Pole tower live-action three-dimensional model reconstruction method and system

Abstract

The invention discloses a tower live-action three-dimensional model reconstruction method and system. The method comprises the following steps: acquiring three-dimensional data of a target area; preprocessing the three-dimensional data of the target area to obtain preprocessed three-dimensional data; performing power cloud processing on the preprocessed three-dimensional data to obtain a power point cloud data set; segmenting the power point cloud data set to obtain segmented three-dimensional data; performing coordinate conversion on the segmented three-dimensional data, and performing correction after coordinate conversion to obtain corrected three-dimensional data; based on the corrected three-dimensional data, reconstructing the preliminary tower live-action three-dimensional model; on the basis of the preliminary tower live-action three-dimensional model, carrying out fine reconstruction to obtain a final tower live-action three-dimensional model; according to the method, point cloud reverse modeling can be realized, the method has the advantages of rapidness and simplicity in operation, no vulnerability of the model, high precision and high spatial resolution, and the purpose of high-precision live-action three-dimensional modeling of the tower pole is realized.

Description

Technical Field

[0001] The invention relates to a method and system for reconstructing a real-scene three-dimensional model of a pole tower, and belongs to the technical field of three-dimensional model reconstruction. Background Art

[0002] The main form of real-life 3D is oblique photography 3D modeling. This measurement technology obtains high-resolution images from five different angles, and then constructs a high-precision, high-spatial-resolution city-level 3D real-life model based on the obtained high-resolution images. However, this method relies on external data acquisition, has high requirements for drone equipment, and obtains a large amount of data, and a long model reconstruction cycle. In addition, for terrain objects with low reflectivity, the modeling effect is poor and further modeling processing is required, which leads to an increase in time and labor costs. The terrain-level 3D real-life model can not only use existing spatial data bodies (such as geographic scene DEM (digital elevation model) data and DOM (digital orthophoto) data) as a data source, but also quickly and conveniently generate a 3D real-life model that meets the accuracy requirements. Therefore, the terrain-level 3D real-life model is gradually being applied to the macro-planning of various industries.

[0003] The current method for reconstructing the real-scene three-dimensional model of the tower pole has complicated steps and is prone to many loopholes, making the model incomplete and the application effect poor. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a method and system for reconstructing a real-life three-dimensional model of a tower. The method can realize point cloud reverse modeling through preprocessing of three-dimensional data (point cloud data), power point cloud processing, segmentation and correction processing. The method has the advantages of fast and simple operation, no model loopholes, high accuracy and high spatial resolution, thereby achieving the purpose of high-precision real-life three-dimensional modeling of towers.

[0005] To achieve the above object, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a method for reconstructing a real-scene three-dimensional model of a tower, comprising: Acquire three-dimensional data of the target area; Preprocessing the three-dimensional data of the target area to obtain preprocessed three-dimensional data; The pre-processed three-dimensional data is subjected to electric power cloud processing to obtain an electric power point cloud data set; Segment the electric power point cloud data set to obtain segmented three-dimensional data; Performing coordinate transformation on the segmented three-dimensional data, and performing correction after the coordinate transformation to obtain corrected three-dimensional data; Based on the corrected 3D data, the preliminary tower real-scene 3D model is reconstructed; Based on the preliminary real-scene 3D model of the tower, a detailed reconstruction is performed to obtain the final real-scene 3D model of the tower.

[0006] Optionally, preprocessing the three-dimensional data of the target area to obtain preprocessed three-dimensional data includes: Based on the SOR algorithm, the three-dimensional data of the target area is first filtered to preliminarily filter out the abnormal values ​​in the three-dimensional data of the target area to obtain the three-dimensional data after the first filtering; The three-dimensional data after the first filtering is filtered for the second time based on the RANSAC algorithm to obtain the preprocessed three-dimensional data.

[0007] Optionally, the pre-processed three-dimensional data is subjected to electric power cloud processing to obtain an electric power point cloud data set, including: Create a new project in the point cloud processing software and upload the pre-processed 3D data to the project; Determine whether there is a pole tower coordinate file in the pre-processed three-dimensional data. If so, import the pole tower coordinates. If not, set the tower and set the position of the pole tower. Crop the preprocessed 3D data to remove areas in the target area that are not related to the tower; The trimmed 3D data is preliminarily classified, and the historical point cloud data is used for auxiliary classification to obtain the preliminary classification results; Based on the preliminary classification results, manual interactive editing is performed to correct the misclassified parts and obtain the final classification results, that is, the three-dimensional data after the final classification; Based on the final classified 3D data, vector wires are generated and edited, and working wires are generated accordingly; According to the generated working condition conductor, the dangerous points are identified and the identified dangers are marked; Based on the above processing steps, an electric power point cloud data set is generated based on point cloud processing software, and the electric power point cloud data set includes: three-dimensional data after final classification and dangerous point data statistics.

