Power transmission line terrain fusion method and system, storage medium and computer equipment

Abstract

According to the power transmission line terrain fusion method and system, the storage medium and the computer equipment provided by the invention, when the power transmission line terrain is fused, the point cloud data of the point cloud channel terrain and the digital elevation model data of the basic terrain are firstly obtained, and the overlapping region of the two kinds of data is divided into the plurality of initial grids, so that the spatial consistency is ensured; then data quality information and topographic information of each initial grid are determined based on the two kinds of data and then comprehensive analysis is carried out to analyze fusion performance of different data, a data evaluation model is constructed and obtained, the model is utilized to carry out ICP algorithm applicability evaluation on each initial grid, a local optimal solution is prevented from being trapped in processing of a complex topography, and the reliability of the data evaluation model is improved. And finally, performing grid optimization on the power transmission line by adopting a particle swarm optimization algorithm according to an evaluation result, thereby realizing data combination of the point cloud channel terrain and the basic terrain of the power transmission line, enabling the point cloud channel terrain and the basic terrain to adapt to different terrain features, and further improving a matching effect and analysis efficiency.

Description

Technical Field

[0001] This application relates to the technical field of terrain fusion, and particularly to a method, system, storage medium, and computer device for terrain fusion of transmission lines. Background Art

[0002] With the development of unmanned aerial vehicle technology, light detection and ranging (LiDAR), and high-precision digital elevation model (DEM), point cloud data has become an important data source for terrain modeling, 3D reconstruction, and precise map construction. Especially in the inspection and maintenance of transmission lines, the fusion application of point cloud data and DEM data can provide more accurate terrain information.

[0003] Current terrain fusion methods for transmission lines can use the iterative closest point (ICP) algorithm to register and optimize the fusion of point cloud data of transmission lines and digital elevation model (DEM) data, and can perform mesh division for different terrain regions to solve the fusion of point cloud data and digital elevation model within the mesh. However, in the prior art, due to the data quality of point cloud data and digital elevation model data and the complexity of the terrain, when dealing with complex and irregular terrains, the ICP algorithm is prone to falling into a local optimal solution. Summary of the Invention

[0004] The purpose of this application aims to solve at least one of the above technical defects, especially the technical defect that due to the data quality of point cloud data and digital elevation model data and the complexity of the terrain in the prior art, when dealing with complex and irregular terrains, the ICP algorithm is prone to falling into a local optimal solution.

[0005] This application provides a method for terrain fusion of transmission lines, and the method includes:

[0006] Obtain the point cloud data of the point cloud channel terrain of the transmission line and the digital elevation model data of the basic terrain, and divide the overlapping area of the point cloud data and the digital elevation model data in the transmission line into multiple initial meshes;

[0007] Determine the data quality information and terrain information of each initial mesh based on the point cloud data and digital elevation model data of each initial mesh;

[0008] Comprehensively analyze the data quality information and terrain information of each initial mesh, and construct a data evaluation model according to the analysis results;

[0009] Use the data evaluation model to evaluate the applicability of the ICP algorithm for each initial mesh, and optimize the mesh of the transmission line using the particle swarm optimization algorithm according to the evaluation results.

[0010] Optionally, dividing the overlapping area of the point cloud data and the digital elevation model data in the transmission line into a plurality of initial grids includes:

[0011] Performing area division in the transmission line according to a preset unit area to obtain a division result;

[0012] Based on the division result, respectively performing data segmentation on the point cloud data and the digital elevation model data to obtain a point cloud sub-region corresponding to the point cloud data and a basic sub-region corresponding to the digital elevation model data;

[0013] Using the grid method to determine the overlapping area in the point cloud sub-region and the basic sub-region, and performing local fusion analysis on each overlapping area to form an initial grid.

[0014] Optionally, determining the data quality information of each initial grid based on the point cloud data and the digital elevation model data of each initial grid includes:

[0015] For each initial grid, respectively calculating a data density difference coefficient and a quality characteristic performance coefficient between the point cloud data and the digital elevation model data of the initial grid, and determining the data quality information of the initial grid according to the data density difference coefficient and the quality characteristic performance coefficient.

[0016] Optionally, the calculation process of the data density difference coefficient includes:

[0017] Using the Euclidean distance measurement method to determine the point cloud nearest neighbor distance of the point cloud data of the initial grid in the point cloud coordinate system and the overall point cloud average nearest neighbor distance, as well as the basic nearest neighbor distance of the digital elevation model data in the basic terrain coordinate system and the overall basic average nearest neighbor distance;

[0018] Calculating a data density difference coefficient between the point cloud data and the digital elevation model data according to the point cloud nearest neighbor distance, the point cloud average nearest neighbor distance, the basic nearest neighbor distance, and the basic average nearest neighbor distance.

[0019] Optionally, the calculation process of the quality characteristic performance coefficient includes:

[0020] Converting the point cloud data of the initial grid into the basic terrain coordinate system, and determining an initial fitting model according to the conversion result, and determining a current fitting model according to the digital elevation model data of the initial grid; wherein, the initial fitting model includes an initial error threshold and an initial number of inliers;

[0021] Calculate the distance from the point cloud data to the current fitting model based on the initial fitting model, and determine the current inlier number and the current average error of the current fitting model according to the distance and the initial error threshold.

[0022] When the current inlier number is greater than the initial inlier number, or when the current inlier number is equal to the initial inlier number and the current average error is less than a preset threshold, iteratively update the current fitting model to form an optimal fitting model.

[0023] Obtain the optimal inlier number, the optimal inlier ratio, and the optimal mean square error of the optimal fitting model, and calculate the quality characteristic performance coefficient between the point cloud data and the digital elevation model data according to the optimal inlier number, the optimal inlier ratio, and the optimal mean square error.

[0024] Optionally, the process of determining the terrain information includes:

[0025] For each initial grid, convert the point cloud data of the initial grid into the basic terrain coordinate system to obtain the elevation coordinate data at each coordinate position in the point cloud data.

[0026] Calculate the average value, the standard deviation, and the coefficient of variation of each coordinate position in the point cloud data based on the elevation coordinate data.

[0027] Calculate the terrain irregularity anomaly coefficient of the initial grid according to the average value, the standard deviation, and the coefficient of variation, and determine the terrain information of the initial grid based on the terrain irregularity anomaly coefficient.

[0028] Optionally, the comprehensive analysis of the data quality information and the terrain information of each initial grid, and the construction of a data evaluation model according to the analysis results include:

[0029] Perform information analysis on the data quality information of each initial grid to obtain the data density difference coefficient and the quality characteristic performance coefficient of each initial grid, and perform information analysis on the terrain information of each initial grid to obtain the terrain irregularity anomaly coefficient of each initial grid.

[0030] For each initial grid, perform weighted summation on the data density difference coefficient, the quality characteristic performance coefficient, and the terrain irregularity anomaly coefficient of the initial grid to obtain the grid data evaluation coefficient.

[0031] Construct a data evaluation model according to the grid data evaluation coefficients of each initial grid.

[0032] Optionally, the grid optimization of the transmission line according to the evaluation result by using the particle swarm optimization algorithm includes:

[0033] Initialize the particle swarm, and determine the grid data evaluation coefficient of the transmission line according to the evaluation result, so as to determine the objective function of the particle swarm according to the grid data evaluation coefficient;

[0034] Use the velocity update formula and the position update formula to perform iterative calculations on the objective function, so as to update the historical optimal position and the global optimal position of each particle in the particle swarm until the maximum iteration end condition is reached, and output the optimal grid division scheme;

[0035] Optimize the grid of the transmission line according to the optimal grid optimization scheme.

