High-voltage transmission line defect identification method and system based on multi-model cooperation

Abstract

The invention discloses a high-voltage transmission line defect identification method and system based on multi-model cooperation. The method comprises the following steps: acquiring a plurality of images corresponding to a plurality of line positions of a target transmission line; determining an identification model corresponding to each line position; determining defect identification information corresponding to each image according to the corresponding identification model; and according to the defect identification information corresponding to all the images, generating a defect judgment result corresponding to the target power transmission line based on an associated continuous defect judgment rule. Therefore, accurate power transmission line defect evaluation based on multi-model matching and continuous defect analysis can be realized, the line inspection efficiency and the defect early warning accuracy are improved, and the line fault risk caused by defect missing detection or misjudgment is reduced.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for identifying defects in high-voltage transmission lines based on multi-model collaboration. Background Technology

[0002] With the rapid growth in demand for smart grid and transmission line inspections, power companies are increasingly focusing on improving line operation safety and inspection efficiency through accurate defect assessment. A key technical issue is how to achieve more accurate line risk monitoring to reduce fault risks. Existing technologies typically collect image data from single locations on transmission lines, use fixed recognition models or simple threshold analysis methods to detect defects, and generate inspection reports based on standard rules to support line maintenance. However, existing solutions lack dynamic model matching for multiple line locations and the ability to analyze and correlate continuous defects. This makes it difficult to accurately assess defect distribution and severity, and they cannot adapt to complex line environments. Consequently, defect warning accuracy is insufficient, and missed or misjudged defects can easily lead to line faults, limiting inspection efficiency and grid operation reliability. Therefore, existing technologies have shortcomings that urgently need to be addressed. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for identifying defects in high-voltage transmission lines based on multi-model collaboration, which can realize accurate defect assessment of transmission lines based on multi-model matching and continuous defect analysis, improve the efficiency of line inspection and the accuracy of defect early warning, and reduce the risk of line failure caused by missed or misjudged defects.

[0004] To address the aforementioned technical problems, the first aspect of this invention discloses a method for identifying defects in high-voltage transmission lines based on multi-model collaboration, the method comprising: Acquire multiple images corresponding to multiple locations of the target transmission line; Determine the identification model corresponding to each of the aforementioned line locations; Based on the corresponding recognition model, the defect recognition information corresponding to each image is determined; Based on the defect identification information corresponding to all the images, and based on the associated continuous defect judgment rule, the defect judgment result corresponding to the target transmission line is generated.

[0005] As an optional implementation, in the first aspect of the present invention, the line location is the location of a line tower, a conductor, an insulator, or a hardware; the image is captured based on multiple different shooting parameters; the shooting parameters are lens position, lens focal length, shooting angle, shooting frame, or shooting light sensitivity parameters.

[0006] As an optional implementation, in the first aspect of the invention, the image is obtained by taking pictures of the location of the line using a drone or monitoring equipment.

[0007] As an optional implementation, in the first aspect of the invention, determining the identification model corresponding to each of the line positions includes: For each of the aforementioned route locations, all the images corresponding to that route location are obtained to obtain a location image set; Determine the matching degree between the set of location images and each candidate recognition model; The candidate recognition model with the highest matching degree is determined as the recognition model corresponding to the line location; the recognition model is trained by a training dataset including multiple training transmission line images and corresponding defect annotations.

[0008] As an optional implementation, in the first aspect of the invention, determining the matching degree between the set of location images and each candidate recognition model includes: For each candidate recognition model, obtain the training image data set corresponding to that candidate recognition model; Calculate the first similarity between the set of location images and the set of training image data; The shooting parameters corresponding to each image in the location image set are determined to be the first image parameter set; The shooting parameters corresponding to each image in the training image dataset are determined as the second image parameter set; Calculate the second similarity between the first image parameter set and the second image parameter set; The product of the first similarity and the second similarity is calculated to obtain the matching degree between the location image set and the candidate recognition model.

[0009] As an optional implementation, in the first aspect of the present invention, determining the defect identification information corresponding to each image based on the corresponding identification model includes: For each image, determine the recognition model corresponding to the line position corresponding to that image; Identify at least two associated images corresponding to this image; After stitching the image together with all the associated images, the image is input into the corresponding recognition model to obtain the defect recognition information corresponding to the image; the defect recognition information includes the defect type and the defect location.

[0010] As an optional implementation, in the first aspect of the invention, determining the at least two associated images corresponding to the image includes: Calculate the shooting location distance and image similarity between the image and any other image mentioned above; Calculate the product of the shooting location distance and the image similarity to obtain the image matching degree between the image and any other image. Filter out at least two other images whose matching degree is greater than a preset matching degree threshold to obtain at least two associated images corresponding to the image.

