Method and system for measuring specific volume value of corroded foil

Abstract

The invention provides a method and system for measuring the specific volume value of a corrosion foil, and relates to the technical field of corrosion foil performance detection.The method includes the steps that a corrosion foil sample is cut, the surface of the corrosion foil sample is cleaned, then collected images are preprocessed, and the boundary contour of each corrosion hole is extracted from the preprocessed images; the method comprises the following steps of: calculating the geometric characteristic parameters of holes, introducing a parallel plate capacitance theory, converting the geometric parameter of each corrosion hole into a corresponding capacitance value by utilizing a formula, further calculating the local specific volume of each hole, then accumulating and averaging the local specific volumes of all holes in an image region, and calculating to obtain the average specific volume value of the region; and finally, constructing a prediction model, taking a plurality of area images as input, taking the corresponding average specific volume value as an output label, and after training is completed, the model can directly perform reasoning prediction on the etched foil image, so that accurate prediction of the specific volume value of the unit area of the whole etched foil is realized under the condition that a formation experiment is not needed.

Description

Technical Field

[0001] This invention relates to the technical field, specifically to a method and system for determining the specific capacitance value of corroded foil. Background Technology

[0002] Etched foil is a functional material widely used in aluminum electrolytic capacitors, and its performance directly affects the energy storage capacity and stability of the capacitor. Among the performance indicators of etched foil, the specific capacitance per unit area is a core parameter for evaluating its quality and manufacturing process. Currently, the specific capacitance of etched foil is commonly tested industrially using electrochemical methods. This method mainly obtains the specific capacitance value by applying voltage to the etched foil sample and measuring its charge-discharge characteristics. However, this type of testing method has obvious limitations: First, the testing process is cumbersome, requiring the sample to be connected to the test circuit and immersed in the electrolyte for a period of time, which is not only time-consuming but also cannot achieve online testing; second, the testing process requires physical contact with the electrodes, which is easily affected by factors such as poor contact of the clamps and fluctuations in electrolyte concentration, causing measurement errors; in addition, for large-area etched foil materials, multiple sampling points and averaging are required, further increasing the risk of human intervention and error accumulation.

[0003] With the development of image processing technology, current methods mostly focus on the two-dimensional features of corrosion pores, lacking consideration for the key parameters affecting specific volume—corrosion pore depth and three-dimensional distribution structure. While this image analysis method, which ignores corrosion pore depth, can achieve a certain degree of non-contact assessment, it is difficult to align with traditional physical measurement results, limiting its reliability and application scope in actual industrial testing. More importantly, most of these methods fail to establish a physical mapping relationship from microstructural features to macroscopic specific volume values, and also fail to form a generalizable specific volume prediction model, resulting in the inability to achieve generalized application under new samples, new processes, or complex image conditions.

[0004] In the prior art, CN11814831A discloses a method for determining the specific capacitance value of etched foil. This method involves acquiring the original image of the etched foil, adjusting its size and normalizing it to form standardized image data. The actual specific capacitance value of the corresponding etched foil sample is then measured using traditional formation experiments. The image and physical test data are matched one-to-one to form image-specific capacitance pairs. However, the image data only relies on the two-dimensional morphological features of the surface and lacks modeling of key factors affecting specific capacitance—the depth of the etched holes and the three-dimensional structure. The training set and test set are randomly divided according to a ratio for model building and performance verification. A deep learning architecture based on convolutional neural networks is adopted, specifically including multiple convolutional layers, pooling layers, activation function layers, batch normalization layers, fully connected layers, and flattening layers. However, convolutional neural networks have a large number of parameters, and the training process consumes high computational resources, which is not conducive to real-time deployment and expansion in a production environment. During network compilation, training parameters such as loss function, optimizer, learning rate, and performance evaluation function are set. During network training, the learning rate is automatically adjusted through callback functions, and the model parameters are continuously optimized using the training set. Finally, the model performance is evaluated using the test set. After training is completed, the network model that passes the evaluation is selected as the final volume prediction model.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a method and system for determining the specific capacitance of etched foil, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A method for determining the specific capacitance of etched foil, comprising the following steps: Step 1: Cut and divide the etched foil into regions and clean it. Collect images of the etched holes on the surface of the etched foil. Preprocess the images of the etched holes. The preprocessing includes using Gaussian filtering to remove noise from the images of the etched holes and using histogram equalization to enhance the contrast of the filtered images of the etched holes. Step 2: Extract each preprocessed corrosion hole image, use edge detection to extract the boundary contour of the enhanced corrosion hole image, use pixel integral of the connected component of the binary image to calculate the corrosion hole area, and use grayscale linear mapping to calculate the depth. Step 3: Based on the parallel plate capacitance theory, convert the geometric parameters of each etched hole into a single hole capacitance value, and calculate the local specific capacitance of the single hole based on the single hole capacitance value. Step 4: Use a weighted average to accumulate the local volume ratio of all holes to obtain the average volume ratio of the entire image region. Repeat the calculation of the average volume ratio for each cropped region. Step 5: Construct a non-destructive prediction model for the specific capacitance of the etched foil. Take multiple region images of the etched foil as input, use the calculated average specific capacitance of each region of the etched foil as a label to train the prediction model, and output the specific capacitance per unit area of ​​the entire etched foil.

