Graphite electrode defect detection method and system

Abstract

This invention provides a method and system for detecting defects in graphite electrodes, belonging to the field of artificial intelligence, to improve the robustness and stability of graphite electrode defect detection. The method includes: an electronic device acquiring M images of the same graphite electrode region, where M is an integer greater than 1; the M images containing content of the same graphite electrode region captured from different angles; the electronic device performing joint processing on the M images using an AI / ML model to obtain a processing result, which indicates whether a defect exists in the same graphite electrode region. The joint processing refers to the combined processing of features extracted from the M images by the AI / ML model.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a method and system for detecting defects in graphite electrodes. Background Technology

[0002] Graphite electrodes, as high-temperature conductive materials, are widely used in electric arc furnace steelmaking, lithium battery anode material production, industrial silicon smelting, and yellow phosphorus preparation. Their core function is to generate high temperatures (up to 3000℃ or higher) through electric arc discharge, enabling metal melting or chemical reactions. In electric arc furnace steelmaking, the consumption of graphite electrodes accounts for 10%-15% of production costs. Surface defects on the electrodes (such as cracks, chipped edges, and oxide peeling) significantly reduce conductivity, exacerbate abnormal discharges, leading to increased smelting energy consumption (approximately 5%-20%), accelerated electrode wear, and even electrode breakage accidents.

[0003] Traditional inspection methods include manual visual inspection, contact measurement, and vibration analysis. These methods have several shortcomings: 1) Manual visual inspection: Relies on operator experience and is affected by factors such as lighting conditions and visual fatigue. The detection rate for microcracks (<1mm) is less than 60%, and it cannot monitor the dynamic production process in real time. 2) Contact measurement: Uses mechanical measuring equipment such as micrometers and profilometers. A single measurement takes 3-5 minutes, the inspection frequency is low (usually 1-2 times per furnace), it cannot meet the needs of continuous production, and may cause secondary damage. 3) Vibration analysis: Judges structural integrity by monitoring the electrode vibration spectrum, but it lacks sensitivity to early surface defects, with a false alarm rate as high as 30%-40%, and is easily affected by mechanical noise during the smelting process. Therefore, the development of modern inspection technology mainly focuses on machine vision and deep learning. Specifically, using a high-resolution CCD camera (resolution up to 2048×2048 pixels) combined with improved Faster R-CNN, YOLOv5, and other neural network architectures can achieve more accurate defect identification.

[0004] However, defect identification using machine vision and deep learning suffers from insufficient stability. Summary of the Invention

[0005] This invention provides a method and system for detecting defects in graphite electrodes, which improves the robustness and stability of defect detection in graphite electrodes.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] In a first aspect, a graphite electrode defect detection method is provided, which is applied to an electronic device. The method includes: the electronic device acquiring M images of the same graphite electrode region, where M is an integer greater than 1, and the M images contain the contents of the same graphite electrode region captured from different angles; the electronic device performing joint processing on the M images through an AI / ML model to obtain a processing result, which is used to indicate whether there is a defect in the same graphite electrode region, and the joint processing refers to the joint processing of the features extracted by the AI / ML model from the M images by the AI / ML model.

[0008] Optionally, the electronic device performs joint processing on the M images using an AI / ML model to obtain the processing result, including: the electronic device inputs the M images into the feature extraction layer of the AI / ML model to obtain the features of the M images output by the feature extraction layer of the AI / ML model; the electronic device fuses the features of the M images to obtain the fused features of the M images; and the electronic device inputs the fused features of the M images into the feature processing layer of the AI / ML model to obtain the processing result output by the feature processing layer of the AI / ML model.

[0009] Optionally, the electronic device inputs M images into the feature extraction layer of the AI / ML model to obtain the features of the M images output by the feature extraction layer of the AI / ML model. This includes: the electronic device inputs the M images one-to-one into the M channels of the AI / ML model, each of the M channels corresponding to a feature extraction layer, for a total of M feature extraction layers, to obtain M features output one-to-one by the M feature extraction layers. Each of the M features corresponds to the feature of one of the M images, thus obtaining the features of the M images. The M feature extraction layers perform convolution operations of the same scale. The electronic device then fuses the features of the M images to obtain the fused features of the M images. This includes: the electronic device iteratively fuses partial features from every two images in the M images to obtain M fused features. Each of the M fused features corresponds to the fused feature of one of the M images, thus obtaining the fused features of the M images.

[0010] Optionally, for the features of the i-th image and the (i+1)-th image in the M images, where i is an integer from 1 to M, and when i=M, the value of i+1 is flipped to 1, the feature of the i-th image is the i-th matrix with a structure of K rows and L columns, and the feature of the (i+1)-th image is the (i+1)-th matrix with a structure of K rows and L columns, where K and L are both integers greater than 2; based on this, the electronic device iteratively fuses partial features of every two images in the M images to obtain M fused features, including: Step 1: The electronic device determines whether the correlation between the elements of the j-th row and the (j+1)-th row in the i-th matrix and the elements of the j-th row and the (j+1)-th row in the (i+1)-th matrix is ​​greater than or equal to the correlation threshold, where j is an integer starting from 1 and changing in increments of 2, j+1=K or j+1=K-1; Step 2: If the i-th image has a feature of the i-th matrix, the electronic device determines whether the correlation between the elements of the j-th row and the (j+1)-th row in the i-th matrix is ​​greater than or equal to the correlation threshold, where j is an integer starting from 1 and changing in increments of 2, j+1=K or j+1=K-1; If the correlation between the elements in row j and row j+1 and the elements in row j and row j+1 of matrix i+1 is greater than or equal to the correlation threshold, the electronic device increments the value of j by 2 and returns to step 1. If the correlation between the elements in row j and row j+1 of matrix i and the elements in row j and row j+1 of matrix i+1 is less than the correlation threshold, the electronic device executes step 3. Step 3: The electronic device merges the elements in row j and row j+1 of matrix i into the elements in row j and row j+1 of matrix i, resulting in 3 merged rows. Then, the electronic device increments the value of j by 2 and returns to step 1. This process continues until j+1=K or j+1=K-1, at which point the features of the merged image i are obtained. When i traverses to M, the features of the merged images M are obtained.

