Power line and environment intelligent identification method and device based on image recognition

Abstract

The present application relates to a method, device and equipment for intelligent identification of power lines and environments based on image recognition. The method includes: obtaining image data of the environment in which the distribution line is located and the distribution line, and identifying the three-dimensional structural data corresponding to the image data of the environment and the distribution line through an image recognition network; constructing a three-dimensional distribution line model corresponding to the distribution line based on the three-dimensional structural data of the environment and the distribution line, and splitting the three-dimensional distribution line model into multiple sub-structure models; identifying the structural information corresponding to each sub-structure model, and analyzing the structural feature information and position feature information of each structural information of the distribution line based on each structural information; identifying the initial non-stop operation feature information of the distribution line based on the structural feature information and position feature information of each structural information, and screening the target non-stop operation feature information that meets the non-stop operation conditions. The use of this method can improve the accuracy of the non-stop operation features of the identified distribution lines.

Description

Technical Field

[0001] The present application relates to the technical field of power grid distribution line survey, and in particular to a method and device for intelligently identifying power lines and environments based on image recognition. Background Art

[0002] Before the non-stop operation of the distribution line, it is necessary to judge the situation of the distribution lines and environmental information in the current scene, so as to determine the non-stop operation method. The current non-stop operation of the distribution line mainly relies on design drawings and manual on-site surveys. This method has many problems, such as long on-site surveys, low efficiency, incomplete information collection, and it is difficult for non-non-stop operation personnel to grasp the key points of the on-site survey during the initial survey. It has high requirements for operators and is difficult to cope with multi-scenario survey tasks for sudden events. Therefore, how to accurately identify the non-stop operation characteristics of the distribution line in the environment is the current research focus.

[0003] The traditional way to identify the non-stop operation characteristics of distribution lines in the current scene is to have multiple workers design drawings and conduct manual on-site surveys, and then manually analyze the non-stop operation characteristics of distribution lines. However, the accuracy of manual drawings is low, and the non-stop operation characteristics analyzed manually are limited by the work experience of workers, resulting in low accuracy of the non-stop operation characteristics of distribution lines. Summary of the invention

[0004] Based on this, it is necessary to provide an intelligent recognition solution for power lines and environments based on image recognition to address the above technical problems.

[0005] In a first aspect, the present application provides a method for intelligently identifying power lines and environments based on image recognition. The method comprises:

[0006] Acquire environmental image data of the environment in which the distribution line is located and power line image data of the distribution line, and identify the environmental three-dimensional structure data corresponding to the environmental image data and the power line three-dimensional structure data corresponding to the power line image data through an image recognition network;

[0007] Based on the three-dimensional structure data of the environment and the three-dimensional structure data of the power line, construct a three-dimensional distribution line model corresponding to the distribution line, and split the three-dimensional distribution line model into multiple sub-structure models;

[0008] Identify the structural information corresponding to each substructure model, and analyze the structural feature information of each structural information of the distribution line and the position feature information of each structural information based on the structural information;

[0009] Based on the structural feature information of each of the structural information and the position feature information of each of the structural information, the initial non-stop operation feature information of the distribution line is identified, and in each of the initial non-stop operation feature information, the target non-stop operation feature information that meets the non-stop operation conditions is screened.

[0010] Optionally, identifying the three-dimensional environmental structure data corresponding to the environmental image data and the three-dimensional power line structure data corresponding to the power line image data through an image recognition network includes:

[0011] Identifying, by an image recognition network, environmental image edge information of each of the environmental image data and power line image edge information corresponding to each of the power line image data;

[0012] Perform three-dimensional stitching on the edge information of each environmental image according to the acquisition angle of each environmental image data to obtain the edge information of the environmental three-dimensional image, and perform three-dimensional stitching on the edge information of each power line image according to the acquisition angle of each power line image data to obtain the edge information of the power line three-dimensional image;

[0013] Relative position information between edge position points in the environmental three-dimensional image edge information is identified to obtain environmental three-dimensional structure data corresponding to the environmental image data, and relative position information between edge position points in the power line three-dimensional image edge information is identified to obtain power line three-dimensional structure data corresponding to the power line image data.

[0014] Optionally, constructing a three-dimensional distribution line model corresponding to the distribution line based on the three-dimensional structure data of the environment and the three-dimensional structure data of the power line includes:

[0015] Based on the relative position relationship between each edge position point in the three-dimensional structure data of the environment, each edge position point is subjected to three-dimensional connection processing to obtain a three-dimensional connection structure of the environment, and based on the relative position relationship between each edge position point in the three-dimensional structure data of the power line, each edge position point is subjected to three-dimensional connection processing to obtain a three-dimensional connection structure of the power line;

[0016] Identifying the environment structure edge connection lines of the environment three-dimensional connection structure and the power line structure edge connection lines of the power line three-dimensional connection structure;

[0017] Based on the edge lines of the environmental structure, the three-dimensional connection structure of the environment is simplified to obtain a three-dimensional structure model of the environment, and based on the edge lines of the power line structure, the three-dimensional connection structure of the power line is simplified to obtain a three-dimensional structure model of the power line;

[0018] The relative position relationship between the three-dimensional structure model of the environment and the three-dimensional structure model of the power line is identified, and based on the relative position relationship, the three-dimensional structure model of the environment and the three-dimensional structure model of the power line are subjected to model splicing processing to obtain a three-dimensional distribution line model of the distribution line.

[0019] Optionally, splitting the three-dimensional power distribution line model into a plurality of substructure models includes:

[0020] Identifying three-dimensional structural features of the three-dimensional distribution line model, and identifying a three-dimensional position information range corresponding to each three-dimensional structural feature;

[0021] Based on the three-dimensional position information range corresponding to each three-dimensional structural feature, the three-dimensional distribution line model is split to obtain a sub-structure model corresponding to each three-dimensional structural feature.

[0022] Optionally, the identifying structure information corresponding to each substructure model includes:

[0023] In response to a structure recognition operation of a user, a structure type corresponding to each substructure model is acquired, and a relative position relationship between each substructure model and the three-dimensional power distribution line model is identified;

[0024] The sample structure information corresponding to each structure type is queried, and based on the substructure model corresponding to each structure type and the relative position relationship between each substructure model and the three-dimensional distribution line model, the sample structure information corresponding to each structure type is adjusted to obtain the structure information corresponding to each substructure model; the structure information includes environmental structure information and power line structure information.

[0025] Optionally, analyzing the structural feature information of each structural information of the distribution line and the location feature information of each structural information based on each structural information includes:

[0026] For each piece of structural information, based on the relative position relationship between the substructure model of the structural information and the three-dimensional distribution line model, the relative position information between the structural information and other structural information is identified, and the interaction influence information between each piece of structural information is collected;

[0027] Based on the interaction information between each structure information and the relative position information between the structure information and other structure information, analyzing the position feature information of the structure information;

[0028] Identify structural edge information of different perspectives of the substructure model of the structural information through an image recognition network, and identify a structural dialysis diagram of each perspective based on the structural edge information of each perspective;

[0029] The graph feature information of the structural dialysis graph of each viewing angle is analyzed to obtain the structural feature information of the structural information.

