End face burr defect detection method and system suitable for pole piece slitting and storage medium

Abstract

The invention provides an end face burr defect detection method suitable for pole piece slitting, and the method comprises the steps: obtaining a collection image of a to-be-detected pole piece slitting end face, extracting an actual foil in the collection image, and generating a skeleton line sub-pixel contour line segment of the actual foil; combining adjacent sub-pixel contour line segments at the fracture into one segment, and fitting into a straight line segment; and according to the straight line segment and the height of the foil, a virtual foil is obtained, the protruding contour of the actual foil protruding out of the virtual foil at the fracture is determined, and a burr defect identification result is generated in combination with a preset burr threshold value. The technical problem that in the prior art, the detection efficiency of pole piece slitting end face burr defects is low is effectively solved.

Description

Technical Field

[0001] This application relates to the field of battery electrode inspection technology, specifically to a method, system, and storage medium for detecting end-face burr defects suitable for electrode cutting. Background Technology

[0002] In the battery manufacturing process, the coated positive and negative electrode sheets need to be slit. Slitting typically involves cutting a wide roll of battery electrode material into multiple sheets of equal width. During the slitting process, burrs are generated at the cutting surfaces along the electrode cut lines, affecting the battery's production quality. Therefore, burr detection on the cut surfaces of the electrode sheets is of great importance. However, existing burr detection methods suffer from low detection efficiency. Summary of the Invention

[0003] In view of this, this application provides a method, system and storage medium for detecting end face burr defects suitable for electrode cutting, so as to solve the technical problem of low detection efficiency in the prior art.

[0004] In a first aspect, this application provides a method for detecting end-face burr defects suitable for electrode cutting, including: Acquire images of the end face of the electrode to be detected after slicing, extract the actual foil from the acquired images, and generate sub-pixel contour segments of the skeleton line of the actual foil. The adjacent sub-pixel contour segments at the merged break point are combined into one segment and fitted as a straight line segment. Based on the height of the straight line segment and the foil, a virtual foil is obtained. The protruding contour of the actual foil protruding from the virtual foil at the break point is determined. Combined with a preset burr threshold, a burr defect identification result is generated.

[0005] Secondly, this application provides a burr defect detection system suitable for electrode cutting, comprising: The extraction module is used to acquire images of the end face of the electrode to be detected, extract the actual foil from the acquired images, and generate sub-pixel contour segments of the skeleton line of the actual foil. The fitting module is used to merge adjacent sub-pixel contour segments at the break point into one segment and fit it into a straight line segment. The detection module is used to obtain the virtual foil based on the height of the straight line segment and the foil, determine the protruding contour of the actual foil protruding from the virtual foil at the break point, and generate burr defect identification results by combining the preset burr threshold.

[0006] Thirdly, this application provides a computer device, including: a memory and a processor, which are interconnected and communicate with each other. The memory stores computer instructions, and the processor executes the computer instructions to realize the end-face burr defect detection method suitable for electrode cutting in the first aspect.

[0007] Fourthly, this application provides a computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the end-face burr defect detection method suitable for electrode cutting according to the first aspect.

[0008] The method, system, and storage medium for detecting end-face burr defects suitable for electrode slitting provided in this application have at least the following advantages: A simplified process of sub-pixel contour extraction, merging and fitting, and virtual foil comparison adapts to high-speed scenarios in electrode slitting, meeting the need for rapid detection. The foil contour is described with sub-pixel precision, and combined with the construction of a benchmark virtual foil, accurate positioning of burr defects is achieved, balancing detection speed and accuracy; the overall detection process eliminates complex and redundant operations, reducing the computational load of the algorithm. Attached Figure Description