[0036] The present application also provides a transmission line terrain fusion system, which is characterized by including:

[0037] A grid division module, configured to obtain the point cloud data of the point cloud channel terrain of the transmission line and the digital elevation model data of the basic terrain, and divide the overlapping area of the point cloud data and the digital elevation model data in the transmission line into multiple initial grids;

[0038] An information determination module, configured to determine the data quality information and terrain information of each initial grid based on the point cloud data and the digital elevation model data of each initial grid;

[0039] A model construction module, configured to perform a comprehensive analysis on the data quality information and terrain information of each initial grid, and construct a data evaluation model according to the analysis result;

[0040] A grid optimization module, configured to use the data evaluation model to evaluate the applicability of the ICP algorithm to each initial grid, and optimize the grid of the transmission line by using the particle swarm optimization algorithm according to the evaluation result.

[0041] The present application also provides a storage medium, which is characterized in that: computer-readable instructions are stored in the storage medium, and when the computer-readable instructions are executed by one or more processors, the one or more processors are caused to execute the steps of the transmission line terrain fusion method as described in any one of the above embodiments.

[0042] The present application also provides a computer device, including: one or more processors, and a memory;

[0043] Computer-readable instructions are stored in the memory, and when the computer-readable instructions are executed by the one or more processors, the steps of the transmission line terrain fusion method as described in any one of the above embodiments are executed.

[0044] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0045] The transmission line terrain fusion method, system, storage medium and computer device provided by this application can, when performing terrain data fusion on a transmission line, first obtain the point cloud data of the point cloud channel terrain of the transmission line and the digital elevation model data of the basic terrain, and divide the overlapping area of the point cloud data and the digital elevation model data in the transmission line into multiple initial grids to ensure the spatial consistency of these two types of data. Then, it can determine the data quality information and terrain information of each initial grid based on the point cloud data and the digital elevation model data of each initial grid to analyze the fusion performance of different data. Next, it can comprehensively analyze the data quality information and terrain information of each initial grid, so as to construct a data evaluation model according to the analysis results, and use this data evaluation model to evaluate the applicability of the ICP algorithm for each initial grid, avoiding falling into a local optimal solution when dealing with complex and irregular terrains. Finally, it can also optimize the grid of the transmission line by using the particle swarm optimization algorithm according to the evaluation results. All in all, this application can automatically adjust the grid division through the particle swarm optimization algorithm, realize the data combination of the point cloud channel terrain and the basic terrain of the transmission line, make it adapt to different terrain features, and thus improve the matching effect and analysis efficiency. Brief Description of the Drawings

[0046] In order to more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0047] Figure 1 It is a flowchart showing a transmission line terrain fusion method provided by an embodiment of this application;

[0048] Figure 2 It is a flowchart showing a process of initial grid division provided by an embodiment of this application;

[0049] Figure 3 It is a flowchart showing a process of determining terrain information provided by an embodiment of this application;

[0050] Figure 4 It is a flowchart showing a process of constructing a data evaluation model provided by an embodiment of this application;

[0051] Figure 5 It is a structural diagram showing a transmission line terrain fusion system provided by an embodiment of this application;

[0052] Figure 6 It is an internal structural diagram of a computer device provided by an embodiment of this application. Specific Embodiments

[0053] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts belong to the scope of protection of the present application.

[0054] The current terrain fusion method for transmission lines can use the ICP algorithm to register and optimize the fusion of the point cloud data of the transmission line and the digital elevation model (DEM) data, and can perform grid division for different terrain regions, thereby solving the fusion of the point cloud data and the digital elevation model within the grid. However, in the prior art, due to the data quality of the point cloud data and the digital elevation model data and the complexity of the terrain, when dealing with complex and irregular terrains, the ICP algorithm is prone to falling into a local optimal solution.

[0055] Based on this, the present application proposes the following technical solutions. For details, please refer to the following text:

[0056] In one embodiment, as Figure 1 shown Figure 1 is a schematic flowchart of a terrain fusion method for a transmission line provided by an embodiment of the present application; the present application provides a terrain fusion method for a transmission line, which specifically includes the following:

[0057] S110: Obtain the point cloud data of the point cloud channel terrain of the transmission line and the digital elevation model data of the basic terrain, and divide the overlapping area of the point cloud data and the digital elevation model data in the transmission line into multiple initial grids.

[0058] In this step, when the computer device performs terrain data fusion on the transmission line, it can first determine the terrain of the data to be fused on the transmission line, that is, the point cloud channel terrain and the basic terrain, and then obtain the point cloud data of the point cloud channel terrain and the digital elevation model data of the basic terrain, and divide the overlapping area of the point cloud data and the digital elevation model data in the transmission line into multiple initial grids to ensure the spatial consistency of these two types of data.

[0059] Among them, the point cloud channel terrain refers to the three-dimensional terrain representation along the transmission line corridor. Such terrain data is usually collected by high-precision sensors such as lidar (LiDAR), digital cameras, and panoramic cameras, and is represented as a series of point cloud data with coordinates (X, Y, Z) in three-dimensional space; while the basic terrain refers to a more extensive three-dimensional terrain representation, representing basic information such as the undulation of the ground surface and landform types, and usually exists in the form of digital elevation model (DEM) data; the digital elevation model data here refers to an elevation data model represented in the form of a grid or mesh, and each initial grid contains the ground elevation value at that location. In addition, the basic terrain data can also exist in the form of existing terrain databases, digital ground models and other basic terrain data, which are not restricted here.

[0060] It should be noted that the point cloud data collected by drones usually adopts a local coordinate system. Especially when collecting data in a small area, it may be located relative to a specific reference point, such as the takeoff point of the drone or a certain ground coordinate; while the digital elevation model data is usually based on a global coordinate system, such as the coordinate system of the digital elevation model data, and there is a problem of inconsistent coordinate systems between it and the local point cloud data collected by drones, especially when converting between different projection methods or coordinate reference systems. Although the digital elevation model data has high precision and wide coverage, there may also be local inaccuracies or data gap areas. The digital elevation model data can provide the terrain information of the entire area, including elevation and slope, but the detailed terrain data of the power line channel that can be provided by the point cloud data obtained by the drone through lidar is mainly concentrated on the transmission line channel, and the coordinate system of this data is not completely consistent with the digital elevation model data. Therefore, this application needs to perform terrain fusion on these two types of data on the transmission line.

[0061] It can be understood that during the data division process in the overlapping area, the computer device can first perform coordinate system conversion on the combined point cloud data and digital elevation model data in the overlapping area to ensure that the two are aligned in the same reference system, and can determine the changes in the transmission line channel terrain on the digital elevation model to obtain a more accurate transmission line channel terrain. By integrating and dividing the point cloud data and digital elevation model data in the overlapping area, the computer device can achieve a comprehensive and accurate description of the transmission line and its surrounding terrain, thereby improving the evaluation accuracy of the transmission line and providing a more scientific decision-making basis for the follow-up.

[0062] S120: Determine the data quality information and terrain information of each initial grid based on the point cloud data and digital elevation model data of each initial grid.

[0063] In this step, after obtaining multiple initial grids through step S110, the computer device can determine the data quality information and terrain information of each initial grid based on the point cloud data and digital elevation model data of each initial grid, so as to analyze the fusion performance of different terrain data.

[0064] Among them, the data quality information refers to the comprehensive index used to evaluate the accuracy, integrity, consistency, and availability of data when processing geospatial data; in this application, it specifically refers to the quality assessment results for the point cloud data and digital elevation model (DEM) data of each initial grid. The terrain information refers to the data set extracted from geospatial data about the surface morphology and characteristics; in this application, it specifically refers to the description of the terrain features extracted for each initial grid.