[0011] Optionally, the electronic device merges the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix into the elements in the j-th row and the (j+1)-th row of the (i)-th matrix, resulting in three merged rows of elements. This includes: the electronic device determining a first correlation between the elements in the j-th row of the (i)-th matrix and the elements in the j-th row of the (i+1)-th matrix; a second correlation between the elements in the j-th row of the (i)-th matrix and the elements in the (i+1)-th matrix; a third correlation between the elements in the j-th row of the (i+1)-th matrix and the elements in the j-th row of the (i+1)-th matrix; and a fourth correlation between the elements in the j-th row of the (i+1)-th matrix and the elements in the (i+1)-th matrix, for a total of four correlations. The electronic device then determines the two rows of elements with the lowest correlation among the four correlations and merges every two elements at the same position in the two rows into one element, resulting in one merged row of elements, thus obtaining three merged rows of elements.

[0012] Optionally, if the two rows with the lowest correlation are the (j+1)th row element and the (j)th row element of the (i+1)th matrix, then the fused three rows of elements in the features of the fused image in the i-th image, in ascending order of their index, are: the (j)th row element of the i-th matrix, the fused row element, and the (j+1)th row element of the (i+1)th matrix; if the two rows with the lowest correlation are the (j)th row element and the (j+1)th row element of the (i+1)th matrix, then the fused three rows of elements in the features of the fused image in the i-th image, in ascending order of their index, are: the (j)th row element of the (i+1)th matrix, the fused row element, and the (j+1)th row element of the (i+1)th matrix. If the two rows with the lowest correlation are the elements in the j-th row of the i-th matrix and the j-th row of the (i+1)-th matrix, then the three rows of elements after fusion, in the features of the fused image in the i-th image, are in ascending order of their numbers: the elements in the fused row, the elements in the (i+1)-th row of the i-th matrix, and the elements in the (i+1)-th row of the (i+1)-th matrix.

[0013] Optionally, for the features of the i-th image and the (i+1)-th image in the M images, where i is an integer from 1 to M, and when i=M, the value of i+1 is flipped to 1, the feature of the i-th image is the i-th matrix with a structure of K rows and L columns, and the feature of the (i+1)-th image is the (i+1)-th matrix with a structure of K rows and L columns, where K and L are both integers greater than 2; based on this, the electronic device iteratively fuses partial features of every two images in the M images to obtain M fused features, including: Step a: The electronic device determines whether the correlation between the elements in the x-th column and the (x+1)-th column of the i-th matrix and the elements in the x-th column and the (x+1)-th column of the (i+1)-th matrix is ​​greater than or equal to the correlation threshold, where x is an integer starting from 1 and changing in increments of 2, x+1=L or x+1=L-1; Step b: If the i-th image has a feature of the i-th matrix, the electronic device determines whether the correlation between the elements in the x-th column and the (x+1)-th column of the i-th matrix is ​​greater than or equal to the correlation threshold, where x is an integer starting from 1 and changing in increments of 2, x+1=L or x+1=L-1; If the correlation between the elements in column x and column x+1 and the elements in column x and column x+1 of matrix i+1 is greater than or equal to the correlation threshold, the electronic device increments the value of x by 2 and returns to step a. If the correlation between the elements in column x and column x+1 of matrix i and the elements in column x and column x+1 of matrix i+1 is less than the correlation threshold, the electronic device executes step c. Step c: The electronic device merges the elements in column x and column x+1 of matrix i into the elements in column x and column x+1 of matrix i, resulting in 3 merged columns. Then, the electronic device increments the value of x by 2 and returns to step a. This process continues until x+1 = L or x+1 = L-1, at which point the features of the merged image i are obtained. When i traverses to M, the features of the merged images M are obtained.

[0014] Optionally, the electronic device merges the elements in the x-th column and the x+1-th column of the (i+1)-th matrix into the elements in the x-th column and the x+1-th column of the i-th matrix, resulting in three merged rows of elements. This includes: the electronic device determining a first correlation between the elements in the x-th column of the i-th matrix and the elements in the x-th column of the (i+1)-th matrix; a second correlation between the elements in the x-th column of the i-th matrix and the elements in the x+1-th column of the (i+1)-th matrix; a third correlation between the elements in the x+1-th column of the i-th matrix and the elements in the x+1-th column of the (i+1)-th matrix; and a fourth correlation between the elements in the x+1-th column of the i-th matrix and the elements in the x+1-th column of the (i+1)-th matrix, for a total of four correlations; the electronic device then determines the two columns of elements with the lowest correlation among the four correlations, and merges every two elements at the same position in the two columns into one element, resulting in one merged column of elements, thus obtaining three merged columns of elements.

