Cylindrical battery cell sealing welding quality detection method, device, electronic device and storage medium

Abstract

The present invention provides a method, device, electronic device and storage medium for detecting the welding quality of cylindrical battery cells. The method includes: obtaining a circumferential image of the weld seam of the target end cap of the target cylindrical battery cell; performing target recognition on the circumferential image based on a target recognition model to determine the weld seam area in the circumferential image; and obtaining the welding quality detection result of the target cylindrical battery cell based on the weld seam area in the circumferential image. The method, device, electronic device and storage medium for detecting the welding quality of cylindrical battery cells provided by the embodiments of the present invention can improve the efficiency of detecting the welding quality of cylindrical battery cells, and the accuracy of the detection result is higher, thereby ensuring the stability of the quality of cylindrical battery cells by identifying the weld seam area in the circumferential image of the weld seam of the target end cap of the target cylindrical battery cell and obtaining the welding quality detection result of the target cylindrical battery cell based on the weld seam area in the circumferential image.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial control, and particularly to a method, device, electronic device and storage medium for detecting the sealing welding quality of cylindrical battery cells. Background Art

[0002] During the production process of batteries, it is necessary to weld the housing and the end cap of the cylindrical battery cell, which is called sealing welding.

[0003] Welding conditions such as missed welding, welding deviation and welding through will cause abnormal operation of the cylindrical battery cell and affect the quality of the battery. Therefore, it is necessary to detect the sealing welding quality of the cylindrical battery cell.

[0004] The existing detection of the sealing welding quality of cylindrical battery cells is usually carried out by manual inspection. Manual inspection has low efficiency, poor accuracy, and cannot ensure the stability of battery quality. Summary of the Invention

[0005] The present invention provides a method, device, electronic device and storage medium for detecting the sealing welding quality of cylindrical battery cells, so as to solve the technical problem of low detection efficiency in the prior art.

[0006] The present invention provides a method for detecting the sealing welding quality of cylindrical battery cells, including:

[0007] Obtaining a circumferential image of the weld seam of the target end cap of the target cylindrical battery cell;

[0008] Based on the target recognition model, performing target recognition on the circumferential image to determine the weld seam area in the circumferential image;

[0009] Based on the weld seam area in the circumferential image, obtaining the detection result of the sealing welding quality of the target cylindrical battery cell.

[0010] According to the method for detecting the sealing welding quality of cylindrical battery cells provided by the present invention, the obtaining of the circumferential image of the weld seam of the target end cap of the target cylindrical battery cell specifically includes:

[0011] Controlling the rotation of the target cylindrical battery cell, and obtaining the original image collected by the image acquisition sensor at the target position during the rotation of the target cylindrical battery cell;

[0012] Based on the original image, obtaining the circumferential image.

[0013] According to the method for detecting the sealing welding quality of cylindrical battery cells provided by the present invention, the obtaining of the circumferential image of the weld seam of the target end cap of the target cylindrical battery cell specifically includes:

[0014] Controlling the image acquisition sensor to rotate around the target cylindrical battery cell, and obtaining the original image collected by the image acquisition sensor during the rotation around the target cylindrical battery cell;

[0015] Based on the original image, obtain the circumferential image.

[0016] According to a cylindrical battery cell sealing quality detection method provided by the present invention, the circumferential image includes multiple images collected under different illumination conditions;

[0017] Correspondingly, obtaining the sealing quality detection result of the target cylindrical battery cell based on the weld region in the circumferential image specifically includes:

[0018] Based on the weld regions in each of the circumferential images, obtain a reflectivity image and a height map;

[0019] Perform defect detection based on the reflectivity map to obtain a first defect image, and perform defect detection based on the height map to obtain a second defect image;

[0020] Obtain the Hamming distance between the first defect image and the second defect image;

[0021] In the case where the Hamming distance is less than the distance threshold, determine the overlapping region between the first defect image and the second defect image as the defect region.

[0022] According to a cylindrical battery cell sealing quality detection method provided by the present invention, obtaining the sealing quality detection result of the target cylindrical battery cell based on the weld region in the circumferential image specifically includes:

[0023] Based on the weld region in the circumferential image, determine the size information and / or position information of the weld;

[0024] Based on the size information and / or position information of the weld, obtain the sealing quality detection result of the target cylindrical battery cell.

[0025] According to a cylindrical battery cell sealing quality detection method provided by the present invention, obtaining the sealing quality detection result of the target cylindrical battery cell based on the weld region in the circumferential image specifically includes:

[0026] Based on each pixel value of the weld region in the circumferential image, obtain the sealing quality detection result of the target cylindrical battery cell.

[0027] According to a cylindrical battery cell sealing quality detection method provided by the present invention, obtaining the sealing quality detection result of the target cylindrical battery cell based on the weld region in the circumferential image specifically includes:

[0028] Compare the number of weld regions with a preset target number, and obtain the sealing quality detection result of the target cylindrical battery cell according to the comparison result.

[0029] The present invention also provides a cylindrical battery cell sealing and welding quality detection device, including:

[0030] An image acquisition module, configured to acquire a circumferential image of a weld seam of a target end cap of a target cylindrical battery cell;

[0031] An image recognition module, configured to perform target recognition on the circumferential image based on a target recognition model to determine a weld seam area in the circumferential image;

[0032] A quality detection module, configured to obtain a sealing and welding quality detection result of the target cylindrical battery cell based on the weld seam area in the circumferential image.

[0033] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the cylindrical battery cell sealing and welding quality detection method as described in any one of the above are implemented.

[0034] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the cylindrical battery cell sealing and welding quality detection method as described in any one of the above are implemented.

[0035] The cylindrical battery cell sealing and welding quality detection method, device, electronic device, and storage medium provided by the present invention can improve the efficiency of cylindrical battery cell sealing and welding quality detection and the accuracy of detection results by identifying the weld seam area in the circumferential image of the weld seam of the target end cap of the target cylindrical battery cell, and based on the weld seam area in the circumferential image, thereby ensuring the stability of the quality of the cylindrical battery cell. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0037] Figure 1 is a flowchart of the cylindrical battery cell sealing and welding quality detection method provided by the present invention;

[0038] Figure 2 is a structural diagram of the cylindrical battery cell sealing and welding quality detection device provided by the present invention;

[0039] Figure 3 is a structural diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0041] In the description of the embodiments of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the embodiments of the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the embodiments of the present invention. In addition, the terms "first", "second", "third" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance and do not involve order.

