Nickel sheet assembly detection method for battery module

Abstract

The present invention belongs to the technical field of nickel sheet assembly of battery modules, and specifically discloses a nickel sheet assembly detection method for battery modules. The method obtains a correctly installed nickel sheet assembly area and an installation block of each nickel sheet by marking a reference image of a standard battery module nickel sheet assembly structure, thereby finding the nickel sheet assembly area and the installation block of each nickel sheet on the image to be detected of the nickel sheet assembly structure of the battery module to be detected through affine transformation, and simultaneously respectively calculates the geometric moments of the reference image and the assembly area, thereby being able to determine the area difference ratio, center of gravity Euclidean distance, and cosine similarity of each nickel sheet between the two through zero-order moment, first-order moment, and second-order moment calculations, and qualitatively and quantitatively analyzes the accurate coverage ratio, horizontal offset degree, and angular deflection degree of the nickel sheet installation, and can effectively determine the problems and degree of the nickel sheet installation. The method is stable and controllable, and the implementation cost is low.

Description

Technical Field

[0001] The invention belongs to the technical field of nickel sheet assembly of battery modules, and in particular relates to a nickel sheet assembly detection method for a battery module. Background Art

[0002] In the assembly process of cylindrical battery modules, the assembly of nickel sheets and cylindrical batteries is a crucial key process. As a commonly used metal material in cylindrical battery modules, nickel sheets are mainly responsible for the conduction of the circuit, and their shape design is usually carefully designed. During the preparation process, the nickel sheets must be assembled to the fixture in strict accordance with the specified position and direction, and then fixed through a welding process to form a complete battery module. Errors in the assembly of nickel sheets may lead to serious consequences, such as short circuits and leakage, and even fires and other safety accidents. Therefore, in the assembly of nickel sheets in battery modules, an effective error-proofing solution is needed to ensure the accuracy and safety of assembly.

[0003] In the existing technology, some nickel sheet assembly solutions use a binary classification AI algorithm for small sample learning, and users independently label good and bad products to provide samples for training. However, the AI ​​algorithm lacks interpretability during the learning and detection process. When a misjudgment occurs, it is difficult to trace the problem, so the controllability is poor. This method often requires repeated attempts and multiple sample learning to achieve a certain accuracy rate, and requires high labeling experience from the operator. In addition, such solutions and equipment are usually expensive, which increases the cost of the process. Summary of the invention

[0004] The technical problem to be solved by the present invention is to overcome the defects in the prior art that nickel sheet assembly of battery modules is prone to errors and the AI ​​learning algorithm has poor controllability and high cost, thereby providing a nickel sheet assembly detection method for battery modules.

[0005] The present invention discloses a nickel sheet assembly detection method for a battery module, comprising the following method steps:

[0006] Step S1: Acquire a reference image of a standard battery module nickel sheet assembly structure;

[0007] Step S2: marking the first nickel sheet assembly area in the reference image, and extracting the first key feature point in the first nickel sheet assembly area;

[0008] Step S3: marking the ROI region of each nickel sheet in the reference image, and calculating the zero-order moment, first-order moment and second-order moment of the first nickel sheet block corresponding to each ROI region in the reference image;

[0009] Step S4: collecting the image to be inspected of the nickel sheet assembly structure of the battery module to be inspected;

[0010] Step S5: performing affine transformation calculation on the image to be detected to obtain an affine transformation image, wherein the second nickel sheet assembly area in the affine transformation image is consistent with the first nickel sheet assembly area in the reference image;

[0011] Step S6: intercepting the affine transformation image through the ROI area to obtain each second nickel sheet block, and calculating the zero-order moment, first-order moment and second-order moment of the second nickel sheet block;

[0012] Step S7: Based on the zero-order moment, first-order moment, and second-order moment of each of the first nickel sheet blocks and the second nickel sheet blocks, the area difference ratio, center of gravity Euclidean distance, and cosine similarity of each of the first nickel sheet blocks and the corresponding second nickel sheet block are calculated to determine the accuracy of nickel sheet assembly of the battery module to be tested.