[0065] It can be understood that after the preliminary division of the transmission line is completed and multiple initial grids are obtained, the computer device can accurately determine the data quality information and terrain information of each initial grid based on the point cloud data and digital elevation model data included in each initial grid. This process not only involves the evaluation of key quality indicators such as the density, accuracy, and integrity of the point cloud data, but may also include the detailed extraction and quantification of terrain features such as slope, elevation change, and surface roughness, so as to comprehensively and deeply understand the performance of different terrain data in the fusion process, including data compatibility, information complementarity, and the overall quality and reliability of the fused data, thereby providing solid data support and scientific basis for subsequent terrain modeling, environmental change monitoring, disaster risk assessment, etc.

[0066] S130: Conduct a comprehensive analysis of the data quality information and terrain information of each initial grid, and construct a data evaluation model based on the analysis results.

[0067] In this step, after obtaining the data quality information and terrain information of each initial grid through step S120, the computer device can conduct a comprehensive analysis of the data quality information and terrain information of each initial grid, and construct a data evaluation model based on the analysis results, so as to use this data evaluation model to evaluate each initial grid and avoid falling into a local optimal solution when dealing with complex and irregular terrains.

[0068] It can be understood that the computer device can consider the mutual relationship between the data quality information and the terrain information and its impact on the overall quality of the terrain data, and then conduct a comprehensive evaluation, and then construct a refined and efficient data evaluation model. This model is an intelligent evaluation tool that integrates the analysis results of various data quality characteristics and terrain characteristics. It can use automated algorithms and technologies to integrate complex information into a unified evaluation framework, so as to accurately and objectively quantify the data quality and terrain complexity of different initial grids, and make more informed decisions in the subsequent data processing and analysis process.

[0069] S140: Use the data evaluation model to evaluate the applicability of the ICP algorithm for each initial grid, and optimize the grid of the transmission line by using the particle swarm optimization algorithm according to the evaluation results.

[0070] In this step, after constructing the data evaluation model through step S130, the computer device can use the data evaluation model to evaluate the applicability of the ICP algorithm for each initial grid, and optimize the grid of the transmission line by using the particle swarm optimization algorithm according to the evaluation results, so as to automatically adjust the grid division through the particle swarm optimization algorithm, realize the data combination of the terrain of the point cloud channel of the transmission line and the basic terrain, make it adapt to different terrain characteristics, and then improve the matching effect and analysis efficiency.

[0071] Among them, the ICP (Iterative Closest Point) algorithm refers to an algorithm based on data registration method, which uses the closest point search method to solve the algorithm based on free-form surfaces. Through iteration, it continuously optimizes the matching error between two point clouds until the error is minimized. The particle swarm optimization algorithm refers to an optimization algorithm based on swarm intelligence. By simulating the foraging behavior of biological groups such as bird flocks, it uses the information sharing and cooperation mechanism in the group to find the optimal solution to the problem; each particle in the algorithm adjusts its position and speed according to its own experience and the experience of the group to gradually approach the optimal solution; each particle in this application can refer to a grid division scheme.

[0072] It can be understood that the computer device can conduct a detailed ICP algorithm applicability evaluation on each initial grid through the data evaluation model to measure the alignment degree between different point cloud data. After obtaining the evaluation results, the computer device can intelligently select the particle swarm optimization algorithm according to the evaluation results to further optimize the grid division of the transmission line, and continuously iteratively search for the optimal solution to the problem through information sharing and cooperation among particles.

[0073] Specifically, in the particle swarm optimization process, each particle represents a possible grid division scheme, and its position is defined by a series of parameters that determine the shape, size, and position of the grid. By calculating the fitness value of each particle, that is, the alignment error of the ICP algorithm or other relevant metrics under this grid division scheme, the computer device can evaluate the advantages and disadvantages of each scheme. Subsequently, based on the information of the individual best position (Pbest) and the global best position (Gbest), the particle swarm optimization algorithm continuously adjusts the speed and position of the particles, gradually approaching the optimal grid division scheme, and thus realizing the grid optimization of the transmission line.

[0074] In the above embodiment, when fusing the terrain data of the transmission line, the point cloud data of the point cloud channel terrain and the digital elevation model data of the basic terrain of the transmission line can be obtained first, and the overlapping area of the point cloud data and the digital elevation model data in the transmission line can be divided into multiple initial grids to ensure the spatial consistency of these two types of data. Then, the data quality information and terrain information of each initial grid can be determined based on the point cloud data and the digital elevation model data of each initial grid to analyze the fusion performance of different data. Next, a comprehensive analysis can be performed on the data quality information and terrain information of each initial grid, so that a data evaluation model can be constructed according to the analysis results, and the applicability of the ICP algorithm to each initial grid can be evaluated using this data evaluation model to avoid falling into a local optimal solution when dealing with complex and irregular terrains. Finally, the particle swarm optimization algorithm can be used to optimize the grid of the transmission line according to the evaluation results. In summary, this application can automatically adjust the grid division through the particle swarm optimization algorithm, realize the data combination of the point cloud channel terrain and the basic terrain of the transmission line, make it adapt to different terrain features, and thus improve the matching effect and analysis efficiency.

[0075] In one embodiment, as Figure 2 shown, Figure 2 is a schematic flowchart of an initial grid division process provided by an embodiment of this application; Figure 2 In, the process of dividing the overlapping area of the point cloud data and the digital elevation model data in the transmission line into multiple initial grids in step S110 may include:

[0076] S111: Perform regional division in the transmission line according to a preset unit area to obtain a division result.

[0077] S112: Based on the division result, perform data segmentation on the point cloud data and the digital elevation model data respectively to obtain a point cloud sub-region corresponding to the point cloud data and a basic sub-region corresponding to the digital elevation model data.

[0078] S113: Determine the overlapping regions in the point cloud sub-region and the base sub-region using the grid method, and perform local fusion analysis on each overlapping region to form an initial grid.

[0079] In this embodiment, the computer device can divide the region in the transmission line according to the preset unit region to obtain the division result. Then, it can perform data segmentation on the point cloud data and the digital elevation model data respectively according to the division result to obtain the point cloud sub-region corresponding to the point cloud data and the base sub-region corresponding to the digital elevation model data. Furthermore, it can use the grid method to determine the overlapping regions in the point cloud sub-region and the base sub-region, and perform local fusion analysis on each overlapping region to form an initial grid.

[0080] It can be understood that the size of the preset unit region in this application is determined by the staff in the professional field, which can be determined in combination with the complexity of the algorithm and according to the actual situation, and is not limited here.

[0081] It should be noted that when the computer device performs local fusion analysis on each overlapping region, it can use the ICP algorithm to combine the point cloud data and the digital elevation model data of the transmission line to obtain the elevation value of the corresponding overlapping region.

[0082] Specifically, before applying the ICP algorithm, the computer device can first perform rough registration on the point cloud data and the digital elevation model data, that is, initially align the point cloud data and the digital elevation model data through known reference information, such as GPS coordinates, flight paths, etc.; for each point in the point cloud data, the computer device can find the nearest point in the digital elevation model data to form a point pair set, and then calculate the optimal rigid transformation matrix according to the point pair set, including the rotation matrix and the translation vector. Using the method of minimizing the error, find the transformation matrix that minimizes the Euclidean distance between the two point sets. Then, the calculated transformation matrix can be applied to the point cloud data to update the position and orientation of the point cloud. Repeat the operation to complete the high-precision fusion of the point cloud data and the digital elevation model data to form an initial grid.