[0008] Further, the etched hole image is preprocessed, specifically through the following steps: Gaussian filtering is used to denoise the images of corrosion holes collected on the surface of the corrosion foil. By constructing a two-dimensional Gaussian kernel and adjusting the weight distribution according to the standard deviation, a weighted average is performed on each pixel in the image through convolution to suppress high-frequency noise interference. Histogram equalization is then performed on the denoised image. By statistically analyzing the pixel gray-level distribution histogram and calculating the cumulative distribution function, the original gray-level values ​​are mapped to the dynamic range, stretching the gray-level difference between the holes and the background in the image and enhancing the contrast.

[0009] Furthermore, edge detection is used to extract the boundary contours of the enhanced erosion hole image. The specific steps are as follows: The Sobel operator is used to calculate the gradient, where the formula for calculating the horizontal gradient is: ; The formula for calculating the gradient in the vertical direction is: ; In the formula, This is the pixel value matrix of the input image; Represents the gradient in the horizontal direction; Represents the gradient in the vertical direction; The formulas for calculating gradient magnitude and direction are as follows: ; In the formula, The gradient magnitude of the image at the current point. The angle of the gradient direction of the current pixel; By preserving the local maximum pixels along the gradient direction, region filling is performed on the image to form a binary mask image of the hole region. The calculation formula is as follows: ; in, Represents pixels Binary mask image at the location.

[0010] Furthermore, the area of ​​the eroded holes is calculated using pixel integration of the connected components of a binary image, and the depth is calculated using gray-level linear mapping. Specific steps include: A specific etched hole is defined as a connected region in a binary image, and its area is the total number of pixels in the region, calculated using the following formula: ; In the formula, For the image belonging to the first Pixel coordinates of the etched hole region For the first A connected region in the image containing eroded holes. This refers to the physical area corresponding to a single pixel. The depth of the etched holes is calculated through grayscale inversion, assuming the original image grayscale value is... The calculation formula is: ; In the formula, This represents the maximum grayscale value of the background region in the image. The physical depth corresponding to a unit grayscale difference. For the first The number of pixels in each etched hole.

[0011] Furthermore, based on the parallel plate capacitance theory, the geometric parameters of each etched hole are converted into a single-hole capacitance value, and the local specific capacitance of a single hole is calculated based on the single-hole capacitance value. The specific formula of the parallel plate capacitance theory is as follows: ; In the formula, The vacuum permittivity, The relative permittivity of the medium inside the hole; The formula for converting the capacitance value of a single hole into the local specific capacitance of a single hole is as follows: ; In the formula, For the first The local specific volume of a corrosion pit. Let be the geometric volume of the hole; The method for calculating the volume of the hole is as follows: ; In the formula, and These represent the area and depth of the hole, respectively.

[0012] Furthermore, the average volume ratio is repeatedly calculated for each cropped region, where the formula for calculating the average volume ratio of the image region is: ; In the formula, This represents the total number of corrosion pores detected within the image area.

