Land surveying method and device based on multi-source data fusion

Abstract

The application discloses a land surveying method and device based on multi-source data fusion, and relates to the technical field of surveying, which comprises the following steps: collecting three-dimensional geodetic coordinate data and three-dimensional space point cloud data, obtaining starting three-dimensional coordinate and ending three-dimensional coordinate of surveying, constructing a triangulated irregular network (TIN) terrain model and a shortest path search model on the ground, searching for a starting node and a target node in a weighted undirected graph, calculating the shortest path on the ground and the ground distance, rendering the TIN terrain model to draw the shortest path on the ground, and setting through points and forbidden areas to update the shortest path on the ground and the ground distance. The application breaks through the dependence of traditional land surveying on the line-of-sight condition by fusing multi-source data and constructing continuous digital terrain, intelligently calculates and visualizes the shortest path on the ground, avoids the error accumulation of manual segmented operation, provides interactive path replanning capability, and significantly improves the accuracy, efficiency and scientific nature of planning and decision-making of ground surveying under complex terrain.

Description

Technical Field

[0001] This invention relates to the field of measurement technology, and specifically to a land measurement method and apparatus based on multi-source data fusion. Background Technology

[0002] Land surveying aims to provide basic geographic information support for spatial planning, real estate registration, infrastructure construction, and resource management by acquiring surface spatial information, including topographic features, ownership boundaries, and engineering elements. Traditional land surveying methods mainly rely on total stations, global positioning systems, and levels, and achieve measurement and positioning by manually setting up control points and observing them point by point.

[0003] To address the challenge of rapidly and accurately obtaining true surface distances in complex terrain conditions, particularly when obstacles obstruct direct line of sight between the starting and ending points, existing technologies employ traverse surveying and intermediate station setups. This involves adding intermediate observation points between the starting and ending points, performing segmented measurements, and accumulating the results. However, this approach suffers from cumbersome procedures, stringent requirements for the field environment, and the accumulation of errors due to numerous intermediate steps. Consequently, the final distance measurements become uncertain in accuracy and reliability, and continuous terrain information along the route cannot be obtained. To resolve these issues, a land surveying method and apparatus based on multi-source data fusion are proposed. Summary of the Invention

[0004] The purpose of this invention is to provide a land surveying method and apparatus based on multi-source data fusion to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: Firstly, a land surveying method based on multi-source data fusion, comprising the following steps 1-7:

[0006] Step 1: Collect and preprocess the three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data between the measurement start point and the end point to obtain the preprocessed three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data.

[0007] Step 2: Obtain the starting point and ending point 3D coordinates of the measurement from the preprocessed 3D geodetic coordinate data, and construct a triangular mesh terrain model based on the preprocessed 3D spatial point cloud data.

[0008] Step 3: Traverse the triangular mesh terrain model to represent the terrain surface within the land survey area as a weighted undirected graph;

[0009] Step 4: Based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, establish a shortest path search model on the ground. In the weighted undirected graph, measure the starting node and the target node that are closest to the starting point and the target node. Based on the starting node and the target node, calculate the shortest path on the ground.

[0010] Step 5: Calculate the surface distance based on the shortest surface path;

[0011] Step 6: Pre-set a typical terrain dataset to compare the surface distance with the actual distance in the typical terrain dataset, and adjust the parameters in the triangular mesh terrain model;

[0012] Step 7: Based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, render the triangular mesh terrain model, draw the shortest surface path, and update the shortest surface path and surface distance by setting necessary points and restricted areas.

[0013] A further improvement to the technical solution of this invention lies in the following: the process of acquiring and preprocessing three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data between the measurement start point and the end point to obtain preprocessed three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data includes:

[0014] Global Navigation Satellite System (GNSS) reference stations are set up at intermediate observation points added within the land survey area. Operators carry measuring devices integrated with mobile stations to conduct static observations of the quasi-physical measurement start and end points within the land survey area in sequence. The static observation process includes using the principle of real-time dynamic differential positioning. The mobile station receives the error correction signal broadcast by the reference station, corrects the satellite signal of the mobile station, and thus calculates the measurement start and end points, as well as the three-dimensional geodetic coordinate data between the measurement start and end points.

[0015] The operator carries a measuring device that integrates a 3D laser scanner and an inertial measurement unit, and moves along a path covering the terrain between the starting point and the end point of the measurement. The laser scanner actively emits laser pulses outward at a preset angular resolution and frequency based on the time of flight and measures the reflected signals to obtain the 3D spatial point cloud data of the ground surface relative to the laser scanner in the device's local coordinate system with the laser scanner as the origin. The inertial measurement unit simultaneously records the attitude data and acceleration change data of the measuring device during the movement.

[0016] The three-dimensional spatial point cloud data and attitude data are synchronized in time, and the attitude data are integrated by applying a state estimation algorithm to calculate the attitude and relative position of the measuring device at each moment.

[0017] The attitude and relative position of each laser point at the time of acquisition are substituted into the preset three-dimensional coordinate transformation model to transform the three-dimensional spatial point cloud data in the local coordinate system of the device to the coordinate system of the survey area, forming three-dimensional spatial point cloud data without motion distortion.

[0018] The motion-distortion-free 3D spatial point cloud data is registered with the selected 3D geodetic coordinate data of the measurement start and end points to the geodetic coordinate system. The registration process includes solving the 3D coordinate transformation model, selecting the already determined measurement start and end points between the survey area coordinate system and the geodetic coordinate system, and using the least squares adjustment method to solve for the transformation parameters that minimize the sum of squared residuals of the measurement start and end points after registration. The transformation parameters are then applied to the 3D spatial point cloud data to obtain preprocessed 3D geodetic coordinate data and 3D spatial point cloud data based on the geodetic coordinate system. The transformation parameters include translation vector, scale factor, and 3D rotation matrix, and their calculation process is as follows:

[0019] ;

[0020] in, The coordinates of a point in the geodetic coordinate system. These are the coordinates of the corresponding point in the source local coordinate system. Let be the translation vector, s be the scale factor, and R be the three-dimensional rotation matrix.