[0042] In the description of the embodiments of the present invention, it should be noted that unless otherwise clearly defined and limited, the terms "installation", "connection", "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present invention can be understood according to specific situations.

[0043] To overcome the above problems of the prior art, the present invention provides a method, device, electronic device and storage medium for detecting the sealing welding quality of cylindrical battery cells. The inventive concept is to determine welding conditions such as welding omission, welding deviation and welding penetration based on the method of image recognition and analysis, and can well complete the detection of the sealing welding quality of cylindrical battery cells.

[0044] Figure 1 It is a schematic flowchart of a method for detecting the sealing welding quality of cylindrical battery cells provided by the present invention. The following combines Figure 1 to describe the method for detecting the sealing welding quality of cylindrical battery cells in the embodiments of the present invention. As Figure 1 shown, the method includes: Step 101, obtaining a circumferential image of the weld seam of the target end cap of the target cylindrical battery cell.

[0045] Specifically, after welding the target end cap to the target cylindrical battery cell, the circumferential image of the weld of the target end cap of the target cylindrical battery cell can be obtained based on the image or video of the target cylindrical battery cell collected by the image acquisition sensor.

[0046] The target end cap can be any one of the two end caps of the target cylindrical battery cell.

[0047] Under normal circumstances, the welding head seals the housing and the end cap of the cylindrical battery cell along a preset trajectory, forming a weld with a certain width and a certain length.

[0048] The image acquisition sensor can collect the images of each part of the weld by taking pictures of each part of the weld or collect a video by taking pictures of each part of the weld, and then perform screenshot processing on the video to obtain the images of each part of the weld.

[0049] After obtaining the images of each part of the weld, the images of each part of the weld can be stitched together to obtain the circumferential image of the weld of the target end cap of the target cylindrical battery cell.

[0050] It can be understood that the weld is a circle with a certain width, and the circumferential image of the weld of the target end cap of the target cylindrical battery cell can reflect the information of the complete weld (i.e., the complete information of the circle). In the circumferential image of the weld of the target end cap of the target cylindrical battery cell, the circle is expanded into a rectangle with this width.

[0051] Step 102: Based on the target recognition model, perform target recognition on the circumferential image to determine the weld area in the circumferential image.

[0052] Specifically, under normal circumstances, the welding head seals the housing and the end cap of the cylindrical battery cell along a preset trajectory, forming a weld with a certain width and a certain length.

[0053] Any target recognition method, such as background removal, template matching, or artificial neural network, etc., can be used to perform target recognition on the circumferential image to identify the weld area in the circumferential image.

[0054] For example, since the material of the weld is significantly different from the materials of the housing and the end cap of the cylindrical battery cell, the housing and the end cap of the cylindrical battery cell in the circumferential image can be used as the background for background removal through image processing to determine the weld area in the circumferential image.

[0055] Preferably, target recognition can be performed on the circumferential image based on the target recognition model. The circumferential image can be input into the target recognition model to obtain the weld area in the circumferential image output by the target recognition model.

[0056] The target recognition model can be a model established based on artificial intelligence algorithms such as artificial neural networks or decision trees. The target recognition model is used to recognize the weld area in the image.

[0057] It can be understood that before step 103, the target recognition model can be trained according to the sample image and the weld area in the pre-labeled sample image to obtain a trained target recognition model.

[0058] Step 103: Based on the weld area in the circumferential image, obtain the sealing quality detection result of the target cylindrical battery cell.

[0059] Specifically, after determining the weld area in the circumferential image, image analysis can be performed on the weld area in the circumferential image to map the features of the weld area in the circumferential image from the image space to the real space, and determine whether the actual value of the weld parameter is consistent with the set value of the weld parameter.

[0060] The weld parameters can include the width, length, position, and trajectory of the weld, etc.

[0061] If the actual value of any weld parameter is inconsistent with the set value of the parameter, when the detection result is a qualitative result, the welding quality detection result of the target soft-pack battery cell can be determined as unqualified.

[0062] If the actual value of each weld parameter is consistent with the set value of the parameter, when the detection result is a qualitative result, the welding quality detection result of the target soft-pack battery cell can be determined as qualified.

[0063] It should be noted that if the weld area in the circumferential image cannot be recognized through step 102, step 103 can be skipped, and the sealing quality detection result of the target cylindrical battery cell can be directly determined as unqualified, specifically, there is a missed weld, that is, the target cylindrical battery cell was not welded in the previous welding process.

[0064] In the embodiment of the present invention, by recognizing the weld area in the circumferential image of the weld of the target end cap of the target cylindrical battery cell and obtaining the sealing quality detection result of the target cylindrical battery cell based on the weld area in the circumferential image, the efficiency of the sealing quality detection of the cylindrical battery cell can be improved, and the accuracy of the detection result is higher, thereby ensuring the stability of the quality of the cylindrical battery cell.

[0065] Based on the content of any of the above embodiments, obtaining the circumferential image of the weld of the target end cap of the target cylindrical battery cell specifically includes: controlling the rotation of the target cylindrical battery cell and obtaining the original image collected by the image acquisition sensor located at the target position during the rotation of the target cylindrical battery cell.

[0066] Specifically, the original images can be collected by rotating the target cylindrical battery cell while keeping the image acquisition sensor stationary. The original images include the images of various parts of the weld seam.

[0067] During the rotation of the target cylindrical battery cell, the image acquisition sensor located at the target position takes pictures of the weld seam to collect multiple original images. The above-mentioned multiple original images can reflect the information of the complete weld seam.

[0068] The rotation of the target cylindrical battery cell can be carried out in a continuous or step-by-step manner. The rotation of the target cylindrical battery cell is carried out with the connection line between the centers of the two end caps of the target cylindrical battery cell as the rotation axis.

