Embedded power transmission line insulator string defect identification method, system and device and storage medium

Abstract

The invention discloses an embedded power transmission line insulator chain defect identification method, system and device and a storage medium, and belongs to the technical field of defect identification, and the method comprises the steps: obtaining an insulator chain image, carrying out the insulator chain region positioning of the insulator chain image through color space change and image processing, and obtaining the coordinates of a target insulator chain; based on the target insulator chain coordinate, feature extraction is carried out on an insulator sheet, and an insulator sheet state vector is constructed; judging the state vector of the insulator sheet to obtain a suspected defective insulator list; calculating the insulator sheets in the suspected defect insulator list and detecting the surfaces of the insulator sheets to obtain the defect types and positions of the surfaces of the insulator sheets; and according to the defect type and position, the associated defect type code and description, generating structured power transmission line insulator defect report data, thereby improving the accuracy of defect identification, the fineness of defect classification and the practicability of a result report.

Description

Technical Field

[0001] The present invention relates to the technical field of defect identification, and in particular to a method, system, device and storage medium for identifying defects in an embedded transmission line insulator string. Background Art

[0002] The embedded transmission line insulator string defect recognition system is a specialized automated inspection device integrated into a specific hardware platform (such as a drone payload). Its primary purpose is to automatically analyze images or video data of transmission line insulator strings captured by drones and other equipment, identifying possible defects such as breakage, cracks, contamination, flashover traces, and spontaneous explosions on the insulators in real time or near real time, and determining the defect type and location.

[0003] Existing technologies lack mechanisms for proactively optimizing image acquisition, resulting in unstable image quality in complex and changing field environments. For example, motion blur caused by flight turbulence or overexposure or underexposure caused by sudden changes in illumination directly impact the reliability of subsequent analysis and may obscure subtle defects. Regarding target positioning, when obscured by tower structures or with a similar background color, positioning errors can introduce non-target areas or omit target areas, compromising the completeness and accuracy of analysis. Therefore, improvements are needed. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is to solve the problem in the existing technology that the quality of the acquired images is unstable in complex and changeable outdoor environments due to the mechanism of actively optimizing image acquisition.

[0006] To solve the above technical problems, the present invention provides the following technical solution: an embedded transmission line insulator string defect identification method, comprising: Acquire an insulator string image, and locate the insulator string region on the insulator string image using color space variation and image processing to obtain the coordinates of the target insulator string; Based on the target insulator string coordinates, feature extraction is performed on the insulator segment to construct an insulator segment state vector; Judging the state vector of the insulator segment to obtain a list of suspected defective insulators; Calculating the insulator pieces in the suspected defective insulator list and inspecting the surfaces of the insulator pieces to obtain the defect types and locations on the surfaces of the insulator pieces; According to the defect type and location, the defect category code and description are associated to generate structured transmission line insulator defect report data.

[0007] As a preferred solution of the embedded transmission line insulator string defect identification method of the present invention, locating the insulator string area includes: Read the channel pixel values of the insulator string image, calculate the color element corresponding to each pixel, and obtain the image in the color space; Setting a first threshold to filter the pixel points of the image in the color space to obtain a target color pixel set; Traverse the coordinate axes of the target color pixel set, determine the maximum and minimum coordinate information, define a rectangular area surrounding all target color pixels, record the vertex coordinates, width and height of the rectangular area, and obtain a preliminary candidate frame; The gradient operator is applied to the image area within the preliminary candidate frame to calculate the gradient of each point, and non-maximum suppression and double threshold processing are performed. The edges are connected to form a continuous edge contour, which is confirmed as the boundary of the insulator string and the coordinates of the target insulator string are established.

[0008] The beneficial effect of this preferred technical solution is that it accurately locates the insulator string area through color screening and edge detection, effectively eliminates background interference, lays the foundation for subsequent defect identification, and significantly improves detection accuracy and efficiency.

[0009] As a preferred solution of the embedded transmission line insulator string defect identification method of the present invention, constructing the insulator segment state vector includes: Within the calibrated insulator string image area, the horizontal projection of each row of pixel intensity is calculated. The gap between the insulator segments is located by detecting the periodic troughs of the projection curve. The hardware is then located by performing sliding window matching using a preset hardware shape template. The independent image area of each insulator segment and the associated hardware area are segmented to obtain a single-segment image area set. For each independent image region of the insulator slice in the single image region set, the pixel values of the color region are converted into grayscale representation, the gray-level co-occurrence matrix is constructed, the texture feature parameters are calculated, and the average value and dispersion degree of the channel pixels of the original region are calculated respectively to obtain the visual parameters of the insulator slice;

[0010] Based on the independent image area and visual parameters of each insulator segment in the single image area set, the edge detection operator is used to obtain the outer boundary point set of the insulator segment, determine the contour line of the shed, calculate the maximum span of the contour line in the horizontal direction as the shed diameter, and measure the thickness of the central part along the axial direction of the insulator. All parameters are integrated to construct the insulator segment state vector.

[0011] The beneficial effect of this preferred technical solution is to comprehensively extract the texture, color and morphological features of the insulator sheet, construct a multi-dimensional state vector, and provide rich and accurate characterization information for defect detection.

[0012] As a preferred solution of the embedded transmission line insulator string defect identification method of the present invention, the determination of the insulator segment state vector includes: Obtaining a database of normal insulator segment sample parameters; The mean and standard deviation are calculated based on the angular second-order matrix and contrast texture features, the threshold range is defined, the upper and lower limits of the color standard deviation and the shed morphological parameters are set as the allowable deviation, and the defect judgment threshold set is established; Based on the defect judgment threshold set and the state vector of each insulator segment, the current value of each parameter in the vector is read in sequence. The angular second-order moment and contrast value are compared with the threshold interval, the color standard deviation is compared with the upper threshold, and the shed diameter and thickness values are compared with the allowable deviation range. If any parameter value falls outside the corresponding normal range, the parameter is marked as abnormal and a parameter deviation indication is obtained. Determine the marking result of each insulator segment based on the parameter deviation indication of each insulator segment; If at least one abnormal parameter mark is included, the insulator segment is determined to be a suspected defective individual. The identifiers of all suspected defective insulator segments and the corresponding insulator segment state vectors are collected and compiled into a list to generate a suspected defective insulator list.