[0021] A further improvement to the technical solution of this invention lies in the following: the process of obtaining the starting point and ending point three-dimensional coordinates of the measurement from the preprocessed three-dimensional geodetic coordinate data, and constructing a triangular mesh terrain model based on the preprocessed three-dimensional spatial point cloud data, includes:

[0022] From the preprocessed 3D geodetic coordinate data, based on the pre-set markers, the 3D geodetic coordinate data of two points marked as the measurement start point and measurement end point are extracted and used as the 3D coordinates of the start point and the 3D coordinates of the end point;

[0023] Based on the preprocessed 3D spatial point cloud data, the Delaunay triangulation algorithm is adopted. The points in the 3D spatial point cloud data are used as vertices. According to the preset construction criteria, the vertices are connected into a network composed of non-overlapping triangular facets, thereby forming a triangular mesh terrain model. The construction criterion is that for the network, the circumcircle formed by the three vertices of the triangular facet does not contain other vertices in the 3D spatial point cloud data.

[0024] Based on the average point cloud density within the land survey area, a preset area threshold for triangular patches is established.

[0025] Traverse the triangular patches in the triangular mesh terrain model, calculate the magnitude of the cross product of two edge vectors in the triangular patch, and take half of the magnitude as the spatial area of ​​the triangular patch;

[0026] The spatial area is compared with an area threshold. If the spatial area is greater than the area threshold, the corresponding triangular facet is marked as an abnormal region.

[0027] A further improvement to the technical solution of this invention lies in the following: the process of traversing the triangular mesh terrain model and representing the terrain surface within the land survey area as a weighted undirected graph includes:

[0028] Traverse the vertices of the triangular patches in the triangular mesh terrain model, map the vertices to nodes in the weighted undirected graph, and store the 3D geodetic coordinate data corresponding to the vertices as attributes in the corresponding nodes.

[0029] Traverse the edges of the triangular patches in the triangular mesh terrain model and map the edge connecting two vertices to the edge connecting the corresponding two nodes in the weighted undirected graph;

[0030] The weight of a node's edge is defined as the 3D spatial length of the edge of the corresponding triangle facet; nodes, nodes' edges, and the weights of nodes' edges constitute a weighted undirected graph.

[0031] A further improvement to the technical solution of this invention lies in: establishing a shortest path search model on the Earth's surface based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point; and in the process of searching a weighted undirected graph, measuring the starting node and the target node with the closest distance between the starting point and the ending point, including:

[0032] The nearest neighbor search algorithm is used to traverse the nodes in the weighted undirected graph, extract the three-dimensional geodetic coordinate data of the nodes, and establish a shortest path search model on the ground.

[0033] The three-dimensional Euclidean distance between the three-dimensional geodetic coordinates of the calculated node and the three-dimensional coordinates of the starting point is selected as the starting node.

[0034] The three-dimensional Euclidean distance between the three-dimensional geodetic coordinates of the node and the three-dimensional coordinates of the endpoint is calculated, and the node with the smallest three-dimensional Euclidean distance is selected as the target node.

[0035] A further improvement to the technical solution of this invention lies in the following: the process of calculating the shortest surface path based on the starting node and the target node includes:

[0036] The A* search algorithm is used to define the estimated total cost value for a node from the starting node through the nodes to the target node. The estimated total cost value is the sum of the actual cost and the estimated heuristic cost. The actual cost is the sum of the weights of the edges from the starting node to the target node along the discovered paths. The estimated heuristic cost is the spatial straight-line distance from the starting node to the target node.

[0037] The A* search algorithm starts from the starting node, iteratively examines neighboring nodes, and calculates the estimated total cost of each neighboring node. In each iteration, it prioritizes expanding the neighboring node with the smallest estimated total cost value until the iterative search expands to the target node.

[0038] If the iterative search process successfully extends to the target node, it means that there is a connected path from the starting node to the target node, and it is considered that there is a shortest surface path. The A* search algorithm constructs the shortest surface path with the minimum actual cost by backtracking.

[0039] If all scalable nodes have been examined before the target node is found, but the target node has not yet been reached, it means that there is no connecting path between the starting node and the target node, and it is considered that there is no shortest surface path.

[0040] A further improvement to the technical solution of this invention lies in the fact that the process of calculating the surface distance based on the shortest surface path includes:

[0041] When a shortest surface path exists, initialize the surface distance to 0. Visit the edges of the nodes that constitute the shortest surface path one by one in the order from the starting node to the target node. Extract the weights of the edges of the nodes from the weighted undirected graph and add the weights of the edges of the nodes to the surface distance. Repeat the accumulation process until all the edges of the nodes of the shortest surface path have been visited, and output the accumulated surface distance.

[0042] A further improvement to the technical solution of this invention lies in the following: the process of pre-setting a typical terrain dataset, comparing surface distances with the actual distances in the typical terrain dataset, and adjusting the parameters in the triangular mesh terrain model includes:

[0043] A pre-set typical terrain dataset is provided, which includes test cases of typical land survey areas, three-dimensional geodetic coordinate data of the start and end points of typical land survey areas, three-dimensional spatial point cloud data, and the actual distance between the start and end points of typical land survey areas. The parameter to be adjusted in the triangular mesh terrain model is set as the maximum side length threshold of the triangle patch.

