Three-dimensional laser scanning data processing and geological logging method for underground caverns

Abstract

The present invention discloses a method for processing three-dimensional laser scanning data and geological logging of underground chambers, comprising the following steps: performing on-site three-dimensional geological scanning by arranging control points, determining the scanning method and scanning route; performing data preprocessing by resolving the original data obtained by the scanner and formatting the POS data of panoramic photos; projecting the panoramic photos into a panoramic sphere, matching the point cloud with the panoramic sphere, drawing structural plane trace lines, measuring the occurrence of structural planes and describing the structural planes, and setting the projection plane and exporting the logging results. The method for processing three-dimensional laser scanning data and geological logging of underground chambers provided by the present invention standardizes the arrangement method of control points, scanning method, and data processing method for on-site three-dimensional laser scanning according to the characteristics of underground chambers, can improve the accuracy and reliability of obtaining three-dimensional laser point clouds, ensure the acquisition of high-quality data, and improve the logging efficiency.

Description

Technical Field

[0001] The present invention relates to the field of underground engineering, and more specifically to a method for processing three-dimensional laser scanning data and geological logging of underground chambers. Background Art

[0002] With the development of social economy, water conservancy and hydropower projects such as pumped storage power stations are constantly under construction. The underground powerhouse is one of the main buildings of such water conservancy and hydropower projects. The underground chambers in the underground powerhouse system generally have characteristics such as large burial depth, large span, high side walls, and obvious group effects. There are a large number of uncertain factors in their occurrence environment and construction process, and there may be adverse geological phenomena such as karst, faults and their fracture zones, and long and large joints and fissures. Geological logging of underground chambers such as exploration adits can provide data support for the research of key issues such as the site selection of the underground powerhouse system, the stability analysis and evaluation of the surrounding rock of the chambers, and the optimization of the chamber design scheme.

[0003] At present, the geological logging of underground chambers is mainly obtained through on-site manual logging, and its accuracy is greatly affected by human factors, and the key rock mass structural planes may be overlooked. Moreover, due to the influence of subsequent construction such as the closure and lining of the chambers, the data accuracy cannot be verified again. There are also risks such as roof collapse, rock burst, water inrush, and toxic gases in some underground chambers. Conducting logging on-site for a long time poses certain risks to the lives and safety of geological personnel.

[0004] With the development of computer technology and new technologies, non-contact acquisition methods such as three-dimensional laser scanning and close-range photogrammetry are gradually developed to collect and analyze the information of rock mass structural planes, relying on their convenience advantages. The original data can be collected on-site at one time, and a large amount of logging work can be transferred to indoor processing. It can reduce the time and risks of on-site logging by geological personnel, reduce the impact of geological logging on the subsequent construction progress, and can save the original three-dimensional data for archiving and subsequent verification. However, the geological logging based on SLAM handheld three-dimensional laser scanning lacks a standardized operation method, resulting in poor data acquisition quality, thus reducing the accuracy of the logging results. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for processing three-dimensional laser scanning data and geological logging of underground chambers in view of the above deficiencies of the prior art, to solve the safety problems and poor data accuracy existing in on-site manual logging, and at the same time improve the data quality of the logging method based on three-dimensional laser scanning and improve the logging efficiency.

[0006] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0007] A method for processing three-dimensional laser scanning data and geological logging of underground chambers, comprising the following steps:

[0008] Step 1: On-site three-dimensional geological scanning;

[0009] Relevant information of the underground cavern is obtained through on-site investigation, the structural characteristics of the geological body are initially grasped, control points are arranged, the scanning method and scanning route are determined, and then a scanner is used to perform laser scanning to obtain raw data; the scanner is a SLAM handheld three-dimensional laser scanner equipped with a panoramic camera;

[0010] Step 2: Data preprocessing;

[0011] The raw data obtained by the scanner is initially solved, point cloud registration, and panoramic photo registration are performed to obtain general point cloud data in las format, panoramic photos in jpg format, and POS data after panoramic photo registration; the POS data of the panoramic photos is formatted and saved as a csv file;

[0012] Step 3: Geological logging;

[0013] The point cloud data in las format, panoramic photos in jpg format, and csv files are imported into three-dimensional logging software, the panoramic photos are projected into a panoramic sphere, the three-dimensional point cloud and the panoramic sphere are fused, and then geological logging work is carried out under the panoramic view to record key geological information. Finally, the projection plane is set and the logging results are exported; the logging results include logging data and logging maps;

[0014] The underground caverns include exploration adits, water conveyance tunnels, traffic tunnels, and underground powerhouses.

