Image-based pole piece notch defect early warning method

Abstract

The invention discloses an image-based pole piece notch defect early warning method, and particularly relates to the technical field of pole piece notches, which comprises the following steps: starting a multi-thread defect detection thread, detecting a burr scrap area, a foil fracture area and a carbon layer cracking area at a pole piece notch, performing corresponding region-of-interest extraction in each thread, and determining the defect of the pole piece notch according to the extracted region-of-interest extraction. The method comprises the following steps of: reducing an original image, quickly positioning an area where a pole piece is located by utilizing the characteristics of the area of the pole piece, locking an area of interest of each thread in the area where the pole piece is located, and collecting defect information about burr and chipping characteristics, defect information about carbon layer cracking characteristics and defect information about foil fracture characteristics in the area of interest of each thread; according to the method, the collected defect information is comprehensively analyzed, potential hazards existing in a pole piece notch are quantified, early warning is carried out in time, the defects of the pole piece after slitting can be rapidly detected, a detection result is output, an alarm is given in time, production loss is reduced, and meanwhile the product quality is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of pole piece incisions, and more specifically, to an image-based early warning method for pole piece incision defects. Background Art

[0002] With the development of automation technology, automated inspection based on machine vision has become an important part of battery manufacturing. Through high-precision image acquisition and multi-threaded parallel processing technology, the incision area of ​​the pole piece can be fully inspected, and early warning can be issued through defect analysis models to prevent substandard product quality. The main difficulty of the existing technology lies in the unclear or fuzzy imaging stage, and the speed and accuracy of the detection stage are not easy to meet at the same time. First, the imaging is fuzzy. In order to make the image bright enough, the exposure time can only be increased under the limitation of hardware level, so the image has motion blur, which leads to the reduction of detection accuracy of the movement direction of the pole piece and the failure to meet the detection accuracy requirements. Secondly, the speed of traditional detection algorithms is relatively low. Therefore, the characteristics of defects such as burrs, debris, carbon layer cracking, and foil breakage are different. The recognition algorithms and feature extraction methods of different defects may vary greatly, resulting in reduced defect detection accuracy in complex incision areas, and it is not easy to provide timely warning of hidden dangers in the pole piece incision.

[0003] In order to solve the above defects, a technical solution is now provided. Summary of the invention

[0004] In order to overcome the above defects of the prior art, an embodiment of the present invention provides an early warning method for pole piece cut defects based on an image to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: An image-based early warning method for pole piece cut defects comprises the following steps: S1: Start the image acquisition thread, start the multi-thread defect detection thread, detect the burr and debris area at the pole piece cut, the foil fracture area, and the carbon layer crack area; S2: extract the corresponding region of interest in each thread, quickly locate the region where the pole piece is located by reducing the original image and using the features of the pole piece region, and lock the region of interest of each thread in the region where the pole piece is located; S3: collecting defect information about burrs and debris features, defect information about carbon layer cracking features, and defect information about foil fracture features in the region of interest of each thread; S4: Conduct a comprehensive analysis of the collected defect information, build a defect warning model, quantify the potential hazards of the pole piece cuts, and issue a warning in a timely manner.

[0006] In a preferred embodiment, the defect information about burrs and debris features includes: Defect information about burrs and debris features is represented by pixel value abnormality coefficients; The logic for obtaining the pixel abnormality coefficient is as follows: for an image containing burrs and debris defects, the pixel value of each pixel point in the region of interest in the thread is obtained by converting the original image into a grayscale image, and the pixel value of each pixel point in the region of interest in the thread is marked as: , where n=1, 2, 3, ..., N, N is a positive integer, and n is the number of the pixel in the grayscale image; Calculate the mean and standard deviation of the pixel values ​​in the grayscale image, and mark the mean and standard deviation of the pixel values ​​in the grayscale image as: and ,in, , ; Set the lowest pixel value threshold and the highest pixel value threshold, and mark the lowest pixel value threshold and the highest pixel value threshold as: and , obtain the number of pixel points below the minimum threshold of pixel value, and mark the number of pixel points below the minimum threshold of pixel value as: a, obtain the number of pixel points above the minimum threshold of pixel value, and mark the number of pixel points above the minimum threshold of pixel value as: b; Calculate the pixel anomaly coefficient, the calculation formula is: ;in, is the pixel abnormality coefficient.