[0044] Set a range for the maximum side length threshold, select candidate parameter values ​​from the range, and use the three-dimensional geodetic coordinate data, three-dimensional spatial point cloud data and actual distance of the starting and ending points of a typical land survey area in a set of test cases as input. Construct a triangular mesh terrain model using candidate parameter values, calculate the surface distance based on the candidate parameter values, and the absolute error between the calculated surface distance and the actual distance. Iterate through the candidate parameter values ​​in the range and test with different test cases to obtain the average error of each candidate parameter value under different terrains.

[0045] By comparing the average error corresponding to each candidate parameter value, the candidate parameter with the smallest average error is selected as the optimal parameter for the triangular mesh terrain model.

[0046] A further improvement to the technical solution of this invention lies in the following: Based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, a three-dimensional graphics rendering pipeline technology is used to render a triangular mesh terrain model, draw the shortest surface path, and update the shortest surface path and surface distance by setting necessary points and restricted areas.

[0047] Using a triangular mesh terrain model as input, the three-dimensional geodetic coordinate data corresponding to the vertices and the topological relationship of the triangular patches are submitted to the graphics rendering pipeline. Through view transformation, projection transformation and viewport transformation, the triangular mesh terrain model is projected onto a two-dimensional display screen. After rasterization, continuous triangular patches are drawn.

[0048] Based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, draw prominent geometric marks at the corresponding positions of the triangular mesh terrain model, connect the nodes that form the shortest surface path in sequence, and draw the shortest surface path.

[0049] The operator uses an interactive device to specify the 3D coordinates of the necessary points on the rendered triangular mesh terrain model. The original single path search problem is decomposed into multiple continuous sub-path search problems. For each sub-path search problem, the A* search algorithm is re-executed to calculate the shortest path of each segment. The shortest paths of each segment are then sequentially spliced ​​together to form the updated shortest surface path. Based on the updated shortest surface path, the updated surface distance is obtained.

[0050] Operators use interactive devices to draw closed polygonal regions representing restricted areas on the rendered triangular mesh terrain model. They analyze the spatial intersection of edges in the weighted undirected graph with the restricted areas, modify the weights of the edges intersecting with the restricted areas to infinity, and re-execute the A* search algorithm with the original starting and target nodes as inputs to avoid edges with infinite weights, thereby obtaining the updated shortest surface path and surface distance.

[0051] Secondly, a land surveying device based on multi-source data fusion is used to realize a land surveying method based on multi-source data fusion. It includes a land surveying data acquisition unit, a terrain feature extraction unit, a land surface morphology representation unit, a shortest land surface path search unit, a land surface distance calculation unit, a parameter optimization unit, and an interactive display unit, wherein the units are connected by electrical signals.

[0052] The land survey data acquisition unit collects and preprocesses the three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data between the starting point and the ending point of the survey, and obtains the preprocessed three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data.

[0053] The terrain feature extraction unit obtains the starting point and ending point three-dimensional coordinates of the measurement from the preprocessed three-dimensional geodetic coordinate data, and constructs a triangular mesh terrain model based on the preprocessed three-dimensional spatial point cloud data.

[0054] The landform representation unit traverses the triangular mesh terrain model and expresses the terrain surface within the land survey area as a weighted undirected graph;

[0055] The shortest surface path search unit establishes a surface shortest path search model based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point. In the weighted undirected graph, it measures the starting node and the target node that are closest to the starting point and the target node, and calculates the shortest surface path based on the starting node and the target node.

[0056] The surface distance calculation unit calculates the surface distance based on the shortest surface path;

[0057] The parameter optimization unit is pre-set with a typical terrain dataset, which is used to compare the surface distance with the actual distance in the typical terrain dataset and adjust the parameters in the triangular mesh terrain model.

[0058] The interactive display unit renders a triangular mesh terrain model based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, draws the shortest surface path, and updates the shortest surface path and surface distance by setting necessary points and restricted areas.

[0059] Due to the adoption of the above technical solution, the technical progress achieved by this invention compared to the prior art is as follows:

[0060] This invention provides a land surveying method and apparatus based on multi-source data fusion. By fusing three-dimensional geodetic coordinate data representing absolute position with three-dimensional spatial point cloud data representing local fine morphology, a continuous digital terrain model including start and end points is constructed. This method breaks through the physical limitation of traditional surveying methods that must rely on direct line of sight between two points. This method eliminates the need for manual segmentation of station setting and point transfer measurements in complex terrain and areas obstructed by obstacles, significantly reducing the complexity and labor intensity of fieldwork and improving the feasibility and efficiency of surveying operations under non-line-of-sight conditions.

[0061] This invention provides a land surveying method and apparatus based on multi-source data fusion. By representing the actual terrain surface as a weighted undirected graph and using a surface shortest path search model for global optimization calculation, it can accurately calculate the true shortest path distance along the surface undulations, rather than the spatial straight-line distance and the cumulative distance of segmented measurements. This integrated algorithm processing mode effectively avoids the error accumulation problem caused by multiple station setups, observations, and calculations in traditional traverse surveying, ensuring the accuracy, objectivity, and reliability of the final measurement results.

[0062] This invention provides a land surveying method and apparatus based on multi-source data fusion. By providing a three-dimensional visualization rendering function that includes terrain, start and end points, and shortest path, users can intuitively review the measurement results and the rationality of the path. Furthermore, this invention empowers users to interactively replan paths by setting mandatory points and restricted areas, making the apparatus not only a simple measurement tool but also a decision support platform that can intelligently optimize paths according to actual engineering and planning needs, greatly enhancing the practicality and adaptability of the apparatus. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0064] Figure 1 A flowchart of the land surveying method based on multi-source data fusion provided by the present invention.

