A method and system for deducing geological structure outcrop lines based on field three-dimensional real scene

Abstract

The application discloses a kind of based on field three-dimensional real scene geological structure outcrop line deducing method and system, belong to computer aided design technical field, comprising: S1, according to the tilt photography of the scene of the outcrop line to be extracted, extract corresponding point cloud, and carry out pretreatment;S2, according to the information of the geological structure outcrop line to be solved, make occurrence plane in point cloud space;S3, the space of point cloud and vector plane is carried out grid division, finds out the small grid that point cloud and vector plane jointly occupy;S4, in the simplified point cloud of all small grid space, find a series of points closest to vector plane;S5, the series of points are fitted into line, obtain the geological structure outcrop line of overall space range.The application can solve the problem of geological structure outcrop line deducing based on field three-dimensional real scene using the related algorithm of tilt photography and point cloud, to quickly interpret and analyze geological structure outcrop line in geological exploration engineering in field work.

Description

Technical Field

[0001] This invention belongs to the field of computer-aided design technology, and in particular relates to a method and system for estimating the outcrop lines of geological structures based on three-dimensional field scenes. Background Technology

[0002] Geological exploration is a fundamental task in the construction of water conservancy and hydropower projects. However, in traditional geological exploration, two-dimensional maps lack intuitiveness, leading to a heavy reliance on engineers' subjective experience in the field analysis process, and the recorded data lacks spatial characteristics. Geological outcrops play an extremely important role in geological exploration and modeling. UAV-based oblique photogrammetry technology and its supporting rapid 3D modeling technology have developed rapidly in recent years. Geological outcrops observed on oblique photogrammetry 3D models can be used to perform profile measurements, lithological analysis, and stratum strike analysis of the study area. Currently, many digital tools for building and visualizing geological 3D scenes are applied in actual engineering projects, and some scholars have visualized geological structural surfaces by generating attitude vector surfaces in oblique photogrammetry scenes. The intersection of a specific attitude with the surface is the geological structure outcrop line.

[0003] However, in scenes built using oblique photogrammetry, it's difficult to directly perform Boolean operations between the oblique photogrammetry and vector surfaces, and geological structure outcrops are not easily derived and displayed directly within the scene. During geological exploration, geological structure outcrops are difficult to extract as line objects in geological digital tools and assign relevant geological attributes, causing inconvenience in cataloging and modeling. For example, the SuperMap desktop system uses oblique photogrammetry to trim 3D surfaces, requiring users to manually delete unnecessary points and lines from the 3D surfaces to leave only the geological structure outcrops, a cumbersome process. The GeoStation geological 3D system focuses on the statistics of geological objects and cannot directly manipulate the topological relationships between oblique photogrammetry and other objects. Currently, specialized geological modeling software worldwide mainly focuses on vector data modeling applications; software such as GOCAD, CATIA, and Earth Volumetric Studio have difficulty directly performing intersection operations between spatial planes and oblique photogrammetry models. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method and system for early fire detection and warning based on the power generation characteristics of photovoltaic modules. By utilizing algorithms related to oblique photography and point clouds, the problem of deducing geological structure outcrop lines based on 3D real-world field scenes can be solved, enabling rapid interpretation and analysis of geological structure outcrop lines in geological exploration projects during fieldwork.

[0005] The specific technical solution adopted in this invention is as follows:

[0006] The primary objective of this patent is to provide a method for estimating geological structure outcrop lines based on three-dimensional real-world field data, including:

[0007] S1. Extract the oblique photography of the exposed line scene as needed, extract the corresponding point cloud, and perform preprocessing.

[0008] S2. Based on the information of the geological structure outcrop line, construct the attitude plane in the point cloud space;

[0009] S3. Divide the space of the point cloud and the vector plane into a grid, and find the small grid that is occupied by both the point cloud and the vector plane.

[0010] S4. Within the simplified point cloud of all the small grid spaces, find a series of points that are closest to the vector plane;

[0011] S5. Fit a series of points extracted in S4 into a line to obtain the geological structure outcrop line of the entire spatial range.

[0012] The second objective of this patent is to provide a system for estimating geological structure outcrop lines based on three-dimensional real-world field scenes, including:

[0013] Preprocessing module: Based on the oblique photography of the exposed line scene that needs to be extracted, extract the corresponding point cloud and perform preprocessing;

[0014] Dip Plane Generation Module: Based on the information of the geological structure outcrop, the dip plane is generated in the point cloud space;

[0015] Fine mesh search module: Divides the space of point cloud and vector plane into a mesh and finds the fine mesh jointly occupied by point cloud and vector plane;

[0016] Series Point Search Module: Within a simplified point cloud of all fine grid spaces, find a series of points that are closest to the vector plane;

[0017] Fitting module: Fits a series of points extracted from the series point search module into a line to obtain the geological structure outcrop line of the entire spatial range.