[0030] Optionally, the identifying the initial non-stop operation characteristic information of the distribution line based on the structural characteristic information of each structural information and the position characteristic information of each structural information includes:

[0031] For each piece of structural information, based on the position feature information of the structural information, structural distance information between the structural information and other structural information is identified, and based on the structural type of each piece of structural information and the structural distance information between the structural information and other structural information, a distance influence feature of the structural information is identified;

[0032] Acquire structural material feature information of each structural information, and identify structural support features of the structural information and uninterrupted operation impact features of the structural information based on feature information of each figure of the structural information and structural material feature information of the structural information;

[0033] The distance influence feature of the structural information, the structural support feature of the structural information, and the non-stop operation influence feature of the structural information are used as the initial non-stop operation feature information of the structural information.

[0034] Optionally, the step of screening target non-stop power operation characteristic information satisfying non-stop power operation conditions from each of the initial non-stop power operation characteristic information includes:

[0035] Obtain multiple types of non-stop operation tasks and environmental requirement information of each non-stop operation task, and analyze the non-stop operation conditions corresponding to each environmental requirement information;

[0036] Analyze each non-stop operation condition to obtain the distance requirement information of each non-stop operation task, the structural support requirement information of each non-stop operation task, and the non-stop operation impact limitation information of each non-stop operation task;

[0037] For each non-stop operation task, identify a first deviation value between a distance impact feature of each structural information and the distance requirement information of the non-stop operation task, a second deviation value between a structural support feature of each structural information and the structural support requirement information of the non-stop operation task, and a third deviation value between a non-stop operation impact feature of each structural information and the non-stop operation impact limitation information of the non-stop operation task;

[0038] The initial non-stop power operation characteristic information of the target structure information whose first deviation value is lower than the first deviation threshold, the second deviation value is lower than the second deviation threshold, and the third deviation value is lower than the third deviation threshold is selected as the target non-stop power operation characteristic information corresponding to the non-stop power operation task.

[0039] In a second aspect, the present application also provides a device for intelligently identifying power lines and environments based on image recognition. The device comprises:

[0040] An acquisition module, used to acquire environmental image data of the environment in which the distribution line is located and power line image data of the distribution line, and to identify environmental three-dimensional structure data corresponding to the environmental image data and power line three-dimensional structure data corresponding to the power line image data through an image recognition network;

[0041] A construction module, configured to construct a three-dimensional distribution line model corresponding to the distribution line based on the three-dimensional structure data of the environment and the three-dimensional structure data of the power line, and split the three-dimensional distribution line model into a plurality of sub-structure models;

[0042] An analysis module, used for identifying the structural information corresponding to each substructure model, and analyzing the structural feature information of each structural information of the distribution line and the position feature information of each structural information based on the structural information;

[0043] A screening module is used to identify the initial non-stop operation characteristic information of the distribution line based on the structural characteristic information of each of the structural information and the position characteristic information of each of the structural information, and to screen the target non-stop operation characteristic information that meets the non-stop operation conditions among each of the initial non-stop operation characteristic information.

[0044] Optionally, the acquisition module is specifically used to:

[0045] Identifying, by an image recognition network, environmental image edge information of each of the environmental image data and power line image edge information corresponding to each of the power line image data;

[0046] Perform three-dimensional stitching on the edge information of each environmental image according to the acquisition angle of each environmental image data to obtain the edge information of the environmental three-dimensional image, and perform three-dimensional stitching on the edge information of each power line image according to the acquisition angle of each power line image data to obtain the edge information of the power line three-dimensional image;

[0047] Relative position information between edge position points in the environmental three-dimensional image edge information is identified to obtain environmental three-dimensional structure data corresponding to the environmental image data, and relative position information between edge position points in the power line three-dimensional image edge information is identified to obtain power line three-dimensional structure data corresponding to the power line image data.

[0048] Optionally, the building module is specifically used to:

[0049] Based on the relative position relationship between each edge position point in the three-dimensional structure data of the environment, each edge position point is subjected to three-dimensional connection processing to obtain a three-dimensional connection structure of the environment, and based on the relative position relationship between each edge position point in the three-dimensional structure data of the power line, each edge position point is subjected to three-dimensional connection processing to obtain a three-dimensional connection structure of the power line;

[0050] Identifying the environment structure edge connection lines of the environment three-dimensional connection structure and the power line structure edge connection lines of the power line three-dimensional connection structure;

[0051] Based on the edge lines of the environmental structure, the three-dimensional connection structure of the environment is simplified to obtain a three-dimensional structure model of the environment, and based on the edge lines of the power line structure, the three-dimensional connection structure of the power line is simplified to obtain a three-dimensional structure model of the power line;

[0052] The relative position relationship between the three-dimensional structure model of the environment and the three-dimensional structure model of the power line is identified, and based on the relative position relationship, the three-dimensional structure model of the environment and the three-dimensional structure model of the power line are subjected to model splicing processing to obtain a three-dimensional distribution line model of the distribution line.

[0053] Optionally, the building module is specifically used to:

[0054] Identifying three-dimensional structural features of the three-dimensional distribution line model, and identifying a three-dimensional position information range corresponding to each three-dimensional structural feature;

[0055] Based on the three-dimensional position information range corresponding to each three-dimensional structural feature, the three-dimensional distribution line model is split to obtain a sub-structure model corresponding to each three-dimensional structural feature.

[0056] Optionally, the analysis module is specifically used to:

[0057] In response to a structure recognition operation of a user, a structure type corresponding to each substructure model is acquired, and a relative position relationship between each substructure model and the three-dimensional power distribution line model is identified;

[0058] The sample structure information corresponding to each structure type is queried, and based on the substructure model corresponding to each structure type and the relative position relationship between each substructure model and the three-dimensional distribution line model, the sample structure information corresponding to each structure type is adjusted to obtain the structure information corresponding to each substructure model; the structure information includes environmental structure information and power line structure information.

[0059] Optionally, the analysis module is specifically used to:

[0060] For each piece of structural information, based on the relative position relationship between the substructure model of the structural information and the three-dimensional distribution line model, the relative position information between the structural information and other structural information is identified, and the interaction influence information between each piece of structural information is collected;

[0061] Based on the interaction information between each structure information and the relative position information between the structure information and other structure information, analyzing the position feature information of the structure information;

[0062] Identify structural edge information of different perspectives of the substructure model of the structural information through an image recognition network, and identify a structural dialysis diagram of each perspective based on the structural edge information of each perspective;

[0063] The graph feature information of the structural dialysis graph of each viewing angle is analyzed to obtain the structural feature information of the structural information.

[0064] Optionally, the screening module is specifically used to:

[0065] For each piece of structural information, based on the position feature information of the structural information, structural distance information between the structural information and other structural information is identified, and based on the structural type of each piece of structural information and the structural distance information between the structural information and other structural information, a distance influence feature of the structural information is identified;

[0066] Acquire structural material feature information of each structural information, and identify structural support features of the structural information and uninterrupted operation impact features of the structural information based on feature information of each figure of the structural information and structural material feature information of the structural information;

[0067] The distance influence feature of the structural information, the structural support feature of the structural information, and the non-stop operation influence feature of the structural information are used as the initial non-stop operation feature information of the structural information.

