A method for correcting large-scale road terrain in a three-dimensional geographic information system

Abstract

The present invention discloses a method for correcting road terrain in a large scene of a three-dimensional geographic information system, including: obtaining a road center vector, a road vector surface, and a road slope protection vector surface; superimposing the road center vector with terrain data to generate a three-dimensional center vector; smoothing and correcting the three-dimensional center vector to obtain a smooth three-dimensional center vector; generating a three-dimensional road vector surface with the corrected three-dimensional center vector as a reference; superimposing the road slope protection vector surface with terrain data to generate a three-dimensional road slope protection vector surface; correcting the terrain data with the three-dimensional road vector surface and the three-dimensional road slope protection vector surface as a reference; superimposing the road corrected terrain with the terrain data to cover the terrain that needs to be corrected in the terrain data to obtain a terrain correction result. The road terrain correction method according to the present invention can solve the problems of road tilt, undulation, and distortion caused by the inconsistency between the terrain and the image in the data source and time phase in the three-dimensional scene of geographic information, and significantly improve the visualization effect of the three-dimensional geographic information system.

Description

Technical Field

[0001] This patented technology relates to the field of geographic information, and in particular to a three-dimensional geographic information terrain correction method. Background Art

[0002] When producing terrain data, road characteristics are not considered at scale.

[0003] It is difficult and costly to update terrain data, and its update speed is slow compared to imagery.

[0004] Geographic information is closely related to human production and life, and many human activities are inseparable from terrain. How to better express the real terrain for human practical activities has always been a goal pursued by people. In the construction of geographic information three-dimensional scenes, due to the inconsistency of terrain data and image data in data source and time phase, the image and terrain often do not match. The roads on the image show inclination, undulation, distortion and other phenomena that are inconsistent with the actual state in the geographic information three-dimensional scene, which seriously affects the visualization effect and user experience of the three-dimensional geographic information system.

[0005] At present, the terrain correction methods mainly include two methods: human-computer interactive editing and Delaunay triangulation construction based on feature constraints.

[0006] Through human-computer interactive editing, terrain data and image data are matched. When editing, it is necessary to combine the actual situation of the editing area and the surrounding terrain, and the three-dimensional scene and the underlying data need to be updated in linkage to display the editing effect in real time. He Zhonghuan et al. (He Zhonghuan, Huang Qingsong. Research on the correction method of complex mountain terrain based on Skyline [J]. Geospatial Information, 2013, 11(6):18-19.) from the Chongqing Institute of Surveying and Mapping of the National Administration of Surveying, Mapping and Geoinformation conducted a study on the correction method of complex mountain terrain based on the Skyline software platform. Irregular polygons were constructed as the editing area in the Skyline platform, and auxiliary tools were developed using the ArcEngine development library to achieve the cutting, reading and correction of terrain data, taking into account the smooth transition between the editing area and the surrounding terrain. The modified results were displayed in real time in the Skyline platform, providing users with a basis for judging whether the modification is correct and providing a reference for the next modification. For the modified area, the direct representation of the data is a matrix of a certain size. The points in the matrix are thinned out to obtain an appropriate number of elevation points, which are used to form a Delaunay triangulation. Then, in Skyline, the Delaunay triangulation is drawn one by one using the terrain excavation function. The number of triangles will affect the drawing time, so its number needs to be controlled.