[0011] As an optional implementation, in the first aspect of the present invention, generating a defect determination result corresponding to the target transmission line based on defect identification information corresponding to all the images and on a rule for determining associated continuous defects includes: By filtering out multiple images of the same defect type, a set of multiple image sets of the same type is obtained; Calculate the average value of the defect locations of all images in each set of images of the same type to obtain the set location corresponding to each set of images of the same type; For any two sets of images of the same type whose positional distance between them is less than a preset distance threshold, the defect type and positional distance corresponding to the two sets of images of the same type are input into the trained associated continuous defect recognition model to obtain the associated continuous defect risk corresponding to the two sets of images of the same type. Determine whether the risk of the associated continuous defects is greater than a preset risk threshold. If so, identify the two sets of images of the same type as a high-risk set pair. All the high-risk sets are displayed in the preset line risk map corresponding to the target transmission line to generate the defect judgment result corresponding to the target transmission line.

[0012] A second aspect of this invention discloses a high-voltage transmission line defect identification system based on multi-model collaboration, the system comprising: The acquisition module is used to acquire multiple images corresponding to multiple locations of the target transmission line; A determination module is used to determine the identification model corresponding to each of the aforementioned line positions; The recognition module is used to determine the defect recognition information corresponding to each image based on the corresponding recognition model. The generation module is used to generate the defect judgment result corresponding to the target transmission line based on the defect identification information corresponding to all the images and the associated continuous defect judgment rule.

[0013] As an optional implementation, in the second aspect of the invention, the line location is the location of a line tower, a conductor, an insulator, or a hardware; the image is captured based on multiple different shooting parameters; the shooting parameters are lens position, lens focal length, shooting angle, shooting frame, or shooting light sensitivity parameters.

[0014] As an optional implementation, in a second aspect of the invention, the image is obtained by taking pictures of the location of the line using a drone or monitoring equipment.

[0015] As an optional implementation, in a second aspect of the invention, the specific method by which the determining module determines the identification model corresponding to each of the line positions includes: For each of the aforementioned route locations, all the images corresponding to that route location are obtained to obtain a location image set; Determine the matching degree between the set of location images and each candidate recognition model; The candidate recognition model with the highest matching degree is determined as the recognition model corresponding to the line location; the recognition model is trained by a training dataset including multiple training transmission line images and corresponding defect annotations.

[0016] As an optional implementation, in a second aspect of the invention, the determining module determines the matching degree between the location image set and each candidate recognition model in the following specific manner: For each candidate recognition model, obtain the training image data set corresponding to that candidate recognition model; Calculate the first similarity between the set of location images and the set of training image data; The shooting parameters corresponding to each image in the location image set are determined to be the first image parameter set; The shooting parameters corresponding to each image in the training image dataset are determined as the second image parameter set; Calculate the second similarity between the first image parameter set and the second image parameter set; The product of the first similarity and the second similarity is calculated to obtain the matching degree between the location image set and the candidate recognition model.

[0017] As an optional implementation, in a second aspect of the invention, the method by which the recognition module determines the specific method of defect recognition information corresponding to each image based on the corresponding recognition model includes: For each image, determine the recognition model corresponding to the line position corresponding to that image; Identify at least two associated images corresponding to this image; After stitching the image together with all the associated images, the image is input into the corresponding recognition model to obtain the defect recognition information corresponding to the image; the defect recognition information includes the defect type and the defect location.

[0018] As an optional implementation, in a second aspect of the invention, the specific method by which the recognition module determines the at least two associated images corresponding to the image includes: Calculate the shooting location distance and image similarity between the image and any other image mentioned above; Calculate the product of the shooting location distance and the image similarity to obtain the image matching degree between the image and any other image. Filter out at least two other images whose matching degree is greater than a preset matching degree threshold to obtain at least two associated images corresponding to the image.

[0019] As an optional implementation, in the second aspect of the present invention, the specific method by which the generation module generates the defect determination result corresponding to the target transmission line based on the defect identification information corresponding to all the images and on the basis of the associated continuous defect judgment rule includes: By filtering out multiple images of the same defect type, a set of multiple image sets of the same type is obtained; Calculate the average value of the defect locations of all images in each set of images of the same type to obtain the set location corresponding to each set of images of the same type; For any two sets of images of the same type whose positional distance between them is less than a preset distance threshold, the defect type and positional distance corresponding to the two sets of images of the same type are input into the trained associated continuous defect recognition model to obtain the associated continuous defect risk corresponding to the two sets of images of the same type. Determine whether the risk of the associated continuous defects is greater than a preset risk threshold. If so, identify the two sets of images of the same type as a high-risk set pair. All the high-risk sets are displayed in the preset line risk map corresponding to the target transmission line to generate the defect judgment result corresponding to the target transmission line.