[0015] Preferably, in Step 1, the control points are fixed by targets and marked with paint; the arrangement of the control points is determined according to the length of the underground cavern, and the determination method is as follows:

[0016] (a) If the length of the underground cavern does not exceed 100m, control points are only arranged outside the cavern;

[0017] (b) If the length of the underground cavern is greater than 100m, control points are arranged both inside and outside the cavern;

[0018] Among them, the coordinates of the control points outside the cavern are measured using RTK equipment or total station, and the distance between the control points is not less than 10m, and the number of control points is 4 - 6; the coordinates of the control points inside the cavern are measured using total station, and the distance between the control points is 50 - 100m.

[0019] Preferably, the distance between the control points outside the cavern is 20 - 30m, preferably 20m, and the number of control points is 5.

[0020] Preferably, in Step 1,

[0021] The specific scanning method is as follows: hold the 3D laser scanner in your right hand, keep the device 20cm away from your body, rotate the scanner so that the front lens faces the direction where you need to take photos; the body is located on the perpendicular midline of the front and rear lenses; when walking to the control point, align the scanner's crosshairs with the control point and wait for 10 seconds to allow the scanner to recognize the control point;

[0022] The specific scanning route is: walk, pass the control points outside the cavern in turn, continue walking, enter the underground cavern, keep the scanner stable in the underground cavern, start scanning along the right wall, reach the end of the cavern, turn slowly to ensure that the scanner can collect relevant data on the upper face, then continue walking to the right and scanning until you return to the starting position and end the scan.

[0023] Preferably, the walking speed is 0.5 m / s.

[0024] Preferably, in step 2, data preprocessing includes data solving and panoramic photo POS data formatting;

[0025] The specific method of data solution is: using the scanner supporting software to perform preliminary solution, point cloud registration, and panoramic photo registration on the original data obtained by the scanner to obtain a three-dimensional laser point cloud and panoramic photo POS data with correct spatial coordinates;

[0026] The specific method for formatting the panoramic photo POS data is as follows: formatting the panoramic photo POS data and saving it as a csv file, wherein the csv file includes at least the following columns: Id, Image, Time, X, Y, Z, Heading, Roll, and Pitch;

[0027] Among them, Id is the code of each row of data, starting from 1 and increasing; Image is the file name of the panoramic photo; Time is the shooting time of the panoramic photo, recorded in Unix timestamp format; X, Y, and Z are the spatial coordinates of the camera after registration when the panoramic photo was taken; Heading, Roll, and Pitch are the angles of rotation of the camera around each axis when the panoramic photo was taken, that is, the three angles of Euler angles; for devices that use quaternions to record camera POS data, the quaternion is converted to Euler angles using the following formula:

[0028]

[0029] θ=asin[2(q0q2-q3q1)]

[0030]

[0031] Among them, ψ, θ, Corresponding to the heading angle, pitch angle, and roll angle of the Euler angles respectively, q0, q1, q2, and q3 are the four components of the quaternion.

[0032] Preferably, in the third step, the geological recording includes point cloud and panoramic image matching, geological recording, and result export;

[0033] The specific method for the point cloud and panoramic image matching is as follows:

[0034] Project each pixel of the panoramic photo into three-dimensional space respectively through the following formula to form a panoramic sphere:

[0035]

[0036] In the formula, m and n are the vertical and horizontal pixel positions of the panoramic photo respectively, R is the radius of the panoramic sphere, and X', Y', and Z' are the coordinates P' of the pixel projected onto the panoramic sphere;

[0037] Use the registration file data obtained in the second step to construct matrices respectively:

[0038]

[0039] where (T x , T y , T z ) is the spatial coordinate after registration of the coordinate where the camera was located during the shooting of the panoramic photo, and a, b, and c are the angles of rotation around the three axes, that is, the three angles of the heading angle, pitch angle, and roll angle;

[0040] Multiply them in sequence to obtain the transformation matrix M = T·R x ·R y ·R z , and use this matrix to transform the pixel coordinates of the panoramic sphere P = M·P', where P' and P are the coordinates of each pixel of the panoramic sphere before and after transformation respectively;