[0013] Furthermore, the step S1 further includes:

[0014] Step S1.1: preprocessing the reference image of the standard battery module nickel sheet assembly structure, wherein the preprocessing method includes at least one of Gaussian filtering noise reduction, color space conversion, and contrast enhancement.

[0015] Furthermore, the step S2 includes:

[0016] Step S2.1: marking the first nickel sheet assembly area in the reference image;

[0017] Step S2.2: extracting the first key feature point in the first nickel sheet assembly area in the reference image by a feature extraction algorithm, and generating the first key point coordinates and the first descriptor.

[0018] Furthermore, the step S3 includes:

[0019] Step S3.1: marking the ROI region of each nickel sheet in the reference image, and dividing the reference image into first nickel sheet blocks based on each ROI region;

[0020] Step S3.2: performing binarization algorithm processing on each of the first nickel sheet blocks to form a first binarized block;

[0021] Step S3.3: performing edge extraction on each of the first binarized blocks to form each first edge block;

[0022] Step S3.4: Calculate the zero-order moment, the first-order moment and the second-order moment of each first edge block.

[0023] Furthermore, in step S3.3:

[0024] The zero-order moment includes M00:

[0025] ;

[0026] The first-order moments include M10 and M01:

[0027] ;

[0028] ;

[0029] The second-order moments include M11, M20 and M02:

[0030] ;

[0031] ;

[0032] ;

[0033] in, represents the grayscale value of the pixel point in the edge block, i represents the horizontal coordinate, and j represents the vertical coordinate.

[0034] Furthermore, the step S4 further includes:

[0035] Step S4.1: preprocessing the image to be inspected of the nickel sheet assembly structure of the battery module to be inspected, wherein the preprocessing method is consistent with the preprocessing method of the reference image of the nickel sheet assembly structure of the standard battery module.

[0036] Furthermore, the step S5 comprises:

[0037] Step S5.1: extracting the second key feature point from the image to be detected by a feature extraction algorithm, and generating the second key point coordinates and the second descriptor;

[0038] Step S5.2: matching the first key point coordinates and the first descriptor with the second key point coordinates and the second descriptor through a feature matcher, and obtaining the second nickel sheet assembly area in the image to be detected through affine transformation calculation;

[0039] Step S5.3: Filter out high-scoring points in the feature matching results, perform affine transformation matrix calculation, and obtain an affine transformation matrix; perform affine transformation calculation on the image to be detected based on the affine transformation matrix to obtain an affine transformation image, wherein the second nickel sheet assembly area in the affine transformation image is consistent with the first nickel sheet assembly area in the reference image.

[0040] Further, the step S6 includes:

[0041] Step S6.1: Segmenting the affine transformed image based on the ROI regions to form second nickel sheet blocks;

[0042] Step S6.2: performing binarization algorithm processing on each of the second nickel sheet blocks to form a second binarized block;

[0043] Step S6.3: performing edge extraction on each of the second binarized blocks to form each second edge block;

[0044] Step S6.4: Calculate the zero-order moment, the first-order moment and the second-order moment of each second edge block.

[0045] Furthermore, the step S7 includes:

[0046] Step S7.1: Calculate the area difference ratio, centroid Euclidean distance, and cosine similarity of each of the first nickel sheet blocks and the corresponding second nickel sheet block based on the zero-order moment, first-order moment, and second-order moment of each of the first nickel sheet blocks and the second nickel sheet block;

[0047] The area difference ratio is expressed as:

[0048] ;

[0049] The centroid coordinates of the first nickel sheet block and the second nickel sheet block are calculated according to the first-order moment and the second-order moment:

[0050] The center of gravity coordinates of the first nickel sheet block are:

[0051] (x,y), where x= , y= ;