[0007] The Delaunay triangulation based on feature constraints is currently the most commonly used terrain correction method, which is to convert terrain data into Delaunay triangulation to create a terrain model, embed the feature boundaries of the object into the terrain model as features, and reconstruct the triangulation locally to complete the fusion of the model and the terrain, ensuring that the model and the terrain are closely linked at the junction. Yao Xiaofei et al. from the PLA Information Engineering University (Yao Xiaofei. Large-scale terrain transformation based on Delaunay triangulation [D]. PLA Information Engineering University, 2008.) analyzed the shortcomings of large-scale virtual terrain environments and summarized the technical methods of virtual terrain environment modeling. In view of the shortcomings of textures replacing objects, the inability to truly integrate terrain and objects, and the lack of realistic scene performance in large-scale virtual terrain environments, Delaunay triangulation was used as the terrain modeling method. The features of the objects that have a transforming effect on the terrain were extracted according to the classification of points, lines, and surfaces, and the features of the objects were inserted into the terrain as constraint lines or polygons. When the elevation of the object does not match the terrain and the terrain needs to adapt to the object, a terrain correction function is constructed according to the object points to correct the terrain. The meshing algorithm uses point-by-point insertion, which is not efficient enough and needs to be further improved. Only the transformation of a single terrain is discussed, and the terrain correction algorithm for massive hierarchical and block data across terrain blocks has not been studied. Jiang Yu et al. from Southwest Jiaotong University (Jiang Yu, Li Yongshu, Lu Heng, Zhang Shuaiyi. Research on the matching method of objects and terrain in three-dimensional scene modeling [J]. Bulletin of Surveying and Mapping, 2013(05):96-100.) aimed at the problem of the processing method of matching objects and terrain in three-dimensional scene modeling, analyzed the causes of the mismatch between objects and terrain in three-dimensional scene modeling, and proposed an improved method to solve the matching of objects and terrain. By embedding the feature constraints of objects in the Delaunay triangulation, seamless matching of objects and terrain is achieved; by local terrain correction, correction values ​​are added to the elevation of the terrain influence points, the sudden change of the terrain after matching is weakened, and the terrain transition is natural. The research and experimental results show that this method can be well applied to the matching of surface objects and terrain in precise three-dimensional scene modeling.

[0008] The above methods can basically correct terrain data for specific work scenes. However, the terrain correction method based on human-computer interactive editing has a low degree of automation, slow speed, and high requirements for the working environment, and cannot meet the terrain correction work in large-scale three-dimensional scene production; the terrain correction method based on the Delaunay triangulation constructed by feature constraints needs to convert terrain data into triangulated networks. In order to obtain the original terrain format, the corrected triangulated network must be interpolated into the original terrain format. In order to improve the calculation speed, it is sometimes necessary to thin the terrain data. The process of data thinning and triangulation interpolation loses certain terrain details to a certain extent. The amount of terrain data used in the construction of geographic information three-dimensional scenes is large, and standard tiling is often used for storage and management. The number of files is as high as thousands, and there is a certain overlap in the terrain data of adjacent sheets. The above methods do not involve terrain data repair technology for large data volumes and multi-file storage, and how to ensure that the overlapping areas of terrain data of adjacent sheets maintain data consistency after terrain correction. Summary of the invention

[0009] Based on this, it is necessary to provide a new terrain correction method that can solve the above problem.

[0010] A method for correcting large-scene road terrain in a three-dimensional geographic information system is provided to achieve the purpose of the present invention, characterized in that it includes the following steps:

[0011] Obtain the road center vector, road vector surface, and road slope protection vector surface;

[0012] Get the combined table of topographic data;

[0013] The road center vector is superimposed with the terrain data to generate a three-dimensional center vector;

[0014] The three-dimensional center vector is smoothed and corrected to obtain a smooth three-dimensional center vector;

[0015] Correcting the terrain according to the smoothed three-dimensional central vector node elevation, the road vector surface, and the road slope protection vector surface to obtain a road corrected terrain;

[0016] The road correction terrain is superimposed on the terrain data to cover the terrain that needs to be corrected in the terrain data to obtain a terrain correction result.

[0017] As an implementable method for correcting road terrain in a large scene of a three-dimensional geographic information system, the road center vector is superimposed with terrain data to generate a three-dimensional center vector, including the following steps:

[0018] Loop through each vector object in the road center vector file, encrypt the nodes of the current vector object, ensure that the distance between adjacent nodes is less than a given threshold, and obtain the node encrypted center line;

[0019] Extract the encrypted centerline nodes, number the nodes, record the corresponding relationship between the nodes and the vector objects and the nodes in the vector objects, and obtain a node vector file;

[0020] Loop through each node in the node vector file, calculate the terrain file corresponding to the current node according to the spatial position relationship between the current node coordinates and the terrain data map sheet combination table, and interpolate and extract the elevation coordinates of the current node in the terrain file according to the spatial coordinates of the current node;

[0021] According to the correspondence between the node and the vector object and the nodes in the vector object, the elevation coordinates of the current node are assigned to the nodes of the vector object to obtain a three-dimensional center vector.

