A battery pack blue film defect detection method, device and electronic equipment

Abstract

The application provides a battery pack blue film defect detection method and device and electronic equipment, including: for each blue film surface of the battery pack, the following processing is performed: extracting a corresponding to-be-detected region image on the depth image of the blue film surface; performing point cloud conversion on the to-be-detected region image to obtain a to-be-processed point cloud image; using a point cloud gradient filtering algorithm to calculate a defect region of the to-be-processed point cloud image to obtain at least one defect region corresponding to the blue film surface and size information corresponding to each defect region; respectively performing defect analysis on the at least one defect region based on a defect detection standard and the size information corresponding to each defect region to determine a defect detection result corresponding to the blue film surface. The application detects defects of the battery pack blue film through a 3D point cloud processing technology, improves detection accuracy, and ensures detection stability.

Description

Technical Field

[0001] This application relates to the field of lithium battery technology, and in particular to a method, apparatus and electronic device for detecting defects in the blue film of a battery pack. Background Technology

[0002] Square aluminum-cased batteries are coated with a blue film during the production process to provide insulation, dustproofing, and waterproofing. However, some batteries develop defects such as bulges, dents, bubbles, and damage to the explosion-proof valve due to bumps and scratches. These defects greatly affect the safety and quality of the batteries. Therefore, the development of automatic detection technology for blue film defects has significant engineering and commercial application value.

[0003] Currently, the detection of defects in blue film coatings relies entirely on manual visual inspection, which wastes human resources. Manual inspection is inefficient and prone to missed detections and over-detection due to visual fatigue and subjective judgment. Existing technologies typically use 2D or 2.5D vision solutions and visual algorithms for detection. However, due to the reflectivity of the blue film surface, it is impossible to obtain images with uniform lighting, resulting in a high false detection rate and unstable detection of traditional algorithms. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide at least one method, device and electronic device for defect detection of the blue film of a battery pack, which uses 3D point cloud processing technology to detect defects in the blue film of the battery pack, thereby improving detection accuracy and ensuring detection stability.

[0005] This application mainly includes the following aspects:

[0006] In a first aspect, embodiments of this application provide a method for detecting defects in the blue film of a battery pack, the method comprising:

[0007] A 3D scan is performed on the blue film surface of the battery pack to obtain a brightness image and a depth image corresponding to each blue film surface. The brightness image indicates the brightness value corresponding to each scan point on the blue film surface, and the depth image indicates the depth value from each scan point on the blue film surface to the viewpoint of the scanning device. For each blue film surface of the battery pack, the following processing is performed: the brightness image corresponding to the blue film surface is located and matched using a preset ROI template, and the target region of interest is extracted within the brightness image; using the coordinate information of the target region of interest, the corresponding depth map of the area to be detected is extracted from the depth image of the blue film surface; the depth map of the area to be detected is converted into a point cloud to obtain a point cloud image to be processed, which includes multiple feature points; a point cloud gradient filtering algorithm is used to calculate the defect region of the point cloud image to be processed, obtaining at least one defect region corresponding to the blue film surface and the size information corresponding to each defect region; based on the defect detection standard and the size information corresponding to each defect region, defect analysis is performed on at least one defect region to determine the defect detection result corresponding to the blue film surface.

[0008] In one possible implementation, the target region of interest (ROI) within the brightness image of each blue film surface is determined as follows: starting from the initial matching scan point within the brightness image of the blue film surface, the brightness image is traversed according to a preset step size; for each matching scan point, the matching degree between the matching region formed by the matching scan point and the preset ROI template is calculated, and the size of the matching region is the same as the size of the region of the preset ROI template; the matching region formed by the matching scan point corresponding to the highest matching degree is determined as the target ROI.

[0009] In one possible implementation, the matching degree between each image point and each scan point is determined by the following formula:

[0010]

[0011] R(x, y) represents the calculated matching value between the matching region formed by the matching scan point (x+x′, y+y′) in the brightness image of the blue film surface and the preset ROI template. The larger the calculated matching value, the greater the matching degree.

[0012] In this formula, T(x′, y′) represents the brightness value at position (x′, y′) in the preset ROI template, ∑ x′，y′ T(x′,y′) 2 This represents the sum of the squares of the brightness values ​​corresponding to each pixel in the preset ROI template. x represents the total step distance in the x-direction from x′ within the brightness image of the blue film surface to the starting scan point, and y represents the total step distance in the y-direction from y′ within the brightness image of the blue film surface to the starting scan point.

[0013]

[0014] In this formula, w represents the width of the preset ROI template, h represents the height of the preset ROI template, and ∑ x′，y′ T(x′, y′) represents the brightness and value of all image points in the preset ROI template.

[0015]

[0016] In this formula, I(x+x′,y+y′) represents the brightness value at position (x+x′,y+y′) in the brightness image of the blue film surface, ∑ x′，y′ I(x+x′,y+y′) represents the brightness and value of all scan points in the w×h region formed by the starting point (x+x′,y+y′) in the brightness image of the blue film surface.