[0083] In one embodiment, the process of determining the data quality information of each initial grid based on the point cloud data and the digital elevation model data of each initial grid in step S120 may include:

[0084] S121: For each initial grid, calculate the data density difference coefficient and the quality characteristic performance coefficient between the point cloud data and the digital elevation model data of this initial grid respectively, and determine the data quality information of this initial grid according to the data density difference coefficient and the quality characteristic performance coefficient.

[0085] In this embodiment, after each initial grid is divided, for each initial grid, the computer device can calculate the data density difference coefficient and the quality characteristic performance coefficient between the point cloud data and the digital elevation model data of the initial grid respectively, and determine the data quality information of the initial grid according to the data density difference coefficient and the quality characteristic performance coefficient.

[0086] Among them, the data density difference coefficient can reflect the difference between the point cloud data and the digital elevation model data; the higher the data density coefficient, the greater the density difference between the two. The quality characteristic performance coefficient can be used to evaluate the overall quality of the point cloud data and the digital elevation model data; the higher the quality characteristic performance coefficient, the better the data quality.

[0087] It can be understood that since the ICP algorithm may be affected by incorrect matching points and lead to inaccurate registration results, the computer device can use the RANSAC (Random Sample Consensus) algorithm before using the ICP algorithm to remove abnormal points or noise points at the initial stage of registration to ensure that the registration process of ICP is based on reliable matching point pairs.

[0088] Specifically, the RANSAC algorithm can iteratively select the best registration model by continuously randomly selecting point pairs, fitting models, and evaluating inliers and outliers, ensuring the removal of noise points and obtaining a robust transformation model that can accurately describe the relationship between the point cloud data and the digital elevation model data. After removing noise points by RANSAC, the point cloud data has a higher matching quality, and in the subsequent ICP registration, the alignment of the source point cloud and the target point cloud will be more accurate.

[0089] Among them, inliers refer to points where the difference between the point cloud data and the digital elevation model data is less than a set threshold as inliers, that is, they conform to the fitting model, and points greater than the set threshold are regarded as outliers, that is, they do not conform to the fitting model and belong to noise points. The best fitting model is determined by obtaining the maximum number of inliers. The fitting models include plane models, line models, and curve models, etc. Generally, the plane model is used in flat areas, the line model is used in gentle slope areas, and the curve model is used in mountainous and complex areas.

[0090] It should be noted that the digital elevation model data usually covers a large range of terrain areas, including all terrain backgrounds where the transmission line corridor is located, while the point cloud data collected by drones usually only contains a relatively narrow area within the corridor. Therefore, when performing model fitting, the digital elevation model data can provide more comprehensive terrain information, which helps to construct a fitting model that better conforms to the actual terrain. Moreover, the density of the point cloud data collected by drones is usually uneven. In some areas with high buildings, vegetation, or sensor occlusion, the point cloud density may be low, while in the overlapping area of the flight path, the density may be high.

[0091] Specifically, the computer device can compare the density differences between the actual point cloud data and the digital elevation model data within the initially divided grid area, and determine the differences between the point cloud data and the fitting model of the digital elevation model data by using the RANSAC algorithm, so as to further determine the data quality information within the initial grid. In this application, the data quality information can be represented by the data density difference coefficient and the quality characteristic performance coefficient. Among them, the density difference will affect the effect of the ICP algorithm. When the density of the point cloud data within the grid is high while the digital elevation model data is sparse, the RANSAC fitting model will be more inclined to be affected by the dense point cloud, which may cause the digital elevation model data to fail to participate in the fitting accurately, increasing the error. When the density of the point cloud data within the grid is low while the digital elevation model data is dense, the ICP algorithm may not find enough matching point pairs, resulting in matching failure or low accuracy.

[0092] In one embodiment, the calculation process of the data density difference coefficient in step S121 may include:

[0093] S1211: Use the Euclidean distance measurement method to determine the nearest neighbor distance of the point cloud data of the initial grid in the point cloud coordinate system and the average nearest neighbor distance of the point cloud as a whole, as well as the basic nearest neighbor distance of the digital elevation model data in the basic terrain coordinate system and the basic average nearest neighbor distance as a whole.

[0094] S1212: Calculate the data density difference coefficient between the point cloud data and the digital elevation model data according to the nearest neighbor distance of the point cloud, the average nearest neighbor distance of the point cloud, the basic nearest neighbor distance, and the basic average nearest neighbor distance.

[0095] In this embodiment, when calculating the data density difference coefficient, the computer device can first use the Euclidean distance measurement method to determine the nearest neighbor distance of the point cloud data of the initial grid in the point cloud coordinate system and the average nearest neighbor distance of the point cloud as a whole, as well as the basic nearest neighbor distance of the digital elevation model data in the basic terrain coordinate system and the basic average nearest neighbor distance as a whole, and then calculate the data density difference coefficient between the point cloud data and the digital elevation model data according to the nearest neighbor distance of the point cloud, the average nearest neighbor distance of the point cloud, the basic nearest neighbor distance, and the basic average nearest neighbor distance.

[0096] Specifically, the computer device can first obtain the position of the point cloud data within the initial grid in the point cloud coordinate system, which is specifically represented as follows:

[0097]

[0098] In the formula, Represent the three-dimensional coordinates of different point cloud data in the point cloud coordinate system; where i = 1, 2, 3, ……, I, I is a positive integer, and i is the number of the point cloud data within the initial grid.

[0099] Next, the computer device can obtain the position of the digital elevation model data within the initial grid in the basic terrain coordinate system, which is specifically represented as follows:

[0100]

[0101] In the formula, are the three-dimensional coordinates of different digital elevation model data in the basic terrain coordinate system; where n = 1, 2, 3, ……, N, N is a positive integer, and n is the number of the digital elevation model data within the initial grid.

[0102] Then, the computer device can use the Euclidean distance metric method to determine the nearest neighbor distance of the point cloud data in the point cloud coordinate system, that is, calculate the distance between each point cloud data within the grid area and other point cloud data, and calculate the overall average nearest neighbor distance of the point cloud data in the point cloud coordinate system. The specific calculation formula is as follows:

[0103]

[0104]

[0105] In the formula, represents the nearest neighbor distance of the point cloud data and all other point cloud data ; is the Euclidean distance between the point cloud data and all other point cloud data ; is the overall average nearest neighbor distance of the point cloud data in the point cloud data coordinate system.

[0106] Similarly, the computer device can also use the Euclidean distance metric method to determine the basic nearest neighbor distance of the digital elevation model data in the basic terrain coordinate system and mark it as , and the overall basic average nearest neighbor distance of the digital elevation model data in the basic terrain coordinate system and mark it as . In addition, the computer device can also obtain the quantities of the point cloud data and the digital elevation model data within the initial grid and mark them as and respectively. Furthermore, based on these data, the data density difference coefficient can be constructed, and the specific formula is as follows:

[0107]

[0108] It can be understood that the larger the data density difference coefficient is, the greater the difference between the number of point cloud data in the initial grid and the number of digital elevation model data in the grid area, and there is a large difference in the degree of dispersion, which may imply that the data distribution in this area is poor. It may be due to problems in the acquisition of point cloud data or the fact that the digital elevation model data fails to fully cover this area. Using the ICP algorithm for point cloud registration may face difficulties. Especially when the data density is inconsistent, the accuracy and robustness of the registration may be affected.

