Power transmission line defect identification method, equipment and medium

Abstract

The invention discloses a power transmission line defect identification method and device and a medium, and belongs to the technical field of defect identification, and the method comprises the steps: collecting an original image of a to-be-identified power transmission line; scanning the original image in the form of a template to obtain a plurality of template images, and performing gray variance normalization on all the template images to obtain a gray variance normalized image of the original image; classifying all normalized gray value variances in the gray variance normalized image through a K-means clustering algorithm so as to identify a lead region and a background region of the gray variance normalized image; and carrying out damage detection on a lead area of the gray variance normalized image so as to identify a defect area of the power transmission line. According to the invention, defects of the power transmission line can be monitored and identified on line, and the accuracy of defect identification is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of defect identification, and specifically relates to a method, device and medium for identifying defects in a power transmission line. Background Art

[0002] With the continuous development of modern society, the stable and reliable operation of transmission lines is an important part of the construction of smart grids. The reliability and operation status of its equipment directly determine the stability and safety of the entire power system, as well as the quality and reliability of power supply. At present, the conductors of long-distance high-voltage transmission lines in my country mostly use steel-core aluminum stranded wires. When the overhead transmission line conductors exposed to the natural environment are online, they are always in a vibrating state in the vertical plane due to the influence of wind in the air (it is generally believed that the transmission line conductors are always in a breeze vibration state). This type of vibration has a significant impact on the stability and service life of the conductor and its accessories, and may cause problems such as damage to the surface of the transmission line conductor or broken strands. In extreme cases, these problems may also pose a threat to public safety.

[0003] Given my country's vast territory and diverse geographical environment, coupled with the influence of macro-climate and local terrain and meteorological conditions, accidents such as conductor and ground wire breakage and damage frequently occur during the long-term operation of transmission lines. The challenges faced by traditional manual inspection methods in assessing the operating conditions of transmission lines are becoming increasingly severe. This has led to the difficulty in accurately grasping the actual operating conditions of transmission lines and their potential faults, and the contradiction between the need to ensure the safe and stable operation of large power grids has become increasingly intensified. In recent years, with the widespread application of helicopters and drones in transmission line inspection tasks, the possibility of using image processing and recognition technology to realize automatic detection of transmission line faults has been significantly improved; however, the data collected by the image acquisition devices carried by drones and helicopters currently rely mainly on manual follow-up analysis, which is not only affected by personal subjective judgment, but also inefficient.

[0004] For some existing automatic identification technologies, since the transmission lines are affected by wind speed and airflow and there are breeze vibrations most of the time, the transmission line images obtained by the monocular camera are very likely to be blurred. At the same time, the surface of the wire presents a darker color characteristic. In addition, the influence of lighting conditions leads to uneven distribution of pixel grayscale values ​​in the wire area, resulting in a high grayscale value dispersion. Background elements such as trees, houses and roads are relatively blurred, which greatly reduces the accuracy of defect identification. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method, device and medium for identifying defects in a power transmission line, which can perform online monitoring and identification of defects in a power transmission line and improve the accuracy of identifying defects in a power transmission line.

[0006] The present invention provides the following technical solutions:

[0007] In a first aspect, a method for identifying defects in a transmission line is provided, comprising:

[0008] Collecting original images of the power transmission lines to be identified;

[0009] The original image is scanned in the form of a template to obtain multiple template images, and grayscale variance normalization is performed on all the template images to obtain a grayscale variance normalized image of the original image;

[0010] All normalized grayscale value variances in the grayscale variance normalized image are classified by K-means clustering algorithm to identify the wire area and background area of ​​the grayscale variance normalized image;

[0011] Damage detection is performed on the conductor area of ​​the grayscale variance normalized image to identify the defective area of ​​the transmission line.