[0020] A third aspect of this invention discloses another high-voltage transmission line defect identification system based on multi-model collaboration, the system comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute some or all of the steps in the high-voltage transmission line defect identification method based on multi-model collaboration disclosed in the first aspect of the present invention.

[0021] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the multi-model collaborative high-voltage transmission line defect identification method disclosed in the first aspect of the present invention.

[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention acquires images of multiple locations of a target transmission line and determines the corresponding identification model. Based on the model, it identifies defect information and generates defect judgment results using associated continuous defect judgment rules. This enables accurate transmission line defect assessment based on multi-model matching and continuous defect analysis, improving line inspection efficiency and defect early warning accuracy, and reducing the risk of line faults caused by missed or misjudged defects. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0024] Figure 1 This is a flowchart illustrating a high-voltage transmission line defect identification method based on multi-model collaboration disclosed in an embodiment of the present invention.

[0025] Figure 2 This is a schematic diagram of the structure of a high-voltage transmission line defect identification system based on multi-model collaboration disclosed in an embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of another high-voltage transmission line defect identification system based on multi-model collaboration disclosed in an embodiment of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0029] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0030] This invention discloses a method and system for identifying defects in high-voltage transmission lines based on multi-model collaboration. By acquiring images of multiple locations along the target transmission line and determining corresponding identification models, the method identifies defect information based on these models and generates defect judgment results using rules for judging related continuous defects. This enables accurate transmission line defect assessment based on multi-model matching and continuous defect analysis, improving line inspection efficiency and defect early warning accuracy, and reducing the risk of line faults caused by missed or misjudged defects. Detailed explanations follow.

[0031] Example 1 Please see Figure 1 , Figure 1 This is a flowchart illustrating a high-voltage transmission line defect identification method based on multi-model collaboration disclosed in an embodiment of the present invention. Figure 1 The described multi-model collaborative high-voltage transmission line defect identification method can be applied to data processing systems / data processing equipment / data processing servers (wherein, the server includes local processing servers or cloud processing servers). Figure 1 As shown, the high-voltage transmission line defect identification method based on multi-model collaboration may include the following operations: 101. Obtain multiple images corresponding to multiple line locations of the target transmission line.

[0032] Optionally, the image can be an aerial image taken by a drone, an infrared thermal image, a visible light image, or a high-resolution image; the present invention does not limit the type of image.

[0033] Optionally, the location of the line can be a pole / tower location, a conductor location, an insulator location, or a hardware location; this invention does not impose any limitations.

[0034] Optionally, this acquisition process can be achieved through drone inspection, fixed camera shooting, satellite remote sensing, or manual collection; this invention does not limit the scope of the acquisition.

[0035] 102. Determine the recognition model corresponding to each line position.

[0036] Optionally, the recognition model can be a deep learning model, an object detection model, or a segmentation model; this invention does not impose any limitations.

[0037] Optionally, this determination process can be implemented based on matching degree calculation, model selection, or dynamic allocation algorithm, and the present invention does not limit it.

[0038] Optionally, the determination of the recognition model can be optimized by combining image features, shooting parameters, or model performance, and this invention does not limit it.

[0039] 103. Based on the corresponding recognition model, determine the defect recognition information corresponding to each image.

[0040] Optionally, the defect identification information may include defect type, defect location, defect size, or defect severity, which is not limited in this invention.

[0041] Optionally, this determination process can be implemented based on model reasoning, result parsing, or post-processing algorithms, and the present invention does not limit it.

[0042] Optionally, the determination of the defect identification information can be optimized by combining confidence thresholds or multi-model fusion, and this invention does not limit it.

[0043] 104. Based on the defect identification information corresponding to all images, and based on the rules for judging related continuous defects, generate the defect judgment result corresponding to the target transmission line.

[0044] Optionally, the rule for determining the associated continuous defects can be a distance rule, a type consistency rule, or a risk assessment rule; this invention does not impose any limitations.

[0045] Optionally, the defect determination result can be a defect distribution map, a risk heat map, or a continuous defect segment identifier; this invention does not impose any limitations on this.