[0022] As an implementable method for correcting large-scene road terrain in a three-dimensional geographic information system, the three-dimensional center vector is smoothly corrected to obtain a smooth three-dimensional center vector, comprising the following steps:

[0023] Taking the accumulated distance from the 3D center vector node to the starting point of the 3D center vector as X and the elevation of the 3D center vector node as Y, a 3D center section vector is created, and the 3D center vector section nodes are numbered, and the corresponding relationship with the 3D center vector nodes is recorded;

[0024] The three-dimensional central section vector is processed based on the Douglas-Peucker algorithm, and some nodes are selected from the three-dimensional central section vector nodes to control the overall fluctuation trend of the three-dimensional central section vector. The selected nodes are called feature points, and the unselected nodes are called non-feature points. The elevation of the feature points remains unchanged, and the elevation of the non-feature points is linearly interpolated using the two nearest feature points before and after. Then, the three-dimensional central section vector is smoothed to obtain a three-dimensional central section vector with a reasonable fluctuation trend;

[0025] According to the correspondence between the three-dimensional central vector section node and the three-dimensional central vector node, the elevation of the three-dimensional central section vector node with a reasonable undulation trend is assigned to the corresponding three-dimensional central vector node to obtain a three-dimensional central vector with a reasonable undulation trend.

[0026] As an implementable method for correcting road terrain in a large scene of a three-dimensional geographic information system, the terrain is corrected according to the three-dimensional center vector with a reasonable trend and the road vector surface and the road slope protection vector surface to obtain the corrected terrain, including the following steps:

[0027] The road vector surface nodes are encrypted to ensure that the distance between adjacent nodes is less than a given threshold, thereby obtaining an encrypted road vector surface;

[0028] Draw a perpendicular line from the node of the encrypted road vector surface to the three-dimensional central vector with a reasonable undulating trend, interpolate the perpendicular foot elevation, assign the elevation to the node of the encrypted road vector surface, and obtain a three-dimensional road vector surface;

[0029] The road slope protection vector surface nodes are encrypted to ensure that the distance between adjacent nodes is less than a given threshold, thereby obtaining an encrypted road slope protection vector surface;

[0030] The encrypted road slope protection vector surface is superimposed on the terrain data to generate a three-dimensional road slope protection vector surface, and the method is consistent with the method of superimposing the road center vector and the terrain data to generate a three-dimensional center vector;

[0031] A TIN is constructed based on the three-dimensional road vector surface and the three-dimensional road slope protection vector surface, and the TIN is converted into a raster to obtain a road correction terrain.

[0032] As an implementable method for correcting road terrain in a large scene of a three-dimensional geographic information system, the road corrected terrain is superimposed on the original terrain to cover the original terrain, and the road terrain repair is completed, including the following steps:

[0033] According to the spatial position relationship between the road-corrected terrain and the terrain data map sheet combination table, the map sheet numbers of all terrain data overlapping with the road-corrected terrain are searched, and a map number list is established.

[0034] The map number list is traversed in a loop, the road correction terrain is superimposed with the terrain data corresponding to the current map number to obtain the intersection, and the terrain data is covered with the road correction terrain in the area with the intersection, thereby completing the road terrain repair.

[0035] Compared with the prior art, the present invention has the following advantages: by encrypting characteristic vector nodes, extracting vector nodes, directly using the original format of terrain data for interpolation to obtain the required elevation, there is no need to convert the terrain data into a Delaunay triangulation in advance, thus avoiding the workload and unnecessary precision loss caused by format conversion; directly repairing the terrain data stored and managed in sections, there is no need to splice the terrain data together in advance and then cut it according to the correction range, and the correction result can ensure the consistency of data in overlapping areas of terrain data of different sheets; the algorithm has low complexity, is easy to program and implement, and has cyclic batch processing and a high degree of automation. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a flow chart of a specific embodiment of a method for correcting large-scene road terrain in a three-dimensional geographic information system according to the present invention;

[0037] FIG2 (a) is a schematic diagram of the effect of three-dimensional road vector smoothing before a specific embodiment of a method for correcting road terrain in a large scene of a three-dimensional geographic information system according to the present invention;