[0009] Figure 1 A schematic diagram of a method for detecting end-face burr defects suitable for electrode cutting is shown; Figure 2 An image of the end face of a slit electrode is shown; Figure 3 It shows the basis Figure 2 A schematic diagram after extracting the coating area; Figure 4 It shows the basis Figure 3 A partial schematic diagram after the foil material has been extracted; Figure 5 A schematic diagram of a local foil skeleton line subpixel contour segment is shown; Figure 6 A schematic diagram of a cut foil is shown; Figure 7 A partial schematic diagram of a fitted line segment is shown; Figure 8 A partial schematic diagram of the expansion of a straight line segment is shown; Figure 9 A partial schematic diagram representing a protruding profile is shown; Figure 10 A schematic diagram showing a burr defect is provided. Figure 11 A schematic diagram of an end-face burr defect detection system suitable for electrode cutting is shown. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, not all of them. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0011] Example 1 In actual electrode slitting scenarios, the electrode slitting speed is approximately 120m per minute, which corresponds to acquiring approximately 212 images per second (4096*1000, 2.25μm = 1px). Existing technologies cannot meet the requirement for rapid and efficient detection of burr defects on the end face of rapidly slitting electrodes.

[0012] In view of this, this application provides a method for detecting end-face burr defects suitable for electrode cutting, comprising the following steps. See [link to relevant documentation]. Figure 1 As shown, S1. Acquire the image of the end face of the electrode to be detected after slicing, extract the actual foil from the acquired image, and generate the sub-pixel contour line segment of the skeleton line of the actual foil. S2. Merge adjacent sub-pixel contour segments at the break point into one segment and fit it as a straight line segment. S3. Based on the height of the straight line segment and the foil, obtain the virtual foil, determine the protruding contour of the actual foil protruding from the virtual foil at the break point, and generate burr defect identification results by combining the preset burr threshold.

[0013] This application acquires images of the electrode cutting end face, accurately extracts the actual foil area from the images, and then generates sub-pixel contour segments of the actual foil skeleton line. These sub-pixel contour segments overcome pixel-level precision limitations, providing a more detailed description of the foil skeleton contour morphology and laying the foundation for subsequent accurate detection. For adjacent sub-pixel contour segments scattered at the break point, a merging process is performed, integrating them into continuous line segments. These segments are then transformed into straight line segments using a fitting algorithm. These straight line segments serve as a reference skeleton contour for the foil break point. Combining the actual height of the foil, the fitted straight line segments are pixel-dilated in the column direction to construct a virtual foil without burr defects, i.e., the ideal foil contour. By comparing the pixel distribution of the actual foil and the virtual foil, protruding contours of the actual foil exceeding those of the virtual foil are selected and compared with a preset burr threshold to finally generate burr defect identification results.

[0014] This application specifically addresses the core pain point of low detection efficiency in existing technologies. Through a simplified process of sub-pixel contour extraction, merging and fitting, and virtual foil comparison, it adapts to high-speed scenarios such as electrode cutting, meeting the need for rapid detection. It describes the foil contour with sub-pixel precision and, combined with the construction of a benchmark virtual foil, achieves accurate localization of burr defects, balancing detection speed and accuracy. The overall detection process eliminates complex and redundant operations, reducing the computational load of the algorithm.

[0015] Extracting the actual foil material from the acquired images includes: The coated area in the acquired image is extracted based on a first dynamic threshold, and the actual foil material in the coated area is extracted based on a second dynamic threshold. The acquired image is shown below. Figure 2 As shown, the extracted coating area is as follows: Figure 3 As shown, the extracted foil material is as follows Figure 4 As shown.

[0016] The acquired image is segmented based on the first dynamic threshold to filter out the coating area of ​​the electrode. The dynamic threshold can be adaptively adjusted in real time according to pixel contrast, etc., to avoid the failure of extraction under complex conditions such as changes in illumination and uneven coating thickness. Within the obtained coating area, it is further segmented based on the second dynamic threshold to accurately extract the actual foil area. By first coarsely locating and then finely extracting, interference from irrelevant areas such as coating layer and background is eliminated.

[0017] Compared to fixed thresholds, dynamic thresholds offer greater adaptability, effectively addressing environmental changes and material variations during electrode slitting and enhancing the stability of foil extraction. Two-step extraction with layer-by-layer focusing effectively eliminates interference from coating layers, background noise, and other factors, improving the extraction accuracy of the actual foil area. This provides reliable foundational data for subsequent contour generation and defect identification. The extraction algorithm is simple, computationally efficient, and effectively improves detection efficiency.