[0046] Optionally, this generation process can be implemented by combining risk models, spatial analysis, or visualization techniques, and this invention does not limit it.

[0047] As can be seen, the above-described embodiments of the invention acquire images of multiple locations of the target transmission line and determine the corresponding identification model, identify defect information based on the model, and generate defect judgment results using the associated continuous defect judgment rules. This enables accurate transmission line defect assessment based on multi-model matching and continuous defect analysis, improves line inspection efficiency and defect early warning accuracy, and reduces the risk of line faults caused by missed or misjudged defects.

[0048] As an optional embodiment, in the above steps, the line location can be the location of a line tower, conductor, insulator, or hardware; the image is captured based on multiple different shooting parameters; the shooting parameters can be the lens position, lens focal length, shooting angle, shooting frame, or shooting light sensitivity parameters.

[0049] As can be seen, the above optional embodiments define the details of the line location and image capture parameters to comprehensively characterize the image-related features of the transmission line, assist in realizing accurate transmission line defect assessment based on multi-model matching and continuous defect analysis, improve the efficiency of line inspection and the accuracy of defect early warning, and reduce the risk of line failure caused by missed or misjudged defects.

[0050] As an optional embodiment, in the above steps, the images are obtained by taking pictures of the line location using a drone or monitoring equipment.

[0051] As can be seen, the above optional embodiments limit the acquisition of relevant images by taking pictures of the line location using drones or monitoring equipment, so as to more conveniently and efficiently take pictures of high-voltage transmission lines, assist in realizing accurate transmission line defect assessment based on multi-model matching and continuous defect analysis, improve the efficiency of line inspection and the accuracy of defect early warning, and reduce the risk of line failure caused by missed defects or misjudgments.

[0052] As an optional embodiment, the step of determining the identification model corresponding to each line position in the above steps includes: For each route location, obtain all images corresponding to that route location to get a location image set; Determine the matching degree between the set of location images and each candidate recognition model; The candidate recognition model with the highest matching degree is determined as the recognition model corresponding to the location of the line; the recognition model is trained by a training dataset that includes multiple training transmission line images and corresponding defect annotations.

[0053] Optionally, the location image set can be a multi-view image set, a multi-temporal image set, or a multi-modal image set; the present invention does not impose any limitations on this.

[0054] Optionally, the acquisition of this set of location images can be achieved based on location index, image metadata, or database query, and this invention does not limit it.

[0055] Optionally, the construction of this location image set can be optimized by combining temporal continuity or shooting integrity, and the present invention does not limit it.

[0056] Optionally, the candidate recognition model can be a pre-trained model, a domain-specific model, or an incrementally trained model; this invention does not impose any limitations.

[0057] Optionally, the matching degree can be a similarity score, a matching index, or a compatibility value; this invention does not impose any limitations on this.

[0058] Optionally, the training dataset may include labeled images, augmented images, or simulated images, and this invention does not impose any limitations.

[0059] Optionally, the selection of the recognition model can be optimized in combination with computing resources or real-time requirements, and this invention does not limit it.

[0060] As can be seen, through the above optional embodiments, by calculating the matching degree between the location image set and the candidate recognition model for each line location and selecting the model with the highest matching degree, the targeting and accuracy of the recognition model selection are improved through the product analysis of image and shooting parameter similarity on the basis of accurate defect assessment, providing reliable model support for defect recognition and reducing the risk of recognition error caused by model mismatch.

[0061] As an optional embodiment, the step of determining the matching degree between the location image set and each candidate recognition model in the above steps includes: For each candidate recognition model, obtain the training image data set corresponding to that candidate recognition model; Calculate the first similarity between the set of location images and the training image data set; The shooting parameters corresponding to each image in the location image set are defined as the first image parameter set; The shooting parameters corresponding to each image in the training image dataset are determined as the second image parameter set; Calculate the second similarity between the first set of image parameters and the second set of image parameters; The product of the first similarity and the second similarity is calculated to obtain the matching degree between the set of location images and the candidate recognition model.

[0062] Optionally, the training image dataset can be a model training set, a validation set, or a feature extraction set; this invention does not impose any limitations.

[0063] Optionally, the acquisition of the training image dataset can be based on model metadata, training logs, or database interfaces, and this invention does not limit the scope of the acquisition.

[0064] Optionally, the first similarity can be image content similarity, feature vector similarity, or distribution similarity; the present invention does not limit this.

[0065] Optionally, the first set of image parameters can be a parameter vector, a parameter matrix, or structured data; this invention does not impose any limitations.