[0017] In one possible implementation, at least one defect region corresponding to each blue film surface and the size information corresponding to each defect region are determined as follows: The point cloud map to be processed is divided into two-dimensional grids according to a preset grid side length; for each grid in the two-dimensional grid, the elevation value corresponding to each feature point in the grid is obtained, and the minimum elevation value in the grid is determined as the grid value corresponding to that grid; an elevation grid is constructed based on the grid values ​​corresponding to each grid in the two-dimensional grid; each grid in the elevation grid is traversed, and the slope value corresponding to that grid is calculated by combining the preset grid side length and the grid value; a slope grid is constructed based on the slope values ​​corresponding to each grid in the elevation grid; for each grid in the slope grid, if the slope value is less than a preset slope value, the grid attribute is determined to be ground; if the slope value is greater than or equal to the preset slope value, the grid attribute is determined to be non-ground; an attribute grid is established based on the grid attributes corresponding to each grid; and at least one defect region and the size information corresponding to each defect region are determined based on the attribute grid.

[0018] In one possible implementation, the slope grid corresponding to the point cloud to be processed is constructed as follows: For each grid in the elevation grid, the following processing is performed: it is determined whether a nine-square grid centered on the current grid can be constructed in the elevation grid; if a nine-square grid centered on the current grid can be constructed in the elevation grid, the slope value corresponding to the current grid is calculated based on the elevation value corresponding to each adjacent grid of the current grid; if a nine-square grid centered on the current grid cannot be constructed in the elevation grid, the slope calculation for the current grid is abandoned, and the next grid is traversed; a slope grid is formed by constructing multiple grids with slope values.

[0019] In one possible implementation, the step of calculating the slope value corresponding to the grid based on the elevation value corresponding to each adjacent grid includes: for each adjacent grid of the grid, performing the following processing: calculating the elevation difference between the grid and the adjacent grid and the grid center distance; calculating the ratio between the elevation difference and the grid center distance corresponding to the grid; calculating the average value among the multiple ratios corresponding to the grid, and determining the average value as the slope value corresponding to the grid.

[0020] In one possible implementation, the size information includes the defect area, defect length, defect diameter, and defect depth. The at least one defect area corresponding to each blue film surface and the size information corresponding to each defect area are determined as follows: multiple feature points in the non-ground grid of the attribute grid are extracted; the multiple feature points in the non-ground grid are mapped to a 2D image, and a connected component extraction algorithm is performed on the 2D image to determine at least one defect area corresponding to the blue film surface; for each defect area, the defect area, defect length, defect diameter, and defect depth are determined based on a preset function.

[0021] In one possible implementation, the defect detection standard includes multiple defect detection rules, each corresponding to a defect type. The defect detection result includes the defect type to which each defect area belongs. The defect detection result for each blue film surface is determined as follows: For each defect area corresponding to the blue film surface, the following processing is performed: The defect area, length, diameter, height, and depth corresponding to the defect area are verified using multiple defect detection rules. For each defect detection rule, if the defect area, length, diameter, height, and depth of the defect area meet the defect size indicated by the defect detection rule, then the defect type corresponding to the defect area is the defect type indicated by the defect detection rule. If the defect area, length, diameter, height, and depth of the defect area do not meet the defect size indicated by any defect detection rule, then the defect area is determined to be a normal area.

[0022] Secondly, this application also provides a defect detection device for the blue film of a battery pack. The device includes: a scanning module for performing 3D scanning on the blue film surface of the battery pack to obtain a brightness image and a depth image corresponding to each blue film surface of the battery pack, wherein the brightness image indicates the brightness value corresponding to each scanning point on the blue film surface, and the depth image indicates the depth value from each scanning point on the blue film surface to the viewpoint of the scanning device; a positioning and matching module for performing positioning and matching on the brightness image corresponding to each blue film surface of the battery pack using a preset ROI template corresponding to that blue film surface, and extracting the target region of interest within the brightness image; and an extraction module for using the coordinate information of the target region of interest on each blue film surface of the battery pack. The system extracts the depth map of the corresponding region to be detected from the depth image of the surface; the point cloud conversion module performs point cloud conversion on the depth map of the region to be detected for each blue film surface of the battery pack to obtain a point cloud image to be processed, which includes multiple feature points; the defect region calculation module calculates the defect region of the point cloud image to be processed for each blue film surface of the battery pack using a point cloud gradient filtering algorithm to obtain at least one defect region corresponding to the blue film surface and the size information of each defect region; the detection module performs defect analysis on at least one defect region for each blue film surface of the battery pack based on defect detection standards and the size information of each defect region to determine the defect detection result corresponding to the blue film surface.

[0023] Thirdly, embodiments of this application also provide an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the memory communicate via the bus. The machine-readable instructions are executed by the processor to perform the steps of the defect detection method for the blue film of the battery pack provided in the first aspect or any possible implementation of the first aspect.

[0024] This application provides a method, apparatus, and electronic device for defect detection of the blue film of a battery pack. The method includes: for each blue film surface of the battery pack, performing the following processing: extracting the corresponding region image to be detected from the depth image of the blue film surface; performing point cloud transformation on the region image to be detected to obtain a point cloud image to be processed; using a point cloud gradient filtering algorithm to calculate the defect region in the point cloud image to be processed, obtaining at least one defect region corresponding to the blue film surface and the size information corresponding to each defect region; and performing defect analysis on at least one defect region based on defect detection standards and the size information corresponding to each defect region to determine the defect detection result corresponding to the blue film surface. This application uses 3D point cloud processing technology to detect defects in the blue film of the battery pack, improving detection accuracy and ensuring detection stability.

