Insulation shell surface defect image detection method for lithium battery charger

Through multi-scale feature extraction module, high-priority screening module and adaptive filtering sparse defect components, combined with deep convolutional neural network, the problems of low efficiency and insufficient accuracy of traditional detection methods are solved, and the insulated shell surface of lithium battery charger is realized The efficient identification of defects improves the safety and reliability of the charger.

CN119027377AActive Publication Date: 2024-11-26BEIJING KETIANHONGRI TECH CO LTD

Patent Information

Application Number
CN202411032089.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-30
Publication Date
2024-11-26
Estimated Expiration
2044-07-30

AI Technical Summary

Technical Problem

The traditional lithium battery charger's insulated shell surface defect detection method is inefficient and insufficient accuracy, making it difficult to efficiently and accurately identify complex defects, affecting the safety and reliability of the charger.

Method used

A multi-scale feature extraction module, a high-priority screening module and adaptive filtering sparse defect components are adopted, combined with a deep convolutional neural network, a surface defect recognition model for the insulated shell of lithium battery chargers is constructed, and efficient detection is achieved through image processing technology.

Benefits of technology

It significantly improves the accuracy and efficiency of defect identification, can accurately identify the defect location and type of the charger's insulated shell surface, enhances quality control, and improves safety and reliability in the production process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119027377A_ABST
    Figure CN119027377A_ABST
Patent Text Reader

Abstract

The invention provides an insulating shell surface defect image detection method for a lithium battery charger, and belongs to the field of image processing. Through multi-scale feature extraction, high-priority screening and self-adaptive filtering technologies, efficient and accurate detection of the surface defects of the insulating shell of the lithium battery charger is realized; the method comprises the following steps: firstly, shooting and marking defect images on the surface of the insulating shell in a production environment to form a data set; then, inputting a defect image, generating three feature maps with different scales, screening features according to priority value confidence, and only retaining features corresponding to high priority values; secondly, enhancing flexibility and accuracy of feature processing through multi-scale feature priority processing, and enhancing features and boundaries of the region of interest through self-adaptive filtering of sparse defect components; and finally constructing a defect identification model according to the module. According to the method, the efficiency and accuracy of surface defect detection of the lithium battery charger are remarkably improved through deep learning and image processing technologies.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of image processing, and in particular relates to an image detection method for surface defects of an insulating shell of a lithium battery charger. Background Art

[0002] Lithium batteries have been widely used in modern life, especially in electric vehicles, consumer electronics and energy storage systems. As an important accessory of lithium batteries, the quality of lithium battery chargers directly affects the performance and safety of lithium batteries. As a protective component, the surface quality of the insulating shell of lithium battery chargers is particularly critical. Surface defects of the insulating shell, such as scratches, cracks and deformation, not only affect the appearance, but may also pose a hidden danger to the function and safety of the charger. In order to ensure the reliability of the charger, detection methods for surface defects of the insulating shell have emerged and become an indispensable part of the production process.

[0003] Traditional surface defect detection methods mainly rely on manual inspection and simple machine vision technology; however, manual inspection is inefficient and its accuracy is limited by the operator's experience and attention; simple machine vision technology often shows its shortcomings when facing complex surface defects, and it is difficult to efficiently and accurately identify various types of defects; with the rapid development of deep learning and image processing technology, multi-scale feature extraction and adaptive filtering technology have gradually been applied to the field of defect detection, greatly improving the accuracy and efficiency of detection; therefore, the development of a lithium battery charger insulating shell surface defect detection method based on advanced image processing technology has become a hot topic in the industry.

[0004] The present invention proposes a method for detecting surface defects of insulating shells of lithium battery chargers by using images, aiming to realize efficient detection of surface defects of insulating shells of lithium battery chargers by constructing multi-scale feature extraction modules, high priority screening modules and adaptive filtering sparse defect components. The method constructs a special surface defect data set by collecting and annotating image data of the insulating shell surface in a production environment. Subsequently, a deep convolutional neural network is used for feature extraction, and the accuracy and efficiency of defect recognition are gradually improved through steps such as multi-scale feature fusion, priority screening and adaptive filtering. Finally, the trained defect recognition model can accurately identify the location and type of surface defects of the insulating shell of the charger, providing strong technical support for the quality control of the charger. Summary of the invention

[0005] The present invention proposes an image detection method for the surface defects of the insulating shell of a lithium battery charger, aiming to improve the safety and reliability of the lithium battery charger during the production and detection process through efficient and automated image processing technology.