[0052] The center of gravity coordinates of the second nickel sheet block are:

[0053] (x,y), where x= , y= ;

[0054] According to the centroid coordinates of the first nickel sheet block and the second nickel sheet block, the centroid Euclidean distance of the corresponding nickel sheet is calculated:

[0055] ;

[0056] Calculate the distribution of the first nickel block along the x-axis respectively , distribution along the y-axis Linkage with xy direction :

[0057] ;

[0058] ;

[0059] ;

[0060] Calculate the distribution of the second nickel block along the x-axis respectively , distribution along the y-axis Linkage with xy direction :

[0061] ;

[0062] ;

[0063] ;

[0064] Calculate the cosine similarity between the first nickel sheet block and the second nickel sheet block:

[0065] ;

[0066] in, represents the geometric moment of the kth nickel block, represents the geometric moment of the kth second nickel sheet block;

[0067] Step S7.2: Set an area difference ratio threshold. When the area difference ratio is greater than the area difference ratio threshold, it is considered that the corresponding nickel sheet area deviation is too large; set a center of gravity Euclidean distance threshold. When the center of gravity Euclidean distance is greater than the center of gravity Euclidean distance threshold, it is considered that the corresponding nickel sheet center of gravity deviates too much; set a cosine similarity threshold. When the cosine similarity is less than the cosine similarity threshold, it is considered that the corresponding nickel sheet contour eccentricity is too large.

[0068] Furthermore, it also includes:

[0069] Step S8: Determine whether the corresponding nickel sheet is misassembled based on the area difference ratio, center of gravity Euclidean distance, and cosine similarity between each of the first nickel sheet blocks and the second nickel sheet blocks; if the assembly is wrong, render a red frame on the corresponding second nickel sheet block of the affine transformation image; if the assembly is correct, render a green frame on the corresponding second nickel sheet block of the affine transformation image.

[0070] Beneficial effect: The present invention discloses a nickel sheet assembly detection method for a battery module, which obtains the nickel sheet assembly area of ​​the correctly installed nickel sheet and the installation block of each nickel sheet by marking a reference image of a standard battery module nickel sheet assembly structure, thereby finding the nickel sheet assembly area and the installation block of each nickel sheet on the image to be detected of the nickel sheet assembly structure of the battery module to be detected through affine transformation, and simultaneously calculates the geometric moments of the reference image and the assembly area respectively, thereby being able to determine the area difference ratio, center of gravity Euclidean distance, and cosine similarity of each nickel sheet between the two through zero-order moment, first-order moment, and second-order moment calculations, and qualitatively and quantitatively analyze the accurate coverage ratio, horizontal offset degree, and angular deflection degree of the nickel sheet installation, which can effectively determine the problems and degree of the nickel sheet installation, and is stable and controllable at a low cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0072] Figure 1 The figure is a schematic block diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0073] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present application, so the present application is not limited by the specific embodiments disclosed below.

[0074] In the description of the present application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present application, "plurality" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0075] Reference Figure 1 As shown, this embodiment discloses a nickel sheet assembly detection method for a battery module, comprising the following method steps:

[0076] Step S1: Acquire a reference image of a standard battery module nickel sheet assembly structure;

[0077] Specifically, the step S1 further includes:

[0078] Step S1.1: Preprocess the reference image of the standard battery module nickel sheet assembly structure, wherein the preprocessing method includes at least one of Gaussian filtering noise reduction, color space conversion, and contrast enhancement. In this embodiment, the preprocessing can reduce the noise in the image, enhance the target features, and reduce the amount of calculation.

[0079] Step S2: marking the first nickel sheet assembly area in the reference image, and extracting the first key feature point in the first nickel sheet assembly area;

[0080] Specifically, the step S2 includes:

[0081] Step S2.1: marking the first nickel sheet assembly area in the reference image; in this embodiment, a rectangular frame is used to surround all nickel sheets to form the first nickel sheet assembly area.