[0018] The third objective of this patent is to provide a computer program for realizing the above-mentioned method of estimating geological structure outcrop lines based on field three-dimensional real-world scenes.

[0019] The fourth objective of this patent is to provide an information data processing terminal that realizes the above-mentioned method for estimating geological structure outcrop lines based on field three-dimensional real-world scenes.

[0020] The fifth objective of this patent is to provide a computer-readable storage medium, including instructions that, when executed on a computer, cause the computer to perform the aforementioned method for estimating geological structure outcrop lines based on field three-dimensional real-world scenes.

[0021] The advantages and positive effects of this invention are as follows:

[0022] This invention utilizes algorithms related to oblique photography and point clouds to solve the problem of estimating geological outcrop lines based on 3D real-world field scenes. Compared with traditional geological outcrop data extraction methods, this invention features better visualization, higher efficiency, and ease of use.

[0023] This invention, based on field 3D real-scene analysis, focuses on the projection mapping relationship between the geological survey site surface and the horizontal plane. By utilizing algorithms related to simplified point clouds and point cloud mesh generation, it simplifies the massive point cloud down to the outcrop line location, extracts points near the desired outcrop line, and finally fits the geological structure outcrop line through minimum distance calculation. Traditional methods for estimating geological structure outcrop lines rely on engineers' subjective experience; oblique photography and vector surfaces cannot be directly used for Boolean operations, requiring manual calibration of the geological structure outcrop line on the oblique photography. The method proposed in this invention can directly perform intersection calculations based on attitude and location information in the oblique photography scene, enabling rapid estimation of the geological structure outcrop line; it simplifies the point cloud computing process, improves computational speed, and maintains a certain level of accuracy, meeting the needs of rapid field analysis of geological outcrop lines. Attached Figure Description

[0024] Figure 1 A flowchart of a preferred embodiment of the present invention;

[0025] Figure 2 This is a schematic diagram of oblique photography in a preferred embodiment of the present invention;

[0026] Figure 3 This is a schematic diagram of a local point cloud projected onto the attitude plane in a preferred embodiment of the present invention;

[0027] Figure 4 This is a diagram of a quadtree structure in a preferred embodiment of the present invention;

[0028] Figure 5 This is a schematic diagram of the extraction of geological structure outcrop lines in a preferred embodiment of the present invention. Detailed Implementation

[0029] To further understand the invention's content, features, and effects, the following embodiments are provided, and detailed descriptions are given in conjunction with the accompanying drawings.

[0030] Geological outcrops play a crucial role in geological exploration and modeling. Geological outcrops observed on oblique photogrammetry 3D models enable analysis of the study area, including profile measurements, lithological analysis, and stratum strike determination. Existing comprehensive geological digital tools struggle to extract the geometry of geological outcrop lines by finding the intersection of spatial vector surfaces with the oblique photogrammetry model. Therefore, this invention, combining hydraulic engineering geological exploration and geological structure deduction, utilizes a quadtree approach to divide the oblique photogrammetry point cloud model into a network, calculates the vector plane distance, and then uses scatter plots to fit the geological outcrop lines. This simplifies the point cloud computing process, maintains a certain level of accuracy, and meets the needs of rapid field analysis of geological outcrops.

[0031] The following is in conjunction with the appendix Figures 1 to 5 The structure of the present invention will be described in detail.

[0032] A method for estimating geological structure outcrop lines based on 3D field simulations includes:

[0033] S1, based on the oblique photography of the scene where the exposed line needs to be extracted, extract the corresponding point cloud and preprocess it; further including:

[0034] S101: Obtain the oblique photography model of the work area and convert the oblique photography into point cloud (.ply) using oblique photography software;

[0035] S102: Remove outliers;

[0036] S103: Remove non-surface points, including vertical features and noise below the surface, so that the spatial point cloud can be projected onto the horizontal plane in subsequent steps.

[0037] S2, based on the information of the geological structure outcrop, construct the attitude plane in the point cloud space; further including:

[0038] S201: Read the location and orientation information of the outcrop line;

[0039] S202: Calculate the plane equation in the coordinate system relative to the point cloud space based on the information of the exposed line, and limit the plane range according to the structural extension distance;

[0040] The plane equation is as follows:

[0041] sinθ·sinα(x-x0)+cosθ·sinα(y-y0)+cosα(z-z0)=0

[0042] In the formula, θ is the dip direction (0°-360°), α is the dip angle (0°-90°), and (x0,y0,z0) is the spatial location of the known geological outcrop.