[0068] Optionally, the screening module is specifically used to:

[0069] Obtain multiple types of non-stop operation tasks and environmental requirement information of each non-stop operation task, and analyze the non-stop operation conditions corresponding to each environmental requirement information;

[0070] Analyze each non-stop operation condition to obtain the distance requirement information of each non-stop operation task, the structural support requirement information of each non-stop operation task, and the non-stop operation impact limitation information of each non-stop operation task;

[0071] For each non-stop operation task, identify a first deviation value between a distance impact feature of each structural information and the distance requirement information of the non-stop operation task, a second deviation value between a structural support feature of each structural information and the structural support requirement information of the non-stop operation task, and a third deviation value between a non-stop operation impact feature of each structural information and the non-stop operation impact limitation information of the non-stop operation task;

[0072] The initial non-stop power operation characteristic information of the target structure information whose first deviation value is lower than the first deviation threshold, the second deviation value is lower than the second deviation threshold, and the third deviation value is lower than the third deviation threshold is selected as the target non-stop power operation characteristic information corresponding to the non-stop power operation task.

[0073] In a third aspect, the present application provides a computer device, wherein the computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of any one of the methods in the first aspect are implemented.

[0074] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0075] The above-mentioned method and device for intelligent recognition of power lines and environments based on image recognition obtains environmental image data of the environment in which the distribution line is located and power line image data of the distribution line, and identifies the environmental three-dimensional structure data corresponding to the environmental image data and the power line three-dimensional structure data corresponding to the power line image data through an image recognition network; constructs a three-dimensional distribution line model corresponding to the distribution line based on the environmental three-dimensional structure data and the power line three-dimensional structure data, and splits the three-dimensional distribution line model into multiple sub-structure models; identifies the structure information corresponding to each sub-structure model, and analyzes the structure feature information of each structure information of the distribution line and the position feature information of each structure information based on each structure information; identifies the initial non-stop operation feature information of the distribution line based on the structure feature information of each structure information and the position feature information of each structure information, and screens the target non-stop operation feature information that meets the non-stop operation conditions from each initial non-stop operation feature information. This solution uses an image recognition network to identify the environmental three-dimensional structure data of the environmental image data and the three-dimensional structure data of the power line of the power line image, thereby constructing a three-dimensional distribution line model corresponding to the distribution line, which not only avoids the inefficiency of manual composition, but also improves the accuracy of constructing the three-dimensional structure model of the distribution line. Then, by identifying the structure information corresponding to each substructure model of the three-dimensional structure model and analyzing the position feature information and the structure feature information of each structure information, the initial non-stop operation feature information of each structure information is obtained, avoiding the problem of analysis result deviation caused by different experience of manual analysis, and improving the comprehensiveness and accuracy of the analysis of the structure information. Finally, by screening the target non-stop operation feature information that meets the non-stop operation conditions, the efficiency of determining the structural features of the structure information corresponding to each non-stop operation task is improved, which is convenient for planning the operation route when executing the non-stop operation task in the subsequent time, thereby comprehensively improving the accuracy of the non-stop operation features of the identified distribution lines. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 A schematic diagram of a flow chart of a method for intelligently identifying power lines and environments based on image recognition in one embodiment;

[0077] Figure 2 It is a structural block diagram of a power line and environment intelligent recognition device based on image recognition in one embodiment;

[0078] Figure 3 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0079] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0080] The method for intelligently identifying power lines and environments based on image recognition provided by the embodiment of the present application can be applied to the application environment of three-dimensional modeling of power distribution lines of power grids. The method can be applied to a terminal, a server, or a system including a terminal and a server, and implemented through the interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, etc. Among them, the terminal identifies the environmental three-dimensional structure data of the environmental image data and the three-dimensional structure data of the power line image through the image recognition network, so as to construct a three-dimensional distribution line model corresponding to the distribution line, which not only avoids the inefficiency of manual composition, but also improves the accuracy of constructing the three-dimensional structure model of the distribution line. Then, by identifying the structural information corresponding to each sub-structure model of the three-dimensional structure model and analyzing the position feature information and the structural feature information of each structure information, the initial non-stop operation feature information of each structure information is obtained, which avoids the problem of analysis result deviation caused by different experience of manual analysis and improves the comprehensiveness and accuracy of the analysis of the structural information. Finally, by screening the target non-stop operation feature information that meets the non-stop operation conditions, the efficiency of determining the structural features of the structural information corresponding to each non-stop operation task is improved, which is convenient for planning the operation route when executing the non-stop operation task in the future, thereby comprehensively improving the accuracy of the non-stop operation features of the identified distribution lines.

[0081] In one embodiment, Figure 1 As shown, a method for intelligently identifying power lines and environments based on image recognition is provided, which is described by taking the method applied to a terminal as an example, and includes the following steps:

[0082] Step S101, obtaining environmental image data of the environment in which the distribution line is located and power line image data of the distribution line, and identifying the environmental three-dimensional structure data corresponding to the environmental image data and the power line three-dimensional structure data corresponding to the power line image data through an image recognition network.

[0083] In this embodiment, the terminal receives environmental image data of different perspectives of the environment in which the distribution line is located, and the power line image data of the distribution line, which are transmitted by an image acquisition device equipped with point cloud scanning technology. Then, the terminal identifies the distance information of each edge position point corresponding to the environmental image data of each perspective, and the distance information of each edge position point corresponding to the power line image data through the image recognition network. Among them, the image recognition network is a convolutional neural network based on an image edge recognition algorithm, and the convolutional neural network is a convolutional neural network composed of multiple convolutional layers, pooling layers, and fully connected layers in a specific sequence. Then, based on the distance information of each edge position point corresponding to the environmental image data, the terminal identifies the environmental three-dimensional structure data corresponding to the environmental image data, and based on the distance information of each edge position point corresponding to the power line image data, identifies the power line three-dimensional structure data corresponding to the power line image data. The specific recognition process will be described in detail later. The distance information is three-dimensional relative distance information.

[0084] Step S102: construct a three-dimensional distribution line model corresponding to the distribution line based on the three-dimensional structure data of the environment and the three-dimensional structure data of the power line, and split the three-dimensional distribution line model into multiple sub-structure models.

[0085] In this embodiment, the terminal constructs a three-dimensional distribution line model corresponding to the distribution line based on the three-dimensional structure data of the environment and the three-dimensional structure data of the power line, and splits the three-dimensional distribution line model into multiple sub-structure models. Among them, the way to construct the three-dimensional distribution line model is to construct the three-dimensional structure model through the finite element algorithm. The three-dimensional distribution line model is a three-dimensional structure model of the distribution line including the environment in which the distribution line is located. The specific construction process will be described in detail later. The terminal splits the distribution line model into sub-structure models of different structures, wherein the sub-structure model includes a sub-structure model corresponding to the environmental structure and a sub-structure model corresponding to the power line structure. The specific splitting process will be described in detail later.

[0086] Step S103, identifying the structural information corresponding to each sub-structure model, and analyzing the structural feature information of each structural information of the distribution line and the position feature information of each structural information based on each structural information.

