Substation three-dimensional modeling method based on laser point cloud and visible light fusion

Abstract

The invention discloses a transformer substation three-dimensional modeling method based on laser point cloud and visible light fusion, and the method comprises the steps: carrying out the scanning and shooting of a transformer substation through employing an unmanned plane, carrying a laser radar and a visible light camera, and obtaining laser point cloud data and visible light image data correspondingly; acquiring laser point cloud data and visible light image data; the obtained data is preprocessed; constructing a three-dimensional curved surface model based on the preprocessed data; executing texture mapping of the three-dimensional curved surface model to generate a transformer substation three-dimensional model; according to the three-dimensional modeling method for the transformer substation, high-precision, high-efficiency and high-reality digital reconstruction of the transformer substation is realized by combining an advanced data acquisition technology and a processing algorithm. According to the method, the modeling efficiency and precision are improved, powerful technical support is provided for intelligent management and optimization of the transformer substation, the problems that in the prior art, the modeling workload is large, the modeling efficiency is low are solved, and the modeling efficiency is improved.

Description

Technical Field

[0001] The invention relates to a safety inspection of a power transmission line, and in particular to a three-dimensional modeling method for a substation based on the fusion of laser point cloud and visible light. Background Art

[0002] Substations are the main components of the power system, and the effectiveness of their daily inspection and maintenance is closely related to the safe operation of the power system. The 3D model of a substation can intuitively and vividly express the 3D spatial position of the equipment and facilities in the station and the internal structure of the equipment. Therefore, it is increasingly widely used in 3D design of substations, intelligent operation and inspection and control of substations, and simulation of substations.

[0003] For example, a 500kV substation covers an area of ​​about 100,000 square meters, including main transformer area, distribution device area, control building and other main functional areas. The substation has a complex equipment layout, including multiple main transformers, circuit breakers, disconnectors, lightning arresters and other equipment. The three-dimensional model in the existing technology is used for equipment management and safety assessment purposes, and its modeling process is often too complicated, consuming more computing resources and time. Summary of the invention

[0004] The purpose of the present invention is to provide a three-dimensional modeling method for a substation based on the fusion of laser point cloud and visible light, which solves the above-mentioned technical problems pointed out in the prior art.

[0005] The present invention provides a three-dimensional modeling method for a substation based on the fusion of laser point cloud and visible light, comprising the following steps:

[0006] Use drones equipped with laser radar and visible light cameras to scan and photograph the substation, and obtain laser point cloud data and visible light image data respectively;

[0007] Acquire laser point cloud data and visible light image data;

[0008] Preprocess the acquired data;

[0009] Constructing a three-dimensional surface model based on the preprocessed data;

[0010] Perform texture mapping of the 3D surface model to generate a 3D model of the substation.

[0011] Preferably, a drone equipped with a laser radar and a visible light camera is used to scan and photograph the substation, and the corresponding laser point cloud data and visible light image data include:

[0012] Plan a preset drone flight path and set multiple scanning sites;

[0013] Execute the current drone flight path and perform a 360-degree rotation scan at each scanning station;

[0014] Perform multi-angle oblique photography of the target device;

[0015] A high-precision GNSS / INS system is used to record the precise position coordinates of each scan and shot. The gyroscope and accelerometer are used to record the pitch angle, yaw angle and roll angle of each scan and shot. The position coordinates and attitude angle information are associated with the corresponding point cloud data and image data to form an information mapping table.

[0016] Preferably, preprocessing the acquired data includes:

[0017] De-noising the laser point cloud data;

[0018] Perform color correction and image enhancement on visible light image data;

[0019] Extract feature points from visible light image data and feature points from laser point cloud data;

[0020] Obtain the formation information mapping table and align the feature points of the laser point cloud data with the visible light image data.

[0021] Preferably, constructing a three-dimensional surface model based on the preprocessed data includes:

[0022] Segment the registered point cloud data and identify different target devices;

[0023] Using the RANSAC algorithm to fit the basic geometric shape of the target device (such as plane, cylinder, etc.);

[0024] Building a wireframe model based on the basic geometric shape of the target device;

[0025] Generate a 3D surface model based on the wireframe model.

