Method, apparatus and device for detecting electrode pole defect of lithium battery

Abstract

Disclosed in the present invention are a method, apparatus and device for detecting an electrode pole defect of a lithium battery, and a storage medium. The method comprises: acquiring a depth map and a 2D grayscale map of an electrode pole, wherein the size of the depth map is the same as that of the 2D grayscale map; performing equalization processing on the depth map to obtain a first image; performing filtering processing and edge detection on the first image to obtain an electrode pole region; according to an edge profile coordinate point of the electrode pole region, locating a detection position corresponding to the 2D grayscale image, so as to obtain a detection region; and detecting different types of defects in the detection region by means of a preset defect detection algorithm, so as to obtain a defect detection result. The detection quality can be improved, the detection speed can be ensured, the requirements of the industry for electrode pole inspection of lithium batteries can be satisfied, and the quality inspection requirements of the new energy industry for electrode pole inspection of lithium batteries can be met.

Description

A method, device and equipment for detecting defects in lithium battery poles

[0001] This invention claims priority date 2023-06-29 and priority number 2023107780802. Technical Field

[0002] The present invention relates to the technical field of lithium batteries, and in particular to a method, device and equipment for detecting defects in lithium battery poles. Background Art

[0003] Lithium batteries offer high energy density, long life, and are environmentally friendly. They are widely used in mobile electronic devices, electric vehicles, energy storage systems, and other fields. The performance and safety of lithium batteries are closely related to their internal structure, with the terminal being a crucial component. The quality and defects of the terminal directly impact the performance and safety of lithium batteries.

[0004] Currently, lithium battery pole defect detection primarily relies on non-destructive testing technologies such as X-ray testing, magnetic particle inspection, and ultrasonic testing. X-ray testing is a commonly used method that can detect defects and foreign matter inside the pole, but it requires specialized equipment and technicians and carries radiation hazards. Magnetic particle inspection and ultrasonic testing can detect surface and internal defects, but are less effective for detecting minor defects. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to propose a method, device and equipment for detecting defects in lithium battery poles, aiming to solve the problems in the prior art of excessively high requirements for equipment and technicians for defect detection of lithium battery poles, and poor detection effect of some small defects.

[0006] To achieve the above objectives, the present invention provides a method for detecting defects in lithium battery poles, the method comprising:

[0007] Obtaining a depth map and a 2D grayscale image of the pole, wherein the depth map and the 2D grayscale image have the same size;

[0008] Performing equalization processing on the depth map to obtain a first image;

[0009] Performing filtering and edge detection on the first image to obtain a polar column area;

[0010] Locating the detection position corresponding to the 2D grayscale image according to the edge contour coordinate points of the pole area to obtain a detection area;

[0011] The detection area is inspected for different defect types using a preset defect detection algorithm to obtain defect detection results.

[0012] Preferably, the method further comprises:

[0013] Determine whether the defect size at the corresponding position in the defect detection result is greater than a detection threshold; if so, determine that it is a real pole defect.

[0014] Preferably, the corresponding position is found in the depth map according to the obtained defect position; the defect size of the corresponding defect is calculated, including the defect length, width and depth, and the length, width and depth are compared with the corresponding thresholds of the detection parameters. When all three are greater than the corresponding thresholds, it is confirmed that the corresponding defect exists in the pole; otherwise, it is considered that the corresponding defect does not exist.

[0015] Preferably, whether the defect is real is determined by combining the concave-convex features of the defect on the depth map and the grayscale features of the defect on the 2D grayscale map.

[0016] Preferably, a set of depth values ​​is obtained on the depth map according to the obtained defect position, and the minimum value is taken as the depth value of the current defect. If the size and depth value of the defect are determined to meet the user-defined defect specifications, it is considered that a real defect exists.

[0017] Preferably, the filtering and edge detection of the first image to obtain the polar column area includes:

[0018] performing mean filtering on the first image to obtain a second image;

[0019] Calculating the second image using an edge detection operator to obtain an initial polar region;

[0020] Performing a closing operation and calculating contour coordinates on the initial pole region to determine a reference depth value of the pole;

[0021] Coordinate points in the first image that meet a preset range of the reference depth value are selected, and a set of all the coordinate points is used as the polar column area.