[0007] In a preferred embodiment, the defect information about the cracking characteristics of the carbon layer includes: The defect information about the cracking characteristics of the carbon layer is represented by a characteristic deviation coefficient; The acquisition logic of the characteristic deviation coefficient is: obtain an image containing a crack area, obtain the characteristics of each crack in the image, and mark the characteristics of each crack in the image as: , ,in, represents the width of each i-th crack in the image, represents the length of the i-th crack in the image, i=1, 2, 3, ..., I, I is a positive integer, and i is the number of the crack in the image; Set the crack width threshold and crack length threshold, and mark the crack width threshold and crack length threshold as: and ; Calculate the characteristic deviation coefficient using the formula: ;in, is the characteristic deviation coefficient.

[0008] In a preferred embodiment, the defect information about the fracture characteristics of the foil includes: The defect information about the fracture characteristics of the foil is expressed by the damage dispersion coefficient; The logic for obtaining the damage dispersion coefficient is: obtain the coordinate point of the broken part according to the coordinates of two adjacent foil regions, and mark the coordinate point of the broken part as: ,in, is the position of the pth pixel point on the x-axis at the fractured part, is the position of the p-th pixel point on the y-axis at the position of the broken part, p=1, 2, 3, ..., P, P is a positive integer, and p is the number of the pixel point at the position of the broken part; The average value and standard deviation of the pixel points at the position of the broken part on the x-axis are calculated, and the average value and standard deviation of the pixel points at the position of the broken part on the x-axis are marked as: and , the mean value and standard deviation of the pixel points at the broken part on the y-axis, and the mean value and standard deviation of the pixel points at the broken part on the x-axis are marked as: and ; in, , , , ; The coefficient of variation of the pixel point at the position of the fractured part on the x-axis and the coefficient of variation of the pixel point at the position of the fractured part on the y-axis are obtained, and the coefficient of variation of the pixel point at the position of the fractured part on the x-axis and the coefficient of variation of the pixel point at the position of the fractured part on the y-axis are marked as: and ; Calculate the damage dispersion coefficient using the following formula: ;in, is the damage dispersion coefficient.

[0009] In a preferred embodiment, building a defect warning model includes: The defect information about burrs and debris features, the defect information about carbon layer cracking features, and the defect information about foil fracture features are comprehensively analyzed. By weighted calculation of pixel abnormality coefficient, feature deviation coefficient, and damaged dispersion coefficient, a defect warning evaluation model is constructed to generate a defect warning evaluation coefficient, which is expressed as follows: ;in, , , are the proportional coefficients of pixel anomaly coefficient, feature deviation coefficient, and damaged dispersion coefficient, respectively. , , Both are greater than 0.

[0010] In a preferred embodiment, the potential hidden dangers of the pole piece cut are quantified and early warning is given in time, including: Set a defect warning assessment coefficient threshold, compare the defect warning assessment coefficient with the defect warning assessment coefficient threshold, if the defect warning assessment coefficient is greater than the defect warning assessment coefficient threshold, generate a warning signal, indicating that the severity of the defect has exceeded the set safety or quality standards, there are potential risks or hidden dangers, if the defect warning assessment coefficient is less than the defect warning assessment coefficient threshold, no warning signal is generated.

[0011] Technical effects and advantages of the present invention: The present invention quantifies the risk degree of missing items in each area by identifying the burr area, debris area, foil fracture area and carbon layer cracking area, obtains the pixel anomaly coefficient, characteristic deviation coefficient and damaged dispersion coefficient, and quantifies the potential risks or hidden dangers by constructing a defect early warning evaluation model, which helps to quickly detect defects after electrode slitting, output the detection results, and issue alarms in time to reduce production losses while ensuring product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings; Figure 1 The figure is a flow chart of an early warning method for electrode cut defects based on images of the present invention. DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0014] Example 1 like Figure 1 A new image-based prediction method for electrode cutout defects is presented. The present invention quantifies the risk degree of missing items in each area by identifying the burr area, debris area, foil fracture area and carbon layer cracking area, obtains the pixel anomaly coefficient, characteristic deviation coefficient and damaged dispersion coefficient, and quantifies the potential risks or hidden dangers by constructing a defect early warning evaluation model, which helps to quickly detect defects after electrode slitting, output the detection results, and issue alarms in time to reduce production losses while ensuring product quality.