[0043] S203: Load the finite plane of the orientation into the same space of the point cloud.

[0044] S3, meshing the space of the point cloud and the vector plane, identifying the small meshes shared by the point cloud and the vector plane; further including:

[0045] S301: Project the point cloud and spatial plane vertically onto the horizontal plane (XOY); the spatial plane, after being projected onto the horizontal plane, forms a parallelogram region Ω; since the actual point cloud in the geological survey site has good horizontal projection properties compared with other 3D point clouds, the point cloud used to solve for geological outcrops corresponds to only one surface elevation (z) under a horizontal (x,y) coordinate system, the octree traversal used in traditional spatial point cloud can be transformed into a two-dimensional quadtree traversal, improving computational efficiency; the projected planar point cloud dataset is denoted as D.

[0046] S302: Determining the outermost domain. Let the planar point cloud dataset be D. Traverse all points in D to obtain the maximum value x in the X and Y directions. min y max With minimum value x max y min The coordinates of the four intersection points are obtained: (x...) max ,y max ), (x max ,y min ), (x min ,y min ), (x min ,y max This ensures that the projected planar point cloud is contained within this rectangular grid.

[0047] S303: Connect the midpoints of opposite sides of the previous rectangular grid, dividing this rectangular grid into 4 smaller grids. The grid is encoded using a linear quadtree. The encoding of any page node is an n-digit code (n is the level of the quadtree). For a regularly partitioned quadtree space, a quaternary code number is introduced. In this example, a grid that is not further subdivided is considered a leaf node in the quadtree. Each node in the quadtree uniquely corresponds to a quaternary number M.

[0048] M = q n-1 ·4 n-1 +q n-2 ·4 n-2 +…+q k ·4 k +…+q1·4 1 +q0·4 0

[0049] Where, q k={0,1,2,3}, representing the geometric position within this level; n represents the level of the quadtree;

[0050] S304: After meshing, determine the intersection of the point cloud dataset D, the parallelogram region Ω, and the edge of each small mesh; record the mesh status at each node: if the mesh contains any point in the point cloud, mark it as a full node for the point cloud; if the mesh intersects with the region Ω, mark it as a full node for the region; if both of the above conditions are met, mark the node as a full node, and the remaining nodes as empty nodes;

[0051] For example, for grid node 0 of a certain layer,

[0052]

[0053] If:

[0054]

[0055] Then mark the grid (node) as a full node.

[0056] S305: Extract each small mesh that contains both point cloud and vector plane (the above full nodes). Other empty node meshes cannot have exposed lines, so they are discarded and not considered.

[0057] S306: Set the grid length and width thresholds, take each small grid from the previous step as the initial rectangular grid, and repeat S303 to C5 until the size of all small grids is smaller than the threshold; record the grids contained in all small grids, and the intersection of the plane and the point cloud will be generated near this part of the point cloud.

[0058] S307: Discretize the selected portion of the point cloud (i.e., all full leaf nodes) at the end of the loop into scattered points.

[0059] S4, within the simplified point cloud of all fine grid spaces, find a set of points closest to the vector plane; further including:

[0060] S401: Map the scattered points output by S3 back to the three-dimensional space according to their original height, denoted as dataset D';

[0061] S402: Set the threshold δ for the distance between the scattered points and the plane according to actual needs;

[0062] S403: In dataset D', traverse every point in every grid cell in breadth-first search, and calculate the distance d from each point to the plane using the plane equation:

[0063] d=sinθ·sinα(x D' -x0)+cosθ·sinα(y D'-y0)+cosα(z D' -z0)

[0064] In the formula, θ represents the dip direction (0°-360°), α represents the dip angle (0°-90°), and (x0, y0, z0) represents the spatial location of the known geological outcrop. D' ,y D' ,z D' ) represents the points in dataset D'.

[0065] S404: Extract a series of scattered points whose distance d is less than the distance threshold δ, denoted as dataset D.

[0066] S5, by fitting a series of points extracted from S4 into a line, obtains the geological structure outcrop line of the entire spatial range, further including:

[0067] S501: Select the point P1 closest to the edge in dataset D, select the point P2 closest to it in the dataset, and connect P1 and P2.

[0068] S502: At the current point P n Search nearby (excluding the previous point P) n-1 (Except for) its nearest point P in the dataset n+1 P n P n+1 Make a connection;

[0069] S503: Repeat S502 until a point is repeatedly connected, then stop connecting the lines. The resulting polyline is the geological outcrop line on the oblique photography model.