[0008] Optionally, the segmenting of the electric power point cloud data set to obtain segmented three-dimensional data includes: Based on the dangerous points in the electric power point cloud data set, the electric power point cloud data set is segmented for the first time according to the dangerous points, and the electric power point cloud data set segmented for the first time is extracted to obtain a safe data set and a dangerous data set; Conduct security monitoring on dangerous data sets and remove garbled, mixed-code data and virus data from dangerous data sets; The dangerous data set and the safe data set after safety monitoring are used to construct a secondary feature matrix to obtain constructed data; The constructed data is segmented for the second time based on the final classification result to obtain multiple sub-data files with different categories.

[0009] Optionally, the step of performing coordinate transformation on the segmented three-dimensional data includes: Get the source coordinate data in multiple sub-data files; Using the plane four-parameter conversion model, the source coordinate data is converted into the target plane coordinate data; The three-dimensional data after coordinate conversion is obtained based on the target plane coordinate data.

[0010] Optionally, performing correction after coordinate conversion to obtain corrected three-dimensional data includes: Acquire relevant data between each data node and corresponding peripheral nodes in the three-dimensional data after coordinate conversion, wherein the peripheral nodes are adjacent nodes of the data node; Based on the relevant data between each data node and the corresponding surrounding nodes, determine the number of correction layers, the weight and offset value of each surrounding node in each correction layer; Based on the weights and offset values ​​of each peripheral node in the correction layer, the three-dimensional data is corrected to obtain corrected three-dimensional data.

[0011] Optionally, the reconstructing the preliminary tower real-scene three-dimensional model based on the corrected three-dimensional data includes: Generate vector conductors representing the transmission line based on the corrected three-dimensional data; On the basis of the vector conductor, confirm the connection relationship between the insulator and the vector conductor, and edit the drainage line of the insulator based on the connection relationship; Generate a preliminary tower vector model based on the corrected 3D data; The vector conductor is split and combined with the preliminary tower vector model and the edited insulator drainage line to perform vector modeling to obtain a preliminary tower real-life 3D model.

[0012] Optionally, the detailed reconstruction based on the preliminary pole tower real scene 3D model to obtain the final pole tower real scene 3D model includes: Obtain data of different tower types, and confirm the type of target tower based on the data of different tower types; Using the existing model library to edit multiple models in the preliminary pole tower real scene 3D model, and define connection points to generate the pole tower real scene 3D model in DML format, wherein the connection points are the connection positions between the various models; The pole tower 3D model, insulator 3D model, ground wire 3D model, spacer rod 3D model and 3D models of related components in the pole tower real scene 3D model in DML format are edited, and the final pole tower real scene 3D model in OBJ format is generated after format conversion.

[0013] In a second aspect, the present invention provides a tower real-scene three-dimensional model reconstruction system, comprising: A data acquisition module, used to acquire three-dimensional data of a target area; A preprocessing module, used for preprocessing the three-dimensional data of the target area to obtain preprocessed three-dimensional data; The power point cloud processing module is used to perform power point cloud processing on the pre-processed three-dimensional data to obtain a power point cloud data set; A segmentation module is used to segment the power point cloud data set to obtain segmented three-dimensional data; A conversion and correction module is used to perform coordinate conversion on the segmented three-dimensional data and perform correction after the coordinate conversion to obtain corrected three-dimensional data; The reconstruction module is used to reconstruct the preliminary pole tower real scene 3D model based on the corrected 3D data; and is used to perform fine reconstruction on the basis of the preliminary pole tower real scene 3D model to obtain the final pole tower real scene 3D model.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The method for reconstructing the real-scene three-dimensional model of the tower proposed in the present invention realizes point cloud reverse modeling through preprocessing of three-dimensional data (point cloud data), power point cloud processing, segmentation and correction processing. It has the advantages of fast and simple operation, no model loopholes, high accuracy and high spatial resolution, and achieves the purpose of high-precision real-scene three-dimensional modeling of the tower.