[0069] Based on the original images, circumferential images are obtained.

[0070] Specifically, based on any image stitching method, the original images can be subjected to image stitching processing to obtain the circumferential image of the weld seam of the target end cap of the target cylindrical battery cell.

[0071] In the embodiment of the present invention, the original images collected by the image acquisition sensor located at the target position during the rotation of the target cylindrical battery cell are used to obtain circumferential images based on the original images, so that the circumferential images of the weld seam of the target end cap of the complete target cylindrical battery cell can be obtained. Thus, the seal welding quality detection result of the target cylindrical battery cell can be obtained based on the circumferential images, the efficiency of the seal welding quality detection of the cylindrical battery cell can be improved, and the accuracy of the detection result is higher, thereby ensuring the stability of the quality of the cylindrical battery cell.

[0072] Based on the content of any of the above embodiments, to obtain the circumferential image of the weld seam of the target end cap of the target cylindrical battery cell, specifically, it includes: controlling the image acquisition sensor to rotate around the target cylindrical battery cell and obtaining the original images collected during the process of the image acquisition sensor rotating around the target cylindrical battery cell.

[0073] Specifically, the original images can be collected by keeping the target cylindrical battery cell stationary and rotating the image acquisition sensor around the target cylindrical battery cell. The original images include the images of various parts of the weld seam.

[0074] The rotation of the image acquisition sensor around the target cylindrical battery cell can be carried out in a continuous or step-by-step manner. The rotation of the image acquisition sensor around the target cylindrical battery cell is carried out with the connection line between the centers of the two end caps of the target cylindrical battery cell as the rotation axis.

[0075] During the process of the image acquisition sensor rotating around the target cylindrical battery cell, pictures of the weld seam can be taken at a preset time interval, or pictures of the weld seam can be taken at each preset position during the rotation to collect multiple original images. The above-mentioned multiple original images can reflect the information of the complete weld seam.

[0076] Based on the original images, obtain the circumferential images.

[0077] Specifically, based on any image stitching method, perform image stitching on each original image to obtain the circumferential image of the weld seam of the target end cap of the target cylindrical battery cell.

[0078] In the embodiment of the present invention, the original images are collected by rotating an image acquisition sensor around the target cylindrical battery cell. Based on the original images, the circumferential images are obtained, and a complete circumferential image of the weld seam of the target end cap of the target cylindrical battery cell can be obtained. Thus, the sealing quality detection result of the target cylindrical battery cell can be obtained based on the circumferential images, the efficiency of the sealing quality detection of the cylindrical battery cell can be improved, and the accuracy of the detection result is higher, thereby ensuring the stability of the quality of the cylindrical battery cell.

[0079] Based on the content of any of the above embodiments, the circumferential images include multiple images collected under different lighting conditions.

[0080] Specifically, multiple lighting conditions can be preset, and the circumferential images obtained by stitching the original images collected under each lighting condition are respectively obtained, so as to obtain multiple circumferential images, which is convenient for detecting the defective area in the case where the sealing quality detection result of the target cylindrical battery cell is unqualified.

[0081] The lighting conditions can include the angle and / or intensity of the light. Different lighting conditions can be formed based on multiple different light sources or by changing the position of a single light source.

[0082] Correspondingly, based on the weld seam area in the circumferential images, obtaining the sealing quality detection result of the target cylindrical battery cell specifically includes: obtaining a reflectivity image and a height map based on the weld seam area in each circumferential image.

[0083] Specifically, based on the photometric stereo method, reconstruction can be performed according to the weld seam area in each circumferential image to obtain the reflectivity map of the weld seam area.

[0084] The reflectivity refers to the percentage of the energy of the light reflected by an object to the energy of the incident light.

[0085] Still based on the photometric stereo method, according to the weld seam area in each circumferential image, the normal line of the surface of the weld seam area can be obtained. According to this normal line, the depth of the surface of the weld seam area can be determined, and thus the height map of the weld seam area can be obtained.

[0086] Perform defect detection based on the reflectivity map to obtain a first defect image, and perform defect detection based on the height map to obtain a second defect image.

[0087] Specifically, performing defect detection based on the reflectivity map can obtain a first defective area.

[0088] Specifically, a band-stop filter can be applied to the reflectivity map of the weld region to extract pixels with abnormal reflectivity, thereby obtaining a first defect region composed of pixels with abnormal reflectivity.

[0089] Abnormal reflectivity includes reflectivity greater than a preset upper reflectivity limit and reflectivity less than a preset lower reflectivity limit.

[0090] Both the upper reflectivity limit and the lower reflectivity limit can be set according to actual needs. The specific values of the upper reflectivity limit and the lower reflectivity limit are not specifically limited in the embodiments of the present invention.

[0091] Taking the first defect region as the foreground, the background of the reflectivity map is removed; after removing the background of the reflectivity map, binarization is performed to obtain a first defect image.

[0092] Defect detection is performed based on the height map to obtain a second defect region.

[0093] Specifically, a band-stop filter can be applied to the height map of the weld region to extract pixels with abnormal height, thereby obtaining a second defect region composed of pixels with abnormal height.

[0094] Abnormal height includes height greater than a preset upper height limit and height less than a preset lower height limit.

[0095] Both the upper height limit and the lower height limit can be set according to actual needs. The specific values of the upper height limit and the lower height limit are not specifically limited in the embodiments of the present invention.

[0096] Taking the second defect region as the foreground, the background of the height map is removed; after removing the background of the height map, binarization is performed to obtain a second defect image.

[0097] It can be understood that the sizes of the first defect image and the second defect image are the same.

[0098] Obtain the Hamming distance between the first defect image and the second defect image.

[0099] Specifically, the Hamming distance between the first defect image and the second defect image can be obtained.

[0100] The Hamming distance between two equal-length strings s1 and s2 is defined as the minimum number of replacements required to change one into the other. For example, the Hamming distance between the strings "1111" and "1001" is 2.

[0101] The Hamming distance between two images of the same size is defined as the number of pixel values of the pixel points at the same position in the two images that are different.