[0013] The beneficial effect of this preferred technical solution is that, by comparing multi-dimensional parameters with thresholds, suspected defective insulator pieces can be accurately screened, effectively reducing the misjudgment rate and improving the accuracy and reliability of defect detection.

[0014] As a preferred solution of the method for identifying defects in an embedded transmission line insulator string according to the present invention, the defect types and locations on the surface of the insulator piece include: Applying a gradient filter to the image blocks of suspected defective insulator pieces, calculating the pixel intensity change rates in the coordinate axis directions, synthesizing the gradient amplitude images, and obtaining a high-contrast surface image;

[0015] Based on the high-contrast surface image and the corresponding insulator sheet state vector, chain code tracking or path search is used to connect the gradient pixel points in the image, identify linear structures whose length and curvature meet the crack conditions, and perform threshold segmentation on the original color image area to extract abnormal dark patches. The geometric parameters and grayscale distribution of the patches are calculated to obtain the defect appearance feature set.

[0016] As a preferred solution of the embedded transmission line insulator string defect identification method of the present invention, it further includes: Based on the defect appearance feature set, the linear structure features, dark patch parameters and the quantitative features of various types of defects in the pre-stored defect sample library are similarly calculated, and the estimated defect category is assigned. The degree of damage is evaluated based on the numerical range of the feature parameters to obtain the defect type and location.

[0017] As a preferred solution of the embedded transmission line insulator string defect identification method of the present invention, generating structured transmission line insulator defect report data includes: Based on the defect type and location, a local image of the defective insulator segment is cropped from the original high-resolution image. At the same time, the installation serial number, three-dimensional spatial coordinates, defect category identifier determined by the system, and corresponding text description of the defect are retrieved to form a single defect holographic file. According to the single defect holographic file, a universal data exchange format is selected, and the image slices, digitized serial numbers and coordinates, standardized defect codes, and coded description text in the file are sequentially filled into the preset key-value pair template to form a record and obtain a standardized defect entry; Based on the standardized defect items, a list or array structure is created, each standardized defect item is added as an element, and the list is sorted according to the tower identification of the inspection task or the severity level of the defect to generate structured transmission line insulator defect report data.

[0018] The present invention provides an embedded transmission line insulator string defect identification system.

[0019] To solve the above technical problems, the present invention provides the following technical solution: an embedded transmission line insulator string defect identification system, comprising:

[0020] A positioning module is used to obtain an insulator string image, locate the insulator string area on the insulator string image using color space change and image processing, and obtain the coordinates of the target insulator string; A feature extraction module is used to extract features of the insulator segment based on the target insulator string coordinates and construct an insulator segment state vector; a potential defect detection module, configured to determine the state vector of the insulator segment and obtain a list of suspected defective insulators; a defect type discrimination module, configured to calculate the insulator pieces in the list of suspected defective insulators and detect the surfaces of the insulator pieces to obtain the defect types and locations on the surfaces of the insulator pieces; The defect information output module is used to associate the defect category code and description according to the defect type and location, and generate structured transmission line insulator defect report data.

[0021] The present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the embedded transmission line insulator string defect identification method are implemented.

[0022] The present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the embedded transmission line insulator string defect identification method are implemented.

[0023] The beneficial effects of the present invention are as follows: by actively regulating the camera posture and focal length, and combining the optimization of contrast and clarity in the image sequence and specific image correction steps, the basic quality of the input image is guaranteed, laying the foundation for subsequent precise analysis; then, color space transformation and threshold segmentation are used to preliminarily screen the target pixels, and the minimum bounding rectangle and gradient edge detection and shape analysis within the frame are combined to achieve the calibration of the insulator string area, effectively eliminating background interference. For the calibration area, the segmentation of individual insulator pieces and hardware is completed by analyzing pixel projection and structural features, and the grayscale co-occurrence matrix of each insulator piece is calculated to extract texture features. Combined with color statistics and morphological parameters such as the diameter and thickness of the umbrella skirt, a comprehensive state vector is constructed to enrich the representation dimension. This vector is used to perform an objective comparison with the preset multi-dimensional threshold to effectively screen out potentially abnormal insulator pieces. Ultimately, by calculating pixel directional derivatives to enhance surface details in suspected defect areas and detecting specific abnormal patterns such as continuous high-gradient line segments or irregular dark areas, accurate identification and positioning of specific defect types such as cracks, flashovers, and self-explosions were achieved. At the same time, combined with severity level assessment, structured data containing image slices, positioning information, category codes, and descriptions was generated, which improved the accuracy of defect identification, the precision of defect classification, and the practicality of result reports. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0025] Figure 1 This is an overall flow chart of an embedded transmission line insulator string defect identification method provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0026] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0027] Example 1, reference Figure 1, is an embodiment of the present invention, which provides an embedded transmission line insulator string defect identification method, comprising: S100: Acquire an insulator string image, locate the insulator string region on the insulator string image using color space change and image processing, and obtain the coordinates of the target insulator string; S200: Based on the coordinates of the target insulator string, extract features of the insulator segment and construct an insulator segment state vector; S300: judging the state vector of the insulator segment and obtaining a list of suspected defective insulators; S400: Calculating the insulator segments in the suspected defective insulator list and inspecting the surfaces of the insulator segments to obtain defect types and locations on the surfaces of the insulator segments; S500: According to the defect type and location, the defect category code and description are associated to generate structured transmission line insulator defect report data.

[0028] It should be noted that by actively optimizing image acquisition, accurately locating the insulator string area, comprehensively extracting features and accurately judging defects, efficient and accurate identification of transmission line insulator defects is achieved, which significantly improves detection efficiency and reliability, reduces manual inspection costs, and ensures the safe operation of the power grid.

[0029] In the embodiment of the present invention, the above step S100 includes the following sub-steps A1-A4; In A1: Read the channel pixel values of the insulator string image, calculate the color element corresponding to each pixel, and obtain the image in color space; In A2: a first threshold is set to filter the pixels of the image in the color space to obtain a target color pixel set; In A3: traverse the coordinate axes of the target color pixel set, determine the maximum and minimum coordinate information, define a rectangular area surrounding all target color pixels, record the vertex coordinates, width, and height of the rectangular area, and obtain a preliminary candidate frame; In A4: Apply the gradient operator to the image area within the preliminary candidate frame to calculate the gradient of each point, perform non-maximum suppression and double threshold processing, connect the edges to form a continuous edge contour, confirm it as the boundary of the insulator string, and establish the coordinates of the target insulator string.

