Automatic flattening method, cataloging method, electronic equipment and storage medium for underground caverns

Abstract

The present invention discloses an automatic flattening method, a cataloging method, an electronic device and a storage medium for an underground cavern. The automatic flattening method uses a plane fitting method to fit a cutting plane, automatically determines a flattening reference point according to the cutting plane, and automatically divides the cavern model into regions according to the flattening reference point and spatial parameters. Compared with the traditional manual method, the method reduces the influence of random errors and realizes full-process automated one-button operation. Based on the theory of unfolding point clouds along broken lines and the principle of three-dimensional space geometric transformation, the cavern top area and the right wall area are rotated to a vertical plane in the elevation direction of the left wall, or the cavern top area and the left wall area are rotated to a vertical plane in the elevation direction of the right wall, thereby realizing automatic flattening of the cavern model and solving the problem that, in the process of geological cataloging on a three-dimensional cavern model, some point clouds of geological elements cannot be selected due to the occlusion of the point cloud viewing angle by the cavern arch structure.

Description

Technical Field

[0001] The present invention belongs to the field of three-dimensional point cloud data processing, and in particular relates to an automatic flattening method, a cataloging method, an electronic device and a storage medium for underground caverns based on three-dimensional point cloud data. Background Art

[0002] Traditional manual geological cataloging methods have major problems such as high risk, time-consuming and labor-intensive, highly subjective, difficult to verify, low accuracy, low efficiency, large manual limitations, and low level of informatization. There is an urgent need for intelligent technical solutions that can reduce the manpower input in engineering geological surveys and improve the efficiency of geological cataloging operations.

[0003] By constructing a three-dimensional model of the cave to record the real geological scene of the cave, we can effectively solve the problems of low efficiency, high risk, insufficient manpower and low recording accuracy of traditional manual geological recording methods, and greatly simplify the on-site workload of geological survey personnel.

[0004] 3D point cloud models offer the advantages of high precision, high density, and a true-to-scale reproduction of cavern scenes. By aligning the geospatial coordinates of reference points, a correspondence can be established between the 3D point cloud model and the cavern spatial coordinates, directly obtaining the 3D spatial coordinates of each location within the cavern. Joints and rock mass structural surfaces constructed on the 3D cavern model can be quickly and accurately extracted from the 3D point cloud data to quickly and accurately extract linear feature strike and dip, geometric parameters, and occurrence information. Therefore, the use of 3D point cloud data for caverns in real-world modeling and geological cataloging is gaining widespread application.

[0005] Because 3D cavern models are typically arched structures, cavern texture details are primarily visible on the inner wall surfaces, limiting the viewing angle. This results in a complex real-world cataloging operation. Furthermore, mapping geological cataloging data requires dividing the cavern into three sections: the left wall, the roof, and the right wall. The geological cataloging information on each section of the 3D cavern model must then be projected onto a 2D cataloging map. Currently, there is no effective solution for preserving the spatial correspondence between the point cloud data of the 3D cavern model, constructing a spatially flattened 3D model visual effect, and simultaneously reducing manual operation workload, minimizing the instability of manual intervention, and rapidly and automatically obtaining a robust flattened cavern model. Summary of the Invention

[0006] The purpose of the present invention is to provide an automatic flattening method, cataloging method, electronic device and storage medium for underground caverns, so as to solve the problem that some point clouds of geological elements cannot be selected due to the occlusion of the point cloud perspective of the arch structure during the cataloging process of the cavern model, and the problem that the cataloging efficiency is low due to the large workload of manual operation.

[0007] The present invention solves the above technical problems through the following technical solutions: a method for automatically flattening an underground cavern, comprising the following steps:

[0008] Acquire the real-scene image data of the cave, and perform 3D reconstruction based on the real-scene image data to obtain the original 3D cave model and its point cloud dataset S;

[0009] Obtain the axis unit vector of the original three-dimensional cave model and calculate the axis unit vector space converted into the normal vector The rotation matrix of

[0010] Rotate each point in the point cloud dataset S according to the rotation matrix to obtain a rotated three-dimensional cave model and its point cloud dataset S′, wherein the coordinates of each point in the point cloud dataset S′ are based on a target coordinate system, wherein the target coordinate system is a coordinate system with an extension line of a cave cross section of the three-dimensional cave model as the X-axis, a cave depth direction as the Y-axis, and a cave elevation direction as the Z-axis;

[0011] Fitting a dividing plane between the cave roof and the cave wall of the three-dimensional cave model;

[0012] Determine the coordinates of the flattening reference point according to the cutting plane and the three-dimensional cave model;

[0013] Calculating the spatial parameters of the three-dimensional cave model according to the flattened reference points;