[0087] In this embodiment, the terminal identifies the structural information corresponding to each substructure model. The structural information is environmental structural information and power line structural information. The environmental structural information includes vegetation structural information, terrain structural information, artificial building structural information, and road structural information. The power line structural information is the structural information of each component structure of the distribution line. Then, based on each structural information, the terminal analyzes the structural feature information of each structural information of the distribution line and the location feature information of each structural information. The specific analysis process and identification process will be described in detail later.

[0088] Step S104, based on the structural feature information of each structural information and the position feature information of each structural information, the initial non-stop operation feature information of the distribution line is identified, and the target non-stop operation feature information that meets the non-stop operation conditions is screened out from each initial non-stop operation feature information.

[0089] In this embodiment, the terminal identifies the initial non-stop operation characteristic information of the distribution line based on the structural characteristic information of each structural information and the location characteristic information of each structural information, and selects the target non-stop operation characteristic information that meets the non-stop operation conditions from each initial non-stop operation characteristic information. The non-stop operation conditions include the non-stop operation conditions of each non-stop operation task, which is the demand information of the non-stop operation task for the distribution line. The specific identification process and screening process will be described in detail later.

[0090] Based on the above scheme, the environmental three-dimensional structure data of the environmental image data and the three-dimensional structure data of the power line of the power line image are identified through the image recognition network, so as to construct a three-dimensional distribution line model corresponding to the distribution line, which not only avoids the inefficiency of manual composition, but also improves the accuracy of constructing the three-dimensional structure model of the distribution line. Then, by identifying the structure information corresponding to each sub-structure model of the three-dimensional structure model and analyzing the position feature information and the structure feature information of each structure information, the initial non-stop operation feature information of each structure information is obtained, which avoids the problem of analysis result deviation caused by different experience of manual analysis and improves the comprehensiveness and accuracy of the analysis of the structure information. Finally, by screening the target non-stop operation feature information that meets the non-stop operation conditions, the efficiency of determining the structural features of the structure information corresponding to each non-stop operation task is improved, which is convenient for planning the operation route when executing the non-stop operation task in the future, thereby comprehensively improving the accuracy of the non-stop operation features of the identified distribution lines.

[0091] Optionally, environmental three-dimensional structure data corresponding to environmental image data and power line three-dimensional structure data corresponding to power line image data are identified through an image recognition network, including: identifying environmental image edge information of each environmental image data and power line image edge information corresponding to each power line image data through an image recognition network; performing three-dimensional splicing on each environmental image edge information according to the acquisition angle of each environmental image data to obtain environmental three-dimensional image edge information, and performing three-dimensional splicing on each power line image edge information according to the acquisition angle of each power line image data to obtain power line three-dimensional image edge information; identifying relative position information between each edge position point in the environmental three-dimensional image edge information to obtain environmental three-dimensional structure data corresponding to the environmental image data, and identifying relative position information between each edge position point in the power line three-dimensional image edge information to obtain power line three-dimensional structure data corresponding to the power line image data.

[0092] In this embodiment, the terminal identifies the environmental image edge information of each environmental image data and the power line image edge information corresponding to each power line image data through an image recognition network. The image edge information is the two-dimensional position information of each connection point of the image edge of the main structure image in the image data. The main structure image includes the environmental structure image in the environmental image data and the power line structure image in the power line image data.

[0093] Then, the terminal performs three-dimensional stitching of the edge information of each environmental image according to the acquisition angle of each environmental image data to obtain the edge information of the environmental three-dimensional image, and performs three-dimensional stitching of the edge information of each power line image according to the acquisition angle of each power line image data to obtain the edge information of the power line three-dimensional image. The specific stitching method is to calculate the three-dimensional relative distance information between the two edge position points by identifying the distance information between the two edge position points in different viewing angles, and then determine the three-dimensional position information between the edge position points in the three-dimensional image based on the three-dimensional relative distance information between the edge position points to obtain the edge information of the environmental three-dimensional image and the edge information of the power line three-dimensional image.

[0094] The terminal identifies the relative position information between each edge position point in the edge information of the environment three-dimensional image, obtains the environment three-dimensional structure data corresponding to the environment image data, and identifies the relative position information between each edge position point in the edge information of the power line three-dimensional image, obtains the power line three-dimensional structure data corresponding to the power line image data. The relative position information is the relative position information between a single edge position point and all other edge position points in each three-dimensional image edge information.

[0095] Based on the above scheme, by identifying the three-dimensional relative distance information between each edge position point in the image, and then stitching the obtained three-dimensional image edge information, and then based on the relative position information between each edge position point, identifying each three-dimensional structure data, the accuracy of the identified three-dimensional structure data is improved.

[0096] Optionally, based on the environmental three-dimensional structure data and the three-dimensional structure data of the power line, a three-dimensional distribution line model corresponding to the distribution line is constructed, including: based on the relative position relationship between each edge position point in the environmental three-dimensional structure data, three-dimensionally connecting each edge position point to obtain an environmental three-dimensional connection structure, and based on the relative position relationship between each edge position point in the three-dimensional structure data of the power line, three-dimensionally connecting each edge position point to obtain a three-dimensional connection structure of the power line; identifying environmental structure edge lines of the environmental three-dimensional connection structure and power line structure edge lines of the three-dimensional connection structure of the power line; based on the environmental structure edge lines, simplifying the environmental three-dimensional connection structure to obtain an environmental three-dimensional structure model, and simplifying the three-dimensional connection structure of the power line based on the power line structure edge lines to obtain a three-dimensional structure model of the power line; identifying the relative position relationship between the environmental three-dimensional structure model and the three-dimensional structure model of the power line, and based on the relative position relationship, model splicing the environmental three-dimensional structure model and the three-dimensional structure model of the power line to obtain a three-dimensional distribution line model of the distribution line.

[0097] In this embodiment, the terminal performs three-dimensional connection processing on each edge position point based on the relative position relationship between each edge position point in the three-dimensional structure data of the environment to obtain the three-dimensional connection structure of the environment, and performs three-dimensional connection processing on each edge position point based on the relative position relationship between each edge position point in the three-dimensional structure data of the power line to obtain the three-dimensional connection structure of the power line. The connection method is to perform spatial straight line connection between each edge position point to obtain the three-dimensional connection structure of the power line corresponding to each edge position point.

[0098] Then, the terminal identifies the edge connection lines of the environment three-dimensional connection structure and the edge connection lines of the power line structure of the power line three-dimensional connection structure. The edge connection lines are the connection information between each edge position point and its adjacent edge position point, wherein the adjacent edge position points are the two edge position points with the closest relative distance.

[0099] Based on the edge connection lines of the environment structure, the three-dimensional connection structure of the environment is simplified to obtain the three-dimensional structure model of the environment, and based on the edge connection lines of the power line structure, the three-dimensional connection structure of the power line is simplified to obtain the three-dimensional structure model of the power line. The simplification process is to eliminate the connection information of each edge position point in each three-dimensional connection structure except the structure edge connection line of the three-dimensional connection structure to obtain the three-dimensional structure model.

[0100] The terminal uses an image recognition network to identify the relative distance between each edge position point in the three-dimensional structure data of the environment and each edge position point in the three-dimensional structure data of the power line of the distribution line, and obtains the three-dimensional relative distance information between each edge position point in the three-dimensional structure data and each edge position point in the three-dimensional structure data of the power line of the distribution line. The terminal then uses the three-dimensional relative distance information as the relative position relationship between the three-dimensional structure model of the environment and the three-dimensional structure model of the power line. Afterwards, based on the relative position relationship, the terminal performs model splicing processing on the three-dimensional structure model of the environment and the three-dimensional structure model of the power line to obtain the three-dimensional distribution line model of the distribution line.