[0070] Example

[0071] The oblique photography used in the experiment is an oblique photography model of a construction area of ​​a water conservancy project in western my country. The geological outcrops and occurrence information in it were all obtained from field measurements. Figure 2 As shown. In general, the site topography is complex with numerous geological outcrops; the oblique photography quality is good.

[0072] A method for estimating geological structure outcrop lines based on 3D field data, with the overall route map shown in Figure 1, includes the following steps:

[0073] S1, based on the oblique photography of the exposed line scene to be extracted, extract the corresponding point cloud and preprocess it, which further includes:

[0074] S101: Obtain the oblique photography model of the work area and convert the oblique photography into point cloud (.ply) using oblique photography software. In general, the point cloud of this work area is relatively dense, with about 1.5 million point clouds in this block.

[0075] S102: Remove outliers to reduce the range of the outermost domain in the algorithm;

[0076] S103: Remove non-surface points, including vertical features and subsurface noise, to facilitate subsequent steps in projecting the spatial point cloud onto a horizontal plane. This process removed point clouds of most houses, trees, and cars in the work area's plains; it also removed all subsurface noise.

[0077] S2, based on the information of the geological structure outcrop, construct the attitude plane in the point cloud space; this further includes:

[0078] S201: Read the location and orientation information of the outcrop line;

[0079] S202: Calculate the plane equation in the coordinate system relative to the point cloud space based on the information of the exposed line, and limit the plane range according to the structural extension distance;

[0080] The plane equation is as follows:

[0081] sinθ·sinα(x-x0)+cosθ·sinα(y-y0)+cosα(z-z0)=0

[0082] In the formula, θ is the dip direction (0°-360°), α is the dip angle (0°-90°), and (x0,y0,z0) is the spatial location of the known geological outcrop.

[0083] S203: Load the finite plane of the orientation into the same space of the point cloud.

[0084] S3, mesh the space of the point cloud and the vector plane, and identify the small meshes jointly occupied by the point cloud and the vector plane, which further includes:

[0085] S301: Project the point cloud and spatial plane vertically onto the horizontal plane (XOY); the spatial plane, after being projected onto the horizontal plane, forms a parallelogram region Ω; since the actual point cloud in the geological survey site has good horizontal projection properties compared with other 3D point clouds, the point cloud used to solve for geological outcrops corresponds to only one surface elevation (z) under a horizontal (x,y) coordinate system, the octree traversal used in traditional spatial point cloud can be transformed into a two-dimensional quadtree traversal, improving computational efficiency; the projected planar point cloud dataset is denoted as D.

[0086] S302: Determining the outermost domain. Let the planar point cloud dataset be D. Traverse all points in D to obtain the maximum value x in the X and Y directions. max y max With minimum value x min y min The coordinates of the four intersection points are obtained: (x...) max ,ymax ), (x max ,y min ), (x min ,y min ),

[0087] (x min ,y max This ensures that the projected planar point cloud is contained within this rectangular grid.

[0088] S303: Connect the midpoints of opposite sides of the previous rectangular grid, dividing this rectangular grid into 4 smaller grids. The grid is encoded using a linear quadtree encoding. The encoding of any page node is an n-digit code (n is the level of the quadtree). For a regularly partitioned quadtree space, a quaternary encoding sequence number is introduced; for example... Figure 4 As shown, in this example of mesh partitioning, a mesh that is no longer partitioned is considered a leaf node in a quadtree; each node in the quadtree uniquely corresponds to a quaternary number M:

[0089] M = q n-1 ·4 n-1 +q n-2 ·4 n-2 +…+q k ·4 k +…+q1·4 1 +q0·4 0

[0090] Where, q k ={0,1,2,3}, representing the geometric position within this level; n represents the level of the quadtree;

[0091] S304: After meshing, determine the intersection of the point cloud dataset D, the parallelogram region Ω, and the edge of each small mesh; record the mesh status at each node: if the mesh contains any point in the point cloud, mark it as a full node for the point cloud; if the mesh intersects with the region Ω, mark it as a full node for the region; if both of the above conditions are met, mark the node as a full node, and the remaining nodes as empty nodes;

[0092] For example, for grid node 0 of a certain layer,

[0093]

[0094] If:

[0095]

[0096] Then mark the grid (node) as a full node. For example... Figure 4As shown, nodes 03, 10, 11, 12, 30, 31, 32, and 33 are the full nodes of the lowest level.

[0097] S305: Extract each small mesh that simultaneously contains both point cloud and vector plane (the aforementioned full nodes). Other empty node meshes cannot have exposed lines and are discarded. Figure 3 As shown, the four grids, top left, top right, bottom left, and bottom right, are denoted as 0, 1, 2, and 3, respectively. Grids 1, 2, and 3 contain both point clouds and vector planes, and require further processing. Grid 0 contains only point clouds but not vector planes, so the points on it are discarded without consideration.