[0025] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0026] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 A flowchart of a defect detection method for the blue film of a battery pack provided in an embodiment of this application is shown;

[0028] Figure 2 A schematic diagram of a preset ROI template provided in an embodiment of this application is shown;

[0029] Figure 3 A schematic diagram of a two-dimensional grid provided in an embodiment of this application is shown;

[0030] Figure 4 A schematic diagram of an elevation grid provided in an embodiment of this application is shown;

[0031] Figure 5 A schematic diagram of a slope grid provided in an embodiment of this application is shown;

[0032] Figure 6 A schematic diagram of an attribute grid provided in an embodiment of this application is shown;

[0033] Figure 7 This illustration shows a structural schematic diagram of a defect detection device for the blue film of a battery pack according to an embodiment of this application;

[0034] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0036] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0037] Currently, defect detection of the battery pack's blue film relies entirely on manual visual inspection, which wastes human resources. Manual inspection is inefficient and prone to missed or over-detection due to visual fatigue and subjective judgment. Some manufacturers use 2D or 2.5D vision solutions and visual algorithms for detection. However, due to the reflective surface of the blue film, it is impossible to obtain images with uniform lighting. Traditional algorithms have a high false detection rate, are unstable, and cannot obtain the depth and size of defects. For example, pits, damage, and wrinkles on the battery pack's blue film need to be determined based on the depth and size to determine whether they are defects.

[0038] Based on this, embodiments of this application provide a method, apparatus, and electronic device for defect detection of the blue film in a battery pack. The method utilizes 3D point cloud processing technology to detect defects in the blue film, improving detection accuracy and ensuring detection stability. The details are as follows:

[0039] Please see Figure 1 , Figure 1 A flowchart illustrating a defect detection method for the blue film of a battery pack according to an embodiment of this application is shown. Figure 1 As shown, the defect detection method provided in this application includes the following steps:

[0040] S100. Perform 3D scanning on the blue film surface of the battery pack to obtain the brightness and depth images corresponding to each blue film surface of the battery pack.

[0041] The brightness image indicates the brightness value corresponding to each scanning point on the blue film surface, and the depth image indicates the depth value from each scanning point on the blue film surface to the viewpoint of the scanning device. The scanning device is a 3D line laser measuring instrument.

[0042] In specific implementation, the battery pack in this application can be a square aluminum-cased battery. A 3D line laser measuring instrument is required to scan the six sides of the square aluminum-cased battery (including two small side surfaces, two large side surfaces, the bottom surface, and the top surface). Each blue film surface is scanned to obtain a brightness image and a depth image.

[0043] The field of view for the top and bottom surfaces can be 71mm, with an XY scanning accuracy of 0.017mm / pixel. The field of view for the large and small sides can be 130mm, with an XY scanning accuracy of 0.02mm / pixel. In this application, the brightness image is used to locate the ROI (region of interest), and the depth image is used to identify and calculate defects such as pits, bubbles, and wrinkles.

[0044] S110. For each blue film surface of the battery pack, use the preset ROI template corresponding to the blue film surface to locate and match the brightness image corresponding to the blue film surface, and extract the target region of interest within the brightness image.

[0045] Preferably, the preset ROI template is a pre-defined brightness image corresponding to a blue film surface without defects. Because image interference from corners needs to be eliminated, the size of the preset ROI template is smaller than the corresponding blue film surface size. For example, if the blue film surface size is 5×5, the size of the preset ROI template can be 3×3. Typically, the preset ROI template is determined with the center point of the blue film surface as the center. Please refer to [link / reference]. Figure 2 , Figure 2 A schematic diagram of a preset ROI template provided in an embodiment of this application is shown. Figure 2As shown, A represents the blue film surface used to determine the defect-free area of ​​the preset ROI template, B represents the preset ROI template extracted from A, and AO represents the center point of A.

[0046] In this application, different blue film surfaces correspond to different preset ROI templates. The top and bottom surfaces can share a preset ROI template, the two large side surfaces can share a preset ROI template, and the two small side surfaces can share a preset ROI template. The preset ROI template is also a rectangular area.

[0047] In a preferred embodiment, the target region of interest within the brightness image of each blue film surface is determined by the following method:

[0048] Starting from the initial matching scan point in the brightness image of the blue film surface, the brightness image is traversed according to a preset step size. For each matching scan point, the matching degree between the matching region formed by the matching scan point and the preset ROI template is calculated. The matching region formed by the matching scan point with the highest matching degree is determined as the target region of interest.

[0049] Specifically, the size of the matching area formed by the matching scan points is the same as the size of the preset ROI template area. Here, "same" means that the width of the matching area is the same as the width of the preset ROI template area, and the height of the matching area is the same as the height of the preset ROI template area.