[0030] Specifically, the acquisition of insulator string images includes sending control commands to the pitch motor and rotation motor of the camera gimbal to adjust the shooting angle according to the UAV flight attitude parameters and lens parameters, and driving the lens focus motor to adjust the relative position of the lens group, change the focal length value, and obtain the camera ready state; based on the camera ready state, with an exposure time of 1 / 800 second and a sensitivity of ISO100, the dynamic image of the transmission line insulator string is continuously captured, and the image pixel data of each frame is stored in the onboard storage unit to generate the original image data set; traversing each frame, the grayscale histogram standard deviation is calculated as the contrast value, the pixel displacement between frames is analyzed to evaluate motion blur, the frame with the largest standard deviation and the smallest displacement is selected, and the lens distortion model parameter geometric correction is applied to obtain a clear insulator string image.

[0031] Specifically, the intensity values of the red, green, and blue channels of the clear insulator string image are read one by one. Usually, the value range of each channel is 0 to 255. For each pixel, the RGB (red, green, blue) value is converted to the HSV (hue, saturation, value) value using a standard color space conversion algorithm. For example, the hue is calculated based on the maximum and minimum values of red, green, and blue, and is usually mapped to a range of 0 to 360 degrees or normalized to a range of 0 to 1. Saturation indicates the purity of color and is calculated as: , in, is the minimum value among red, green and blue. It is the maximum value among red, green and blue; Brightness is The normalized value of is used to complete the conversion of the entire image to form an HSV image; The first threshold is the preset hue range and saturation minimum threshold of the insulator string material. Pixels in the HSV image are screened based on the first threshold. The hue range and saturation minimum threshold are obtained through color statistical analysis of a large number of known insulator sample images. For common brown or cyan ceramic insulators, the preset hue range is set to, for example, hue values of 15 to 35 for brown (for example, a hue range of 0-360), or hue values of 170 to 190 for cyan. The minimum saturation threshold is empirically set, for example, to be greater than 0.3 (saturation range of 0-1) to exclude low-saturation background objects such as the sky or concrete towers. The brightness is usually set to a wider range, such as 0.2 to 0.9, to accommodate lighting changes and shadows. All pixels whose hue, saturation, and brightness fall within the preset ranges are marked as belonging to the target color. All these marked pixels are aggregated to obtain the target color pixel set. Traverse each pixel in the target color pixel set and read the pixel's position in the original image (horizontally) and (Vertical) coordinate value, before the traversal begins, initialize four variables: minimum coordinate( ) is a value greater than the maximum width of the image (for example, the image width plus 1), the maximum coordinate( ) is -1, the minimum coordinate( ) is a value greater than the maximum height of the image (for example, the image height plus 1), the maximum coordinate( ) is -1, during the traversal process, for each pixel point of the target color pixel set ,Will Coordinates and current and For comparison, if , then update ,like , then update , similarly, Coordinates and current and For comparison, if , then update ,like , then update , when all the pixels in the target color pixel set are traversed and compared, the obtained 、 、 、 The four values constitute the boundary of the minimum bounding rectangle that can completely surround all target color pixels. The system then records the coordinates of the upper left corner of the rectangular area as , and calculate the width and height of the rectangle to get the preliminary candidate frame. The width and height of the rectangle are expressed as: , , in, for The maximum and minimum values of the coordinates, for The maximum and minimum values of the coordinates, is the width, is the height; Extract the image sub-region corresponding to the preliminary candidate frame, apply a gradient operator to this sub-region, such as the Sobel operator, and calculate the gradient of each pixel in the image sub-region by convolving the 3x3 Sobel horizontal and vertical convolution kernels with the image sub-region respectively. Directional gradient component and Directional gradient component , and then the gradient amplitude and gradient direction of the point are expressed as: , , in, is the azimuth function, is the gradient direction, is the gradient amplitude; Next, a non-maximum suppression operation is performed to check whether the gradient amplitude is greater than the gradient amplitude of the two adjacent pixels along the gradient direction of each pixel. If it is not the maximum, the gradient amplitude of the pixel is set to zero, thereby refining the edge. Subsequently, a double threshold method is used for edge connection, and a high threshold and a low threshold are set. For example, the high threshold is set to the value at the 90th percentile of the cumulative distribution histogram after statistics of all non-zero gradient amplitudes in the image after non-maximum suppression, such as 120, and the low threshold is set to a specific ratio of the high threshold, such as 0.4×120, that is, 48. Pixels with gradient amplitudes higher than the high threshold are regarded as strong edge points and are directly retained. Pixels with gradient amplitudes lower than the low threshold are suppressed and not regarded as edge points, while pixels with gradient amplitudes between the low threshold and the high threshold are regarded as Weak edge points are retained as edge points only when they are connected to strong edge points through an 8-neighborhood. All qualified edge pixels are traced and connected to form a set of continuous edge contours. Then, these edge contours are subjected to connected domain analysis to calculate the properties of each connected domain, such as area, perimeter, aspect ratio, and circularity. For example, the lower limit of the area of the valid contour is set to 0.5% of the total pixels of the image (excluding small noise), and the aspect ratio (major axis length / minor axis length) is greater than 2.5 (insulator strings are usually slender structures). This aspect ratio threshold is obtained by counting the geometric features of 100 typical insulator string sample images, taking the 10th percentile value of the distribution, and selecting the connected domain that best meets these preset geometric constraints as the boundary of the insulator string. Finally, the geometric center coordinates are calculated based on the confirmed boundary pixel point set. The coordinates of the target insulator string are established by using the rough center of the insulator string or by determining the main axis direction as the center line through principal component analysis (PCA) of the boundary points.

[0032] It should be noted that by actively adjusting the shooting parameters and optimizing the image processing process, the acquisition quality and positioning accuracy of the insulator string image are significantly improved.