[0015] The preprocessing process in the tower real-scene three-dimensional model reconstruction method proposed in the present invention can perform noise reduction and filtering processing on the three-dimensional data, thereby improving the accuracy of subsequent data analysis.

[0016] The segmentation process in the tower real-scene 3D model reconstruction method proposed in the present invention can segment huge data, and through data risk assessment before segmenting the data, it can effectively avoid the mixing of garbled codes, mixed codes and virus data, thereby improving the security of data segmentation.

[0017] The correction process in the tower real-scene three-dimensional model reconstruction method proposed in the present invention can make the corrected three-dimensional data closer to the data that should be measured, thereby reducing errors in subsequent application processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative labor, among which: Figure 1 The flowchart of the method for reconstructing the real-scene three-dimensional model of the tower in the embodiment of the present invention is shown; Figure 2 The power point cloud processing flow chart in the embodiment of the present invention is shown; Figure 3 The flowchart of the segmentation process in the embodiment of the present invention is shown; Figure 4 The flowchart of the reconstruction of the preliminary tower real-scene 3D model in the embodiment of the present invention is shown; Figure 5 The figure shows a flowchart of reconstructing the final real-scene 3D model of the tower in an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0020] The application principle of the present invention is described in detail below in conjunction with the accompanying drawings. Example 1

[0021] like Figure 1 As shown, an embodiment of the present invention provides a method for reconstructing a real-scene three-dimensional model of a tower, comprising the following steps: S01: Acquire three-dimensional data of the target area; S02: preprocessing the three-dimensional data of the target area to obtain preprocessed three-dimensional data; S03: Performing electric power cloud processing on the pre-processed three-dimensional data to obtain an electric power point cloud data set; S04: Segmenting the electric power point cloud data set to obtain segmented three-dimensional data; S05: performing coordinate transformation on the segmented three-dimensional data, and performing correction after the coordinate transformation to obtain corrected three-dimensional data; S06: Reconstructing the preliminary tower real-scene 3D model based on the corrected 3D data; S07: Based on the preliminary pole tower real scene 3D model, perform detailed reconstruction to obtain the final pole tower real scene 3D model.

[0022] Specifically, both step S06 and step S07 use Metasequoia3D metasequoia modeling software to reconstruct the model.

[0023] The method for reconstructing the real-scene three-dimensional model of the tower proposed in the present invention realizes point cloud reverse modeling through preprocessing of three-dimensional data (point cloud data), power point cloud processing, segmentation and correction processing. It has the advantages of fast and simple operation, no model loopholes, high accuracy and high spatial resolution, and achieves the purpose of high-precision real-scene three-dimensional modeling of the tower.

[0024] In this embodiment, the step S01 of acquiring the three-dimensional data of the target area specifically includes: Aerial photography drones are used to obtain three-dimensional data of the target area multiple times, and the aerial photography drones are equipped with laser radar (LiDAR) cameras.

[0025] In this embodiment, the three-dimensional data of the target area is preprocessed in step S02 to obtain preprocessed three-dimensional data, which specifically includes: The first filtering process is performed on the three-dimensional data of the target area based on the SOR (Statistical Outline Remove) algorithm to preliminarily filter out the abnormal values ​​in the three-dimensional data of the target area to obtain the three-dimensional data after the first filtering; The three-dimensional data after the first filtering is filtered for the second time based on the RANSAC (Random Sample Consensus) algorithm to obtain the preprocessed three-dimensional data.

[0026] Specifically, the SOR algorithm can effectively identify and remove points that are significantly different from the main body of the data set. These outliers may be caused by measurement errors, equipment failures, or other external factors. By removing these outliers, not only can the accuracy and stability of subsequent processing steps be improved, but the amount of calculations in subsequent processing steps can also be significantly reduced, the overall processing efficiency can be improved, and the robustness and reliability of the model can be improved. The RANSAC algorithm can further identify and remove data points that do not meet the model assumptions through random sampling and consensus set estimation. These data points may include outliers that are not completely removed by the SOR algorithm, as well as some outliers that are difficult to identify through simple statistical methods, thereby improving the accuracy of subsequent model reconstruction; In summary, combining the SOR algorithm and the RANSAC algorithm to filter the 3D data can fully utilize the advantages of the two algorithms to achieve efficient, accurate and robust processing of the 3D data, which provides strong support for subsequent 3D modeling and analysis.