[0066] Optionally, the process of determining the first set of image parameters can be based on image metadata parsing, parameter extraction, or data cleaning, and this invention does not limit this process.

[0067] Optionally, the second set of image parameters can be a parameter vector, a parameter matrix, or structured data; this invention does not impose any limitations.

[0068] Optionally, the process of determining the second set of image parameters can be based on training data annotation, metadata reading, or parameter reconstruction, and the present invention does not limit this.

[0069] Optionally, the second similarity can be parameter vector similarity, normalized distance, or weighted similarity, and this invention does not limit it.

[0070] As can be seen, through the above optional embodiments, the matching degree is obtained by calculating the product of image similarity and shooting parameter similarity between the location image set and the training image data set. Based on the accurate identification model, the comprehensiveness and accuracy of model matching are improved by two-dimensional similarity analysis, which provides a high-quality model basis for the identification of line location defects and reduces the risk of matching deviation caused by single similarity dependence.

[0071] As an optional embodiment, the step above, determining the defect identification information corresponding to each image based on the corresponding recognition model, includes: For each image, determine the recognition model corresponding to the line position in that image; Identify at least two associated images corresponding to this image; After stitching the image together with all associated images, the image is input into the corresponding recognition model to obtain the defect recognition information corresponding to the image; the defect recognition information includes the defect type and defect location.

[0072] Optionally, the associated image can be an adjacent captured image, an overlapping area image, or an image from a similar perspective; the present invention does not impose any limitations on this.

[0073] Optionally, the process of determining the associated image can be based on image matching degree, spatial relationship or time series, and the present invention does not limit it.

[0074] Optionally, the stitching process can be panoramic stitching, partial stitching, or feature-level stitching; this invention does not impose any limitations.

[0075] As can be seen, through the above optional embodiments, the defect type and location are obtained by inputting the stitched image and its associated image into the recognition model. Thus, on the basis of accurate defect recognition, the continuity and accuracy of defect recognition are improved by multi-image stitching and correlation analysis, providing reliable recognition information for continuous defect judgment and reducing the risk of missed defect detection due to the limitations of a single image.

[0076] As an optional embodiment, the step of determining at least two associated images corresponding to the image in the above steps includes: Calculate the shooting location distance and image similarity between this image and any other image; Calculate the product of the shooting location distance and the image similarity to obtain the image matching degree between the image and any other image; Filter out at least two other images whose image matching degree is greater than a preset matching degree threshold, and obtain at least two associated images corresponding to that image.

[0077] Optionally, the shooting location distance can be a three-dimensional spatial distance, a projection distance, or a geographical distance; this invention does not impose any limitations.

[0078] Optionally, the image similarity can be pixel-level similarity, feature-level similarity, or semantic similarity; this invention does not impose any limitations.

[0079] Optionally, the matching threshold can be a fixed threshold, a dynamic threshold, or a threshold adjusted based on image quality; this invention does not impose any limitations.

[0080] As can be seen, through the above optional embodiments, by calculating the product of the shooting location distance and image similarity between images, highly matching associated images are filtered. Thus, on the basis of accurate defect identification, the targeting and reliability of associated image selection are improved through location and visual dual-factor analysis, providing optimized input for image stitching and defect identification, and reducing the risk of identification deviation caused by improper associated images.

[0081] As an optional embodiment, the step above, generating the defect determination result corresponding to the target transmission line based on the defect identification information corresponding to all images and the associated continuous defect judgment rule, includes: By filtering out multiple images with the same corresponding defect type, a set of multiple images of the same type is obtained; Calculate the average defect location of all images in each type of image set to obtain the set location corresponding to each type of image set; For any two sets of images of the same type whose positional distance is less than a preset distance threshold, the defect type and positional distance of the two sets of images of the same type are input into the trained associated continuous defect recognition model to obtain the associated continuous defect risk of the two sets of images of the same type. Determine whether the risk of associated continuous defects is greater than a preset risk threshold. If so, identify the two sets of images of the same type as a high-risk set pair. All high-risk sets are displayed on the line risk map corresponding to the preset target transmission line to generate the defect judgment result corresponding to the target transmission line.

[0082] Optionally, the associated continuous defect identification model can be a classification model, a regression model, or a graph neural network model, and the present invention does not limit it.

[0083] Optionally, the risk threshold can be a fixed threshold, a dynamic threshold, or a threshold adjusted based on historical data; this invention does not impose any limitations.

[0084] Optionally, the route risk map can be a two-dimensional risk map, a three-dimensional risk map, or a heat map; the present invention does not impose any limitations on this.

