An intelligent evaluation method and system for ultraviolet discharge of insulators based on digital image processing

Abstract

The application discloses an insulator ultraviolet discharge intelligent evaluation method and system based on digital image processing, and comprises the following steps: acquiring an ultraviolet imaging video sequence of an insulator to be measured, performing color threshold segmentation frame by frame to extract an ultraviolet light spot area, and generating a frame-by-frame light spot binary image through morphological dilation processing; selecting a reference frame, performing adaptive threshold segmentation and morphological close-open operation, and implementing directional morphological corrosion along the main shaft direction of the insulator to acquire an insulator body segmentation mask; accumulating the frame-by-frame binary image to construct a frequency accumulation matrix, obtaining a logic mask through threshold denoising and morphological filtering, multiplying the frequency matrix pseudo-color mapping and the logic mask to fuse and construct a discharge persistence atlas; identifying an independent discharge source area based on the atlas, extracting three-dimensional features of each area and performing normalization grading; determining a discharge preset state grade according to the combination logic of the graded features, and outputting an insulator operation state evaluation result. The objective quantitative evaluation of the insulator operation state is realized.

Description

Technical Field

[0001] This invention relates to the technical field of corona discharge monitoring, and in particular to an intelligent assessment method and system for ultraviolet discharge of insulators based on digital image processing. Background Technology

[0002] Corona discharge is a common partial discharge phenomenon in high-voltage power equipment, manifesting as transient or continuous discharge in a non-uniform electric field, accompanied by characteristic ultraviolet radiation and audible noise. This phenomenon may indicate insulation defects, the causes of which are diverse, including defects in the metal tip structure and the insulation material itself. Corona discharge leads to power loss, insulation material degradation, and interference with communication systems, and in severe cases, may cause insulation flashover accidents.

[0003] Currently, ultraviolet (UV) imaging technology has become an important means of monitoring corona discharge, enabling non-contact diagnosis by detecting the UV radiation generated by the discharge. UV imagers superimpose the detected UV radiation onto a visible light image as a spot, with the spot size positively correlated with the radiation intensity. Due to the intermittent nature of corona discharge, actual detection requires video recording to capture its temporal behavior. However, existing UV imaging diagnostic technologies have significant limitations: detection results are highly dependent on operator experience and subjective judgment, lacking quantitative evaluation standards; the analysis efficiency of massive video data is low, hindering automated processing; existing methods struggle to accurately locate the spatial relationship between the discharge source and the insulator, leading to high false alarm rates; and for complex conditions such as multi-source discharges and intermittent discharges, effective feature extraction and state classification methods are lacking. Summary of the Invention

[0004] In view of the aforementioned existing problems, this invention is proposed. Therefore, this invention provides an intelligent assessment method and system for ultraviolet discharge of insulators based on digital image processing, solving the problems of subjective diagnosis, inaccurate positioning, coarse classification standards, low automation, and difficulty in fusion of multiple frames in existing ultraviolet imaging diagnostic technologies.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, embodiments of the present invention provide an intelligent assessment method for ultraviolet discharge of insulators based on digital image processing, comprising: A video sequence of ultraviolet imaging of the insulator to be tested is obtained. Color thresholding is performed on each frame of the video sequence to extract the ultraviolet spot region. Morphological dilation processing is performed on the segmentation result to obtain a frame-by-frame binary image of the spot. The frame with the fewest spot pixels in the ultraviolet imaging video sequence is selected as the reference frame. Adaptive threshold segmentation and morphological closing and opening operations are performed on the reference frame, and directional morphological erosion is performed in combination with the main axis direction of the insulator to obtain the insulator body segmentation mask. The frame-by-frame binary image of the light spot is matrix-superimposed to generate a frequency accumulation matrix. The frequency accumulation matrix is ​​then subjected to threshold denoising and morphological filtering to obtain a logical mask. The frequency accumulation matrix is ​​then fused with the logical mask after pseudo-color mapping to construct a discharge persistence spectrum. Based on the discharge persistence map, each independent discharge source region is identified, and three-dimensional features are extracted for each independent discharge source region. The three-dimensional features are then normalized and graded. Based on the combined logic of the normalized and graded three-dimensional features, the discharge phenomenon is divided into multiple preset state levels, and the insulator operation status evaluation results are output.