[0015] Optionally, if the two columns with the lowest correlation are the (x+1)th column element of the i-th matrix and the (x)th column element of the (i+1)th matrix, then the three fused columns, in ascending order of their indices, represent the features of the fused image in the i-th image: the (x)th column element of the i-th matrix, the fused column element, and the (x+1)th column element of the (i+1)th matrix. If the two columns with the lowest correlation are the (x)th column element of the i-th matrix and the (x+1)th column element of the (i+1)th matrix, then the three fused columns, in ascending order of their indices, represent the features of the fused image in the i-th image: the (x)th column element of the (i+1)th matrix, the fused column element, and the (x+1)th column element of the (i-th matrix). If the two columns with the lowest correlation are the x-th column of the i-th matrix and the x-th column of the (i+1)-th matrix, then the three columns of elements after fusion, in the features of the fused i-th image, are in ascending order of their numbers as follows: the first column of elements after fusion, the (x+1)-th column of the i-th matrix, and the (x+1)-th column of the (i+1)-th matrix.

[0016] Secondly, a graphite electrode defect detection system is provided. The system includes an electronic device configured to: acquire M images of the same graphite electrode region, where M is an integer greater than 1; the M images contain the contents of the same graphite electrode region captured from different angles; and the electronic device performs joint processing on the M images using an AI / ML model to obtain a processing result, which is used to indicate whether there is a defect in the same graphite electrode region. The joint processing refers to the joint processing of the features extracted by the AI / ML model from the M images by the AI / ML model.

[0017] The electronic devices in the system described in the second aspect are specifically configured to perform the method described in the first aspect. For details, please refer to the relevant introduction in the first aspect above, which will not be repeated here.

[0018] Thirdly, a computer-readable storage medium is provided, comprising: a computer program or instructions; when the computer program or instructions are executed on a computer, the computer causes the computer to perform the method described in the first aspect.

[0019] In summary, the above methods and systems have the following technical effects:

[0020] For the same graphite electrode region, the same region can be photographed from different angles, resulting in M ​​images. Due to image reflection, the image quality of the same location in the region may differ when photographed from different angles. Based on this characteristic, electronic devices can use an AI / ML model to jointly process the M images. That is, the features extracted by the AI / ML model from the M images are jointly processed by the AI / ML model to fuse the features of different image qualities at the same location. The resulting processing result indicating whether there is a defect in the same graphite electrode region can be more robust. Attached Figure Description

[0021] Figure 1 A schematic flowchart of the graphite electrode defect detection method provided in an embodiment of the present invention;

[0022] Figure 2 A schematic diagram of a scenario for the graphite electrode defect detection method provided in an embodiment of the present invention;

[0023] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0024] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0025] This invention will be presented in relation to systems that may include multiple devices, components, modules, etc. It should be understood and appreciated that various systems may include additional devices, components, modules, etc., and / or may not include all the devices, components, modules, etc. discussed in conjunction with the accompanying drawings. Furthermore, combinations of these approaches are also possible.

[0026] Furthermore, in embodiments of the present invention, words such as "exemplary" and "for example" are used to indicate that something is presented as an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Rather, the use of the term "exemplary" is intended to present the concept in a specific manner.

[0027] In the embodiments of this invention, "of", "corresponding (relevant)" and "corresponding" can sometimes be used interchangeably. It should be noted that when their distinction is not emphasized, their intended meaning is matching. In addition, the " / " mentioned in this invention can be used to indicate an "or" relationship.

[0028] The network architecture and business scenarios described in the embodiments of this invention are for the purpose of more clearly illustrating the technical solutions of the embodiments of this invention, and do not constitute a limitation on the technical solutions provided by the embodiments of this invention. As those skilled in the art will know, with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided by the embodiments of this invention are also applicable to similar technical problems.

[0029] For example, Figure 1 A schematic flowchart of the graphite electrode defect detection method provided in this embodiment of the invention. This method can be applied to electronic devices.

[0030] like Figure 1 As shown, the flowchart of this graphite electrode defect detection method is as follows:

[0031] S101, the electronic device acquires M images of the same graphite electrode region, where M is an integer greater than 1.

[0032] The M images contain the same graphite electrode region captured from different angles, and can be high-resolution images taken by a high-resolution CCD camera.

[0033] For example, such as Figure 2 As shown, assume that the same graphite electrode region ab lies in the xy plane of the xyz rectangular coordinate system. The shooting direction is located in the xq plane, the q direction is located in the zy plane, and the q direction makes an angle of 45-60° with the z-axis. There can be 5 shooting directions, namely direction #1 to direction #5, where direction #1 is along... Figure 2 The arrows shown are parallel to the q direction. The angle between direction #2 and direction #1 can be +8°, ​​the angle between direction #3 and direction #2 can be +8°, ​​the angle between direction #4 and direction #1 can be -8°, and the angle between direction #5 and direction #4 can be -8°.

[0034] M images are images of the same size (or resolution).

[0035] S102, the electronic device performs joint processing on M images using an AI / ML model to obtain the processing result.

[0036] The processing results are used to indicate whether there are defects in the same graphite electrode area, such as the size, direction and trend of the wave.