[0098] S306: Set the grid length and width thresholds, take each small grid from the previous step as the initial rectangular grid, and repeat S303 to C5 until the size of all small grids is smaller than the threshold; record the grids contained in all small grids, and the intersection of the plane and the point cloud will be generated near this part of the point cloud.

[0099] S307: Discretize the selected portion of the point cloud (i.e., all full leaf nodes) at the end of the loop into scattered points.

[0100] S4, within the simplified point cloud of all fine grid spaces, find a set of points closest to the vector plane, which further includes:

[0101] S401: Map the scattered points output by S3 back to the three-dimensional space according to their original height, denoted as dataset D';

[0102] S402: Set the threshold δ for the distance between the scattered points and the plane according to actual needs;

[0103] S403: In dataset D', traverse every point in every grid cell in breadth-first search, and calculate the distance d from each point to the plane using the plane equation:

[0104] d=sinθ·sinα(x D' -x0)+cosθ·sinα(y D' -y0)+cosα(z D' -z0)

[0105] In the formula, θ represents the dip direction (0°-360°), α represents the dip angle (0°-90°), and (x0, y0, z0) represents the spatial location of the known geological outcrop. D' ,y D' ,z D' ) represents the points in dataset D'.

[0106] S404: Extract a series of scattered points whose distance d is less than the distance threshold δ, denoted as dataset D.

[0107] S5, by fitting a series of points extracted in S4 into a line, yields the geological structure outcrop line for the entire spatial range. Further, it includes:

[0108] S501: Select the point P1 closest to the edge in dataset D, select the point P2 closest to it in the dataset, and connect P1 and P2.

[0109] S502: At the current point P n Search nearby (excluding the previous point P) n-1 (Except for) its nearest point P in the dataset n+1 P n P n+1 Make a connection;

[0110] S503: Repeat S502 until a point is repeatedly connected, then stop connecting the lines. The resulting polyline is the geological outcrop line on the oblique photography model.

[0111] Finally, after the E-step fitting, the geological structure outcrop lines are formed; after being loaded into the oblique photography scene, as shown... Figure 5 As shown, the method of using discrete point clouds and quadtree meshes to deduce geological structure outcrop lines yields results that are close to the displayed values, demonstrating a certain degree of accuracy; the calculation process is also relatively short, exhibiting a certain level of efficiency.

[0112] A system for estimating geological structure outcrop lines based on 3D field simulations includes:

[0113] Preprocessing module: Based on the oblique photography of the scene from which the exposed line needs to be extracted, extract the corresponding point cloud and perform preprocessing; further including:

[0114] S101: Obtain the oblique photography model of the work area and convert the oblique photography into point cloud (.ply) using oblique photography software;

[0115] S102: Remove outliers;

[0116] S103: Remove non-surface points, including vertical features and noise below the surface, so that the spatial point cloud can be projected onto the horizontal plane in subsequent steps.

[0117] The attitude plane generation module: Based on the information of the outcrop lines of the desired geological structure, it generates the attitude plane in the point cloud space; it further includes:

[0118] S201: Read the location and orientation information of the outcrop line;

[0119] S202: Calculate the plane equation in the coordinate system relative to the point cloud space based on the information of the exposed line, and limit the plane range according to the structural extension distance;

[0120] The plane equation is as follows:

[0121] sinθ·sinα(x-x0)+cosθ·sinα(y-y0)+cosα(z-z0)=0

[0122] In the formula, θ is the dip direction (0°-360°), α is the dip angle (0°-90°), and (x0,y0,z0) is the spatial location of the known geological outcrop.

[0123] S203: Load the finite plane of the orientation into the same space of the point cloud.

[0124] Fine mesh finding module: This module performs mesh generation on the point cloud and vector plane, identifying the fine mesh areas shared by both. Further components include:

[0125] S301: Project the point cloud and spatial plane vertically onto the horizontal plane (XOY); the spatial plane, after being projected onto the horizontal plane, forms a parallelogram region Ω; since the actual point cloud in the geological survey site has good horizontal projection properties compared with other 3D point clouds, the point cloud used to solve for geological outcrops corresponds to only one surface elevation (z) under a horizontal (x,y) coordinate system, the octree traversal used in traditional spatial point cloud can be transformed into a two-dimensional quadtree traversal, improving computational efficiency; the projected planar point cloud dataset is denoted as D.

[0126] S302: Determining the outermost domain. Let the planar point cloud dataset be D. Traverse all points in D to obtain the maximum value x in the X and Y directions. max y max With minimum value x min y min The coordinates of the four intersection points are obtained: (x...) max ,y max ), (x max ,y min ), (x min ,y min ),

[0127] (x min ,y max This ensures that the projected planar point cloud is contained within this rectangular grid.