[0006] As a preferred embodiment of the intelligent assessment method for ultraviolet discharge of insulators based on digital image processing described in this invention, the step of performing color threshold segmentation on each frame of the video sequence to extract the ultraviolet spot region, and performing morphological dilation processing on the segmentation result to obtain a frame-by-frame binary image of the spot, includes: For each frame of the image, a threshold range is set for the three RGB channels. Pixels that meet the magenta spot characteristics are marked as spot areas and assigned a value of 1, while pixels that do not meet the conditions are marked as background and assigned a value of 0. A cross-shaped structural element is used to perform morphological dilation on the marked spot area to compensate for the missing edge pixels. The cross-shaped structural element includes a central 3×3 pixel area and pixel areas extending in four directions: up, down, left, and right.

[0007] As a preferred embodiment of the intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing described in this invention, the method includes: performing adaptive threshold segmentation and morphological opening / closing operations on the reference frame, and combining this with directional morphological erosion along the principal axis of the insulator to obtain an insulator body segmentation mask, comprising: Extract the blue channel image of the reference frame and perform adaptive threshold segmentation using the Otsu algorithm; The segmentation results are sequentially subjected to morphological closing and opening operations using circular structuring elements, and small noise regions are removed by connected component analysis to obtain the preliminary insulator region. The principal axis tilt angle of the preliminary region of the insulator is calculated, and a morphological erosion operation is performed along the principal axis tilt angle using linear structural elements to remove interfering structures perpendicular to the insulator body, thereby obtaining the segmentation mask of the insulator body.

[0008] As a preferred embodiment of the intelligent assessment method for ultraviolet discharge of insulators based on digital image processing described in this invention, the method includes: obtaining a logic mask by threshold denoising and morphological filtering of the frequency accumulation matrix; fusing the frequency accumulation matrix with the logic mask after pseudo-color mapping to construct a discharge persistence map. Binarization and morphological opening and closing operations are performed on the denoised matrix to obtain a logic mask that identifies the independent power supply region. The frequency accumulation matrix is ​​pseudo-color encoded to generate a frequency color map, and the frequency color map is multiplied with the logic mask to obtain the discharge persistence spectrum.

[0009] As a preferred embodiment of the intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing described in this invention, the method includes: identifying each independent discharge source region based on the discharge persistence spectrum, and extracting three-dimensional features for each independent discharge source region, including: Extract the maximum durability, equivalent area, and minimum spatial distance of each independent discharge power region; Wherein, the maximum persistence is the ratio of the maximum grayscale value of the pixel in the region to the total number of frames in the video sequence; the equivalent area is the ratio of the total number of pixels in the region to the pixel area of ​​the insulator skirt; the minimum spatial distance is the minimum Euclidean distance between the centroid of the discharge region and the segmentation mask of the insulator body; Before the normalization and grading process, a conversion relationship between pixels and actual physical dimensions is established based on the known physical dimensions of the insulator skirt, and the equivalent area and the minimum spatial distance are converted into actual physical quantities.

[0010] As a preferred embodiment of the intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing described in this invention, the normalization and hierarchical processing of the three-dimensional features includes mapping the three-dimensional features to the discrete interval [0,2]. When the maximum durability is less than the first durability threshold, the maximum durability level is 0; when the maximum durability is not less than the first durability threshold and less than the second durability threshold, the maximum durability level is 1; when the maximum durability is not less than the second durability threshold, the maximum durability level is 2. When the equivalent area is less than the third equivalent area threshold, the equivalent area level is 0; when the equivalent area is not less than the third equivalent area threshold and is less than the fourth equivalent area threshold, the equivalent area level is 1; when the equivalent area is not less than the fourth equivalent area threshold, the equivalent area level is 2. When the minimum spatial distance is greater than the fifth spatial distance threshold, the minimum spatial distance level is 0; when the minimum spatial distance is not greater than the fifth spatial distance threshold but greater than the sixth spatial distance threshold, the minimum spatial distance level is 1; when the minimum spatial distance is not greater than the sixth spatial distance threshold, the minimum spatial distance level is 2.

[0011] As a preferred embodiment of the intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing described in this invention, the discharge phenomenon is divided into multiple preset state levels based on the combination logic of the normalized and graded three-dimensional features, and the insulator operation status evaluation results are output, including: the preset state levels include no discharge, external discharge, mild body discharge, moderate body discharge and severe body discharge. When the maximum durability level is 0 and the equivalent area level is 0, it is determined that there is no discharge; When the minimum spatial distance level is 0, it is determined to be an external discharge; When the minimum spatial distance level is 1 and the sum of the maximum durability level and the equivalent area level is less than 3, it is determined to be a mild body discharge; when the minimum spatial distance level is 1 and the sum of the maximum durability level and the equivalent area level is greater than or equal to 3, or when the minimum spatial distance level is 2 and the sum of the maximum durability level and the equivalent area level is less than 2, it is determined to be a moderate body discharge. When the minimum spatial distance level is 2 and the sum of the maximum durability level and the equivalent area level is greater than or equal to 2, it is determined to be a severe body discharge. After performing the aforementioned determination independently on each independent discharge source region, a structured evaluation report is generated and output, which includes the discharge source number, spatial coordinates, three-dimensional feature parameters, and corresponding state level.