[0014] Dividing the three-dimensional cave model into a left wall area, a cave top area, and a right wall area according to the flattening reference point and the spatial parameter;

[0015] The left wall area remains unchanged, and the cave roof area and the right wall area are flattened to the plane where the left wall area is located according to the target coordinate system and the flattening reference point. The left wall area, the flattened cave roof area and the right wall area are integrated to obtain a flattened three-dimensional cave model and its point cloud dataset S″;

[0016] Alternatively, the right wall area remains unchanged, and the cave roof area and the left wall area are flattened to the plane where the right wall area is located according to the target coordinate system and the flattening reference point. The right wall area, the flattened cave roof area, and the left wall area are integrated to obtain the flattened three-dimensional cave model and its point cloud dataset S″.

[0017] Furthermore, the calculation formula of the rotation matrix is:

[0018]

[0019] Where R represents the rotation matrix; I represents the identity matrix; Represents the axis unit vector With normal vector The unit vector corresponding to the cross product result; cosθ represents the axis unit vector With normal vector The cosine value of the rotation angle obtained by dot multiplication,

[0020] Furthermore, the cutting plane is fitted using the method of minimizing the sum of squared errors. The specific implementation process of the fitting is as follows:

[0021] Assume that any point p in the point cloud dataset S′ i 'The coordinates in the target coordinate system are (x i ,y i ,z i ), where i = 1, 2, 3, ..., n, and n represents the number of points in the point cloud dataset S′;

[0022] The general form of the plane equation Ax+By+Cz+D=0 is transformed into z=a0x+a1y+a2, and the objective function Q is constructed. The expression of the objective function Q is:

[0023]

[0024] Where a0, a1, and a2 are the coefficients of the partition plane equation, and (x, y, z) are the coordinates of the points on the partition plane.

[0025] Solving minQ, we can obtain a0, a1, and a2, and thus obtain the equation of the partition plane z = a0x + a1y + a2.

[0026] Furthermore, the flattening reference points include the upper left point of the cave, the upper left rear point of the cave, the upper right point of the cave, the lower right point of the cave, and the cave vertex. The coordinates of the flattening reference points are determined as follows:

[0027] Calculate the coordinates of all intersection points between the segmentation plane and the point cloud dataset S′;

[0028] In the intersection point data set, the point that satisfies the minimum X-axis coordinate value and the minimum Y-axis coordinate value is the upper left point of the cave;

[0029] In the intersection point data set, the point with the maximum X-axis coordinate value and the minimum Y-axis coordinate value is the upper right point of the cave;

[0030] In the intersection point data set, the point with the minimum X-axis coordinate value and the maximum Y-axis coordinate value is the upper left rear point of the cave;

[0031] The point with the largest elevation value or the largest Z-axis coordinate value at the front end of the three-dimensional cave model is the cave vertex;

[0032] The point with the smallest elevation value of the right wall at the front end of the cavern or the smallest Z-axis coordinate value in the three-dimensional cavern model is the lower right point of the cavern.

[0033] Preferably, the space parameters include the height H of the cave wall, the height H of the cave top, and the height H of the cave wall. t , cavern width W, cavern length L, cavern ground elevation B; the specific calculation formula of the spatial parameters is:

[0034] H=(z1+z2) / 2-z min , H t =z max -Hz min

[0035] L=y3-y1,B=z min

[0036] Among them, (x1, y1, z1) is the coordinate of the upper left point of the cave in the target coordinate system, (x2, y2, z2) is the coordinate of the upper right point of the cave in the target coordinate system, (x3, y3, z3) is the coordinate of the upper left rear point of the cave in the target coordinate system, and z max is the maximum value of the Z-axis coordinate value in the point cloud dataset S′, z min is the minimum value of the Z-axis coordinate in the point cloud dataset S′.

[0037] Furthermore, the specific implementation process of flattening the cave top area to the surface where the left wall area is located is as follows:

[0038] In the XY plane of the target coordinate system, a general equation of an ellipse is constructed based on the flattened reference point so that all points in the cave roof area satisfy the general equation of the ellipse. The expression of the general equation of the ellipse is:

[0039]

[0040] Where W represents the width of the cave, H represents the height of the cave wall, B represents the elevation of the cave floor, and (x, y, z) represents the coordinates of a point in the cave ceiling area.