[0013] Furthermore, a non-destructive prediction model for the specific capacitance of the etched foil is constructed. Multiple region images of the etched foil are used as input, and the calculated average specific capacitance value of each region of the etched foil is used as a label to train the prediction model. The output is the specific capacitance value per unit area of ​​the entire etched foil. The specific training process of the non-destructive prediction model for the specific capacitance of the etched foil is as follows: the preprocessed image samples are divided into training set, validation set and test set; the images of the training set and validation set are used as input to the prediction model, and the average specific capacitance of each region of the etched foil is calculated as a label to train the prediction model; After training, the non-destructive prediction model for the specific volume of the etched foil is input into the entire image of the etched foil and outputs the overall unit area specific volume prediction value of the etched foil.

[0014] The present invention also provides a system for determining the specific capacitance value of an etched foil, the system being used to perform the above-described method for determining the specific capacitance value of an etched foil, comprising: The image preprocessing module is used to cut and divide the etched foil into regions and clean it, acquire images of the etched holes on the surface of the etched foil, and preprocess the images of the etched holes. The preprocessing includes using Gaussian filtering to remove noise from the images of the etched holes and using histogram equalization to enhance the contrast of the filtered images of the etched holes. The corrosion hole geometry extraction module is used to extract each corrosion hole image after preprocessing. It uses edge detection to extract the boundary contour of the enhanced corrosion hole image, uses pixel integral of the connected component of the binary image to calculate the corrosion hole area, and uses grayscale linear mapping to calculate the depth. The single-hole capacitance calculation module is used to convert the geometric parameters of each corroded hole into a single-hole capacitance value based on the parallel plate capacitance theory, and to calculate the local specific capacitance of a single hole based on the single-hole capacitance value. The region volume ratio accumulation module is used to accumulate the local volume ratio of all holes using a weighted average to obtain the average volume ratio value of the entire image region. The average volume ratio value is calculated repeatedly for each cropped region. The specific capacity prediction modeling module is used to construct a non-destructive prediction model for the specific capacity value of the etched foil. It takes multiple region images of the etched foil as input, uses the calculated average specific capacity value of each region of the etched foil as a label to train the prediction model, and outputs the specific capacity value per unit area of ​​the entire etched foil.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention involves cutting and cleaning the surface of a etched foil sample, then preprocessing the acquired images. The boundary contours of each etched hole are extracted from the preprocessed images, and the geometric feature parameters of the holes, such as area, perimeter, and depth, are calculated. Parallel plate capacitance theory is introduced, and the geometric parameters of each etched hole are converted into corresponding capacitance values ​​using formulas. Furthermore, the local specific capacitance of each hole is calculated. Then, the local specific capacitance of all holes within the image region is accumulated and averaged to calculate the average specific capacitance value of the region. Finally, a prediction model is constructed. Using multiple regional images as input and the corresponding average specific capacitance value as the output label, after training, the model can directly perform inference and prediction on etched foil images, thus achieving accurate prediction of the specific capacitance per unit area of ​​the entire etched foil without the need for formation experiments. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the overall method flow of the present invention; Figure 2 This is a graph showing the changes in volume parameters and average grayscale value of a conventional technical solution in an embodiment of the present invention. Figure 3 This is a graph showing the variation of specific capacitance value in a conventional technical solution according to an embodiment of the present invention; Figure 4 This is a graph showing the changes in volume parameters and average grayscale values ​​of the present invention. Figure 5 This is a graph showing the variation of specific volume values ​​in this invention; Figure 6 This is a comparison chart of the specific capacitance value changes between the conventional technical solution and this embodiment of the invention; Figure 7 This is a block diagram of the overall system structure of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0018] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0019] Example: Please see Figures 1 to 6 The present invention provides a technical solution: A method for determining the specific capacitance of etched foil, comprising the following steps: Step 1: Cut and divide the etched foil into regions and clean it. Collect images of the etched holes on the surface of the etched foil. Preprocess the images of the etched holes. The preprocessing includes using Gaussian filtering to remove noise from the images of the etched holes and using histogram equalization to enhance the contrast of the filtered images of the etched holes. The etched foil is cut and cleaned, and images of the corrosion holes on the surface of the etched foil are acquired. The corrosion hole images are preprocessed, specifically: the etched foil is cut into multiple small square areas; the surface of the etched foil is ultrasonically cleaned; and high-resolution images of the corrosion holes on the surface of the etched foil are acquired. The preprocessing steps for the etched hole image are as follows: Gaussian filtering is used to denoise the images of corrosion holes collected on the surface of the corrosion foil. By constructing a two-dimensional Gaussian kernel and adjusting the weight distribution according to the standard deviation, a weighted average is performed on each pixel in the image through convolution to suppress high-frequency noise interference. Histogram equalization is then performed on the denoised image. By statistically analyzing the pixel gray-level distribution histogram and calculating the cumulative distribution function, the original gray-level values ​​are mapped to the dynamic range, stretching the gray-level difference between the holes and the background in the image and enhancing the contrast.