[0012] Secondly, the present invention provides an intelligent assessment system for ultraviolet discharge of insulators based on digital image processing, comprising: The spot segmentation module is used to acquire the ultraviolet imaging video sequence of the insulator under test, perform color threshold segmentation on each frame of the video sequence to extract the ultraviolet spot region, and perform morphological dilation processing on the segmentation result to obtain a frame-by-frame binary image of the spot. The body localization module is used to select the frame with the fewest number of light spot pixels from the ultraviolet imaging video sequence as a reference frame, perform adaptive threshold segmentation and morphological closing and opening operations on the reference frame, and perform directional morphological erosion in combination with the insulator main axis direction to obtain an insulator body segmentation mask. The spectrum construction module is used to perform matrix superposition on the frame-by-frame binary image of light spots to generate a frequency accumulation matrix, perform threshold denoising and morphological filtering on the frequency accumulation matrix to obtain a logic mask, and then perform pseudo-color mapping on the frequency accumulation matrix and fuse it with the logic mask to construct a discharge persistence spectrum. The feature extraction module is used to identify each independent discharge source region based on the discharge persistence map, extract three-dimensional features for each independent discharge source region, and perform normalization and hierarchical processing on the three-dimensional features respectively. The status assessment module is used to classify the discharge phenomenon into multiple preset status levels based on the combination logic of the normalized and graded three-dimensional features, and output the insulator operation status assessment results.

[0013] Thirdly, the present invention provides an electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing.

[0014] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention replaces manual visual inspection with digital image processing algorithms, transforming subjective descriptions such as spot area, discharge duration, and spatial distance into precise values, eliminating diagnostic biases caused by differences in operator experience, and ensuring repeatable and traceable evaluation results; it effectively integrates temporal information through persistent image processing technology, solving the problem of missed intermittent discharges, and can simultaneously identify and distinguish multiple independent discharge sources in the video, achieving pixel-level positioning accuracy and significantly better spatial resolution than single-frame analysis; further, it introduces the distance parameter between the discharge source and the insulator body, effectively distinguishing between discharges caused by defects in the insulator body and discharges caused by external interference, reducing false alarm rates; the five-level classification standard comprehensively considers intensity, duration, and spatial location, achieving refined classification of insulation status and providing a basis for differentiated operation and maintenance strategies; the algorithm is based on classic image processing technology, with low computational complexity and short video processing time, and can be deployed on field detection terminals or online monitoring systems to achieve real-time assessment and early warning of insulator status; it is applicable to ultraviolet detection data of polymer insulators with voltage levels of 500kV and below, and can be adapted to different camera models and shooting conditions by adjusting the size of structural elements and threshold parameters, making it highly practical for engineering applications. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Fig. 1 This is a schematic diagram of the process flow of an intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing, according to an embodiment of the present invention. Fig. 2 This is a schematic diagram of the structural elements used in the expansion calculation of an intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing, according to an embodiment of the present invention. Fig. 3 This is a flowchart of an insulator segmentation algorithm for an intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing, as described in one embodiment of the present invention. Detailed Implementation

[0017] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0018] Example 1, referring to Figs. 1-3 As one embodiment of the present invention, this embodiment provides an intelligent assessment method for ultraviolet discharge of insulators based on digital image processing, such as... Fig. 1 As shown, it includes: S100: Acquire the ultraviolet imaging video sequence of the insulator to be tested, perform color threshold segmentation on each frame of the video sequence to extract the ultraviolet spot region, and perform morphological dilation processing on the segmentation result to obtain frame-by-frame binary image of the spot. S200: Select the frame with the fewest spot pixels from the ultraviolet imaging video sequence as the reference frame, perform adaptive threshold segmentation and morphological closing and opening operations on the reference frame, and perform directional morphological erosion in combination with the main axis direction of the insulator to obtain the insulator body segmentation mask. S300: The frequency accumulation matrix is ​​generated by matrix superposition of the frame-by-frame binary image of the light spot. Threshold denoising and morphological filtering are performed on the frequency accumulation matrix to obtain the logic mask. The frequency accumulation matrix is ​​then fused with the logic mask after pseudo-color mapping to construct the discharge persistence spectrum. S400: Identify each independent discharge source region based on the discharge persistence spectrum, extract three-dimensional features for each independent discharge source region, and perform normalization and hierarchical processing on the three-dimensional features respectively. S500: Based on the combinational logic of the normalized and graded three-dimensional features, the discharge phenomenon is divided into multiple preset state levels, and the insulator operation status evaluation results are output.