[0041] Load and display the three-dimensional point cloud and the transformed panoramic sphere at the same time, and lock the viewing angle at the center of the panoramic sphere, then the point cloud and the panoramic image can be matched;

[0042] The geological recording includes the drawing of structural plane traces, the measurement of structural plane attitudes, and the description of structural planes; the specific method is as follows:

[0043] Under the panoramic view, draw the structural plane trace according to the morphological characteristics of the structural plane shown on the panoramic sphere: Use the graphic drawing tool to click on the starting point and the ending point of the structural plane in sequence. The software automatically captures the cloud point closest to the cursor, and uses the three-dimensional coordinates of the cloud point as the coordinates of the starting point and the ending point of the structural plane trace to draw the structural plane trace; for different types of structural planes, distinguish them by different colors and numbers; if the structural plane is long or there are turning deformations, add nodes at the turning points and in the middle to make the drawn trace more conform to the morphology reflected in the panoramic photo.

[0044] Under the panoramic view, select multiple points belonging to the structural plane to fit a plane for measuring the attitude of the structural plane; for a structural plane for which more than 3 nodes have been selected during the process of drawing the trace, a plane can be determined, and the nodes of the structural plane trace can be directly read to fit a plane for measuring the attitude of the structural plane.

[0045] Describe the structural plane according to the characteristics of the structural plane in the panoramic sphere. The description content includes the type of the structural plane, number, dip direction, dip angle, position in the underground cavern, roughness, filling condition, aperture, remarks.

[0046] The obtained results are exported by projecting the three-dimensional structural plane trace onto a regular plane, and then unfolding and splicing the plane into a two-dimensional geological record display map of the underground cavern, including projection plane definition, structural plane trace projection, and result export; the specific method is: define the projection plane by defining the shape, size, and axis of the cross-section; the structural plane trace projection adopts different projection methods according to different types of cross-section shapes; the projected structural plane trace, together with the projection plane, is unfolded along the top axis to obtain a two-dimensional structural plane display map, including the top of the cavern, left and right sidewall parts; export the structural plane display map to the dxf format supported by AutoCAD, and the exported structural plane display map retains the colors of different types of structural planes; export the structural plane description information in the structural plane description step to the csv format supported by Excel.

[0047] Preferably, if the point cloud density is high, blocking the display of the panoramic photo or affecting the computer performance, the point cloud can be thinned. Use the point cloud resampling function to perform resampling through the spatial relationship between points, and specify the minimum distance between two points; the minimum distance is 0.01 - 0.05 m.

[0048] Preferably, the specific measurement method for the attitude of the structural plane is as follows:

[0049] Use the least squares method for fitting to obtain the structural plane expression in the form of AX + BY + CZ + D = 0;

[0050] (a) Calculate the dip angle: The dip angle is defined as the angle between the plane along the inclined direction and the horizontal plane, and numerically it is equal to the angle between the normal vector and the Z-axis;

[0051] The normal vector N of the structural plane is obtained according to the structural plane expression, and then the dip angle of the structural plane is obtained through the following formula:

[0052]

[0053] In the formula, d is the dip angle of the structural plane, N is the normal vector of the structural plane, and N z is the component of the normal vector in the Z-axis direction;

[0054] (b) Calculate the trend: The trend is defined as the azimuth angle between the inclined direction of the plane and the due north direction on the horizontal plane; take the xy components of the vector N, project the normal vector onto the XY plane to obtain the vector Nxy, and the Y-axis vector is Y(0,1). The following formula can be used to calculate the angle between it and the Y-axis:

[0055]

[0056] Since the calculation result of the above formula is 0 to 180 degrees, it is also necessary to judge whether Nxy is in the clockwise direction of the Y-axis according to the positive or negative of the cross product result |Nxy|×|Y| of the two vectors. If it is in the counterclockwise direction, the trend is 360 - t.

[0057] Preferably, the method for projecting the trace of the structural plane is as follows: for an underground chamber with a circular cross-section, polar projection is used. Taking the center of the circle as the center, rays are emitted to the endpoints and nodes of the trace of the structural plane, and the intersection points of the rays and the circle are the projected points; for an underground chamber with a rectangular cross-section, orthographic projection is used for the side walls and the top surface respectively, and projection is carried out perpendicular to the plane to obtain the projected points; for an underground chamber with a horseshoe-shaped cross-section, orthographic projection is used for the side walls and polar projection is used for the arched roof to obtain the projected points.