[0065] Figure 2 The structural block diagram of the land surveying device based on multi-source data fusion provided by the present invention. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0067] Example 1, as Figure 1 As shown, this invention provides a land surveying method based on multi-source data fusion, including the following steps 1-7:

[0068] Step 1: Collect and preprocess the three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data between the measurement start point and the end point to obtain the preprocessed three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data.

[0069] In some embodiments, a global navigation satellite system reference station is set up at an intermediate observation point added within the land survey area. The operator carries a measuring device integrated with a mobile station and conducts static observations of the quasi-physical measurement start and end points within the land survey area in sequence. The static observation process includes using the principle of real-time dynamic differential positioning, where the mobile station receives the error correction signal broadcast by the reference station, corrects the satellite signal of the mobile station, and thus calculates the measurement start and end points, as well as the three-dimensional geodetic coordinate data between the measurement start and end points.

[0070] In some embodiments, an operator carries a measuring device integrating a 3D laser scanner and an inertial measurement unit, and moves along a path covering the terrain between the starting and ending points of the measurement. The laser scanner actively emits laser pulses outward at a preset angular resolution and frequency based on the time of flight and measures the reflected signals to obtain 3D spatial point cloud data of the ground surface relative to the laser scanner in the device's local coordinate system with the laser scanner as the origin. The inertial measurement unit simultaneously records the attitude data and acceleration change data of the measuring device during the movement.

[0071] In some embodiments, the three-dimensional spatial point cloud data and attitude data are synchronized in time, and a state estimation algorithm is applied to perform integral calculations on the attitude data to deduce the attitude and relative position of the measuring device at each moment.

[0072] In some embodiments, the attitude and relative position of each laser point at the acquisition time are substituted into a preset three-dimensional coordinate transformation model, so that the three-dimensional spatial point cloud data in the local coordinate system of the device is transformed into the coordinate system of the measurement area, forming three-dimensional spatial point cloud data without motion distortion.

[0073] In some embodiments, motion-distortion-free 3D spatial point cloud data is registered with the selected 3D geodetic coordinate data of the measurement start and end points to a geodetic coordinate system. The registration process includes solving a 3D coordinate transformation model, selecting the predetermined measurement start and end points between the survey area coordinate system and the geodetic coordinate system, using the least squares adjustment method to solve for the transformation parameters that minimize the sum of squared residuals between the measurement start and end points after registration, and applying the transformation parameters to the 3D spatial point cloud data to obtain preprocessed 3D geodetic coordinate data and 3D spatial point cloud data based on the geodetic coordinate system. The transformation parameters include a translation vector, a scale factor, and a 3D rotation matrix, and the specific calculation formulas are as follows:

[0074] ;

[0075] in, The coordinates of a point in the geodetic coordinate system. These are the coordinates of the corresponding point in the source local coordinate system. Let be the translation vector, s be the scale factor, and R be the three-dimensional rotation matrix.

[0076] Step 2: Obtain the starting point and ending point 3D coordinates of the measurement from the preprocessed 3D geodetic coordinate data, and construct a triangular mesh terrain model based on the preprocessed 3D spatial point cloud data.

[0077] In some embodiments, the three-dimensional geodetic coordinate data of two points marked as the measurement start point and the measurement end point are extracted from the preprocessed three-dimensional geodetic coordinate data according to the pre-set markers, and used as the three-dimensional coordinates of the start point and the three-dimensional coordinates of the end point.

[0078] In some embodiments, based on preprocessed 3D spatial point cloud data, the Delaunay triangulation algorithm is used to connect the points in the 3D spatial point cloud data into a network composed of non-overlapping triangular facets according to a preset construction criterion, thereby forming a triangular mesh terrain model. The construction criterion is that for the network, the circumcircle formed by the three vertices of the triangular facets does not contain other vertices in the 3D spatial point cloud data.

[0079] In some embodiments, an area threshold for triangular patches is preset based on the average point cloud density within the land survey area.

[0080] In some embodiments, the triangular patches in the triangular mesh terrain model are traversed, the magnitude of the cross product of two edge vectors in the triangular patch is calculated, and half of the magnitude is taken as the spatial area of ​​the triangular patch.

[0081] In some embodiments, the spatial area is compared with an area threshold. If the spatial area is greater than the area threshold, the corresponding triangular patch is marked as an abnormal region.

[0082] Step 3: Traverse the triangular mesh terrain model to represent the terrain surface within the land survey area as a weighted undirected graph.

[0083] In some embodiments, the vertices of the triangular patches in the triangular mesh terrain model are traversed, and the vertices are mapped to nodes in a weighted undirected graph. The three-dimensional geodetic coordinate data corresponding to the vertices are stored as attributes in the corresponding nodes.

[0084] In some embodiments, the edges of the triangular patches in the triangular mesh terrain model are traversed, and the edges connecting two vertices are mapped to the edges connecting the corresponding two nodes in the weighted undirected graph.

[0085] In some embodiments, the weight of the edge of a node is defined as the three-dimensional spatial length of the edge of the corresponding triangular facet.

[0086] In some embodiments, nodes, edges of nodes, and weights of edges of nodes constitute a weighted undirected graph.

[0087] Step 4: Based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, establish a shortest path search model on the ground. In the weighted undirected graph, measure the starting node and the target node that are closest to the starting point and the ending point. Based on the starting node and the target node, calculate the shortest path on the ground.

[0088] In some embodiments, the nearest neighbor search algorithm is used to traverse the nodes in the weighted undirected graph, extract the three-dimensional geodetic coordinate data of the nodes, and establish a shortest path search model on the ground.

[0089] In some embodiments, the three-dimensional geodetic coordinates of the calculated node are compared with the three-dimensional coordinates of the starting point in three-dimensional space. The node with the smallest three-dimensional Euclidean distance is selected as the starting node.