[0019] It should be noted that traditional ultraviolet imaging diagnosis relies on manual observation of spot size, brightness, and flicker frequency. Different operators may have significantly different judgments on the same discharge phenomenon, lacking objective and reproducible quantitative evaluation indicators. A single frame image can only reflect the instantaneous discharge state and cannot characterize the cumulative effect and persistence of the discharge over time. Manually reviewing long-term videos (usually over 30 seconds, containing 720 frames) is inefficient and easily misses key information about intermittent discharges. Existing technologies cannot automatically distinguish between discharges on the insulator itself and discharges on surrounding hardware and conductors, and cannot accurately calculate the spatial distance between the discharge source and the polymer surface of the insulator, thus limiting the accuracy of insulation status assessment. Traditional methods only use simple threshold classification based on spot area or photon count, without comprehensively considering multi-dimensional characteristics such as discharge intensity, persistence, and spatial location, and cannot achieve refined hierarchical assessment of insulator operating status. There is a lack of fully automated algorithms from video preprocessing and feature extraction to status classification, making it difficult to meet the application requirements of smart grids for real-time monitoring and batch diagnosis of equipment status. This invention enables an objective quantitative assessment of the insulator's operating status and improves the ability to identify multi-source discharges by constructing a discharge persistence spectrum and a multi-parameter fusion classification model.

[0020] In this embodiment of the invention, step S100 involves performing color thresholding on each frame of the video sequence to extract the ultraviolet spot region, and then performing morphological dilation on the segmentation result to obtain a frame-by-frame binary image of the spot, including: For each frame of the image, a threshold range is set for the three RGB channels. Pixels that meet the magenta spot characteristics are marked as spot areas and assigned a value of 1, while pixels that do not meet the conditions are marked as background and assigned a value of 0. A cross-shaped structural element is used to perform morphological dilation on the marked spot area to compensate for the missing edge pixels. The cross-shaped structural element includes a central 3×3 pixel area and pixel areas extending in four directions: up, down, left, and right.

[0021] It should be noted that in this embodiment of the invention, the threshold range for the RGB three channels is set to R>200, G<100, B>200, which is used to segment the magenta light spot.

[0022] Reference Fig. 2 In this embodiment of the invention, morphological dilation operation is used to enhance the edge of the segmentation result, and cross-shaped structural elements are used to process the edge of the light spot to compensate for the edge pixel loss that may be caused by threshold segmentation, thus ensuring the integrity of the discharge area.

[0023] In this embodiment of the invention, step S200 involves performing adaptive threshold segmentation and morphological closing / opening operations on the reference frame, and combining this with directional morphological erosion along the principal axis of the insulator to obtain an insulator body segmentation mask, including: The blue channel image of the reference frame is extracted, and adaptive threshold segmentation is performed using the Otsu algorithm; The segmentation results are sequentially subjected to morphological closing and opening operations using circular structuring elements, and small noise regions are removed by connected component analysis to obtain the preliminary insulator region. The principal axis tilt angle of the initial region of the insulator is calculated, and a morphological erosion operation is performed along the principal axis tilt angle using linear structural elements to remove interfering structures perpendicular to the insulator body, thus obtaining a segmentation mask for the insulator body.

[0024] Reference Fig. 3 In one feasible approach, an ultraviolet imaging video sequence of the insulator under test is acquired at a frame rate of 24 fps and a duration of approximately 30 seconds. The frame with the fewest spot pixels is selected from the video sequence as the reference frame, typically the frame with no discharge or the weakest discharge. Otsu adaptive thresholding is applied to the blue channel of the reference frame to initially separate the insulator region. Morphological closing and opening operations are performed sequentially, using 25×25 circular structural elements to eliminate small-area noise and connect fractured regions. The closing operation is defined as dilation followed by erosion. The opening operation is defined as erosion followed by expansion. The expansion operation fills small holes inside the insulator region and connects broken parts, while the erosion operation suppresses over-extension and removes small protrusions. The combined use of closing and opening operations effectively eliminates noise spots in binary images, fills tiny holes, and smooths insulator boundaries, while maintaining the area and shape of the insulator body essentially unchanged. Eight-connected region labeling identifies individual regions, and non-target regions with fewer than a threshold of pixels are deleted, preserving the polymer insulator body.