[0033] In the embodiment of the present invention, the above step S200 includes the following sub-steps B1-B3; In B1: Within the calibrated insulator string image region, the horizontal projection of each row of pixel intensity is calculated. The gap between the insulator segments is located by detecting the periodic troughs of the projection curve. The hardware is then located by performing sliding window matching using a preset hardware shape template. The independent image region of each insulator segment and the associated hardware region are segmented to obtain a single-segment image region set. In B2: for each independent image region of the insulator slice in the single image region set, the pixel values of the color region are converted into grayscale representation, a gray-level co-occurrence matrix is constructed, the texture feature parameters are calculated, and the average value and dispersion of the channel pixels of the original region are calculated respectively to obtain the visual parameters of the insulator slice; In B3: Based on the independent image area and visual parameters of each insulator segment in the single image area set, the edge detection operator is used to obtain the outer boundary point set of the insulator segment, the contour line of the shed is determined, the maximum span of the contour line in the horizontal direction is calculated as the shed diameter, and the thickness of the central part along the axial direction of the insulator is measured. All parameters are integrated to construct the insulator segment state vector.

[0034] Specifically, the calibrated insulator string image area is cropped from the complete aerial image. If the insulator string is tilted in the image, the main axis direction is calculated and rotation correction is performed to make the insulator string vertical or horizontal. In the corrected insulator string image area, it is converted into a grayscale image, and then the sum of pixel intensities is calculated row by row to form a one-dimensional horizontal projection curve, where the abscissa of the curve represents the row number and the ordinate represents the total grayscale value of the row pixels. Due to the material and structure of the insulator sheet, the total grayscale value of the gap area between the sheets is usually higher or lower (depending on the background and the insulator color), so it is represented on the projection curve. The periodic peaks and troughs are found, and the gap between the insulator segments is located by finding these troughs. Specifically, the projection curve is smoothed, for example, using Gaussian smoothing with a window size of 3 to 5 pixels to remove noise. Then, a peak-valley detection algorithm is used. A point is identified as a trough if the projection value is less than the projection values of the two adjacent points on the left and right, and the projection value is lower than the preset valley depth threshold. The threshold is set to 60% of the average value of all local peaks on the projection curve. This percentage is determined by statistically analyzing the projection data of 50 groups of different types of insulator strings and selecting the proportion that maximizes the gap detection accuracy. For example, if the average peak If the value of the trough depth is 20,000, the valley depth threshold is 12,000. A continuous trough sequence that meets a specific distance constraint (for example, the spacing is between 30 and 80 pixels, which is estimated based on the typical insulator piece height and shooting resolution) is used to preliminarily divide the insulator pieces. At the same time, the system loads a variety of preset hardware shape templates, such as U-shaped hanging ring templates, ball head hanging ring templates, etc. These templates are small-size (for example, 30x40 pixels) grayscale images obtained by cropping and averaging a large number of standard hardware images. A sliding window is used at the top and bottom of the insulator string image area and at the location where the hardware connection exists according to the insulator string model. In this method, the normalized cross correlation (NCC) is calculated between each fitting template and the image sub-region within the corresponding window. When the NCC value exceeds the preset matching similarity threshold, for example, 0.7, it is considered that a fitting is located. The threshold of 0.7 is obtained by performing a receiver operating characteristic curve (ROC) analysis on a test set containing 100 known fittings and 100 non-fitting areas, and selecting the NCC value corresponding to the maximum Youden index point. Based on the identified gap position and fitting position, the insulator string image is finally segmented into a series of independent insulator segment image regions and the associated fitting image regions to obtain a single-piece image region set; Based on the independent image area of each segmented insulator piece in the single image area set, the pixel values in the color image area are converted to grayscale representation through the grayscale value = 0.299×red channel value + 0.587×green channel value + 0.114×blue channel, and the grayscale image of the single insulator is obtained. A gray level co-occurrence matrix (GLCM) is constructed on the grayscale image to compress the grayscale level from 0-255 to a smaller range, for example, 0-31, a total of 32 grayscale levels, and the compression method is linear mapping, that is, the new grayscale level , select the statistical directions of 0 degrees, 45 degrees, 90 degrees and 135 degrees, fix the statistical distance to 1 pixel, calculate the GLCM in these four directions respectively, add the corresponding elements and normalize them to get an average GLCM, based on this average GLCM, calculate the angular second moment, which is calculated as the sum of the squares of all element values in the matrix. For example, if the GLCM is a 2x2 matrix , then the second-order angular moment is , also calculate the contrast, that is, the weighted sum of the squares of the differences between different grayscale pairs in GLCM, and the entropy value, that is, the probability value of each element in GLCM multiplied by the logarithm and then summed and negative, return to the original color independent image area, respectively for the red (R), green (G), and blue (B) channels, calculate the arithmetic mean and standard deviation of all pixel values in each channel, combine the texture parameters (angular second moment, contrast, entropy) and color statistical parameters (R channel average, R channel standard deviation, G channel average, G channel standard deviation, B channel average, B channel standard deviation) to obtain the visual parameters of the insulator sheet; Based on the independent image areas of each insulator piece in the single image area set and the visual parameters of the insulator piece, an edge detection operator is applied to the independent image area, such as the Canny edge detection algorithm, including using a Gaussian filter (for example, a kernel size of 5x5 and a standard deviation of 1.4) to smooth the image, calculate the gradient amplitude and direction, perform non-maximum suppression, and finally connect the edges through a double threshold method, where the setting of high and low thresholds is crucial for extracting accurate umbrella skirt contours. The high threshold can be set to the 85th percentile of the gradient amplitude histogram of the independent image area, for example, calculated to be 100, and the low threshold is set to 0.4 times the high threshold, that is, 40. This method is used to obtain the set of boundary points of the outermost layer of the insulator piece, and the longest continuous contour is selected from it as the contour line of the umbrella skirt. Next, in order to calculate the umbrella skirt diameter, it is necessary to determine the main axis direction of the insulator piece, which is obtained by performing principal component analysis (PCA) on the contour point set. The contour line is rotated to be vertical or horizontal to the main axis direction, and then rotated perpendicular to the main axis direction. In the direction of the principal axis (i.e., the radial direction of the shed), the maximum span of the contour line point set is calculated. For example, if the principal axis is vertical, the difference between the maximum and minimum abscissas of all points on the contour line is calculated. This difference is the shed diameter. To measure the central thickness of the insulator along the axial direction, first exclude the shed edge area in a direction parallel to the principal axis (for example, exclude 20% of the area on both ends of the contour line in the principal axis direction). Within the remaining central area, multiple distances between contour lines are measured in a direction perpendicular to the principal axis, and the average or minimum value is taken as the central thickness. For example, in the central area of a vertically aligned insulator segment, five equally spaced horizontal lines are selected. The distance between the left and right points of the contour on each horizontal line is calculated. The average of these five distances is the central thickness. Finally, the calculated shed diameter and central thickness are integrated with all parameters such as the second-order angular moment, contrast, entropy, and the mean and standard deviation of each color channel to form a multi-dimensional vector, which is used to construct the insulator segment state vector.