[0037] Joint processing refers to the joint processing of features extracted from M images by an artificial intelligence (AI) / machine learning (ML) model, which will be described in detail below.

[0038] First, the electronic device can input M images into the feature extraction layer of an AI / ML model, obtaining the features of the M images output by the AI / ML model's feature extraction layer. For example, the electronic device can input the M images one-to-one into the M channels of the AI / ML model, with each of the M channels corresponding to a feature extraction layer, for a total of M feature extraction layers. This results in M ​​features output by the M feature extraction layers, each of which corresponds to a feature of one of the M images. In other words, the electronic device obtains the features of the M images. The M feature extraction layers perform convolution operations of the same scale, meaning the size of the convolution kernel and the stride of the convolution are the same.

[0039] Secondly, electronic devices can fuse the features of M images to obtain the fused features of M images. For example, electronic devices can poll and fuse partial features of every two images in the M images to obtain M fused features. Each of the M fused features corresponds to the fused feature of one image in the M images, thus obtaining the fused features of M images. The following are two cases.

[0040] Scenario 1:

[0041] For the features of the i-th image and the (i+1)-th image in M ​​images, where i is an integer from 1 to M, and when i=M, the value of i+1 is flipped to 1, the feature of the i-th image is the i-th matrix with a structure of K rows and L columns, and the feature of the (i+1)-th image is the (i+1)-th matrix with a structure of K rows and L columns, where K and L are both integers greater than 2.

[0042] Based on this, the electronic device performs the following steps:

[0043] Step 1: The electronic device can determine whether the correlation between the elements in the j-th row and the (j+1)-th row of the i-th matrix and the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix is ​​greater than or equal to the correlation threshold, where j is an integer starting from 1 and changing in increments of 2, and j+1=K or j+1=K-1.

[0044] The correlation between the elements in the j-th and (j+1)-th rows of the i-th matrix and the elements in the (i+1)-th rows of the i-th matrix can be calculated using methods such as Euclidean distance, Manhattan distance, Minkowski distance, and Pearson coefficient. Taking Euclidean distance as an example, and similar interpretations for other methods, a larger Euclidean distance indicates a lower correlation (i.e., correlation greater than or equal to a correlation threshold can be understood as Euclidean distance less than or equal to a distance threshold). Conversely, a smaller Euclidean distance indicates a higher correlation (i.e., correlation less than a correlation threshold can be understood as Euclidean distance greater than a distance threshold). Each element in the j-th row can be a vector, and similarly, each element in the (j+1)-th row can also be a vector.

[0045] Step 2: If the correlation between the elements in row j and row (j+1) of matrix i and the elements in row j and row (j+1) of matrix i+1 is greater than or equal to the correlation threshold, the electronic device increments the value of j by 2 and returns to step 1; if the correlation between the elements in row j and row (j+1) of matrix i and the elements in row j and row (j+1) of matrix i+1 is less than the correlation threshold, the electronic device executes step 3.

[0046] In this context, a correlation between the elements in the j-th row and the (j+1)-th row of the i-th matrix and the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix is ​​greater than or equal to a correlation threshold, indicating high similarity. That is, for the i-th image corresponding to the i-th matrix and the (i+1)-th image corresponding to the (i+1)-th matrix, the region in the i-th image mapped from the elements in the j-th row and the (j+1)-th row of the i-th matrix (which can also be understood as the pattern of the elements in the j-th row and the (j+1)-th row of the i-th matrix obtained by convolution) is similar to the region in the (i+1)-th image mapped from the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix (which can also be understood as the pattern of the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix) obtained by convolution. Neither of the patterns in these two regions exhibits inconsistent gloss patterns (which can be understood as a small amount of reflection or flooding) due to sunlight reflection. Conversely, if the correlation between the elements in the j-th row and the (j+1)-th row of the i-th matrix and the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix is ​​less than the correlation threshold, it means that the patterns in the two regions exhibit inconsistent gloss patterns due to sunlight reflection during photography. Therefore, feature fusion is required to improve the robustness of the model.

[0047] Step 3: The electronic device merges the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix into the elements in the j-th row and the (j+1)-th row of the i-th matrix, resulting in a merged set of 3 rows. Then, the electronic device increments the value of j by 2 and returns to execute step 1.

[0048] The electronic device can determine four correlations: a first correlation between an element in row j of matrix i and an element in row j of matrix i+1; a second correlation between an element in row j of matrix i and an element in row j+1 of matrix i; a third correlation between an element in row j+1 of matrix i and an element in row j of matrix i+1; and a fourth correlation between an element in row j+1 of matrix i and an element in row j+1 of matrix i. The electronic device can then determine the two rows with the lowest correlation among these four correlations and merge every two elements at the same position in these two rows into one element (e.g., adding every two vectors at the same position, i.e., adding them digit by digit, such as adding the first vector of row A to the first vector of row B, adding the second vector of row A to the second vector of row B, and so on), resulting in a merged row of elements, thus obtaining a merged three-row element set.