[0015] Collect threads and start multi-threaded defect detection threads to detect burr and debris areas at the pole piece cut, foil fracture areas, and carbon layer cracking areas; S2: extract the corresponding region of interest in each thread, quickly locate the region where the pole piece is located by reducing the original image and using the features of the pole piece region, and lock the region of interest of each thread in the region where the pole piece is located; S3: collecting defect information about burrs and debris features, defect information about carbon layer cracking features, and defect information about foil fracture features in the region of interest of each thread; S4: Conduct a comprehensive analysis on the collected defect information and build a defect warning model to quantify the potential hazards of the pole piece cut.

[0016] In step 1, multiple image acquisition threads are defined. Each image acquisition thread is used for one type of defect detection. High-resolution camera equipment is used to capture images of the electrode cut in real time from the production line. The acquired images are sequentially stored in a linked list structure. Specifically, the images are added to the end of the linked list, and it is ensured that each new image acquired can be correctly added and recorded. Each image acquisition thread works independently and processes multiple images in parallel; Each detection thread takes an image from the head of the linked list (the first image collected) for processing. After successful acquisition, the head of the linked list moves downward, indicating that a node has been processed. After acquiring the image, the thread deletes the head node of the linked list to avoid repeated processing of the detected image, thus ensuring the synchronization and coordination of multiple threads when using the linked list. After each thread acquires the image, it starts the defect detection algorithm of its own thread to obtain the defects existing at the electrode cut. When the detection is completed, the defect detection result will be sent to the host computer through the callback function (for result summary, analysis or display). The callback function mechanism allows the detection thread to return the result immediately after completing the task to avoid blocking the main thread. After the detection is completed, if there is no new image, the thread will enter a waiting state until a new image is collected at the end of the linked list. Once the new image enters the linked list, the thread will obtain the image from the head of the linked list again and start a new defect detection process. The whole process is executed in a loop to ensure that the newly collected image can be processed in time.

[0017] Among them, this embodiment provides three defect detection methods, that is, three defect detection algorithms for image acquisition threads are designed, including the burr and debris area at the pole piece cut, the foil fracture area, and the carbon layer crack area.

[0018] In step 2, for each thread, the original image containing the polar slice is input, and the input image is reduced according to the set reduction ratio; It should be noted that the purpose of reducing the input image is to speed up subsequent processing and reduce the amount of calculation by lowering the image resolution, making subsequent area positioning more efficient. The reduction ratio is determined by professional staff.

[0019] Using the features of the pole piece area to quickly locate the pole piece area, including using a specific image processing algorithm to extract the features of the pole piece area, and determining the approximate position of the pole piece in the image through the extracted features; Among them, in order to quickly locate the area where the pole piece is located, the following steps and image processing algorithms can be used: Edge detection: Edge detection algorithms are used to identify significant boundaries and contours in an image, which helps to extract the contour features of the pole piece. Commonly used algorithms include Canny edge detection, Sobel operator, and Prewitt operator. Morphological operations: Morphological operations are used to process the shape features of the image, helping to further extract and optimize the polar region, reducing the boundaries of objects in the image through erosion, helping to remove small noise points, and increasing the boundaries of objects in the image through expansion, helping to fill small holes and connect broken areas; Region segmentation: Segment the image into different regions to better identify the polar regions and divide the image into different regions based on the pixel values.

[0020] Each thread is responsible for processing a specific region of interest (ROI) and performing defect detection. Multiple threads work simultaneously, and each thread executes the corresponding detection algorithm within its own ROI to detect the specific location and type of defects.

[0021] In step 3, for the burr area and debris area, the characteristics of the burrs and debris can be used to identify the characteristics of these defects using image processing technology, and the defect information about the burr and debris characteristics can be collected. After collection, the defect information about the burr and debris characteristics is represented by the pixel value abnormality coefficient.

[0022] Among them, since the burrs may extend out of the pole piece area, the found pole piece area is enlarged to prevent the burrs from being missed. According to the detection accuracy requirements, the area expanded by 10 pixels perpendicular to the movement direction can meet the detection requirements. Finally, the corresponding area is taken out from the original image according to the proportion and the ROI area is locked.