[0090] In some embodiments, the three-dimensional Euclidean distance between the three-dimensional geodetic coordinates of the calculation node and the three-dimensional coordinates of the endpoint is calculated, and the node with the smallest three-dimensional Euclidean distance is selected as the target node.

[0091] In some embodiments, the A* search algorithm is used to define the estimated total cost value for a node from the starting node through nodes to the target node. The estimated total cost value is the sum of the actual cost and the estimated heuristic cost. The actual cost is the sum of the weights of the edges from the starting node to the target node along the discovered path. The estimated heuristic cost is the spatial straight-line distance from the target node to the target node.

[0092] In some embodiments, the A* search algorithm starts from the starting node, iteratively examines neighboring nodes, and calculates the estimated total cost value of each neighboring node. In each iteration, it prioritizes expanding the neighboring node with the smallest estimated total cost value until the iterative search expands to the target node.

[0093] In some embodiments, if the iterative search process successfully extends to the target node, it indicates that there is a connected path from the starting node to the target node, and it is considered that there is a shortest surface path. The A* search algorithm constructs the shortest surface path with the minimum actual cost by backtracking.

[0094] In some embodiments, if all scalable nodes have been examined before the target node is found, but the target node has not yet been reached, it indicates that there is no connecting path between the starting node and the target node, and it is considered that there is no shortest surface path.

[0095] Step 5: Calculate the surface distance based on the shortest surface path.

[0096] In some embodiments, when a shortest surface path exists, the surface distance is initialized to 0. The edges of the nodes that constitute the shortest surface path are visited one by one in the order from the starting node to the target node. The weights of the edges of the nodes are extracted from the weighted undirected graph, and the weights of the edges of the nodes are accumulated into the surface distance. The accumulation process is repeated until all the edges of the nodes of the shortest surface path have been visited, and the accumulated surface distance is output.

[0097] Step 6: Preset a typical terrain dataset to compare the surface distance with the actual distance in the typical terrain dataset, and adjust the parameters in the triangular mesh terrain model.

[0098] In some embodiments, a typical terrain dataset is pre-set. The typical terrain dataset includes test cases of typical land survey areas, three-dimensional geodetic coordinate data of the start and end points of typical land survey areas, three-dimensional spatial point cloud data, and the actual distance between the start and end points of typical land survey areas. The parameter to be adjusted in the triangular mesh terrain model is set as the maximum side length threshold of the triangle patch.

[0099] In some embodiments, a range of values ​​for the maximum side length threshold is set, and candidate parameter values ​​are selected from the range. Using the three-dimensional geodetic coordinate data, three-dimensional spatial point cloud data, and true distance of the starting and ending points of a typical land survey area in a set of test cases as input, a triangular mesh terrain model is constructed using the candidate parameter values. The surface distance based on the candidate parameter values ​​is calculated, as well as the absolute error between the calculated surface distance and the true distance. The candidate parameter values ​​in the range are traversed, and different test cases are used for testing to obtain the average error of each candidate parameter value under different terrains.

[0100] In some embodiments, the average error corresponding to each candidate parameter value is compared, and the candidate parameter value with the smallest average error is selected as the optimal parameter of the triangular mesh terrain model.

[0101] Step 7: Based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, render the triangular mesh terrain model, draw the shortest surface path, and update the shortest surface path and surface distance by setting necessary points and restricted areas.

[0102] In some embodiments, a triangular mesh terrain model is used as input, and the three-dimensional geodetic coordinate data corresponding to the vertices and the topological relationship of the triangular patches are submitted to the graphics rendering pipeline. Through view transformation, projection transformation and viewport transformation, the triangular mesh terrain model is projected onto a two-dimensional display screen, and after rasterization processing, continuous triangular patches are drawn.

[0103] In some embodiments, based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, prominent geometric marks are drawn at the corresponding positions of the triangular mesh terrain model, and the nodes that constitute the shortest surface path are connected in sequence to draw the shortest surface path.

[0104] In some embodiments, the operator uses an interactive device to specify the three-dimensional coordinates of the necessary points on the rendered triangular mesh terrain model. The original single path search problem is decomposed into multiple consecutive sub-path search problems. For each sub-path search problem, the A* search algorithm is re-executed to calculate the shortest path of each segment. The shortest paths of each segment are then sequentially spliced ​​together to form the updated shortest surface path. Based on the updated shortest surface path, the updated surface distance is obtained.

[0105] In some embodiments, the operator uses an interactive device to draw closed polygonal regions representing restricted areas on the rendered triangular mesh terrain model, analyzes the spatial intersection of edges in the weighted undirected graph with the restricted areas, modifies the weights of the edges intersecting with the restricted areas to infinity, and re-executes the A* search algorithm with the original starting node and target node as input to avoid edges with infinite weights, thereby obtaining the updated shortest surface path and surface distance.

[0106] Example 2, as Figure 2 As shown, based on Embodiment 1, the present invention also provides a technical solution: a land surveying device based on multi-source data fusion, used to realize a land surveying method based on multi-source data fusion, including a land surveying data acquisition unit, a terrain feature extraction unit, a land surface morphology representation unit, a shortest land surface path search unit, a land surface distance calculation unit, a parameter optimization unit, and an interactive display unit, wherein the units are electrically connected.

[0107] The land survey data acquisition unit collects and preprocesses the three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data between the starting point and the ending point of the survey, and obtains the preprocessed three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data.

[0108] The terrain feature extraction unit obtains the starting point and ending point three-dimensional coordinates of the measurement from the preprocessed three-dimensional geodetic coordinate data, and constructs a triangular mesh terrain model based on the preprocessed three-dimensional spatial point cloud data.

[0109] The landform representation unit traverses the triangular mesh terrain model and expresses the terrain surface within the land survey area as a weighted undirected graph;

[0110] The shortest surface path search unit establishes a surface shortest path search model based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point. In the weighted undirected graph, it measures the starting node and the target node that are closest to the starting point and the target node. Based on the starting node and the target node, it calculates the shortest surface path.