[0050] In another preferred embodiment, the matching degree between each image point and each scan point is determined by the following formula:

[0051]

[0052] R(x, y) represents the calculated matching value between the matching region formed by the matching scan point (x+x′, y+y′) in the brightness image of the blue film surface and the preset ROI template. The larger the calculated matching value, the greater the matching degree.

[0053] In this formula, T(x′, y′) represents the brightness value at position (x′, y′) in the preset ROI template, ∑ x′，y′ T(x′,y′) 2 This represents the sum of the squares of the brightness values ​​corresponding to each pixel in the preset ROI template. x represents the total step length in the x-direction from x′ in the brightness image of the blue film surface to the starting scan point, and y represents the total step length in the y-direction from y′ in the brightness image of the blue film surface to the starting scan point.

[0054] in,

[0055]

[0056] In this formula, w represents the width of the preset ROI template, h represents the height of the preset ROI template, and ∑ x′，y′ T(x′, y′) represents the brightness and value of all image points in the preset ROI template.

[0057]

[0058] In this formula, I(x+x′,y+y′) represents the brightness value at position (x+x′,y+y′) in the brightness image of the blue film surface, ∑ x′，y′ I(x+x′,y+y′) represents the brightness and value of all scan points in the w×h region formed by the starting point (x+x′,y+y′) in the brightness image of the blue film surface.

[0059] In this process, the region in the brightness image of the blue film surface that best matches the brightness of the corresponding preset ROI template is found, and the region with the best brightness match is determined as the target region of interest. This method can effectively solve the problem of the influence caused by the difference in brightness between the ROI template and the brightness image of the blue film surface.

[0060] S120. For each blue film surface of the battery pack, using the coordinate information of the target region of interest, extract the corresponding depth map of the region to be detected from the depth image of the blue film surface.

[0061] In this application, for each blue film surface, while determining the target region of interest, the coordinate information corresponding to the target region of interest is also determined. Since the preset ROI template is rectangular, the determined target region of interest is also a rectangular region. Its corresponding coordinate information includes the coordinates of the four vertex points and the center coordinates of the rectangular region. Based on the coordinates of the four vertex points and the center coordinates of the target region of interest, the corresponding positions can be directly extracted on the depth image to form a depth map of the region to be detected.

[0062] S130. For each blue film surface of the battery pack, perform point cloud conversion on the depth map of the area to be detected to obtain the point cloud map to be processed.

[0063] The point cloud map to be processed includes multiple feature points.

[0064] S140. For each blue film surface of the battery pack, the point cloud gradient filtering algorithm is used to calculate the defect region of the point cloud image to be processed, so as to obtain at least one defect region corresponding to the blue film surface and the size information corresponding to each defect region.

[0065] In a preferred embodiment, at least one defect region corresponding to each blue film surface and the size information corresponding to each defect region are determined in the following manner:

[0066] The point cloud map to be processed is divided into two-dimensional grids according to the preset grid side length. For each grid in the two-dimensional grid, the elevation value corresponding to each feature point in the grid is obtained. The minimum elevation value in the grid is determined as the grid value corresponding to the grid. Based on the grid value corresponding to each grid in the two-dimensional grid, an elevation grid is constructed.

[0067] For details, please refer to Figure 3 , Figure 3 A schematic diagram of a two-dimensional grid provided in an embodiment of this application is shown. Figure 3 As shown, taking a 3×3 nine-square grid as an example, each grid can be understood as a small cube containing multiple feature points. In this application, the elevation value refers to the distance from the feature point to the absolute datum plane corresponding to the scanning device, such as... Figure 3 As shown, for grid a in the two-dimensional grid Grid1 13 It contains 3 feature points, with elevation values ​​of 15, 26, and 28 respectively. The minimum elevation value is 15, and 15 is defined as grid a. 13 The corresponding grid value.

[0068] Please see Figure 4 , Figure 4 A schematic diagram of an elevation grid provided in an embodiment of this application is shown. Figure 3 The example of Grid1, a 3×3 two-dimensional grid, shown below, ultimately yields the following result: Figure 4 The elevation grid shown is Grid2.

[0069] Traverse each grid in the elevation grid, calculate the slope value corresponding to the grid by combining the preset grid side length and grid value, and construct a slope grid based on the slope value corresponding to each grid in the elevation grid.

[0070] In a preferred embodiment, the slope grid corresponding to the point cloud to be processed is constructed in the following manner:

[0071] For each grid cell in the elevation grid, perform the following processing:

[0072] Determine whether a nine-square grid centered on the current elevation grid can be constructed. If it can, calculate the slope value of the grid based on the elevation values ​​of each adjacent grid. If it cannot be constructed, abandon the slope calculation for the current grid and continue traversing the next grid. A slope grid is formed by constructing multiple grids with slope values.

[0073] In another preferred embodiment, the step of calculating the slope value corresponding to the grid based on the elevation value corresponding to each adjacent grid cell includes:

[0074] For each adjacent grid cell of this grid, perform the following processing:

[0075] Calculate the elevation difference between the grid and its adjacent grid, as well as the distance between the grid centers. Calculate the ratio between the elevation difference and the distance between the grid centers for the grid. Calculate the average of the multiple ratios for the grid and determine the average value as the slope value for the grid.