[0027] like Figure 2 As shown, in this embodiment, in step S03, the pre-processed three-dimensional data is subjected to power cloud processing to obtain a power point cloud data set, which specifically includes: Create a project and upload files: create a new project in the point cloud processing software and upload the pre-processed 3D data to the project; specifically, Pointscene point cloud processing software is used in this embodiment; Tower position acquisition: determine whether there is a tower coordinate file in the pre-processed 3D data. If so, import the tower coordinates. If not, set the tower and set the tower position. Point cloud clipping: clip the pre-processed 3D data to remove the areas irrelevant to the tower in the target area, reduce the data volume and improve processing efficiency; Preliminary classification: The clipped 3D data is preliminarily classified (automatically classified), and the historical point cloud data is used for auxiliary classification to obtain the preliminary classification results; among them, the preliminary classification distinguishes different types of land objects such as buildings, vegetation, insulators, etc., reducing the workload of manual editing Type interaction: Based on the preliminary classification results, manual interactive editing is performed to correct the misclassified parts and obtain the final classification results, that is, the final classified 3D data; Generate working condition conductor: Generate and edit vector conductors based on the final classified 3D data, and generate working condition conductors, i.e. conductor models that reflect actual working conditions; Danger point identification: According to the generated working condition wire, the dangerous point is identified and the identified danger is marked (danger point mapping); Specifically, the identification of dangerous points includes: Identification of line-to-ground distance danger: First determine the type of ground object at the two-dimensional ground object point to be inspected, and then calculate the plane distance between the two-dimensional power line point to be inspected of the working conductor and the two-dimensional ground object point to be inspected, which is the line-to-ground distance. If the line-to-ground distance is less than the line-to-ground safety distance threshold, the partition and the corresponding power line phase are also considered to be dangerous points.

[0028] Generate a power point cloud data set: Based on the above processing steps, generate a power point cloud data set based on the point cloud processing software, the power point cloud data set includes: the three-dimensional data after the final classification and the dangerous point data statistics. Specifically, a report can also be directly generated.

[0029] like Figure 3 As shown, in this embodiment, in step S04, the power point cloud data set is segmented to obtain segmented three-dimensional data, which specifically includes: S031: based on the dangerous points in the electric power point cloud data set, the electric power point cloud data set is segmented for the first time according to the dangerous points, and the electric power point cloud data set segmented for the first time is extracted to obtain a safe data set and a dangerous data set; S032: Perform security monitoring on the dangerous data set and remove garbled, mixed code data and virus data in the dangerous data set; S033: construct a secondary feature matrix for the dangerous data set and the safe data set after safety monitoring to obtain constructed data; S034: The constructed data is segmented for the second time according to the data category based on the final classification result to obtain a plurality of sub-data files with different categories.

[0030] The three-dimensional data segmentation process adopted by the present invention can segment and process huge data, segment and extract the data based on the dangerous points, effectively avoid the mixing of garbled codes, mixed codes and virus data, and improve the security of data segmentation.

[0031] In this embodiment, in step S05, coordinate transformation is performed on the segmented three-dimensional data, and correction is performed after the coordinate transformation to obtain the corrected three-dimensional data, which specifically includes: (1) Coordinate transformation Get the source coordinate data in multiple sub-data files; The source coordinate data is converted into the target plane coordinate data using the plane four-parameter conversion model; the plane four-parameter conversion model is as follows:

[0032] Wherein, A is the source coordinate system of the coordinate data in multiple sub-data files; B is the target plane coordinate system; x and y are coordinate parameters; k is the scale factor for converting the A coordinate system to the B coordinate system; α is the rotation angle parameter for converting the A coordinate system to the B coordinate system; x0 and y0 represent the translation amounts in the x-axis and y-axis directions respectively.

[0033] Obtaining three-dimensional data after coordinate conversion based on target plane coordinate data; (2) Correction Acquire relevant data between each data node and corresponding peripheral nodes in the three-dimensional data after coordinate conversion, wherein the peripheral nodes are adjacent nodes of the data node; Based on the relevant data between each data node and the corresponding surrounding nodes, determine the number of correction layers, the weight and offset value of each surrounding node in each correction layer; Based on the weights and offset values ​​of each peripheral node in the correction layer, the three-dimensional data is corrected to obtain corrected three-dimensional data.