[0049] If the two rows with the lowest correlation are the (j+1)th row of matrix i and the (j)th row of matrix i+1, then the three rows of elements after fusion, in the features of the fused image i, are in ascending order of their indices: the (j)th row of matrix i, the fused row of elements, and the (j+1)th row of matrix i+1. If the two rows with the lowest relevance are the (j+1)th row of matrix i and the (j+1)th row of matrix i+1, then the three rows of elements after fusion, in the fused features of the i-th image, are ordered from smallest to largest as follows: the (j)th row of matrix i, the (j)th row of matrix i+1, and the fused row of elements. This fusion and sorting method avoids disrupting the feature structure; that is, the fused features still maintain the structural order of each row as much as possible.

[0050] Until j+1=K or j+1=K-1, the features of the fused i-th image are obtained.

[0051] When i traverses to M, the features of M images after fusion are obtained.

[0052] Scenario 2:

[0053] For the features of the i-th image and the (i+1)-th image in M ​​images, where i is an integer from 1 to M, and when i=M, the value of i+1 is flipped to 1, the feature of the i-th image is the i-th matrix with a structure of K rows and L columns, and the feature of the (i+1)-th image is the (i+1)-th matrix with a structure of K rows and L columns, where K and L are both integers greater than 2.

[0054] Based on this, the electronic device performs the following steps:

[0055] Step a: The electronic device determines whether the correlation between the elements in column x and column x+1 of matrix i and the elements in column x and column x+1 of matrix i+1 is greater than or equal to the correlation threshold, where x is an integer starting from 1 and changing in increments of 2, and x+1=L or x+1=L-1.

[0056] Step b: If the correlation between the elements in column x and column x+1 of matrix i and the elements in column x and column x+1 of matrix i+1 is greater than or equal to the correlation threshold, the electronic device increases the value of x by 2 and returns to execute step a; if the correlation between the elements in column x and column x+1 of matrix i and the elements in column x and column x+1 of matrix i+1 is less than the correlation threshold, the electronic device executes step c.

[0057] Step c: The electronic device merges the elements in the x-th column and the x+1-th column of the (i+1)-th matrix into the elements in the x-th column and the x+1-th column of the i-th matrix, resulting in 3 merged columns. Then, the electronic device increments the value of x by 2 and returns to execute step a.

[0058] The electronic device can determine four correlations: a first correlation between an element in column x of matrix i and an element in column x of matrix (i+1); a second correlation between an element in column x of matrix i and an element in column x of matrix (i+1); a third correlation between an element in column x of matrix i and an element in column x of matrix (i+1); and a fourth correlation between an element in column x of matrix i and an element in column x of matrix (i+1). The electronic device then identifies the two columns with the lowest correlation among these four correlations and merges every two elements at the same position in these two columns into a single merged column, resulting in a final three-column composition.

[0059] Where the two columns with the lowest correlation are the (x+1)th column element of the i-th matrix and the (x)th column element of the (i+1)th matrix, then the three fused columns of elements in the features of the i-th image, in ascending order of their indices, are: the (x)th column element of the i-th matrix, the fused column element, and the (x+1)th column element of the (i+1)th matrix. If the two columns with the lowest correlation are the x-th column of the i-th matrix and the x-th column of the (i+1)-th matrix, then the three columns of elements after fusion, in the features of the fused i-th image, are in ascending order of their numbers as follows: the first column of elements after fusion, the (x+1)-th column of the i-th matrix, and the (x+1)-th column of the (i+1)-th matrix.

[0060] Until x+1=L or x+1=L-1, the features of the fused i-th image are obtained;

[0061] When i traverses to M, the features of the M images after fusion are obtained.

[0062] It is understandable that the implementation principle of case 2 is similar to that of case 1, and can be referred to for understanding, so it will not be elaborated here.

[0063] In addition, whether to use Case 1 or Case 2 depends on the characteristics of the gloss pattern. For example, if the gloss pattern is similar to a horizontal stripe pattern, that is, along the row direction, then Case 1 is used. If the gloss pattern is similar to a vertical stripe pattern, that is, along the column direction, then Case 2 is used.

[0064] Finally, the electronic device can input the fused features of the M images into the feature processing layer of the AI / ML model, and obtain the output of the feature processing layer of the AI / ML model. The feature processing layer can be a sampling layer followed by a fully connected layer.

[0065] In summary: For the same graphite electrode region, the same region can be photographed from different angles, resulting in M ​​images. Due to image reflection, the image quality of the same location in the region may differ when photographed from different angles. Based on this characteristic, electronic devices can use an AI / ML model to jointly process the M images. That is, the features extracted by the AI / ML model from the M images are jointly processed by the AI / ML model to fuse the features of different image qualities at the same location. The resulting processing result indicating whether there is a defect in the same graphite electrode region can be more robust.

[0066] The above combination Figure 1 This invention provides a detailed description of a graphite electrode defect detection method according to embodiments of the present invention. The following details a graphite electrode defect detection system for implementing the method provided in the embodiments of the present invention.

[0067] The system includes an electronic device configured to: acquire M images of the same graphite electrode region, where M is an integer greater than 1; the M images contain the content of the same graphite electrode region captured from different angles; the electronic device performs joint processing on the M images using an AI / ML model to obtain a processing result, which is used to indicate whether there are defects in the same graphite electrode region; the joint processing refers to the joint processing of the features extracted by the AI / ML model from the M images by the AI / ML model.

[0068] The electronic devices in this system are specifically configured to perform the above. Figure 1 The methods shown are for reference and understanding only, and will not be elaborated upon further.