[0035] It should be noted that through precise image segmentation and multi-dimensional feature extraction, the insulator slice state vector is comprehensively constructed to provide rich and accurate characterization information for defect detection.

[0036] In the embodiment of the present invention, the above step S300 includes the following sub-steps C1-C5; In C1: obtaining a database of normal insulator segment sample parameters; In C2: Calculate the mean and standard deviation based on the angular second-order matrix and contrast texture features, define the threshold range, set the fixed upper and lower limits of the color standard deviation and shed morphology parameters as the allowable deviation, and establish the defect judgment threshold set; In C3: Based on the defect judgment threshold set and the state vector of each insulator segment, the current value of each parameter in the vector is read in sequence. The angular second-order moment and contrast value are compared with the threshold range, the color standard deviation is compared with the upper threshold, and the shed diameter and thickness values are compared with the allowable deviation range. If any parameter value falls outside the corresponding normal range, the parameter is marked as abnormal and a parameter deviation indication is obtained. In C4: judging the marking result of each insulator segment according to the parameter deviation indication of each insulator segment; In C5: If at least one abnormal parameter mark is included, the insulator segment is determined to be a suspected defective individual, and the identifiers of all suspected defective insulator segments and the corresponding insulator segment state vectors are collected and compiled into a list to generate a suspected defective insulator list.

[0037] Specifically, the values corresponding to all normal samples are extracted from the database of normal insulator sample parameters, the arithmetic mean and standard deviation are calculated respectively, and the threshold interval is defined and set as the mean plus or minus times the standard deviation, e.g. , determined by optimizing the balance between defect detection rate and false alarm rate on the validation set, ensuring coverage of about 98.8% of normal sample fluctuations. Taking the second-order moment of angle as an example, if the mean of normal samples is 0.075 and the standard deviation is 0.01, the threshold interval is set to ,Right now For contrast, if the mean of normal samples is 30 and the standard deviation is 8, the dynamic threshold interval is ,Right now As for the color standard deviation (e.g., the standard deviation of each R, G, and B channel), since it often indicates contamination or uneven material, only an upper limit is usually set. This upper limit is the 98th percentile of the corresponding color standard deviation value in the normal sample database. For example, if the 98th percentile value of the R channel standard deviation of the normal sample is 18 (grayscale 0-255), then the upper limit of the R channel standard deviation is 18. For shed morphological parameters such as shed diameter and center thickness, fixed upper and lower limits are set as the allowable deviation range based on the nominal dimensions and allowable tolerances in the design specifications corresponding to the insulator model, or combined with the distribution of these parameters in the normal sample database (e.g., taking the 1% and 99% percentiles). For example, if the nominal shed diameter of a certain model of insulator is 250 mm and the allowable tolerance is plus or minus 5 mm, the allowable deviation range can be set to [245 mm, 255 mm]. The normal range or threshold value of each key parameter in the state vector is obtained, and a defect judgment threshold set is established. According to the defect judgment threshold set and the insulator state vector of each insulator segment to be detected, the system reads the current measurement value of each parameter in the state vector in sequence and compares it with the normal range of the corresponding parameter in the defect judgment threshold set. The normal range is the allowable value interval preset for each parameter in the defect judgment threshold set. For example, for the angular second-order moment parameter, if the angular second-order moment measurement value of the current insulator segment is 0.045, and the angular second-order moment normal range obtained from the defect judgment threshold set is [0.050, 0.100], since 0.045 is less than 0.050, the parameter value falls outside the normal range, and the angular second-order moment parameter of this insulator segment is marked as abnormal. Similarly, if the contrast measurement value is 55, and the normal range of contrast is [10, 50], since 55 is greater than 50, the contrast parameter is also marked as abnormal. For the color standard deviation, R is used as the reference. For example, if the current value is 22 and the preset upper limit of the R channel standard deviation is 18, since 22 is greater than 18, this parameter is also marked as abnormal. For the shed diameter, if the measured value is 240 mm and the allowable deviation range is [245 mm, 255 mm], since 240 mm is less than 245 mm, the shed diameter is marked as abnormal. Similar comparison operations are performed on the center thickness and all other parameters included in the state vector. The current value of each parameter is checked against the normal interval or unilateral threshold defined in the defect judgment threshold set. As long as the current value of any parameter exceeds the corresponding normal range boundary, the specific parameter is assigned an abnormal flag. After completing the item-by-item comparison of all parameters in the insulator segment state vector, the system integrates this flag information and generates a parameter deviation indication for the current insulator segment, including the normal or non-normal status of each parameter. The marking results of each insulator piece involved in the inspection image are checked. During the inspection process, if at least one parameter in the parameter deviation indication of an insulator piece is marked as abnormal, then the insulator piece is preliminarily judged as an individual suspected of having defects. For example, for insulator piece A, if the angular second-order moment is within the normal range, but the contrast is marked as abnormal, the color standard deviation is normal, the shed diameter is normal, and the center part thickness is normal, due to the abnormal contrast parameter, insulator piece A is classified as a suspected defective individual. For example, for insulator piece B, if the shed diameter exceeds the lower limit of the allowable deviation range, and the R channel color standard deviation also exceeds If the upper limit is exceeded, even if all other parameters are within the normal range, insulator segment B is also judged as a suspected defective individual due to the presence of two abnormal parameter marks, the shed diameter and the R channel color standard deviation. For insulator segment C, if the parameter deviation indication shows that all parameters are within their respective normal threshold ranges and there are no abnormal marks, the insulator segment is not judged as a suspected defective individual. The unique identifiers of all insulator segments judged as suspected defective individuals and the corresponding complete insulator segment state vectors are collected, and the information is recorded one by one and compiled into a structured list. This list is the generated list of suspected defective insulators.

[0038] It should be noted that by scientifically setting thresholds and strictly comparing parameters, suspected defective insulators can be accurately screened, effectively improving detection accuracy and reliability.