[0058] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0059] A method for processing three-dimensional laser scanning data and geological logging of an underground chamber provided by the present invention standardizes the layout method, scanning method, and data processing method of control points for on-site three-dimensional laser scanning according to the characteristics of the underground chamber, can improve the accuracy and reliability of obtaining three-dimensional laser point clouds, ensure the acquisition of high-quality data, and improve the logging efficiency.

[0060] The method of the present invention conducts logging work indoors, reducing on-site working hours, lowering relevant safety risks in underground caverns, reducing the impact on the progress of subsequent construction, and improving overall work efficiency. The method of the present invention preprocesses data, enabling data obtained by scanners of different manufacturers and models to be used for logging work with 3D logging software after processing; the method of the present invention standardizes the processes and formats of data processing, facilitating standardized management and import into software, and the data processing process is simple with high accuracy of processing results; the present invention conducts structural plane logging through panoramic photos and 3D laser point clouds, and the traced line positions drawn are more accurate than the measurements on-site using station numbers and tape measures. At the same time, the attitudes of structural planes can be automatically calculated and projected into corresponding logging maps according to different underground cavern morphologies. The method of the present invention can implement geological logging and graphic processing in one program, export geological logging maps and logging data, reduce operations in different software, and improve logging efficiency. The logging results of the method of the present invention have a unified data format, including geological information such as the positions, attitudes, roughness, filling conditions, and aperture degrees of structural plane traces, providing a basis for subsequent design, construction, etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 is a technical flow chart of a method for processing 3D laser scanning data and geological logging of an underground cavern according to the present invention;

[0062] Figure 2 is a schematic diagram of the layout of control points for a method for processing 3D laser scanning data and geological logging of an underground cavern according to the present invention;

[0063] Figure 3 is a schematic diagram of POS data formatting for a method for processing 3D laser scanning data and geological logging of an underground cavern according to the present invention;

[0064] Figure 4 is a schematic diagram of trace projection for a method for processing 3D laser scanning data and geological logging of an underground cavern according to the present invention;

[0065] Figure 5 is a schematic diagram of the logging data format for a method for processing 3D laser scanning data and geological logging of an underground cavern according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0066] The following describes in detail the implementation of the present invention with reference to the accompanying drawings. However, they do not constitute a limitation to the present invention and are only for illustration purposes. At the same time, the advantages of the present invention will become clearer and easier to understand.

[0067] As Figure 1 shown, a method for processing 3D laser scanning data and geological logging of an underground cavern provided by the present invention includes the following steps:

[0068] I. On-site three-dimensional geological scanning.

[0069] In this step, first, control points are arranged. By reasonably arranging the control points, accurate coordinate information can be obtained. Then, a scanner is used for laser scanning. This scanner can be equipped with a panoramic camera to take panoramic photos while scanning. The scanning work is carried out along a suitable route and at a suitable speed. The panoramic camera is connected to the scanner in a supporting manner, and the data of both the panoramic camera and the scanner are recorded by the scanner. This scanner is a SLAM handheld three-dimensional laser scanner. The arrangement of control points and laser scanning together constitute the important content of three-dimensional geological scanning, providing basic data for subsequent data processing and geological logging.

[0070] II. Data preprocessing.

[0071] In this step, the original data obtained by the scanner in the above step I is solved and converted into point cloud data in the general las format and panoramic photos in the jpg format. At the same time, the POS data of the panoramic photos is processed and converted into a specific format to facilitate subsequent three-dimensional fusion and geological logging work.

[0072] III. Geological logging.

[0073] In this step, the processed laser point cloud data and panoramic photos are imported into three-dimensional logging software for the fusion of three-dimensional point clouds and panoramic photos. Then, geological logging work is carried out under the panoramic view to record key geological information such as the position, attitude, roughness, filling condition, and aperture of the structural plane trace. Finally, the projection plane is set and the logging results are exported, including logging data and logging maps, etc., providing comprehensive geological information support for the design, construction, and operation of underground projects.

[0074] The above method specifically includes the following steps:

[0075] I. On-site three-dimensional geological scanning

[0076] In this specific embodiment, the scanner uses a GeoSLAM ZEB Horizon three-dimensional laser scanner and is equipped with a ZEBVision panoramic camera. The underground chamber is an exploration adit.