[0076] Please see Figure 5 , Figure 5 A schematic diagram of a slope grid provided in an embodiment of this application is shown. Figure 4 The local example of the elevation grid Grid2 shown below results in the following: Figure 5 The slope grid Grid3 shown is specifically designed to be... Figure 4 The central grid a of the elevation grid 22 For example, suppose the preset grid side length is 8m × 8m, a 22 The corresponding grid value is 11, and it can be used with the central grid a. 22 Constructing a 3x3 grid, the central grid a 22 The adjacent grids and their corresponding grid values ​​include a 11 =14, a 12 =16, a 13 =15, a 21 =15, a 23 =14, a 31 =12, a 32 =14 and a 33 =14, where a 22 Adjacent grid a 12 For example, a 22 With a 12 The slope value S formed between them L2 = (16-11) / 8 = 0.62, (16-11) represents a 12 With a 22 The elevation difference between them, 8 represents a 12 With a 22 The distance between the center of the grid.

[0077] We can calculate a in sequence. 11 With a 22 The slope value S between L1 =(14-11) / (8×1.42)=0.26,a 13 With a 22 The slope value S between L3 =(15-11) / (8×1.42)=0.35,a 21 With a 22 The slope value S between L4=(15-11) / 8=0.5, a 23 With a 22 The slope value S between L5 =(14-11) / 8=0.37, a 31 With a 22 The slope value S between L6 =(12-11) / (8×1.42)=0.08,a 32 With a 22 The slope value S between L7 = (14-11) / 8 = 0.37 and a 33 With a 22 The slope value S between L8 = (14-11) / (8×1.42) = 0.26.

[0078]

[0079] Similarly, for each grid in the elevation grid, if it can form a nine-square grid, a slope value can be determined. Multiple grids with slope values ​​can be used to form a slope grid.

[0080] For each grid in the slope grid, if the slope value is less than the preset slope value, the grid attribute is determined to be ground; if the slope value is greater than or equal to the preset slope value, the grid attribute is determined to be non-ground. Based on the grid attribute corresponding to each grid, an attribute grid is established.

[0081] Please see Figure 6 , Figure 6 A schematic diagram of an attribute grid provided in an embodiment of this application is shown.

[0082] like Figure 5 and Figure 6 As shown, the slope grid Grid3 includes grids and corresponding slope values, including b. 11 =0.25, b 12 =0.15, b 13 =0.19, b 21 =0.25, b 22 =0.28, b 23 =0.29, b 31 =0.5, b 32 =0.20 and b 33 =0.22, where, if the preset slope value is 0.4, then an attribute grid Grid4 is constructed, whose grid and corresponding attribute values ​​include c 11 =Ground, c 12 =Ground, c 13 =Ground, c 21 =Ground, c 22=Ground, c 23 =Ground, c 31 =Non-ground, c 32 = Ground and c 33 = Ground.

[0083] Based on the attribute grid, at least one defect region and the corresponding size information of each defect region are determined.

[0084] Specifically, the dimensional information includes the defect area, defect length, defect diameter, and defect depth. The dimensional information for each defect area is determined by the following method:

[0085] Multiple feature points in the non-ground grid of the attribute grid are extracted, and the multiple feature points in the non-ground grid are mapped into a 2D image. The connected component algorithm is then used to extract the 2D image to determine at least one defect region corresponding to the blue film surface. For each defect region, the defect region area, defect region length, defect region diameter, and defect region depth are determined based on a preset function.

[0086] S150. For each blue film surface of the battery pack, perform defect analysis on at least one defect area based on the defect detection standard and the size information corresponding to each defect area, and determine the defect detection result corresponding to the blue film surface.

[0087] In one example, the defect detection standard includes multiple defect detection rules, each corresponding to a defect type. The defect detection result includes the defect type to which each defect area belongs. Defect types include bubbles, pits, breaks, wrinkles, and scratches. Specifically, the defect size corresponding to bubbles can be a defect area diameter ≥ 8 mm, the defect area depth corresponding to pits can be a defect area depth ≥ 0.2 mm and a defect area diameter ≥ 8 mm, the defect size corresponding to breaks can be a defect area height ≥ 0.8 mm and a defect area length of 1.5 mm ≤ 5 mm, the defect size corresponding to wrinkles can be a defect area height ≥ 0.8 mm and a defect area length < 5 mm, and the defect size corresponding to scratches can be a defect area length of 10 mm ≤ 20 mm.

[0088] In this step of the application, the defect detection standard can be invoked first. When the detection standard and rules change, the newly set defect detection standard can be used for subsequent defect detection.

[0089] The defect detection results for each blue film surface are determined using the following method:

[0090] For each defect area corresponding to the blue film surface, perform the following processing:

[0091] Multiple defect detection rules are used to verify the defect area, defect length, defect diameter, defect height, and defect depth corresponding to the defect area. For each defect detection rule, if the defect area, defect length, defect diameter, defect height, and defect depth of the defect area meet the defect size indicated by the defect detection rule, then the defect type of the defect area is the defect type indicated by the defect detection rule. If the defect area, defect length, defect diameter, defect height, and defect depth of the defect area do not meet the defect size indicated by any defect detection rule, then the defect area is determined to be a normal area.