[0022] Preferably, the calculating the second image using an edge detection operator to obtain an initial polar region includes:

[0023] Calculating edge gradients in the X and Y directions of the second image using an edge detection operator to obtain edge gradient information;

[0024] extracting edge information from the first image according to the edge gradient information to obtain a third image having an edge gradient value;

[0025] The third image is binarized according to a preset gradient threshold to obtain the initial polar region.

[0026] Preferably, performing a closing operation on the initial pole region and calculating contour coordinates to determine a reference depth value of the pole includes:

[0027] Performing a closing operation on the initial pole region to obtain a first region;

[0028] Calculating the area of ​​the circumscribed rectangle of the first region and selecting the outline coordinates of the pole region;

[0029] shrinking the outline coordinates according to a preset distance to obtain a plurality of second areas;

[0030] A plurality of depth values ​​of the second area are counted, and a median of the depth values ​​is taken as the reference depth value.

[0031] Preferably, the detecting of different defect types in the detection area by using a preset defect detection algorithm includes:

[0032] Detecting bump-type defects and scratch-type defects in the detection area using a detection model built based on YOLO;

[0033] The notch type defects in the detection area are detected by a detection algorithm based on bandpass filtering and morphology.

[0034] Preferably, the detecting of notch-type defects in the detection area by using a detection algorithm based on bandpass filtering and morphology comprises:

[0035] Obtain the location of the aluminum hole in the center of the pole and shield it to determine the first defect signal;

[0036] Determine a suitable cutoff frequency by using bandpass filtering to filter out high-frequency and low-frequency components of the first defect signal, thereby obtaining a second defect signal having a specific frequency range;

[0037] The second defect signal is subjected to dilation and erosion operations through morphology to obtain the position of the target defect.

[0038] To achieve the above object, the present invention further provides a device for detecting defects in a lithium battery electrode, the device comprising:

[0039] an acquisition unit, configured to acquire a depth map and a 2D grayscale map of the pole, wherein the depth map and the 2D grayscale map have the same size;

[0040] A first processing unit, configured to perform equalization processing on the depth map to obtain a first image;

[0041] a second processing unit, configured to perform filtering and edge detection on the first image to obtain a polar column area;

[0042] a positioning unit, configured to locate the detection position corresponding to the 2D grayscale image according to the edge contour coordinate points of the pole area to obtain a detection area;

[0043] The detection unit is used to detect different defect types in the detection area using a preset defect detection algorithm to obtain a defect detection result.

[0044] In order to achieve the above-mentioned objectives, the present invention also proposes a device for detecting defects in lithium battery poles, comprising a processor, a memory, and a computer program stored in the memory, wherein the computer program is executed by the processor to implement the steps of a method for detecting defects in lithium battery poles as described in the above-mentioned embodiment.

[0045] In order to achieve the above objectives, the present invention also proposes a computer-readable storage medium, on which a computer program is stored. The computer program is executed by a processor to implement the steps of a method for detecting lithium battery pole defects as described in the above embodiment.

[0046] Beneficial effects:

[0047] The above solution, by detecting different types of defects through preset defect detection algorithms, can optimize detection efficiency, improve detection quality while ensuring detection speed. By combining the advantages of traditional image algorithms and AI, it can meet the industry's demand for lithium battery pole detection and achieve the quality inspection requirements of lithium battery pole detection in the new energy industry.

[0048] The above solution determines the true defect by combining the depth map and 2D grayscale image of the battery pole. It judges defects from multiple dimensions, meets the various specifications defined by users for a defect, meets user needs to the greatest extent, and can greatly reduce the misjudgment rate.

[0049] The above scheme determines the actual area of ​​the pole by obtaining the reference depth value of the pole, which can filter out the surrounding interference information and achieve accurate positioning of defects in the detection area.