[0069] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Exemplarily, the electronic device may be a terminal device, or a chip (system) or other component or assembly that can be disposed in the terminal device. Figure 3 As shown, the electronic device 400 may include a processor 401. Optionally, the electronic device 400 may also include a memory 402 and / or a transceiver 403. The processor 401 is coupled to the memory 402 and the transceiver 403, for example, they can be connected via a communication bus. Alternatively, the electronic device 400 may also be a chip, such as including the processor 401; in this case, the transceiver may be the chip's input / output interface.

[0070] The following is combined Figure 3 The various components of electronic device 400 are described in detail below:

[0071] The processor 401 is the control center of the electronic device 400. It can be a single processor or a collective term for multiple processing elements. For example, the processor 401 can be one or more central processing units (CPUs), application-specific integrated circuits (ASICs), or one or more integrated circuits configured to implement embodiments of the present invention, such as one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs).

[0072] Optionally, the processor 401 can perform various functions of the electronic device 400, such as the aforementioned functions, by running or executing software programs stored in the memory 402 and calling scientific data stored in the memory 402. Figure 1 The graphite electrode defect detection method shown.

[0073] In a specific implementation, as one example, processor 401 may include one or more CPUs, for example... Figure 3 CPU0 and CPU1 are shown in the diagram.

[0074] In a specific implementation, as one example, the electronic device 400 may also include multiple processors. Each of these processors may be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process scientific data (such as computer programs or instructions).

[0075] The memory 402 is used to store the software program that executes the present invention, and is controlled by the processor 401 to execute it. The specific implementation method can be referred to the above method embodiment, and will not be repeated here.

[0076] Optionally, the memory 402 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or scientific data structures and accessible by a computer, but not limited thereto. The memory 402 may be integrated with the processor 401 or may exist independently and be accessible through the interface circuit of the electronic device 400. Figure 3 (Not shown in the image) is coupled to processor 401, and this embodiment of the invention does not specifically limit this.

[0077] Transceiver 403 is used for communication with other electronic devices. For example, if electronic device 400 is a terminal device, transceiver 403 can be used to communicate with a network device or with another terminal device. As another example, if electronic device 400 is a network device, transceiver 403 can be used to communicate with a terminal device or with another network device.

[0078] Alternatively, transceiver 403 may include a receiver and a transmitter. Figure 3 (Not shown separately). The receiver is used to implement the receiving function, and the transmitter is used to implement the transmitting function.

[0079] Alternatively, the transceiver 403 can be integrated with the processor 401, or it can exist independently and be connected via the interface circuit of the electronic device 400. Figure 3 (Not shown in the image) is coupled to processor 401, and this embodiment of the invention does not specifically limit this.

[0080] Understandable, Figure 3 The structure of the electronic device 400 shown does not constitute a limitation on the electronic device. Actual electronic devices may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0081] Furthermore, the technical effects of the electronic device 400 can be referred to the technical effects of the methods described in the above method embodiments, and will not be repeated here.

[0082] It should be understood that the processor in the embodiments of the present invention can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0083] It should also be understood that the memory in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0084] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer program or instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions can be transferred from one website, computer, server, or scientific data center to another website, computer, server, or scientific data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a scientific data storage device such as a server or scientific data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0085] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0086] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0087] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0088] 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 in 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. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0089] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0090] In the embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0091] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0092] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0093] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0094] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for detecting defects in a graphite electrode, characterized in that, Applied to electronic devices, the method includes: The electronic device acquires M images of the same graphite electrode region, where M is an integer greater than 1; the M images include content captured from different angles of the same graphite electrode region. The electronic device performs joint processing on the M images using an AI / ML model to obtain a processing result. The processing result is used to indicate whether there is a defect in the same graphite electrode region. The joint processing refers to the joint processing of the features extracted by the AI / ML model from the M images by the AI / ML model. The electronic device performs joint processing on the M images using an AI / ML model to obtain the processing results, including: The electronic device inputs the M images into the feature extraction layer of the AI / ML model to obtain the features of the M images output by the feature extraction layer of the AI / ML model; The electronic device fuses the features of the M images to obtain the fused features of the M images; The electronic device inputs the features of the M images after fusion into the feature processing layer of the AI / ML model, and obtains the processing result output by the feature processing layer of the AI / ML model. The electronic device inputs the M images into the feature extraction layer of the AI / ML model to obtain the features of the M images output by the feature extraction layer of the AI / ML model, including: The electronic device inputs the M images one-to-one into the M channels of the AI / ML model. Each of the M channels corresponds to a feature extraction layer, for a total of M feature extraction layers, resulting in M ​​features output by the M feature extraction layers. Each of the M features corresponds to the feature of one of the M images, thus obtaining the features of the M images. The M feature extraction layers perform convolution operations of the same scale. The electronic device fuses the features of the M images to obtain the fused features of the M images, including: The electronic device polls and fuses partial features of every two images in the M images to obtain M fused features. Each of the M fused features corresponds to the fused feature of one image in the M images, thus obtaining the fused features of the M images. For the features of the i-th image and the (i+1)-th image in the M images, i is an integer from 1 to M. When i=M, the value of i+1 is flipped to 1. The feature of the i-th image is the i-th matrix with a structure of K rows and L columns, and the feature of the (i+1)-th image is the (i+1)-th matrix with a structure of K rows and L columns, where K and L are both integers greater than 2. Based on this, the electronic device polls and fuses partial features from every two images in the M images to obtain M fused features, including: Step 1: The electronic device determines whether the correlation between the elements in the j-th row and the (j+1)-th row of the i-th matrix and the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix is ​​greater than or equal to the correlation threshold, where j is an integer starting from 1 and increasing in increments of 2, and j+1=K or j+1=K-1; Step 2: If the correlation between the elements in the j-th row and the (j+1)-th row of the i-th matrix and the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix is ​​greater than or equal to the correlation threshold, the electronic device increments the value of j by 2 and returns to step 1; if the correlation between the elements in the j-th row and the (j+1)-th row of the i-th matrix and the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix is ​​less than the correlation threshold, the electronic device executes step 3. Step 3: The electronic device merges the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix into the elements in the j-th row and the (j+1)-th row of the i-th matrix, resulting in 3 merged rows. Then, the electronic device increments the value of j by 2 and returns to execute step 1. Until j+1=K or j+1=K-1, the features of the fused i-th image are obtained; When i traverses to M, the features of the M images after fusion are obtained.