[0085] Optionally, the display process can be implemented based on graphic rendering, color encoding, or an interactive interface; this invention does not limit the scope of the demonstration.

[0086] As can be seen, through the above optional embodiments, by filtering image sets with the same defect type and calculating the set location, and using the location distance and defect type input association continuous defect identification model to determine high-risk set pairs and display them in the risk map, the comprehensiveness of defect determination and early warning efficiency are improved through continuous risk assessment and visualization, based on the generation of accurate defect determination results. This provides scientific support for transmission line maintenance and reduces the line safety risks caused by ignoring continuous defects.

[0087] Example 2 Please see Figure 2 , Figure 2 This is a schematic diagram of a high-voltage transmission line defect identification system based on multi-model collaboration disclosed in an embodiment of the present invention. Figure 2 The described multi-model collaborative high-voltage transmission line defect identification system can be applied to data processing systems / data processing equipment / data processing servers (wherein, the server includes a local processing server or a cloud processing server). For example... Figure 2 As shown, the high-voltage transmission line defect identification system based on multi-model collaboration may include: The acquisition module 201 is used to acquire multiple images corresponding to multiple line locations of the target transmission line.

[0088] The determination module 202 is used to determine the identification model corresponding to each line position.

[0089] The recognition module 203 is used to determine the defect recognition information corresponding to each image based on the corresponding recognition model.

[0090] The generation module 204 is used to generate the defect judgment result corresponding to the target transmission line based on the defect identification information corresponding to all images and the associated continuous defect judgment rule.

[0091] As can be seen, the above-described embodiments of the invention acquire images of multiple locations of the target transmission line and determine the corresponding identification model, identify defect information based on the model, and generate defect judgment results using the associated continuous defect judgment rules. This enables accurate transmission line defect assessment based on multi-model matching and continuous defect analysis, improves line inspection efficiency and defect early warning accuracy, and reduces the risk of line faults caused by missed or misjudged defects.

[0092] As an optional embodiment, the line location is the location of the line tower, conductor, insulator, or hardware; the image is captured based on multiple different shooting parameters; the shooting parameters are the lens position, lens focal length, shooting angle, shooting frame, or shooting light sensitivity parameters.

[0093] As can be seen, the above optional embodiments define the details of the line location and image capture parameters to comprehensively characterize the image-related features of the transmission line, assist in realizing accurate transmission line defect assessment based on multi-model matching and continuous defect analysis, improve the efficiency of line inspection and the accuracy of defect early warning, and reduce the risk of line failure caused by missed or misjudged defects.

[0094] As an alternative embodiment, the images are obtained by taking pictures of the line location using drones or monitoring equipment.

[0095] As can be seen, the above optional embodiments limit the acquisition of relevant images by taking pictures of the line location using drones or monitoring equipment, so as to more conveniently and efficiently take pictures of high-voltage transmission lines, assist in realizing accurate transmission line defect assessment based on multi-model matching and continuous defect analysis, improve the efficiency of line inspection and the accuracy of defect early warning, and reduce the risk of line failure caused by missed defects or misjudgments.

[0096] As an optional embodiment, the specific method by which the determining module determines the identification model corresponding to each line position includes: For each route location, obtain all images corresponding to that route location to get a location image set; Determine the matching degree between the set of location images and each candidate recognition model; The candidate recognition model with the highest matching degree is determined as the recognition model corresponding to the location of the line; the recognition model is trained by a training dataset that includes multiple training transmission line images and corresponding defect annotations.

[0097] As can be seen, through the above optional embodiments, by calculating the matching degree between the location image set and the candidate recognition model for each line location and selecting the model with the highest matching degree, the targeting and accuracy of the recognition model selection are improved through the product analysis of image and shooting parameter similarity on the basis of accurate defect assessment, providing reliable model support for defect recognition and reducing the risk of recognition error caused by model mismatch.

[0098] As an optional embodiment, the specific method by which the determining module determines the matching degree between the set of location images and each candidate recognition model includes: For each candidate recognition model, obtain the training image data set corresponding to that candidate recognition model; Calculate the first similarity between the set of location images and the training image data set; The shooting parameters corresponding to each image in the location image set are defined as the first image parameter set; The shooting parameters corresponding to each image in the training image dataset are determined as the second image parameter set; Calculate the second similarity between the first set of image parameters and the second set of image parameters; The product of the first similarity and the second similarity is calculated to obtain the matching degree between the set of location images and the candidate recognition model.