[0111] The surface distance calculation unit calculates the surface distance based on the shortest surface path;

[0112] The parameter optimization unit is pre-set with a typical terrain dataset, which is used to compare the surface distance with the actual distance in the typical terrain dataset and adjust the parameters in the triangular mesh terrain model.

[0113] The interactive display unit renders a triangular mesh terrain model based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, draws the shortest surface path, and updates the shortest surface path and surface distance by setting necessary points and restricted areas.

[0114] It should be noted that the aforementioned Figure 1 The explanations and effects of the method embodiments shown are also applicable to the method of this embodiment, and the principle is the same. Therefore, this embodiment will not be limited thereto.

Claims

1. A land surveying method based on multi-source data fusion, characterized in that, Includes the following steps: Collect and preprocess three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data between the measurement start point and the end point to obtain the preprocessed three-dimensional geodetic coordinate data and the three-dimensional spatial point cloud data; The starting point and ending point three-dimensional coordinates of the measurement are obtained from the preprocessed three-dimensional geodetic coordinate data. Based on the preprocessed three-dimensional spatial point cloud data, a triangular mesh terrain model is constructed. By traversing the triangular mesh terrain model, the terrain surface within the land survey area is represented as a weighted undirected graph; Based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, a shortest path search model for the Earth's surface is established. The model searches the weighted undirected graph and measures the starting node and the target node that are closest to the starting point and the ending point. Based on the starting node and the target node, the shortest path on the Earth's surface is calculated. Calculate the surface distance based on the shortest surface path; A pre-set typical terrain dataset is used to compare the surface distance with the actual distance in the typical terrain dataset and to adjust the parameters in the triangular mesh terrain model; Based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, the triangular mesh terrain model is rendered, the shortest surface path is drawn, and the shortest surface path and the surface distance are updated by setting necessary points and restricted areas. The process of acquiring and preprocessing the three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data between the measurement start point and the end point to obtain the preprocessed three-dimensional coordinate data and the three-dimensional spatial point cloud data includes: Operators carry a measuring device integrated with a mobile station to conduct static observations of the quasi-physical measurement start and end points within the land survey area. The static observation process includes using the real-time dynamic differential positioning principle to calculate the measurement start and end points, as well as the three-dimensional geodetic coordinate data between the measurement start and end points. The operator carries the measuring device, which integrates a 3D laser scanner and an inertial measurement unit, and moves along a path covering the terrain between the starting point and the ending point of the measurement. The laser scanner acquires 3D spatial point cloud data of the ground surface relative to the laser scanner in the device's local coordinate system with the laser scanner as the origin. The inertial measurement unit simultaneously records the attitude data and acceleration change data of the measuring device during the movement. The three-dimensional spatial point cloud data and the attitude data are synchronized in time, and the attitude data is integrated to calculate the attitude and relative position of the measuring device at each moment. The attitude and relative position of each laser point at the time of acquisition are substituted into a preset three-dimensional coordinate transformation model to transform the three-dimensional spatial point cloud data in the local coordinate system of the device to the coordinate system of the survey area, thereby forming three-dimensional spatial point cloud data without motion distortion. The motion-distortion-free 3D spatial point cloud data is registered with the selected 3D geodetic coordinate data of the measurement start and end points to the geodetic coordinate system. The registration process includes using the least squares adjustment method to solve for the transformation parameters that minimize the sum of squared residuals of the measurement start and end points after registration, and applying the transformation parameters to the 3D spatial point cloud data to obtain the preprocessed 3D geodetic coordinate data and the 3D spatial point cloud data. The transformation parameters include translation vector, scale factor and 3D rotation matrix. The process of obtaining the starting point and ending point 3D coordinates of the measurement from the preprocessed 3D geodetic coordinate data, and constructing a triangular mesh terrain model based on the preprocessed 3D spatial point cloud data includes: From the preprocessed three-dimensional geodetic coordinate data, according to the pre-set markers, the three-dimensional geodetic coordinate data of two points marked as the measurement start point and measurement end point are extracted and used as the three-dimensional coordinates of the start point and the three-dimensional coordinates of the end point; Based on the preprocessed 3D spatial point cloud data, the Delaunay triangulation algorithm is used to connect the points in the 3D spatial point cloud data as vertices according to the preset construction criteria to form a network composed of non-overlapping triangular facets, thereby forming a triangular mesh terrain model. The construction criterion is that for the network, the circumcircle formed by the three vertices of the triangular facets does not contain the other vertices in the 3D spatial point cloud data. Based on the average point cloud density within the land survey area, a preset area threshold for the triangular patch is established; Traverse the triangular patches in the triangular mesh terrain model, calculate the magnitude of the cross product of two edge vectors in the triangular patch, and take half of the magnitude as the spatial area of ​​the triangular patch. The spatial area is compared with the area threshold. If the spatial area is greater than the area threshold, the corresponding triangular facet is marked as an abnormal region. The process of traversing the triangular mesh terrain model to represent the terrain surface within the land survey area as a weighted undirected graph includes: Traverse the vertices of the triangular patches in the triangular mesh terrain model, map the vertices to nodes in a weighted undirected graph, and store the three-dimensional geodetic coordinate data corresponding to the vertices as attributes in the corresponding nodes; Traverse the edges of the triangular patches in the triangular mesh terrain model, and map the edge connecting two vertices to the edge connecting the corresponding two nodes in the weighted undirected graph; The weight of the edge of the node is defined as the three-dimensional spatial length of the edge of the corresponding triangle facet; the node, the edge of the node, and the weight of the edge of the node constitute the weighted undirected graph; The process of establishing a shortest path search model based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, and searching the weighted undirected graph to measure the starting node and the target node with the closest distance between the starting point and the ending point, includes: The nearest neighbor search algorithm is used to traverse the nodes in the weighted undirected graph, extract the three-dimensional geodetic coordinate data of the nodes, and establish a shortest path search model on the ground. Calculate the three-dimensional Euclidean distance between the three-dimensional geodetic coordinates of the node and the three-dimensional coordinates of the starting point, and select the node with the smallest three-dimensional Euclidean distance as the starting node; Calculate the three-dimensional Euclidean distance between the three-dimensional geodetic coordinates of the node and the three-dimensional coordinates of the endpoint, and select the node with the smallest three-dimensional Euclidean distance as the target node.