[0109] In one embodiment, the calculation process of the quality feature performance coefficient in step S121 may include:

[0110] S1213: Convert the point cloud data of the initial grid into the basic terrain coordinate system, and determine the initial fitting model according to the conversion result, and, determine the current fitting model according to the digital elevation model data of the initial grid; wherein, the initial fitting model includes an initial error threshold and an initial inlier number.

[0111] S1214: Calculate the distance from the point cloud data to the current fitting model based on the initial fitting model, and determine the current inlier number and the current average error of the current fitting model through the distance and the initial error threshold.

[0112] S1215: When the current inlier number is greater than the initial inlier number, or when the current inlier number is equal to the initial inlier number and the current average error is less than the preset threshold, iteratively update the current fitting model to form the optimal fitting model.

[0113] S1216: Obtain the optimal inlier number, the optimal inlier ratio, and the optimal mean square error of the optimal fitting model, and calculate the quality feature performance coefficient between the point cloud data and the digital elevation model data according to the optimal inlier number, the optimal inlier ratio, and the optimal mean square error.

[0114] In this embodiment, when calculating the quality characteristic performance coefficient, the computer device may convert the point cloud data of the initial grid into the basic terrain coordinate system, determine the initial fitting model according to the conversion result, and determine the current fitting model according to the digital elevation model data of the initial grid. The initial fitting model includes an initial error threshold and an initial inlier number. Then, based on the initial fitting model, calculate the distance from the point cloud data to the current fitting model, and determine the current inlier number and the current average error of the current fitting model through the distance and the initial error threshold. When the current inlier number is greater than the initial inlier number, or when the current inlier number is equal to the initial inlier number and the current average error is less than the preset threshold, iteratively update the current fitting model to form an optimal fitting model. Finally, obtain the optimal inlier number, the optimal inlier ratio, and the optimal mean square error of the optimal fitting model, and calculate the quality characteristic performance coefficient between the point cloud data and the digital elevation model data according to the optimal inlier number, the optimal inlier ratio, and the optimal mean square error.

[0115] It can be understood that this application can determine the difference between the point cloud data and the fitting model of the digital elevation model data according to the RANSAC algorithm, mainly to evaluate the matching effect and fitting accuracy of the data. By randomly selecting a small subset and performing model fitting, maximize the number of inliers (i.e., points that conform to the model), obtain the inlier ratio within the initial grid and the average error of the fitting process, so as to judge the matching effect between the point cloud data and the digital elevation model data.

[0116] It should be noted that the number of inliers reflects the consistency between the fitting model and the data. If the number of inliers is large, it means that the matching effect between the point cloud data and the digital elevation model data is good, that is, most data points conform to the selected fitting model, such as a plane model, a straight line model, or a curve model, etc.

[0117] Among them, a large number of inliers means fewer noise points, and the outliers and outliers in the data set do not dominate, which helps to improve the stability and accuracy of model fitting; a large number of inliers means better data quality, the data itself is not affected by too much error or noise, and the data has high reliability; and the increase in the number of inliers indicates that the selected model can accurately describe the relationship between the point cloud data and the digital elevation model data, so the fitted model is more reliable. A small average error means that the fitting model can accurately describe the relationship between the point cloud data and the digital elevation model data, the data points are at a small distance from the fitting model, indicating that the fitting process is of high precision; a small average error indicates good consistency between the point cloud data and the digital elevation model data, high matching degree between the two, the fitting model can well capture the structural characteristics of the data, and the small error also means fewer noise points in the data, better quality of the point cloud data, and the fitting of the model will not be seriously affected by the noise data points.

[0118] Specifically, the computer device can convert the point cloud data into the basic terrain coordinate system, select an initial fitting model according to the conversion result, determine its initial inlier number, set an initial error threshold for judging whether a point is an inlier, and select the maximum number of iterations N of the RANSAC algorithm and the success probability. .

[0119] Of course, when performing coordinate system conversion on the point cloud data, the coordinate conversion tools that can be used include: tools such as GDAL, Proj4, and PyProj. The computer device can convert the point cloud data and the digital elevation model data under the initial same coordinate system through the coordinate conversion tools. The selection of the fitting model is usually determined through continuous iteration and optimization based on the existing data.

[0120] In the process of calculating the quality characteristic performance coefficient, the computer device can randomly select a small subset from the digital elevation model data, use the selected subset to fit the current fitting model, calculate the distance from the point cloud data to the current fitting model, compare the distance from the point cloud data to the current fitting model with the initial error threshold, obtain the current inlier number and the current inlier ratio of the current fitting model, and obtain the current average error of the current fitting model, that is, the average distance from the inliers to the model. If the current inlier number of the current fitting model is greater than the current inlier number of the previous initial fitting model, or the inlier numbers of the two are equal but the current average error is less than the preset threshold, then update the current fitting model to the optimal model. After multiple iterations, the optimal fitting model is output.

[0121] Furthermore, the computer device can obtain the optimal inlier number of the optimal fitting model according to the output optimal fitting model and mark it as , and obtain the optimal inlier ratio of the model and mark it as , where . At the same time, the computer device can also obtain the optimal mean square error of the optimal fitting model and mark it as . The specific expression of the optimal mean square error is as follows:

[0122]

[0123] In the formula, p = 1, 2, 3, ……, P, P is a positive integer, and p is the number of inliers of the optimal fitting model in the grid area; represents the coordinates of the p-th inlier; represents the coordinates of the optimal fitting model at p; represents the Euclidean distance from the coordinates of the optimal fitting model at p to the coordinates of the p-th inlier.

[0124] And the calculation formula of the quality characteristic performance coefficient The specific representation is as follows:

[0125]

[0126] It can be seen from the above formula that the larger the mass characteristic performance coefficient is, the more the number of inliers and the proportion of inliers in the optimal fitting model within the initial grid are, and the smaller the error between the inliers and the optimal fitting model is, indicating that the fitting effect between the point cloud data and the digital elevation model data within the initial grid is better, the model fitting is more accurate, and the error is smaller, which is more suitable for the application of the ICP algorithm.

[0127] In one embodiment, as Figure 3 shown, Figure 3 is a schematic flow chart of a process for determining terrain information provided by an embodiment of the present application; Figure 3 In, the process of determining the terrain information in step S120 may include:

[0128] S122: For each initial grid, convert the point cloud data of the initial grid into the basic terrain coordinate system to obtain elevation coordinate data at each coordinate position in the point cloud data.

[0129] S123: Calculate the average value, standard deviation, and coefficient of variation of each coordinate position in the point cloud data based on the elevation coordinate data.

[0130] S124: Calculate the terrain irregularity anomaly coefficient of the initial grid according to the average value, standard deviation, and coefficient of variation, and determine the terrain information of the initial grid based on the terrain irregularity anomaly coefficient.

[0131] In this embodiment, in determining the terrain information, for each initial grid, the computer device may convert the point cloud data of the initial grid into the basic terrain coordinate system to obtain elevation coordinate data at each coordinate position in the point cloud data, and then calculate the average value, standard deviation, and coefficient of variation of each coordinate position in the point cloud data based on the elevation coordinate data, so that the terrain irregularity anomaly coefficient of the initial grid can be calculated according to the average value, standard deviation, and coefficient of variation, and the terrain information of the initial grid can be determined based on the terrain irregularity anomaly coefficient.

[0132] It can be understood that the adaptability of the ICP algorithm to the terrain is closely related to the regularity of local terrain changes. Therefore, it is necessary to judge the rationality of the regional division based on the terrain information within the grid area. The error between the point cloud data and the digital elevation model data in the irregular terrain area is large, and the distribution of the point cloud is uneven, which may lead to inaccurate matching points during the matching process.