[0023] The specific steps of obtaining chips and burrs include: Perform rectangular expansion on the foil image to eliminate pixels where the foil affects the optical properties of the surrounding area; Invert the pixel values ​​of the rectangular expanded foil image to obtain an inverted image; Combine the foil image with the negated image, extract the difference between the foil image and the negated image, that is, perform a bitwise AND operation on the negated image and the original foil image, and the resulting image shows that the overlapping part of the foil area and its negated image is a possible defect area; Define Blob, apply Blob analysis algorithm to the result image after bitwise AND operation, identify the connected pixel areas in the image, and use blob analysis to extract the defect area; Apply high-order horizontal closing operation to the defect area extracted by blob6 to remove small horizontal gaps and noise to obtain a more coherent defect area image; The image processed by the closing operation is further subjected to blob extraction to obtain the final defect area.

[0024] The steps of obtaining the foil image include: By stretching the image of the region of interest in the thread, typically by image processing techniques such as image scaling or transformation, to enhance the contrast and definition of the foil features; Using a rectangular structure element to perform an opening operation on the stretched image to remove smaller burrs and debris, wherein the opening operation includes first performing an erosion operation and then performing a dilation operation; Performing a closing operation on the image after the rectangular opening operation to close the non-linked area of ​​the foil in a short distance, thereby obtaining a more coherent foil area, wherein the closing operation includes first performing a dilation operation and then performing an erosion operation; Blob analysis algorithm is applied to extract connected areas in the image, extract areas containing foil features, and eliminate non-foil areas by comparing Blob features (such as area and shape) to screen out real foil areas.

[0025] Burrs and debris are very important for early warning of defects in pole piece cuts for the following reasons: If burrs exist at the cut of the electrode, it may cause a short circuit inside the battery. The burrs may penetrate other electrodes or diaphragms, causing electrical performance problems of the battery, and cause mechanical damage during the assembly process of the battery, affecting the long-term stability and reliability of the battery. Debris may cause the dispersion of materials inside the battery, affect the contact area of ​​the pole pieces, reduce battery performance, and may cause uneven chemical reactions inside the battery, shortening the battery's service life.

[0026] By effectively detecting burrs and debris at the electrode cuts, potential production problems can be discovered early, thereby reducing subsequent rework and repairs, improving production efficiency, reducing the probability of battery failure in actual use, and improving the overall quality of the product.

[0027] The logic for obtaining the pixel abnormality coefficient is as follows: for an image containing burrs and debris defects, the pixel value of each pixel point in the region of interest in the thread is obtained by converting the original image into a grayscale image, and the pixel value of each pixel point in the region of interest in the thread is marked as: , where n=1, 2, 3, ..., N, N is a positive integer, and n is the number of the pixel in the grayscale image; Calculate the mean and standard deviation of the pixel values ​​in the grayscale image, and mark the mean and standard deviation of the pixel values ​​in the grayscale image as: and ,in, , ; Set the lowest pixel value threshold and the highest pixel value threshold, and mark the lowest pixel value threshold and the highest pixel value threshold as: and , obtain the number of pixel points below the minimum threshold of pixel value, and mark the number of pixel points below the minimum threshold of pixel value as: a, obtain the number of pixel points above the minimum threshold of pixel value, and mark the number of pixel points above the minimum threshold of pixel value as: b; It should be noted that the pixel threshold is used to identify burrs and debris. Burrs usually appear at the edge of the pole piece and are often brighter than the surrounding background or other parts of the pole piece. Therefore, setting a higher pixel value range can better cover all manifestations of burrs. Debris is often small, randomly distributed objects that may be slightly darker than burrs because they are usually located on the surface or edge of the pole piece and do not reflect as much light as burrs. Therefore, the pixel value may be low. The normal area of ​​the pole piece has a more uniform brightness in the image and is within a medium pixel value range.

[0028] Calculate the pixel anomaly coefficient, the calculation formula is: ;in, is the pixel abnormality coefficient.

[0029] It can be seen from the formula that the larger the pixel anomaly coefficient is, the more burr areas and debris areas there may be in the image, and the higher the possibility of defects.

[0030] For the pole piece and the carbon layer cracking area in the pole piece, the characteristics of the carbon layer cracking can be used to identify the characteristics of these defects using image processing technology, and the defect information about the carbon layer cracking characteristics can be collected. After collection, the defect information about the carbon layer cracking characteristics is represented by the characteristic deviation coefficient.