[0082] Step S2.2: extracting the first key feature point in the first nickel sheet assembly area in the reference image by a feature extraction algorithm, and generating first key point coordinates and a first descriptor. In this embodiment, the feature extraction algorithm is preferably a SIFT feature extraction algorithm.

[0083] Step S3: Mark the ROI area of ​​each nickel sheet in the reference image, and calculate the zero-order moment, first-order moment and second-order moment of the first nickel sheet block corresponding to each ROI area in the reference image; in this embodiment, each first nickel sheet block is a rectangle that just surrounds one nickel sheet.

[0084] Specifically, step S3 includes:

[0085] Step S3.1: marking the ROI region of each nickel sheet in the reference image, and dividing the reference image into first nickel sheet blocks based on each ROI region;

[0086] Step S3.2: performing binarization algorithm processing on each of the first nickel sheet blocks to form a first binarized block;

[0087] Step S3.3: performing edge extraction on each of the first binarized blocks to form each first edge block;

[0088] Step S3.4: Calculate the zero-order moment, the first-order moment and the second-order moment of each first edge block.

[0089] In step S3.3:

[0090] The zero-order moment includes M00:

[0091] ;

[0092] The first-order moments include M10 and M01:

[0093] ;

[0094] ;

[0095] The second-order moments include M11, M20 and M02:

[0096] ;

[0097] ;

[0098] ;

[0099] in, represents the grayscale value of the pixel point in the edge block, i represents the horizontal coordinate, and j represents the vertical coordinate.

[0100] Step S4: collecting the image to be inspected of the nickel sheet assembly structure of the battery module to be inspected;

[0101] Specifically, the step S4 further includes:

[0102] Step S4.1: Preprocess the image to be detected of the nickel sheet assembly structure of the battery module to be detected, and the preprocessing method is consistent with the preprocessing method of the reference image of the nickel sheet assembly structure of the standard battery module. In this embodiment, by controlling the preprocessing method of the image to be detected of the nickel sheet assembly structure of the battery module to be detected and the reference image of the nickel sheet assembly structure of the standard battery module to be consistent, it is possible to ensure that the image features after preprocessing remain consistent, avoid the differences introduced by the preprocessing, and thus cause errors in the subsequent affine transformation and the calculation of the area difference ratio, the center of gravity Euclidean distance, and the cosine similarity.

[0103] Step S5: performing affine transformation calculation on the image to be detected to obtain an affine transformation image, wherein the second nickel sheet assembly area in the affine transformation image is consistent with the first nickel sheet assembly area in the reference image;

[0104] Specifically, step S5 includes:

[0105] Step S5.1: extracting the second key feature point from the image to be detected by a feature extraction algorithm, and generating the second key point coordinates and the second descriptor;

[0106] Step S5.2: Match the first key point coordinates and the first descriptor with the second key point coordinates and the second descriptor through a feature matcher, and obtain the second nickel sheet assembly area in the image to be detected through affine transformation calculation; as a preferred embodiment of this embodiment, the feature matcher is a FLANN feature matcher, which can efficiently process large-scale feature point data sets and provides higher speed and good accuracy in approximate matching.

[0107] Step S5.3: Filter out high-scoring points in the feature matching results, perform affine transformation matrix calculation, and obtain an affine transformation matrix; perform affine transformation calculation on the image to be detected based on the affine transformation matrix to obtain an affine transformation image, wherein the second nickel sheet assembly area in the affine transformation image is consistent with the first nickel sheet assembly area in the reference image.