2. The land surveying method based on multi-source data fusion according to claim 1, characterized in that: The process of calculating the shortest surface path based on the starting node and the target node includes: Using the A* search algorithm, the estimated total cost value for reaching the target node from the starting node is defined. The estimated total cost value is the sum of the actual cost and the estimated heuristic cost. The actual cost is the sum of the weights of the edges from the starting node to the target node along the discovered path. The estimated heuristic cost is the spatial straight-line distance from the target node to the target node. The A* search algorithm starts from the starting node, iteratively examines neighboring nodes, and calculates the estimated total cost value of each neighboring node. In each iteration, the neighboring node with the smallest estimated total cost value is preferentially expanded until the iterative search expands to the target node. If the iterative search process is successfully extended to the target node, it indicates that there is a connected path from the starting node to the target node, and it is considered that there is a shortest surface path. The A* search algorithm constructs the shortest surface path with the minimum actual cost by backtracking. If all scalable nodes have been examined before the target node is found, but the target node has not yet been reached, it means that there is no connecting path between the starting node and the target node, and it is considered that there is no shortest surface path.

3. The land surveying method based on multi-source data fusion according to claim 2, characterized in that: The process of calculating the surface distance based on the shortest surface path includes: When the shortest surface path exists, the surface distance is initialized to 0. Following the order from the starting node to the target node, the edges of the nodes that constitute the shortest surface path are visited one by one. The weights of the edges of the nodes are extracted from the weighted undirected graph, and the weights of the edges of the nodes are accumulated into the surface distance. The accumulation process is repeated until all the edges of the nodes of the shortest surface path have been visited, and the accumulated surface distance is output.

4. The land surveying method based on multi-source data fusion according to claim 3, characterized in that: The process of comparing the surface distances with the actual distances in the pre-set typical terrain dataset and adjusting the parameters in the triangular mesh terrain model includes: A pre-set typical terrain dataset is provided, which includes test cases of typical land survey areas, three-dimensional geodetic coordinate data of the start and end points of typical land survey areas, three-dimensional spatial point cloud data, and the actual distance between the start and end points of typical land survey areas. The parameter to be adjusted in the triangular mesh terrain model is set as the maximum side length threshold of the triangular patch. The maximum side length threshold is set to a range, and candidate parameter values ​​are selected from the range. Using the three-dimensional geodetic coordinate data, three-dimensional spatial point cloud data, and actual distance of the starting and ending points of the typical land survey area in a set of test cases as input, the triangular mesh terrain model is constructed using the candidate parameter values. The surface distance based on the candidate parameter values ​​is calculated, as well as the absolute error between the surface distance and the actual distance is calculated. The candidate parameter values ​​in the range are traversed, and different test cases are used for testing to obtain the average error of each candidate parameter value under different terrains. By comparing the average errors corresponding to each candidate parameter value, the candidate parameter value with the smallest average error is selected as the optimal parameter of the triangular mesh terrain model.

5. The land surveying method based on multi-source data fusion according to claim 3, characterized in that: The process of rendering the triangular mesh terrain model based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, using three-dimensional graphics rendering pipeline technology, drawing the shortest surface path, and updating the shortest surface path and the surface distance by setting mandatory points and restricted areas includes: Using the triangular mesh terrain model as input, the three-dimensional geodetic coordinate data corresponding to the vertex and the topological relationship of the triangular facets are submitted to the graphics rendering pipeline. Through view transformation, projection transformation and viewport transformation, the triangular mesh terrain model is projected onto a two-dimensional display screen. After rasterization processing, continuous triangular facets are drawn. Based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, prominent geometric marks are drawn at the corresponding positions of the triangular mesh terrain model, and the nodes that constitute the shortest surface path are connected in sequence to draw the shortest surface path. The operator uses an interactive device to specify the three-dimensional coordinates of the necessary points on the rendered triangular mesh terrain model. The original single path search problem is decomposed into multiple consecutive sub-path search problems. For each sub-path search problem, the A* search algorithm is re-executed to calculate the shortest path of each segment. The shortest paths of each segment are then sequentially concatenated to form the updated shortest surface path. Based on the updated shortest surface path, the updated surface distance is obtained. The operator uses an interactive device to draw closed polygonal regions representing restricted areas on the rendered triangular mesh terrain model. They analyze the spatial intersection of edges in the weighted undirected graph with the restricted areas, modify the weights of the edges intersecting with the restricted areas to infinity, and re-execute the A* search algorithm using the original starting node and target node as inputs to avoid the edges with infinite weights, thereby obtaining the updated shortest surface path and surface distance.