[0031] The specific cracking areas of the pole piece and the carbon layer in the pole piece include: According to the threads in the cracked area of ​​the carbon layer, the image of the region of interest in the thread is obtained, the image of the region of interest is horizontally opened, and the small burrs and interference in the horizontal direction of the image are removed using the horizontal structure element to accurately extract the pole piece area; Perform a rectangular closing operation on the image after the horizontal opening operation to reduce the influence of the dark area in the image and make the polar area more coherent; The image after the rectangular closing operation is opened again horizontally to further remove horizontal interference in the image and further smooth the image, making the features of the polar region more obvious and consistent; Blob analysis algorithm is applied to extract connected areas in the image, and polar regions containing carbon layer cracking information are extracted. These regions represent the parts containing carbon layer cracking, and the Blob extraction operation marks these regions; Perform a rectangular closing operation on the image extracted from the Blob to remove the cracks and make the pole area more coherent; Apply Blob analysis algorithm to extract connected areas in the image, extract polar regions without holes, and mark these areas; Perform an XOR operation on the image with cracks and the image without cracks to separate the area with cracks and the area without cracks; The Blob analysis algorithm is applied to extract the crack area in the image after XOR operation.

[0032] Early warning of cracks in pole piece cuts is very important, mainly because cracks may cause or indicate the following problems: Cracks will reduce the strength of the pole piece structure, which may make the pole piece unable to withstand the expected mechanical stress, and cracks may be early signs of fatigue or damage to the pole piece structure; Cracks in the pole piece may affect the conductivity of the battery, reduce the charging and discharging efficiency of the battery, and may cause poor contact between the pole piece and other battery components, thereby shortening the service life of the battery; Cracks may expose materials inside the pole piece, increase the risk of short circuit, and pose a threat to battery safety. Cracks may also cause electrolyte leakage inside the battery, causing environmental pollution or damage to equipment.

[0033] The acquisition logic of the characteristic deviation coefficient is: obtain an image containing a crack area, obtain the characteristics of each crack in the image, and mark the characteristics of each crack in the image as: , ,in, represents the width of each i-th crack in the image, represents the length of the i-th crack in the image, i=1, 2, 3, ..., I, I is a positive integer, and i is the number of the crack in the image; Set the crack width threshold and crack length threshold, and mark the crack width threshold and crack length threshold as: and ; It should be noted that the crack width threshold and the crack length threshold are specifically set by personnel in professional fields, and the crack width threshold and the crack length threshold refer to the limits on the width and length of a crack.

[0034] Calculate the characteristic deviation coefficient using the formula: ;in, is the characteristic deviation coefficient.

[0035] It can be seen from the formula that the larger the characteristic deviation coefficient is, the larger the area of ​​the crack region may be and the higher the possibility of defects.

[0036] For the broken area of ​​the foil, the characteristics of the broken foil can be used to identify the characteristics of these defects using image processing technology, and to collect defect information about the broken foil characteristics. After collection, the defect information about the broken foil characteristics is represented by the damaged dispersion coefficient.

[0037] The specific steps of obtaining the foil fracture area include: Extract the foil area from the image, determine the upper and lower edges of the foil in the extracted foil area, count the pixel values ​​of these edges, calculate the pixel distance between the upper and lower edges, and take the median of all measured values ​​as the thickness of the foil; Extract the pole piece area from the image, determine the upper and lower edges of the pole piece, count the pixel distances of these edges, and take the median value as the thickness of the pole piece; In the area where the foil exists, calculate the distance from the upper edge of the foil to the upper edge of the pole piece, sum and average all the calculation results, calculate the distance from the lower edge of the foil to the lower edge of the pole piece, sum and average them; Extract the broken foil area in the image, determine two adjacent foil areas, record the upper right corner coordinates of the previous foil area and the lower left corner coordinates of the next foil area, and calculate the size of the broken part based on the coordinates of the two adjacent foil areas.

[0038] The foil fracture area is of great significance for the early warning of the pole piece cut defect, mainly because the foil fracture is closely related to the structural integrity of the pole piece cut. The following are the reasons: Foil fracture may mean that the material has been subjected to external forces such as stretching, squeezing or bending, resulting in structural damage. Foil is an important component of the battery electrode, maintaining the mechanical strength and electrical conductivity of the electrode. Once the foil breaks, the overall strength of the material will be reduced, increasing the probability of defects at the cut; The pole piece cutout needs to remain intact and smooth to ensure normal battery function. If the foil is broken or damaged near the cutout, this may aggravate the weakening of the pole piece cutout and lead to further defects. The pole piece and foil in the battery jointly affect the current conduction efficiency. Foil breakage will cause uneven current distribution, which will affect battery performance. In addition, breakage may also cause local heating, increasing the risk of thermal runaway of the battery.