[0039] In the embodiment of the present invention, the above step S400 includes the following sub-steps D1-D3; In D1: A gradient filter is applied to the image block of the suspected defective insulator piece, the pixel intensity change rate in the coordinate axis direction is calculated, and the gradient amplitude image is synthesized to obtain a high-contrast surface image; In D2: Based on the high-contrast surface image and the corresponding insulator segment state vector, chain code tracking or path search is used to connect the gradient pixel points in the image, identify linear structures whose length and curvature meet the crack conditions, and perform threshold segmentation on the original color image area to extract abnormal dark patches. The geometric parameters and grayscale distribution of the patches are calculated to obtain the defect appearance feature set; In D3: Based on the defect appearance feature set, the linear structure features and dark patch parameters are similar to the quantitative features of various types of defects in the pre-stored defect sample library. The estimated defect category is assigned, and the degree of damage is evaluated based on the numerical range of the feature parameters to obtain the defect type and location.

[0040] Specifically, the image block of the insulator piece is extracted from the original image pixel data corresponding to each suspected defective insulator piece in the suspected defective insulator list. If the original image is in color, it is converted into a grayscale image. A gradient filter is applied to the grayscale image block. The Sobel operator can be used. The 3x3 Sobel horizontal convolution kernel is used, for example , and vertical convolution kernels, such as Perform two-dimensional convolution operations with the image blocks respectively to calculate the pixel intensity change rate of each pixel in the image block in the x direction, which is recorded as , and the pixel intensity change rate in the y direction, recorded as , the gradient amplitude is synthesized for each pixel, and the gradient amplitude is expressed as: , in That is, the gradient amplitude of the pixel point, and a new image is obtained. The pixel values of the image are composed of the corresponding gradient amplitude, which significantly highlights the areas where the pixel intensity changes sharply in the original image, such as object edges, texture details, and potential surface irregularities such as cracks or scratches, to obtain a high-contrast surface image; For binarization of high-contrast surface images, the Otsu adaptive threshold algorithm can be used to automatically select the optimal segmentation threshold by maximizing the inter-class variance. For example, if the calculated Otsu threshold is 70 (grayscale range 0-255), pixels with gradient amplitudes greater than 70 are marked as foreground, and the rest are background. The chain code tracking algorithm is used for the foreground pixels in the binarized image. Starting from a foreground point, the next foreground point is searched along 8 neighborhoods and encoded with a number representing the direction, for example, 0 represents east, 1 represents northeast, and so on to 7 represents southeast. The path is recorded until it returns to the starting point or there are no foreground points to connect to, forming a closed or open chain, and then the linear structure is identified. For each linear structure, the length (the length of the chain code sequence) and the average curvature (for example, obtained by calculating the cumulative value of the chain code direction change divided by the length) are calculated. The length must exceed a minimum length threshold, for example, set to 15 pixels. This value is based on experience, that is, linear structures with a length of less than 15 pixels are usually noise or insignificant. The 15-pixel threshold is based on an analysis of 200 known micro-crack sample images. The minimum length that can eliminate more than 90% of non-crack interference is selected. At the same time, its average curvature must be less than a maximum curvature threshold, such as 0.2 radians / pixel, to ensure the straightness or slow curvature of the linear structure. This curvature threshold is set by analyzing the curvature distribution of 50 known real cracks to the maximum curvature value covering 95% of the real crack samples. Linear structures that meet these two conditions are preliminarily judged as crack candidates. At the same time, the system returns to the original color image area of the insulator sheet and converts it to the HSV color space. The V (brightness) channel is thresholded to extract abnormal dark patches. The segmentation threshold of the dark patches is set to the 15th percentile value of the insulator sheet V channel pixel value histogram. For example, if this value is calculated to be 60, pixels with a V channel value less than 60 are considered dark area candidates. The connected domain analysis is performed on the resulting dark areas to calculate the area, perimeter, and circularity of each connected dark patch ( ) and the average gray value, constitute the defect appearance feature set; The similarity calculation is performed on the features extracted in real time from the defect appearance feature set and the quantitative features of each type of defect in the pre-stored defect sample library. The pre-stored defect sample library is established by extracting features and performing statistical analysis on a large number of images (for example, more than 1,000) of defective insulators (such as self-explosion, flashover, ice coating, contamination, damage, etc.) that have been confirmed by experts. Each defect type in the library corresponds to one or more feature vector templates or statistical distribution models of feature parameters (for example, mean and covariance matrix); the similarity calculation can use a trained classifier model, such as a support vector machine (SVM) or a deep learning network (such as a convolutional neural network CNN). The model takes the defect appearance feature set as input and outputs the defect category. For example, the training process of an SVM classifier is as follows: at least 80% of the data in the pre-stored defect sample library is used as the training set, the defect appearance feature set of each sample is extracted to form a feature vector, and the real defect category label is assigned. The radial basis function (RBF) is selected as the kernel function. The optimal penalty coefficient C (e.g., C=10) and kernel parameter gamma (e.g., gamma=0.01) are determined through grid search and ten-fold cross-validation. A classification model is then trained. The model input is the apparent defect feature set of the insulator under test, and the output is the probability of each defect category or a direct category determination. When the probability of a category exceeds a preset confidence threshold (e.g., 0.75, which is adjusted on the remaining 20% of the validation set to achieve the optimal balance between precision and recall), it is assigned the estimated defect category. After the defect category is determined, the damage severity is assessed based on the numerical range of key characteristic parameters related to the defect type. For example, if a crack is identified, the crack length is classified as a minor crack (e.g., less than 20 pixels), a general crack (20-50 pixels), or a severe crack (greater than 50 pixels). The estimated defect category and assessed damage severity are combined with the defect's pixel coordinate location in the image to determine the confirmed defect type and location.

[0041] It should be noted that by enhancing image features and intelligent classification algorithms, the type and location of insulator defects can be accurately identified, significantly improving detection accuracy and efficiency.