[0020] Step 2: Extract each preprocessed corrosion hole image, use edge detection to extract the boundary contour of the enhanced corrosion hole image, use pixel integral of the connected component of the binary image to calculate the corrosion hole area, and use grayscale linear mapping to calculate the depth. The specific steps for extracting the boundary contours of the enhanced erosion hole image using edge detection are as follows: The Sobel operator is used to calculate the gradient, where the formula for calculating the horizontal gradient is: ; The formula for calculating the gradient in the vertical direction is: ; In the formula, This is the pixel value matrix of the input image; Represents the gradient in the horizontal direction; Indicates the gradient in the vertical direction; The formulas for calculating gradient magnitude and direction are: ; In the formula, The gradient magnitude of the image at the current point. The angle of the gradient direction of the current pixel; By preserving the local maximum pixels along the gradient direction, region filling is performed on the image to form a binary mask image of the hole region. The calculation formula is as follows: ; in, Represents pixels Binary mask image at the location.

[0021] The method of calculating the area of ​​eroded holes using pixel integration of connected components in a binary image and calculating the depth using grayscale linear mapping includes the following specific steps: A specific etched hole is defined as a connected region in a binary image, and its area is the total number of pixels in the region, calculated using the following formula: ; In the formula, For the image belonging to the first Pixel coordinates of the etched hole region For the first A connected region in the image containing eroded pits. This refers to the physical area corresponding to a single pixel. The depth of the etched holes is calculated through grayscale inversion, assuming the original image grayscale value is... The calculation formula is: ; In the formula, This represents the maximum grayscale value of the background region in the image. The physical depth corresponding to a unit grayscale difference. For the first The number of pixels in each etched hole.

[0022] Step 3: Based on the parallel plate capacitance theory, convert the geometric parameters of each etched hole into a single hole capacitance value, and calculate the local specific capacitance of the single hole based on the single hole capacitance value. Based on the parallel plate capacitance theory, the geometric parameters of each etched hole are converted into a single-hole capacitance value. Then, the local specific capacitance of a single hole is calculated based on this single-hole capacitance value. The specific formula for the parallel plate capacitance theory is as follows: ; In the formula, The vacuum permittivity, The relative permittivity of the medium inside the hole; The formula for converting the capacitance value of a single hole into the local specific capacitance of a single hole is as follows: ; In the formula, For the first The local specific volume of a corrosion pit. Let be the geometric volume of the hole; The method for calculating the volume of the hole is as follows: ; In the formula, and These represent the area and depth of the hole, respectively.

[0023] 6. The method for determining the specific capacitance value of etched foil according to claim 1, characterized in that: the average specific capacitance value is repeatedly calculated for each region after cropping, wherein the formula for calculating the average specific capacitance value of the image region is: ; In the formula, This represents the total number of corrosion pores detected within the image area.

[0024] Step 5: Construct a non-destructive prediction model for the specific capacitance of the etched foil. The images of multiple regions of the etched foil are used as input. The average specific capacitance of each region of the etched foil is calculated as a label to train the prediction model. The output is the specific capacitance per unit area of ​​the entire etched foil.