[0133] Specifically, the computer device can convert the point cloud data into the basic terrain coordinate system, determine the relative elevation at each coordinate position of the point cloud data, and form elevation coordinate data. ; The relative elevation here refers to the height difference of a certain location relative to a reference point, such as a nearby mountain peak, river, or any known terrain point, or a reference plane. By determining the relative elevation within the transmission line corridor, the undulation and irregularity of the terrain in this area can be understood more clearly. Then, the computer device can calculate the average value of the relative elevation at each coordinate position in the point cloud data and the standard deviation , and its expression can be as follows:

[0134]

[0135]

[0136] It should be noted that the standard deviation of the relative elevation at each coordinate position in the point cloud data can help analyze the degree of elevation change in this area and reflect the undulation and irregularity of the terrain. If the standard deviation is small, it means that the elevation in this area is relatively stable, the terrain is relatively flat, and the elevation change of the point cloud data is small. If the standard deviation is large, it means that the terrain in this area changes greatly, with obvious undulations, and the elevation change of the point cloud data is relatively drastic, and it may contain more terrain features such as high peaks and deep valleys.

[0137] In addition, the computer device can also calculate the coefficient of variation of the relative elevation at each coordinate position in the point cloud data based on the elevation coordinate data , and its expression can be as follows:

[0138]

[0139] Furthermore, the computer device can calculate the terrain irregularity anomaly coefficient of the initial grid based on the above calculated average value, standard deviation, and coefficient of variation , and its expression can be as follows:

[0140]

[0141] From the above formula, it can be seen that the larger the terrain irregularity anomaly coefficient, the more irregular the terrain within the initial grid, and there are uneven situations. Therefore, it is necessary to re-divide the initial grid to ensure the accuracy of data processing and analysis.

[0142] In one embodiment, as Figure 4 shown, Figure 4 is a schematic flow chart of the process of constructing a data evaluation model provided by an embodiment of the present application; Figure 4In step S130, the process of comprehensively analyzing the data quality information and terrain information of each initial grid and constructing a data evaluation model based on the analysis results may include:

[0143] S131: Parse the data quality information of each initial grid to obtain the data density difference coefficient and quality characteristic performance coefficient of each initial grid, and, parse the terrain information of each initial grid to obtain the terrain irregularity anomaly coefficient of each initial grid.

[0144] S132: For each initial grid, perform a weighted sum of the data density difference coefficient, quality characteristic performance coefficient, and terrain irregularity anomaly coefficient of this initial grid to obtain a grid data evaluation coefficient.

[0145] S133: Construct a data evaluation model based on the grid data evaluation coefficients of each initial grid.

[0146] In this embodiment, when constructing the data evaluation model, the computer device can parse the data quality information of each initial grid to obtain the data density difference coefficient and quality characteristic performance coefficient of each initial grid, and, parse the terrain information of each initial grid to obtain the terrain irregularity anomaly coefficient of each initial grid. Then, for each initial grid, perform a weighted sum of the data density difference coefficient, quality characteristic performance coefficient, and terrain irregularity anomaly coefficient of this initial grid to obtain a grid data evaluation coefficient. Finally, a data evaluation model can be constructed based on the grid data evaluation coefficients of each initial grid.

[0147] Specifically, the present application can perform a weighted calculation on the data density difference coefficient, quality characteristic performance coefficient, and terrain irregularity anomaly coefficient to obtain a grid data evaluation coefficient, and further construct a data evaluation model for the grid. Among them, the calculation formula of the grid data evaluation coefficient can be specifically expressed as follows:

[0148]

[0149] In the formula, represents the grid data evaluation coefficient of the qth grid area, q = 1, 2, 3..., Q, Q is a positive integer, and q is the number of the initial grid; , , are respectively the proportionality coefficients of the data density difference coefficient, quality characteristic performance coefficient, and terrain irregularity anomaly coefficient, and are all greater than 0.

[0150] It can be understood that, as can be seen from the above formula, the smaller the data density difference coefficient and the terrain irregularity anomaly coefficient, and the larger the quality characteristic performance coefficient, the larger the grid data evaluation coefficient, indicating a higher matching degree between the point cloud data and the digital elevation model data within the initial grid. Further precise matching of the point cloud data and the digital elevation model data using the ICP algorithm in the initial grid will be more efficient and accurate. On the contrary, the larger the data density difference coefficient and the terrain irregularity anomaly coefficient, and the smaller the quality characteristic performance coefficient, the smaller the grid data evaluation coefficient, indicating a lower matching degree between the point cloud data and the digital elevation model data within the initial grid. It is best to re-divide the initial grid and use the ICP algorithm to perform further precise matching of the point cloud data and the digital elevation model data in the updated initial grid.

[0151] In one embodiment, the process of optimizing the grid of the transmission line using the particle swarm optimization algorithm according to the evaluation result in step S140 may include:

[0152] S141: Initialize the particle swarm, and determine the grid data evaluation coefficient of the transmission line according to the evaluation result, so as to determine the objective function of the particle swarm according to the grid data evaluation coefficient.

[0153] S142: Use the velocity update formula and the position update formula to perform iterative calculations on the objective function to update the historical optimal position and the global optimal position of each particle in the particle swarm until the maximum iteration end condition is reached, and output the optimal grid division scheme.

[0154] S143: Optimize the grid of the transmission line according to the optimal grid optimization scheme.

[0155] In this embodiment, when the computer device optimizes the grid of the transmission line, it can first initialize the particle swarm, determine the grid data evaluation coefficient of the transmission line according to the evaluation result, so as to determine the objective function of the particle swarm according to the grid data evaluation coefficient, and then use the velocity update formula and the position update formula to perform iterative calculations on the objective function to update the historical optimal position and the global optimal position of each particle in the particle swarm until the maximum iteration end condition is reached, output the optimal grid division scheme, and finally optimize the grid of the transmission line according to the optimal grid optimization scheme.

[0156] It can be understood that this application can use the particle swarm optimization algorithm to dynamically adjust the division of the initial grid in the transmission line, maximize the terrain evaluation coefficient of each grid, so as to ensure that the grid division adapts to different terrain backgrounds and is conducive to subsequent point cloud matching processing. It should be noted that each particle in the particle swarm represents a grid division scheme.

[0157] Specifically, before grid optimization, the computer device can first initialize the particle swarm, including the size, shape, and number of grids, and set the parameters of the particles, including the number of particles, the maximum number of iterations, the inertia weight, and the learning factor; then define the objective function according to the evaluation results of each initial grid divided by the transmission line , where .

[0158] In addition, the velocity update formula adopted in this application is as follows:

[0159]

[0160] In the formula, represents the velocity vector of the h-th particle at the (t + 1)-th iteration, h = 1, 2, 3,..., H, H is a positive integer, and h is the particle number; ω represents the inertia weight, which is used to control the continuity of the velocity; represents the velocity vector of the h-th particle at the t-th iteration; and represent the learning factors; and represent two random numbers, ranging from [0, 1]; represents the historical optimal position of the h-th particle; represents the position of the h-th particle at the t-th iteration; represents the global optimal position.

[0161] And the position update formula is as follows:

[0162]

[0163] In the formula, represents the new position of the h-th particle at the (t + 1)-th iteration.

[0164] Through the above formulas, the computer device can update the historical optimal position and the global optimal position of each particle, and then repeat the process of updating the particle position and calculating the objective function value until the maximum number of iterations is reached or the change in the objective function value is very small, and finally converge to the optimal grid division scheme.