[0077] (1.1) Arrangement of control points

[0078] In order to make the three-dimensional laser point cloud have correct spatial coordinates for calculating data such as the attitude of the structural plane, control points need to be set before the start of scanning, and the coordinates of the control points are measured.

[0079] According to the on-site conditions and equipment requirements, several control points are set inside and outside the underground chamber, and their coordinates are measured.

[0080] Specifically, the arrangement of control points is asFigure 2 As shown, the layout of the control points is determined according to the length of the exploration adit, and the layout method is as follows:

[0081] (a) If the length of the exploration adit does not exceed 100 m, the control points can be arranged only outside the adit.

[0082] (b) If the length of the exploration adit is greater than 100 m, the control points need to be arranged both inside and outside the adit.

[0083] The control points are fixed by targets and marked with paint, and their coordinates are obtained through surveying instruments.

[0084] Among them, for the control points outside the adit, RTK equipment or total station is used to measure the coordinates, and 5 control points are arranged near the entrance of the adit. The distance between the control points should be not less than 10 m, preferably 20 - 30 m, and most preferably 20 m. The distances between the control points can be the same and can be adjusted according to the actual on-site conditions. In particular, all the control points cannot be set on a straight line.

[0085] For the control points inside the adit, total station is used to measure the coordinates, and 1 control point is arranged on each of the left and right sidewalls of the adit every 50 - 100 m.

[0086] This control point layout method makes the control points sufficiently dispersed and can better register the point cloud. For the case without control points inside the adit, the corresponding control points can be easily identified in the figure during control point matching.

[0087] (1.2) Laser scanning

[0088] Hold the laser scanner with the right hand, keep the device 20 cm away from the body, rotate the scanner orientation so that the front lens faces the direction where photos need to be taken. Keep the body on the perpendicular bisector of the front and rear lenses to reduce the occlusion of the panoramic image. At the control point, align the sight of the scanner with the control point and keep it still for 10 s to enable the scanner to identify the control point.

[0089] Specifically, it is divided into two cases:

[0090] (a) When the length of the exploration adit does not exceed 100 m: Turn on the machine at the entrance of the adit, walk at a normal walking speed, pass through 5 control points outside the adit in sequence, then enter the underground adit, turn back at the end of the underground adit, and return to the starting position to end the scanning.

[0091] (b) When the length of the exploration tunnel exceeds 100m: Turn on the scanner at the entrance of the tunnel and walk at a normal walking speed. After passing the five control points outside the cave, enter the underground cave. Keep the scanner stable in the underground cave and start scanning along the right wall. When passing the control point, turn around slowly, align the scanner's crosshairs with the control point on the right, stand still for 10 seconds, then resume the scanning posture and continue walking. At the end of the cave, turn around slowly counterclockwise to ensure that the scanner can collect relevant data on the upper face. Continue walking to the right and scanning. When passing the control point, use the same method to obtain the control point until you return to the position where the scan started and end the scan.

[0092] The walking speed is preferably 0.5 m / s and should be kept as stable as possible to ensure the quality of the laser point cloud and panoramic photos.

[0093] For important geological boundaries, geological structures, typical faults and other areas, you can stop for a while and point the camera at that area to obtain richer photo data for subsequent cataloging.

[0094] 2. Data Processing

[0095] (2.1) Data solution

[0096] Use Connect View software to perform data solution, which includes three steps: preliminary solution, point cloud registration, and panoramic photo registration.

[0097] Import the original data into Connect View software, which automatically converts it into point cloud data in las format and panoramic photos in jpg format.

[0098] After the initial solution is completed, the control point coordinate information is used to align the point cloud so that the point cloud has the correct spatial coordinates. The scanner trajectory data is read and the control point coordinates are imported. The software automatically identifies the control points in the trajectory and matches them with the imported coordinates. If the automatic matching of control points cannot be completed, the control points can be easily identified by referring to the layout specifications of the control points in step (1.1) to manually match the point cloud with the control points.

[0099] Read the registered point cloud file, trajectory file, panoramic photo, shooting time and other data, determine the deformation matrix according to the coordinate change of the point cloud, modify the POS data of the panoramic photo, and re-register the panoramic photo.