2. The method according to claim 1, characterized in that, The electronic device merges the elements of row j and row j+1 of the (i+1)th matrix into the elements of row j and row j+1 of the i-th matrix, resulting in 3 merged rows, including: The electronic device determines a first correlation degree between the element in the j-th row of the i-th matrix and the element in the j-th row of the (i+1)-th matrix, a second correlation degree between the element in the j-th row of the i-th matrix and the element in the (i+1)-th matrix, a third correlation degree between the element in the (i+1)-th row of the i-th matrix and the element in the j-th row of the (i+1)-th matrix, and a fourth correlation degree between the element in the (i+1)-th row of the i-th matrix and the element in the (i+1)-th matrix, for a total of four correlation degrees; The electronic device determines the two rows of elements with the lowest correlation among the four correlations, and merges every two elements at the same position in the two rows into one element to obtain a merged row of elements, thereby obtaining the merged three rows of elements.

3. The method according to claim 2, characterized in that, If the two rows of elements with the lowest correlation are the (j+1)th row of the i-th matrix and the (j)th row of the (i+1)th matrix, then the three rows of elements after fusion are in the following order from smallest to largest in the features of the fused i-th image: the (j)th row of the i-th matrix, the elements of the fused row, and the (j+1)th row of the (i+1)th matrix. If the two rows of elements with the lowest correlation are the element in the j-th row of the i-th matrix and the element in the (j+1)-th row of the i+1-th matrix, then the three rows of elements after fusion are in the following order from smallest to largest in the features of the fused i-th image: the element in the j-th row of the (i+1)-th matrix, the element in the fused row, and the element in the (j+1)-th row of the i-th matrix; If the two rows of elements with the lowest correlation are the element in the j-th row of the i-th matrix and the element in the j-th row of the (i+1)-th matrix, then the three rows of elements after fusion are in the following order from smallest to largest in the features of the fused i-th image: the element in the fused row, the element in the (j+1)-th row of the i-th matrix, and the element in the (j+1)-th row of the (i+1)-th matrix. If the two rows with the lowest correlation are the (j+1)th row element of the i-th matrix and the (j+1)th row element of the (i+1)th matrix, then the three rows of elements after fusion, in the features of the fused i-th image, are in ascending order of their numbers as follows: the j-th row element of the i-th matrix, the j-th row element of the (i+1)th matrix, and the fused row element.

4. The method according to claim 1, characterized in that, For the features of the i-th image and the (i+1)-th image in the M images, i is an integer from 1 to M. When i=M, the value of i+1 is flipped to 1. The feature of the i-th image is the i-th matrix with a structure of K rows and L columns. The feature of the (i+1)-th image is the (i+1)-th matrix with a structure of K rows and L columns. K and L are both integers greater than 2. Based on this, the electronic device polls and fuses partial features from every two images in the M images to obtain M fused features, including: Step a: The electronic device determines whether the correlation between the elements in the x-th column and the (x+1)-th column of the i-th matrix and the elements in the x-th column and the (x+1)-th column of the (i+1)-th matrix is ​​greater than or equal to the correlation threshold, where x is an integer starting from 1 and changing in increments of 2, and x+1=L or x+1=L-1; Step b: If the correlation between the elements in the x-th column and the (x+1)-th column of the i-th matrix and the elements in the x-th column and the (x+1)-th column of the (i+1)-th matrix is ​​greater than or equal to the correlation threshold, the electronic device increments the value of x by 2 and returns to step a; if the correlation between the elements in the x-th column and the (x+1)-th column of the i-th matrix and the elements in the x-th column and the (x+1)-th column of the (i+1)-th matrix is ​​less than the correlation threshold, the electronic device executes step c. Step c: The electronic device merges the elements in the x-th column and the x+1-th column of the (i+1)-th matrix into the elements in the x-th column and the x+1-th column of the i-th matrix, resulting in 3 merged columns. Then, the electronic device increments the value of x by 2 and returns to execute step a. Until x+1=L or x+1=L-1, the features of the fused i-th image are obtained; When i traverses to M, the features of the M images after fusion are obtained.

