A method for detecting corrosion of steel components of a power transmission line

Abstract

The application is a kind of transmission line steel component corrosion detection method, the method comprises: obtaining the corrosion image of high-voltage transmission line, and removing the background by using Depth Anything V2 model to extract the close-range image;According to the corrosion degree, the close-range image is artificially labeled to obtain the labeled data, and the completely corroded pixel set is divided into different types of subsets by K-means clustering algorithm, so as to calculate the center point of the corrosion pixel and define the color space triangle;Get the image of the transmission line to be detected, and remove the background and component segmentation to get the regional image of each component, on this basis, carry out depth calculation and screen out the foreground image;The component region image screened out is projected on color value, the corrosion degree quantitative value is calculated, and the statistical evaluation parameters including the average corrosion degree, the maximum corrosion degree and the corrosion degree proportion are calculated;According to the evaluation parameters, the corresponding corrosion grade is formulated, and finally the scientific corrosion evaluation and monitoring of the transmission line is provided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power transmission line detection, and particularly relates to a method for detecting corrosion of steel components of a power transmission line. BACKGROUND

[0002] Because the power transmission line is exposed to the outdoor environment for a long time, it is easily affected by moisture, salt and other corrosive substances in the atmosphere, especially in humid or rainy areas, these factors will accelerate the oxidation reaction of metal components, thereby causing corrosion. Corrosion not only reduces the mechanical strength of the power transmission line and its components, but also can cause poor electrical contact, leading to a decrease in power transmission efficiency, and even causing safety accidents such as wire breakage and short circuit. Therefore, regular corrosion detection of the power transmission line is crucial to ensure the safe and stable operation of the power system.

[0003] CN110726725A "A method and device for detecting corrosion of power transmission line fittings" discloses "a method and device for detecting corrosion of power transmission line fittings, the method first acquires a power transmission line image, processes the target fitting image obtained to obtain a foreground image and a background image; the foreground image is subjected to gamma transformation, the foreground image after gamma transformation is subjected to gray scale processing by using a super red algorithm to obtain a gray scale image of the foreground image; finally, the corrosion condition of the fitting is determined according to the gray scale image of the foreground image. The present application focuses on gamma transformation of the foreground image to strengthen the fitting in the target fitting image, which can adjust the influence of too strong or too weak light on the fitting in the target fitting image, so as to more accurately judge the corrosion degree, area and other corrosion conditions of the fitting, provide comprehensive corrosion defect information for the operation and maintenance personnel, and provide an effective technical means for realizing intelligent power operation and maintenance", but the application needs to set a brightness threshold to adjust the gamma value of the gamma transformation, this threshold-dependent method may not be stable under different environmental conditions, and if the threshold is not properly set, the accuracy of the corrosion detection may be affected, therefore, there is an urgent need for a method for detecting corrosion of steel components of a power transmission line that maintains detection accuracy under different environmental conditions. SUMMARY

[0004] In order to solve the above-mentioned problems existing in the prior art, the present application provides a method for detecting corrosion of steel components of a power transmission line.

[0005] The technical scheme of the present application is as follows:

[0006] A method for detecting corrosion of steel components of a power transmission line, the method comprising:

[0007] Obtain a corroded image of a high-voltage transmission line and, using the Depth Anything V2 model, remove the background and distant view from the image to obtain the corresponding close-up image.

[0008] Manually labeling the rust in the close-up image according to the degree of rust to obtain labeling data; obtaining a non-corroded pixel set and a completely corroded pixel set based on the labeling data, dividing the completely corroded pixel set into two disjoint completely corroded pixel subsets of different types using a Kmean clustering algorithm, calculating the center points of the non-corroded pixel set and the completely corroded pixel subsets, and defining a color space triangle based on the center points;

[0009] Acquire an image of a high-voltage transmission line to be inspected, remove the background and distant view from the image based on the Depth Anything V2 model, and obtain a close-up image. Use skeleton generation technology to process the close-up image to obtain a target structure, obtain prompt points of the Segment Anything 2 model from the target structure, and segment the image of the high-voltage transmission line to be inspected based on the prompt points to obtain regional images corresponding to various components in the high-voltage transmission line to be inspected. Perform depth calculation on the segmented regional images to filter out the regional images of components located in the foreground.