[0006] A method for detecting surface defects of an insulating shell of a lithium battery charger by image detection, comprising the following steps: S1. Preparation of the surface dataset of the insulating shell of a lithium battery charger. In the production environment of a lithium battery charger, the entire surface of a lithium battery charger with defects on the surface of the insulating shell is photographed, and the images obtained by the photography are annotated. The annotated targets are the surface defect areas and five defect types of the insulating shell of the lithium battery charger. After the annotation is completed, the surface dataset of the insulating shell of the lithium battery charger is obtained. S2, construct a multi-scale feature extraction module, input the defect image, and obtain feature maps of three different scales; S3. Build a high priority screening module to screen queries by priority value confidence, and only retain queries corresponding to priority values ​​higher than the priority value confidence; S4. Construct a multi-scale feature priority processing module, so that the priority processing module is more flexible and accurate in processing multi-scale features, thereby improving the performance of defect recognition; S5, construct AFSD adaptive filtering sparse defect components to enhance the features and boundaries of the region of interest; S6. Construct a surface defect recognition model for the insulation shell of a lithium battery charger, including an input module, a multi-scale feature generation module, a high priority screening module, a multi-scale feature priority processing module, a CCRS module, and an AFSD adaptive filtering sparse defect component; S7. Train and detect the surface defect recognition model of the insulating shell of the lithium battery charger. Use the lithium battery charger for detection. Use the insulating shell surface defect dataset of the lithium battery charger to train the insulating shell surface defect recognition model of the lithium battery charger. After the training is completed, input the new insulating shell surface image of the lithium battery charger to be tested into the insulating shell surface defect recognition model of the lithium battery charger to obtain the test results. The test results include the defect location and type in the insulating shell surface image of the lithium battery charger.

[0007] Preferably, in the method for detecting surface defects of an insulating shell of a lithium battery charger, in step S1, five types of defects on the surface of the insulating shell of the lithium battery charger specifically include five major types of defects: scratches on the shell surface, color difference on the shell surface, cracks on the shell surface, bumps on the shell surface and deformation of the shell surface.

[0008] Preferably, the method for detecting surface defects of an insulating shell of a lithium battery charger, in step S2, for the multi-scale feature extraction module, specifically comprises the following steps: A set of images of surface defects of the insulating shell of a lithium battery charger is received, denoted as X, where the size of each image is H×W×C, where H is the height, W is the width, and C is the number of channels, which is equal to 3; feature extraction is performed on each image X, and it is input into a deep convolutional neural network, and the convolution layer and pooling layer in the network extract the initial feature value of the image; forward propagation is performed in the network to extract multi-layer feature maps, and the initial feature map is converted into three scale feature maps F through a feature fusion strategy. l 、F m 、F h , specifically F h =H×W×C3, where C1, C2 and C3 are the new number of channels; the multi-scale feature maps are scaled uniformly by using interpolation and pooling operations to obtain a set of feature maps with the same resolution; the scaled feature map set is input into the feature selection module, important features are screened out through the attention mechanism, and features are reorganized to form the final feature representation.

[0009] Preferably, in the above-mentioned method for detecting surface defects of insulating shell of lithium battery charger, in step S3, the high priority screening module is specifically as follows: Input F obtained by the multi-scale feature generation module l 、F m 、F h Feature maps of three scales, each scale feature contains n query points Q i = {q i1 ,q i2 , ..., q in}, where i represents the scale level; for each query point Q i , calculate its position and related priority value in the grid terminal defect image, and the priority value calculation formula is changed to: Among them, W and b are learnable parameters. By setting the dynamic priority value confidence interval and dynamically adjusting it in the range of 0.6 to 1, query points with priority values ​​higher than the confidence level are screened out to form a feature set S after high priority screening.

[0010] Preferably, in the above-mentioned method for detecting surface defects of insulating shell of lithium battery charger, the multi-scale feature priority processing module in step S4 is specifically designed as follows: In the multi-scale feature priority processing module, the priority value calculation method of the query is redefined. The feature value set S calculated in step S3 is judged, and the queries higher than the set threshold t are called priority queries. The threshold t is dynamically adjusted during the training process to adapt to different feature distributions. An improved attention mechanism is applied to the priority query to enhance its feature representation capability. The formula is as follows: Where Q i is the query point, K i is the key, V i is the value, d k is the scaling factor, T is the setting threshold; The selected priority features S will be given priority in the subsequent processing steps, thereby improving the accuracy and efficiency of defect recognition; other unselected queries will remain unchanged and used as auxiliary information in the subsequent processing stages, thereby improving the overall recognition accuracy and robustness.