[0026] Preferably, a three-dimensional surface model is generated based on the wireframe model, specifically including: converting the wireframe model into a triangular mesh through a triangulation algorithm; using a B-spline interpolation method to generate a smooth three-dimensional surface; applying a high-resolution visible light image to the three-dimensional surface for texture mapping.

[0027] Preferably, the color correction and image enhancement processing performed on the visible light image data is used to perform color enhancement processing on the acquired visible light image data to distinguish between suspected target data and real target data.

[0028] Preferably, performing texture mapping of the three-dimensional surface model to generate the three-dimensional model of the substation includes:

[0029] Performing mesh processing on the three-dimensional surface model to obtain mesh points;

[0030] The texture mapping coordinates of each grid point are calculated using the principle of multi-view geometry;

[0031] A high-resolution visible light image is mapped onto the surface of a three-dimensional surface model, and the texture is spliced ​​to obtain a mapped three-dimensional surface model.

[0032] Preferably, the high-resolution visible light image is mapped to the surface of the three-dimensional surface model to achieve texture splicing to obtain the mapped three-dimensional surface model. The specific steps are: according to the texture mapping coordinates, the corresponding image blocks are cut out from the high-resolution image; and the image blocks are mapped to the corresponding mesh vertices.

[0033] Preferably, after obtaining the mapped three-dimensional surface model, the following steps are further included:

[0034] Identify different target devices on the three-dimensional surface model and recognize their geometric information;

[0035] and adding semantic information of the target device to each target device according to the three-dimensional surface model;

[0036] A comprehensive three-dimensional model including geometric information and semantic information of target equipment, namely, a three-dimensional model of the substation, is generated.

[0037] A storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned substation three-dimensional modeling method based on the fusion of laser point cloud and visible light.

[0038] Compared with the prior art, the embodiments of the present invention have at least the following technical advantages:

[0039] From the analysis of the above-mentioned substation three-dimensional modeling method based on the fusion of laser point cloud and visible light provided by the present invention, it can be known that the main processing process of the substation three-dimensional modeling method based on the fusion of laser point cloud and visible light includes: acquiring laser point cloud data and visible light image data; preprocessing the acquired data; constructing a three-dimensional surface model based on the preprocessed data; performing texture mapping of the three-dimensional surface model to generate a three-dimensional model of the substation.

[0040] In specific applications, drones equipped with laser radar and visible light cameras are used to achieve all-round, high-precision scanning and shooting of substations. By simultaneously collecting laser point cloud data and visible light image data, the combination of geometric information and visual information is achieved. This fusion improves the accuracy and realism of the model, so that the final three-dimensional model has both high-precision geometric features and realistic visual effects. At the same time, an information mapping table is introduced; the information mapping table associates laser point cloud data, visible light image data with corresponding position coordinates and attitude angle information. This association allows each set of data to have a clear spatial reference, ensuring consistency and traceability between data.

[0041] The 3D surface model was constructed based on the preprocessed data, which enabled efficient digital reconstruction of the complex structure of the substation. This method can accurately capture the geometric features of various equipment and structures in the substation, providing precise spatial information for subsequent analysis and management. By performing texture mapping on the 3D surface model, a highly realistic 3D model of the substation was generated.

[0042] The final generated 3D model of the substation contains not only precise geometric information, but also rich visual details. Modeling in this way can ensure faster modeling speed and richer details, which helps to improve the operational efficiency and management level of the substation.

[0043] In general, the substation 3D modeling method based on the fusion of laser point cloud and visible light realizes the high-precision, high-efficiency and high-realism digital reconstruction of the substation by combining advanced data acquisition technology and processing algorithms. It not only improves the efficiency and accuracy of modeling, but also provides strong technical support for the intelligent management and optimization of substations, and has significant practical value and application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic diagram of the main process of a substation 3D modeling method based on the fusion of laser point cloud and visible light;

[0045] Figure 2 A schematic diagram of the operation steps for scanning and photographing a substation using a laser radar and a visible light camera mounted on a drone in a three-dimensional modeling method of a substation based on the fusion of laser point cloud and visible light, and obtaining laser point cloud data and visible light image data respectively;