[0010] The color values ​​of the pixels in the screened component area image are projected onto the color space triangle to obtain the corresponding projection point; the quantitative value of the rust degree of the corresponding projection point is calculated, and the evaluation parameters in the component area image are counted. The evaluation parameters include the average rust degree, the maximum rust degree, and the rust degree ratio of all pixels. The rust grade of the component corresponding to the component area image is evaluated based on the evaluation parameters.

[0011] As a preferred embodiment of the present invention, the background and distant view in the high-voltage transmission line corrosion image are removed based on the Depth Anything V2 model to obtain a close-up image. Specifically, the depth of each pixel in the high-voltage transmission line image is calculated. Corresponding depth , sort the depths of all pixels corresponding to the high-voltage transmission line image from large to small, and regard the pixels in the last 50% of the order as the background, and obtain the close-up image after removing the background and the corresponding rust pixel set .

[0012] As a preferred embodiment of the present invention, the rust in the close-up image is manually labeled according to the degree of rust to obtain labeled data, wherein:

[0013] The range of the rust degree is [0,1], and the rust degree is marked on the pixels in the area without rust in the close-up image. The corresponding value is 0; the most severely corroded area in the close-up image includes pixels with the degree of corrosion. The corresponding value is 1; get the rust pixel set Corresponding annotation data, including the degree of corrosion corresponding to the pixel and color value coordinates .

[0014] As a preferred embodiment of the present invention, the center points of the non-corroded pixel set and the completely corroded pixel subset are calculated, and a color space triangle is defined based on the center points, specifically:

[0015] Based on rust pixel set Get the uncorroded pixel set and a completely rusted pixel set , expressed as:

[0016] ;

[0017] ;

[0018] Calculate the set of completely corroded pixels The red component corresponding to any pixel in The mean , expressed as:

[0019] ;

[0020] Where, Represents a completely corroded pixel set The number of pixels in ;

[0021] Will completely rust pixel set According to the red component Divide into subsets and , expressed as:

[0022] ;

[0023] ;

[0024] Calculate the subset and Center point and , expressed as:

[0025] ;

[0026] ;

[0027] Re-divide the subsets according to the center point, and iterate the division process until convergence, which can be expressed as:

[0028] ;

[0029] ;

[0030] ;

[0031] ;

[0032] Where, is the final number of iterations, ;

[0033] Get a subset and , and the corresponding center points are and ;

[0034] Set of uncorroded pixels , a subset of completely corroded pixels and To reduce, specifically, repeatedly calculate the center point of the set and delete the one farthest from the center point in the set points, among which To reduce the ratio; repeat the set reduction process until the center point of the set no longer changes, and obtain the final set of uncorroded pixels and its corresponding center point , a subset of completely corroded pixels and and its corresponding center point and ;

[0035] The center point 、 and As the three vertices of the triangle, get the color space triangle in color space ,in, 、 and The color values ​​of the three vertices are respectively recorded as 、 and .

[0036] As a preferred embodiment of the present invention, the prompt points of the Segment Anything 2 model are obtained from the target structure, and the components of the high-voltage transmission line image to be detected are segmented according to the prompt points to obtain the regional images corresponding to the components in the high-voltage transmission line to be detected, specifically:

[0037] Select the node with the highest degree from the target structure. If the target structure has no node, select any point on the longest edge of the target structure as a node.

[0038] Set the node's ID to 1, the background points around the node and the key points of the node's corresponding connecting edges to 0, use the IDs as the cue points of the current component, and call the Segment Anything 2 model to generate the mask of the current component;

[0039] Extracting the corresponding part of the high-voltage transmission line image to be detected according to the mask and saving it as the regional image of the current component;

[0040] The masks generated by each component are traversed in a loop to obtain the regional images corresponding to each component in the high-voltage transmission line to be detected.

[0041] As a preferred embodiment of the present invention, depth calculation is performed on the segmented regional image to filter out the component regional image located in the foreground, specifically:

[0042] Calculate the depth of the regional image corresponding to each component, expressed as follows:

[0043] ;

[0044] Where, For components The corresponding regional image depth;

[0045] Get the maximum depth of all components according to the depth of the regional image corresponding to each component , filter out the component area image located in the foreground, wherein the component area image located in the foreground corresponds to the depth .