[0099] As can be seen, through the above optional embodiments, the matching degree is obtained by calculating the product of image similarity and shooting parameter similarity between the location image set and the training image data set. Based on the accurate identification model, the comprehensiveness and accuracy of model matching are improved by two-dimensional similarity analysis, which provides a high-quality model basis for the identification of line location defects and reduces the risk of matching deviation caused by single similarity dependence.

[0100] As an optional embodiment, the recognition module determines the specific method for identifying defect information corresponding to each image based on the corresponding recognition model, including: For each image, determine the recognition model corresponding to the line position in that image; Identify at least two associated images corresponding to this image; After stitching the image together with all associated images, the image is input into the corresponding recognition model to obtain the defect recognition information corresponding to the image; the defect recognition information includes the defect type and defect location.

[0101] As can be seen, through the above optional embodiments, the defect type and location are obtained by inputting the stitched image and its associated image into the recognition model. Thus, on the basis of accurate defect recognition, the continuity and accuracy of defect recognition are improved by multi-image stitching and correlation analysis, providing reliable recognition information for continuous defect judgment and reducing the risk of missed defect detection due to the limitations of a single image.

[0102] As an optional embodiment, the recognition module determines the specific method by which it identifies at least two associated images corresponding to the image, including: Calculate the shooting location distance and image similarity between this image and any other image; Calculate the product of the shooting location distance and the image similarity to obtain the image matching degree between the image and any other image; Filter out at least two other images whose image matching degree is greater than a preset matching degree threshold, and obtain at least two associated images corresponding to that image.

[0103] As can be seen, through the above optional embodiments, by calculating the product of the shooting location distance and image similarity between images, highly matching associated images are filtered. Thus, on the basis of accurate defect identification, the targeting and reliability of associated image selection are improved through location and visual dual-factor analysis, providing optimized input for image stitching and defect identification, and reducing the risk of identification deviation caused by improper associated images.

[0104] As an optional embodiment, the specific method by which the generation module generates the defect judgment result corresponding to the target transmission line based on the defect identification information corresponding to all images and the associated continuous defect judgment rule includes: By filtering out multiple images with the same corresponding defect type, a set of multiple images of the same type is obtained; Calculate the average defect location of all images in each type of image set to obtain the set location corresponding to each type of image set; For any two sets of images of the same type whose positional distance is less than a preset distance threshold, the defect type and positional distance of the two sets of images of the same type are input into the trained associated continuous defect recognition model to obtain the associated continuous defect risk of the two sets of images of the same type. Determine whether the risk of associated continuous defects is greater than a preset risk threshold. If so, identify the two sets of images of the same type as a high-risk set pair. All high-risk sets are displayed on the line risk map corresponding to the preset target transmission line to generate the defect judgment result corresponding to the target transmission line.

[0105] As can be seen, through the above optional embodiments, by filtering image sets with the same defect type and calculating the set location, and using the location distance and defect type input association continuous defect identification model to determine high-risk set pairs and display them in the risk map, the comprehensiveness of defect determination and early warning efficiency are improved through continuous risk assessment and visualization, based on the generation of accurate defect determination results. This provides scientific support for transmission line maintenance and reduces the line safety risks caused by ignoring continuous defects.

[0106] Example 3 Please see Figure 3 , Figure 3 This is another high-voltage transmission line defect identification system based on multi-model collaboration disclosed in the embodiments of the present invention. Figure 3 The described multi-model collaborative high-voltage transmission line defect identification system is applied in a data processing system / data processing equipment / data processing server (wherein, the server includes a local processing server or a cloud processing server). Figure 3 As shown, the high-voltage transmission line defect identification system based on multi-model collaboration may include: Memory 301 storing executable program code; Processor 302 coupled to memory 301; The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the high-voltage transmission line defect identification method based on multi-model collaboration described in Embodiment 1.

[0107] Example 4 This invention discloses a computer read storage medium that stores a computer program for electronic data exchange, wherein the computer program causes a computer to execute the steps of the high-voltage transmission line defect identification method based on multi-model collaboration described in Embodiment 1.

[0108] Example 5 This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps of the high-voltage transmission line defect identification method based on multi-model collaboration described in Embodiment 1.

[0109] The foregoing has described specific embodiments of this specification; other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than those shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily have to follow the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0110] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0111] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware.

[0112] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0113] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0114] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0115] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0116] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0117] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0118] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0119] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0120] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0121] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0122] Finally, it should be noted that the high-voltage transmission line defect identification method and system based on multi-model collaboration disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention, and are only used to illustrate the technical solutions of the present invention, not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying defects in high-voltage transmission lines based on multi-model collaboration, characterized in that, The method includes: Acquire multiple images corresponding to multiple locations of the target transmission line; Determine the recognition model corresponding to each of the aforementioned line locations; Based on the corresponding recognition model, the defect recognition information corresponding to each image is determined; Based on the defect identification information corresponding to all the images, and based on the associated continuous defect judgment rule, the defect judgment result corresponding to the target transmission line is generated.