[0046] Figure 3 A schematic diagram of the operation steps of data preprocessing in a substation 3D modeling method based on the fusion of laser point cloud and visible light;

[0047] Figure 4A schematic diagram of the operation steps for constructing a three-dimensional surface model in a three-dimensional modeling method for a substation based on the fusion of laser point cloud and visible light;

[0048] Figure 5 A schematic diagram of the operation steps for generating a three-dimensional model of a substation in a three-dimensional modeling method of a substation based on the fusion of laser point cloud and visible light;

[0049] Figure 6 The present invention is a schematic diagram of the operation steps for generating a three-dimensional model of a substation in a three-dimensional modeling method of a substation based on the fusion of laser point cloud and visible light. DETAILED DESCRIPTION

[0050] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] The present invention is further described in detail below through specific embodiments in conjunction with the accompanying drawings.

[0052] like Figure 1 As shown, an embodiment of the present invention provides a substation three-dimensional modeling method based on the fusion of laser point cloud and visible light, including the following operation steps:

[0053] S1. Use a drone equipped with a laser radar and a visible light camera to scan and photograph the substation, and obtain laser point cloud data and visible light image data respectively;

[0054] S2, obtaining laser point cloud data and visible light image data;

[0055] S3, preprocessing the acquired data;

[0056] S4, constructing a three-dimensional surface model based on the preprocessed data;

[0057] S5. Execute texture mapping of the three-dimensional surface model to generate a three-dimensional model of the substation.

[0058] Analysis of the above technical solution shows that the main processing procedures of the above-mentioned substation 3D modeling method based on the fusion of laser point cloud and visible light include: acquiring laser point cloud data and visible light image data; preprocessing the acquired data; constructing a 3D surface model based on the preprocessed data; performing texture mapping of the 3D surface model to generate a 3D model of the substation.

[0059] In specific applications, drones equipped with laser radar and visible light cameras are used to achieve all-round, high-precision scanning and shooting of substations. By simultaneously collecting laser point cloud data and visible light image data, the combination of geometric information and visual information is achieved. This fusion improves the accuracy and realism of the model, so that the final three-dimensional model has both high-precision geometric features and realistic visual effects. At the same time, an information mapping table is introduced; the information mapping table associates laser point cloud data, visible light image data with corresponding position coordinates and attitude angle information. This association allows each set of data to have a clear spatial reference, ensuring consistency and traceability between data.

[0060] The 3D surface model was constructed based on the preprocessed data, which enabled efficient digital reconstruction of the complex structure of the substation. This method can accurately capture the geometric features of various equipment and structures in the substation, providing precise spatial information for subsequent analysis and management. By performing texture mapping on the 3D surface model, a highly realistic 3D model of the substation was generated.

[0061] The final generated 3D model of the substation contains not only precise geometric information, but also rich visual details. Modeling in this way can ensure faster modeling speed and richer details, which helps to improve the operational efficiency and management level of the substation.

[0062] In general, the substation 3D modeling method based on the fusion of laser point cloud and visible light realizes the high-precision, high-efficiency and high-realism digital reconstruction of the substation by combining advanced data acquisition technology and processing algorithms. It not only improves the efficiency and accuracy of modeling, but also provides strong technical support for the intelligent management and optimization of substations, and has significant practical value and application prospects.

[0063] like Figure 2 As shown in the figure, the substation is scanned and photographed using a drone equipped with a laser radar and a visible light camera, and the corresponding laser point cloud data and visible light image data include:

[0064] S11. Plan a preset UAV flight path and set multiple scanning sites;

[0065] S12, execute the current UAV flight path and perform a 360-degree rotation scan at each scanning site;

[0066] S13, performing multi-angle tilt photography on the target device;

[0067] S14. Use a high-precision GNSS / INS system to record the precise position coordinates of each scan and shot; use a gyroscope and accelerometer to record the pitch angle, yaw angle and roll angle of each scan and shot; associate the position coordinates and attitude angle information with the corresponding point cloud data and image data to form an information mapping table.