[0102] The calculation formula for the Hamming distance between the first defect image and the second defect image is as follows:

[0103]

[0104] Among them, H represents the Hamming distance between the first defect image and the second defect image; A represents the first defect image; B represents the second defect image; represents the exclusive OR operation.

[0105] The smaller the Hamming distance, the higher the similarity between the first defect image and the second defect image, and the higher the similarity between the first defect region and the second defect region; the larger the Hamming distance, the lower the similarity between the first defect image and the second defect image, and the lower the similarity between the first defect region and the second defect region.

[0106] When the Hamming distance is less than the distance threshold, the overlapping region between the first defect image and the second defect image is determined as the defect region.

[0107] Specifically, the size of the Hamming distance H and the distance threshold can be compared.

[0108] If the Hamming distance H is less than the distance threshold, it indicates that the first defect region and the second defect region are relatively close, and the results of the above two defect detections are close, both with high detection accuracy. Therefore, the overlapping part of the first defect region and the second defect region, and the overlapping region between the first defect image and the second defect image can be determined as the defect region, which is used as the final defect detection result.

[0109] The distance threshold can be set according to the actual situation such as the size of the image. For example, the distance threshold is 5 or 10. The specific value of the distance threshold is not specifically limited in the embodiments of the present invention.

[0110] It should be noted that when the distance is less than the distance threshold, the first defect region or the second defect region can also be directly used as the defect region (i.e., the final defect detection result).

[0111] When there is a defect region, the sealing quality detection result of the target cylindrical battery cell can be determined as unqualified, and the detected defect region is also output as the sealing quality detection result of the target cylindrical battery cell.

[0112] It should be noted that if the first defect region is not extracted based on the emissivity map and the second defect region is not extracted based on the height map, the sealing quality detection result of the target cylindrical battery cell can be determined as qualified.

[0113] In an embodiment of the present invention, by obtaining multiple circumferential images collected under different lighting conditions, extracting the weld regions in each circumferential image, obtaining a reflectivity map and a height map based on the weld regions in each circumferential image, performing defect detection based on the reflectivity image and the height map respectively, and when the similarity of the detection results obtained by the two defect detection methods is greater than the similarity threshold, determining the overlapping part of the defect regions obtained by the two defect detection methods as the finally output defect region, a more accurate defect region can be obtained.

[0114] Based on the content of any of the above embodiments, obtaining the sealing quality detection result of the target cylindrical battery cell based on the weld region in the circumferential image specifically includes: determining the size information and / or position information of the weld based on the weld region in the circumferential image.

[0115] Specifically, for each weld region in the circumferential image, the size information and / or position information of the weld region in the circumferential image can be obtained.

[0116] The size information may include length, width, area, etc.

[0117] The position information may be the trajectory information of the weld region. The trajectory information can be represented by the coordinates of each point on the trajectory relative to a preset feature point on the target cylindrical battery cell in the image.

[0118] Based on the imaging parameters of the original image (such as focal length and object distance, etc.), the size information and position information of each weld region in the circumferential image can be mapped from the image space to the real space to obtain the size information and position information of the weld in the real space.

[0119] For any weld region, the calculation formulas for the two pixel distances of the length and width of the weld region in the circumferential image can be expressed as:

[0120]

[0121]

[0122] where, pixel_distance 1 represents the length of the weld region in the circumferential image, and the coordinates of the two end pixel points in the length direction of the weld region are respectively (X 1 , Y 1 ) and (X 2 , Y 2 ); pixel_distance 2 represents the width of the weld region in the circumferential image, and the coordinates of the two end pixel points in the width direction of the weld region are respectively (X 3 , Y 3 ) and (X 4 , Y4 )。

[0123] The calculation formulas for the length and width of the weld in the real space can be expressed as:

[0124] L = k 1 *pixel_distance 1

[0125] W = k 1 *pixel_distance 2

[0126] Wherein, L represents the length of the weld in the real space; W represents the width of the weld in the real space; k 1 represents the first conversion coefficient. The first conversion coefficient k 1 , which is used for the conversion between the distance in the circumferential image and the distance in the real space.

[0127] For any weld area, the calculation formula for the area of the weld in the real space is

[0128] S = k 2 *Sum

[0129] Wherein, S represents the area of the weld in the real space; Sum represents the number of pixel points included in the corresponding weld area in the circumferential image; k 2 represents the second conversion coefficient. The second conversion coefficient k 2 , which is used for the conversion between the area in the circumferential image and the area in the real space.

[0130] Based on the size information and / or position information of the weld, obtain the sealing quality detection result of the target cylindrical battery cell.

[0131] Specifically, the size information and position information of the weld in the above real space are respectively the actual values of the size and position of the weld.

[0132] The size information of the weld can be compared with the preset value of the size of the weld to determine whether the actual value of the size of the weld is consistent with the preset value.

[0133] The position information of the weld can be compared with the set value of the position of the weld to determine whether the actual value of the position of the weld is consistent with the preset value.

[0134] It can be understood that when only the dimensional information of the weld seam is determined, the sealing welding quality inspection result of the target cylindrical battery cell can be obtained only based on the dimensional information of the weld seam; when only the position information of the weld seam is determined, the sealing welding quality inspection result of the target cylindrical battery cell can be obtained only based on the position information of the weld seam; when the dimensional information and the position information of the weld seam are determined, the sealing welding quality inspection result of the target cylindrical battery cell can be obtained based on the dimensional information and the position information of the weld seam.

[0135] When only the dimensional information of the weld seam is determined, if the actual value of the dimension of the weld seam is consistent with the preset value, the welding quality inspection result of the target soft-pack battery cell can be determined to be qualified; if the actual value of the dimension of the weld seam is inconsistent with the preset value, the welding quality inspection result of the target soft-pack battery cell can be determined to be unqualified.

[0136] When only the position information of the weld seam is determined, if the actual value of the position of the weld seam is consistent with the preset value, the welding quality inspection result of the target soft-pack battery cell can be determined to be qualified; if the actual value of the position of the weld seam is inconsistent with the preset value, the welding quality inspection result of the target soft-pack battery cell can be determined to be unqualified, and it can be further determined that there is a welding deviation.