[0128] S303: Connect the midpoints of opposite sides of the previous rectangular grid, dividing this rectangular grid into 4 smaller grids. The grid is encoded using a linear quadtree. The encoding of any page node is an n-digit code (n is the level of the quadtree). For a regularly partitioned quadtree space, a quaternary code number is introduced. In this example, a grid that is not further subdivided is considered a leaf node in the quadtree. Each node in the quadtree uniquely corresponds to a quaternary number M.

[0129] M = q n-1 ·4 n-1 +q n-2 ·4 n-2 +…+q k ·4 k +…+q1·4 1 +q0·4 0

[0130] Where, q k ={0,1,2,3}, representing the geometric position within this level; n represents the level of the quadtree;

[0131] S304: After meshing, determine the intersection of the point cloud dataset D, the parallelogram region Ω, and the edge of each small mesh; record the mesh status at each node: if the mesh contains any point in the point cloud, mark it as a full node for the point cloud; if the mesh intersects with the region Ω, mark it as a full node for the region; if both of the above conditions are met, mark the node as a full node, and the remaining nodes as empty nodes;

[0132] For example, for grid node 0 of a certain layer,

[0133]

[0134] If:

[0135]

[0136] Then mark the grid (node) as a full node.

[0137] S305: Extract each small mesh that contains both point cloud and vector plane (the above full nodes). Other empty node meshes cannot have exposed lines, so they are discarded and not considered.

[0138] S306: Set the grid length and width thresholds, take each small grid from the previous step as the initial rectangular grid, and repeat S303 to C5 until the size of all small grids is smaller than the threshold; record the grids contained in all small grids, and the intersection of the plane and the point cloud will be generated near this part of the point cloud.

[0139] S307: Discretize the selected portion of the point cloud (i.e., all full leaf nodes) at the end of the loop into scattered points.

[0140] Series Point Search Module: Within a simplified point cloud of all fine grid spaces, find a series of points that are closest to the vector plane;

[0141] S4, within the simplified point cloud of all fine grid spaces, find a set of points closest to the vector plane; further including:

[0142] S401: Map the scattered points output by S3 back to the three-dimensional space according to their original height, denoted as dataset D';

[0143] S402: Set the threshold δ for the distance between the scattered points and the plane according to actual needs;

[0144] S403: In dataset D', traverse every point in every grid cell in breadth-first search, and calculate the distance d from each point to the plane using the plane equation:

[0145] d=sinθ·sinα(x D' -x0)+cosθ·sinα(y D' -y0)+cosα(z D' -z0)

[0146] In the formula, θ represents the dip direction (0°-360°), α represents the dip angle (0°-90°), and (x0, y0, z0) represents the spatial location of the known geological outcrop. D' ,y D' ,z D' ) represents the points in dataset D'.

[0147] S404: Extract a series of scattered points whose distance d is less than the distance threshold δ, denoted as dataset D.

[0148] Fitting module: Fits a series of points extracted by the series point search module into a line to obtain the geological structure outcrop line of the entire spatial range. Further includes:

[0149] S501: Select the point P1 closest to the edge in dataset D, select the point P2 closest to it in the dataset, and connect P1 and P2.

[0150] S502: At the current point P n Search nearby (excluding the previous point P) n-1 (Except for) its nearest point P in the dataset n+1 P n P n+1 Make a connection;

[0151] S503: Repeat S502 until a point is repeatedly connected, then stop connecting the lines. The resulting polyline is the geological outcrop line on the oblique photography model.

[0152] A computer program that implements the above-mentioned method for estimating geological structure outcrop lines based on field 3D real-world scenes.

[0153] An information data processing terminal for realizing the above-mentioned method of inferring geological structure outcrop lines based on field three-dimensional real-world scenes.

[0154] A computer-readable storage medium includes instructions that, when executed on a computer, cause the computer to perform the above-described method for estimating geological structure outcrop lines based on field 3D real-world scenes.

[0155] A computer program for implementing the method for estimating geological structure outcrop lines based on field 3D real-world scenes in the above preferred embodiment.

[0156] An information data processing terminal that implements the method for estimating geological structure outcrop lines based on field three-dimensional real-world scenes in the above preferred embodiment.

[0157] A computer-readable storage medium includes instructions that, when executed on a computer, cause the computer to perform the method for estimating geological structure outcrop lines based on field three-dimensional real-world scenes as described in the preferred embodiment above.

[0158] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented, in whole or in part, as a computer program product, the computer program product includes one or more computer instructions. When the computer program instructions are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0159] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention shall fall within the scope of the technical solution of the present invention.