[0050] The above solution uses image detection to detect defects without manual visual inspection, which not only greatly saves labor costs, but also has quantitative standards for defect judgment compared to the human eye. This ensures that the products produced will not have uneven quality due to human subjective judgment, thus ensuring product production quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0052] FIG1 is a flow chart of a method for detecting defects in a lithium battery pole provided by an embodiment of the present invention.

[0053] FIG2 is a schematic structural diagram of a device for detecting defects in a lithium battery pole provided by an embodiment of the present invention.

[0054] The realization of the objectives of the invention, the functional features and advantages will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Modes for Carrying Out the Invention

[0055] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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 making creative work are within the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention for which protection is sought, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0056] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, features specified as "first" or "second" may explicitly or implicitly include one or more of the features.

[0057] The present invention is described in detail below with reference to the embodiments.

[0058] 1 is a flow chart of a method for detecting defects in a lithium battery electrode provided by an embodiment of the present invention.

[0059] In this embodiment, the method includes:

[0060] S11, obtaining a depth map and a 2D grayscale map of the pole, wherein the depth map and the 2D grayscale map have the same size.

[0061] S12: Perform equalization processing on the depth map to obtain a first image.

[0062] The equalization processing performed on the depth map includes equalization processing achieved by limiting a contrast adaptive histogram.

[0063] In this embodiment, a set of depth maps and 2D grayscale images of the same size are obtained, and the depth maps are equalized using a limited contrast adaptive histogram. Compared with ordinary histogram equalization, limited contrast adaptive histogram equalization pays more attention to improving local contrast and better retains more details of the image.

[0064] S13: Perform filtering and edge detection on the first image to obtain a polar column area.

[0065] Furthermore, in step S13, filtering and edge detection are performed on the first image to obtain a polar region, including:

[0066] S13-1, performing mean filtering on the first image to obtain a second image;

[0067] S13-2, using an edge detection operator to calculate the second image to obtain an initial polar region;

[0068] S13-3, performing a closing operation on the initial pole region and calculating contour coordinates to determine a reference depth value of the pole;

[0069] S13-4, selecting coordinate points in the first image that meet a preset range of the reference depth value, and using a set of all the coordinate points as the polar column area.

[0070] In step S13-2, the edge detection operator is used to calculate the second image to obtain the initial polar region, including:

[0071] Calculating edge gradients in the X and Y directions of the second image using an edge detection operator to obtain edge gradient information;

[0072] extracting edge information from the first image according to the edge gradient information to obtain a third image having an edge gradient value;

[0073] The third image is binarized according to a preset gradient threshold to obtain the initial polar region.

[0074] Wherein, in step S13-3, performing a closing operation on the initial pole region and calculating contour coordinates to determine a reference depth value of the pole includes:

[0075] Performing a closing operation on the initial pole region to obtain a first region;

[0076] Calculating the area of ​​the circumscribed rectangle of the first region and selecting the outline coordinates of the pole region;

[0077] shrinking the outline coordinates according to a preset distance to obtain a plurality of second areas;

[0078] A plurality of depth values ​​of the second area are counted, and a median of the depth values ​​is taken as the reference depth value.

[0079] In this embodiment, although the contrast-limited adaptive histogram equalization operation on the depth map can enhance the local contrast, it will also amplify the noise in the original image. Therefore, further processing through mean filtering can suppress the noise, which is more beneficial to the subsequent processing of the image.

[0080] Since there are more than just pole regions in the image, in order to accurately locate the pole regions, the depth map region is further divided using a reference depth value standard. To obtain the actual pole region, an approximate initial pole region is first solved. The reason why only an approximate initial pole region can be obtained here is that directly calculating the pole edge cannot accurately locate the actual pole region due to issues such as blurred edges and close surrounding objects. The obtained area may be stuck to surrounding objects. Therefore, in this embodiment, an approximate initial pole region is first obtained.