[0039] If the fracture area is near the cut, burrs or debris on the pole piece are more likely to accumulate, leading to short circuits or other electrochemical reactions. These problems pose a potential threat to battery safety. Therefore, foil fracture can serve as an early warning of pole piece cut defects and help prevent more serious failures.

[0040] The logic for obtaining the damage dispersion coefficient is: obtain the coordinate point of the broken part according to the coordinates of two adjacent foil regions, and mark the coordinate point of the broken part as: ,in, is the position of the pth pixel point on the x-axis at the fractured part, is the position of the p-th pixel point on the y-axis at the position of the broken part, p=1, 2, 3, ..., P, P is a positive integer, and p is the number of the pixel point at the position of the broken part; The average value and standard deviation of the pixel points at the position of the broken part on the x-axis are calculated, and the average value and standard deviation of the pixel points at the position of the broken part on the x-axis are marked as: and , the mean value and standard deviation of the pixel points at the broken part on the y-axis, and the mean value and standard deviation of the pixel points at the broken part on the x-axis are marked as: and ; in, , , , ; The coefficient of variation of the pixel point at the position of the fractured part on the x-axis and the coefficient of variation of the pixel point at the position of the fractured part on the y-axis are obtained, and the coefficient of variation of the pixel point at the position of the fractured part on the x-axis and the coefficient of variation of the pixel point at the position of the fractured part on the y-axis are marked as: and ; Calculate the damage dispersion coefficient using the following formula: ;in, is the damage dispersion coefficient.

[0041] From the formula, we can see that the larger the damage dispersion coefficient, the more fracture points are distributed in different positions, and the crack propagation directions are inconsistent, which may indicate extensive damage to the material or that the material structure is affected by multiple external forces. The risk is higher and may cause damage to a wider range.

[0042] In step 4, the defect information about burrs and debris features, the defect information about carbon layer cracking features, and the defect information about foil fracture features are comprehensively analyzed. By weighted calculation of pixel abnormality coefficient, feature deviation coefficient, and damaged dispersion coefficient, a defect warning evaluation model is constructed to generate a defect warning evaluation coefficient, which is expressed as follows: ;in, , , are the proportional coefficients of pixel anomaly coefficient, feature deviation coefficient, and damaged dispersion coefficient, respectively. , , Both are greater than 0.

[0043] Set a defect warning assessment coefficient threshold, compare the defect warning assessment coefficient with the defect warning assessment coefficient threshold, if the defect warning assessment coefficient is greater than the defect warning assessment coefficient threshold, generate a warning signal, indicating that the severity of the defect has exceeded the set safety or quality standards, there are potential risks or hidden dangers, if the defect warning assessment coefficient is less than the defect warning assessment coefficient threshold, no warning signal is generated.

[0044] The present invention quantifies the risk degree of missing items in each area by identifying the burr area, debris area, foil fracture area and carbon layer cracking area, obtains the pixel anomaly coefficient, characteristic deviation coefficient and damaged dispersion coefficient, and quantifies the potential risks or hidden dangers by constructing a defect early warning evaluation model, which helps to quickly detect defects after electrode slitting, output the detection results, and issue alarms in time to reduce production losses while ensuring product quality.

[0045] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0046] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0047] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0048] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0049] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0050] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0051] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An image-based early warning method for pole piece cut defects, characterized in that: The following steps are involved: S1: Start the image acquisition thread, start the multi-thread defect detection thread, detect the burr and debris area at the pole piece cut, the foil fracture area, and the carbon layer crack area; S2: extract the corresponding region of interest in each thread, quickly locate the region where the pole piece is located by reducing the original image and using the features of the pole piece region, and lock the region of interest of each thread in the region where the pole piece is located; S3: collecting defect information about burrs and debris features, defect information about carbon layer cracking features, and defect information about foil fracture features in the region of interest of each thread; S4: Conduct a comprehensive analysis of the collected defect information, build a defect warning model, quantify the potential hazards of the pole piece cuts, and issue a warning in a timely manner.