[0012] Optionally, scanning the original image in the form of a template to obtain multiple template images, and performing grayscale variance normalization on all the template images to obtain a grayscale variance normalized image of the original image specifically includes:

[0013] Each pixel point is used as the template center one by one, and the original image is scanned to obtain multiple template images;

[0014] Based on the grayscale values ​​of all pixels in each template image, the mean grayscale value of each template image is obtained;

[0015] According to the grayscale values ​​and grayscale mean values ​​of all pixels in the template image, the grayscale value variance of the template center is obtained;

[0016] Based on the grayscale value variances of all template centers, the grayscale value variance of each template center is normalized to obtain a grayscale variance normalized image of the original image.

[0017] Optionally, the grayscale value variance of each template center is normalized based on the grayscale value variance of all template centers. The specific formula is:

[0018]

[0019] Among them, q (x,y) is the gray value variance of the pixel point (x, y) corresponding to the template center, q min and q max is the minimum and maximum value of the gray value variance of all pixels; H(x,y) is the normalized gray value variance of the pixel (x,y) corresponding to the template center.

[0020] Optionally, the step of classifying all normalized grayscale value variances in the grayscale variance normalized image by using a K-means clustering algorithm to identify a wire region and a background region of the grayscale variance normalized image specifically includes:

[0021] Each normalized grayscale value variance in the grayscale variance normalized image is input as a sample into the K-means clustering algorithm, and any two normalized grayscale value variances are selected as the initial codebook;

[0022] The initial codebook is separated to obtain two codes with doubled dimension, and the two codes with doubled dimension form the current codebook;

[0023] For each sample H(x,y), find the code with the closest Euclidean distance in the current codebook and assign the sample H(x,y) to the corresponding cluster;

[0024] For each cluster, the code of the cluster is updated according to the modulus values ​​of all samples in the cluster to form an updated codebook;

[0025] Reclassify all samples according to the updated codebook;

[0026] It is determined whether the codebook update termination condition is met. If not, the codebook is continuously updated until the codebook update termination condition is met. If it is met, two clusters after classification are output, where the cluster with small code is the wire area, and the cluster with large code is the background area.

[0027] Optionally, the initial codebook is separated to obtain two codes with doubled dimension, and the specific formula is:

[0028] Q1=Q0(1+ε)

[0029] Q2=Q0(1-ε)

[0030] Among them, Q1 and Q2 are two codes after the dimension is doubled, Q0 is the initial codebook, and ε is the set error variable threshold.

[0031] Optionally, in the determining whether a codebook update termination condition is met, the codebook update termination condition is:

[0032]

[0033]

[0034] Where D′ is the distortion value before the codebook is updated, D is the distortion value after the codebook is updated, ε0 is the preset threshold, N and M are the length and width of the grayscale variance normalized image, respectively, and q c(x,y) is the code of the cluster to which the sample H(x,y) belongs, d(H(x,y),qc(x,y) ) is H(x,y) and q c(x,y) The distance between.

[0035] Optionally, the performing damage detection on the conductor area of ​​the grayscale variance normalized image to identify the defective area of ​​the transmission line specifically includes:

[0036] Project the wire area of ​​the grayscale variance normalized image in the horizontal direction to obtain the normalized grayscale value variance of each column of pixels;

[0037] The single strand diameter of the transmission line is used as the minimum length, and the normalized grayscale value of the pixel corresponding to the minimum length is used as the correction item to correct the variance of the normalized grayscale value of each column of pixels, so as to obtain the optimized normalized grayscale value variance of each column of pixels of the transmission line;

[0038] Through the threshold segmentation method, according to the optimized normalized gray value variance of each column of the transmission line pixel points, the damage threshold of whether there is a defect in the transmission line is obtained;

[0039] It is determined in turn whether the variance of the normalized grayscale value of each column of pixels of the transmission line after optimization is greater than the damage threshold. If so, there is no defect in the pixel column; otherwise, there is a defect in the pixel column.