[0081] Specifically, edge detection is first performed using the Sobel operator. The Sobel operator is a combination of Gaussian smoothing and differential operations, which can meet basic edge detection requirements with high efficiency. In this embodiment, the Sobel operator is used to calculate edge gradient information in the X and Y directions. The X and Y gradient information is then multiplied by 0.5 and added together to obtain the final edge gradient information. The resulting image containing edge gradient information is binarized using a preset gradient threshold to produce an edge-only image containing only values ​​of 0 and 255. In other words, the image contains edge gradient information, with the value at each pixel being a gradient value. By selecting an appropriate gradient threshold for binarization (i.e., pixel positions above the gradient threshold are set to 255, and pixel positions below the gradient threshold are set to 0), this operation can roughly determine the initial polar region.

[0082] When solving the pole area, it cannot be guaranteed that only the edges of the poles are included. The edges of some other objects around will also be solved at the same time, so it is necessary to screen all edge contours. In this embodiment, some small interferences are removed by the corrosion operation in the morphological operation, and then all edges are processed by the closing operation, and the positions where the edges are disconnected are connected to form complete closed edges. After that, the existing contour operator in OpenCV is used to calculate the circumscribed rectangular area of ​​all contours, and the contour with the largest area is the contour of the required pole area. According to the coordinates of the contour, the coordinates of the entire pole area can be known. Four pole areas are selected from the four directions of up, down, left and right. The contour coordinates are retracted a certain distance, and the areas of several poles are selected. The depth values ​​of these areas are counted and sorted, and the median is taken to further select the reference depth value. Specifically, the coordinate points are traversed toward the center in the four directions of up, down, left, and right, and the pixel value of the current point is determined. If it is 255, it belongs to the polar column area, otherwise it does not belong to the polar column area. After determining which points belong to the polar column area, the coordinate positions of some points in the area are selected, that is, the row and column of the image, and then the depth values ​​of the corresponding positions on the depth map are taken as statistical data. After sorting, the median is obtained as the benchmark depth value of the polar column.

[0083] Further, a depth value range is obtained by using the reference depth value and a certain value of deviation. The coordinate points that meet the requirements are screened out on the depth map with this depth value range. The set of all points is the real pole area. Since the depth value of the entire pole plane cannot be completely consistent, there will be some slight unevenness, which is reflected in the depth value as a tiny numerical deviation. Therefore, after the reference depth value is obtained, a value is subtracted from the reference depth value and a value is added to obtain a depth value range. The depth values ​​within this range can be determined as pole areas. By judging the depth value of each point on the depth map according to this standard, it can be determined which points actually belong to the pole area. In this way, the real pole area can be obtained and the area that actually needs to be detected can be accurately located.

[0084] S14, locating the detection position corresponding to the 2D grayscale image according to the edge contour coordinate points of the pole area to obtain a detection area.

[0085] S15, performing detection of different defect types on the detection area using a preset defect detection algorithm to obtain a defect detection result.

[0086] Furthermore, in step S15, the detection of different defect types in the detection area is performed using a preset defect detection algorithm, including:

[0087] S15-1, detecting bump type defects and scratch type defects in the detection area through a detection model built based on YOLO.

[0088] Furthermore, the construction process of the detection model based on YOLO includes:

[0089] Based on the characteristics of bumps and scratches, a large number of 2D grayscale images of defects containing bump areas and scratch areas are collected and annotated, including the images and corresponding annotation information, that is, the location and category of the defects, to obtain a training dataset;

[0090] The detection model built using YOLO is trained on the training dataset. The model parameters are updated through multiple iterations so that the model can better predict the target location and category, thus obtaining a detection model for defect detection.

[0091] Use the validation set to evaluate the model's effectiveness and ensure the most appropriate precision and recall for current defect detection.

[0092] The packaged model is used for actual pole defect detection to obtain the locations of bump defects and scratch defects.

[0093] S15-2, detecting notch-type defects in the detection area using a detection algorithm based on bandpass filtering and morphology.

[0094] Wherein, in step S15-1, the detection of notch-type defects in the detection area by using a detection algorithm based on bandpass filtering and morphology includes:

[0095] Obtain the location of the aluminum hole in the center of the pole and shield it to determine the first defect signal;

[0096] Determine a suitable cutoff frequency by using bandpass filtering to filter out high-frequency and low-frequency components of the first defect signal, thereby obtaining a second defect signal having a specific frequency range;

[0097] The second defect signal is subjected to dilation and erosion operations through morphology to obtain the position of the target defect.