[0034] like Figure 4 As shown, in this embodiment, in step S06, the preliminary tower real scene 3D model is reconstructed based on the corrected 3D data, which specifically includes: Generate vector conductors: Generate vector conductors representing transmission lines based on the corrected 3D data; Insulator drainage line editing: Based on the vector conductor, confirm the connection relationship between the insulator and the vector conductor, and edit the insulator drainage line based on the connection relationship; Generate a preliminary pole tower vector model: Generate a preliminary pole tower vector model based on the corrected three-dimensional data; Conductor splitting - vector modeling: Split the vector conductor and combine it with the preliminary tower vector model and the edited insulator drainage line to perform vector modeling to obtain a preliminary tower real-life 3D model in OBJ format.

[0035] like Figure 5 As shown, in this embodiment, step S07 performs fine reconstruction on the basis of the preliminary pole tower real scene 3D model to obtain the final pole tower real scene 3D model, which specifically includes: Tower type statistics: obtain data of different tower types, and confirm the type of target tower based on the data of different tower types; Model library model and connection point editing: Use the existing model library to edit multiple models in the preliminary tower real-scene 3D model, and define connection points to generate a tower real-scene 3D model in DML format. The connection points are the connection positions between the various models. 3D editing processing: edit the 3D tower model, insulator model, ground wire model, spacer rod model and 3D models of related components in the 3D tower real-life model in DML format, and generate the final 3D tower real-life model in OBJ format after format conversion. Example 2

[0036] The embodiment of the present invention provides a tower real-scene three-dimensional model reconstruction system, comprising: A data acquisition module, used to acquire three-dimensional data of a target area; A preprocessing module, used for preprocessing the three-dimensional data of the target area to obtain preprocessed three-dimensional data; The power point cloud processing module is used to perform power point cloud processing on the pre-processed three-dimensional data to obtain a power point cloud data set; A segmentation module is used to segment the power point cloud data set to obtain segmented three-dimensional data; A conversion and correction module is used to perform coordinate conversion on the segmented three-dimensional data and perform correction after the coordinate conversion to obtain corrected three-dimensional data; The reconstruction module is used to reconstruct the preliminary pole tower real scene 3D model based on the corrected 3D data; and is used to perform fine reconstruction on the basis of the preliminary pole tower real scene 3D model to obtain the final pole tower real scene 3D model. Example 3

[0037] An embodiment of the present invention provides a computer-readable storage medium.

[0038] A computer-readable storage medium stores a program thereon, which, when executed by a processor, implements the steps in a method for reconstructing a real-scene three-dimensional model of a tower as described in Example 1 of the present disclosure.

[0039] The detailed steps are the same as those of the method for reconstructing a real-scene three-dimensional model of a tower provided in Example 1, and will not be repeated here.

[0040] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A method for reconstructing a real-scene three-dimensional model of a tower, characterized in that: include: Acquire three-dimensional data of the target area; Preprocessing the three-dimensional data of the target area to obtain preprocessed three-dimensional data; The pre-processed three-dimensional data is subjected to electric power cloud processing to obtain an electric power point cloud data set; Segment the electric power point cloud data set to obtain segmented three-dimensional data; Performing coordinate transformation on the segmented three-dimensional data, and performing correction after the coordinate transformation to obtain corrected three-dimensional data; Based on the corrected 3D data, the preliminary tower real-scene 3D model is reconstructed; Based on the preliminary real-scene 3D model of the tower, a detailed reconstruction is performed to obtain the final real-scene 3D model of the tower.

2. The method for reconstructing a real-scene three-dimensional model of a tower according to claim 1, characterized in that: Preprocessing the three-dimensional data of the target area to obtain preprocessed three-dimensional data includes: Based on the SOR algorithm, the three-dimensional data of the target area is first filtered to preliminarily filter out the abnormal values ​​in the three-dimensional data of the target area to obtain the three-dimensional data after the first filtering; The three-dimensional data after the first filtering is filtered for the second time based on the RANSAC algorithm to obtain the preprocessed three-dimensional data.

3. The method for reconstructing a real-scene three-dimensional model of a tower according to claim 2, characterized in that: The pre-processed 3D data is processed into electric power cloud data set to obtain electric power point cloud data set, including: Create a new project in the point cloud processing software and upload the pre-processed 3D data to the project; Determine whether there is a pole tower coordinate file in the pre-processed three-dimensional data. If so, import the pole tower coordinates. If not, set the tower and set the position of the pole tower. Crop the preprocessed 3D data to remove areas in the target area that are not related to the tower; The trimmed 3D data is preliminarily classified, and the historical point cloud data is used for auxiliary classification to obtain the preliminary classification results; Based on the preliminary classification results, manual interactive editing is performed to correct the misclassified parts and obtain the final classification results, that is, the three-dimensional data after the final classification; Based on the final classified three-dimensional data, vector wires are generated and edited, and working condition wires are generated accordingly.