[0046] As a preferred embodiment of the present invention, the color values ​​of the pixels in the filtered component area image are projected onto the color space triangle to obtain the corresponding projection points, specifically:

[0047] along Towards and Draw a perpendicular line intersecting the point , is the rust change direction corresponding to the component area image pixel in the color space triangle, which is expressed as:

[0048] ;

[0049] Compute color space triangle Normal vector , expressed as:

[0050] ;

[0051] According to the normal vector Calculate the color value of any pixel in the component area image In color space triangle Projection point of the plane , expressed as:

[0052] ;

[0053] According to the projection point and the direction of rust change Get the corresponding projection point , expressed as:

[0054] .

[0055] As a preferred embodiment of the present invention, the quantitative value of the degree of corrosion of the corresponding projection point is calculated. The formula is:

[0056] .

[0057] As a preferred embodiment of the present invention, evaluation parameters in the component region image are counted, and the evaluation parameters include the average corrosion degree, the maximum corrosion degree, and the corrosion degree ratio of all pixels, which are expressed as follows:

[0058] ;

[0059] ;

[0060] ;

[0061] Where, is the average degree of rust, The maximum degree of rust, is the percentage of rust degree, For components The number of pixels, The degree of rust.

[0062] As a preferred embodiment of the present invention, the rust level of the component corresponding to the component region image is evaluated according to the evaluation parameters as follows:

[0063] The range of preset rust grades and their corresponding quantitative values ​​of rust degree, wherein the rust grades include light rust, moderate rust and severe rust, wherein the light rust is divided into level 1 rust and level 2 rust, the moderate rust is divided into level 3 rust and level 4 rust, and the severe rust is divided into level 5 rust and level 6 rust, wherein:

[0064] When 0≤maximum corrosion degree quantitative value<0.3, 0<average corrosion degree quantitative value<0.1, the component is in level 1 corrosion; when 0≤maximum corrosion degree quantitative value<0.3, 0.1≤average corrosion degree quantitative value<0.2, the component is in level 2 corrosion;

[0065] When the maximum rust degree quantitative value is ≥0.3 and the rust degree proportion quantitative value is <3%, the component is at level 3 rust;

[0066] When the maximum rust degree quantitative value is ≥0.3 and the rust degree proportion quantitative value is 3%≤<10%, the component is at level 4 rust;

[0067] When the maximum rust degree quantitative value is ≥0.3 and the rust degree proportion quantitative value is 10%≤<33%, the component is at level 5 rust;

[0068] When the maximum rust degree quantitative value is ≥0.3 and 33% is ≤ the rust degree ratio quantitative value, the component is at level 6 rust.

[0069] Compared with the prior art, the present invention has the following beneficial effects:

[0070] 1. This invention provides a method for detecting corrosion of steel components in power transmission lines. It uses the Depth Anything V2 model to extract the image foreground, making subsequent corrosion detection more targeted and reducing the impact of the image background on the detection results. It also generates prompt points for the Segment Anything 2 model based on skeleton generation technology, improving the accuracy of component segmentation.

[0071] 2. The present invention provides a method for detecting corrosion of steel components in power transmission lines. This method uses a K-means clustering algorithm to classify the set of completely corroded pixels, accurately defines the color space, and performs further analysis, making corrosion detection more detailed and reliable.

[0072] 3. The present invention provides a method for detecting rust on steel components of power transmission lines. By performing depth calculation on the segmented regional images, foreground component regional images are screened out, and the pixel color values ​​of these regional images are further projected into a color space triangle to quantitatively evaluate the degree of rust. The rust grade is determined based on the evaluation parameters. This method not only quantifies the degree of rust, but also enables a comprehensive evaluation based on multiple dimensions such as the proportion of rust, average rust degree, and maximum rust degree, making rust detection more comprehensive and accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0074] Figure 2 2 is a schematic diagram of a color space triangle according to an embodiment of the present invention. DETAILED DESCRIPTION

[0075] The specific embodiments of the present invention are described below to facilitate understanding of the present invention by those skilled in the art. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the appended claims, these changes are obvious, and all inventions and creations utilizing the concepts of the present invention are protected.