[0098] In this embodiment, since there is an aluminum hole in the center of the pole, which is similar to a notch defect in that the grayscale values ​​are very close, this situation will cause interference when using a 2D grayscale image to detect notches. In order to avoid the need for post-processing to misjudge the aluminum hole as a notch when detecting defects, this embodiment finds the aluminum hole location before detection and then shields the aluminum hole area. Based on the depth map and the reference depth value of the pole, it can be determined that the depth value of the aluminum hole is lower than a certain value. The aluminum hole area is thus screened out, and the corresponding position in the 2D grayscale image is divided into regions. The average grayscale value of the pole area is selected to replace the grayscale value of the region. In this way, the region will not be judged as a defect when detecting defects. Specifically, by using a bandpass filter to select an appropriate cutoff frequency, the high-frequency and low-frequency components in the image signal are filtered out, and only the signal within a specific frequency range is retained; this can remove noise and interference while highlighting the defect signal, making the defect signal more obvious. The corresponding performance in the image is to highlight the defect area. Through morphological operations such as dilation and erosion, some interference is removed and possible adhesions between defects are disconnected to make the defects clearer. Then, the circumscribed rectangle of the obtained defect point set is calculated to obtain the location coordinates of the defects.

[0099] By applying a mean filter to the image using two filter kernels of different sizes, one large and one small, two filtered images are obtained. Subtracting the two images yields the same result as bandpass filtering (i.e., bandpass filtering, in the frequency domain, removes both high-frequency and low-frequency signals from the image). Noise and background are removed, leaving only the gap. Because pixel distribution cannot achieve the desired ideal state, the resulting gap may exhibit interference, including dark edges, gaps clumping together, and clumping noise around the gap. To address this issue, this embodiment utilizes morphological operations, including dilation, erosion, opening, and closing, to obtain the final gap outline.

[0100] Furthermore, the method further comprises:

[0101] Determine whether the defect size at the corresponding position in the defect detection result is greater than a detection threshold; if so, determine that it is a real pole defect.

[0102] Specifically, according to the defect position obtained above, the corresponding position is found in the depth map; the defect size of the corresponding defect (including defect length, width and depth) is calculated, and the three are compared with the corresponding thresholds of the detection parameters. When all three are greater than the corresponding thresholds, it is confirmed that the corresponding defect exists in the pole; otherwise, it is considered that the corresponding defect does not exist.

[0103] Specifically, the authenticity of the defect is determined by combining the concave and convex features of the defect in the depth map and the grayscale features in the 2D grayscale map. Since the positions of objects in the depth map and grayscale map are consistent, when determining whether the defect is real, we check whether the current area meets the characteristics of the defect in both the depth map and the grayscale map. The concave and convex features of the defect are manifested in the depth map as depth values ​​that are smaller or larger than the surrounding values, which can be used to determine which points are below or above the polar surface; the grayscale features of the defect are manifested in the grayscale map as pixel values ​​that are different from the surrounding pixels, usually with a smaller grayscale value. Only when the defect characteristics are present in both maps can it be determined to be a real defect. By calculating the size of the defect, the part that may contain welding dust (which is more similar to the defect in the grayscale map) in the actual production process can be further eliminated to obtain a more accurate true defect.

[0104] Based on the defect location determined above, a set of depth values ​​is calculated on the depth map, and the minimum value is taken as the depth value of the current defect. If the defect size and depth value both meet the user-defined defect specifications, the defect is considered to be a real one. This embodiment thus achieves accurate measurement of defect data and identification of real defects, improving defect recognition accuracy.

[0105] 2 is a schematic structural diagram of a device for detecting defects in a lithium battery pole provided by an embodiment of the present invention.