[0041] The cave roof fan-shaped broken line is determined according to the general equation of the ellipse and the flattened reference point. The fitting equation of the cave roof fan-shaped broken line is:

[0042] x1<x<x2,y≥y1

[0043] Among them, (x1, y1, z1) is the coordinate of the upper left point of the cave in the target coordinate system, and (x2, y2, z2) is the coordinate of the upper right point of the cave in the target coordinate system;

[0044] Calculate the vertical distance h from each point in the cave roof area to the cave roof fan-shaped broken line and the arc length l from the foot of the perpendicular corresponding to the point to the left starting point of the cave roof fan-shaped broken line;

[0045] The cave roof fan-shaped polyline is stretched and projected onto the XY plane, and the spatial relationship between each point in the cave roof area and the stretched and projected cave chamber fan-shaped polyline is restored according to the vertical distance h and the arc length l to obtain a flattened cave roof area point cloud dataset;

[0046] Each point in the flattened cave roof area point cloud dataset is rotated 90° counterclockwise along the Y axis and translated above the left wall area to obtain the final flattened cave roof area.

[0047] Furthermore, the specific implementation process of flattening the right wall area to the surface where the left wall area is located is:

[0048] The points in the right wall area are flattened to above the flattened cave top area according to the coordinate transformation formula. The coordinate transformation formula is:

[0049] (x',y',z')=(Wx,y,B+H+len+H-z+B)

[0050]

[0051] Where (x', y', z') represents the coordinates of the point in the right wall area after flattening, (x, y, z) represents the coordinates of the point in the right wall area before flattening, W represents the width of the cavern, H represents the height of the cavern wall, B represents the elevation of the cavern ground, and len represents the total arc length of the section in the cavern top area along the X-axis.

[0052] Furthermore, the automatic flattening method further includes:

[0053] The flattened three-dimensional cave model is reversely rotated according to the rotation matrix so that the unchanged area completely overlaps with the corresponding area of the original three-dimensional cave model, thereby obtaining the final cave model and its point cloud dataset S′″.

[0054] Based on the same inventive concept, the present invention further provides a geological cataloging method, which performs geological cataloging based on the flattened three-dimensional cavern model and its point cloud dataset S″ obtained by the above-mentioned automatic flattening method for underground caverns.

[0055] Based on the same inventive concept, the present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor runs the computer program, it executes the steps of the above-mentioned automatic flattening method of underground caverns or the geological cataloging method.

[0056] Based on the same inventive concept, the present invention also provides a computer-readable storage medium, which is a non-volatile storage medium or a non-transient storage medium, on which a computer program is stored. When the computer program is run by a processor, it executes the steps of the above-mentioned automatic flattening method of underground caverns or the geological cataloging method.

[0057] Beneficial effects

[0058] Compared with the prior art, the advantages of the present invention are:

[0059] The automatic flattening and cataloging method for underground caverns provided by the present invention automatically determines the flattening reference points based on the cutting plane, and automatically divides the cavern model into regions based on the flattening reference points and spatial parameters. Compared with the traditional method that requires prior conditions and manual intervention, this method reduces the impact of random errors and realizes one-click automation of the entire process.

[0060] Based on the theory of unfolding point clouds along broken lines and the principle of three-dimensional spatial geometric transformation, the cave roof area and right wall area are rotated to the vertical elevation in the left wall elevation direction, or the cave roof area and left wall area are rotated to the vertical elevation in the right wall elevation direction, realizing automatic flattening of the cave model. This solves the problem of some geological element point clouds being unable to be selected due to the perspective occlusion of the cave roof arch structure point cloud during geological cataloging on the three-dimensional cave model, making it easier for survey workers to identify and draw geological structural elements in real time more quickly. The cave model flattening process is also the geological element flattening process. After drawing geological elements on the flattened cave model, a two-dimensional geological cataloging sketch can be automatically generated, avoiding the problem of inconsistency between cataloging on the three-dimensional cave model and the traditional two-dimensional geological cataloging map. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only one embodiment of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0062] Figure 1 This is a flow chart of the automatic flattening method for underground caverns according to an embodiment of the present invention;

[0063] Figure 2 is a front view of the original three-dimensional cave model in an embodiment of the present invention;

[0064] Figure 3 is a side view of the original three-dimensional cave model in an embodiment of the present invention;

[0065] Figure 4is a front view of the dividing plane between the cave ceiling and the cave wall in an embodiment of the present invention;

[0066] Figure 5 is a side view of a cutting plane between a cave ceiling and a cave wall in an embodiment of the present invention;

[0067] Figure 6 2 is a schematic diagram of the position of the flattening reference point in an embodiment of the present invention;

[0068] Figure 7 This is a fitting effect diagram of the cave roof fan-shaped broken line in an embodiment of the present invention;

[0069] Figure 8 This is a flattening effect diagram of the cave roof area along the cave roof fan-shaped broken line in an embodiment of the present invention;

[0070] Figure 9 This is a diagram of a cave model after flattening in an embodiment of the present invention;

[0071] Figure 10 3 is a comparison diagram of the flattened cave model and the original three-dimensional cave model in an embodiment of the present invention. DETAILED DESCRIPTION

[0072] The following is a clear and complete description of the technical solutions of the present invention in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts are within the scope of protection of the present invention.