[0038] FIG2( b ) is a schematic diagram of the effect of three-dimensional road vector smoothing in a specific embodiment of a method for correcting large-scene road terrain in a three-dimensional geographic information system of the present invention;

[0039] Figure 3 Schematic diagram of Douglas-Peucker algorithm;

[0040] Figure 4 A schematic diagram of a terrain constructed by characteristic lines in a specific embodiment of a method for correcting a large-scene road terrain in a three-dimensional geographic information system according to the present invention;

[0041] FIG5 (a) is a schematic diagram of the surface morphology of a road before terrain restoration in a specific embodiment of a method for correcting road terrain in a large scene of a three-dimensional geographic information system according to the present invention;

[0042] FIG5( b ) is a schematic diagram of the surface morphology of a road terrain after restoration in a specific embodiment of a method for correcting a road terrain in a large scene of a three-dimensional geographic information system according to the present invention;

[0043] FIG6 (a) is a schematic diagram of the effect of road terrain before correction in a specific embodiment of a method for correcting road terrain in a large scene of a three-dimensional geographic information system according to the present invention;

[0044] FIG6( b ) is a schematic diagram of the effect of road terrain correction in a specific embodiment of a method for correcting road terrain in a large scene of a three-dimensional geographic information system according to the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the specific implementation of the embodiments of the present invention is described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the described 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.

[0046] Unless otherwise defined, the technical or scientific terms used herein shall have the general meaning understood by a person with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the patent application, specification and claims of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "a" do not indicate a quantity limitation, but indicate the existence of at least one.

[0047] According to an embodiment of the present invention, a method for correcting large-scale road terrain in a three-dimensional geographic information system is provided. Figure 1 As shown, the following steps are included:

[0048] S100, obtaining a road center vector, a road vector surface, and a road slope protection vector surface.

[0049] All vectors mentioned in the embodiments of the present invention are stored in the ESRI Shapefile (hereinafter referred to as: shp) file format. SHP is a vector graphics format that can save the spatial position of geometric figures, and comes with a two-dimensional table for storing the attribute information of each geometric body in SHP. Many geometric bodies in SHP can represent complex geographical objects and provide them with powerful and accurate computing capabilities. In ordinary use, SHP usually contains points, polylines and polygons. Shapes with Z coordinates are three-dimensional, and shapes with M values ​​contain a user-specified measurement value, which is defined on each point coordinate.

[0050] S110, the road center vector is the existing basic data, which is a vector polyline stored in the shp format, and the attribute field information records the width and type (bridge, tunnel, etc.) of the road.

[0051] S120, the road vector surface is obtained by taking the road center vector as a reference and using the width attribute of the road center vector as a distance to make a buffer zone.

[0052] Buffer:

[0053] A buffer zone is a kind of influence range or service range of a geographic spatial target (including point targets, line targets, and surface targets). Specifically, it refers to a polygon with a certain width that is automatically established around a point, line, or surface entity. The basic idea of ​​buffer zone generation is to determine its domain given a spatial target, and the domain size is determined by the domain radius. The buffer zone of a spatial target is:

[0054]

[0055] In the formula, is the domain radius, i.e. the buffer distance; Generally refers to the minimum Euclidean distance, but can also be other defined distances; for The buffer distance is The buffer is to Distance Less than or equal to The set of all points of .

[0056] S130, the road slope protection vector surface is obtained by taking the road vector surface as a reference and taking the slope protection width (here uniformly set to 20 meters according to actual conditions) as a buffer distance to make a buffer zone.

[0057] S200, obtaining a terrain data sheet combination table.

[0058] The terrain data used in the embodiment of the present invention is a 1:50,000 digital elevation model DEM, the grid size is 5 meters, the data format is ASCII format, and it is stored and managed according to the 1:50,000 standard, and named according to the sheet number. The method for creating a terrain data sheet combination table is: loop reading each DEM, obtaining the four-range range of the DEM, that is, the coordinates of the four corner points, creating a shp vector surface according to the corner point coordinates, adding a Name attribute field, and the field value is the sheet number used.

[0059] S300, the road center vector is superimposed with the terrain data to generate a three-dimensional center vector.