[0165] Next, the transmission line terrain fusion system provided by the embodiments of this application will be described. The transmission line terrain fusion system described below can be mutually referred to the transmission line terrain fusion method described above.

[0166] In one embodiment, as Figure 5 shown, Figure 5Schematic structural diagram of a transmission line terrain fusion system provided by an embodiment of the present application; the present application also provides a transmission line terrain fusion system, including a grid division module 210, an information determination module 220, a model construction module 230, and a grid optimization module 240, specifically including the following:

[0167] The grid division module 210 is configured to obtain the point cloud data of the point cloud channel terrain of the transmission line and the digital elevation model data of the basic terrain, and divide the overlapping area of the point cloud data and the digital elevation model data in the transmission line into multiple initial grids.

[0168] The information determination module 220 is configured to determine the data quality information and terrain information of each initial grid based on the point cloud data and the digital elevation model data of each initial grid.

[0169] The model construction module 230 is configured to comprehensively analyze the data quality information and terrain information of each initial grid, and construct a data evaluation model according to the analysis results.

[0170] The grid optimization module 240 is configured to evaluate the applicability of the ICP algorithm to each initial grid by using the data evaluation model, and perform grid optimization on the transmission line by using the particle swarm optimization algorithm according to the evaluation results.

[0171] In the above embodiment, when performing terrain data fusion on the transmission line, the point cloud data of the point cloud channel terrain of the transmission line and the digital elevation model data of the basic terrain can be obtained first, and the overlapping area of the point cloud data and the digital elevation model data in the transmission line can be divided into multiple initial grids to ensure the spatial consistency of these two types of data. Then, the data quality information and terrain information of each initial grid can be determined based on the point cloud data and the digital elevation model data of each initial grid to analyze the fusion performance of different data. Next, the data quality information and terrain information of each initial grid can be comprehensively analyzed, so that a data evaluation model can be constructed according to the analysis results, and the applicability of the ICP algorithm to each initial grid can be evaluated by using the data evaluation model to avoid falling into a local optimal solution when dealing with complex and irregular terrains. Finally, the transmission line can be grid-optimized by using the particle swarm optimization algorithm according to the evaluation results. In summary, the present application can automatically adjust the grid division through the particle swarm optimization algorithm to realize the data combination of the point cloud channel terrain of the transmission line and the basic terrain, adapt it to different terrain features, and thus improve the matching effect and analysis efficiency.

[0172] In one embodiment, the grid division module 210 may include:

[0173] The area division sub-module is configured to perform area division in the transmission line according to a preset unit area to obtain a division result.

[0174] A data segmentation sub-module, configured to perform data segmentation on the point cloud data and the digital elevation model data respectively based on the division result, so as to obtain a point cloud sub-region corresponding to the point cloud data and a basic sub-region corresponding to the digital elevation model data.

[0175] A data fusion sub-module, configured to use the grid method to determine the overlapping regions in the point cloud sub-region and the basic sub-region, and perform local fusion analysis on each overlapping region to form an initial grid.

[0176] In one embodiment, the information determination module 220 may include:

[0177] A coefficient calculation sub-module, configured to calculate, for each initial grid, the data density difference coefficient and the quality characteristic performance coefficient between the point cloud data and the digital elevation model data of the initial grid respectively, and determine the data quality information of the initial grid according to the data density difference coefficient and the quality characteristic performance coefficient.

[0178] In one embodiment, the calculation process of the data density difference coefficient in step S121 may include:

[0179] A distance measurement unit, configured to use the Euclidean distance measurement method to determine the point cloud nearest neighbor distance of the point cloud data of the initial grid in the point cloud coordinate system and the point cloud average nearest neighbor distance in the whole, as well as the basic nearest neighbor distance of the digital elevation model data in the basic terrain coordinate system and the basic average nearest neighbor distance in the whole.

[0180] A first coefficient calculation unit, configured to calculate the data density difference coefficient between the point cloud data and the digital elevation model data according to the point cloud nearest neighbor distance, the point cloud average nearest neighbor distance, the basic nearest neighbor distance and the basic average nearest neighbor distance.

[0181] In one embodiment, the calculation process of the quality characteristic performance coefficient in step S121 may include:

[0182] A model fitting unit, configured to convert the point cloud data of the initial grid into the basic terrain coordinate system, determine an initial fitting model according to the conversion result, and determine a current fitting model according to the digital elevation model data of the initial grid; wherein, the initial fitting model includes an initial error threshold and an initial inlier number.

[0183] A data calculation unit, configured to calculate the distance from the point cloud data to the current fitting model based on the initial fitting model, and determine the current inlier number and the current average error of the current fitting model through the distance and the initial error threshold.

[0184] A model update unit, configured to iteratively update the current fitting model to form an optimal fitting model when the current number of inlier points is greater than the initial number of inlier points, or when the current number of inlier points is equal to the initial number of inlier points and the current average error is less than a preset threshold.

[0185] A second coefficient calculation unit, configured to obtain the optimal number of inlier points, the optimal inlier ratio, and the optimal mean square error of the optimal fitting model, and calculate a quality characteristic performance coefficient between the point cloud data and the digital elevation model data based on the optimal number of inlier points, the optimal inlier ratio, and the optimal mean square error.

[0186] In one embodiment, the information determination module 220 may further include:

[0187] A data conversion sub-module, configured to convert the point cloud data of each initial grid into the basic terrain coordinate system to obtain elevation coordinate data at each coordinate position in the point cloud data.

[0188] A coordinate calculation sub-module, configured to calculate the average value, standard deviation, and coefficient of variation of each coordinate position in the point cloud data based on the elevation coordinate data.

[0189] A terrain information calculation sub-module, configured to calculate a terrain irregularity anomaly coefficient of the initial grid based on the average value, standard deviation, and coefficient of variation, and determine the terrain information of the initial grid based on the terrain irregularity anomaly coefficient.

[0190] In one embodiment, the model construction module 230 may include:

[0191] An information parsing sub-module, configured to parse the data quality information of each initial grid to obtain the data density difference coefficient and the quality characteristic performance coefficient of each initial grid, and parse the terrain information of each initial grid to obtain the terrain irregularity anomaly coefficient of each initial grid.

[0192] A weighted summation sub-module, configured to perform a weighted summation of the data density difference coefficient, the quality characteristic performance coefficient, and the terrain irregularity anomaly coefficient of each initial grid to obtain a grid data evaluation coefficient for each initial grid.

[0193] A model construction sub-module, configured to construct a data evaluation model based on the grid data evaluation coefficients of each initial grid.

[0194] In one embodiment, the grid optimization module 240 may include:

[0195] A function determination sub-module, configured to initialize a particle swarm, determine the grid data evaluation coefficient of the transmission line according to the evaluation result, and determine the objective function of the particle swarm according to the grid data evaluation coefficient.

[0196] The solution output sub-module is used to perform iterative calculations on the objective function using the velocity update formula and the position update formula to update the historical optimal position and the global optimal position of each particle in the particle swarm until the maximum iteration end condition is reached, and output the optimal grid division solution.

[0197] The grid optimization sub-module is used to optimize the grid of the transmission line according to the optimal grid optimization solution.

[0198] In one embodiment, the present application also provides a storage medium, in which computer-readable instructions are stored. When the computer-readable instructions are executed by one or more processors, one or more processors are caused to execute the steps of the transmission line terrain fusion method as described in any one of the above embodiments.

[0199] In one embodiment, the present application also provides a computer device, in which computer-readable instructions are stored. When the computer-readable instructions are executed by one or more processors, one or more processors are caused to execute the steps of the transmission line terrain fusion method as described in any one of the above embodiments.