Claims

1. A method for estimating geological structure outcrop lines based on three-dimensional field scenes, characterized in that, include: S1. Extract the oblique photography of the exposed line scene as needed, extract the corresponding point cloud, and perform preprocessing. S2. Based on the information of the geological structure outcrop line, construct the attitude plane in the point cloud space; S3. Mesh the space of the point cloud and the vector plane, identifying the small mesh areas shared by both; specifically including: S301. Project the point cloud and spatial plane onto the horizontal plane XOY in a vertical direction; after the spatial plane is projected onto the horizontal plane, a parallelogram-shaped region is formed. Traverse the spatial point cloud using a two-dimensional quadtree; denoted as D for the projected planar point cloud dataset. S302. Determining the outermost region: Traverse all points in D to obtain the maximum values ​​in the X and Y directions. , and minimum value , The coordinates of the four intersection points are obtained: , , , The projected planar point cloud is included within this rectangular grid. S303. Connect the midpoints of opposite sides of the previous rectangular grid, dividing this rectangular grid into 4 smaller grids. The grid is encoded using a linear quadtree encoding. Any page node is encoded using an n-digit code. For a regularly partitioned quadtree space, a quaternary encoding sequence number is introduced. In grid partitioning, a grid that is no longer further subdivided is considered a leaf node in the quadtree. Each node in the quadtree uniquely corresponds to a quaternary number M. in, , representing the geometric position within that layer; n Represents the level of the quadtree; S304. After meshing, determine the planar point cloud dataset D and the parallelogram region. The intersection with the edges of each small mesh; recording the mesh status at each node: if the mesh contains any point in the point cloud, it is marked as a full node for the point cloud; if the mesh intersects with the region... If there is an intersection, it is marked as a full node for the region; if both of the above conditions are met, the node is marked as a full node, and the rest of the nodes are empty nodes. S305. Extract each small mesh that contains both point cloud and vector plane, and discard other empty node meshes; S306. Set the grid length and width thresholds. Take each small grid from the previous step as the initial rectangular grid. Repeat S303 to S305 until the size of all small grids is smaller than the threshold. Record the grids contained within all small grids. The intersection of the plane and the point cloud will be generated near this part of the point cloud. S307. Discretize the selected point cloud at the end of the loop into scattered points; S4. Within the simplified point cloud of all the small grid spaces, find a series of points that are closest to the vector plane; S5. Fit a series of points extracted in S4 into a line to obtain the geological structure outcrop line of the entire spatial range.

2. The method for estimating geological structure outcrop lines based on three-dimensional field scenes according to claim 1, characterized in that, S1 includes: S101. Obtain the oblique photography of the scene from which the exposed line needs to be extracted, and convert the oblique photography into a point cloud (.ply); S102, Remove outliers; S103. Remove non-surface points, including vertical features and noise points below the surface.

3. The method for estimating geological structure outcrop lines based on three-dimensional field scenes according to claim 2, characterized in that, S2 include: S201. Read the location and orientation information of the outcrop line; S202. Based on the location and attitude information of the outcrop line, calculate the plane equation in the coordinate system relative to the point cloud space, and limit the plane range according to the structural extension distance. The plane equation is as follows: ； In the formula, For the tendency (0°-360°), The tilt angle is (0°-90°). The spatial location of a known geological outcrop; S203. Load the finite plane of the orientation into the same space of the point cloud.

4. The method for estimating geological structure outcrop lines based on three-dimensional field scenes according to claim 3, characterized in that, S4 include: S401. Map the scattered points output by S3 back to the three-dimensional space according to their original height, and denote it as dataset D'. S402. Based on actual needs, set a threshold for the distance between the scattered points and the plane. ; S403. In dataset D', traverse every point in each grid in breadth-first search, and calculate the distance d from each point to the plane using the plane equation: ； In the formula, For the tendency (0°-360°), The tilt angle is (0°-90°). Given the spatial location of a known geological outcrop. For points in dataset D'; S404, Extracting cases where the calculated distance d is less than the distance threshold. A series of scattered points, denoted as dataset D''.

5. The method for estimating geological structure outcrop lines based on three-dimensional field scenes according to claim 4, characterized in that, S5 include: S501. Select P1, the point closest to the edge in dataset D'', and select P2, the point closest to it in dataset D''. Connect P1 and P2. S502, divided by the previous point P n-1 In addition, at the current point P n Find the nearest point P in dataset D''. n+1 P n P n+1 Make a connection; S503, repeat S502 until all points are connected repeatedly, then stop connecting the lines. The resulting polyline is the geological outcrop line on the oblique photography model.