[0025] The tilt angle of the insulator spindle is calculated. A 1×200 linear structure element is used to perform a morphological etching operation at angle θ to remove interfering structures such as fittings and supports that are perpendicular to the insulator, thereby achieving precise segmentation of the insulator body.

[0026] In one possible implementation, the insulator region contour point set can be extracted using the minimum bounding rectangle method, and the rectangle with the smallest area that can completely contain the region can be found using the rotating caliper algorithm. The angle between the long side of this rectangle and the horizontal axis is θ.

[0027] In another possible implementation, the geometric moments of the insulator region can be calculated using the image moment method. The direction angle θ = 0.5∙arctan2(2μ 11 , μ 20 -μ 02 ), where μ11, μ20, and μ02 are the central moments.

[0028] Furthermore, a 1×200 linear structuring element, oriented at θ+90° (perpendicular to the insulator's main axis), is used to perform morphological erosion. The center pixel is retained as foreground only if all pixels covered by this linear structuring element are foreground elements. This operation effectively removes interfering structures such as fittings and supports orthogonal to the insulator's main axis, while the polymer body of the insulator, with a width greater than 200 pixels perpendicular to the main axis, is preserved.

[0029] Furthermore, all frames after ultraviolet video spot segmentation and enhancement are cumulatively superimposed to generate a frequency accumulation matrix. The pixel value in the matrix represents the frequency at which that location is detected as a spot during the entire observation period; a higher value indicates stronger discharge persistence.

[0030] In this embodiment of the invention, step S300 involves performing threshold denoising and morphological filtering on the frequency accumulation matrix to obtain a logic mask, and then fusing the frequency accumulation matrix with the logic mask after pseudo-color mapping to construct a discharge persistence map. Noise points with pixel values ​​lower than 2% of the total number of frames in the frequency accumulation matrix are removed. The denoised matrix is ​​then binarized and subjected to morphological opening and closing operations to obtain a logic mask that identifies the independent power supply region. The frequency accumulation matrix is ​​pseudo-color encoded to generate a frequency color map. The frequency color map is then multiplied with the logic mask to obtain the discharge persistence spectrum.

[0031] Specifically, the frequency accumulation matrix is ​​thresholded to remove noise points with frequencies lower than 2% of the total frame count. If the video has 720 frames, the threshold is 15. Then, binarization is performed, setting pixels greater than 0 to 1. Next, morphological opening and closing operations are performed sequentially, using the same cross-shaped structuring element as in step S100. The opening operation breaks the weak connections between different discharge sources and removes isolated noise, while the closing operation fills the small holes inside each discharge source region, making each region a complete connected component. The processed binary image is the logical mask, where each eight-connected region corresponds to an independent discharge source.

[0032] It should be noted that pseudo-color mapping is applied to the frequency accumulation matrix to convert the discharge frequency into color information, such as low-frequency blue and high-frequency red. This color information is then multiplied with a logic mask to generate a persistent spectrum that integrates spatial distribution and duration. This spectrum condenses the discharge information of the entire video into a single image, visually displaying the location, intensity, and duration of the discharge.

[0033] In this embodiment of the invention, step S400 involves identifying each independent discharge source region based on the discharge persistence spectrum and extracting three-dimensional features for each independent discharge source region, including: Extract the maximum durability, equivalent area, and minimum spatial distance of each independent discharge power region; Among them, the maximum persistence is the ratio of the maximum gray value of the pixel in the region to the total number of frames in the video sequence; the equivalent area is the ratio of the total number of pixels in the region to the pixel area of ​​the insulator skirt; and the minimum spatial distance is the minimum Euclidean distance between the centroid of the discharge area and the segmentation mask of the insulator body. Before normalization and hierarchical processing, a conversion relationship between pixels and actual physical dimensions is established based on the known physical dimensions of the insulator skirts, converting the equivalent area and minimum spatial distance into actual physical quantities.

[0034] Furthermore, the maximum durability characterizes the persistence of the strongest discharge point; the equivalent area is established by using the known insulator skirt diameter (130mm) as the calibration benchmark to establish a pixel-to-millimeter conversion relationship; the minimum Euclidean distance between the centroid of the spatial distance calculation area and the insulator segmentation mask is converted into the actual distance after calibration to distinguish between the discharge of the main body and the discharge of the surrounding hardware.