[0108] Step S6: intercepting the affine transformation image through the ROI area to obtain each second nickel sheet block, and calculating the zero-order moment ( ), first-order moment ( , ), and the second moment ( , , );

[0109] Specifically, step S6 includes:

[0110] Step S6.1: Segmenting the affine transformed image based on the ROI regions to form second nickel sheet blocks;

[0111] Step S6.2: performing binarization algorithm processing on each of the second nickel sheet blocks to form a second binarized block;

[0112] Step S6.3: performing edge extraction on each of the second binarized blocks to form each second edge block;

[0113] Step S6.4: Calculate the zero-order moment, first-order moment and second-order moment of each second edge block. In this embodiment, the calculation method of the zero-order moment, first-order moment and second-order moment of the second edge block is the same as that of the first edge block.

[0114] Step S7: Based on the zero-order moment, first-order moment, and second-order moment of each of the first nickel sheet blocks and the second nickel sheet blocks, the area difference ratio, center of gravity Euclidean distance, and cosine similarity of each of the first nickel sheet blocks and the corresponding second nickel sheet block are calculated to determine the accuracy of nickel sheet assembly of the battery module to be tested.

[0115] Specifically, the step S7 includes:

[0116] Step S7.1: Calculate the area difference ratio, centroid Euclidean distance, and cosine similarity of each of the first nickel sheet blocks and the corresponding second nickel sheet block based on the zero-order moment, first-order moment, and second-order moment of each of the first nickel sheet blocks and the second nickel sheet block;

[0117] The area difference ratio is expressed as:

[0118] ;

[0119] The centroid coordinates of the first nickel sheet block and the second nickel sheet block are calculated according to the first-order moment and the second-order moment:

[0120] The center of gravity coordinates of the first nickel sheet block are:

[0121] (x,y), where x= , y= ;

[0122] The center of gravity coordinates of the second nickel sheet block are:

[0123] (x,y), where x= , y= ;

[0124] According to the centroid coordinates of the first nickel sheet block and the second nickel sheet block, the centroid Euclidean distance of the corresponding nickel sheet is calculated:

[0125] ;

[0126] Calculate the distribution of the first nickel block along the x-axis respectively , distribution along the y-axis Linkage with xy direction :

[0127] ;

[0128] ;

[0129] ;

[0130] Calculate the distribution of the second nickel block along the x-axis respectively , distribution along the y-axis Linkage with xy direction :

[0131] ;

[0132] ;

[0133] ;

[0134] Calculate the cosine similarity between the first nickel sheet block and the second nickel sheet block:

[0135] ;

[0136] in, represents the geometric moment of the kth nickel block, represents the geometric moment of the kth second nickel sheet block;

[0137] Step S7.2: Set an area difference ratio threshold. When the area difference ratio is greater than the area difference ratio threshold, it is considered that the corresponding nickel sheet area deviation is too large; set a center of gravity Euclidean distance threshold. When the center of gravity Euclidean distance is greater than the center of gravity Euclidean distance threshold, it is considered that the corresponding nickel sheet center of gravity deviates too much; set a cosine similarity threshold. When the cosine similarity is less than the cosine similarity threshold, it is considered that the corresponding nickel sheet contour eccentricity is too large.

[0138] In this embodiment, the zero-order moment represents the sum of the grayscale values ​​in the image. In a binary contour image, its physical property is also equivalent to the area of ​​the contour, and its difference ratio can be used to quantify the difference in area. The first-order moment represents the center of gravity of the contour. When the nickel sheet is installed incorrectly, the Euclidean distance will quantify the offset of the center of gravity of the contour. The second-order moment means the eccentricity of the contour, and the cosine similarity is extremely sensitive to direction, and can effectively measure whether the direction of the contour has changed. Therefore, different types of assembly errors of the nickel sheet are effectively quantified through the zero-order moment, the first-order moment and the second-order moment.

[0139] As a further improvement of this embodiment, it also includes:

[0140] Step S8: Determine whether the corresponding nickel sheet is misassembled based on the area difference ratio, center of gravity Euclidean distance, and cosine similarity between each of the first nickel sheet blocks and the second nickel sheet blocks; if the assembly is wrong, render a red frame on the corresponding second nickel sheet block of the affine transformation image; if the assembly is correct, render a green frame on the corresponding second nickel sheet block of the affine transformation image.