[0040] Optionally, the correction term n y for:

[0041]

[0042] The obtaining of the normalized grayscale value variance of each column of pixels of the transmission line after optimization includes obtaining the normalized grayscale value variance of the pixels of the yth column after optimization;

[0043] The normalized gray value variance M of the optimized pixel points in the yth column y for:

[0044] M=mn

[0045] yyy

[0046] The damage threshold T of whether the transmission line has defects is:

[0047] T=(M′ max +M′ min ) / 2

[0048] Among them, m y is the normalized grayscale value variance of the yth column of pixels in the conductor area; i is the number of pixels occupied by a single-strand conductor in the length corresponding to its radius value, r is the number of pixels occupied by a single-strand conductor in the width corresponding to its radius value, X is the number of pixels occupied by the width of the conductor area, and M′max and M′ min are respectively the maximum and minimum values ​​of the normalized gray value variance after optimization in all columns of the wire area.

[0049] In a second aspect, a computer device is provided, comprising a processor and a memory; wherein, when the processor executes a computer program stored in the memory, the steps of the power transmission line defect identification method described in any one of the first aspects are implemented.

[0050] In a third aspect, a computer-readable storage medium is provided for storing a computer program; when the computer program is executed by a processor, the steps of the power transmission line defect identification method described in any one of the first aspects are implemented.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] The present invention enables the operator to accurately perform online wire defect identification without electrical contact with the transmission line, thereby greatly improving the efficiency and reliability of transmission line safety detection; in addition, the present invention takes into account the highly dispersed grayscale value of the wire area in the transmission line image, and adopts grayscale variance normalization technology to process these complex image features, so that the features of the wire area are more prominent, thereby improving the accuracy of defect identification; in addition, the present invention also avoids the identification error caused by illumination changes, thereby further improving the accuracy of transmission line defect identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a flow chart of the steps of the power transmission line defect identification method of the present invention;

[0054] Figure 2 is a flow chart of the K-means clustering algorithm of the present invention;

[0055] Figure 3 is an original image of the power transmission line of the present invention;

[0056] Figure 4 It is the binary segmentation image of the K-means clustering algorithm of the present invention;

[0057] Figure 5 is the wire area of ​​the grayscale variance normalized image of the present invention;

[0058] Figure 6 This is a potential damage area identification diagram of the conductor of the present invention. DETAILED DESCRIPTION

[0059] The present invention will be further described below in conjunction with the accompanying drawings. The following examples are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the scope of protection of the present invention. It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the term "comprising" and any of its variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0060] Example 1

[0061] like Figure 1 As shown, a method for identifying defects in a transmission line comprises the following steps:

[0062] S1: Collect the original image of the power transmission line to be identified.

[0063] like Figure 3 As shown, the data is collected by a drone equipped with a monocular camera.

[0064] S2: Scan the original image in the form of a template to obtain multiple template images, and perform grayscale variance normalization on all the template images to obtain a grayscale variance normalized image of the original image.

[0065] Through a process with the center of the template as the reference point, each pixel of the entire image is scanned one by one. In this process, the gray value variance of all pixels in the template is used to update the pixel value at the center of the template.

[0066] Specifically, step S2 includes the following sub-steps:

[0067] S21: Taking each pixel point as the template center one by one, scanning the original image to obtain multiple template images.

[0068] According to the resolution of the actual image, the most suitable template size is flexibly selected. In view of the small size of most transmission line images obtained by cameras, the present invention recommends using a 3×3 template for full-image scanning. At the same time, with the advancement of imaging technology and computing performance, when processing larger images, the size of the template can also be appropriately expanded to further improve the processing effect.

[0069] S22: Based on the grayscale values ​​of all pixels in each template image, obtain the grayscale value mean of each template image.

[0070]

[0071] Among them, Q is the mean gray value of each pixel in a template image, Q i is the gray value of the i-th pixel in the template image.

[0072] S23: Obtain the grayscale value variance of the template center according to the grayscale values ​​and grayscale mean values ​​of all pixels in the template image.

[0073]

[0074] Among them, q k is the variance of the grayscale value of the template center of the kth template image.

[0075] S24: Based on the grayscale value variances of all template centers, normalize the grayscale value variance of each template center to obtain a grayscale variance normalized image of the original image.

[0076] The specific formula is:

[0077]

[0078] Among them, q (x,y) is the gray value variance of the pixel point (x, y) corresponding to the template center, q min and q max is the minimum and maximum value of the gray value variance of all pixels; H(x,y) is the normalized gray value variance of the pixel (x,y) corresponding to the template center.