[0025] A non-destructive prediction model for the specific capacitance of etched foil is constructed. Multiple region images of the etched foil are used as input, and the calculated average specific capacitance of each region of the etched foil is used as a label to train the prediction model. The output is the specific capacitance per unit area of ​​the entire etched foil. The specific training process of the non-destructive prediction model for the specific capacitance of the etched foil is as follows: the preprocessed image samples are divided into training set, validation set and test set; the images of the training set and validation set are used as input to the prediction model, and the average specific capacitance of each region of the etched foil is calculated as a label to train the prediction model; After training, the non-destructive prediction model for the specific volume of the etched foil is input into the entire image of the etched foil and outputs the overall unit area specific volume prediction value of the etched foil.

[0026] Table 1: Statistical Table of Data Measured Using Traditional Techniques

[0027] like Figures 2-3 As shown, the average grayscale value of the image gradually increases with the increase of the volume parameter, especially when the area parameter is higher than 1000, the grayscale value is concentrated between 110 and 130. This indicates that the image grayscale value is related to the light transmittance of the aperture structure, and the increase in area reflects more light transmission paths. However, the perimeter distribution is more dense, mostly concentrated in the area with lower volume parameters, and the corresponding grayscale value changes are not significant. This shows that the perimeter of the display contour cannot effectively reflect the true optical characteristics or depth structure of the aperture. In the range of specific capacitance value between 55 and 58 μF / cm³, the area parameter fluctuates drastically between 1200 and 1400, while the total capacitance is relatively concentrated between 750 and 850 μF. This phenomenon indicates that there is no linear or stable mapping between area and specific capacitance. Especially in high-area samples, there are still low capacitance values, indicating that the image area seriously overestimates the effective energy storage area when blind apertures or non-through apertures are present, making the specific capacitance calculated by empirical formulas unreliable.

[0028] Table 2: Statistical Table of Image Recognition and Modeling Prediction Methods

[0029] like Figures 4-5 As shown, the volume parameter gradually increases from low to high, with the average gray value mainly distributed between 90 and 130. The average gray value generally shows an upward trend with increasing volume. The three features—depth, perimeter, and area—are represented by different horizontal lines in the figure, clearly corresponding to the gray-level response of each image group. Points corresponding to high volume parameters are mostly concentrated in the high gray-level range, indicating that the gray-level features extracted in this scheme have a significant correlation with the three-dimensional structure volume, demonstrating the effectiveness of image semantic information in reflecting microstructure. The performance of image gray-level average and area features in the volumetric prediction process is shown. The horizontal axis represents the area parameter, and the vertical axis represents the image gray-level average. The points show a clear clustering trend. It can be seen that in areas with an area greater than 1200, the average gray value is concentrated between 110 and 130, exhibiting strong concentration and stability, reflecting the consistency between image gray-level and pore structure features. This embodiment uses the image gray-level average as a bridge to fuse volume parameters to achieve lossless modeling of volumetric parameters, enhancing the physical interpretability and accuracy of the prediction, demonstrating the significant advantages of the fusion modeling method.

[0030] like Figure 6As shown, the specific volume value measured by the traditional method fluctuated significantly in 35 samples, especially dropping sharply to 50 at point 10, and continuing to deteriorate to 50 after point 35. This reflects the sensitivity of the traditional process to fluctuations in the concentration of the corrosive solution and the merging of pores. In contrast, this embodiment accurately obtains the area and depth of pores through image preprocessing and boundary contour extraction, avoiding errors from manual measurement. The data range of the image recognition and modeling method is only 6, and the standard deviation is reduced by 40%. In the range of 20-30, the traditional value fluctuates by ±3 due to pore overlap, while this embodiment compresses the fluctuation to ±1 by accumulating the specific volume of a single pore and generalizing the model.