[0137] When the dimensional information and the position information of the weld seam are determined, if the actual value of the dimension of the weld seam is consistent with the preset value and the actual value of the position of the weld seam is consistent with the preset value, the welding quality inspection result of the target soft-pack battery cell can be determined to be qualified; if the actual value of the dimension of the weld seam is inconsistent with the preset value, or the actual value of the position of the weld seam is inconsistent with the preset value, the welding quality inspection result of the target soft-pack battery cell can be determined to be unqualified.

[0138] Based on the weld seam area in the circumferential image, the embodiments of the present invention determine the dimensional information and / or the position information of the weld seam, and obtain the sealing welding quality inspection result of the target cylindrical battery cell based on the dimensional information and / or the position information of the actual weld seam, which can improve the efficiency of the sealing welding quality inspection of the cylindrical battery cell, and the accuracy of the inspection result is higher, so as to ensure the stability of the quality of the cylindrical battery cell.

[0139] Based on the content of any of the above embodiments, obtaining the sealing welding quality inspection result of the target cylindrical battery cell based on the weld seam area in the circumferential image specifically includes: obtaining the sealing welding quality inspection result of the target cylindrical battery cell based on the pixel values of each pixel in the weld seam area in the circumferential image.

[0140] Specifically, the pixel values of each pixel in the weld seam area in the circumferential image refer to the pixel values of each pixel included in the weld seam area in the circumferential image.

[0141] By comparing the pixel values of each pixel in the weld seam area in the circumferential image, it can be determined whether there are pixels in each pixel included in the weld seam area in the circumferential image whose pixel values are significantly different from those of other pixels.

[0142] Under normal circumstances, the welding seam flow of the welding seam head is constant, so the pixel values of each pixel in the welding seam area in the circumferential image are close. If there is an obvious difference between the pixel value of a certain pixel and other pixels among the pixels included in the welding seam area, it indicates that there may be a bright spot or a dark spot in the welding seam at a certain position in the real space corresponding to this pixel.

[0143] If among the pixels included in the welding seam area, the pixel value of a certain pixel is significantly greater than the pixel values of other pixels, this pixel is a bright spot, indicating that there is a pinhole in the welding seam at a certain position in the real space corresponding to this pixel; if among the pixels included in the welding seam area, the pixel value of a certain pixel is significantly less than the pixel values of other pixels, this pixel is a dark spot, indicating that there is a phenomenon of burn-through in the welding seam at a certain position in the real space corresponding to this pixel.

[0144] Based on the pixel values of each pixel in the welding seam area in the circumferential image in the embodiments of the present invention, the sealing welding quality detection result of the target cylindrical battery cell is obtained, which can more quickly, efficiently and accurately determine whether there is a pinhole or burn-through, improve the efficiency of the sealing welding quality detection of the cylindrical battery cell, and the accuracy of the detection result is higher, so as to ensure the stability of the quality of the cylindrical battery cell.

[0145] Based on the content of any of the above embodiments, based on the welding seam area in the circumferential image, the sealing welding quality detection result of the target cylindrical battery cell is obtained, specifically including: comparing the number of welding seam areas with a preset target number, and obtaining the sealing welding quality detection result of the target cylindrical battery cell according to the comparison result.

[0146] Specifically, missed welding means that a certain position should be welded but is actually not welded.

[0147] Welding deviation means that a certain position should not be welded but is actually welded.

[0148] Burn-through means that the welding energy at a certain position is too large and penetrates the metal shell.

[0149] After determining the welding seam area in the circumferential image, the number of welding seam areas can be determined.

[0150] The target number can be a value set in advance. The target number is the number of times the welding seam head welds on a certain surface of the cylindrical battery cell.

[0151] The target number can be set according to the actual situation. For the specific value of the target number, the embodiments of the present invention do not make specific limitations.

[0152] Preferably, the target number is 1.

[0153] Compare the number of welding seam areas with the preset target number.

[0154] If the number of weld regions is greater than the target number, it indicates that the number of welds formed on a certain surface of the target cylindrical battery cell is greater than the preset number of welding times on that surface. This means that in a certain weld, multiple welds were formed instead of a single weld, and it can be determined that there is a missed weld.

[0155] If the number of weld regions is less than the target number, it indicates that the number of welds formed on a certain surface of the target cylindrical battery cell is greater than the preset number of welding times on that surface. This means that the welds formed in one instance are connected to the welds formed in another instance to form a single weld (i.e., welding deviation), or that no weld was formed in a certain weld (i.e., missed weld), and it can be determined that there is a welding deviation or a missed weld.

[0156] If the number of weld regions is equal to the target number, it indicates that the number of welds formed on a certain surface of the target cylindrical battery cell is equal to the preset number of welding times on that surface. It is temporarily not possible to determine whether there is a missed weld or a welding deviation.

[0157] It should be noted that based on at least two of the size information and / or position information of the weld, the pixel values of each pixel in the weld region in the circumferential image, and the number of weld regions, after respectively obtaining the preliminary results of the sealed welding quality detection, the preliminary results can be fused to obtain the final sealed welding quality detection result.

[0158] For example, welding through, pinholes, missed welds, welding deviations, etc. can be classified as a single detection index, and a full score is assigned to each detection index. By determining the actual score of each detection index as the preliminary result and obtaining the sum of the actual scores as the final sealed welding quality detection result.

[0159] It should be noted that the actual score of the detection index can be determined based on whether the phenomenon corresponding to the detection index exists, or based on the degree of the phenomenon corresponding to the detection index (e.g., the ratio of welding deviation).

[0160] In the embodiments of the present invention, by comparing the number of weld regions with the preset target number to obtain the sealed welding quality detection result of the target cylindrical battery cell, it is possible to more quickly, efficiently, and accurately determine whether there is a missed weld or a welding deviation, improve the efficiency of the sealed welding quality detection of the cylindrical battery cell, and the accuracy of the detection result is higher, thereby ensuring the stability of the quality of the cylindrical battery cell.