[0018] See Figure 5 As shown, generating the sub-pixel contour segments of the skeleton line of the actual foil includes: determining the center of the contour of each column of the actual foil as a skeleton point; wherein, there are no skeleton points at the break point; and smoothing all skeleton points to obtain the sub-pixel contour segments of the skeleton line.

[0019] The extracted actual foil area is scanned column by column. The center position of the foil outline in each column is a skeleton point. The set of skeleton points initially forms the core outline of the foil. All skeleton points are processed by smoothing algorithms, such as Gaussian smoothing and mean filtering, to eliminate image noise and skeleton point fluctuations caused by minor unevenness on the foil surface, ultimately forming continuous and smooth subpixel skeleton line outline segments.

[0020] Using the center of the column contour as the skeleton point, the core contour shape of the foil is accurately captured. Compared with the pixel level, the sub-pixel level accuracy significantly improves the delicacy and accuracy of the contour description, while reducing the amount of data processing. The smoothing process effectively filters noise interference and avoids local small fluctuations from affecting the overall continuity of the contour, providing stable basic data for subsequent merging of adjacent line segments and straight line fitting. The process logic for generating the skeleton line sub-pixel contour line segments of the actual foil is simple and efficient, balancing accuracy and speed, and is suitable for high-speed detection scenarios.

[0021] Before merging adjacent sub-pixel contour segments at the break point into one segment, it is determined whether the horizontal projection length of any two adjacent sub-pixel contour segments at the break point exceeds a preset projection length threshold. For sub-pixel contour segments exceeding the preset projection length threshold, they are trimmed according to the preset projection length threshold to become one of the target segments for merging. See the foil material corresponding to the trimmed segment for details. Figure 6 As shown, by filtering and cutting through the port, the foil at the break point can be inspected in a targeted manner, without having to inspect all foil areas, thus improving inspection efficiency.

[0022] The horizontal projection length is the projected distance of a line segment in the horizontal direction. The calculated horizontal projection length is compared with a preset projection length threshold. If the horizontal projection length of a line segment exceeds this threshold, it indicates that the line segment may be too long, and it is trimmed, retaining only the portion that meets the threshold requirement as one of the target line segments for subsequent merging operations. The trimming operation only applies to line segments exceeding the threshold, without destroying the integrity of valid contour line segments. This effectively reduces computational load and improves detection efficiency while ensuring data validity.

[0023] For subpixel contour segments that are not greater than a preset projection length threshold, the adjacent subpixel contour segments on the other side are included in one of the merged segments; the horizontal projection length of the newly included subpixel contour segments is repeatedly judged to be greater than the preset projection length threshold, so as to trim the segments or add more included segments.

[0024] For line segments whose horizontal projection length does not exceed the threshold, the adjacent sub-pixel contour line segments on the other side are included in the set of line segments to be merged; for newly included adjacent line segments, the judgment-trimming / continuing inclusion process is repeated. In simple terms, this means merging multiple breaks that are close to each other, eliminating the need for multiple separate processing steps, enhancing the versatility and robustness of the method, further reducing computational load, and improving detection efficiency.

[0025] Based on the straight line segment and the height of the foil, a virtual foil is obtained, and the protruding contour of the actual foil protruding from the virtual foil at the break point is determined, including: The straight line segments are expanded pixel by pixel in the column direction according to the height of the foil to obtain the virtual foil; The formula for determining the protrusion profile of the actual foil material protruding from the virtual foil material is: In this definition, pixels within the foil area are assigned a value of 1, while pixels outside the foil area are assigned a value of 0. The pixels that protrude from the virtual foil from the actual foil. For the actual foil pixels, These are the pixels of the virtual foil.

[0026] See Figure 7 and Figure 8 As shown, using the fitted straight line segment as a reference, pixel expansion is performed in the column direction according to the actual height of the foil, that is, expanding the pixel range along the column direction to construct a virtual foil area without burrs in an ideal state. Pixels within the foil area are defined as 1, and those outside the area as 0. By calculating the difference between the actual foil pixels and the virtual foil pixels and removing negative values ​​(i.e., removing depressions), the set of pixels where the actual foil protrudes from the virtual foil is obtained. This set is the protruding contour corresponding to the burrs, as shown in the figure. Figure 9 As shown. It should be noted here that, Figure 5 , Figure 7 , Figure 8 and Figure 9 The images are presented in a way that retains the coating and foil, but this is only for the purpose of making the technical solution of this application more intuitive. In fact, they are not retained in the actual processing. That is to say, the number of image pixels that actually need to be detected is very small, which can effectively improve the detection efficiency.