6. A land surveying device based on multi-source data fusion, characterized in that, It includes a land survey data acquisition unit, a terrain feature extraction unit, a land surface morphology representation unit, a shortest surface path search unit, a surface distance calculation unit, a parameter optimization unit, and an interactive display unit, wherein the units are connected by electrical signals. The land survey data acquisition unit is used to collect and preprocess three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data between the measurement start point and the end point to obtain the preprocessed three-dimensional geodetic coordinate data and the three-dimensional spatial point cloud data. The terrain feature extraction unit is used to obtain the starting point and ending point three-dimensional coordinates of the measurement from the preprocessed three-dimensional geodetic coordinate data, and to construct a triangular mesh terrain model based on the preprocessed three-dimensional spatial point cloud data. The landform representation unit is used to traverse the triangular mesh terrain model and express the terrain surface within the land survey area as a weighted undirected graph. The shortest surface path search unit is used to establish a surface shortest path search model based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, search the weighted undirected graph, measure the starting node and the target node with the closest distance between the starting point and the ending point, and calculate the shortest surface path based on the starting node and the target node. The surface distance calculation unit is used to calculate the surface distance based on the shortest surface path; The parameter optimization unit is used to pre-set a typical terrain dataset, compare the surface distance with the actual distance in the typical terrain dataset, and adjust the parameters in the triangular mesh terrain model. The interactive display unit is used to render the triangular mesh terrain model based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, draw the shortest surface path, and update the shortest surface path and the surface distance by setting necessary points and restricted areas. The process of acquiring and preprocessing the three-dimensional geodetic coordinate data and three-dimensional spatial point cloud data between the measurement start point and the end point to obtain the preprocessed three-dimensional coordinate data and the three-dimensional spatial point cloud data includes: Operators carry a measuring device integrated with a mobile station to conduct static observations of the quasi-physical measurement start and end points within the land survey area. The static observation process includes using the real-time dynamic differential positioning principle to calculate the measurement start and end points, as well as the three-dimensional geodetic coordinate data between the measurement start and end points. The operator carries the measuring device, which integrates a 3D laser scanner and an inertial measurement unit, and moves along a path covering the terrain between the starting point and the ending point of the measurement. The laser scanner acquires 3D spatial point cloud data of the ground surface relative to the laser scanner in the device's local coordinate system with the laser scanner as the origin. The inertial measurement unit simultaneously records the attitude data and acceleration change data of the measuring device during the movement. The three-dimensional spatial point cloud data and the attitude data are synchronized in time, and the attitude data is integrated to calculate the attitude and relative position of the measuring device at each moment. The attitude and relative position of each laser point at the time of acquisition are substituted into a preset three-dimensional coordinate transformation model to transform the three-dimensional spatial point cloud data in the local coordinate system of the device to the coordinate system of the survey area, thereby forming three-dimensional spatial point cloud data without motion distortion. The motion-distortion-free 3D spatial point cloud data is registered with the selected 3D geodetic coordinate data of the measurement start and end points to the geodetic coordinate system. The registration process includes using the least squares adjustment method to solve for the transformation parameters that minimize the sum of squared residuals of the measurement start and end points after registration, and applying the transformation parameters to the 3D spatial point cloud data to obtain the preprocessed 3D geodetic coordinate data and the 3D spatial point cloud data. The transformation parameters include translation vector, scale factor and 3D rotation matrix. The process of obtaining the starting point and ending point 3D coordinates of the measurement from the preprocessed 3D geodetic coordinate data, and constructing a triangular mesh terrain model based on the preprocessed 3D spatial point cloud data includes: From the preprocessed three-dimensional geodetic coordinate data, according to the pre-set markers, the three-dimensional geodetic coordinate data of two points marked as the measurement start point and measurement end point are extracted and used as the three-dimensional coordinates of the start point and the three-dimensional coordinates of the end point; Based on the preprocessed 3D spatial point cloud data, the Delaunay triangulation algorithm is used to connect the points in the 3D spatial point cloud data as vertices according to the preset construction criteria to form a network composed of non-overlapping triangular facets, thereby forming a triangular mesh terrain model. The construction criterion is that for the network, the circumcircle formed by the three vertices of the triangular facets does not contain the other vertices in the 3D spatial point cloud data. Based on the average point cloud density within the land survey area, a preset area threshold for the triangular patch is established; Traverse the triangular patches in the triangular mesh terrain model, calculate the magnitude of the cross product of two edge vectors in the triangular patch, and take half of the magnitude as the spatial area of ​​the triangular patch. The spatial area is compared with the area threshold. If the spatial area is greater than the area threshold, the corresponding triangular facet is marked as an abnormal region. The process of traversing the triangular mesh terrain model to represent the terrain surface within the land survey area as a weighted undirected graph includes: Traverse the vertices of the triangular patches in the triangular mesh terrain model, map the vertices to nodes in a weighted undirected graph, and store the three-dimensional geodetic coordinate data corresponding to the vertices as attributes in the corresponding nodes; Traverse the edges of the triangular patches in the triangular mesh terrain model, and map the edge connecting two vertices to the edge connecting the corresponding two nodes in the weighted undirected graph; The weight of the edge of the node is defined as the three-dimensional spatial length of the edge of the corresponding triangle facet; the node, the edge of the node, and the weight of the edge of the node constitute the weighted undirected graph; The process of establishing a shortest path search model based on the three-dimensional coordinates of the starting point and the three-dimensional coordinates of the ending point, and searching the weighted undirected graph to measure the starting node and the target node with the closest distance between the starting point and the ending point, includes: The nearest neighbor search algorithm is used to traverse the nodes in the weighted undirected graph, extract the three-dimensional geodetic coordinate data of the nodes, and establish a shortest path search model on the ground. Calculate the three-dimensional Euclidean distance between the three-dimensional geodetic coordinates of the node and the three-dimensional coordinates of the starting point, and select the node with the smallest three-dimensional Euclidean distance as the starting node; Calculate the three-dimensional Euclidean distance between the three-dimensional geodetic coordinates of the node and the three-dimensional coordinates of the endpoint, and select the node with the smallest three-dimensional Euclidean distance as the target node.

Images