[0073] The following specific embodiments are used to describe the technical solution of the present application in detail. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0074] The present invention provides an automatic flattening method for an underground cavern, which uses a point cloud plane fitting algorithm to automatically divide the left wall area, the right wall area and the cave roof area, and then uses the general equation of an ellipse and the unfolding method along a broken line to automatically rotate and transform the cave roof area and the right wall area into the vertical plane of the left wall area in the Z-axis direction, or rotate and transform the cave roof area and the left wall area into the vertical plane of the right wall area in the Z-axis direction. The entire flattening process is automated, greatly improving efficiency.

[0075] This embodiment takes the rotation transformation of the cave top area and the right wall area to the vertical plane of the left wall area in the Z-axis direction as an example, and uses the real cave scene modeling data for flattening processing, such as Figure 1 As shown, the automatic flattening method for underground caverns specifically includes the following steps:

[0076] 1. Construction of the original 3D cave model:

[0077] The real-scene image data of the cave is obtained, and the original 3D cave model and its point cloud dataset S are obtained based on the 3D reconstruction of the real-scene image data of the cave. Figure 2 and 3 The original 3D cave model from different perspectives is shown, where the coordinates of each point in the point cloud dataset S are based on the WGS-84 coordinate system.

[0078] The point cloud dataset S can be expressed as S = {p1,p2,p3,…,p i ,…,p n}, where n is the number of points in the point cloud dataset S, p i is the i-th point in the point cloud dataset S. In this embodiment, the point cloud dataset S has 314183 points.

[0079] The three-dimensional reconstruction of the cave model is an existing technology, and reference may be made to a method and device for three-dimensional modeling of an underground cave (authorization announcement number CN104933763B).

[0080] 2. Calculate the rotation matrix:

[0081] Get the axis unit vector of the original 3D cave model and calculate the axis unit vector space into the normal vector The rotation matrix of .

[0082] Before modeling the cavern, the axis needs to be designed first. Therefore, the axis unit vector can be directly obtained from the original three-dimensional cavern model, which is recorded as is the axis unit vector Coordinates in the WGS-84 coordinate system. (0,1,0) is the normal vector Coordinates in the target coordinate system, normal vector Points to the Y axis of the target coordinate system.

[0083] In this embodiment, the calculation formula of the rotation matrix is:

[0084]

[0085] Where R represents the rotation matrix; I represents the identity matrix; Represents the axis unit vector With normal vector The unit vector corresponding to the cross product result; because are all unit vectors, and their modulus is 1, so cosθ represents the axis unit vector With normal vector The cosine value of the rotation angle obtained by dot multiplication,

[0086] 3. Rotation transformation of the original 3D cave model:

[0087] According to the rotation matrix R, each point in the point cloud dataset S is rotated, that is, the original three-dimensional cave model is rotated to obtain the rotated three-dimensional cave model and its point cloud dataset S′, which is expressed as S′={p′1,p′2,p′3,…,p′ i ,…,p′ n}, each point p′ in the point cloud dataset S′ i The coordinates of are based on the target coordinate system. The target coordinate system takes the extension line of the cavern section of the three-dimensional cavern model as the X-axis, the cavern depth direction as the Y-axis, and the cavern elevation direction as the Z-axis. That is, the extension line of the cavern section of the rotated three-dimensional cavern model overlaps or is parallel to the X-axis of the target coordinate system, the cavern depth direction overlaps or is parallel to the Y-axis of the target coordinate system, and the cavern elevation direction overlaps or is parallel to the Z-axis of the target coordinate system.

[0088] The subsequent steps are all based on the target coordinate system.

[0089] 4. Fitting of the cutting plane:

[0090] In this embodiment, the method of minimizing the sum of squared errors is used to fit the dividing plane between the top and the wall of the three-dimensional cave model. The specific implementation process of the fitting is as follows:

[0091] (4.1) Assume that any point p in the point cloud dataset S′ i 'The coordinates in the target coordinate system are (x i ,y i ,z i ), where i = 1, 2, 3, ..., n, and n represents the number of points in the point cloud dataset S′.

[0092] (4.2) The general form of the plane equation Ax+By+Cz+D=0 is transformed into z=a0x+a1y+a2, and the objective function Q is constructed. The expression of the objective function Q is:

[0093]

[0094] Where a0, a1, and a2 represent the coefficients of the partition plane equation, and (x, y, z) represent the coordinates of a point on the partition plane.