[0092] Based on the same application concept, this application also provides a battery pack blue film defect detection device corresponding to the battery pack blue film defect detection method provided in the above embodiments. Since the principle of the device in this application is similar to the battery pack blue film defect detection method in the above embodiments of this application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0093] Please see Figure 7 , Figure 7 A schematic diagram of a defect detection device for the blue film of a battery pack, provided in an embodiment of this application, is shown. Figure 7 As shown, the device includes

[0094] The scanning module 200 is used to perform 3D scanning on the blue film surface of the battery pack to obtain a brightness image and a depth image corresponding to each blue film surface of the battery pack. The brightness image indicates the brightness value corresponding to each scanning point on the blue film surface, and the depth image indicates the depth value from each scanning point on the blue film surface to the viewpoint of the scanning device.

[0095] The positioning and matching module 210 is used to perform positioning and matching on the brightness image corresponding to each blue film surface of the battery pack using a preset ROI template corresponding to that blue film surface, and extract the target region of interest within the brightness image.

[0096] The extraction module 220 is used to extract the corresponding depth map of the region to be detected from the depth image of each blue film surface of the battery pack, using the coordinate information of the target region of interest.

[0097] The point cloud conversion module 230 is used to perform point cloud conversion on the depth map of the area to be detected for each blue film surface of the battery pack, so as to obtain a point cloud map to be processed, which includes multiple feature points.

[0098] The defect region calculation module 240 is used to calculate the defect region of the point cloud map to be processed for each blue film surface of the battery pack using a point cloud gradient filtering algorithm, so as to obtain at least one defect region corresponding to the blue film surface and the size information of each defect region.

[0099] The detection module 250 is used to perform defect analysis on at least one defect area for each blue film surface of the battery pack based on defect detection standards and the size information corresponding to each defect area, and determine the defect detection result corresponding to the blue film surface.

[0100] Preferably, the positioning and matching module 210 is further configured to: start from the initial matching scanning point in the brightness image of the blue film surface, traverse the brightness image according to a preset step size; for each matching scanning point, calculate the matching degree between the matching region formed by the matching scanning point and the preset ROI template, wherein the size of the matching region is the same as the size of the region of the preset ROI template; and determine the matching region formed by the matching scanning point corresponding to the highest matching degree as the target region of interest.

[0101] Preferably, the positioning and matching module 210 is further configured to determine the matching degree between each image point and each scan point using the following formula:

[0102]

[0103] R(x, y) represents the calculated matching value between the matching region formed by the matching scan point (x+x′, y+y′) in the brightness image of the blue film surface and the preset ROI template. The larger the calculated matching value, the greater the matching degree.

[0104] In this formula, T(x′, y′) represents the brightness value at position (x′, y′) in the preset ROI template, ∑ x′，y′ T(x′,y′) 2 This represents the sum of the squares of the brightness values ​​corresponding to each pixel in the preset ROI template. x represents the total step distance in the x-direction from x′ within the brightness image of the blue film surface to the starting scan point, and y represents the total step distance in the y-direction from y′ within the brightness image of the blue film surface to the starting scan point.

[0105]

[0106] In this formula, w represents the width of the preset ROI template, and h represents the height of the preset ROI template. x′，y′ T(x′, y′) represents the brightness and value of all image points in the preset ROI template.

[0107]

[0108] In this formula, I(x+x′,y+y′) represents the brightness value at position (x+x′,y+y′) in the brightness image of the blue film surface, ∑ x′，y′ I(x+x′,y+y′) represents the brightness and value of all scan points in the w×h region formed by the starting point (x+x′,y+y′) in the brightness image of the blue film surface.

[0109] The defect area calculation module 240 is further configured to: divide the point cloud map to be processed according to the preset grid side length to obtain a two-dimensional grid; for each grid in the two-dimensional grid, obtain the elevation value corresponding to each feature point in the grid, and determine the minimum elevation value in the grid as the grid value corresponding to the grid; construct an elevation grid based on the grid value corresponding to each grid in the two-dimensional grid; traverse each grid in the elevation grid, calculate the slope value corresponding to the grid by combining the preset grid side length and grid value, and construct a slope grid based on the slope value corresponding to each grid in the elevation grid; for each grid in the slope grid, if the slope value is less than the preset slope value, determine the grid attribute as ground; if the slope value is greater than or equal to the preset slope value, determine the grid attribute as non-ground; establish an attribute grid based on the grid attribute corresponding to each grid; and determine at least one defect area and the size information corresponding to each defect area based on the attribute grid.

[0110] Preferably, the defect area calculation module 240 is further configured to: for each grid in the elevation grid, perform the following processing: determine whether a nine-square grid centered on the current grid can be constructed in the elevation grid; if a nine-square grid centered on the current grid can be constructed in the elevation grid, calculate the slope value corresponding to the current grid based on the elevation value corresponding to each adjacent grid of the current grid; if a nine-square grid centered on the current grid cannot be constructed in the elevation grid, abandon the slope calculation of the current grid and continue to traverse the next grid; and construct a slope grid from multiple grids with slope values.