[0035] In this embodiment of the invention, step S400, which involves normalizing and classifying the three-dimensional features, includes mapping the three-dimensional features to the discrete interval [0,2]. When the maximum durability is less than the first durability threshold, the maximum durability level is 0; when the maximum durability is not less than the first durability threshold and less than the second durability threshold, the maximum durability level is 1; when the maximum durability is not less than the second durability threshold, the maximum durability level is 2. When the equivalent area is less than the third equivalent area threshold, the equivalent area level is 0; when the equivalent area is not less than the third equivalent area threshold and is less than the fourth equivalent area threshold, the equivalent area level is 1; when the equivalent area is not less than the fourth equivalent area threshold, the equivalent area level is 2. When the minimum spatial distance is greater than the fifth spatial distance threshold, the minimum spatial distance level is 0; when the minimum spatial distance is not greater than the fifth spatial distance threshold but greater than the sixth spatial distance threshold, the minimum spatial distance level is 1; when the minimum spatial distance is not greater than the sixth spatial distance threshold, the minimum spatial distance level is 2.

[0036] In this embodiment of the invention, in step S500, the combined logic based on the normalized and graded three-dimensional features divides the discharge phenomenon into multiple preset state levels and outputs the insulator operation status evaluation results, including: the preset state levels include no discharge, external discharge, mild body discharge, moderate body discharge and severe body discharge. When the maximum durability level is 0 and the equivalent area level is 0, it is determined that there is no discharge; when the minimum spatial distance level is 0, it is determined that there is external discharge. When the minimum spatial distance level is 1 and the sum of the maximum endurance level and the equivalent area level is less than 3, it is judged as mild body discharge; when the minimum spatial distance level is 1 and the sum of the maximum endurance level and the equivalent area level is greater than or equal to 3, or when the minimum spatial distance level is 2 and the sum of the maximum endurance level and the equivalent area level is less than 2, it is judged as moderate body discharge. When the minimum spatial distance level is 2 and the sum of the maximum durability level and the equivalent area level is greater than or equal to 2, it is judged as severe body discharge. After independently determining the status of each discharge source region, a structured evaluation report is generated and output, which includes the discharge source number, spatial coordinates, three-dimensional feature parameters, and corresponding status level, as shown in Table 1: Table 1. Features extracted from insulators 1 3.5 0.07 26 1 0 0 2 2 8.8 0.31 22 1 1 0 2 3 8.9 0.21 56 1 1 0 2

[0037] It should be noted that this invention replaces manual visual inspection with digital image processing algorithms, transforming subjective descriptions such as spot area, discharge duration, and spatial distance into precise values. This eliminates diagnostic biases caused by differences in operator experience, and the evaluation results are repeatable and traceable. By effectively integrating temporal information through persistent image processing technology, it solves the problem of missed intermittent discharges, enabling simultaneous identification and differentiation of multiple independent discharge sources in the video. Positioning accuracy reaches the pixel level, and spatial resolution is significantly better than single-frame analysis. Furthermore, the introduction of the distance parameter between the discharge source and the insulator body effectively distinguishes between discharges caused by defects in the insulator body and external interference discharges, reducing false alarm rates. The five-level classification standard comprehensively considers intensity, duration, and spatial location, achieving refined grading of insulation status and providing a basis for differentiated operation and maintenance strategies. The algorithm is based on classic image processing technology, has low computational complexity, and short video processing time. It can be deployed on field detection terminals or online monitoring systems to achieve real-time assessment and early warning of insulator status. It is applicable to ultraviolet detection data of polymer insulators with voltage levels of 500kV and below. By adjusting the structural element size and threshold parameters, it can be adapted to different camera models and shooting conditions, making it highly practical for engineering applications.

[0038] Example 2: The above example is an illustrative scheme of an intelligent assessment method for ultraviolet discharge of insulators based on digital image processing. It should be noted that the technical solution of this intelligent assessment system for ultraviolet discharge of insulators based on digital image processing belongs to the same concept as the above-described intelligent assessment method for ultraviolet discharge of insulators based on digital image processing. Details not described in detail in this example can be found in the description of the above-described intelligent assessment method for ultraviolet discharge of insulators based on digital image processing.