[0027] The method of generating virtual foil by pixel dilation accurately restores the ideal foil outline, providing a clear benchmark for the quantitative comparison of burr defects and avoiding subjective judgment bias; the mathematical formula-defined method of convex contour calculation enables accurate quantitative extraction of burr areas, allowing parameters such as burr height and range to be directly obtained through pixel calculation, improving the objectivity and data support of defect detection and ensuring detection efficiency.

[0028] The burr defect identification results are generated by combining a preset burr threshold, including: Determine whether the height of the protruding contour is greater than the preset burr threshold. If the height of the protruding contour is greater than the preset burr threshold, the protruding foil at the break point is determined to be a burr defect. Otherwise, filter.

[0029] Based on battery production quality standards and electrode material characteristics, a pre-set maximum allowable burr height, i.e., a preset burr threshold, is determined. The actual burr height is calculated by pixel-wise to obtain the height of the protruding contour of the actual foil protruding from the virtual foil. This protrusion height is then compared to the preset burr threshold. If the protrusion height exceeds the preset burr threshold, it indicates that the protrusion will affect battery production quality and is thus identified as a burr defect. (See also...) Figure 10 As shown, the foil material outlined in the diagram represents the detected burr defects. If the defect does not exceed the threshold, it is considered a normal deviation and is filtered out.

[0030] Establish quantitative defect judgment standards to avoid misjudgment and omission caused by subjective judgment, and improve the objectivity and consistency of test results; through a threshold filtering mechanism, accurately distinguish between valid defects and harmless deviations, reduce invalid alarms, and lower the cost of subsequent manual review; the preset threshold can be flexibly adjusted to adapt to the quality requirements of different electrode models and different production processes, and enhance the versatility of the method.

[0031] Example 2 This embodiment provides an end-face burr defect detection system suitable for electrode cutting, which is applied to the end-face burr defect detection method suitable for electrode cutting provided in Embodiment 1 above for illustration. See also Figure 11 As shown, the end-face burr defect detection system suitable for electrode cutting provided in this application may include the following multiple modules.

[0032] The extraction module is used to acquire images of the end face of the electrode to be detected, extract the actual foil from the acquired images, and generate sub-pixel contour segments of the skeleton line of the actual foil. The fitting module is used to merge adjacent sub-pixel contour segments at the break point into one segment and fit it into a straight line segment. The detection module is used to obtain the virtual foil based on the height of the straight line segment and the foil, determine the protruding contour of the actual foil protruding from the virtual foil at the break point, and generate burr defect identification results by combining the preset burr threshold.

[0033] The end-face burr defect detection system suitable for electrode cutting provided in this application embodiment can be applied to the end-face burr defect detection method suitable for electrode cutting provided in Embodiment 1 above. For relevant details, please refer to the above method embodiment. Its implementation principle and technical effect are similar, and will not be repeated here.

[0034] It should be noted that the end-face burr defect detection system suitable for electrode slicing provided in this embodiment is only illustrated by the above-described division of functional modules / units. In practical applications, the above functions can be assigned to different functional modules / units as needed, that is, the internal structure of the end-face burr defect detection system suitable for electrode slicing can be divided into different functional modules / units to complete all or part of the functions described above. Furthermore, the implementation method of the end-face burr defect detection method suitable for electrode slicing provided in the above-described embodiment 1 and the implementation system of the end-face burr defect detection system suitable for electrode slicing provided in this embodiment 2 belong to the same concept. The specific implementation process of the end-face burr defect detection system suitable for electrode slicing provided in this embodiment 2 is detailed in the above-described embodiment 1, and will not be repeated here.