4. According to the generated working condition wire, identify the dangerous points and mark the identified dangers; Based on the above processing steps, an electric power point cloud data set is generated based on the point cloud processing software, and the electric power point cloud data set includes: Statistics of the final classified three-dimensional data and dangerous point data.

5. The method for reconstructing a real-scene three-dimensional model of a tower according to claim 3, characterized in that: The electric power point cloud data set is segmented to obtain segmented three-dimensional data, including: Based on the dangerous points in the electric power point cloud data set, the electric power point cloud data set is segmented for the first time according to the dangerous points, and the electric power point cloud data set segmented for the first time is extracted to obtain a safe data set and a dangerous data set; Conduct security monitoring on dangerous data sets and remove garbled, mixed-code data and virus data from dangerous data sets; The dangerous data set and the safe data set after safety monitoring are used to construct a secondary feature matrix to obtain constructed data; The constructed data is segmented for the second time based on the final classification result to obtain multiple sub-data files with different categories.

6. The method for reconstructing a real-scene three-dimensional model of a tower according to claim 4, characterized in that: The step of performing coordinate transformation on the segmented three-dimensional data comprises: Get the source coordinate data in multiple sub-data files; Using the plane four-parameter conversion model, the source coordinate data is converted into the target plane coordinate data; Obtaining three-dimensional data after coordinate conversion based on target plane coordinate data; The method for reconstructing a real-scene three-dimensional model of a tower according to claim 5, characterized in that the correction is performed after the coordinate conversion to obtain the corrected three-dimensional data, comprising: Acquire relevant data between each data node and corresponding peripheral nodes in the three-dimensional data after coordinate conversion, wherein the peripheral nodes are adjacent nodes of the data node; Based on the relevant data between each data node and the corresponding surrounding nodes, determine the number of correction layers, the weight and offset value of each surrounding node in each correction layer; Based on the weights and offset values ​​of each peripheral node in the correction layer, the three-dimensional data is corrected to obtain corrected three-dimensional data.

7. The method for reconstructing a real-scene three-dimensional model of a tower according to claim 6, characterized in that: The reconstruction of the preliminary tower real-scene three-dimensional model based on the corrected three-dimensional data includes: Generate vector conductors representing the transmission line based on the corrected three-dimensional data; On the basis of the vector conductor, confirm the connection relationship between the insulator and the vector conductor, and edit the drainage line of the insulator based on the connection relationship; Generate a preliminary tower vector model based on the corrected 3D data; The vector conductor is split and combined with the preliminary tower vector model and the edited insulator drainage line to perform vector modeling to obtain a preliminary tower real-life 3D model.

8. The method for reconstructing a real-scene three-dimensional model of a tower according to claim 7, characterized in that: The method of reconstructing the final tower real scene 3D model based on the preliminary tower real scene 3D model includes: Obtain data of different tower types, and confirm the type of target tower based on the data of different tower types; Using the existing model library to edit multiple models in the preliminary pole tower real scene 3D model, and define connection points to generate the pole tower real scene 3D model in DML format, wherein the connection points are the connection positions between the various models; The pole tower 3D model, insulator 3D model, ground wire 3D model, spacer rod 3D model and 3D models of related components in the pole tower real scene 3D model in DML format are edited, and the final pole tower real scene 3D model in OBJ format is generated after format conversion.

9. A tower real-scene 3D model reconstruction system, characterized in that: include: A data acquisition module, used to acquire three-dimensional data of a target area; A preprocessing module, used for preprocessing the three-dimensional data of the target area to obtain preprocessed three-dimensional data; The power point cloud processing module is used to perform power point cloud processing on the pre-processed three-dimensional data to obtain a power point cloud data set; A segmentation module is used to segment the power point cloud data set to obtain segmented three-dimensional data; A conversion and correction module is used to perform coordinate conversion on the segmented three-dimensional data and perform correction after the coordinate conversion to obtain corrected three-dimensional data; A reconstruction module is used to reconstruct the preliminary tower real-scene 3D model based on the corrected 3D data; It is used to perform fine reconstruction on the basis of the preliminary real-scene 3D model of the tower to obtain the final real-scene 3D model of the tower.

Images