[0011] Preferably, in the above-mentioned method for detecting surface defects of insulating shell of lithium battery charger, the AFSD adaptive filtering sparse defect component in step S5 is specifically designed as follows: first, an initial thermal map is generated, and the specific formula is as follows: Especially when the defect is less than the edge When , assuming the margin is L, the window width from the defect to the non-edge side should be kL; therefore, we need to correct it and move the window center position to obtain a window of the same size. The specific formula is: In these two formulas, G is the two-dimensional matrix of the original wafer image; G[i,j] is the value at the coordinate (i,j) in the heat map, which represents the defect density in the area; k is the size of the window of interest; [m,n] is the coordinate in the window of interest, traversing the pixels near i and j; then the density eigenvalue mapping operation is performed, and the specific formula is: G[i,j]=(G[i,j]-min(G))·Dist; Among them, Dist is the density eigenvalue scaling factor. The threshold Q is determined by the cumulative number of regions with different types of defects in the production batch. Finally, the sparse defect areas are filtered out and the eigenvalues ​​of the dense defect areas are retained. The specific formula is as follows: Among them, G[i,j] is the final thermal value and Q is the adaptive density threshold.

[0012] Preferably, the insulating shell surface defect image detection method for a lithium battery charger constructs an insulating shell surface defect recognition network model for the lithium battery charger in step S6, firstly inputs the grid terminal image into a multi-scale feature generation module to generate features of three different scales; then, inputs these features into a high priority screening module to screen them in three scales by priority value confidence; subsequently, inputs the screened features into multiple multi-scale feature priority processing modules connected in series to perform multiple high priority screening and self-attention calculations; inputs the detected defect area into the AFSD adaptive filtering sparse defect component to perform defect type classification operations; and finally outputs the insulating shell surface defect position and defect type information of the lithium battery charger.

[0013] Compared with the prior art, the method for detecting defects on the surface of the insulating shell of a lithium battery charger proposed in the present invention has the following advantages: the present invention adopts deep learning and multi-scale feature extraction technology to construct a deep convolutional neural network, which can efficiently extract and accurately identify various defect types; the multi-scale feature extraction module of the present invention generates feature maps of different scales, thereby enhancing the ability to identify different defects; the high priority screening module improves detection accuracy by screening by priority value confidence; the AFSD adaptive filtering sparse defect component effectively filters unimportant defect areas, thereby improving detection accuracy and efficiency; the present invention is significantly superior to the prior art in terms of detection efficiency, accuracy and adaptability to complex environments, and provides a more reliable defect detection solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The present invention is a flow chart of an image detection method for surface defects of an insulating shell of a lithium battery charger.

[0015] Figure 2 Flowchart of screening and processing for multi-scale feature extraction and prioritization.

[0016] Figure 3 Schematic diagram of the network for the surface defect identification model of the insulating shell of the lithium battery charger constructed.

[0017] Figure 4 This is a schematic diagram of the output results of the detection model for the surface defect recognition of the insulating shell of a lithium battery charger. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0019] See also Figure 1-Figure 4 , the present invention provides a technical solution.

[0020] The present invention proposes an image detection method for the surface defects of the insulating shell of a lithium battery charger, aiming to improve the safety and reliability of the lithium battery charger during the production and detection process through efficient and automated image processing technology.