[0101] Based on the above scheme, a three-dimensional connection model is established and then simplified to obtain a three-dimensional structural model. The model simplification process enables the simplified three-dimensional model to occupy a smaller storage space and reduce storage requirements. Then, through the image recognition network, by identifying the edge position points, the relative distance relationship between the models is further determined, and the models are spliced ​​to obtain a three-dimensional distribution line model. The correspondence between different models is judged by matching the position of edge points and other features, which improves the efficiency and accuracy of model construction.

[0102] Optionally, the three-dimensional distribution line model is split into multiple sub-structure models, including: identifying the three-dimensional structural features of the three-dimensional distribution line model, and identifying the three-dimensional position information range corresponding to each three-dimensional structural feature; based on the three-dimensional position information range corresponding to each three-dimensional structural feature, the three-dimensional distribution line model is split to obtain a sub-structure model corresponding to each three-dimensional structural feature.

[0103] In this embodiment, the terminal identifies the three-dimensional structural features of the three-dimensional distribution line model, and identifies the three-dimensional position information range corresponding to each three-dimensional structural feature. The three-dimensional structural feature is the image feature information of the structural information corresponding to each component structure of the three-dimensional distribution line model. The method of identifying the structural features is to identify the image features of each perspective of the three-dimensional structural features through an image feature recognition network to obtain multiple three-dimensional structural features, wherein the image feature recognition network can be a BP (back propagation) neural network.

[0104] The terminal splits the three-dimensional distribution line model based on the three-dimensional position information range corresponding to each three-dimensional structural feature to obtain a sub-structure model corresponding to each three-dimensional structural feature.

[0105] Based on the above scheme, the three-dimensional structural features of the three-dimensional distribution line model are identified to split the three-dimensional distribution line model, thereby improving the accuracy of splitting the three-dimensional distribution line model.

[0106] Optionally, identifying the structural information corresponding to each substructure model includes: in response to the user's structural identification operation, obtaining the structural type corresponding to each substructure model, and identifying the relative position relationship between each substructure model and the three-dimensional distribution line model; querying the sample structural information corresponding to each structural type, and adjusting the sample structural information corresponding to each structural type based on the substructure model corresponding to each structural type and the relative position relationship between each substructure model and the three-dimensional distribution line model to obtain the structural information corresponding to each substructure model; the structural information includes environmental structural information and power line structural information.

[0107] In this embodiment, the terminal obtains the structural type corresponding to each substructure model in response to the user's structural recognition operation. The structural recognition operation process of the user is that the terminal sends each substructure model to the user's client, and then receives the user's recognition result of each substructure model. The terminal extracts the structural type corresponding to each recognition result to obtain the structural type corresponding to each substructure model.

[0108] Then, the terminal identifies the relative position relationship between each substructure model and the three-dimensional power distribution line model, wherein the relative position relationship includes the spatial position of the substructure model in the three-dimensional power distribution line model and the relative distance information between the substructure model and each substructure model.

[0109] The terminal queries the sample structure information corresponding to each structure type. Specifically, the database includes the correspondence between each structure type and the structure. Then, the terminal obtains the sample structure information corresponding to each structure type by querying the parameter information of the structure corresponding to each structure type in the database. Among them, the sample structure information includes the template three-dimensional structure parameters of each component of the environmental structure or power line structure corresponding to the structure type, and the template position parameters of the relative position relationship of each component of each structure. That is, the parameter information includes the three-dimensional structure parameters of each template and the position parameters of each template.

[0110] The terminal identifies the new three-dimensional structural parameters of the different components corresponding to the sub-structure model corresponding to each structure type, and the new position parameters of the relative position relationship of the different components corresponding to the sub-structure model corresponding to each structure type. Then, the terminal replaces the template three-dimensional structural parameters and template position parameters in the sample structure information corresponding to each structure type with the new three-dimensional structural parameters of the different components corresponding to the sub-structure model corresponding to each structure type, and the new position parameters of the relative position relationship of the different components corresponding to the sub-structure model corresponding to each structure type, to obtain the structure information corresponding to each sub-structure model.

[0111] Based on the substructure model corresponding to each structure type and the relative position relationship between each substructure model and the three-dimensional power distribution line model, the sample structure information corresponding to each structure type is adjusted to obtain the structure information corresponding to each substructure model. The structure information includes environmental structure information and power line structure information.

[0112] Based on the above scheme, by receiving the structure type of each substructure model manually identified, the structural information of each substructure model is analyzed, thereby improving the accuracy of the structural information corresponding to the identified substructure model.

[0113] Optionally, based on each structural information, structural feature information of each structural information of the distribution line and position feature information of each structural information are analyzed, including: for each structural information, based on the relative position relationship between the substructure model of the structural information and the three-dimensional distribution line model, the relative position information between the structural information and other structural information is identified, and the interaction information between each structural information is collected; based on the interaction information between each structural information and the relative position information between the structural information and other structural information, the position feature information of the structural information is analyzed; through an image recognition network, the structural edge information of the substructure model of the structural information at different perspectives is identified, and based on the structural edge information of each perspective, the structural perspective diagram of each perspective is identified; and the graph feature information of the structural perspective diagram of each perspective is analyzed to obtain the structural feature information of the structural information.

[0114] In this embodiment, the terminal identifies the relative position information between the structure information and other structure information based on the relative position relationship between the substructure model of the structure information and the three-dimensional distribution line model for each structure information. The relative position information is used to characterize the relative distance between the three-dimensional space of the structure information and the structure information in the straight-line distance between the structure information and other structure information.

[0115] Then, the terminal collects the interaction information between each structure information, wherein the interaction information is the correspondence between the distance information between two structure information and the degree of influence on the non-stop operation, which is the correspondence obtained by the staff based on a large amount of work experience, expert analysis, and Internet examples.

[0116] Then, the terminal analyzes the location feature information of the structural information based on the interactive impact information between each structural information and the relative position information between the structural information and other structural information. Specifically, the terminal identifies the relative position information between the structural information and other structural information, and identifies the degree of influence of the distance between the structural information and other structural information on the non-stop operation based on the correspondence between the distance information between the two structural information and the degree of influence on the non-stop operation. Then, the terminal identifies the interference factor characteristics between the structural information and other structural information based on the structural information in the straight-line distance between the structural information and other structural information. The terminal uses the degree of influence of the distance between the structural information and other structural information on the non-stop operation and the interference factor characteristics between the structural information and other structural information as the location feature information of the structural information.

[0117] The terminal uses an image recognition network to identify the structural edge information of different perspectives of the substructure model of the structural information. The perspectives include the main perspective, left and right perspectives, top perspective, bottom perspective, back perspective, etc. Then, based on the structural edge information of each perspective, the terminal identifies the structural dialysis graph of each perspective, analyzes the graph feature information of the structural dialysis graph of each perspective, and obtains the structural feature information of the structural information.