[0031] like Figure 7 As shown, the present invention also provides a system for determining the specific capacitance value of an etched foil, the system for determining the specific capacitance value of an etched foil being used to perform the above-described method for determining the specific capacitance value of an etched foil, comprising: The image preprocessing module is used to cut and divide the etched foil into regions and clean it, acquire images of the etched holes on the surface of the etched foil, and preprocess the images of the etched holes. The preprocessing includes using Gaussian filtering to remove noise from the images of the etched holes and using histogram equalization to enhance the contrast of the filtered images of the etched holes. The corrosion hole geometry extraction module is used to extract each corrosion hole image after preprocessing. It uses edge detection to extract the boundary contour of the enhanced corrosion hole image, uses pixel integral of the connected component of the binary image to calculate the corrosion hole area, and uses grayscale linear mapping to calculate the depth. The single-hole capacitance calculation module is used to convert the geometric parameters of each corroded hole into a single-hole capacitance value based on the parallel plate capacitance theory, and to calculate the local specific capacitance of a single hole based on the single-hole capacitance value. The region volume ratio accumulation module is used to accumulate the local volume ratio of all holes using a weighted average to obtain the average volume ratio value of the entire image region. The average volume ratio value is calculated repeatedly for each cropped region. The specific capacity prediction modeling module is used to construct a non-destructive prediction model for the specific capacity value of the etched foil. It takes multiple region images of the etched foil as input, uses the calculated average specific capacity value of each region of the etched foil as a label to train the prediction model, and outputs the specific capacity value per unit area of ​​the entire etched foil.

[0032] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0033] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0034] 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; 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.

[0035] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes 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.

Claims

1. A method for determining the specific capacitance value of corroded foil, characterized in that, The specific steps include: Step 1: Cut and divide the etched foil into regions and clean it. Collect images of the etched holes on the surface of the etched foil. Preprocess the images of the etched holes. The preprocessing includes using Gaussian filtering to remove noise from the images of the etched holes and using histogram equalization to enhance the contrast of the filtered images of the etched holes. Step 2: Extract each preprocessed corrosion hole image, use edge detection to extract the boundary contour of the enhanced corrosion hole image, use pixel integral of the connected component of the binary image to calculate the corrosion hole area, and use grayscale linear mapping to calculate the depth. Step 3: Based on the parallel plate capacitance theory, convert the geometric parameters of each etched hole into a single hole capacitance value, and calculate the local specific capacitance of the single hole based on the single hole capacitance value. Step 4: Use a weighted average to accumulate the local volume ratio of all holes to obtain the average volume ratio of the entire image region. Repeat the calculation of the average volume ratio for each cropped region. Step 5: Construct a non-destructive prediction model for the specific capacitance of the etched foil. Take multiple region images of the etched foil as input, use the calculated average specific capacitance of each region of the etched foil as a label to train the prediction model, and output the specific capacitance per unit area of ​​the entire etched foil.

2. The method for determining the specific capacitance value of an etched foil according to claim 1, characterized in that: The preprocessing steps for the etched hole image are as follows: Gaussian filtering is used to denoise the images of corrosion holes collected on the surface of the corrosion foil. By constructing a two-dimensional Gaussian kernel and adjusting the weight distribution according to the standard deviation, a weighted average is performed on each pixel in the image through convolution to suppress high-frequency noise interference. Histogram equalization is then performed on the denoised image. By statistically analyzing the pixel gray-level distribution histogram and calculating the cumulative distribution function, the original gray-level values ​​are mapped to the dynamic range, stretching the gray-level difference between the holes and the background in the image and enhancing the contrast.

3. The method for determining the specific capacitance value of corroded foil according to claim 1, characterized in that: The specific steps for extracting the boundary contours of the enhanced erosion hole image using edge detection are as follows: The Sobel operator is used to calculate the gradient, where the formula for calculating the horizontal gradient is: ； The formula for calculating the gradient in the vertical direction is: ； In the formula, This is the pixel value matrix of the input image; Represents the gradient in the horizontal direction; Indicates the gradient in the vertical direction; The formulas for calculating gradient magnitude and direction are: ； In the formula, The gradient magnitude of the image at the current point. The angle of the gradient direction of the current pixel; By preserving the local maximum pixels along the gradient direction, region filling is performed on the image to form a binary mask image of the hole region. The calculation formula is as follows: ； in, Represents pixels Binary mask image at the location.