[0076] The present invention provides the following technical solution: a method for detecting corrosion of steel components of a power transmission line.

[0077] Example 1:

[0078] S1. Obtain a high-voltage transmission line corrosion image, remove the background and distant view from the high-voltage transmission line corrosion image based on the Depth Anything V2 model, and obtain a corresponding close-up image.

[0079] Preferably, the high voltage transmission line image is calculated for each pixel Corresponding depth , sort the depths of all pixels corresponding to the high-voltage transmission line image from large to small, and regard the pixels in the last 50% of the order as the background, and obtain the close-up image after removing the background and the corresponding rust pixel set ;

[0080] S2. Manually labeling the rust in the near-view image according to the degree of rust to obtain labeled data; obtaining a set of uncorroded pixels and a set of completely corroded pixels based on the labeled data; using a Kmean clustering algorithm to divide the set of completely corroded pixels into two disjoint subsets of completely corroded pixels of different types; calculating the center points of the set of uncorroded pixels and the set of completely corroded pixels; and defining a color space triangle based on the center points.

[0081] S21, manual annotation:

[0082] The rust degree value range is [0, 1], and the rust degree of the pixel contained in the area without rust in the close-up image is annotated The corresponding value is 0; the rust degree of the pixel contained in the area with the most serious rust in the close-up image is annotated The corresponding value is 1; the rust pixel set is obtained The corresponding annotation data, the annotation data includes the rust degree of the pixel And the color value coordinates ;

[0083] Preferably, the area selection during manual annotation can be any closed shape such as a polygon, a circle, etc., and it is not necessary to annotate the area without rust completely, and only a part of the area is selected for annotation; similarly, in the image, only a part of the area with the most serious rust is selected for annotation

[0084] S22, non-rust pixel set and completely rusted pixel set:

[0085] The rust on the high-voltage line is mainly composed of ferrous oxide (black), diiron trioxide (red-brown), triiron tetroxide (black), and iron hydroxide (red-brown), etc. Therefore, the area with serious rust usually appears black or red-brown. In this embodiment, in order to increase the detection accuracy, the K-Means algorithm is used to distinguish the black and red-brown colors, and the number of clusters is set to 2, and the clustering is performed on the rust pixel set, and the non-rust pixel set and the completely rusted pixel set are obtained according to the rust pixel set The formula is expressed as:

[0086] ;

[0087] ;

[0088] The average value of the red component of any pixel point in the completely rusted pixel set is calculated The formula is expressed as:

[0089] ;

[0090] In the formula, represents the number of pixels in the completely rusted pixel set

[0091] The completely rusted pixel set is divided into red components ​​​​​​​​Divide into subsets and , expressed as:

[0092] ;

[0093] ;

[0094] Calculate the subset and Center point and , expressed as:

[0095] ;

[0096] ;

[0097] Re-divide the subsets according to the center point, and iterate the division process until convergence, which can be expressed as:

[0098] ;

[0099] ;

[0100] ;

[0101] ;

[0102] Where, is the final number of iterations, ;

[0103] Get a subset and , and the corresponding center points are and ;

[0104] Set of uncorroded pixels , a subset of completely corroded pixels and To reduce, specifically, repeatedly calculate the center point of the set and delete the one farthest from the center point in the set points, among which To reduce the ratio, in this embodiment, Repeat the set reduction process until the center point of the set no longer changes, and obtain the final set of uncorroded pixels and its corresponding center point , a subset of completely corroded pixels and and its corresponding center point and ;

[0105] S23, color space triangle:

[0106] The center point 、 and As the three vertices of the triangle, get the color space triangle in color space ,in, 、 and The color values ​​of the three vertices are respectively recorded as 、 and ;

[0107] S3. Acquire an image of the high-voltage transmission line to be detected, remove the background and distant view in the image of the high-voltage transmission line to be detected based on the Depth Anything V2 model, and obtain a close-up image to be detected; process the close-up image to be detected using a skeleton generation technique to obtain a target structure, obtain prompt points of the Segment Anything 2 model from the target structure, perform component segmentation on the image of the high-voltage transmission line to be detected based on the prompt points, and obtain regional images corresponding to various components in the high-voltage transmission line to be detected; perform depth calculation on the segmented regional images, and filter out the component regional images located in the foreground;

[0108] S31, the step of obtaining the near-view image to be detected is the same as that in step S1, and will not be repeated here;

[0109] S32, component segmentation:

[0110] Select the highest-degree node from the target structure. If the target structure has no node, select any point on the longest edge of the target structure as a node. Preferably, in this embodiment, if there are multiple highest-degree nodes with the same degree, a node can be randomly selected.