[0039] This embodiment presents an intelligent assessment system for ultraviolet discharge of insulators based on digital image processing, comprising: The spot segmentation module is used to acquire the ultraviolet imaging video sequence of the insulator under test, perform color threshold segmentation on each frame of the video sequence to extract the ultraviolet spot region, and perform morphological dilation processing on the segmentation result to obtain frame-by-frame binary image of the spot. The body localization module is used to select the frame with the fewest spot pixels from the ultraviolet imaging video sequence as the reference frame, perform adaptive threshold segmentation and morphological closing and opening operations on the reference frame, and perform directional morphological erosion in combination with the main axis direction of the insulator to obtain the insulator body segmentation mask. The spectrum construction module is used to generate a frequency accumulation matrix by matrix superposition of frame-by-frame binary images of light spots, to obtain a logic mask by threshold denoising and morphological filtering of the frequency accumulation matrix, and to fuse the frequency accumulation matrix with the logic mask after pseudo-color mapping to construct a discharge persistence spectrum. The feature extraction module is used to identify each independent discharge source region based on the discharge persistence map, extract three-dimensional features for each independent discharge source region, and perform normalization and hierarchical processing on the three-dimensional features respectively. The status assessment module is used to classify discharge phenomena into multiple preset status levels based on the combinational logic of normalized and graded three-dimensional features, and output the insulator operation status assessment results.

[0040] This embodiment also provides an electronic device applicable to the intelligent assessment method for ultraviolet discharge of insulators based on digital image processing, including: The system includes a memory and a processor. The memory stores computer-executable instructions, and the processor executes these instructions to implement the intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing, as proposed in the above embodiments.

[0041] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing as proposed in the above embodiments.

[0042] The storage medium proposed in this embodiment and the intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0043] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0044] 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 it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A smart evaluation method for ultraviolet discharge of insulators based on digital image processing, characterized in that, include: A video sequence of ultraviolet imaging of the insulator to be tested is obtained. Color thresholding is performed on each frame of the video sequence to extract the ultraviolet spot region. Morphological dilation processing is performed on the segmentation result to obtain a frame-by-frame binary image of the spot. The frame with the fewest spot pixels in the ultraviolet imaging video sequence is selected as the reference frame. Adaptive threshold segmentation and morphological closing and opening operations are performed on the reference frame, and directional morphological erosion is performed in combination with the main axis direction of the insulator to obtain the insulator body segmentation mask. The frame-by-frame binary image of the light spot is matrix-superimposed to generate a frequency accumulation matrix. The frequency accumulation matrix is ​​then subjected to threshold denoising and morphological filtering to obtain a logical mask. The frequency accumulation matrix is ​​then fused with the logical mask after pseudo-color mapping to construct a discharge persistence spectrum. Based on the discharge persistence map, each independent discharge source region is identified, and three-dimensional features are extracted for each independent discharge source region. The three-dimensional features are then normalized and graded. Based on the combined logic of the normalized and graded three-dimensional features, the discharge phenomenon is divided into multiple preset state levels, and the insulator operation status evaluation results are output.

2. The intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing as described in claim 1, characterized in that, The step of performing color thresholding segmentation on each frame of the video sequence to extract the ultraviolet spot region, and then performing morphological dilation processing on the segmentation result to obtain a frame-by-frame binary image of the spot, includes: For each frame of the image, a threshold range is set for the three RGB channels. Pixels that meet the magenta spot characteristics are marked as spot areas and assigned a value of 1, while pixels that do not meet the conditions are marked as background and assigned a value of 0. A cross-shaped structural element is used to perform morphological dilation on the marked spot area to compensate for the missing edge pixels. The cross-shaped structural element includes a central 3×3 pixel area and pixel areas extending in four directions: up, down, left, and right.

3. The intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing as described in claim 2, characterized in that, Adaptive threshold segmentation and morphological closing / opening operations are performed on the reference frame, and directional morphological erosion is combined with the insulator principal axis direction to obtain an insulator body segmentation mask, including: Extract the blue channel image of the reference frame and perform adaptive threshold segmentation using the Otsu algorithm; The segmentation results are sequentially subjected to morphological closing and opening operations using circular structuring elements, and small noise regions are removed by connected component analysis to obtain the preliminary insulator region. The principal axis tilt angle of the preliminary region of the insulator is calculated, and a morphological erosion operation is performed along the principal axis tilt angle using linear structural elements to remove interfering structures perpendicular to the insulator body, thereby obtaining the segmentation mask of the insulator body.

4. The intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing as described in claim 3, characterized in that, A logic mask is obtained by performing threshold denoising and morphological filtering on the frequency accumulation matrix. The frequency accumulation matrix is ​​then fused with the logic mask after pseudo-color mapping to construct a discharge persistence map, including: Binarization and morphological opening and closing operations are performed on the denoised matrix to obtain a logic mask that identifies the independent power supply region. The frequency accumulation matrix is ​​pseudo-color encoded to generate a frequency color map, and the frequency color map is multiplied with the logic mask to obtain the discharge persistence spectrum.