[0095] (4.3) To solve minQ, it should satisfy k=0,1,2 to get a0, a1, a2, and thus get the equation of the cutting plane z=a0x+a1y+a2. It is required that the average distance from all points in the point cloud dataset S′ to the cutting plane is the smallest (that is, the sum of the distances is the smallest), that is, to solve minQ, such as Figure 4 and 5The dividing plane between the cave ceiling and the cave wall is shown from different perspectives.

[0096] 5. Automatic determination of flattening reference point:

[0097] The coordinates of the flattening reference point can be automatically determined based on the cutting plane and the three-dimensional cave model.

[0098] In this embodiment, the flattening reference points include the upper left point of the cave, the upper left rear point of the cave, the upper right point of the cave, the lower right point of the cave, and the cave vertex (such as Figure 6 As shown), the process of determining the coordinates of the flattened reference point is:

[0099] (5.1) Based on the dividing plane equation z = a0x + a1y + a2, traverse all points in the point cloud dataset S′ and find the coordinates of the points where the dividing plane intersects with the point cloud dataset S′, that is, the coordinates of the intersection points of the dividing plane and the point cloud dataset S′. The coordinates of the intersection points are based on the target coordinate system.

[0100] (5.2) In the intersection point data set, the point with the smallest X-axis coordinate value and the smallest Y-axis coordinate value is the upper left point of the cavern. That is, the point with the smallest X-axis coordinate value and the smallest Y-axis coordinate value is selected from the intersection point data set as the upper left point of the cavern.

[0101] The upper left point of the cave is the upper left point where the front wall and the top of the cave meet, denoted as P zs (x1, y1, z1), (x1, y1, z1) represents the upper left point P of the cave zs In the target coordinate system, x1=x min , y1=y min , x min Indicates the minimum value of the X-axis coordinate value in the intersection point data set, y min Indicates the minimum Y-axis coordinate value in the intersection point dataset.

[0102] (5.3) In the intersection point data set, the point with the maximum X-axis coordinate value and the minimum Y-axis coordinate value is the upper right point of the cavern. That is, the point with the maximum X-axis coordinate value and the minimum Y-axis coordinate value is selected from the intersection point data set as the upper right point of the cavern.

[0103] The upper right point of the cave is the upper right point P where the front wall of the cave and the top of the cave meet. ys , denoted as P ys (x2, y2, z2), (x2, y2, z2) represents the upper right point P of the cave ys In the target coordinate system, x2=x max , y2=y min , x max Indicates the maximum X-axis coordinate value in the intersection point dataset.

[0104] (5.4) In the intersection point data set, the point with the minimum X-axis coordinate value and the maximum Y-axis coordinate value is the upper left rear point of the cavern. That is, the point with the minimum X-axis coordinate value and the maximum Y-axis coordinate value is selected from the intersection point data set as the upper left rear point of the cavern.

[0105] The left upper rear point of the cave is the left upper rear point P where the rear wall and the cave roof meet. hs , denoted as P hs (x3, y3, z3), (x3, y3, z3) represents the upper left rear point P of the cave ys In the target coordinate system, x3 = x min , y3=y max ,y max Indicates the maximum Y-axis coordinate value in the intersection point dataset.

[0106] (5.5) In the three-dimensional cave model, the point with the maximum elevation value at the front end of the cave or the maximum Z-axis coordinate value is the cave vertex, denoted by P dd (avg(x1,x2),avg(y1,y2),z max ), where avg() means finding the average value, z max is the maximum value of the Z-axis coordinate value in the point cloud dataset S′, z min is the minimum value of the Z-axis coordinate in the point cloud dataset S′.

[0107] (5.6) In the three-dimensional cave model, the point with the smallest elevation value on the right wall of the front end of the cave or the smallest Z-axis coordinate value is the lower right point of the cave, denoted as P yx (x2,y2,z min ), the X-axis coordinate value and Y-axis coordinate value of the lower right point of the cave are the same as those of the upper right point of the cave.

[0108] 6. Calculation of spatial parameters:

[0109] The spatial parameters of the three-dimensional cave model are calculated based on the flattened reference point. The spatial parameters include the height of the cave wall H, the height of the cave top H t , cavern width W, cavern length L, cavern ground elevation B. The specific calculation formula for spatial parameters is:

[0110] H=(z1+z2) / 2-z min (3)

[0111] H t =z max -Hz min (4)

[0112]

[0113] L=y3-y1 (6)

[0114] B=zmin (7)

[0115] 7. Regional division:

[0116] The 3D cave model is divided into the left wall area, the cave roof area, and the right wall area based on the flattened reference points and spatial parameters. Among them, all points with Z-axis coordinate values ≤ H+B and X-axis coordinate values < W / 2 are divided into the left wall area; all points with Z-axis coordinate values ≤ H+B and X-axis coordinate values > W / 2 are divided into the right wall area; and the remaining points are divided into the cave roof area.