[0106] In this embodiment, the device 20 includes:

[0107] An acquisition unit 21 is configured to acquire a depth map and a 2D grayscale map of the pole, wherein the depth map and the 2D grayscale map have the same size;

[0108] A first processing unit 22 is configured to perform equalization processing on the depth map to obtain a first image;

[0109] A second processing unit 23 is configured to perform filtering and edge detection on the first image to obtain a polar column area;

[0110] A positioning unit 24 is configured to locate the detection position corresponding to the 2D grayscale image according to the edge contour coordinate points of the pole area to obtain a detection area;

[0111] The detection unit 25 is configured to detect different defect types in the detection area using a preset defect detection algorithm to obtain a defect detection result.

[0112] In another embodiment, the apparatus 20 further comprises:

[0113] The determination unit is used to determine whether the defect size at the corresponding position in the defect detection result is greater than a detection threshold, and if so, determine it as a real pole defect.

[0114] Each unit module of the device 20 can respectively execute the corresponding steps in the above method embodiment, so each unit module will not be described in detail here. Please refer to the description of the corresponding steps above for details.

[0115] An embodiment of the present invention also provides a device, which includes the lithium battery pole defect detection device as described above, wherein the lithium battery pole defect detection device can adopt the structure of the embodiment of Figure 2, and correspondingly, it can execute the technical solution of the method embodiment shown in Figure 1, and its implementation principle and technical effect are similar. For details, please refer to the relevant records in the above embodiments, which will not be repeated here.

[0116] The device includes: a mobile phone, digital camera, tablet computer, or other device with a camera function, or a device with an image processing function, or a device with an image display function. The device may include components such as a memory, a processor, an input unit, a display unit, and a power supply.

[0117] Among them, the memory can be used to store software programs and modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory may mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as an image playback function, etc.), etc.; the data storage area can store data created according to the use of the device, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory may also include a memory controller to provide the processor and the input unit with access to the memory.

[0118] The input unit can be used to receive input digital, character, or image information, and generate keyboard, mouse, joystick, optical, or trackball signal input related to user settings and function control. Specifically, the input unit of this embodiment can include not only a camera, but also a touch-sensitive surface (such as a touch display) and other input devices.

[0119] The display unit can be used to display information input by the user or information provided to the user and various graphical user interfaces of the device, which can be composed of graphics, text, icons, videos and any combination thereof. The display unit may include a display panel. Optionally, the display panel can be configured in the form of an LCD (Liquid Crystal Display), an OLED (Organic Light-Emitting Diode), etc. Furthermore, the touch-sensitive surface can cover the display panel. When the touch-sensitive surface detects a touch operation on or near it, it is transmitted to the processor to determine the type of touch event. The processor then provides a corresponding visual output on the display panel based on the type of touch event.

[0120] An embodiment of the present invention further provides a computer-readable storage medium. This computer-readable storage medium may be included in the memory described in the above embodiment, or may be a separate computer-readable storage medium not incorporated into a device. The computer-readable storage medium stores at least one instruction, which is loaded and executed by a processor to implement the lithium battery terminal defect detection method shown in FIG1 . The computer-readable storage medium may be a read-only memory, a magnetic disk, an optical disk, or the like.

[0121] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. For similar or identical parts between the various embodiments, reference can be made to each other. For the apparatus embodiments, device embodiments, and storage medium embodiments, since they are generally similar to the method embodiments, their descriptions are relatively simple. For relevant parts, reference can be made to the descriptions of the method embodiments.

[0122] Furthermore, in this document, the terms "comprises," "comprising," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0123] While the foregoing description shows and describes preferred embodiments of the present invention, it should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. Rather, the present invention can be used in various other combinations, modifications, and environments, and can be modified within the scope of the present invention by the teachings herein or by techniques or knowledge in the relevant art. Modifications and variations made by those skilled in the art without departing from the spirit and scope of the present invention are intended to be within the scope of the appended claims.