[0118] Based on the above scheme, the position feature information is determined by identifying the relative position information between the structural information and the degree of influence of the structural information on the non-stop operation, and the structural feature information of the structural information is obtained by identifying the structural features from different perspectives, thereby improving the comprehensiveness of the identified position feature information and structural feature information.

[0119] Optionally, based on the structural characteristic information of each structural information and the location characteristic information of each structural information, the initial non-stop operation characteristic information of the distribution line is identified, including: for each structural information, based on the location characteristic information of the structural information, the structural spacing information between the structural information and other structural information is identified, and based on the structural type of each structural information and the structural spacing information between the structural information and other structural information, the distance influence characteristic of the structural information is identified; the structural material characteristic information of each structural information is acquired, and based on the graph characteristic information of the structural information and the structural material characteristic information of the structural information, the structural support characteristic of the structural information and the non-stop operation influence characteristic of the structural information are identified; the distance influence characteristic of the structural information, the structural support characteristic of the structural information and the non-stop operation influence characteristic of the structural information are used as the initial non-stop operation characteristic information of the structural information.

[0120] In this embodiment, for each structure information, the terminal identifies the structure spacing information between the structure information and other structure information based on the location feature information of the structure information, and identifies the distance impact feature of the structure information based on the structure type of each structure information and the structure spacing information between the structure information and other structure information. The distance impact feature mainly refers to the degree of impact of the distance on the non-stop operation.

[0121] The terminal obtains the structural material characteristic information of each structural information, and identifies the structural support characteristics of the structural information and the non-stop operation influence characteristics of the structural information based on the characteristic information of each figure of the structural information and the structural material characteristic information of the structural information. The structural support characteristics are the load-bearing limit value of the structural information and the load-bearing distribution information of each position information of the structural information.

[0122] The terminal uses the distance influence feature of the structure information, the structure support feature of the structure information, and the non-stop operation influence feature of the structure information as the initial non-stop operation feature information of the structure information.

[0123] Based on the above solution, by identifying the characteristic information related to each structural information and the non-stop power operation, the accuracy of identifying the impact of each structural information on the non-stop power operation is improved.

[0124] Optionally, in each initial non-stop power operation characteristic information, target non-stop power operation characteristic information that meets the non-stop power operation conditions is screened, including: obtaining multiple types of non-stop power operation tasks and environmental requirement information of each non-stop power operation task, and analyzing the non-stop power operation conditions corresponding to each environmental requirement information; parsing each non-stop power operation condition to obtain the distance requirement information of each non-stop power operation task, the structural support requirement information of each non-stop power operation task, and the non-stop power operation impact limitation information of each non-stop power operation task; for each non-stop power operation task, identifying the first deviation value between the distance impact characteristic of each structural information and the distance requirement information of the non-stop power operation task, the second deviation value between the structural support characteristic of each structural information and the structural support requirement information of the non-stop power operation task, and the third deviation value between the non-stop power operation impact characteristic of each structural information and the non-stop power operation impact limitation information of the non-stop power operation task; screening the initial non-stop power operation characteristic information of the target structural information whose first deviation value is lower than the first deviation threshold, the second deviation value is lower than the second deviation threshold, and the third deviation value is lower than the third deviation threshold as the target non-stop power operation characteristic information corresponding to the non-stop power operation task.

[0125] In this embodiment, the terminal obtains multiple types of non-stop power supply operation tasks and the environmental requirement information of each non-stop power supply operation task, and analyzes the non-stop power supply operation conditions corresponding to each environmental requirement information. Then, the terminal analyzes each non-stop power supply operation condition to obtain the distance requirement information of each non-stop power supply operation task, the structural support requirement information of each non-stop power supply operation task, and the non-stop power supply operation impact limitation information of each non-stop power supply operation task.

[0126] Then, for each non-stop operation task, the terminal identifies the first deviation value between the distance impact feature of each structural information and the distance requirement information of the non-stop operation task, the second deviation value between the structural support feature of each structural information and the structural support requirement information of the non-stop operation task, and the third deviation value between the non-stop operation impact feature of each structural information and the non-stop operation impact limitation information of the non-stop operation task. The deviation value is the deviation value between the specific data values ​​obtained after digitizing each information, for example, the impact degree value on non-stop operation in the distance impact feature, the limit load-bearing value in the structural support feature, the impact range value of the non-stop operation impact feature, etc.

[0127] The terminal presets a first deviation threshold, a second deviation threshold, and a third deviation threshold, and filters the initial non-stop power operation characteristic information of the target structure information whose first deviation value is lower than the first deviation threshold, the second deviation value is lower than the second deviation threshold, and the third deviation value is lower than the third deviation threshold, as the target non-stop power operation characteristic information corresponding to the non-stop power operation task.

[0128] Based on the above scheme, by calculating each deviation value, the target non-stop power operation characteristic information corresponding to the non-stop power operation task is screened, thereby improving the efficiency of determining the structural characteristics of the structural information corresponding to each non-stop power operation task, and facilitating the planning of the operation route and other issues when executing the non-stop power operation task in the subsequent time.

[0129] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0130] Based on the same inventive concept, the embodiment of the present application also provides an image-recognition-based power line and environment intelligent identification device for implementing the image-recognition-based power line and environment intelligent identification method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more embodiments of the image-recognition-based power line and environment intelligent identification device provided below can refer to the limitations of the image-recognition-based power line and environment intelligent identification method above, and will not be repeated here.

[0131] In one embodiment, Figure 2 As shown, a device for intelligently identifying power lines and environments based on image recognition is provided, comprising: an acquisition module 210, a construction module 220, an analysis module 230 and a screening module 240, wherein:

[0132] An acquisition module 210 is used to acquire environmental image data of the environment in which the distribution line is located and power line image data of the distribution line, and to identify environmental three-dimensional structure data corresponding to the environmental image data and power line three-dimensional structure data corresponding to the power line image data through an image recognition network;

[0133] A construction module 220, configured to construct a three-dimensional distribution line model corresponding to the distribution line based on the three-dimensional structure data of the environment and the three-dimensional structure data of the power line, and split the three-dimensional distribution line model into a plurality of sub-structure models;

[0134] An analysis module 230, configured to identify the structural information corresponding to each substructure model, and analyze structural feature information of each structural information of the distribution line and position feature information of each structural information based on the structural information;

[0135] The screening module 240 is used to identify the initial non-stop operation characteristic information of the distribution line based on the structural characteristic information of each of the structural information and the position characteristic information of each of the structural information, and to screen the target non-stop operation characteristic information that meets the non-stop operation conditions among each of the initial non-stop operation characteristic information.

[0136] Optionally, the acquisition module 210 is specifically configured to:

[0137] Identifying, by an image recognition network, environmental image edge information of each of the environmental image data and power line image edge information corresponding to each of the power line image data;

[0138] Perform three-dimensional stitching on the edge information of each environmental image according to the acquisition angle of each environmental image data to obtain the edge information of the environmental three-dimensional image, and perform three-dimensional stitching on the edge information of each power line image according to the acquisition angle of each power line image data to obtain the edge information of the power line three-dimensional image;

[0139] Relative position information between edge position points in the environmental three-dimensional image edge information is identified to obtain environmental three-dimensional structure data corresponding to the environmental image data, and relative position information between edge position points in the power line three-dimensional image edge information is identified to obtain power line three-dimensional structure data corresponding to the power line image data.