[0021] A method for detecting surface defects of an insulating shell of a lithium battery charger using an image, the process is as follows: Figure 1 As shown, the specific steps include: S1. Preparation of the surface dataset of the insulating shell of a lithium battery charger. In the production environment of a lithium battery charger, the entire surface of a lithium battery charger with defects on the surface of the insulating shell is photographed, and the images obtained by the photography are annotated. The annotated targets are the surface defect areas and five defect types of the insulating shell of the lithium battery charger. After the annotation is completed, the surface dataset of the insulating shell of the lithium battery charger is obtained. S2, construct a multi-scale feature extraction module, input the defect image, and obtain feature maps of three different scales; S3. Build a high priority screening module to screen queries by priority value confidence, and only retain queries corresponding to priority values ​​higher than the priority value confidence; S4. Construct a multi-scale feature priority processing module, so that the priority processing module is more flexible and accurate in processing multi-scale features, thereby improving the performance of power grid terminal defect identification; S5, construct AFSD adaptive filtering sparse defect components to enhance the features and boundaries of the region of interest; S6. Construct a surface defect recognition model for the insulation shell of a lithium battery charger, including an input module, a multi-scale feature generation module, a high priority screening module, a multi-scale feature priority processing module, a CCRS module, and an AFSD adaptive filtering sparse defect component; S7. Train and detect the surface defect recognition model of the insulating shell of the lithium battery charger. Use the lithium battery charger for detection. Use the insulating shell surface defect dataset of the lithium battery charger to train the insulating shell surface defect recognition model of the lithium battery charger. After the training is completed, input the new insulating shell surface image of the lithium battery charger to be tested into the insulating shell surface defect recognition model of the lithium battery charger to obtain the test results. The test results include the defect location and type in the insulating shell surface image of the lithium battery charger.

[0022] Furthermore, in step S1 of a method for detecting surface defects of an insulating shell of a lithium battery charger, the five types of surface defects of the insulating shell of the lithium battery charger specifically include five major types of defects: scratches on the shell surface, color difference on the shell surface, cracks on the shell surface, bumps on the shell surface and deformation on the shell surface.

[0023] Further, in step S2 of a method for detecting surface defects of an insulating shell of a lithium battery charger, the multi-scale feature extraction module is as follows: Figure 2 As shown on the left, the following steps are included: A set of grid terminal defect images is received, denoted as X, where the size of each image is H×W×C, where H is the height, W is the width, and C is the number of channels, which is equal to 3; feature extraction is performed on each image X, and it is input into a deep convolutional neural network, where the convolution layer and pooling layer in the network extract the initial feature value of the image; forward propagation is performed in the network to extract multi-layer feature maps, and the initial feature map is converted into three scale feature maps F through a feature fusion strategy. l 、F m 、F h , specifically F h =H×W×C3, where C1, C2 and C3 are the new number of channels; the multi-scale feature maps are scaled uniformly by using interpolation and pooling operations to obtain a set of feature maps with the same resolution; the scaled feature map set is input into the feature selection module, important features are screened out through the attention mechanism, and features are reorganized to form the final feature representation.

[0024] Further, in step S3 of a method for detecting surface defects of an insulating shell of a lithium battery charger, the high priority screening module is as follows: Figure 2 As shown in the middle, the specific design is as follows: Input F obtained by the multi-scale feature generation module l 、F m 、F h Feature maps of three scales, each scale feature contains n query points Q i = {q i1 ,q i2 , ..., q in}, where i represents the scale level; for each query point Q i , calculate its position and related priority value in the grid terminal defect image, and the priority value calculation formula is changed to: Among them, W and b are learnable parameters. By setting the dynamic priority value confidence interval and dynamically adjusting it in the range of 0.6 to 1, query points with priority values ​​higher than the confidence level are screened out to form a feature set S after high priority screening.

[0025] Further, in step S4 of a method for detecting surface defects of an insulating shell of a lithium battery charger, the multi-scale feature priority processing module is as follows: Figure 2As shown on the right, the specific design is as follows: In the multi-scale feature priority processing module, the priority value calculation method of the query is redefined. The feature value set S calculated in step S3 is judged, and the queries higher than the set threshold t are called priority queries. The threshold t is dynamically adjusted during the training process to adapt to different feature distributions. An improved attention mechanism is applied to the priority query to enhance its feature representation capability. The formula is as follows: Where Q i is the query point, K i is the key, V i is the value, d k is the scaling factor, T is the setting threshold; The selected priority features S will be given priority in the subsequent processing steps, thereby improving the accuracy and efficiency of defect recognition; other unselected queries will remain unchanged and used as auxiliary information in the subsequent processing stages, thereby improving the overall recognition accuracy and robustness.