[0111] Preferably, the defect area calculation module 240 is further configured to: for each adjacent grid of the grid, calculate the elevation difference between the grid and the adjacent grid and the grid center distance; calculate the ratio between the elevation difference and the grid center distance corresponding to the grid; calculate the average value among the multiple ratios corresponding to the grid, and determine the average value as the slope value corresponding to the grid.

[0112] Preferably, the size information includes the defect area, defect length, defect diameter, and defect depth. The defect area calculation module 240 is further configured to: extract multiple feature points from the non-ground grid in the attribute grid; map the multiple feature points from the non-ground grid to a 2D image, and perform connected component extraction on the 2D image to determine at least one defect area corresponding to the blue film surface; and for each defect area, determine the defect area, defect length, defect diameter, and defect depth based on a preset function.

[0113] Preferably, the defect detection standard includes multiple defect detection rules, each corresponding to a defect type. The defect detection result includes the defect type to which each defect area belongs. The detection module 250 is further configured to: for each defect area corresponding to the blue film surface, perform the following processing: verify the defect area area, defect area length, defect area diameter, defect area height, and defect area depth corresponding to the defect area using multiple defect detection rules; for each defect detection rule, if the defect area area, defect area length, defect area diameter, defect area height, and defect area depth corresponding to the defect area meet the defect size indicated by the defect detection rule, then the defect type corresponding to the defect area is the defect type indicated by the defect detection rule; if the defect area area, defect area length, defect area diameter, defect area height, and defect area depth corresponding to the defect area do not meet the defect size indicated by any defect detection rule, then the defect area is determined to be a normal area.

[0114] Based on the same application concept, please refer to Figure 8 , Figure 8 This diagram illustrates the structure of an electronic device according to an embodiment of this application. The electronic device 300 includes a processor 310, a memory 320, and a bus 330. The memory 320 stores machine-readable instructions executable by the processor 310. When the electronic device 300 is running, the processor 310 and the memory 320 communicate via the bus 330. The machine-readable instructions are executed by the processor 310 to perform the steps of the defect detection method for the blue film of the battery pack as provided in any of the above embodiments.

[0115] Based on the same concept, this application also provides a computer-readable storage medium storing a computer program, which, when run by a processor, executes the steps of the defect detection method for the blue film of the battery pack provided in the above embodiments.

[0116] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and devices described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division; in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection may be through some communication interfaces; the indirect coupling or communication connection of devices or units may be electrical, mechanical, or other forms.

[0117] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0118] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0119] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0120] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for detecting defects in the blue film of a battery pack, characterized in that, The method includes: 3D scanning is performed on the blue film surface of the battery pack to obtain the brightness image and depth image corresponding to each blue film surface of the battery pack. The brightness image indicates the brightness value corresponding to each scanning point on the blue film surface, and the depth image indicates the depth value from each scanning point on the blue film surface to the viewpoint of the scanning device. For each blue film surface of the battery pack, perform the following processing: The preset ROI template corresponding to the blue film surface is used to locate and match the brightness image corresponding to the blue film surface, and the target region of interest is extracted in the brightness image. Using the coordinate information of the target region of interest, the corresponding depth map of the region to be detected is extracted from the depth image of the blue film surface; The depth map of the region to be detected is converted into a point cloud to obtain a point cloud map to be processed, which includes multiple feature points. The point cloud gradient filtering algorithm is used to calculate the defect region of the point cloud image to be processed, so as to obtain at least one defect region corresponding to the blue film surface and the size information of each defect region. Based on the defect detection standard and the size information corresponding to each defect area, defect analysis is performed on the at least one defect area to determine the defect detection result corresponding to the blue film surface. The target region of interest within the brightness image of each blue film surface is determined using the following method: Starting from the initial matching scan point within the brightness image of the blue film surface, traverse the brightness image according to a preset step size; For each matching scan point, calculate the matching degree between the matching region formed by the matching scan point and the preset ROI template. The size of the matching region is the same as the size of the region of the preset ROI template. The matching region formed by the matching scan point corresponding to the highest matching degree is determined as the target region of interest.

2. The method according to claim 1, characterized in that, The matching degree between each image point and each scan point is determined by the following formula: The image representing the brightness of the blue film surface is in the form of The calculated matching value is the matching region formed by the initial matching scan point and the preset ROI template. The larger the calculated matching value, the greater the matching degree. In this formula, Indicates the preset ROI template Brightness value at the location This represents the sum of the squares of the brightness values ​​corresponding to each pixel in the preset ROI template. The brightness of the blue film surface is represented in the image. The total step size in the x-direction from the starting scan point, and y represents the brightness of the blue film surface within the image. Total step size in the y-direction from the starting scan point In this formula, This indicates the width of the preset ROI template. Indicates the height of the preset ROI template. This represents the brightness and value of all pixels in the preset ROI template. In this formula, The image representing the brightness of the blue film surface Brightness value at the location The image representing the brightness of the blue film surface is in the form of Formed from the starting point The brightness and value corresponding to all scan points in the region.