[0079] S3: Classify all normalized grayscale value variances in the grayscale variance normalized image through a K-means clustering algorithm to identify the wire area and the background area of ​​the grayscale variance normalized image.

[0080] like Figure 2 As shown, step S3 specifically includes the following sub-steps:

[0081] S31: Each normalized grayscale value variance in the grayscale variance normalized image is input as a sample into the K-means clustering algorithm, and any two normalized grayscale value variances are selected as an initial codebook.

[0082] Q0=(q1,q2)

[0083] q1 and q2 are two codes of the initial codebook.

[0084] S32: Separate the initial codebook to obtain two codes with doubled dimensions, and the two codes with doubled dimensions form the current codebook.

[0085] The specific formula is:

[0086] Q1=Q0(1+ε)

[0087] Q2=Q0(1-ε)

[0088] Among them, Q1 and Q2 are two codes after the dimension is doubled, Q0 is the initial codebook, and ε is the set error variable threshold.

[0089] S33: For each sample H(x,y), find the code with the closest Euclidean distance in the current codebook, and assign the sample H(x,y) to the corresponding cluster.

[0090]

[0091] Where c(x,y) is the cluster to which H(x,y) belongs, c(x,y)∈{1,2}, x∈[1,N], y∈[1,M], Q j is the codeword corresponding to the jth cluster in the existing codebook, d(H(x,y),Q j ) is H(x,y) and Q j Distance, usually Euclidean distance.

[0092] S34: For each cluster, update the code of the cluster according to the module values ​​of all samples in the cluster to form an updated codebook.

[0093] The specific method of updating the clustered code can refer to the existing technology.

[0094] S35: Reclassify all samples according to the updated codebook.

[0095] S36: Determine whether the codebook update termination condition is met. If not, continuously update the codebook until the codebook update termination condition is met. If met, output two classified clusters, wherein the cluster with a small code is the conductor area, and the cluster with a large code is the background area.

[0096] The codebook update termination condition is:

[0097]

[0098]

[0099] Where D′ is the distortion value before the codebook is updated, D is the distortion value after the codebook is updated, ε0 is the preset threshold, N and M are the length and width of the grayscale variance normalized image, respectively, and q c(x,y)is the code of the cluster to which the sample H(x,y) belongs, d(H(x,y),q c(x,y) ) is H(x,y) and q c(x,y) The distance between.

[0100] Each time all data objects are processed, a new median is generated that more accurately reflects the characteristics of the clusters. Subsequently, all data objects are reallocated to the best matching clusters based on these updated medians. With each iteration, the position of the median is gradually adjusted until a stable state is reached, where the median no longer shifts, and the iteration process ends. Through this series of operations, the grayscale variance normalized image is converted into a segmented binary image according to the preset goal of distinguishing the wire area from the background area, as shown in Figure 1. Figure 4 shown.

[0101] S4: Perform damage detection on the conductor area of ​​the grayscale variance normalized image to identify the defective area of ​​the transmission line.

[0102] like Figure 5 and Figure 6 As shown, step S4 specifically includes the following sub-steps:

[0103] S41: Project the wire area of ​​the grayscale variance normalized image in the horizontal direction to obtain the normalized grayscale value variance of each column of pixels.

[0104]

[0105] Among them, m y is the normalized grayscale value variance of the pixels in the yth column of the wire area.

[0106] S42: Using the single strand diameter of the transmission line as the minimum length, using the normalized grayscale value of the pixel corresponding to the minimum length as a correction item, correcting the normalized grayscale value variance of each column of pixels, and obtaining the optimized normalized grayscale value variance of each column of pixels of the transmission line.

[0107] In order to reduce the difference between the normalized grayscale variance values ​​of the pixels of each section of the transmission line caused by the uneven illumination on the surface of the transmission line, the single-strand diameter of the transmission line is used as the minimum length, and the normalized grayscale value of the pixel corresponding to the minimum length is used as the correction item, thereby reducing the impact of changes in lighting conditions on the detection results and improving the accuracy of damage identification.