[0026] Furthermore, in step S5 of the method for detecting surface defects of an insulating shell of a lithium battery charger, the AFSD adaptive filtering sparse defect component is specifically designed as follows: first, an initial thermal map is generated, and the specific formula is as follows: Especially when the defect is less than the edge When , assuming the margin is L, the window width from the defect to the non-edge side should be kL; therefore, we need to correct it and move the window center position to obtain a window of the same size. The specific formula is: In these two formulas, G is the two-dimensional matrix of the original wafer image; G[i,j] is the value at the coordinate (i,j) in the heat map, which represents the defect density in the area; k is the size of the window of interest; [m,n] is the coordinate in the window of interest, traversing the pixels near i and j; then the density eigenvalue mapping operation is performed, and the specific formula is: G[i,j]=(G[i,j]-min(G))·Dist; Among them, Dist is the density eigenvalue scaling factor. The threshold Q is determined by the cumulative number of regions with different types of defects in the production batch. Finally, the sparse defect areas are filtered out and the eigenvalues ​​of the dense defect areas are retained. The specific formula is as follows: Among them, G[i,j] is the final thermal value and Q is the adaptive density threshold.

[0027] Further, in step S6 of a method for detecting surface defects of an insulating shell of a lithium battery charger, as Figure 3 The figure shows a schematic diagram of the designed and constructed network model for surface defect recognition of the insulating shell of a lithium battery charger. The specific process is as follows: first, the grid terminal image is input into the multi-scale feature generation module to generate features of three different scales; then, these features are input into the high priority screening module to screen them in three scales by the priority value confidence; subsequently, the screened features are input into multiple multi-scale feature priority processing modules connected in series to perform multiple high priority screening and self-attention calculations; the detected defect area is input into the AFSD adaptive filtering sparse defect component to perform defect type classification operations; and finally, the defect location and defect type information of the insulating shell surface of the lithium battery charger are output.

[0028] Furthermore, the image detection results of the surface defects of the insulating shell of the lithium battery charger are as follows: Figure 4 As shown, it can be seen that the insulating shell surface defect detection model of the lithium battery charger can well locate the defect area on the insulating shell surface of the lithium battery charger and output accurate defect classification results on the program side.

Claims

1. A method for detecting surface defects of insulating shell of lithium battery charger using image. It is characterized in that The following steps are involved: S1. Preparation of the surface dataset of the insulating shell of a lithium battery charger. In the production environment of a lithium battery charger, the entire surface of a lithium battery charger with defects on the surface of the insulating shell is photographed, and the images obtained by the photography are annotated. The annotated targets are the surface defect areas and five defect types of the insulating shell of the lithium battery charger. After the annotation is completed, the surface dataset of the insulating shell of the lithium battery charger is obtained. S2, construct a multi-scale feature extraction module, input the defect image, and obtain feature maps of three different scales; S3. Build a high priority screening module to screen queries by priority value confidence, and only retain queries corresponding to priority values ​​higher than the priority value confidence; S4. Construct a multi-scale feature priority processing module, so that the priority processing module is more flexible and accurate in processing multi-scale features, thereby improving the performance of surface defect recognition of the insulating shell of the lithium battery charger; S5, construct AFSD adaptive filtering sparse defect components to enhance the features and boundaries of the region of interest; S6. Construct a surface defect recognition model for the insulation shell of a lithium battery charger, including an input module, a multi-scale feature generation module, a high priority screening module, a multi-scale feature priority processing module, a CCRS module, and an AFSD adaptive filtering sparse defect component; S7. Train and detect the surface defect recognition model of the insulating shell of the lithium battery charger. Use the lithium battery charger for detection. Use the insulating shell surface defect dataset of the lithium battery charger to train the insulating shell surface defect recognition model of the lithium battery charger. After the training is completed, input the new insulating shell surface image of the lithium battery charger to be tested into the insulating shell surface defect recognition model of the lithium battery charger to obtain the test results. The test results include the defect location and type in the insulating shell surface image of the lithium battery charger.

2. The method for detecting surface defects of an insulating shell of a lithium battery charger according to claim 1, It is characterized in that In step S1, the five types of surface defects of the insulating shell of the lithium battery charger include scratches on the shell surface, color difference on the shell surface, cracks on the shell surface, bumps on the shell surface and deformation of the shell surface.

3. The method for detecting surface defects of an insulating shell of a lithium battery charger according to claim 2, It is characterized in that In step S2, the multi-scale feature extraction module specifically includes the following steps: receiving a group of images of surface defects of the insulating shell of a lithium battery charger, denoted as X, wherein the size of each image is H×W×C, wherein H is the height, W is the width, and C is the number of channels equal to 3; extracting features from each image X, and inputting it into a deep convolutional neural network, wherein the convolutional layer and the pooling layer in the network extract initial feature values ​​from the image; performing forward propagation in the network, extracting multi-layer feature maps, and converting the initial feature maps into three scale feature maps F through a feature fusion strategy. l 、F m 、F h , specifically F h =H×W×C 3 , where C 1 , C 2 and C 3 is the new number of channels; the multi-scale feature maps are scaled uniformly by interpolation and pooling operations to obtain a set of feature maps with the same resolution; the set of feature maps after unification is input into the feature selection module, important features are screened out through the attention mechanism, and features are reorganized to form the final feature representation.