[0161] Based on the content of any of the above embodiments, based on the size information and / or position information of the weld, obtaining the sealed welding quality detection result of the target cylindrical battery cell specifically includes: comparing the length and width in the size information of the weld with the preset target length and target width respectively, and obtaining the sealed welding quality detection result of the target cylindrical battery cell according to the comparison result; and / or comparing the position information of the weld with the preset weld position, and obtaining the sealed welding quality detection result of the target cylindrical battery cell according to the comparison result.

[0162] Specifically, the length and width in the dimensional information of the weld seam are the actual length of the weld seam and the actual width at each location.

[0163] For each weld seam, the actual length of the weld seam can be compared with the target length, and the actual width at each location of the weld seam can be compared with the target width at each location.

[0164] The target length refers to the theoretical value (i.e., the preset value) of the length of the weld seam formed by each welding. The target length can be determined based on the weld seam flow rate, weld seam duration, etc. of this weld seam.

[0165] The target width refers to the theoretical value (i.e., the preset value) of the width at a certain position of the weld seam formed by each welding. The target width can be determined based on the weld seam flow rate, etc. of this weld seam.

[0166] If the actual length of any weld seam does not match the target length, it indicates that the length in the dimensional information of the weld seam does not match the target length, and the sealing weld quality inspection result can be determined as unqualified.

[0167] If the actual width at any location of any weld seam does not match the target width, it indicates that the length in the dimensional information of the weld seam does not match the target width, and the sealing weld quality inspection result can be determined as unqualified.

[0168] If the actual length of each weld seam matches the target length, and the actual width at each location of each weld seam matches the target width, it indicates that the length and width in the dimensional information of the weld seam match the preset target length and target width respectively, and the sealing weld quality inspection result can be determined as qualified.

[0169] If the actual length of any weld seam is greater than the target length, it can be determined that there is welding deviation.

[0170] If the actual length of any weld seam is less than the target length, it can be determined that there is missed welding.

[0171] If the actual width at any location of any weld seam is greater than the target width, it can be determined that there is welding deviation.

[0172] If the actual width (not zero) at any location of any weld seam is less than the target width, it can be determined that there is missed welding.

[0173] The preset weld seam position represents the theoretical area covered by the weld seam, that is, the set value of the weld seam position.

[0174] The position information of the weld seam represents the actual area covered by the weld seam, that is, the actual value of the weld seam position.

[0175] Compare the position information of the weld seam with the preset weld seam position, that is, compare the actual area covered by the weld seam with the theoretical area.

[0176] If they are consistent, the sealing welding quality inspection result can be determined as qualified; if they are consistent, the sealing welding quality inspection result can be determined as unqualified.

[0177] If there is a point in the actual area covered by the weld seam but not in the theoretical area, it can be determined that there is a welding deviation.

[0178] If there is a point that does not exist in the actual area covered by the weld seam but exists in the theoretical area, it can be determined that there is a missed weld.

[0179] In the embodiment of the present invention, by comparing the length and width in the size information of the weld seam with the preset target length and target width respectively, and / or comparing the position information of the weld seam with the preset weld seam position, the sealing welding quality inspection result of the target cylindrical battery cell is obtained, and it can more quickly, efficiently and accurately determine whether there is a missed weld or a welding deviation, improve the efficiency of the sealing welding quality inspection of the cylindrical battery cell, and the accuracy of the inspection result is higher, so as to ensure the stability of the quality of the cylindrical battery cell.

[0180] Based on the content of any one of the above embodiments, based on the pixel values of the weld seam area in the circumferential image, the sealing welding quality inspection result of the target cylindrical battery cell is obtained, specifically including: comparing the pixel values of the weld seam area in the circumferential image with the target pixel value interval, and obtaining the sealing welding quality inspection result of the target cylindrical battery cell according to the comparison result.

[0181] Specifically, for each pixel in the weld seam area of the circumferential image, it can be determined whether the pixel value of the pixel falls into the target pixel value interval by comparing the pixel value of the pixel with the target pixel value interval.

[0182] If any pixel value does not fall into the target pixel value interval, it means that the pixel value of this pixel has an obvious difference from other pixels, and there may be a pinhole or a weld-through at a certain position in the real space corresponding to this pixel.

[0183] If each pixel value falls into the target pixel value interval, it means that the pixel value of any pixel has no obvious difference from other pixels, and the weld thickness at a certain position in the real space corresponding to each pixel is basically the same as that at other positions, and it can be determined that there is no possible pinhole or weld-through.

[0184] The target pixel value interval can be a preset target pixel value interval determined in advance, or can be determined according to the pixel values of the weld seam area in the circumferential image.

[0185] Preferably, the target pixel value interval can be determined through the following steps:

[0186] The average value and standard deviation of the pixel values of the weld seam area in the circumferential image can be obtained;

[0187] Based on the average value, subtract and add the product of the standard deviation and a preset coefficient respectively to determine the lower and upper limits of the target pixel value range, thereby determining the target pixel value range.

[0188] The preset coefficient is a positive number and can be set according to the actual situation. The specific value of this coefficient is not specifically limited in the embodiments of the present invention.

[0189] In the embodiments of the present invention, by comparing each pixel value in the weld area of the circumferential image with the target pixel value range, the seal welding quality detection result of the target cylindrical battery cell is obtained, which can more quickly, efficiently, and accurately determine whether there are pinholes or weld-throughs, improve the efficiency of the seal welding quality detection of the cylindrical battery cell, and the accuracy of the detection result is higher, thereby ensuring the stability of the quality of the cylindrical battery cell.

[0190] Next, the seal welding quality detection device provided by the present invention will be described. The seal welding quality detection device described below can be mutually corresponding and referred to with the seal welding quality detection method described above.