[0117] The present invention can automatically fit the cutting plane, automatically determine the flattening reference point and spatial parameters, and automatically divide the model into the left wall area, the right wall area and the cave top area according to the flattening reference point and spatial parameters. Compared with the traditional method of manually determining the flattening reference point and dividing the area, the present invention reduces the impact of random errors and realizes full-process automated one-click operation.

[0118] 8. Flattening of the cave roof and right wall:

[0119] For the point cloud data of different areas, spatial transformation processing is performed separately. The main operation is to keep the coordinates of the points in the left wall area unchanged, expand the points in the cave roof area along the cave roof fan-shaped line and project them above the left wall area, and rotate the points in the right wall area to the top of the expanded cave roof area. Finally, all points are rotated and flattened to the YZ plane with relatively small spatial deformation. The flattening effect is as follows: Figure 9 and 10 shown.

[0120] For the flattening of the cave roof area, the specific implementation process is as follows:

[0121] (8.1) In the XY plane of the target coordinate system, the general equation of the ellipse is constructed based on the upper left point, the upper right point and the top point of the cave, so that all points in the cave top area satisfy the general equation of the ellipse. The expression of the general equation of the ellipse is:

[0122]

[0123] Where W represents the width of the cave, H represents the height of the cave wall, B represents the elevation of the cave floor, (x, y, z) represents the coordinates of the point in the cave ceiling area, and the center point of the ellipse is (W / 2, 0, H+B).

[0124] (8.2) Determine the cave roof fan-shaped line based on the general equation of the ellipse and the coordinates of the upper left point, upper right point and cave vertex (the fitting effect is as follows Figure 7 As shown in Figure 2), the fitting equation of the cave roof fan-shaped broken line is:

[0125] x1<x<x2,y≥y1

[0126] Among them, (x1, y1, z1) is the coordinate of the upper left point of the cave in the target coordinate system, and (x2, y2, z2) is the coordinate of the upper right point of the cave in the target coordinate system.

[0127] (8.3) Calculate the vertical distance h from each point in the cave roof area to the cave roof fan-shaped broken line and the arc length l from the foot of the perpendicular corresponding to the point to the left starting point of the cave roof fan-shaped broken line; the left starting point refers to the starting point on the cave roof fan-shaped broken line on the same side as the left wall area.

[0128] (8.4) The cave roof fan-shaped polyline is stretched and projected onto the XY plane, and the spatial relationship between each point in the cave roof area and the stretched and projected cave chamber fan-shaped polyline is restored based on the vertical distance h and the arc length l to obtain the flattened cave roof area point cloud dataset, as shown in Figure 8 At this point, although the cave roof is flattened, it is not on the same plane as the left wall.

[0129] (8.5) Each point in the flattened cave roof area point cloud dataset is rotated 90° counterclockwise along the Y axis and translated to the top of the left wall area to obtain the final flattened cave roof area (as shown in Figure 8.5). Figure 9 As shown in Figure 2, at this time, the cave top area and the left wall area are on the same plane.

[0130] After final flattening, the point cloud of the cave ceiling area still maintains the corresponding connection relationship in the three-dimensional space, that is, the spatial topological relationship of the point cloud remains unchanged. In the process of flattening the points in the arch-shaped cave ceiling area into a two-dimensional plane in the three-dimensional space, the deformation is maintained at a small level.

[0131] The present invention fits the arched structure of the cave roof area into an elliptical structure, and achieves the purpose of rotating and unfolding the three-dimensional cave model to a plane while ensuring that the deformation error of the local three-dimensional structural characteristics is small. This makes it convenient for survey workers to catalog geological elements on the three-dimensional cave model, realizes portable interactive operation of the three-dimensional cave model, and finally the cataloged geological structure elements can be displayed back to the original three-dimensional cave model.

[0132] For the flattening of the right wall area, the specific implementation process is:

[0133] Traverse all points in the right wall area and flatten the points in the right wall area to the top of the flattened cave roof area by flipping and translating. The coordinate transformation formula of each point in the right wall area is:

[0134] (x',y',z')=(Wx,y,B+H+len+H-z+B) (9)

[0135]

[0136] Where (x', y', z') represents the coordinates of the point in the right wall area after flattening, (x, y, z) represents the coordinates of the point in the right wall area before flattening, W represents the width of the cavern, H represents the height of the cavern wall, B represents the elevation of the cavern ground, and len represents the total arc length of the section in the cavern top area along the X-axis.

[0137] The left wall area remains unchanged, and the left wall area, the flattened cave top area and the right wall area are integrated to obtain the flattened three-dimensional cave model (such as Figure 9 As shown) and its point cloud dataset S″, expressed as S″={p″1,p″2,p″3,…,p″ i ,…,p″ n}.