[0111] The node's identifier is set to 1, and the identifiers of the background points around the node and the key points of the node's corresponding connecting edges are set to 0. The identifiers are used as the prompt points of the current component, and the Segment Anything 2 model is called to generate a mask for the current component. Preferably, in this embodiment, the node is surrounded by a neighborhood with a preset radius.

[0112] Extracting the corresponding part of the high-voltage transmission line image to be detected according to the mask and saving it as the regional image of the current component;

[0113] The masks generated by each component are traversed in a loop to obtain the regional images corresponding to each component in the high-voltage transmission line to be detected. If the generated mask overlaps with the masks of other components, the overlapping parts are assigned to the corresponding components based on the principle of proximity.

[0114] S33, the component region image located in the foreground:

[0115] The depth of each component corresponding region image is calculated, which is expressed in the formula as:

[0116] ;

[0117] In the formula, is the component corresponding region image depth;

[0118] According to the depth of each component corresponding region image, the maximum depth in all components is obtained , and the component region image located in the foreground is screened out, wherein the component region image located in the foreground corresponds to the depth ;

[0119] S4, the color value of the pixel in the screened component region image is projected to the color space triangle to obtain the corresponding projection point; the rust degree quantitative value of the corresponding projection point is calculated, and the evaluation parameters in the component region image are counted, the evaluation parameters including the average rust degree, the maximum rust degree and the rust degree proportion of all pixels, and the rust grade of the component corresponding to the component region image is evaluated according to the evaluation parameters;

[0120] S41, as shown in Figure 2 , a vertical line is drawn along the direction of and to intersect at point , is the rust change direction of the component region image pixel in the color space triangle, which is expressed in the formula as:

[0121] ; The normal vector of the color space triangle

[0122] is calculated, that is, the cross product of the two sides of the triangle composed of the color value is calculated, which is expressed in the formula as: ;

[0123] According to the normal vector , the projection point of the color value of any pixel point in the component region image in the color space triangle

[0124] plane is calculated , which is expressed in the formula as: ;

[0125] ;

[0126] According to the projection point​​ and the direction of rust change Get the corresponding projection point , expressed as:

[0127] ;

[0128] If the corresponding projection point If the value is close to 0, it means that there is no rust; if the value is close to 1, it means that the rust is serious.

[0129] S42. Calculate the quantitative value of the degree of corrosion at the corresponding projection point The formula is:

[0130] ;

[0131] S43. The evaluation parameters include the average corrosion degree, the maximum corrosion degree, and the corrosion degree ratio of all pixels, which are expressed as follows:

[0132] ;

[0133] ;

[0134] ;

[0135] Where, is the average degree of rust, The maximum degree of rust, is the percentage of rust degree, For components The number of pixels, is the degree of rust;

[0136] S44. Based on the ASTM D610 international standard, preset rust grades and their corresponding rust degree quantitative value ranges, wherein the rust grades include light rust, moderate rust, and severe rust. Light rust is divided into grade 1 and grade 2, moderate rust is divided into grade 3 and grade 4, and severe rust is divided into grade 5 and grade 6, wherein:

[0137] When 0≤maximum corrosion degree quantitative value<0.3, 0<average corrosion degree quantitative value<0.1, the component is in level 1 corrosion; when 0≤maximum corrosion degree quantitative value<0.3, 0.1≤average corrosion degree quantitative value<0.2, the component is in level 2 corrosion;

[0138] When the maximum rust degree quantitative value is ≥0.3 and the rust degree proportion quantitative value is <3%, the component is at level 3 rust;

[0139] When the maximum rust degree quantification value is greater than or equal to 0.3 and the rust degree proportion quantification value is greater than or equal to 10% and less than 33%, the component is in the 5th rust level.