[0068] like Figure 3 As shown, the preprocessing of the acquired data includes:

[0069] S31, performing denoising processing on the laser point cloud data;

[0070] S32, performing color correction and image enhancement processing on the visible light image data;

[0071] S33, extracting feature points in the visible light image data and feature points in the laser point cloud data;

[0072] S34, obtaining a formation information mapping table to align the laser point cloud data with the feature points of the visible light image data.

[0073] The information mapping table is used to realize data association, spatial positioning and the basis of data fusion; the information mapping table associates the laser point cloud data and visible light image data with the corresponding position coordinates and attitude angle information. This association makes each set of data have a clear spatial reference, ensuring the consistency and traceability between the data and forming the role of data association.

[0074] By recording the precise position coordinates (provided by the GNSS / INS system) and attitude angle information (provided by the gyroscope and accelerometer) of each scan and shot, the information map provides an accurate 3D spatial position for each data point. This is essential for subsequent 3D reconstruction.

[0075] Data fusion basis: The information mapping table provides a unified framework for integrating data from different sensors (lidar, visible light camera, GNSS / INS system, gyroscope, accelerometer), so that these heterogeneous data can be understood and processed in the same reference system. The information mapping table provides a basis for the fusion of laser point cloud data and visible light image data. In step S34, this mapping table can be used to more accurately align the feature points of the two data types, thereby achieving more accurate data fusion.

[0076] In summary, as an intermediate data structure, the information mapping table plays a bridging role between original data collection and subsequent data processing in the entire substation 3D modeling process, and is a key component to ensure the accuracy, consistency and traceability of the entire modeling process.

[0077] like Figure 4As shown, the construction of a three-dimensional surface model based on the preprocessed data includes:

[0078] S41, segmenting the registered point cloud data to identify different target devices;

[0079] S42, fitting the basic geometric shape (such as a plane, a cylinder, etc.) of the target device using a RANSAC algorithm;

[0080] S43, constructing a wireframe model according to the basic geometric shape of the target device;

[0081] S44. Generate a three-dimensional surface model based on the wireframe model.

[0082] Preferably, a three-dimensional surface model is generated based on the wireframe model, specifically including: converting the wireframe model into a triangular mesh through a triangulation algorithm; using a B-spline interpolation method to generate a smooth three-dimensional surface; applying a high-resolution visible light image to the three-dimensional surface for texture mapping (applying a high-resolution visible light image to the surface for texture mapping).

[0083] Step S41: Point cloud data segmentation, which is necessary to segment the registered point cloud data in order to identify different target devices. A variety of segmentation algorithms (such as region growing or clustering-based methods) can be used to achieve this.

[0084] Step S42: Use the RANSAC algorithm to fit the geometric shape. RANSAC (Random Sampling Consensus Algorithm) is an effective method for extracting specific geometric shapes (such as planes, cylinders, spheres, etc.) from data containing noise. In this process, it is necessary to select appropriate models and parameters to ensure the accuracy of the fitting.

[0085] Step S43: constructing a wireframe model. It is reasonable to construct a wireframe model based on the basic geometric shape of the target device. The wireframe model provides a basic structure for the subsequent generation of a three-dimensional surface model.

[0086] Step S44: Generate a three-dimensional surface model. This step should include a specific method of how to generate a surface model from a wireframe model, such as using a meshing technique or a surface interpolation method.

[0087] The specific instructions for generating a 3D surface model are as follows:

[0088] 1. Mesh Generation: Convert the wireframe model to a mesh model, usually using a triangulation algorithm (such as Delaunay triangulation) to generate triangular patches. Formula: Given a set of points P, it can be converted into a triangular mesh T;

[0089] T = Delaunay(P);

[0090] 2. Surface interpolation: Use interpolation methods to generate smooth surfaces on the mesh. For example, you can use B-spline or NURBS methods to create smoother surfaces.

[0091] For a given set of control points C, the expression of the B-spline surface is:

[0092]

[0093] Where N i,k (u) and N j,l (v) is the B-spline basis function.