2. The high-voltage transmission line defect identification method based on multi-model collaboration according to claim 1, characterized in that, The line location refers to the location of the line tower, conductor, insulator, or hardware; the image is captured based on multiple different shooting parameters; the shooting parameters include lens position, lens focal length, shooting angle, shooting frame, or shooting light sensitivity parameters.

3. The high-voltage transmission line defect identification method based on multi-model collaboration according to claim 1, characterized in that, The images were obtained by taking pictures of the location of the line using drones or monitoring equipment.

4. The high-voltage transmission line defect identification method based on multi-model collaboration according to claim 1, characterized in that, The determination of the identification model corresponding to each of the aforementioned line locations includes: For each of the aforementioned route locations, all the images corresponding to that route location are obtained to obtain a location image set; Determine the matching degree between the set of location images and each candidate recognition model; The candidate recognition model with the highest matching degree is determined as the recognition model corresponding to the line location; the recognition model is trained by a training dataset including multiple training transmission line images and corresponding defect annotations.

5. The high-voltage transmission line defect identification method based on multi-model collaboration according to claim 4, characterized in that, Determining the matching degree between the location image set and each candidate recognition model includes: For each candidate recognition model, obtain the training image data set corresponding to that candidate recognition model; Calculate the first similarity between the set of location images and the set of training image data; The shooting parameters corresponding to each image in the location image set are determined to be the first image parameter set; The shooting parameters corresponding to each image in the training image dataset are determined as the second image parameter set; Calculate the second similarity between the first image parameter set and the second image parameter set; The product of the first similarity and the second similarity is calculated to obtain the matching degree between the location image set and the candidate recognition model.

6. The high-voltage transmission line defect identification method based on multi-model collaboration according to claim 1, characterized in that, The step of determining the defect identification information corresponding to each image based on the corresponding recognition model includes: For each image, determine the recognition model corresponding to the line position corresponding to that image; Identify at least two associated images corresponding to this image; After stitching the image together with all the associated images, the image is input into the corresponding recognition model to obtain the defect recognition information corresponding to the image; the defect recognition information includes the defect type and the defect location.

7. The high-voltage transmission line defect identification method based on multi-model collaboration according to claim 6, characterized in that, The determination of at least two associated images corresponding to the image includes: Calculate the shooting location distance and image similarity between the image and any other image mentioned above; Calculate the product of the shooting location distance and the image similarity to obtain the image matching degree between the image and any other image. Filter out at least two other images whose matching degree is greater than a preset matching degree threshold to obtain at least two associated images corresponding to the image.

8. The high-voltage transmission line defect identification method based on multi-model collaboration according to claim 6, characterized in that, The step of generating a defect determination result for the target transmission line based on defect identification information corresponding to all the images and on a rule for determining associated continuous defects includes: By filtering out multiple images of the same defect type, a set of multiple image sets of the same type is obtained; Calculate the average value of the defect locations of all images in each set of images of the same type to obtain the set location corresponding to each set of images of the same type; For any two sets of images of the same type whose positional distance between them is less than a preset distance threshold, the defect type and positional distance corresponding to the two sets of images of the same type are input into the trained associated continuous defect recognition model to obtain the associated continuous defect risk corresponding to the two sets of images of the same type. Determine whether the risk of the associated continuous defects is greater than a preset risk threshold. If so, identify the two sets of images of the same type as a high-risk set pair. All the high-risk sets are displayed in the preset line risk map corresponding to the target transmission line to generate the defect judgment result corresponding to the target transmission line.

9. A high-voltage transmission line defect identification system based on multi-model collaboration, characterized in that, The system includes: The acquisition module is used to acquire multiple images corresponding to multiple locations of the target transmission line; A determination module is used to determine the identification model corresponding to each of the aforementioned line positions; The recognition module is used to determine the defect recognition information corresponding to each image based on the corresponding recognition model. The generation module is used to generate the defect judgment result corresponding to the target transmission line based on the defect identification information corresponding to all the images and the associated continuous defect judgment rule.

10. A high-voltage transmission line defect identification system based on multi-model collaboration, characterized in that, The system includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the high-voltage transmission line defect identification method based on multi-model collaboration as described in any one of claims 1-8.

Images