[0141] The present embodiment provides a nickel sheet assembly detection method for a battery module, which obtains the nickel sheet assembly area of ​​the correctly installed nickel sheet and the installation block of each nickel sheet by marking a reference image of the nickel sheet assembly structure of a standard battery module, thereby finding the nickel sheet assembly area and the installation block of each nickel sheet on the image to be detected of the nickel sheet assembly structure of the battery module to be detected through affine transformation, and simultaneously calculates the geometric moments of the reference image and the assembly area respectively, thereby being able to determine the area difference ratio, center of gravity Euclidean distance, and cosine similarity of each nickel sheet between the two through zero-order moment, first-order moment, and second-order moment calculations, and qualitatively and quantitatively analyze the accurate coverage ratio, horizontal offset degree, and angular deflection degree of the nickel sheet installation, which can effectively determine the problems and degree of the nickel sheet installation, and is stable and controllable, with low implementation cost.

[0142] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0143] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the patent application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent application shall be subject to the attached claims.

Claims

1. A nickel sheet assembly detection method for a battery module, characterized in that: The method comprises the following steps: Step S1: Acquire a reference image of a standard battery module nickel sheet assembly structure; Step S2: marking the first nickel sheet assembly area in the reference image, and extracting the first key feature point in the first nickel sheet assembly area; The step S2 comprises: Step S2.1: marking the first nickel sheet assembly area in the reference image; Step S2.2: extracting a first key feature point in the first nickel sheet assembly area in the reference image by a feature extraction algorithm, and generating first key point coordinates and a first descriptor; Step S3: marking the ROI region of each nickel sheet in the reference image, and calculating the zero-order moment, first-order moment and second-order moment of the first nickel sheet block corresponding to each ROI region in the reference image; Step S4: collecting the image to be inspected of the nickel sheet assembly structure of the battery module to be inspected; Step S5: performing affine transformation calculation on the image to be detected to obtain an affine transformation image, wherein the second nickel sheet assembly area in the affine transformation image is consistent with the first nickel sheet assembly area in the reference image; The step S5 comprises: Step S5.1: extracting the second key feature point from the image to be detected by a feature extraction algorithm, and generating the second key point coordinates and the second descriptor; Step S5.2: matching the first key point coordinates and the first descriptor with the second key point coordinates and the second descriptor through a feature matcher, and obtaining the second nickel sheet assembly area in the image to be detected through affine transformation calculation; Step S5.3: Filter out high-scoring points in the feature matching results, perform affine transformation matrix calculation, and obtain an affine transformation matrix; perform affine transformation calculation on the image to be detected based on the affine transformation matrix to obtain an affine transformation image, wherein the second nickel sheet assembly area in the affine transformation image is consistent with the first nickel sheet assembly area in the reference image; Step S6: intercepting the affine transformation image through the ROI area to obtain each second nickel sheet block, and calculating the zero-order moment, first-order moment and second-order moment of the second nickel sheet block; Step S7: Based on the zero-order moment, first-order moment, and second-order moment of each of the first nickel sheet blocks and the second nickel sheet blocks, the area difference ratio, center of gravity Euclidean distance, and cosine similarity of each of the first nickel sheet blocks and the corresponding second nickel sheet block are calculated to determine the accuracy of nickel sheet assembly of the battery module to be tested.

2. The nickel sheet assembly detection method for a battery module according to claim 1, characterized in that: The step S1 further comprises: Step S1.1: preprocessing the reference image of the standard battery module nickel sheet assembly structure, wherein the preprocessing includes at least one of Gaussian filtering noise reduction, color space conversion, and contrast enhancement.