[0108] Correction Item n y for:

[0109]

[0110] Among them, i is the number of pixels occupied by the length of a single-strand wire corresponding to its radius value, r is the number of pixels occupied by the width of a single-strand wire corresponding to its radius value, and X is the number of pixels occupied by the width of the wire area.

[0111] The normalized gray value variance M of the optimized pixel points in the yth column y for:

[0112] M y =m y -n y

[0113] S43: By using a threshold segmentation method, according to the optimized normalized gray value variance of each column of pixel points of the transmission line, a damage threshold value for determining whether the transmission line has defects is obtained.

[0114] The damage threshold T of whether the transmission line has defects is:

[0115] T=(M′ max +M′ min ) / 2

[0116] Among them, M′ max and M′ min are respectively the maximum and minimum values ​​of the normalized gray value variance after optimization in all columns of the wire area.

[0117] S44: determining in turn whether the variance of the normalized grayscale values ​​of the pixels in each column of the transmission line after optimization is greater than the damage threshold; if so, the pixels in this column do not have defects; otherwise, the pixels in this column do have defects.

[0118] If M y >T, then the pixel point in the yth column has no defect, otherwise, the pixel point in the yth column has a defect.

[0119] The present invention is no longer limited to directly using the grayscale value of each pixel in the image for processing, but turns to analyzing the dispersion degree of these grayscale values, which can overcome the influence of lighting conditions and improve the accuracy of transmission line defect recognition.

[0120] In some other embodiments, the lateral projection result map can convert image information into chart information, and can further highlight the damaged area of ​​the transmission line after threshold processing.

[0121] Example 2

[0122] The present invention provides a computer device, comprising a processor and a memory; wherein the processor implements the steps of the above-mentioned power transmission line defect identification method when executing a computer program stored in the memory.

[0123] For more specific processes of the above method, please refer to the corresponding contents disclosed in the aforementioned embodiments, which will not be repeated here.

[0124] Example 3

[0125] The present invention provides a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, the steps of the above-mentioned power transmission line defect identification method are implemented.

[0126] For more specific processes of the above method, please refer to the corresponding contents disclosed in the aforementioned embodiments, which will not be repeated here.

[0127] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device and storage medium disclosed in the embodiment, since they correspond to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0128] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution in the embodiments of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a disk, an optical disk, etc., and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention or some parts of the embodiments.

[0129] The above are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims

1. A method for identifying defects in a power transmission line, characterized in that: include: Collecting original images of the power transmission lines to be identified; The original image is scanned in the form of a template to obtain multiple template images, and grayscale variance normalization is performed on all the template images to obtain a grayscale variance normalized image of the original image; All normalized grayscale value variances in the grayscale variance normalized image are classified by K-means clustering algorithm to identify the wire area and background area of ​​the grayscale variance normalized image; Damage detection is performed on the conductor area of ​​the grayscale variance normalized image to identify the defective area of ​​the transmission line.

2. The method for identifying power transmission line defects according to claim 1, characterized in that: The method of scanning the original image in the form of a template to obtain a plurality of template images, and performing grayscale variance normalization on all the template images to obtain a grayscale variance normalized image of the original image specifically includes: Each pixel point is used as the template center one by one, and the original image is scanned to obtain multiple template images; Based on the grayscale values ​​of all pixels in each template image, the mean grayscale value of each template image is obtained; According to the grayscale values ​​and grayscale mean values ​​of all pixels in the template image, the grayscale value variance of the template center is obtained; Based on the grayscale value variances of all template centers, the grayscale value variance of each template center is normalized to obtain a grayscale variance normalized image of the original image.

3. The method for identifying power transmission line defects according to claim 2, characterized in that: Based on the grayscale value variance of all template centers, the grayscale value variance of each template center is normalized. The specific formula is: Among them, q (x,y) is the gray value variance of the pixel (x, y) corresponding to the template center, q min and q max is the minimum and maximum value of the grayscale value variance of all pixels; H(x,y) is the normalized grayscale value variance of the pixel (x,y) corresponding to the template center.