4. The method for detecting surface defects of an insulating shell of a lithium battery charger according to claim 3, It is characterized in that In step S3, the high priority screening module is specifically designed as follows: Input F obtained by the multi-scale feature generation module l 、F m 、F h Feature maps of three scales, each scale feature contains n query points Q i = {q i1 ,q i2 , ..., q in }, where i represents the scale level; for each query point Q i , calculate its position and related priority value in the surface defect picture of the insulating shell of the lithium battery charger, and the priority value calculation formula is changed to: Among them, W and b are learnable parameters. By setting the dynamic priority value confidence interval and dynamically adjusting it in the range of 0.6 to 1, query points with priority values ​​higher than the confidence level are screened out to form a feature set S after high priority screening.

5. The method for detecting surface defects of an insulating shell of a lithium battery charger according to claim 4, It is characterized in that The multi-scale feature priority processing module described in step S4 is specifically designed as follows: In the multi-scale feature priority processing module, the priority value calculation method of the query is redefined. The feature value set S calculated in step S3 is judged, and the queries higher than the set threshold t are called priority queries. The threshold t is dynamically adjusted during the training process to adapt to different feature distributions. An improved attention mechanism is applied to the priority query to enhance its feature representation capability. The formula is as follows: Where Q i is the query point, K i is the key, V i is the value, d k is the scaling factor, T is the setting threshold; The selected priority features S will be given priority in the subsequent processing steps, thereby improving the accuracy and efficiency of defect recognition; other unselected queries will remain unchanged and used as auxiliary information in the subsequent processing stages, thereby improving the overall recognition accuracy and robustness.

6. A method for detecting surface defects of an insulating shell of a lithium battery charger according to claim 5, It is characterized in that The AFSD adaptive filtering sparse defect component described in step S5 is specifically designed as follows: First, an initial thermal map is generated, and the specific formula is as follows: Especially when the defect is less than the edge When , assuming the margin is L, the window width from the defect to the non-edge side should be kL; therefore, we need to correct it and move the window center position to obtain a window of the same size. The specific formula is: In these two formulas, G is the two-dimensional matrix of the original wafer image; G[i,j] is the value at the coordinate (i,j) in the heat map, which represents the defect density in the area; k is the size of the window of interest; [m,n] is the coordinate in the window of interest, traversing the pixels near i and j; then the density eigenvalue mapping operation is performed, and the specific formula is: G[i,j]=(G[i,j]-min(G))·Dist; Among them, Dist is the density eigenvalue scaling factor. The threshold Q is determined by the cumulative number of regions with different types of defects in the production batch. Finally, the sparse defect areas are filtered out and the eigenvalues ​​of the dense defect areas are retained. The specific formula is as follows: Among them, G[i,j] is the final thermal value and Q is the adaptive density threshold.

7. A method for detecting surface defects of an insulating shell of a lithium battery charger according to claim 6, It is characterized in that The insulating shell surface defect recognition network model of the lithium battery charger constructed in step S6 first inputs the insulating shell surface image of the lithium battery charger into the multi-scale feature generation module to generate features of three different scales; then, these features are input into the high priority screening module to screen them in three scales by the priority value confidence; then, the screened features are input into the multi-scale feature priority processing module in series to perform multiple high priority screening and self-attention calculations; the detected defect area is input into the AFSD adaptive filtering sparse defect component to perform defect type classification operations; and finally, the insulating shell surface defect position and defect type information of the lithium battery charger are output.

Citation Information

Patent Citations

  • Steel plate surface defect classification method based on double-flow neural network model

    CN111627018A

  • Real-time high-precision detection method for appearance defects of power adapter

    CN113205136A

  • Improved YOLOv5-based battery case end face defect real-time detection method

    CN114972316A

  • Face Authentication Using Recognition-by-Parts, Boosting, and Transduction

    US20110135166A1

  • Multi-sensor test device for quality control scanning

    US20240068952A1

Cited By

  • Lithium battery shell defect detection method based on machine vision

    CN121169917A