[0060] S310, in order to ensure that the spatial form of the three-dimensional center vector generated by superimposing the road center vector and the terrain data is closely matched with the terrain data and can truly reflect the true form of the terrain along the road center vector, the road center vector is node-encrypted with the terrain data grid size (here 5 meters) as a threshold, and the distance between adjacent nodes is ensured to be less than a given threshold, thereby obtaining a node-encrypted center line;

[0061] S320, extracting the node coordinates of the node encryption center line, creating a node vector file, numbering the nodes, and recording the corresponding relationship between the nodes and the vector objects and the nodes in the vector objects;

[0062] S330, looping through each node in the node vector file, calculating the terrain file corresponding to the current node according to the spatial position relationship between the current node coordinates and the terrain data map sheet combination table, and interpolating and extracting the elevation coordinates of the current node in the terrain file according to the spatial coordinates of the current node;

[0063] S340, assigning the elevation coordinates of the current node to the nodes of the vector object according to the correspondence between the node and the vector object and the nodes in the vector object, to obtain a three-dimensional center vector.

[0064] S400, the three-dimensional center vector is smoothed and corrected to obtain a smooth three-dimensional center vector.

[0065] S410, taking the accumulated distance from the 3D center vector node to the starting point of the 3D center vector as X and the elevation of the 3D center vector node as Y, creating a 3D center section vector, numbering the 3D center vector section nodes, storing them in the shp file as the M value of the shp file node, recording the corresponding relationship with the 3D center vector node, and creating a 3D center section vector;

[0066] S420, processing the three-dimensional central section vector based on the Douglas-Peucker algorithm, selecting some nodes from the three-dimensional central section vector nodes to control the overall fluctuation trend of the three-dimensional central section vector, the selected nodes are called feature points, and the unselected nodes are called non-feature points, the elevation of the feature points remains unchanged, and the elevation of the non-feature points is linearly interpolated using the two closest feature points, and then the three-dimensional central section vector is smoothed. The smoothing algorithm used here is a piecewise cubic polynomial interpolation method, and a three-dimensional central section vector with a reasonable fluctuation trend is obtained after smoothing;

[0067] S430, according to the correspondence between the three-dimensional central vector section node and the three-dimensional central vector node, the elevation of the three-dimensional central section vector node with a reasonable undulation trend is assigned to the corresponding three-dimensional central vector node to obtain a three-dimensional central vector with a reasonable undulation trend; the comparison of the three-dimensional central vector before and after smoothing correction is shown in Figure 2 (a) and Figure 2 (b).

[0068] Douglas-Peucker algorithm:

[0069] The Douglas-Peukcer algorithm was proposed by D. Douglas and T. Peueker in 1973, referred to as the DP algorithm. It is currently recognized as a classic algorithm for simplifying linear features. A considerable number of existing line simplification algorithms are improved on the basis of this algorithm. Its advantage is that it is translationally and rotationally invariant. After the curve and threshold are given, the sampling result is certain.

[0070] The basic idea of ​​the algorithm is: connect a straight line between the first and last points of each curve, calculate the distance between all points and the straight line, and find the maximum distance value dmax, and compare dmax with the limit difference D: if dmax < D, all the middle points on this curve are discarded; if dmax ≥ D, keep the coordinate point corresponding to dmax, and use this point as the boundary to divide the curve into two parts, and repeat this method for these two parts.

[0071] The detailed steps of the algorithm are as follows:

[0072] A. Draw a straight line between the first and last points of the curve, and find the distances from the remaining points to the straight line. Figure 3 Middle (a).

[0073] B. Select the largest one and compare it with the threshold. If it is greater than the threshold, the point with the largest distance from the straight line is retained, otherwise all points between the two end points of the straight line are discarded. Figure 3 In (b), point 4 is retained.

[0074] C. Divide the known curve into two parts according to the retained points, repeat steps 1 and 2, and iterate, that is, still select the one with the largest distance to compare with the threshold, and select and discard in turn until there is no point to discard, and finally obtain the coordinates of the curve points that meet the given accuracy limit, such as Figure 3 Middle (c), Figure 3 In (d), keep the 6th and 7th points respectively and discard the other points to complete the simplification of the line.