3. The method according to claim 1, characterized in that, The following methods are used to determine at least one defect region corresponding to each blue film surface and the size information corresponding to each defect region: The point cloud map to be processed is divided according to the preset grid side length to obtain a two-dimensional grid; For each grid in the two-dimensional grid, the elevation value corresponding to each feature point in the grid is obtained, and the minimum elevation value in the grid is determined as the grid value corresponding to the grid. An elevation grid is constructed based on the grid value corresponding to each grid in the two-dimensional grid. Traverse each grid in the elevation grid, calculate the slope value corresponding to the grid by combining the preset grid side length and grid value, and construct a slope grid based on the slope value corresponding to each grid in the elevation grid. For each grid in the slope grid, if the slope value is less than the preset slope value, the grid attribute is determined to be ground; if the slope value is greater than or equal to the preset slope value, the grid attribute is determined to be non-ground. Based on the grid attribute corresponding to each grid, an attribute grid is established. Based on the attribute grid, at least one defect region and the corresponding size information of each defect region are determined.

4. The method according to claim 3, characterized in that, Construct the slope grid corresponding to the point cloud to be processed using the following method: For each grid cell in the elevation grid, perform the following processing: Determine whether a nine-square grid centered on the given elevation grid can be constructed. If a nine-square grid centered on the elevation grid can be constructed in the elevation grid, then the slope value corresponding to the grid can be calculated based on the elevation value corresponding to each adjacent grid of the grid. If a nine-square grid centered on the elevation grid cannot be constructed, the slope calculation for that grid is abandoned, and the next grid is traversed. The slope grid is constructed from multiple grids with slope values.

5. The method according to claim 4, characterized in that, The step of calculating the slope value corresponding to the grid based on the elevation value corresponding to each adjacent grid includes: For each adjacent grid cell of this grid, perform the following processing: Calculate the elevation difference between this grid and its adjacent grids, as well as the distance between the grid centers; Calculate the ratio between the elevation difference corresponding to the grid and the distance from the grid center; Calculate the average value among the multiple ratios corresponding to the grid, and determine the average value as the slope value corresponding to the grid.

6. The method according to claim 4, characterized in that, The dimensional information includes the defect area, defect length, defect diameter, and defect depth. Specifically, at least one defect region corresponding to each blue film surface and the size information corresponding to each defect region are determined in the following way: Extract multiple feature points from the non-ground grid in the attribute grid; Multiple feature points in the non-ground grid are mapped to a 2D image, and the connected component algorithm is used to extract the 2D image to determine at least one defect region corresponding to the blue film surface. For each defective region, the area, length, diameter, and depth of the defective region are determined based on a preset function.

7. The method according to claim 6, characterized in that, The defect detection standard includes multiple defect detection rules, each corresponding to a defect type. The defect detection result includes the defect type to which each defect area belongs. The defect detection results for each blue film surface are determined using the following method: For each defect area corresponding to the blue film surface, perform the following processing: Multiple defect detection rules were used to verify the defect area, defect length, defect diameter, defect height, and defect depth corresponding to the defect area. For each defect detection rule, if the defect area, defect length, defect diameter, defect height, and defect depth of the defect area meet the defect size indicated by the defect detection rule, then the defect type corresponding to the defect area is the defect type indicated by the defect detection rule. If the defect area, length, diameter, height, and depth of the defect area do not meet the defect size indicated by any defect detection rule, then the defect area is determined to be a normal area.

8. A defect detection device for the blue film of a battery pack, characterized in that, The device includes: The scanning module is used to perform 3D scanning on the blue film surface of the battery pack to obtain a brightness image and a depth image corresponding to each blue film surface of the battery pack. The brightness image indicates the brightness value corresponding to each scanning point on the blue film surface, and the depth image indicates the depth value from each scanning point on the blue film surface to the viewpoint of the scanning device. The localization and matching module is used to locate and match the brightness image corresponding to each blue film surface of the battery pack using a preset ROI template corresponding to that blue film surface, and extract the target region of interest within the brightness image. The extraction module is used to extract the corresponding depth map of the region to be detected from the depth image of each blue film surface of the battery pack, using the coordinate information of the target region of interest. The point cloud conversion module is used to perform point cloud conversion on the depth map of the area to be detected for each blue film surface of the battery pack to obtain a point cloud map to be processed, which includes multiple feature points. The defect region calculation module is used to calculate the defect region of the point cloud map to be processed for each blue film surface of the battery pack using a point cloud gradient filtering algorithm, so as to obtain at least one defect region corresponding to the blue film surface and the size information of each defect region. The detection module is used to perform defect analysis on at least one defect area for each blue film surface of the battery pack based on defect detection standards and the size information corresponding to each defect area, and determine the defect detection result corresponding to the blue film surface. The positioning and matching module is further used for: Starting from the initial matching scan point within the brightness image of the blue film surface, traverse the brightness image according to a preset step size; For each matching scan point, calculate the matching degree between the matching region formed by the matching scan point and the preset ROI template. The size of the matching region is the same as the size of the region of the preset ROI template. The matching region formed by the matching scan point corresponding to the highest matching degree is determined as the target region of interest.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. The machine-readable instructions are executed by the processor to perform the steps of the defect detection method for the blue film of the battery pack as described in any one of claims 1 to 7.

Images