[0100] At this point, we have obtained a three-dimensional laser point cloud with correct spatial coordinates, a panoramic photo in jpg format, and the registered panoramic photo POS data.

[0101] (2.2) Panoramic photo POS data formatting

[0102] Scanners of different brands and models record POS data during panoramic photo shooting using different data formats. In terms of data storage methods, there are formats such as tables, texts, and JSON. The data content is also divided into quaternions, Euler angles, etc. To facilitate the subsequent matching of 3D laser point clouds and panoramic photos, the POS data of panoramic photos is uniformly formatted and processed according to the following content and format, and saved as a csv file.

[0103] As Figure 3 shown in the schematic diagram of the content in the csv file after formatting the POS data of the panoramic photo, the data includes columns such as Id, Image, Time, X, Y, Z, Heading, Roll, Pitch, etc. Among them:

[0104] Id is the encoding of each row of data, increasing from 1.

[0105] Image is the file name of the panoramic photo.

[0106] Time is the shooting time of the panoramic photo, recorded in Unix timestamp format.

[0107] X, Y, and Z are the spatial coordinates of the camera during panoramic photo shooting after registration. Among them, X and Y are in the horizontal direction, and Z is in the height direction.

[0108] Heading, Roll, and Pitch are the angles of the camera's rotation around each axis during photo shooting, that is, the three angles of the Euler angle.

[0109] For devices that record camera POS data using quaternions, the quaternion can be converted to Euler angles through the following formula:

[0110]

[0111] θ = asin[2(q0q2 - q3q1)]

[0112]

[0113] Among them, ψ, θ, correspond to the heading angle (Heading), pitch angle (Pitch), and roll angle (Roll) of the Euler angle respectively, and q0, q1, q2, q3 are the four components of the quaternion.

[0114] It should be noted that when the pitch angle is ±90°, that is, when the nose is vertically downward or upward, there is a singular pose. At this time, the angle value needs to be directly set according to the machine pose. In addition, since the results of arctan and arcsin cannot cover all orientations, atan2 needs to be used to replace arctan in actual calculations.

[0115] 3. Geological Recording

[0116] A simple 3D laser point cloud lacks color information and is difficult to identify geological phenomena such as rock boundaries, seepage points, and cracks. Panoramic photos can be displayed in a panoramic sphere, which can be rotated 360° for observation and simulate geological surveys in underground caverns. The 3D laser point cloud can be fused and matched with the panoramic sphere, and the panoramic sphere can be used for observation, and the spatial data of the laser point cloud can be used to carry out geological cataloging.

[0117] Import the three-dimensional laser point cloud data registered in step (2.1), the panoramic photo in jpg format, and the csv file processed in step (2.2) into the three-dimensional geological cataloging software, and start geological cataloging in the panoramic view. The three-dimensional geological cataloging software used in this specific embodiment is Changyan three-dimensional real-scene geological cataloging software, which is a real-scene three-dimensional geological cataloging software developed by Changjiang Geotechnical Engineering Co., Ltd.

[0118] (3.1) Point cloud and panoramic image matching

[0119] Through the formula:

[0120]

[0121] Each pixel of the panoramic photo is projected into three-dimensional space to form a panoramic sphere. In the formula, m and n are the vertical and horizontal pixel positions of the panoramic photo, R is the radius of the panoramic sphere, and X', Y', and Z' are the coordinates P' of the pixel projected onto the panoramic sphere.

[0122] Use the registration file data processed in step (2.2) to construct the matrices respectively:

[0123]

[0124] Among them, (T x , T y , T z ) is the spatial coordinate value of the camera's coordinates after registration when the panoramic photo is taken, a, b, c are the angles of rotation of the camera around the three axes when the panoramic photo is taken, namely the Heading, Roll, and Pitch angles.

[0125] Multiply them in sequence to get the transformation matrix M = T·R x ·R y ·R z Using this matrix, the coordinates of each pixel of the panoramic sphere are transformed into P = M·P'. Wherein, P'(X', Y', Z') and P(X, Y, Z) are the coordinates of each pixel of the panoramic sphere before transformation and the spatial coordinates after transformation, respectively.

[0126] Load and display the 3D point cloud and the transformed panoramic sphere at the same time, and lock the viewing angle at the center of the panoramic sphere to match the point cloud with the panoramic image.