[0138] 9. Reverse rotation:

[0139] The flattened 3D cave model is reversely rotated in the XY plane according to the rotation matrix R, so that the unchanged area completely coincides with the corresponding area of the original 3D cave model. The final cave model and its point cloud dataset S′′ are obtained, which can be expressed as S′′={p′′1,p′′2,p′′3,…,p′′ i ,…,p″′ n}.

[0140] For example, when the left wall area remains unchanged, the cave roof area and the right wall area are flattened to the plane where the left wall area is located. Then, during the reverse rotation, the flattened cave point cloud model is reversely rotated in the XOY plane of the target coordinate system so that the left wall area completely overlaps with the left wall area of the original 3D cave model, as shown in FIG. Figure 10 shown.

[0141] Based on the same inventive concept, an embodiment of the present invention further provides a geological cataloging method, which performs geological cataloging based on the flattened three-dimensional cavern model and its point cloud dataset S″ obtained by the above-mentioned automatic flattening method for underground caverns.

[0142] The present invention provides survey workers with tools for drawing geological points, lines, and surface elements. It fits geological elements such as joints, cracks, and structural surfaces through point clouds, and simultaneously quickly and accurately calculates the inclination and dip information of linear geological elements and the occurrence information of geological structural surface elements.

[0143] The above disclosure is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or modifications within the technical scope disclosed in the present invention, and they should all be covered by the scope of protection of the present invention.

Claims

1. A method for automatically flattening an underground cavern, characterized in that: The following steps are involved: Acquire the real-scene image data of the cave, and perform 3D reconstruction based on the real-scene image data to obtain the original 3D cave model and its point cloud dataset S; Obtain the axis unit vector of the original three-dimensional cave model and calculate the axis unit vector space converted into the normal vector The rotation matrix of Rotate each point in the point cloud dataset S according to the rotation matrix to obtain a rotated three-dimensional cave model and its point cloud dataset S′, wherein the coordinates of each point in the point cloud dataset S′ are based on a target coordinate system, wherein the target coordinate system is a coordinate system with an extension line of a cave cross section of the three-dimensional cave model as the X-axis, a cave depth direction as the Y-axis, and a cave elevation direction as the Z-axis; Fitting a dividing plane between the cave roof and the cave wall of the three-dimensional cave model; Determine the coordinates of the flattening reference point according to the cutting plane and the three-dimensional cave model; Calculating the spatial parameters of the three-dimensional cave model according to the flattened reference points; Dividing the three-dimensional cave model into a left wall area, a cave top area, and a right wall area according to the flattening reference point and the spatial parameter; The left wall area remains unchanged, and the cave roof area and the right wall area are flattened to the plane where the left wall area is located according to the target coordinate system and the flattening reference point. The left wall area, the flattened cave roof area and the right wall area are integrated to obtain a flattened three-dimensional cave model and its point cloud dataset S″; Alternatively, the right wall area remains unchanged, and the cave roof area and the left wall area are flattened to the plane where the right wall area is located according to the target coordinate system and the flattening reference point. The right wall area, the flattened cave roof area, and the left wall area are integrated to obtain the flattened three-dimensional cave model and its point cloud dataset S″.

2. The automatic flattening method of underground caverns according to claim 1, characterized in that: The calculation formula of the rotation matrix is: Where R represents the rotation matrix; I represents the identity matrix; Represents the axis unit vector With normal vector The unit vector corresponding to the cross product result; cosθ represents the axis unit vector With normal vector The cosine value of the rotation angle obtained by dot multiplication, 3. The automatic flattening method of underground caverns according to claim 1, characterized in that: The cutting plane is fitted using the method of minimizing the sum of squared errors. The specific implementation process of the fitting is as follows: Assume that any point p in the point cloud dataset S′ i 'The coordinates in the target coordinate system are (x i ,y i ,z i ), where i = 1, 2, 3, ..., n, and n represents the number of points in the point cloud dataset S′; The general form of the plane equation Ax+By+Cz+D=0 is transformed into z=a0x+a1y+a2, and the objective function Q is constructed. The expression of the objective function Q is: Where a0, a1, and a2 are the coefficients of the partition plane equation, and (x, y, z) are the coordinates of the points on the partition plane. Solving minQ, we can obtain a0, a1, and a2, and thus obtain the equation of the partition plane z = a0x + a1y + a2.