[0094] 3. Texture mapping: Apply texture images on the generated 3D surface to enhance the visual effect. For each vertex V (x, y, z) on the mesh, its texture coordinates (u, v);

[0095] Calculated by texture mapping formula: T(V) = I(u, v);

[0096] In summary, in step S4, the process of generating a 3D surface model needs to be described in more detail, including specific steps such as meshing, surface interpolation and texture mapping. By refining these steps, we can more clearly understand how to generate the final 3D surface model from the wireframe model.

[0097] The basic geometric shape (such as plane, cylinder, etc.) of the target device is fitted using the RANSAC algorithm; a wireframe model is constructed based on the basic geometric shape of the target device; and a three-dimensional surface model is generated based on the wireframe model. The above execution process is conducive to accurately identifying and classifying various equipment in the substation, improving the accuracy and robustness of geometric shape fitting, and providing technical support for generating smooth and accurate three-dimensional surface models.

[0098] Preferably, the color correction and image enhancement processing performed on the visible light image data is used to perform color enhancement processing on the acquired visible light image data to distinguish between suspected target data and real target data.

[0099] like Figure 5 As shown, texture mapping of the three-dimensional surface model is performed to generate a three-dimensional model of the substation, including:

[0100] S51, meshing the three-dimensional surface model to obtain grid points;

[0101] S52, calculating the texture mapping coordinates of each grid point using the multi-view geometry principle;

[0102] S53, mapping the high-resolution visible light image to the surface of the three-dimensional surface model, and splicing the texture to obtain the mapped three-dimensional surface model.

[0103] Preferably, the high-resolution visible light image is mapped to the surface of the three-dimensional surface model to achieve texture splicing to obtain the mapped three-dimensional surface model. The specific steps are: according to the texture mapping coordinates, the corresponding image blocks are cut out from the high-resolution image; and the image blocks are mapped to the corresponding mesh vertices.

[0104] The execution process of S51-S53 is beneficial to improving the visual realism of the model, ensuring the accuracy and continuity of texture mapping, and enhancing the detail expression of the model.

[0105] like Figure 6 As shown, after obtaining the mapped three-dimensional surface model, step S5 further includes the following steps:

[0106] S54, marking different target devices on the three-dimensional surface model and identifying their geometric information;

[0107] S55, and adding semantic information of the target device to each target device according to the three-dimensional surface model;

[0108] S56. Generate a comprehensive three-dimensional model including geometric information and semantic information of target equipment, namely, a three-dimensional model of the substation.

[0109] S51: In the meshing process of the three-dimensional surface model, the main steps include:

[0110] Create a mesh structure: Select a triangle or quadrilateral mesh as the basic unit.

[0111] Determine the resolution of the grid and set the grid size as required.

[0112] Compute Mesh Vertices: Calculates the positions of mesh vertices using surface equations.

[0113] Execution formula: For a given surface equation z = f(x, y), the position of the mesh vertex V(i, j) is: V(i, j) = (x i ,y j ,f(x i ,y j )); where x i and j are discrete points generated according to the grid step size.

[0114] S52: Calculate texture mapping coordinates: The calculation process of texture mapping coordinates includes:

[0115] S521. Extract feature points: extract feature points (such as corner points, edges) from the visible light image.

[0116] S522. Establishing a corresponding relationship: matching the feature points with the points in the three-dimensional space and calculating the texture coordinates.

[0117] Execution formula: For each feature point P's coordinates (X, Y, Z) in 3D space, its corresponding texture coordinates (u, v) can be calculated by the following transformation:

[0118]

[0119] Where W and H are the width and height of the texture image respectively.

[0120] S53: Mapping high-resolution visible light image: Mapping the acquired high-resolution image to the mesh model to achieve texture stitching. Image cutting: Cutting the high-resolution image into small blocks according to the texture coordinates. Texture stitching: Mapping the cut image blocks to the corresponding mesh vertices.

[0121] Execution formula:

[0122] Assuming I is a high-resolution visible light image, the formula for mapping to the grid is:

[0123] M(u,v)=I(u,v);

[0124] This means mapping the pixel values ​​of the image I to the corresponding texture coordinates of the 3D model.