[0127] If the point cloud density is high and blocks the display of the panoramic photo or affects the computer performance, the point cloud can be thinned. Use the point cloud resampling function to resample the spatial relationship between points and specify the minimum distance between two points.

[0128] Setting the distance between two points in the point cloud to 0.01-0.05m can reduce the occlusion of the point cloud on the panoramic photo and the demand for computer performance while ensuring the cataloging accuracy.

[0129] (3.2) Geological Logging

[0130] ① Drawing of structural surface traces

[0131] In the panoramic view, the structural surface traces are drawn according to the structural surface morphological features displayed by the panoramic sphere.

[0132] Use the graphic drawing tool to click the starting point and end point of the structural surface in turn. The software will automatically capture the cloud point closest to the cursor, and use the three-dimensional coordinates of the cloud point as the coordinates of the starting point and end point of the structural surface trace to draw the structural surface trace.

[0133] After drawing is completed, right-click the mouse to end drawing.

[0134] If the structural surface is long or has certain turning deformations, nodes can be added at the turning points and appropriate places in the middle to make the drawn traces more consistent with the shape reflected in the panoramic photo.

[0135] For a long and large structural surface, a panoramic photo is difficult to show its full picture. When drawing a model, you can switch to an adjacent panoramic sphere perspective to continue drawing without interrupting the drawing.

[0136] Different types of structural surfaces, such as joints, fissures, faults, etc., are distinguished by different colors and numbers.

[0137] ② Structural surface occurrence measurement

[0138] In the panoramic view, select multiple points belonging to the structural surface to fit the plane and measure the structural surface orientation.

[0139] The specific measurement method is:

[0140] The least square method is used for fitting, and a structural surface expression in the form of AX+BY+CZ+D=0 is obtained.

[0141] (a) Calculate the inclination angle: The inclination angle is defined as the angle between the plane and the horizontal plane along the inclination direction, and its value is equal to the angle between the normal vector and the Z axis.

[0142] Calculate the normal vector N of the structural plane according to the structural plane expression, and then calculate the dip angle of the structural plane through the following formula:

[0143]

[0144] In the formula, d is the dip angle of the structural plane, N is the normal vector of the structural plane, and N z is the component of the normal vector in the Z-axis direction.

[0145] (b) Calculate the trend: The definition of the trend is the azimuth angle between the inclined direction of the plane on the horizontal plane and the due north direction. Take the xy components of the vector N, project the normal vector onto the XY plane to obtain the vector Nxy, and the Y-axis vector is Y(0,1). The following formula can be used to calculate the angle between it and the Y-axis:

[0146]

[0147] Since the calculation result of the above formula is 0 to 180 degrees, it is also necessary to judge whether Nxy is in the clockwise direction of the Y-axis according to the positive and negative of the cross product result |Nxy|×|Y| of the two vectors. If it is in its counterclockwise direction, the trend is 360 - t.

[0148] For relatively long structural planes, more than 3 nodes have been selected during the process of drawing the trace line, and a plane can be determined. The nodes of the structural plane trace can be directly read for calculation, omitting the step of selecting the points belonging to the structural plane, and the attitude of the structural plane is automatically calculated after the trace line is drawn.

[0149] For some structural planes such as fissures, etc., the degree of opening is small, and excavation has not formed a large enough surface to fit the plane, nor is there a spatial turning to form a three-dimensional trace. For such structural planes, a plane passing through the trace can be constructed and the plane can be rotated around the trace. Rotate the plane to a suitable position, switch different perspectives for observation, and use the attitude of the plane to represent the attitude of the structural plane.

[0150] ③ Description of the structural plane

[0151] Describe the structural plane according to the characteristics of the structural plane in the panoramic sphere.

[0152] The description content includes number, trend, dip angle, position in the underground cavern, roughness, filling condition, degree of opening, remarks, etc.

[0153] The type of the structural plane can be selected, including fissure (L), joint (J), long fault (F), jointed fault (f), geological boundary (D), other (Q), etc. Different types correspond to different number prefixes and display colors.

[0154] The number can be an incrementing number and is counted separately according to different prefixes.

[0155] The trend and dip angle can be automatically filled according to the measurement results.

[0156] For the location, roughness, filling condition, aperture, etc. of the underground cavern, built-in entries can be selected through the drop-down menu, or input can be made according to the on-site situation.