4. The automatic flattening method of underground caverns according to claim 1, characterized in that: The flattening reference points include the upper left point of the cave, the upper left rear point of the cave, the upper right point of the cave, the lower right point of the cave, and the cave vertex. The coordinates of the flattening reference points are determined as follows: Calculate the coordinates of all intersection points between the segmentation plane and the point cloud dataset S′; In the intersection point data set, the point that satisfies the minimum X-axis coordinate value and the minimum Y-axis coordinate value is the upper left point of the cave; In the intersection point data set, the point with the maximum X-axis coordinate value and the minimum Y-axis coordinate value is the upper right point of the cave; In the intersection point data set, the point with the minimum X-axis coordinate value and the maximum Y-axis coordinate value is the upper left rear point of the cave; The point with the largest elevation value or the largest Z-axis coordinate value at the front end of the three-dimensional cave model is the cave vertex; The point with the smallest elevation value or the smallest Z-axis coordinate value on the right wall of the front end of the three-dimensional cave model is the lower right point of the cave; Preferably, the space parameters include the height H of the cave wall, the height H of the cave top, and the height H of the cave wall. t , cavern width W, cavern length L, cavern ground elevation B; the specific calculation formula of the spatial parameters is: H=(z1+z2) / 2-z min ,H t =z max -Hz min L=y3-y1,B=z min Among them, (x1, y1, z1) is the coordinate of the upper left point of the cave in the target coordinate system, (x2, y2, z2) is the coordinate of the upper right point of the cave in the target coordinate system, (x3, y3, z3) is the coordinate of the upper left rear point of the cave in the target coordinate system, and z max is the maximum value of the Z-axis coordinate value in the point cloud dataset S′, z min is the minimum value of the Z-axis coordinate in the point cloud dataset S′.

5. The automatic flattening method for underground caverns according to any one of claims 1 to 4, characterized in that: The specific implementation process of flattening the cave top area to the surface where the left wall area is located is as follows: In the XY plane of the target coordinate system, a general equation of an ellipse is constructed based on the flattened reference point so that all points in the cave roof area satisfy the general equation of the ellipse. The expression of the general equation of the ellipse is: Where W represents the width of the cave, H represents the height of the cave wall, B represents the elevation of the cave floor, and (x, y, z) represents the coordinates of a point in the cave ceiling area. The cave roof fan-shaped broken line is determined according to the general equation of the ellipse and the flattened reference point. The fitting equation of the cave roof fan-shaped broken line is: x1<x<x2,y≥y1 Among them, (x1, y1, z1) is the coordinate of the upper left point of the cave in the target coordinate system, and (x2, y2, z2) is the coordinate of the upper right point of the cave in the target coordinate system; Calculate the vertical distance h from each point in the cave roof area to the cave roof fan-shaped broken line and the arc length l from the foot of the perpendicular corresponding to the point to the left starting point of the cave roof fan-shaped broken line; The cave roof fan-shaped polyline is stretched and projected onto the XY plane, and the spatial relationship between each point in the cave roof area and the stretched and projected cave chamber fan-shaped polyline is restored according to the vertical distance h and the arc length l to obtain a flattened cave roof area point cloud dataset; Each point in the flattened cave roof area point cloud dataset is rotated 90° counterclockwise along the Y axis and translated above the left wall area to obtain the final flattened cave roof area.

6. The automatic flattening method for underground caverns according to any one of claims 1 to 4, characterized in that: The specific implementation process of flattening the right wall area to the surface where the left wall area is located is as follows: The points in the right wall area are flattened to above the flattened cave top area according to the coordinate transformation formula. The coordinate transformation formula is: (x',y',z')=(Wx,y,B+H+len+H-z+B) Where (x', y', z') represents the coordinates of the point in the right wall area after flattening, (x, y, z) represents the coordinates of the point in the right wall area before flattening, W represents the width of the cavern, H represents the height of the cavern wall, B represents the elevation of the cavern ground, and len represents the total arc length of the section in the cavern top area along the X-axis.

7. The automatic flattening method of underground caverns according to claim 1, characterized in that: The automatic flattening method further includes: The flattened three-dimensional cave model is reversely rotated according to the rotation matrix so that the unchanged area completely overlaps with the corresponding area of the original three-dimensional cave model, thereby obtaining the final cave model and its point cloud dataset S′″.

8. A geological logging method, characterized in that: The flattened three-dimensional cavern model and its point cloud dataset S″ obtained based on the automatic flattening method of underground caverns according to any one of claims 1 to 7 are used for geological cataloging.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, wherein: When the processor runs the computer program, the processor executes the steps of the automatic flattening method for underground caverns according to any one of claims 1 to 7 or the geological logging method according to claim 8.

10. A computer-readable storage medium, wherein the computer-readable storage medium is a non-volatile storage medium or a non-transient storage medium, and a computer program is stored thereon, wherein: When the computer program is executed by a processor, the steps of the automatic flattening method for underground caverns according to any one of claims 1 to 7 or the geological logging method according to claim 8 are executed.

Images