Claims

1. A method for detecting defects in lithium battery poles, characterized in that: The method comprises: Obtain a depth map and a 2D grayscale map of the pole, wherein the depth map and the 2D grayscale map have the same size; Performing equalization processing on the depth map by using a limited contrast adaptive histogram to obtain a first image; Performing filtering and edge detection on the first image to obtain a polar column area; comprising: Performing mean filtering on the first image to obtain a second image; Calculating the second image using a Sobel edge detection operator to obtain an initial polar region; Performing a closing operation on the initial pole region and calculating contour coordinates to determine a reference depth value of the pole; including: Performing a closing operation on the initial pole region to obtain a first region; Calculate the area of ​​the circumscribed rectangle of the first region and filter out the contour coordinates of the pole region; The outline coordinates are indented according to a preset distance to obtain a plurality of second areas; Counting a plurality of depth values ​​of the second area, and taking a median of the depth values ​​as the reference depth value; Selecting coordinate points in the first image that meet a preset range of the reference depth value, and taking a set of all the coordinate points as the pole region; Locate the detection position corresponding to the 2D grayscale image according to the edge contour coordinate points of the pole area to obtain the detection area; Performing detection of different defect types on the detection area by using a preset defect detection algorithm to obtain defect detection results; including: Detecting bump type defects and scratch type defects in the detection area by using a detection model built based on YOLO; The notch type defects in the detection area are detected by a detection algorithm based on bandpass filtering and morphology.

2. A method for detecting defects in a lithium battery pole according to claim 1, characterized in that: The method further comprises: It is determined whether the defect size at the corresponding position in the defect detection result is greater than the detection threshold, and if so, it is determined to be a real pole defect.

3. A method for detecting defects in a lithium battery pole according to claim 2, characterized in that: Find the corresponding position in the depth map according to the obtained defect position; Calculate the defect size of the corresponding defect, including the defect length, width and depth, and compare the length, width and depth with the corresponding threshold of the detection parameter. When all three are greater than the corresponding threshold, it is confirmed that the pole has the corresponding defect, otherwise, it is considered that the corresponding defect does not exist.

4. A method for detecting defects in a lithium battery pole according to claim 3, characterized in that: By combining the concave-convex features of the defect in the depth map and the grayscale features in the 2D grayscale map, it is determined whether the defect is real.

5. A method for detecting defects in a lithium battery pole according to claim 4, characterized in that: A set of depth values ​​is obtained on the depth map according to the obtained defect position, and the minimum value is taken as the depth value of the current defect. If the size and depth value of the defect are determined to meet the user-defined defect specifications, it is considered that a real defect exists.

6. The method for detecting a lithium battery pole defect according to claim 1, characterized in that: The step of calculating the second image using a Sobel edge detection operator to obtain an initial polar region includes: Calculating edge gradients in the X direction and the Y direction of the second image using an edge detection operator to obtain edge gradient information; Extracting edge information in the first image according to the edge gradient information to obtain a third image with edge gradient values; The third image is binarized according to a preset gradient threshold to obtain the initial polar column region.

7. The method for detecting a lithium battery pole defect according to claim 1, characterized in that: The detecting of the notch type defects in the detection area by using a detection algorithm based on bandpass filtering and morphology comprises: Obtain the position of the aluminum hole in the center of the pole and shield it to determine the first defect signal; Determine a suitable cutoff frequency by using bandpass filtering to filter out the high-frequency part and the low-frequency part of the first defect signal, so as to obtain a second defect signal with a specific frequency range; The second defect signal is expanded and eroded by morphology to obtain the position of the target defect.

8. A lithium battery pole defect detection device, characterized in that: The device comprises: An acquisition unit, used to acquire a depth map and a 2D grayscale map of the pole, wherein the depth map and the 2D grayscale map have the same size; A first processing unit, configured to perform equalization processing on the depth map by using a limited contrast adaptive histogram to obtain a first image; The second processing unit is used to perform filtering and edge detection on the first image to obtain a polar column area; including: A positioning unit, used for positioning the detection position corresponding to the 2D grayscale image according to the edge contour coordinate points of the pole area to obtain a detection area; The detection unit is used to detect different defect types in the detection area through a preset defect detection algorithm to obtain a defect detection result.

9. A lithium battery pole defect detection device, characterized in that: It comprises a processor, a memory and a computer program stored in the memory, wherein the computer program is executed by the processor to implement the steps of a method for detecting a lithium battery pole defect as described in any one of claims 1 to 7.

Images