[0035] Example 3 One embodiment of this application also provides a computer device, including but not limited to a processor and a memory, wherein the memory stores computer instructions, and the processor executes the computer instructions to implement the method described above.

[0036] One embodiment of this application also provides a computer-readable storage medium for storing a computer program that, when executed by a processor, implements the method described in the above-described method embodiments.

Claims

1. A method for detecting end-face burr defects suitable for electrode cutting, characterized in that, include: Acquire images of the end face of the electrode to be detected after slicing, extract the actual foil material from the acquired images, and generate the skeleton line subpixel contour line segment of the actual foil material. The adjacent sub-pixel contour segments at the merged break point are combined into one segment and fitted as a straight line segment. Based on the height of the straight line segment and the foil, a virtual foil is obtained, and the protruding contour of the actual foil protruding from the virtual foil at the break point is determined. Combined with a preset burr threshold, a burr defect identification result is generated.

2. The method for detecting end-face burr defects suitable for electrode cutting according to claim 1, characterized in that, The step of generating burr defect identification results by combining a preset burr threshold includes: Determine whether the height of the protruding contour is greater than a preset burr threshold; If the height of the protruding contour is greater than the preset burr threshold, the protruding foil at the break point is determined to be a burr defect. Otherwise, filter.

3. The method for detecting end-face burr defects suitable for electrode cutting according to claim 2, characterized in that, Extracting the actual foil material from the acquired image includes: The coating area in the acquired image is extracted based on a first dynamic threshold. The actual foil material in the coated area is extracted based on the second dynamic threshold.

4. The method for detecting end-face burr defects suitable for electrode cutting according to claim 3, characterized in that, The generation of the skeleton line subpixel contour segment of the actual foil includes: The center of the outline of each column of the actual foil is determined by a skeleton point; Smooth all skeleton points to obtain subpixel contour segments of the skeleton line.

5. The method for detecting end-face burr defects suitable for electrode cutting according to claim 4, characterized in that, Before the adjacent sub-pixel contour segments at the merged break point are considered as one segment, the following is also included: Determine whether the horizontal projection length of two adjacent sub-pixel contour segments at the break point is greater than the preset projection length threshold. For sub-pixel contour lines that are larger than the preset projection length threshold, they are trimmed according to the preset projection length threshold to be used as one of the target lines for merging.

6. The method for detecting end-face burr defects suitable for electrode cutting according to claim 5, characterized in that, Also includes: For sub-pixel contour segments that are not greater than the preset projection length threshold, the adjacent sub-pixel contour segments on the other side are included in one of the merged segments. Repeatedly determine whether the horizontal projection length of the newly included subpixel contour line segment is greater than the preset projection length threshold, in order to trim the line segment or add the included line segment.

7. The method for detecting end-face burr defects suitable for electrode cutting according to claim 6, characterized in that, The process of obtaining a virtual foil based on the height of the straight line segment and the foil material, and determining the protruding contour of the actual foil protruding from the virtual foil at the break point, includes: The straight line segments are pixel-expanded in the column direction according to the height of the foil to obtain a virtual foil; The formula for determining the protrusion profile of the actual foil material protruding from the virtual foil material is: In this definition, pixels within the foil area are assigned a value of 1, while pixels outside the foil area are assigned a value of 0. The pixels that protrude from the virtual foil from the actual foil. For the actual foil pixels, These are the pixels of the virtual foil.

8. A system for detecting end-face burr defects suitable for electrode cutting, characterized in that, include: The extraction module is used to acquire images of the end face of the electrode to be detected, extract the actual foil in the acquired images, and generate the skeleton line sub-pixel contour line segment of the actual foil. The fitting module is used to merge adjacent sub-pixel contour segments at the break point into one segment and fit it into a straight line segment. The detection module is used to obtain the virtual foil based on the height of the straight line segment and the foil, determine the protruding contour of the actual foil protruding from the virtual foil at the break point, and generate burr defect identification results by combining the preset burr threshold.

9. A computer device, characterized in that, include: The device includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes the computer instructions to implement the end-face burr defect detection method suitable for electrode cutting as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the end-face burr defect detection method suitable for electrode cutting as described in any one of claims 1-7.

Images