[0140] Optionally, the construction module 220 is specifically used to:

[0141] Based on the relative position relationship between each edge position point in the three-dimensional structure data of the environment, each edge position point is subjected to three-dimensional connection processing to obtain a three-dimensional connection structure of the environment, and based on the relative position relationship between each edge position point in the three-dimensional structure data of the power line, each edge position point is subjected to three-dimensional connection processing to obtain a three-dimensional connection structure of the power line;

[0142] Identifying the environment structure edge connection lines of the environment three-dimensional connection structure and the power line structure edge connection lines of the power line three-dimensional connection structure;

[0143] Based on the edge lines of the environmental structure, the three-dimensional connection structure of the environment is simplified to obtain a three-dimensional structure model of the environment, and based on the edge lines of the power line structure, the three-dimensional connection structure of the power line is simplified to obtain a three-dimensional structure model of the power line;

[0144] The relative position relationship between the three-dimensional structure model of the environment and the three-dimensional structure model of the power line is identified, and based on the relative position relationship, the three-dimensional structure model of the environment and the three-dimensional structure model of the power line are subjected to model splicing processing to obtain a three-dimensional distribution line model of the distribution line.

[0145] Optionally, the construction module 220 is specifically used to:

[0146] Identifying three-dimensional structural features of the three-dimensional distribution line model, and identifying a three-dimensional position information range corresponding to each three-dimensional structural feature;

[0147] Based on the three-dimensional position information range corresponding to each three-dimensional structural feature, the three-dimensional distribution line model is split to obtain a sub-structure model corresponding to each three-dimensional structural feature.

[0148] Optionally, the analysis module 230 is specifically used for:

[0149] In response to a structure recognition operation of a user, a structure type corresponding to each substructure model is acquired, and a relative position relationship between each substructure model and the three-dimensional power distribution line model is identified;

[0150] The sample structure information corresponding to each structure type is queried, and based on the substructure model corresponding to each structure type and the relative position relationship between each substructure model and the three-dimensional distribution line model, the sample structure information corresponding to each structure type is adjusted to obtain the structure information corresponding to each substructure model; the structure information includes environmental structure information and power line structure information.

[0151] Optionally, the analysis module 230 is specifically used for:

[0152] For each piece of structural information, based on the relative position relationship between the substructure model of the structural information and the three-dimensional distribution line model, the relative position information between the structural information and other structural information is identified, and the interaction influence information between each piece of structural information is collected;

[0153] Based on the interaction information between each structure information and the relative position information between the structure information and other structure information, analyzing the position feature information of the structure information;

[0154] Identify structural edge information of different perspectives of the substructure model of the structural information through an image recognition network, and identify a structural dialysis diagram of each perspective based on the structural edge information of each perspective;

[0155] The graph feature information of the structural dialysis graph of each viewing angle is analyzed to obtain the structural feature information of the structural information.

[0156] Optionally, the screening module 240 is specifically used for:

[0157] For each piece of structural information, based on the position feature information of the structural information, structural distance information between the structural information and other structural information is identified, and based on the structural type of each piece of structural information and the structural distance information between the structural information and other structural information, a distance influence feature of the structural information is identified;

[0158] Acquire structural material feature information of each structural information, and identify structural support features of the structural information and uninterrupted operation impact features of the structural information based on feature information of each figure of the structural information and structural material feature information of the structural information;

[0159] The distance influence feature of the structural information, the structural support feature of the structural information, and the non-stop operation influence feature of the structural information are used as the initial non-stop operation feature information of the structural information.

[0160] Optionally, the screening module 240 is specifically used for:

[0161] Obtain multiple types of non-stop operation tasks and environmental requirement information of each non-stop operation task, and analyze the non-stop operation conditions corresponding to each environmental requirement information;

[0162] Analyze each non-stop operation condition to obtain the distance requirement information of each non-stop operation task, the structural support requirement information of each non-stop operation task, and the non-stop operation impact limitation information of each non-stop operation task;

[0163] For each non-stop operation task, identify a first deviation value between a distance impact feature of each structural information and the distance requirement information of the non-stop operation task, a second deviation value between a structural support feature of each structural information and the structural support requirement information of the non-stop operation task, and a third deviation value between a non-stop operation impact feature of each structural information and the non-stop operation impact limitation information of the non-stop operation task;

[0164] The initial non-stop power operation characteristic information of the target structure information whose first deviation value is lower than the first deviation threshold, the second deviation value is lower than the second deviation threshold, and the third deviation value is lower than the third deviation threshold is selected as the target non-stop power operation characteristic information corresponding to the non-stop power operation task.

[0165] Each module in the above-mentioned power line and environment intelligent identification device based on image recognition can be implemented in whole or in part by software, hardware and their combination. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0166] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 3As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a method for intelligent recognition of power lines and environments based on image recognition is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0167] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0168] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of any one of the methods in the first aspect are implemented.

[0169] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the methods in the first aspect are implemented.

[0170] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0171] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0172] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0173] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for intelligent identification of power lines and environments based on image recognition, characterized in that: The method comprises: Acquire environmental image data of the environment in which the distribution line is located and power line image data of the distribution line, and identify the environmental three-dimensional structure data corresponding to the environmental image data and the power line three-dimensional structure data corresponding to the power line image data through an image recognition network; Based on the three-dimensional structure data of the environment and the three-dimensional structure data of the power line, construct a three-dimensional distribution line model corresponding to the distribution line, and split the three-dimensional distribution line model into multiple sub-structure models; Identify the structural information corresponding to each substructure model, and analyze the structural feature information of each structural information of the distribution line and the position feature information of each structural information based on the structural information; Based on the structural feature information of each structural information and the position feature information of each structural information, identifying the initial non-stop operation feature information of the distribution line, and screening the target non-stop operation feature information that meets the non-stop operation conditions from each initial non-stop operation feature information; The identifying the initial non-stop operation characteristic information of the distribution line based on the structural characteristic information of each structural information and the position characteristic information of each structural information comprises: For each piece of structural information, based on the position feature information of the structural information, structural distance information between the structural information and other structural information is identified, and based on the structural type of each piece of structural information and the structural distance information between the structural information and other structural information, a distance influence feature of the structural information is identified; Acquire structural material feature information of each structural information, and identify structural support features of the structural information and uninterrupted operation impact features of the structural information based on feature information of each figure of the structural information and structural material feature information of the structural information; Using the distance influence feature of the structural information, the structural support feature of the structural information, and the non-stop operation influence feature of the structural information as initial non-stop operation feature information of the structural information; The step of screening target non-stop power operation characteristic information that meets the non-stop power operation condition from each of the initial non-stop power operation characteristic information includes: Obtain multiple types of non-stop operation tasks and environmental requirement information of each non-stop operation task, and analyze the non-stop operation conditions corresponding to each environmental requirement information; Analyze each non-stop operation condition to obtain the distance requirement information of each non-stop operation task, the structural support requirement information of each non-stop operation task, and the non-stop operation impact limitation information of each non-stop operation task; For each non-stop operation task, identify a first deviation value between a distance impact feature of each structural information and the distance requirement information of the non-stop operation task, a second deviation value between a structural support feature of each structural information and the structural support requirement information of the non-stop operation task, and a third deviation value between a non-stop operation impact feature of each structural information and the non-stop operation impact limitation information of the non-stop operation task; The initial non-stop power operation characteristic information of the target structure information whose first deviation value is lower than the first deviation threshold, the second deviation value is lower than the second deviation threshold, and the third deviation value is lower than the third deviation threshold is selected as the target non-stop power operation characteristic information corresponding to the non-stop power operation task.