2. The image-based early warning method for electrode cut defects according to claim 1, characterized in that: Defect information about burr and chip characteristics, including: Defect information about burrs and debris features is represented by pixel value abnormality coefficients; The logic for obtaining the pixel abnormality coefficient is as follows: for an image containing burrs and debris defects, the pixel value of each pixel point in the region of interest in the thread is obtained by converting the original image into a grayscale image, and the pixel value of each pixel point in the region of interest in the thread is marked as: , where n=1, 2, 3, ..., N, N is a positive integer, and n is the number of the pixel in the grayscale image; Calculate the mean and standard deviation of the pixel values ​​in the grayscale image, and mark the mean and standard deviation of the pixel values ​​in the grayscale image as: and ,in, , ; Set the lowest pixel value threshold and the highest pixel value threshold, and mark the lowest pixel value threshold and the highest pixel value threshold as: and , obtain the number of pixel points below the minimum threshold of pixel value, and mark the number of pixel points below the minimum threshold of pixel value as: a, obtain the number of pixel points above the minimum threshold of pixel value, and mark the number of pixel points above the minimum threshold of pixel value as: b; Calculate the pixel anomaly coefficient, the calculation formula is: ;in, is the pixel abnormality coefficient.

3. The image-based early warning method for electrode cut defects according to claim 1, characterized in that: Defect information about the characteristics of carbon layer cracking, including: The defect information about the cracking characteristics of the carbon layer is represented by a characteristic deviation coefficient; The acquisition logic of the characteristic deviation coefficient is: obtain an image containing a crack area, obtain the characteristics of each crack in the image, and mark the characteristics of each crack in the image as: , ,in, represents the width of each i-th crack in the image, represents the length of the i-th crack in the image, i=1, 2, 3, ..., I, I is a positive integer, and i is the number of the crack in the image; Set the crack width threshold and crack length threshold, and mark the crack width threshold and crack length threshold as: and ; Calculate the characteristic deviation coefficient using the formula: ;in, is the characteristic deviation coefficient.

4. The image-based early warning method for electrode cut defects according to claim 1, characterized in that: Defect information about the foil fracture characteristics, including: The defect information about the fracture characteristics of the foil is expressed by the damage dispersion coefficient; The logic for obtaining the damage dispersion coefficient is: obtain the coordinate point of the broken part according to the coordinates of two adjacent foil regions, and mark the coordinate point of the broken part as: ,in, is the position of the pth pixel point on the x-axis at the fractured part, is the position of the p-th pixel point on the y-axis at the position of the fractured part, p=1, 2, 3, ..., P, P is a positive integer, and p is the number of the pixel point at the position of the fractured part; The average value and standard deviation of the pixel points at the position of the broken part on the x-axis are calculated, and the average value and standard deviation of the pixel points at the position of the broken part on the x-axis are marked as: and , the mean value and standard deviation of the pixel points at the broken part on the y-axis, and the mean value and standard deviation of the pixel points at the broken part on the x-axis are marked as: and ; in, , , , ; The coefficient of variation of the pixel point at the position of the fractured part on the x-axis and the coefficient of variation of the pixel point at the position of the fractured part on the y-axis are obtained, and the coefficient of variation of the pixel point at the position of the fractured part on the x-axis and the coefficient of variation of the pixel point at the position of the fractured part on the y-axis are marked as: and ; Calculate the damage dispersion coefficient using the following formula: ;in, is the damage dispersion coefficient.

5. The image-based early warning method for electrode cut defects according to claim 1, characterized in that: Build a defect warning model, including: The defect information about burrs and debris features, the defect information about carbon layer cracking features, and the defect information about foil fracture features are comprehensively analyzed. By weighted calculation of pixel abnormality coefficient, feature deviation coefficient, and damaged dispersion coefficient, a defect warning evaluation model is constructed to generate a defect warning evaluation coefficient, which is expressed as follows: ;in, , , are the proportional coefficients of pixel anomaly coefficient, feature deviation coefficient, and damaged dispersion coefficient, respectively. , , Both are greater than 0.

6. The image-based early warning method for electrode cut defects according to claim 1, characterized in that: Quantify the potential hazards of pole piece incisions and provide timely warnings, including: Set a defect warning assessment coefficient threshold, compare the defect warning assessment coefficient with the defect warning assessment coefficient threshold, if the defect warning assessment coefficient is greater than the defect warning assessment coefficient threshold, generate a warning signal, indicating that the severity of the defect has exceeded the set safety or quality standards, there are potential risks or hidden dangers, if the defect warning assessment coefficient is less than the defect warning assessment coefficient threshold, no warning signal is generated.

Images