6. A system for estimating geological structure outcrop lines based on three-dimensional field scenes, characterized in that, include: Preprocessing module: Based on the oblique photography of the exposed line scene that needs to be extracted, extract the corresponding point cloud and perform preprocessing; Dip Plane Generation Module: Based on the information of the geological structure outcrop, the dip plane is generated in the point cloud space; Fine Mesh Finding Module: This module performs mesh generation on the point cloud and vector plane, identifying the fine mesh areas shared by both. Specifically, it includes: S301. Project the point cloud and spatial plane onto the horizontal plane XOY in a vertical direction; after the spatial plane is projected onto the horizontal plane, a parallelogram-shaped region is formed. Traverse the spatial point cloud using a two-dimensional quadtree; denoted as D for the projected planar point cloud dataset. S302. Determining the outermost region: Traverse all points in D to obtain the maximum values ​​in the X and Y directions. , and minimum value , The coordinates of the four intersection points are obtained: , , , The projected planar point cloud is included within this rectangular grid. S303. Connect the midpoints of opposite sides of the previous rectangular grid, dividing this rectangular grid into 4 smaller grids. The grid is encoded using a linear quadtree encoding. Any page node is encoded using an n-digit code. For a regularly partitioned quadtree space, a quaternary encoding sequence number is introduced. In grid partitioning, a grid that is no longer further subdivided is considered a leaf node in the quadtree. Each node in the quadtree uniquely corresponds to a quaternary number M. in, , representing the geometric position within that layer; n Represents the level of the quadtree; S304. After meshing, determine the planar point cloud dataset D and the parallelogram region. The intersection with the edges of each small mesh; recording the mesh status at each node: if the mesh contains any point in the point cloud, it is marked as a full node for the point cloud; if the mesh intersects with the region... If there is an intersection, it is marked as a full node for the region; if both of the above conditions are met, the node is marked as a full node, and the rest of the nodes are empty nodes. S305. Extract each small mesh that contains both point cloud and vector plane, and discard other empty node meshes; S306. Set the grid length and width thresholds. Take each small grid from the previous step as the initial rectangular grid. Repeat S303 to S305 until the size of all small grids is smaller than the threshold. Record the grids contained within all small grids. The intersection of the plane and the point cloud will be generated near this part of the point cloud. S307. Discretize the selected point cloud at the end of the loop into scattered points; Series Point Search Module: Within a simplified point cloud of all fine grid spaces, find a series of points that are closest to the vector plane; Fitting module: Fits a series of points extracted from the series point search module into a line to obtain the geological structure outcrop line of the entire spatial range.

7. The geological structure outcrop line estimation system based on three-dimensional field reality as described in claim 6, characterized in that, The preprocessing module includes: S101. Obtain the oblique photography of the scene from which the exposed line needs to be extracted, and convert the oblique photography into a point cloud (.ply); S102, Remove outliers; S103. Remove non-surface points, including vertical features and noise points below the surface. The orientation plane generation module includes: S201. Read the location and orientation information of the outcrop line; S202. Based on the location and attitude information of the outcrop line, calculate the plane equation in the coordinate system relative to the point cloud space, and limit the plane range according to the structural extension distance. The plane equation is as follows: ； In the formula, For the tendency (0°-360°), The tilt angle is (0°-90°). The spatial location of a known geological outcrop; S203. Load the finite plane of the orientation into the same space of the point cloud.

8. The geological structure outcrop line estimation system based on three-dimensional field reality as described in claim 7, characterized in that, The series of point search modules include: S401. Map the scattered points output by the fine grid search module back to the three-dimensional space according to their original height, and denote it as dataset D'. S402. Based on actual needs, set a threshold for the distance between the scattered points and the plane. ; S403. In dataset D', traverse every point in each grid in breadth-first search, and calculate the distance d from each point to the plane using the plane equation: ； In the formula, For the tendency (0°-360°), The tilt angle is (0°-90°). Given the spatial location of a known geological outcrop. For points in dataset D'; S404, Extracting cases where the calculated distance d is less than the distance threshold. A series of scattered points, denoted as dataset D''.

9. The geological structure outcrop line estimation system based on three-dimensional field reality as described in claim 8, characterized in that, The fitting module includes: S501. Select P1, the point closest to the edge in dataset D'', and select P2, the point closest to it in dataset D''. Connect P1 and P2. S502, divided by the previous point P n-1 In addition, at the current point P n Find the nearest point P in dataset D''. n+1 P n P n+1 Make a connection; S503, repeat S502 until all points are connected repeatedly, then stop connecting the lines. The resulting polyline is the geological outcrop line on the oblique photography model.

Images