2. The method according to claim 1, characterized in that The identifying, by means of an image recognition network, the environmental three-dimensional structure data corresponding to the environmental image data and the power line three-dimensional structure data corresponding to the power line image data comprises: Identifying, by an image recognition network, environmental image edge information of each of the environmental image data and power line image edge information corresponding to each of the power line image data; Perform three-dimensional stitching on the edge information of each environmental image according to the acquisition angle of each environmental image data to obtain the edge information of the environmental three-dimensional image, and perform three-dimensional stitching on the edge information of each power line image according to the acquisition angle of each power line image data to obtain the edge information of the power line three-dimensional image; Relative position information between edge position points in the environmental three-dimensional image edge information is identified to obtain environmental three-dimensional structure data corresponding to the environmental image data, and relative position information between edge position points in the power line three-dimensional image edge information is identified to obtain power line three-dimensional structure data corresponding to the power line image data.

3. The method according to claim 2, characterized in that The constructing of a three-dimensional distribution line model corresponding to the distribution line based on the three-dimensional structure data of the environment and the three-dimensional structure data of the power line includes: Based on the relative position relationship between each edge position point in the three-dimensional structure data of the environment, each edge position point is subjected to three-dimensional connection processing to obtain a three-dimensional connection structure of the environment, and based on the relative position relationship between each edge position point in the three-dimensional structure data of the power line, each edge position point is subjected to three-dimensional connection processing to obtain a three-dimensional connection structure of the power line; Identifying the environment structure edge connection lines of the environment three-dimensional connection structure and the power line structure edge connection lines of the power line three-dimensional connection structure; Based on the edge lines of the environmental structure, the three-dimensional connection structure of the environment is simplified to obtain a three-dimensional structure model of the environment, and based on the edge lines of the power line structure, the three-dimensional connection structure of the power line is simplified to obtain a three-dimensional structure model of the power line; The relative position relationship between the three-dimensional structure model of the environment and the three-dimensional structure model of the power line is identified, and based on the relative position relationship, the three-dimensional structure model of the environment and the three-dimensional structure model of the power line are subjected to model splicing processing to obtain a three-dimensional distribution line model of the distribution line.

4. The method according to claim 2, characterized in that: The step of splitting the three-dimensional power distribution line model into a plurality of substructure models comprises: Identifying three-dimensional structural features of the three-dimensional distribution line model, and identifying a three-dimensional position information range corresponding to each three-dimensional structural feature; Based on the three-dimensional position information range corresponding to each three-dimensional structural feature, the three-dimensional distribution line model is split to obtain a sub-structure model corresponding to each three-dimensional structural feature.

5. The method according to claim 1, characterized in that The identifying structure information corresponding to each substructure model includes: In response to a structure recognition operation of a user, a structure type corresponding to each substructure model is acquired, and a relative position relationship between each substructure model and the three-dimensional power distribution line model is identified; The sample structure information corresponding to each structure type is queried, and based on the substructure model corresponding to each structure type and the relative position relationship between each substructure model and the three-dimensional distribution line model, the sample structure information corresponding to each structure type is adjusted to obtain the structure information corresponding to each substructure model; the structure information includes environmental structure information and power line structure information.

6. The method according to claim 5, characterized in that The step of analyzing the structural feature information of each structural information of the power distribution line and the position feature information of each structural information based on each structural information includes: For each piece of structural information, based on the relative position relationship between the substructure model of the structural information and the three-dimensional distribution line model, the relative position information between the structural information and other structural information is identified, and the interaction influence information between each piece of structural information is collected; Based on the interaction information between each structure information and the relative position information between the structure information and other structure information, analyzing the position feature information of the structure information; Identify structural edge information of different perspectives of the substructure model of the structural information through an image recognition network, and identify a structural dialysis diagram of each perspective based on the structural edge information of each perspective; The graph feature information of the structural dialysis graph of each viewing angle is analyzed to obtain the structural feature information of the structural information.

7. A device for intelligently identifying power lines and environments based on image recognition, characterized in that: The device comprises: An acquisition module, used to acquire environmental image data of the environment in which the distribution line is located and power line image data of the distribution line, and to identify environmental three-dimensional structure data corresponding to the environmental image data and power line three-dimensional structure data corresponding to the power line image data through an image recognition network; A construction module, configured to construct a three-dimensional distribution line model corresponding to the distribution line based on the three-dimensional structure data of the environment and the three-dimensional structure data of the power line, and split the three-dimensional distribution line model into a plurality of sub-structure models; An analysis module, used for identifying the structural information corresponding to each substructure model, and analyzing the structural feature information of each structural information of the distribution line and the position feature information of each structural information based on the structural information; A screening module, for identifying the initial non-stop operation characteristic information of the distribution line based on the structural characteristic information of each structural information and the position characteristic information of each structural information, and screening the target non-stop operation characteristic information that meets the non-stop operation conditions from each initial non-stop operation characteristic information; The identifying the initial non-stop operation characteristic information of the distribution line based on the structural characteristic information of each structural information and the position characteristic information of each structural information comprises: For each piece of structural information, based on the position feature information of the structural information, structural distance information between the structural information and other structural information is identified, and based on the structural type of each piece of structural information and the structural distance information between the structural information and other structural information, a distance influence feature of the structural information is identified; Acquire structural material feature information of each structural information, and identify structural support features of the structural information and uninterrupted operation impact features of the structural information based on feature information of each figure of the structural information and structural material feature information of the structural information; Using the distance influence feature of the structural information, the structural support feature of the structural information, and the non-stop operation influence feature of the structural information as initial non-stop operation feature information of the structural information; The step of screening target non-stop power operation characteristic information that meets the non-stop power operation condition from each of the initial non-stop power operation characteristic information includes: Obtain multiple types of non-stop operation tasks and environmental requirement information of each non-stop operation task, and analyze the non-stop operation conditions corresponding to each environmental requirement information; Analyze each non-stop operation condition to obtain the distance requirement information of each non-stop operation task, the structural support requirement information of each non-stop operation task, and the non-stop operation impact limitation information of each non-stop operation task; For each non-stop operation task, identify a first deviation value between a distance impact feature of each structural information and the distance requirement information of the non-stop operation task, a second deviation value between a structural support feature of each structural information and the structural support requirement information of the non-stop operation task, and a third deviation value between a non-stop operation impact feature of each structural information and the non-stop operation impact limitation information of the non-stop operation task; The initial non-stop power operation characteristic information of the target structure information whose first deviation value is lower than the first deviation threshold, the second deviation value is lower than the second deviation threshold, and the third deviation value is lower than the third deviation threshold is selected as the target non-stop power operation characteristic information corresponding to the non-stop power operation task.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

Images