[0191] Figure 2 It is a schematic structural diagram of the seal welding quality detection device provided by the embodiments of the present invention. Based on the content of any of the above embodiments, as Figure 2 shown, the device includes an image acquisition module 201, an image recognition module 202, and a quality detection module 203, where:

[0192] The image acquisition module 201 is configured to acquire a circumferential image of the weld of the target end cap of the target cylindrical battery cell;

[0193] The image recognition module 202 is configured to perform target recognition on the circumferential image based on a target recognition model to determine the weld area in the circumferential image;

[0194] The quality detection module 203 is configured to obtain the seal welding quality detection result of the target cylindrical battery cell based on the weld area in the circumferential image.

[0195] Specifically, the image acquisition module 201, the image recognition module 202, and the quality detection module 203 are electrically connected in sequence.

[0196] The image acquisition module 201 can acquire a circumferential image of the weld of the target end cap of the target cylindrical battery cell based on the image or video of the target cylindrical battery cell collected by an image acquisition sensor.

[0197] The image recognition module 202 can perform target recognition on the circumferential image based on any target recognition method based on a target recognition model, such as background removal, template matching, or artificial neural network, etc., to identify the weld area in the circumferential image.

[0198] The quality inspection module 203 can perform image analysis on the weld area in the circumferential image, map the features of the weld area in the circumferential image from the image space to the real space, and determine whether the actual value of the weld parameter is consistent with the set value of the weld parameter.

[0199] The parameters of the weld can include the width, length, position, and trajectory of the weld, etc.

[0200] If the actual value of any parameter of the weld is inconsistent with the set value of the parameter, in the case where the detection result is a qualitative result, the welding quality detection result of the target soft-pack battery cell can be determined to be unqualified.

[0201] If the actual value of each parameter of the weld is consistent with the set value of the parameter, in the case where the detection result is a qualitative result, the welding quality detection result of the target soft-pack battery cell can be determined to be qualified.

[0202] Optionally, the image acquisition module 201 includes:

[0203] A first control unit for controlling the rotation of the target cylindrical battery cell and acquiring the original image collected by the image acquisition sensor located at the target position during the rotation of the target cylindrical battery cell;

[0204] An image stitching unit for acquiring a circumferential image based on the original image.

[0205] Optionally, the image acquisition module 201 includes:

[0206] A second control unit for controlling the rotation of the image acquisition sensor around the target cylindrical battery cell and acquiring the original image collected during the rotation of the image acquisition sensor around the target cylindrical battery cell;

[0207] An image stitching unit for acquiring a circumferential image based on the original image.

[0208] Optionally, the circumferential image includes multiple images collected under different lighting conditions;

[0209] Correspondingly, the quality inspection module 203 can specifically be used for:

[0210] Based on the weld area in each circumferential image, acquiring a reflectivity image and a height map;

[0211] Performing defect detection based on the reflectivity map to obtain a first defect image, and performing defect detection based on the height map to obtain a second defect image;

[0212] Obtaining the Hamming distance between the first defect image and the second defect image;

[0213] When the Hamming distance is less than the distance threshold, the overlapping area between the first defective image and the second defective image is determined as the defective area.

[0214] Optionally, the quality detection module 203 includes:

[0215] A size acquisition sub-module for determining the size information and / or position information of the weld seam based on the weld seam area in the circumferential image;

[0216] A first detection sub-module for obtaining the hermetic welding quality detection result of the target cylindrical battery cell based on the size information and / or position information of the weld seam.

[0217] Optionally, the quality detection module 203 includes:

[0218] A second detection sub-module for obtaining the hermetic welding quality detection result of the target cylindrical battery cell based on the pixel values of each pixel in the weld seam area in the circumferential image.

[0219] Optionally, the quality detection module 203 includes:

[0220] A third detection sub-module for comparing the number of weld seam areas with a preset target number and obtaining the hermetic welding quality detection result of the target cylindrical battery cell according to the comparison result.

[0221] Optionally, the first detection sub-module may include: a first detection unit and / or a second detection unit;

[0222] A first detection unit for comparing the length and width in the size information of the weld seam with a preset target length and target width respectively, and obtaining the hermetic welding quality detection result of the target cylindrical battery cell according to the comparison result;

[0223] A second detection unit for comparing the position information of the weld seam with a preset weld seam position and obtaining the hermetic welding quality detection result of the target cylindrical battery cell according to the comparison result.

[0224] Optionally, the second detection sub-module may specifically be used to compare the pixel values of each pixel in the weld seam area in the circumferential image with a target pixel value range, and obtain the hermetic welding quality detection result of the target cylindrical battery cell according to the comparison result.

[0225] The hermetic welding quality detection device for cylindrical battery cells provided by the embodiments of the present invention is used to execute the above-mentioned hermetic welding quality detection method for cylindrical battery cells of the present invention. Its implementation manner is consistent with the implementation manner of the hermetic welding quality detection method for cylindrical battery cells provided by the present invention, and the same beneficial effects can be achieved, which will not be elaborated here.

[0226] The cylindrical battery cell sealing quality detection device is used for the cylindrical battery cell sealing quality detection method of each of the foregoing embodiments. Therefore, the descriptions and definitions in the cylindrical battery cell sealing quality detection method in each of the foregoing embodiments can be used for the understanding of each execution module in the embodiments of the present invention.

[0227] In the embodiment of the present invention, by identifying the weld region in the circumferential image of the weld of the target end cap of the target cylindrical battery cell, and based on the weld region in the circumferential image, obtaining the sealing quality detection result of the target cylindrical battery cell, the efficiency of the cylindrical battery cell sealing quality detection can be improved, and the accuracy of the detection result is higher, so as to ensure the stability of the cylindrical battery cell quality.

[0228] Figure 3 An example of a schematic physical structure diagram of an electronic device is shown as Figure 3 As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330, and a communication bus 340. Among them, the processor 310, the communication interface 320, and the memory 330 complete mutual communication through the communication bus 340. The processor 310 can call the logical instructions stored in the memory 330 and executable on the processor 310 to execute the cylindrical battery cell sealing quality detection method provided in each of the foregoing method embodiments. The method includes: obtaining a circumferential image of the weld of the target end cap of the target cylindrical battery cell; performing target recognition on the circumferential image based on a target recognition model to determine the weld region in the circumferential image; and obtaining the sealing quality detection result of the target cylindrical battery cell based on the weld region in the circumferential image.