4. The method for identifying power transmission line defects according to claim 1, characterized in that: The method of classifying all normalized grayscale value variances in the grayscale variance normalized image by using a K-means clustering algorithm to identify the wire area and background area of ​​the grayscale variance normalized image specifically includes: Each normalized grayscale value variance in the grayscale variance normalized image is input as a sample into the K-means clustering algorithm, and any two normalized grayscale value variances are selected as the initial codebook; The initial codebook is separated to obtain two codes with doubled dimensions, and the two codes with doubled dimensions form the current codebook; For each sample H(x,y), find the code with the closest Euclidean distance in the current codebook and assign the sample H(x,y) to the corresponding cluster; For each cluster, the code of the cluster is updated according to the modulus values ​​of all samples in the cluster to form an updated codebook; Reclassify all samples according to the updated codebook; It is determined whether the codebook update termination condition is met. If not, the codebook is continuously updated until the codebook update termination condition is met. If it is met, two clusters after classification are output, where the cluster with small code is the wire area, and the cluster with large code is the background area.

5. The method for identifying power transmission line defects according to claim 1, characterized in that: The initial codebook is separated to obtain two codes with doubled dimension. The specific formula is: Q1=Q0(1+ε) Q2=Q0(1-ε) Among them, Q1 and Q2 are two codes after the dimension is doubled, Q0 is the initial codebook, and ε is the set error variable threshold.

6. The method for identifying power transmission line defects according to claim 1, characterized in that: In the determination of whether the codebook update termination condition is met, the codebook update termination condition is: Where D′ is the distortion value before the codebook is updated, D is the distortion value after the codebook is updated, ε0 is the preset threshold, N and M are the length and width of the grayscale variance normalized image, respectively, and q c(x,y) is the code of the cluster to which the sample H(x,y) belongs, d(H(x,y),q c(x,y) ) is H(x,y) and q c(x,y) The distance between.

7. The method for identifying power transmission line defects according to claim 1, characterized in that: The damage detection of the conductor area of ​​the grayscale variance normalized image to identify the defective area of ​​the transmission line specifically includes: Project the wire area of ​​the grayscale variance normalized image in the horizontal direction to obtain the normalized grayscale value variance of each column of pixels; The single strand diameter of the transmission line is used as the minimum length, and the normalized grayscale value of the pixel corresponding to the minimum length is used as the correction item to correct the variance of the normalized grayscale value of each column of pixels, so as to obtain the optimized normalized grayscale value variance of each column of pixels of the transmission line; Through the threshold segmentation method, according to the optimized normalized gray value variance of each column of the transmission line pixel points, the damage threshold of whether there is a defect in the transmission line is obtained; It is determined in turn whether the variance of the normalized grayscale value of each column of pixels of the transmission line after optimization is greater than the damage threshold. If so, there is no defect in the pixel column; otherwise, there is a defect in the pixel column.

8. The method for identifying power transmission line defects according to claim 1, characterized in that: The correction term n y for: The obtaining of the normalized grayscale value variance of each column of pixels of the transmission line after optimization includes obtaining the normalized grayscale value variance of the pixels of the yth column after optimization; The normalized gray value variance M of the optimized pixel points in the yth column y for: M y =m y -n y The damage threshold T of whether the transmission line has defects is: T=(M′ max +M′ min ) / 2 Among them, m y is the normalized grayscale value variance of the yth column of pixels in the conductor area; i is the number of pixels occupied by a single-strand conductor in the length corresponding to its radius value, r is the number of pixels occupied by a single-strand conductor in the width corresponding to its radius value, X is the number of pixels occupied by the width of the conductor area, and M′ max and M′ min are respectively the maximum and minimum values ​​of the normalized gray value variance after optimization in all columns of the wire area.

9. A computer device, characterized in that: It comprises a processor and a memory; wherein, when the processor executes the computer program stored in the memory, the steps of the power transmission line defect identification method described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium, characterized in that: Used to store computer programs; when the computer programs are executed by a processor, the steps of the power transmission line defect identification method according to any one of claims 1 to 8 are implemented.

Images