3. The nickel sheet assembly detection method for a battery module according to claim 2, characterized in that: The step S3 comprises: Step S3.1: marking the ROI region of each nickel sheet in the reference image, and segmenting the reference image into first nickel sheet blocks based on each ROI region; Step S3.2: performing binarization algorithm processing on each of the first nickel sheet blocks to form a first binarized block; Step S3.3: performing edge extraction on each of the first binarized blocks to form each first edge block; Step S3.4: Calculate the zero-order moment, the first-order moment and the second-order moment of each first edge block.

4. The nickel sheet assembly detection method for a battery module according to claim 3, characterized in that: In step S3.4: The zero-order moment includes M 00 : ； The first-order moment includes M 10 and M 01 : ； ； The second-order moment includes M 11 、M 20 and M 02 : ； ； ； in, represents the grayscale value of the pixel point in the edge block, i represents the horizontal coordinate, and j represents the vertical coordinate.

5. The nickel sheet assembly detection method for a battery module according to claim 4, characterized in that: The step S4 further comprises: Step S4.1: preprocessing the image to be inspected of the nickel sheet assembly structure of the battery module to be inspected, wherein the preprocessing method is consistent with the preprocessing method of the reference image of the nickel sheet assembly structure of the standard battery module.

6. A nickel sheet assembly detection method for a battery module according to claim 5, characterized in that: The step S6 comprises: Step S6.1: Segmenting the affine transformed image into second nickel sheet blocks based on the ROI regions; Step S6.2: performing binarization algorithm processing on each of the second nickel sheet blocks to form a second binarized block; Step S6.3: performing edge extraction on each of the second binarized blocks to form each second edge block; Step S6.4: Calculate the zero-order moment, the first-order moment and the second-order moment of each second edge block.

7. A nickel sheet assembly detection method for a battery module according to claim 6, characterized in that: The step S7 comprises: The step S7 comprises: Step S7.1: Calculate the area difference ratio, centroid Euclidean distance, and cosine similarity of each of the first nickel sheet blocks and the corresponding second nickel sheet block based on the zero-order moment, first-order moment, and second-order moment of each of the first nickel sheet blocks and the second nickel sheet block; The area difference ratio is expressed as: ； The centroid coordinates of the first nickel sheet block and the second nickel sheet block are calculated according to the first-order moment and the second-order moment: The coordinates of the center of gravity of the kth first nickel block are: (x,y), where x= , y= ; The coordinates of the center of gravity of the kth second nickel block are: (x,y), where x= , y= ; According to the centroid coordinates of the kth first nickel sheet block and the kth second nickel sheet block, the centroid Euclidean distance of the corresponding nickel sheet is calculated: ； Calculate the distribution of the kth nickel block along the x-axis respectively , distribution along the y-axis Linkage with xy direction : ； ； ； Calculate the distribution of the kth second nickel block along the x-axis respectively , distribution along the y-axis Linkage with xy direction : ； ； ； Calculate the cosine similarity between the first nickel sheet block and the second nickel sheet block: ； Step S7.2: Set an area difference ratio threshold. When the area difference ratio is greater than the area difference ratio threshold, it is considered that the corresponding nickel sheet area deviation is too large; set a center of gravity Euclidean distance threshold. When the center of gravity Euclidean distance is greater than the center of gravity Euclidean distance threshold, it is considered that the corresponding nickel sheet center of gravity deviates too much; set a cosine similarity threshold. When the cosine similarity is less than the cosine similarity threshold, it is considered that the corresponding nickel sheet contour eccentricity is too large.

8. The nickel sheet assembly detection method for a battery module according to claim 1, characterized in that: Also includes: Step S8: Determine whether the corresponding nickel sheet is misassembled based on the area difference ratio, center of gravity Euclidean distance, and cosine similarity between each of the first nickel sheet blocks and the second nickel sheet blocks; if the assembly is wrong, render a red frame on the corresponding second nickel sheet block of the affine transformation image; if the assembly is correct, render a green frame on the corresponding second nickel sheet block of the affine transformation image.

Images