[0042] In the embodiment of the present invention, the above step S500 includes the following sub-steps E1-E3; In E1: Based on the defect type and location, a partial image of the defective insulator segment is cropped from the original high-resolution image. Simultaneously, the installation serial number, 3D spatial coordinates, defect category identifier determined by the system, and corresponding text description of the defect are retrieved to form a single defect holographic file. In E2: Based on the single defect holographic file, a universal data exchange format is selected, and the image slices, digitized serial numbers and coordinates, standardized defect codes, and encoded description text in the file are sequentially filled into the preset key-value pair template to form a record and obtain a standardized defect entry; In E3: Create a list or array structure based on the standardized defect items, add each standardized defect item as an element, and sort the list based on the tower identification of the inspection task or the severity level of the defect to generate structured transmission line insulator defect report data.

[0043] Specifically, based on the positioning information in the insulator segment image, a local image slice of the insulator segment containing the defect is cropped from the stored original high-resolution aerial image. During cropping, a certain margin is added to the defect range, for example, 30 pixels in each direction, to ensure the integrity of the defect context. Simultaneously, the installation serial number of the defective insulator segment in the insulator string is retrieved from the image processing stage or metadata associated with the inspection task. The 3D spatial coordinates are extracted from the Exchangeable Image File Format (EXIF) recorded when the drone captured the image or from the synchronized flight log. Based on the determined defect type, the corresponding defect category identifier is searched from a preset defect code-description mapping table. For example, "self-destruction" corresponds to the code "ZBA01." Each code in the table is associated with a standard text description of the defect condition, for example, "ZBA01" corresponds to "complete rupture of the insulator disc, loss of insulation function." The image slice, installation serial number, GPS 3D spatial coordinates, defect category identifier, defect description, original image file name, and shooting timestamp are integrated to form an independent single defect holographic file for each confirmed defect. According to the single defect holographic archive of each defective insulator piece, a data exchange format is selected, such as JSON format (text data exchange format), and the data in the single defect holographic archive is organized according to a pre-defined key-value pair template, and the field name (key) and data type of each information element in the output data are clarified. For example, a preset JSON format template structure is as follows: {"defect_id": "unique defect number", "tower_id": "tower identification", "string_id": "serial number", "piece_index": installation serial number, "gps_longitude": longitude value, "gps_latitude": latitude value, "gps_altitude": altitude value, "capture_timestamp": "shooting time", "defect_code": "defect category identifier", "defect_description": "defect situation text description", "severity_ level": "severity level code", "image_slice_base64": "Image slice Base64 encoding", "original_image_ref": "Original image file name"}, convert the image slice in the single defect holographic file into a Base64-encoded string and fill it into the image_slice_base64 field, fill the digitized installation serial number and coordinates into the corresponding numeric fields respectively, fill the standardized defect category identifier into the defect_code field, and fill the UTF-8 encoded defect description into the defect_description field. Other information such as tower identification, serial number, shooting time, severity level code (for example, "01" for critical, "02" for serious, and "03" for general) is also filled in. Each time the information of a defective insulator slice is filled in, a structured JSON record is formed, thereby obtaining a standardized defect entry set containing multiple independent defect records; Create a list or array structure in memory and add each standardized defect entry as an independent element to this list or array. After all standardized defect entries are added, sort the list according to the preset or user-selected sorting rules. There can be multiple sorting rules. For example, you can prioritize ascending order based on the tower identification of the inspection task. Under the same tower identification, you can then sort in descending order based on the severity level code of the defect (for example, "01" is critical at the front and "03" is general at the end). Finally, under the same severity level, you can sort in ascending order based on the insulator serial number and the piece serial number. This sorting method (For example, the primary key is the tower identification, the secondary key is the severity level, and the tertiary key is the serial number and piece number) This makes it easy for operation and maintenance personnel to plan maintenance work by area and urgency. The severity level codes are sorted based on the degree of threat they pose to the safe operation of the line. For example, "01-Critical" requires immediate processing, "02-Serious" requires processing as soon as possible, and "03-General" can be included in regular maintenance. Alternatively, the severity level code of the defect can be directly used as the primary sorting key for global descending sorting, placing all critical defects at the front, regardless of which tower they belong to. After the sorting is completed, the final generated structured transmission line insulator defect report data is obtained.

[0044] It should be noted that through the generation of structured data, efficient organization and output of defect information is achieved, which facilitates operation and maintenance personnel to quickly locate and handle defects and improve the efficiency of power grid operation and maintenance.

[0045] The above is a schematic diagram of the embedded transmission line insulator string defect identification method of this embodiment. It should be noted that the technical solution of this embedded transmission line insulator string defect identification system and the technical solution of the embedded transmission line insulator string defect identification method described above are based on the same concept. For details not described in detail in the technical solution of the embedded transmission line insulator string defect identification system of this embodiment, please refer to the description of the technical solution of the embedded transmission line insulator string defect identification method described above.

[0046] The embedded transmission line insulator string defect identification system in this embodiment includes: A positioning module is used to obtain an insulator string image, locate the insulator string area on the insulator string image using color space change and image processing, and obtain the coordinates of the target insulator string; A feature extraction module is used to extract features of the insulator segment based on the target insulator string coordinates and construct an insulator segment state vector; a potential defect detection module, configured to determine the state vector of the insulator segment and obtain a list of suspected defective insulators; a defect type discrimination module, configured to calculate the insulator pieces in the list of suspected defective insulators and detect the surfaces of the insulator pieces to obtain the defect types and locations on the surfaces of the insulator pieces; The defect information output module is used to associate the defect category code and description according to the defect type and location, and generate structured transmission line insulator defect report data.

[0047] This embodiment also provides a computer device suitable for the embedded transmission line insulator string defect identification method, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the embedded transmission line insulator string defect identification method proposed in the above embodiment.

[0048] This embodiment further provides a storage medium having a computer program stored thereon. When the program is executed by a processor, the embedded transmission line insulator string defect identification method proposed in the above embodiment is implemented.

[0049] The storage medium proposed in this embodiment and the method for implementing embedded transmission line insulator string defect identification proposed in the above embodiment belong to the same inventive concept. Technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0050] From the above description of the embodiments, those skilled in the art will clearly understand that the present invention can be implemented using software and necessary general-purpose hardware. Of course, it can also be implemented using hardware, but in many cases the former is the preferred embodiment. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This software product can be stored on a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk, or optical disk, and includes instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods of various embodiments of the present invention.