[0229] In addition, when the logical instructions in the foregoing memory 330 are implemented in the form of software function units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc that can store program codes.

[0230] The processor 310 in the electronic device provided by an embodiment of the present invention can call the logical instructions in the memory 330. Its implementation manner is consistent with the implementation manner of the cylindrical battery cell sealing welding quality detection method provided by the present invention, and the same beneficial effects can be achieved, so details are not described herein again.

[0231] On the other hand, an embodiment of the present invention further provides a computer program product. The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the cylindrical battery cell sealing welding quality detection method provided by each of the above method embodiments. The method includes: obtaining a circumferential image of a weld seam of a target end cap of a target cylindrical battery cell; performing target recognition on the circumferential image based on a target recognition model to determine a weld seam region in the circumferential image; and obtaining a sealing welding quality detection result of the target cylindrical battery cell based on the weld seam region in the circumferential image.

[0232] When the computer program product provided by an embodiment of the present invention is executed, the above-mentioned cylindrical battery cell sealing welding quality detection method is implemented. Its specific implementation manner is consistent with the implementation manner described in the embodiments of the foregoing method, and the same beneficial effects can be achieved, so details are not described herein again.

[0233] On another aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is configured to execute the cylindrical battery cell sealing welding quality detection method provided by each of the above embodiments. The method includes: obtaining a circumferential image of a weld seam of a target end cap of a target cylindrical battery cell; performing target recognition on the circumferential image based on a target recognition model to determine a weld seam region in the circumferential image; and obtaining a sealing welding quality detection result of the target cylindrical battery cell based on the weld seam region in the circumferential image.

[0234] When the computer program stored on the non-transitory computer-readable storage medium provided by an embodiment of the present invention is executed, the above-mentioned cylindrical battery cell sealing welding quality detection method is implemented. Its specific implementation manner is consistent with the implementation manner described in the embodiments of the foregoing method, and the same beneficial effects can be achieved, so details are not described herein again.

[0235] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative labor.

[0236] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0237] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting the welding quality of cylindrical battery cells, characterized in that, it includes: Obtaining a circumferential image of the weld of the target end cap of the target cylindrical battery cell; Based on the target recognition model, performing target recognition on the circumferential image to determine the weld area in the circumferential image; Based on the weld area in the circumferential image, obtaining the welding quality detection result of the target cylindrical battery cell; Wherein, the circumferential image includes multiple images collected under different lighting conditions; Correspondingly, the obtaining the welding quality detection result of the target cylindrical battery cell based on the weld area in the circumferential image specifically includes: Based on the weld areas in each of the circumferential images, obtaining a reflectivity image and a height map; Performing defect detection based on the reflectivity map to obtain a first defect image, and performing defect detection based on the height map to obtain a second defect image; Obtaining the Hamming distance between the first defect image and the second defect image; In the case where the Hamming distance is less than the distance threshold, determining the overlapping area between the first defect image and the second defect image as the defect area.

2. The method for detecting the welding quality of cylindrical battery cells according to claim 1, characterized in that, The obtaining a circumferential image of the weld of the target end cap of the target cylindrical battery cell specifically includes: Controlling the rotation of the target cylindrical battery cell, and obtaining the original images collected by the image acquisition sensor located at the target position during the rotation of the target cylindrical battery cell; Based on the original images, obtaining the circumferential image.

3. The method for detecting the welding quality of cylindrical battery cells according to claim 1, characterized in that, The obtaining a circumferential image of the weld of the target end cap of the target cylindrical battery cell specifically includes: Controlling the image acquisition sensor to rotate around the target cylindrical battery cell, and obtaining the original images collected during the rotation of the image acquisition sensor around the target cylindrical battery cell; Based on the original images, obtaining the circumferential image.

4. The method for detecting the welding quality of cylindrical battery cells according to claim 1, characterized in that, The obtaining the welding quality detection result of the target cylindrical battery cell based on the weld area in the circumferential image specifically includes: Based on the weld area in the circumferential image, determining the size information and / or position information of the weld; Based on the size information and / or position information of the weld, obtaining the welding quality detection result of the target cylindrical battery cell.

5. The method for detecting the welding quality of cylindrical battery cells according to claim 1, characterized in that, The obtaining the welding quality detection result of the target cylindrical battery cell based on the weld area in the circumferential image specifically includes: Based on the pixel values of the weld area in the circumferential image, obtaining the welding quality detection result of the target cylindrical battery cell.

6. The method for detecting the welding quality of cylindrical battery cells according to claim 1, characterized in that, The obtaining the welding quality detection result of the target cylindrical battery cell based on the weld area in the circumferential image specifically includes: Comparing the number of the weld areas with a preset target number, and obtaining the welding quality detection result of the target cylindrical battery cell according to the comparison result.

7. A cylindrical battery cell sealing and welding quality detection device, characterized in that, it includes: an image acquisition module for acquiring a circumferential image of the weld of the target end cap of the target cylindrical battery cell; an image recognition module for performing target recognition on the circumferential image based on a target recognition model to determine the weld area in the circumferential image; a quality detection module for obtaining a sealing and welding quality detection result of the target cylindrical battery cell based on the weld area in the circumferential image; wherein the circumferential image includes multiple images collected under different lighting conditions; correspondingly, the obtaining of the sealing and welding quality detection result of the target cylindrical battery cell based on the weld area in the circumferential image specifically includes: obtaining a reflectivity image and a height map based on the weld area in each circumferential image; performing defect detection based on the reflectivity map to obtain a first defect image, and performing defect detection based on the height map to obtain a second defect image; obtaining the Hamming distance between the first defect image and the second defect image; in the case where the Hamming distance is less than a distance threshold, determining the overlapping area between the first defect image and the second defect image as a defect area.

8. An electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, when the processor executes the program, it implements the steps of the cylindrical battery cell sealing and welding quality detection method according to any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium, on which a computer program is stored, characterized in that, when the computer program is executed by a processor, it implements the steps of the cylindrical battery cell sealing and welding quality detection method according to any one of claims 1 to 6.

Images