference.

7. The internal abnormality detection method of a mesh battery according to claim 1, characterized in that: The calculating the alignment difference according to the coordinates of the highest point and the coordinates of the lowest point includes: Taking the coordinates of the highest point and the coordinates of the lowest point as a set of data, performing straight line fitting on the set of data, and obtaining a minimum circumscribed rectangle based on the fitted straight line and the set of data; The length of the short side of the minimum circumscribed rectangle in the vertical direction is the alignment difference.

8. The method for detecting internal abnormality of a mesh battery according to claim 1, characterized in that: The method of judging whether the battery is defective according to the coordinate difference and judging the quality of the battery according to the alignment difference includes: Setting a first difference interval and a second difference interval; If the coordinate difference is not within the first difference interval, it is determined that the battery is defective and the process ends; If the coordinate difference is within the first difference range, the next process is entered; if the alignment difference is not within the second difference range, the battery is determined to be defective and the process ends; If the coordinate difference is within the first difference range, the process proceeds to the next process; if the alignment difference is within the second difference range, the process proceeds to the next process to continue to judge the quality of the battery.

9. An internal abnormality detection device for a mesh battery, characterized in that: include: A memory for storing a program for constructing the internal abnormality detection method of the mesh battery according to any one of claims 1 to 8; A processor, configured to implement the internal abnormality detection method of the mesh battery according to any one of claims 1 to 8 by executing the program stored in the memory.

10. A computer-readable storage medium, characterized in that: The invention comprises a program which can be executed by a processor to implement the internal abnormality detection method of the mesh battery according to any one of claims 1 to 8.

Images