[0125] S54: Identify different target devices

[0126] In the mapped 3D surface model, different target devices are identified and their geometric information is extracted.

[0127] Segmentation model: Use segmentation algorithms (such as region-based segmentation) to identify different devices.

[0128] Geometric information extraction: Extract geometric features (such as bounding box, volume) of each device.

[0129] Execution formula:

[0130] Assume the bounding box of the target device is B, and the geometric information can be expressed as:

[0131] Volume(B)=L·W·H;

[0132] Where L, W, and H are the length, width, and height of the bounding box respectively.

[0133] S55: Add semantic information of target device: Add relevant semantic information (such as device name, function, etc.) to each target device. Semantic information database: Establish a database to store semantic information of devices. Associate devices with information: Associate geometric information with corresponding semantic information.

[0134] Execution formula:

[0135] Assume that the semantic information of the target device is SSS, which can be expressed as:

[0136] S(i) = {name, type, status, ...}; where i represents the index of the device;

[0137] S56: Generate comprehensive 3D model: Combine geometric information and semantic information to generate a complete 3D model of the substation. Integrate data: Integrate the geometric information and semantic information of all target devices into a comprehensive model. Output 3D model: Save in a standard format (such as OBJ, FBX).

[0138] Formula: The comprehensive three-dimensional model M can be expressed as:

[0139] Where n is the number of target devices.

[0140] The above steps describe in detail the process from meshing of the 3D surface model to generating a comprehensive 3D model.

[0141] For example, the above implementation steps:

[0142] Data collection: Use drones equipped with laser radar and high-precision cameras. Design 15 scanning sites to cover the entire substation area. Perform 360-degree rotation scanning at each site, and perform multi-angle tilt photography of key equipment. Record the precise location and posture information of each scan and shot.

[0143] Data preprocessing: CloudCompare software was used to denoise and preliminarily register the point cloud data. Adobe Lightroom was used to perform color correction and enhancement on the visible light image. SIFT algorithm was used to extract image feature points, and 3DSIFT was used to extract point cloud feature points. RANSAC algorithm was used for feature point matching and data registration.

[0144] 3D surface model construction: PointCNN is used to perform semantic segmentation of point clouds and identify major equipment such as transformers and circuit breakers. The RANSAC algorithm is used to fit basic geometric shapes for each type of equipment. Rhino software is used to build a wireframe model and generate a NURBS surface.

[0145] Texture mapping and comprehensive model generation: Meshing: Use Meshlab software to perform quadrilateral meshing on the NURBS surface.

[0146] Texture mapping: Structure from Motion (SfM) technology is used to calculate camera parameters and texture mapping coordinates. Agisoft Metashape software is used for image stitching and texture mapping.

[0147] Device identification and geometric information extraction: Use the deep learning model PointNet++ to finely classify devices and extract geometric information such as size and volume of each device.

[0148] Adding semantic information: Establishing a device database, including information such as model, rated voltage, capacity, etc. Associate the extracted geometric information with the semantic information in the database.

[0149] Comprehensive 3D model generation: Use Autodesk Revit software to integrate geometric models, textures, and semantic information. Build hierarchical structures and spatial relationships between devices. Generate a final model that includes geometric, texture, and semantic information. This case creates a 3D model of a substation that is both accurate and rich in information, providing strong support for substation management and optimization.

[0150] In summary, the example of the present invention proposes a three-dimensional modeling method for substations based on the fusion of laser point cloud and visible light, which uses mobile laser scanning technology to scan and model existing physical equipment and buildings, and has the technical characteristics of high modeling efficiency, highly realistic scenes, and high precision. By combining high-precision geometric information and high-resolution texture information, accurate three-dimensional reconstruction of the substation is achieved. At the same time, by adding semantic information, the model not only has a high degree of visual authenticity, but also contains rich attribute information, which provides strong support for the digital management, analysis, and decision-making of the substation.

[0151] Embodiment 2

[0152] The third embodiment of the present invention provides a storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for three-dimensional modeling of a substation based on the fusion of laser point cloud and visible light are implemented.