[0051] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An embedded transmission line insulator string defect identification method, characterized in that: include: Acquire an insulator string image, and locate the insulator string region on the insulator string image using color space variation and image processing to obtain the coordinates of the target insulator string; Based on the target insulator string coordinates, feature extraction is performed on the insulator segment to construct an insulator segment state vector; Judging the state vector of the insulator segment to obtain a list of suspected defective insulators; Calculating the insulator pieces in the suspected defective insulator list and inspecting the surfaces of the insulator pieces to obtain the defect types and locations on the surfaces of the insulator pieces; According to the defect type and location, the defect category code and description are associated to generate structured transmission line insulator defect report data.

2. The embedded transmission line insulator string defect identification method according to claim 1, characterized in that: Locating the insulator string area includes: Read the channel pixel values of the insulator string image, calculate the color element corresponding to each pixel, and obtain the image in the color space; Setting a first threshold to filter the pixel points of the image in the color space to obtain a target color pixel set; Traverse the coordinate axes of the target color pixel set, determine the maximum and minimum coordinate information, define a rectangular area surrounding all target color pixels, record the vertex coordinates, width and height of the rectangular area, and obtain a preliminary candidate frame; The gradient operator is applied to the image area within the preliminary candidate frame to calculate the gradient of each point, and non-maximum suppression and double threshold processing are performed. The edges are connected to form a continuous edge contour, which is confirmed as the boundary of the insulator string and the coordinates of the target insulator string are established.

3. The embedded transmission line insulator string defect identification method according to claim 2, characterized in that: Constructing the insulator segment state vector includes: Within the calibrated insulator string image area, the horizontal projection of each row of pixel intensity is calculated. The gap between the insulator segments is located by detecting the periodic troughs of the projection curve. The hardware is then located by performing sliding window matching using a preset hardware shape template. The independent image area of each insulator segment and the associated hardware area are segmented to obtain a single-segment image area set. For each independent image region of the insulator slice in the single image region set, the pixel values of the color region are converted into grayscale representation, the gray-level co-occurrence matrix is constructed, the texture feature parameters are calculated, and the average value and dispersion degree of the channel pixels of the original region are calculated respectively to obtain the visual parameters of the insulator slice; Based on the independent image area and visual parameters of each insulator segment in the single image area set, the edge detection operator is used to obtain the outer boundary point set of the insulator segment, determine the contour line of the shed, calculate the maximum span of the contour line in the horizontal direction as the shed diameter, and measure the thickness of the central part along the axial direction of the insulator. All parameters are integrated to construct the insulator segment state vector.

4. The embedded transmission line insulator string defect identification method according to claim 3, characterized in that: Determining the insulator segment state vector includes: Obtaining a database of normal insulator segment sample parameters; The mean and standard deviation are calculated based on the angular second-order matrix and contrast texture features, the threshold range is defined, the upper and lower limits of the color standard deviation and the shed morphological parameters are set as the allowable deviation, and the defect judgment threshold set is established; Based on the defect judgment threshold set and the state vector of each insulator segment, the current value of each parameter in the vector is read in sequence. The angular second-order moment and contrast value are compared with the threshold interval, the color standard deviation is compared with the upper threshold, and the shed diameter and thickness values are compared with the allowable deviation range. If any parameter value falls outside the corresponding normal range, the parameter is marked as abnormal and a parameter deviation indication is obtained. Determine the marking result of each insulator segment based on the parameter deviation indication of each insulator segment; If at least one abnormal parameter mark is included, the insulator segment is determined to be a suspected defective individual. The identifiers of all suspected defective insulator segments and the corresponding insulator segment state vectors are collected and compiled into a list to generate a suspected defective insulator list.

5. The embedded transmission line insulator string defect identification method according to claim 4, characterized in that: The types and locations of defects on the surface of insulator sheets include: Applying a gradient filter to the image blocks of suspected defective insulator pieces, calculating the pixel intensity change rates in the coordinate axis directions, synthesizing the gradient amplitude images, and obtaining a high-contrast surface image; Based on the high-contrast surface image and the corresponding insulator sheet state vector, chain code tracking or path search is used to connect the gradient pixel points in the image, identify linear structures whose length and curvature meet the crack conditions, and perform threshold segmentation on the original color image area to extract abnormal dark patches. The geometric parameters and grayscale distribution of the patches are calculated to obtain the defect appearance feature set.

6. The embedded transmission line insulator string defect identification method according to claim 5, characterized in that: Also includes: Based on the defect appearance feature set, the linear structure features, dark patch parameters and the quantitative features of various types of defects in the pre-stored defect sample library are similarly calculated, and the estimated defect category is assigned. The degree of damage is evaluated based on the numerical range of the feature parameters to obtain the defect type and location.

7. The embedded transmission line insulator string defect identification method according to claim 6, characterized in that: The data for generating structural transmission line insulator defect report includes: Based on the defect type and location, a local image of the defective insulator segment is cropped from the original high-resolution image. At the same time, the installation serial number, three-dimensional spatial coordinates, defect category identifier determined by the system, and corresponding text description of the defect are retrieved to form a single defect holographic file. According to the single defect holographic file, a universal data exchange format is selected, and the image slices, digitized serial numbers and coordinates, standardized defect codes, and coded description text in the file are sequentially filled into the preset key-value pair template to form a record and obtain a standardized defect entry; Based on the standardized defect items, a list or array structure is created, each standardized defect item is added as an element, and the list is sorted according to the tower identification of the inspection task or the severity level of the defect to generate structured transmission line insulator defect report data.

8. An embedded transmission line insulator string defect identification system, applying the embedded transmission line insulator string defect identification method according to any one of claims 1 to 7, characterized in that: A positioning module is used to obtain an insulator string image, locate the insulator string area on the insulator string image using color space change and image processing, and obtain the coordinates of the target insulator string; A feature extraction module is used to extract features of the insulator segment based on the target insulator string coordinates and construct an insulator segment state vector; a potential defect detection module, configured to determine the state vector of the insulator segment and obtain a list of suspected defective insulators; a defect type discrimination module, configured to calculate the insulator pieces in the list of suspected defective insulators and detect the surfaces of the insulator pieces to obtain the defect types and locations on the surfaces of the insulator pieces; The defect information output module is used to associate the defect category code and description according to the defect type and location, and generate structured transmission line insulator defect report data.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the embedded transmission line insulator string defect identification method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the embedded transmission line insulator string defect identification method according to any one of claims 1 to 7 are implemented.

Images