4. The method for determining the specific capacitance value of etched foil according to claim 1, characterized in that: The method of calculating the area of ​​eroded holes using pixel integration of connected components in a binary image and calculating the depth using grayscale linear mapping includes the following specific steps: A specific etched hole is defined as a connected region in a binary image, and its area is the total number of pixels in the region, calculated using the following formula: ； In the formula, For the image belonging to the first Pixel coordinates of the etched hole region For the first A connected region in the image containing eroded pits. This refers to the physical area corresponding to a single pixel. The depth of the etched holes is calculated through grayscale inversion, assuming the original image grayscale value is... The calculation formula is: ； In the formula, This represents the maximum grayscale value of the background region in the image. The physical depth corresponding to a unit grayscale difference. For the first The number of pixels in each etched hole.

5. The method for determining the specific capacitance value of etched foil according to claim 1, characterized in that: Based on the parallel plate capacitance theory, the geometric parameters of each etched hole are converted into a single-hole capacitance value. Then, the local specific capacitance of a single hole is calculated based on this single-hole capacitance value. The specific formula for the parallel plate capacitance theory is as follows: ； In the formula, The vacuum permittivity, The relative permittivity of the medium inside the hole; The formula for converting the capacitance value of a single hole into the local specific capacitance of a single hole is as follows: ； In the formula, For the first The local specific volume of a corrosion pit. Let be the geometric volume of the hole; The method for calculating the volume of the hole is as follows: ； In the formula, and These represent the area and depth of the hole, respectively.

6. The method for determining the specific capacitance value of an etched foil according to claim 1, characterized in that: The average volume ratio is repeatedly calculated for each cropped region, and the formula for calculating the average volume ratio of the image region is as follows: ； In the formula, This represents the total number of corrosion pores detected within the image area.

7. The method for determining the specific capacitance value of etched foil according to claim 1, characterized in that: A non-destructive prediction model for the specific capacitance of etched foil is constructed. Multiple region images of the etched foil are used as input, and the calculated average specific capacitance of each region of the etched foil is used as a label to train the prediction model. The output is the specific capacitance per unit area of ​​the entire etched foil. The specific training process of the non-destructive prediction model for the specific capacitance of the etched foil is as follows: the preprocessed image samples are divided into training set, validation set and test set; the images of the training set and validation set are used as input to the prediction model, and the average specific capacitance of each region of the etched foil is calculated as a label to train the prediction model; After training, the non-destructive prediction model for the specific volume of the etched foil is input into the entire image of the etched foil and outputs the overall unit area specific volume prediction value of the etched foil.

8. A system for determining the specific capacitance of corroded foil, characterized in that: The measuring system is used to perform the measuring method according to any one of claims 1-7, comprising: The image preprocessing module is used to cut and divide the etched foil into regions and clean it, acquire images of the etched holes on the surface of the etched foil, and preprocess the images of the etched holes. The preprocessing includes using Gaussian filtering to remove noise from the images of the etched holes and using histogram equalization to enhance the contrast of the filtered images of the etched holes. The corrosion hole geometry extraction module is used to extract each corrosion hole image after preprocessing. It uses edge detection to extract the boundary contour of the enhanced corrosion hole image, uses pixel integral of the connected component of the binary image to calculate the corrosion hole area, and uses grayscale linear mapping to calculate the depth. The single-hole capacitance calculation module is used to convert the geometric parameters of each corroded hole into a single-hole capacitance value based on the parallel plate capacitance theory, and to calculate the local specific capacitance of a single hole based on the single-hole capacitance value. The region volume ratio accumulation module is used to accumulate the local volume ratio of all holes using a weighted average to obtain the average volume ratio value of the entire image region. The average volume ratio value is calculated repeatedly for each cropped region. The specific capacity prediction modeling module is used to construct a non-destructive prediction model for the specific capacity value of the etched foil. It takes multiple region images of the etched foil as input, uses the calculated average specific capacity value of each region of the etched foil as a label to train the prediction model, and outputs the specific capacity value per unit area of ​​the entire etched foil.

Images