[0153] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0154] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0155] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0156] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. A person skilled in the art may modify the technical solutions described in the above embodiments, or replace part or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A three-dimensional modeling method for substation based on the fusion of laser point cloud and visible light, characterized in that: The steps are as follows: Use drones equipped with laser radar and visible light cameras to scan and photograph the substation, and obtain laser point cloud data and visible light image data respectively; Acquire laser point cloud data and visible light image data; Preprocess the acquired data; Constructing a three-dimensional surface model based on the preprocessed data; Perform texture mapping of the 3D surface model to generate a 3D model of the substation.

2. The three-dimensional modeling method of substation based on the fusion of laser point cloud and visible light according to claim 1 is characterized in that: The substation is scanned and photographed using a drone equipped with a laser radar and a visible light camera. The corresponding laser point cloud data and visible light image data include: Plan a preset drone flight path and set multiple scanning sites; Execute the current drone flight path and perform a 360-degree rotation scan at each scanning station; Perform multi-angle oblique photography of the target device; A high-precision GNSS / INS system is used to record the precise position coordinates of each scan and shot. The gyroscope and accelerometer are used to record the pitch angle, yaw angle and roll angle of each scan and shot. The position coordinates and attitude angle information are associated with the corresponding point cloud data and image data to form an information mapping table.

3. The substation three-dimensional modeling method based on laser point cloud and visible light fusion according to claim 2 is characterized in that: The preprocessing of the acquired data includes: De-noising the laser point cloud data; Perform color correction and image enhancement on visible light image data; Extract feature points from visible light image data and feature points from laser point cloud data; Obtain the formation information mapping table and align the feature points of the laser point cloud data with the visible light image data.

4. The three-dimensional modeling method of substation based on laser point cloud and visible light fusion according to claim 3 is characterized in that: The construction of 3D surface model based on preprocessed data includes: Segment the registered point cloud data and identify different target devices; Using a RANSAC algorithm to fit the basic geometric shape of the target device; Building a wireframe model based on the basic geometric shape of the target device; Generate a 3D surface model based on the wireframe model.

5. The three-dimensional modeling method of substation based on laser point cloud and visible light fusion according to claim 4 is characterized in that: Generating a three-dimensional surface model based on a wireframe model specifically includes: converting the wireframe model into a triangular mesh through a triangulation algorithm; generating a smooth three-dimensional surface using a B-spline interpolation method; and applying a high-resolution visible light image to the three-dimensional surface for texture mapping.

6. The three-dimensional modeling method of substation based on laser point cloud and visible light fusion according to claim 5 is characterized in that: The color correction and image enhancement processing performed on the visible light image data is used to perform color enhancement processing on the acquired visible light image data to distinguish between suspected target data and real target data.

7. The three-dimensional modeling method of substation based on laser point cloud and visible light fusion according to claim 6 is characterized in that: Execute texture mapping of the 3D surface model to generate a 3D model of the substation including: Performing mesh processing on the three-dimensional surface model to obtain mesh points; The texture mapping coordinates of each grid point are calculated using the principle of multi-view geometry; A high-resolution visible light image is mapped onto the surface of a three-dimensional surface model, and the texture is spliced ​​to obtain a mapped three-dimensional surface model.

8. The three-dimensional modeling method of substation based on laser point cloud and visible light fusion according to claim 7 is characterized in that: The high-resolution visible light image is mapped to the surface of the three-dimensional surface model, and the texture is spliced ​​to obtain the mapped three-dimensional surface model. The specific steps are: according to the texture mapping coordinates, the corresponding image blocks are cut out from the high-resolution image; and the image blocks are mapped to the corresponding mesh vertices.

9. The three-dimensional modeling method of substation based on laser point cloud and visible light fusion according to claim 1 is characterized in that: After obtaining the mapped three-dimensional surface model, the following steps are also included: Identify different target devices on the three-dimensional surface model and recognize their geometric information; and adding semantic information of the target device to each target device according to the three-dimensional surface model; Generate a 3D substation model containing geometric information and semantic information of target equipment.

10. A storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, implements the steps of the substation three-dimensional modeling method based on the fusion of laser point cloud and visible light as described in any one of claims 1 to 9.

Images