[0075] Piecewise cubic polynomial interpolation:

[0076] The piecewise cubic polynomial interpolation method, also known as the five-point method, has the following basic ideas:

[0077] A. Establish a cubic polynomial curve equation between every two data points;

[0078] B. The curve has a continuous first-order derivative;

[0079] C. The derivative of each node is determined by taking one point as the center and two adjacent points on the left and right sides, a total of five points.

[0080] According to condition C, we can assume that the first-order derivative of the curve at each point has been obtained (the method is described later). Then, between any two adjacent discrete points, we can obtain the following four equations:

[0081] Let the curve equation be , the derivatives of each point have been found to be .

[0082]

[0083] set up , then the above equations can be used to find , , , The value of , that is, the function expression between the two points.

[0084] Determine the first-order derivative of a point based on the two points before and after it:

[0085] Figure 4 Middle (a) Middle is the point where the first-order derivative is to be taken, AB is the tangent line of the curve, and from the AKIMA condition we can get:

[0086]

[0087] It can be found The first derivative of a point.

[0088] For the first point and the second point, as well as the second to last point and the first point, the method of filling points is used to find their derivatives. The filling points are filled using the principle of equal increments, such as Figure 4 As shown in (b), For the first point, For the second point, For the third point, if I ask The derivative of dot, so that there are two dots in front of it. and The coordinates of are known, then: Increment - Increment Increment - Incremental replenishment Point B, similarly fill in point B, so , The first-order derivative of can be found using the five-point method.

[0089] S500, the terrain is corrected according to a three-dimensional central vector with a reasonable undulation trend, a road vector surface, and a road slope protection vector surface.

[0090] The terrain is corrected according to the three-dimensional central vector with a reasonable undulating trend, the road vector plane, and the road slope protection vector plane to obtain a corrected terrain, including the following steps:

[0091] S510, the road vector surface nodes are encrypted to ensure that the distance between adjacent nodes is less than a given threshold, and an encrypted road vector surface is obtained. The purpose and method of node encryption are the same as those of the road center vector node encryption;

[0092] S520, drawing a perpendicular line from the encrypted road vector surface node to the three-dimensional central vector with a reasonable trend, interpolating the perpendicular foot elevation, assigning the elevation to the encrypted road vector surface node, and obtaining a three-dimensional road vector surface;

[0093] S530, the road slope protection vector surface nodes are encrypted to obtain the road slope protection encrypted vector surface, and the node encryption purpose and method are the same as the road center vector node encryption;

[0094] S540, the road slope protection encrypted vector surface is superimposed with the terrain data to generate a three-dimensional center vector, the method is the same as the method of superimposing the road center vector with the terrain data to generate a three-dimensional center vector;

[0095] S550, the three-dimensional road vector surface and the three-dimensional road slope protection vector surface are used to create an irregular triangulated network (TIN), and the TIN is converted into a raster terrain to obtain a road correction terrain. The comparison of the road terrain before and after correction is shown in Figure 5 (a) and Figure 5 (b).

[0096] S600: The road correction terrain is superimposed on the original terrain to cover the original terrain.

[0097] The road correction terrain is superimposed on the original terrain to cover the original terrain, completing the road terrain repair, including the following steps:

[0098] S610, according to the spatial position relationship between the road-corrected terrain and the terrain data map sheet combination table, search for the map sheet numbers of all terrain data overlapping with the road-corrected terrain, and create a map number list.

[0099] S620, loop through the list of map numbers, superimpose the road correction terrain with the terrain data corresponding to the current map number to obtain the intersection, and use the road correction terrain to cover the terrain data in the intersection area to complete the road terrain repair. The comparison of the effect before and after the road terrain repair is shown in Figure 6 (a) and Figure 6 (b).