5. The method according to claim 4, characterized in that, The electronic device merges the elements in the x-th column and the x+1-th column of the (i+1)-th matrix into the elements in the x-th column and the x+1-th column of the i-th matrix, resulting in 3 rows of merged elements, including: The electronic device determines a first correlation degree between the element in column x of the i-th matrix and the element in column x of the (i+1)-th matrix, a second correlation degree between the element in column x of the i-th matrix and the element in column x+1 of the (i+1)-th matrix, a third correlation degree between the element in column x+1 of the i-th matrix and the element in column x of the (i+1)-th matrix, and a fourth correlation degree between the element in column x+1 of the i-th matrix and the element in column x+1 of the (i+1)-th matrix, for a total of four correlation degrees; The electronic device determines the two columns of elements with the lowest correlation among the four correlations, and merges every two elements at the same position in the two columns into one element to obtain a merged column of elements, thereby obtaining the merged three columns of elements.

6. The method according to claim 5, characterized in that, If the two columns with the lowest correlation are the (x+1)th column element of the i-th matrix and the (x)th column element of the (i+1)th matrix, then the three fused columns of elements in the features of the fused i-th image, in ascending order of their numbers, are: the (x)th column element of the i-th matrix, the fused column element, and the (x+1)th column element of the (i+1)th matrix; If the two columns with the lowest correlation are the x-th column element in the i-th matrix and the x+1-th column element in the (i+1)-th matrix, then the three fused columns of elements in the features of the fused i-th image, in ascending order of their numbers, are: the x-th column element in the (i+1)-th matrix, the fused column element, and the x+1-th column element in the i-th matrix; If the two columns with the lowest correlation are the x-th column element in the i-th matrix and the x-th column element in the (i+1)-th matrix, then the three columns of elements after fusion are, in ascending order of their serial numbers, the following: the fused column element, the (x+1)-th column element in the i-th matrix, and the (x+1)-th column element in the (i+1)-th matrix. If the two columns with the lowest correlation are the (x+1)th column element of the i-th matrix and the (x+1)th column element of the (i+1)th matrix, then the three columns of elements after fusion, in the features of the fused i-th image, are in ascending order of their numbers as follows: the (x)th column element of the i-th matrix, the (x)th column element of the (i+1)th matrix, and the elements of the fused row.

7. A graphite electrode defect detection system, characterized in that, The system includes electronic equipment, and the system is configured to: The electronic device acquires M images of the same graphite electrode region, where M is an integer greater than 1; the M images include content captured from different angles of the same graphite electrode region. The electronic device performs joint processing on the M images using an AI / ML model to obtain a processing result. The processing result is used to indicate whether there is a defect in the same graphite electrode region. The joint processing refers to the joint processing of the features extracted by the AI / ML model from the M images by the AI / ML model. The electronic device performs joint processing on the M images using an AI / ML model to obtain the processing results, including: The electronic device inputs the M images into the feature extraction layer of the AI / ML model to obtain the features of the M images output by the feature extraction layer of the AI / ML model; The electronic device fuses the features of the M images to obtain the fused features of the M images; The electronic device inputs the features of the M images after fusion into the feature processing layer of the AI / ML model, and obtains the processing result output by the feature processing layer of the AI / ML model. The electronic device inputs the M images into the feature extraction layer of the AI / ML model to obtain the features of the M images output by the feature extraction layer of the AI / ML model, including: The electronic device inputs the M images one-to-one into the M channels of the AI / ML model. Each of the M channels corresponds to a feature extraction layer, for a total of M feature extraction layers, resulting in M ​​features output by the M feature extraction layers. Each of the M features corresponds to the feature of one of the M images, thus obtaining the features of the M images. The M feature extraction layers perform convolution operations of the same scale. The electronic device fuses the features of the M images to obtain the fused features of the M images, including: The electronic device polls and fuses partial features of every two images in the M images to obtain M fused features. Each of the M fused features corresponds to the fused feature of one image in the M images, thus obtaining the fused features of the M images. For the features of the i-th image and the (i+1)-th image in the M images, i is an integer from 1 to M. When i=M, the value of i+1 is flipped to 1. The feature of the i-th image is the i-th matrix with a structure of K rows and L columns, and the feature of the (i+1)-th image is the (i+1)-th matrix with a structure of K rows and L columns, where K and L are both integers greater than 2. Based on this, the electronic device polls and fuses partial features from every two images in the M images to obtain M fused features, including: Step 1: The electronic device determines whether the correlation between the elements in the j-th row and the (j+1)-th row of the i-th matrix and the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix is ​​greater than or equal to the correlation threshold, where j is an integer starting from 1 and increasing in increments of 2, and j+1=K or j+1=K-1; Step 2: If the correlation between the elements in the j-th row and the (j+1)-th row of the i-th matrix and the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix is ​​greater than or equal to the correlation threshold, the electronic device increments the value of j by 2 and returns to step 1; if the correlation between the elements in the j-th row and the (j+1)-th row of the i-th matrix and the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix is ​​less than the correlation threshold, the electronic device executes step 3. Step 3: The electronic device merges the elements in the j-th row and the (j+1)-th row of the (i+1)-th matrix into the elements in the j-th row and the (j+1)-th row of the i-th matrix, resulting in 3 merged rows. Then, the electronic device increments the value of j by 2 and returns to execute step 1. Until j+1=K or j+1=K-1, the features of the fused i-th image are obtained; When i traverses to M, the features of the M images after fusion are obtained.

Images