5. The intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing as described in claim 4, characterized in that, Based on the discharge persistence map, each independent discharge source region is identified, and three-dimensional features are extracted for each independent discharge source region, including: Extract the maximum durability, equivalent area, and minimum spatial distance of each independent discharge power region; Wherein, the maximum persistence is the ratio of the maximum grayscale value of the pixel in the region to the total number of frames in the video sequence; the equivalent area is the ratio of the total number of pixels in the region to the pixel area of ​​the insulator skirt; the minimum spatial distance is the minimum Euclidean distance between the centroid of the discharge region and the segmentation mask of the insulator body; Before the normalization and grading process, a conversion relationship between pixels and actual physical dimensions is established based on the known physical dimensions of the insulator skirt, and the equivalent area and the minimum spatial distance are converted into actual physical quantities.

6. The intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing as described in claim 5, characterized in that, The normalization and hierarchical processing of the three-dimensional features includes mapping the three-dimensional features to the discrete interval [0,2]. When the maximum durability is less than the first durability threshold, the maximum durability level is 0; when the maximum durability is not less than the first durability threshold and less than the second durability threshold, the maximum durability level is 1; when the maximum durability is not less than the second durability threshold, the maximum durability level is 2. When the equivalent area is less than the third equivalent area threshold, the equivalent area level is 0; when the equivalent area is not less than the third equivalent area threshold and is less than the fourth equivalent area threshold, the equivalent area level is 1; when the equivalent area is not less than the fourth equivalent area threshold, the equivalent area level is 2. When the minimum spatial distance is greater than the fifth spatial distance threshold, the minimum spatial distance level is 0; when the minimum spatial distance is not greater than the fifth spatial distance threshold but greater than the sixth spatial distance threshold, the minimum spatial distance level is 1; when the minimum spatial distance is not greater than the sixth spatial distance threshold, the minimum spatial distance level is 2.

7. The intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing as described in claim 6, characterized in that, Based on the combination logic of the normalized and graded three-dimensional features, the discharge phenomenon is divided into multiple preset state levels, and the insulator operation status evaluation results are output, including: preset state levels include no discharge, external discharge, mild body discharge, moderate body discharge and severe body discharge. When the maximum durability level is 0 and the equivalent area level is 0, it is determined that there is no discharge; When the minimum spatial distance level is 0, it is determined to be an external discharge; When the minimum spatial distance level is 1 and the sum of the maximum durability level and the equivalent area level is less than 3, it is determined to be a mild body discharge; when the minimum spatial distance level is 1 and the sum of the maximum durability level and the equivalent area level is greater than or equal to 3, or when the minimum spatial distance level is 2 and the sum of the maximum durability level and the equivalent area level is less than 2, it is determined to be a moderate body discharge. When the minimum spatial distance level is 2 and the sum of the maximum durability level and the equivalent area level is greater than or equal to 2, it is determined to be a severe body discharge. After performing the aforementioned determination independently on each independent discharge source region, a structured evaluation report is generated and output, which includes the discharge source number, spatial coordinates, three-dimensional feature parameters, and corresponding state level.

8. A smart insulator ultraviolet discharge assessment system based on digital image processing, applied to the method described in any one of claims 1-7, characterized in that, include: The spot segmentation module is used to acquire the ultraviolet imaging video sequence of the insulator under test, perform color threshold segmentation on each frame of the video sequence to extract the ultraviolet spot region, and perform morphological dilation processing on the segmentation result to obtain a frame-by-frame binary image of the spot. The body localization module is used to select the frame with the fewest number of light spot pixels from the ultraviolet imaging video sequence as a reference frame, perform adaptive threshold segmentation and morphological closing and opening operations on the reference frame, and perform directional morphological erosion in combination with the insulator main axis direction to obtain an insulator body segmentation mask. The spectrum construction module is used to perform matrix superposition on the frame-by-frame binary image of light spots to generate a frequency accumulation matrix, perform threshold denoising and morphological filtering on the frequency accumulation matrix to obtain a logic mask, and then perform pseudo-color mapping on the frequency accumulation matrix and fuse it with the logic mask to construct a discharge persistence spectrum. The feature extraction module is used to identify each independent discharge source region based on the discharge persistence map, extract three-dimensional features for each independent discharge source region, and perform normalization and hierarchical processing on the three-dimensional features respectively. The status assessment module is used to classify the discharge phenomenon into multiple preset status levels based on the combination logic of the normalized and graded three-dimensional features, and output the insulator operation status assessment results.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the intelligent evaluation method for ultraviolet discharge of insulators based on digital image processing as described in any one of claims 1 to 7.

Images