Claims

1. A method for correcting large-scale road terrain in a three-dimensional geographic information system, characterized in that: The following steps are involved: Obtain the road center vector, road vector surface, and road slope protection vector surface; Get the combined table of topographic data; The road center vector is superimposed with the terrain data to generate a three-dimensional center vector; specifically, the method includes: looping through each vector object in the road center vector file, encrypting the nodes of the current vector object, ensuring that the distance between adjacent nodes is less than a given threshold, and obtaining a node encryption center line; extracting the node coordinates of the encrypted center line to create a node vector file, numbering the nodes, and recording the corresponding relationship between the nodes and the vector objects and the nodes in the vector objects; looping through each node in the node vector file, calculating the terrain data corresponding to the current node according to the spatial position relationship between the current node coordinates and the map sheet combination table of the terrain data, and extracting the elevation coordinates of the current node by interpolating in the terrain data according to the spatial coordinates of the current node; assigning the elevation coordinates of the node to the node of the vector object according to the corresponding relationship between the node and the vector object and the nodes in the vector object, and obtaining a three-dimensional center vector; The three-dimensional center vector is smoothed and corrected to obtain a smooth three-dimensional center vector; wherein the smooth three-dimensional center vector is a three-dimensional center vector with a reasonable trend; specifically comprising: taking the accumulated distance from the three-dimensional center vector node to the starting point of the three-dimensional center vector as X, and the elevation of the three-dimensional center vector node as Y, creating a three-dimensional center profile vector, and numbering the three-dimensional center vector profile nodes, and recording the corresponding relationship with the three-dimensional center vector nodes; processing the three-dimensional center profile vector based on the Douglas-Peucker algorithm, selecting some nodes from the three-dimensional center profile vector nodes to control the overall fluctuation trend of the three-dimensional center profile vector, the selected nodes are called feature points, and the unselected nodes are called non-feature points, the elevation of the feature points remains unchanged, and the elevation of the non-feature points is linearly interpolated using the nearest two feature points before and after, and then smoothing the three-dimensional center profile vector to obtain a three-dimensional center profile vector with a reasonable fluctuation trend; according to the corresponding relationship between the three-dimensional center vector profile nodes and the three-dimensional center vector nodes, assigning the node elevation of the three-dimensional center profile vector with a reasonable fluctuation trend to the corresponding node of the three-dimensional center vector to obtain a three-dimensional center vector with a reasonable trend; Correcting the terrain data according to the smoothed three-dimensional central vector node elevation, the road vector surface, and the slope protection vector surface to obtain a road corrected terrain; The road correction terrain is superimposed on the terrain data to cover the terrain that needs to be corrected in the terrain data to obtain a terrain correction result.

2. The method for correcting large-scene road terrain in a three-dimensional geographic information system according to claim 1, characterized in that: The terrain data is corrected according to the smoothed three-dimensional central vector node elevation, the road vector surface, and the slope protection vector surface to obtain a road corrected terrain, including the following steps: The road vector surface nodes are encrypted to ensure that the distance between adjacent nodes is less than a given threshold, thereby obtaining an encrypted road vector surface; Draw a perpendicular line from the encrypted road vector surface node to the three-dimensional central vector with a reasonable undulating trend, interpolate the perpendicular foot elevation, assign the elevation to the encrypted road vector surface node, and obtain a three-dimensional road vector surface; The road slope protection vector surface nodes are encrypted to ensure that the distance between adjacent nodes is less than a given threshold, thereby obtaining an encrypted road slope protection vector surface; The encrypted road slope protection vector surface is superimposed on the terrain data to generate a three-dimensional road slope protection vector surface, and the method is consistent with the method of superimposing the road center vector and the terrain data to generate a three-dimensional center vector; A TIN is constructed based on the three-dimensional road vector surface and the three-dimensional road slope protection vector surface, and the TIN is converted into a raster to obtain a road correction terrain.

3. The method for correcting large-scene road terrain in a three-dimensional geographic information system according to claim 1, characterized in that: The road correction terrain is superimposed on the terrain data to cover the terrain that needs to be corrected in the terrain data to obtain a terrain correction result, including the following steps: According to the spatial position relationship between the road-corrected terrain and the map sheet combination table of the terrain data, the map sheet numbers of all terrain data overlapping with the road-corrected terrain are searched, and a map number list is established; The map number list is traversed in a loop, the road correction terrain is superimposed with the terrain data corresponding to the current map number to obtain the intersection, and the terrain data is covered with the road correction terrain in the area with the intersection, thereby completing the road terrain repair.

Images