[0140] When the maximum rust degree quantification value is greater than or equal to 0.3 and the rust degree proportion quantification value is greater than or equal to 10% and less than 33%, the component is in the 5th rust level.

[0141] When the maximum rust degree quantification value is greater than or equal to 0.3 and the rust degree proportion quantification value is greater than or equal to 10% and less than 33%, the component is in the 5th rust level.

[0142] The above merely illustrates the embodiments of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent flow transformation, or direct or indirect application in other related technical fields, which are made by using the content of the present application specification and drawings, are also included in the patent protection scope of the present application.

Claims

1. A method for detecting corrosion of steel components of a transmission line, characterized in that: The method comprises: Obtain a corroded image of a high-voltage transmission line and, using the Depth Anything V2 model, remove the background and distant view from the image to obtain the corresponding close-up image. Manually labeling the rust in the close-up image according to the degree of rust to obtain labeling data; obtaining a non-corroded pixel set and a completely corroded pixel set based on the labeling data, dividing the completely corroded pixel set into two disjoint completely corroded pixel subsets of different types using a Kmean clustering algorithm, calculating the center points of the non-corroded pixel set and the completely corroded pixel subsets, and defining a color space triangle based on the center points; Acquire an image of a high-voltage transmission line to be inspected, remove the background and distant view from the image based on the Depth Anything V2 model, and obtain a close-up image. Use skeleton generation technology to process the close-up image to obtain a target structure, obtain prompt points of the Segment Anything 2 model from the target structure, and segment the image of the high-voltage transmission line to be inspected based on the prompt points to obtain regional images corresponding to various components in the high-voltage transmission line to be inspected. Perform depth calculation on the segmented regional images to filter out the regional images of components located in the foreground. The color values ​​of the pixels in the screened component area image are projected onto the color space triangle to obtain the corresponding projection point; the quantitative value of the rust degree of the corresponding projection point is calculated, and the evaluation parameters in the component area image are counted. The evaluation parameters include the average rust degree, the maximum rust degree, and the rust degree ratio of all pixels. The rust grade of the component corresponding to the component area image is evaluated based on the evaluation parameters.

2. A method for detecting corrosion of steel components of a power transmission line according to claim 1, characterized in that: Based on the Depth Anything V2 model, the background and distant view in the high-voltage transmission line corrosion image are removed to obtain the close-up image. Specifically, the depth of each pixel in the high-voltage transmission line image is calculated. Corresponding depth , sort the depths of all pixels corresponding to the high-voltage transmission line image from large to small, and regard the pixels in the last 50% of the order as the background, and obtain the close-up image after removing the background and the corresponding rust pixel set .

3. A method for detecting corrosion of steel components of a power transmission line according to claim 2, characterized in that: The rust in the close-up image is manually annotated according to the degree of rust to obtain annotated data, wherein: The range of the rust degree is [0,1], and the rust degree is marked on the pixels in the area without rust in the close-up image. The corresponding value is 0; the most severely corroded area in the close-up image includes pixels with the degree of corrosion. The corresponding value is 1; get the rust pixel set Corresponding annotation data, including the degree of corrosion corresponding to the pixel and color value coordinates .

4. A method for detecting corrosion of steel components of a power transmission line according to claim 3, characterized in that: The center points of the non-corroded pixel set and the completely corroded pixel subset are calculated, and a color space triangle is defined based on the center points, specifically: Based on rust pixel set Get the uncorroded pixel set and a completely rusted pixel set , expressed as: ； ； Calculate the set of completely corroded pixels The red component corresponding to any pixel in The mean , expressed as: ； Where, Represents a completely corroded pixel set The number of pixels in ; Will completely rust pixel set According to the red component Divide into subsets and , expressed as: ； ； Calculate the subset and Center point and , expressed as: ； ； Re-divide the subsets according to the center point, and iterate the division process until convergence, which can be expressed as: ； ； ； ； Where, is the final number of iterations, ; Get a subset and , and the corresponding center points are and ; Set of uncorroded pixels , a subset of completely corroded pixels and To reduce, specifically, repeatedly calculate the center point of the set and delete the one farthest from the center point in the set points, among which To reduce the ratio; repeat the set reduction process until the center point of the set no longer changes, and obtain the final set of uncorroded pixels and its corresponding center point , a subset of completely corroded pixels and and its corresponding center point and ; The center point 、 and As the three vertices of the triangle, get the color space triangle in color space ,in, 、 and The color values ​​of the three vertices are respectively recorded as 、 and .