[0200] Schematically, as Figure 6 shown, Figure 6 is an internal structure schematic diagram of a computer device provided by an embodiment of the present application. The computer device 300 can be provided as a server. Referring to Figure 4 , the computer device 300 includes a processing component 302, which further includes one or more processors, and memory resources represented by a memory 301 for storing instructions executable by the processing component 302, such as application programs. The application programs stored in the memory 301 can include one or more modules each corresponding to a set of instructions. In addition, the processing component 302 is configured to execute instructions to perform the transmission line terrain fusion method of any of the above embodiments.

[0201] The computer device 300 may further include a power supply component 303 configured to perform power management of the computer device 300, a wired or wireless network interface 304 configured to connect the computer device 300 to a network, and an input / output (I / O) interface 305. The computer device 300 can operate based on an operating system stored in the memory 301, such as WindowsServer TM, Mac OS XTM, Unix TM, Linux TM, Free BSDTM or the like.

[0202] Those skilled in the art can understand, Figure 6The structure shown is only a block diagram of some of the structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this 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 component layout.

[0203] Finally, it should also be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0204] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0205] The above description of the disclosed embodiments enables those skilled in the art to implement or use this application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for fusion of power transmission line terrain, characterized in that: The method comprises: Acquire point cloud data of the point cloud channel terrain of the transmission line and digital elevation model data of the basic terrain, and divide the overlapping area of ​​the point cloud data and the digital elevation model data in the transmission line into a plurality of initial grids; Determine data quality information and terrain information of each initial grid based on point cloud data and digital elevation model data of each initial grid; Comprehensively analyze the data quality information and terrain information of each initial grid, and construct a data evaluation model based on the analysis results; The data evaluation model is used to evaluate the suitability of the ICP algorithm for each initial grid, and the particle swarm optimization algorithm is used to optimize the grid of the transmission line according to the evaluation results.

2. The method for fusion of power transmission line terrain according to claim 1, characterized in that: The step of dividing the overlapping area of ​​the point cloud data and the digital elevation model data in the transmission line into a plurality of initial grids comprises: Performing area division in the transmission line according to preset unit areas to obtain a division result; Based on the division results, the point cloud data and the digital elevation model data are respectively segmented to obtain a point cloud sub-region corresponding to the point cloud data and a basic sub-region corresponding to the digital elevation model data; A grid method is used to determine the overlapping area between the point cloud sub-area and the basic sub-area, and a local fusion analysis is performed on each overlapping area to form an initial grid.

3. The method for fusion of power transmission line terrain according to claim 1, characterized in that: The step of determining the data quality information of each initial grid based on the point cloud data and the digital elevation model data of each initial grid includes: For each initial grid, the data density difference coefficient and the quality feature expression coefficient between the point cloud data and the digital elevation model data of the initial grid are calculated respectively, and the data quality information of the initial grid is determined according to the data density difference coefficient and the quality feature expression coefficient.

4. The method for merging power transmission line terrain according to claim 3, characterized in that: The calculation process of the data density difference coefficient includes: The Euclidean distance measurement method is used to determine the point cloud nearest neighbor distance of the point cloud data of the initial grid in the point cloud coordinate system and the overall average nearest neighbor distance of the point cloud, as well as the basic nearest neighbor distance of the digital elevation model data in the basic terrain coordinate system and the overall basic average nearest neighbor distance; The data density difference coefficient between the point cloud data and the digital elevation model data is calculated according to the point cloud nearest neighbor distance, the point cloud average nearest neighbor distance, the basic nearest neighbor distance and the basic average nearest neighbor distance.

5. The method for merging power transmission line terrain according to claim 3, characterized in that: The calculation process of the quality characteristic performance coefficient includes: The point cloud data of the initial grid is converted into a basic terrain coordinate system, and an initial fitting model is determined according to the conversion result, and a current fitting model is determined according to the digital elevation model data of the initial grid; wherein the initial fitting model includes an initial error threshold and an initial number of inliers; Calculating the distance from the point cloud data to the current fitting model based on the initial fitting model, and determining the current number of inliers and the current average error of the current fitting model through the distance and the initial error threshold; When the current number of inliers is greater than the initial number of inliers, or the current number of inliers is equal to the initial number of inliers and the current average error is less than a preset threshold, iteratively updating the current fitting model to form an optimal fitting model; The optimal number of inliers, the optimal ratio of inliers and the optimal mean square error of the optimal fitting model are obtained, and the quality feature performance coefficient between the point cloud data and the digital elevation model data is calculated according to the optimal number of inliers, the optimal ratio of inliers and the optimal mean square error.

6. The method for merging power transmission line terrain according to claim 1, characterized in that: The process of determining the terrain information includes: For each initial grid, the point cloud data of the initial grid is converted into a basic terrain coordinate system to obtain elevation coordinate data at each coordinate position in the point cloud data; Calculate the mean value, standard deviation and coefficient of variation of each coordinate position in the point cloud data based on the elevation coordinate data; The terrain irregularity anomaly coefficient of the initial grid is calculated according to the mean value, the standard deviation and the coefficient of variation, and the terrain information of the initial grid is determined based on the terrain irregularity anomaly coefficient.

7. The method for merging power transmission line terrain according to claim 1, characterized in that: The data quality information and terrain information of each initial grid are comprehensively analyzed, and a data evaluation model is constructed based on the analysis results, including: The data quality information of each initial grid is analyzed to obtain the data density difference coefficient and the quality characteristic performance coefficient of each initial grid, and the terrain information of each initial grid is analyzed to obtain the terrain irregular anomaly coefficient of each initial grid; For each initial grid, the data density difference coefficient, quality feature performance coefficient and terrain irregular anomaly coefficient of the initial grid are weighted and summed to obtain the grid data evaluation coefficient; A data evaluation model is constructed based on the grid data evaluation coefficients of each initial grid.

8. The method for merging power transmission line terrain according to claim 1, characterized in that: The step of optimizing the grid of the transmission line using a particle swarm optimization algorithm according to the evaluation result includes: Initializing a particle swarm, and determining a grid data evaluation coefficient of the transmission line according to an evaluation result, so as to determine an objective function of the particle swarm according to the grid data evaluation coefficient; Iteratively calculating the objective function using a velocity update formula and a position update formula to update the historical optimal position and the global optimal position of each particle in the particle swarm until the maximum iteration end condition is reached, and outputting the optimal grid division scheme; The transmission line is grid optimized according to the optimal grid optimization scheme.

9. A transmission line terrain fusion system, characterized in that: include: A grid division module, used to obtain point cloud data of the point cloud channel terrain of the transmission line and digital elevation model data of the basic terrain, and divide the overlapping area of ​​the point cloud data and the digital elevation model data in the transmission line into a plurality of initial grids; An information determination module, for determining data quality information and terrain information of each initial grid based on point cloud data and digital elevation model data of each initial grid; The model building module is used to conduct a comprehensive analysis of the data quality information and terrain information of each initial grid, and to construct a data evaluation model based on the analysis results; The grid optimization module is used to evaluate the applicability of the ICP algorithm on each initial grid using the data evaluation model, and to optimize the grid of the transmission line using the particle swarm optimization algorithm according to the evaluation results.

10. A storage medium, characterized in that: The storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the transmission line terrain fusion method as described in any one of claims 1 to 8.

11. A computer device, characterized in that: include: one or more processors, and memory; The memory stores computer-readable instructions, and when the computer-readable instructions are executed by the one or more processors, the steps of the transmission line terrain fusion method as claimed in any one of claims 1 to 8 are performed.

Images