5. A method for detecting corrosion of steel components of a power transmission line according to claim 4, characterized in that: Obtain the cue points of the Segment Anything 2 model from the target structure, perform component segmentation on the image of the high-voltage transmission line to be detected based on the cue points, and obtain the regional images corresponding to the components in the high-voltage transmission line to be detected, specifically: Select the node with the highest degree from the target structure. If the target structure has no node, select any point on the longest edge of the target structure as a node. Set the node's ID to 1, the background points around the node and the key points of the node's corresponding connecting edges to 0, use the IDs as the cue points of the current component, and call the Segment Anything 2 model to generate the mask of the current component; Extracting the corresponding part of the high-voltage transmission line image to be detected according to the mask and saving it as the regional image of the current component; The masks generated by each component are traversed in a loop to obtain the regional images corresponding to each component in the high-voltage transmission line to be detected.

6. A method for detecting corrosion of steel components of a power transmission line according to claim 5, characterized in that: Perform depth calculation on the segmented region image and filter out the component region image in the foreground, specifically: Calculate the depth of the regional image corresponding to each component, expressed as follows: ； Where, For components The corresponding regional image depth; Get the maximum depth of all components according to the depth of the regional image corresponding to each component , filter out the component area image located in the foreground, wherein the component area image located in the foreground corresponds to the depth .

7. A method for detecting corrosion of steel components of a power transmission line according to claim 6, characterized in that: Project the color values ​​of the pixels in the filtered component area image onto the color space triangle to obtain the corresponding projection points, specifically: along Towards and Draw a perpendicular line intersecting the point , is the rust change direction corresponding to the component area image pixel in the color space triangle, which is expressed as: ； Compute color space triangle Normal vector , expressed as: ； According to the normal vector Calculate the color value of any pixel in the component area image In color space triangle Projection point of the plane , expressed as: ； According to the projection point and the direction of rust change Get the corresponding projection point , expressed as: 。 8. A method for detecting corrosion of steel components of a power transmission line according to claim 7, characterized in that: Calculate the quantitative value of the degree of corrosion at the corresponding projection point The formula is: 。 9. A method for detecting corrosion of steel components of a power transmission line according to claim 8, characterized in that: The evaluation parameters in the component area image are counted. The evaluation parameters include the average corrosion degree, the maximum corrosion degree, and the corrosion degree ratio of all pixels, which are expressed as follows: ； ； ； Where, is the average degree of rust, The maximum degree of rust, is the percentage of rust degree, For components The number of pixels, The degree of rust.

10. A method for detecting corrosion of steel components of a power transmission line according to claim 9, characterized in that: The rust level of the component corresponding to the component region image evaluated according to the evaluation parameters is specifically: The range of preset rust grades and their corresponding quantitative values ​​of rust degree, wherein the rust grades include light rust, moderate rust and severe rust, wherein the light rust is divided into level 1 rust and level 2 rust, the moderate rust is divided into level 3 rust and level 4 rust, and the severe rust is divided into level 5 rust and level 6 rust, wherein: When 0≤maximum corrosion degree quantitative value<0.3, 0<average corrosion degree quantitative value<0.1, the component is in level 1 corrosion; when 0≤maximum corrosion degree quantitative value<0.3, 0.1≤average corrosion degree quantitative value<0.2, the component is in level 2 corrosion; When the maximum rust degree quantitative value is ≥0.3 and the rust degree proportion quantitative value is <3%, the component is at level 3 rust; When the maximum rust degree quantitative value is ≥0.3 and the rust degree proportion quantitative value is 3%≤<10%, the component is at level 4 rust; When the maximum rust degree quantitative value is ≥0.3 and the rust degree proportion quantitative value is 10%≤<33%, the component is at level 5 rust; When the maximum rust degree quantitative value is ≥0.3 and 33% is ≤ the rust degree ratio quantitative value, the component is at level 6 rust.

Images