[0157] (3.3) Result export

[0158] During the logging process, the structural plane trace lines drawn are line segments in the three-dimensional space based on the three-dimensional laser point cloud and cannot be directly exported as a geological logging display map of the underground cavern in the general sense.

[0159] It is necessary to first project the three-dimensional structural plane trace lines onto a regular plane, and then unfold and splice the plane into a two-dimensional geological logging display map of the underground cavern.

[0160] ① Projection plane definition

[0161] The projection plane is defined by defining the shape, size, and axis of the cross-section.

[0162] The cross-section shapes include circular, rectangular, horseshoe-shaped, etc.

[0163] For a circular cross-section, the radius of the circle needs to be defined; for a rectangular cross-section, the width and height need to be defined; for a horseshoe-shaped cross-section, the width, height, and the radius of the arc segment need to be defined.

[0164] In the present invention, the axis of the projection plane is defined as the bottom central axis, rather than the central axis. The axis is specified by clicking the starting point and the ending point on the ground in the point cloud.

[0165] According to the above input content, a projection plane is generated, and the projection plane is the theoretical design form of the underground cavern.

[0166] Due to excavation errors and errors in specifying the axis, etc., the generated projection plane may have a certain angle with the three-dimensional point cloud of the underground cavern. The cross-section size can be adjusted by re-entering the parameters to generate the projection plane, and the projection plane angle, projection plane position, etc. can be adjusted manually to make the projection plane reflect the true form of the underground cavern as much as possible.

[0167] ② Projection of structural plane trace lines

[0168] Since the excavated underground cavern is not a theoretical curved surface and has undulations, the drawn structural plane trace lines also need to be projected onto the projection plane for further processing. Different types of cross-section shapes adopt different projection methods.

[0169] Such as Figure 4As shown in the figure, for an underground chamber with a circular cross-section, using stereographic projection, with the center of the circle as the center, rays are emitted towards the endpoints and nodes of the structural plane traces, and the intersection points of the rays and the circle are the projected points. For an underground chamber with a rectangular cross-section, orthographic projection is used for the side walls and the top surface respectively, projecting perpendicular to the plane to obtain the projected points. For an underground chamber with an arch-shaped cross-section, orthographic projection is used for the side walls and stereographic projection is used for the arched roof to obtain the projected points.

[0170] ③ Result export

[0171] The projected structural plane traces, together with the projection plane, are unfolded along the top axis to obtain a two-dimensional display diagram of the structural plane, including parts such as the roof, left and right side walls, etc.

[0172] The above two-dimensional display diagram of the structural plane can be exported in the dxf format supported by AutoCAD for further editing and beautification in AutoCAD or other graphic editing software.

[0173] The exported display diagram of the structural plane retains the colors of different types of structural planes, facilitating viewing and differentiation.

[0174] Each graphic includes side wall boundaries, etc., and has its own layer, which can be differentiated according to the layer name during subsequent editing, and is also convenient for subsequent editing of different layers.

[0175] The recorded description information of the structural plane can also be exported in the csv format supported by Excel, including content such as number, name, dip direction, dip angle, category, roughness, aperture, filling condition, location, remarks, line width, color, coordinates, etc., as Figure 5 shown.

[0176] The coordinates record the X, Y, and Z coordinates of each endpoint and node in sequence, separated by spaces between X, Y, and Z, and separated by semicolons between different points.

[0177] Other parts not described herein belong to the prior art.

Claims

1. A method for processing three-dimensional laser scanning data and geological logging of underground caverns, characterized in that: It includes the following steps: Step 1, on-site three-dimensional geological scanning; Obtain relevant information about the underground chamber through on-site investigation, initially master the structural characteristics of the geological body, arrange control points, determine the scanning method and scanning route, and then use a scanner to perform laser scanning to obtain raw data; the scanner is a SLAM handheld three-dimensional laser scanner equipped with a panoramic camera; Step 2, data preprocessing; Perform preliminary calculation, point cloud registration, and panoramic photo registration on the raw data obtained by the scanner to obtain point cloud data in the common las format, panoramic photos in the jpg format, and POS data after panoramic photo registration; perform formatting processing on the POS data of the panoramic photos and save it as a csv file; The data preprocessing includes data calculation and formatting of POS data of panoramic photos; The specific method of the data calculation is: use the software supporting the scanner to perform preliminary calculation, point cloud registration, and panoramic photo registration on the raw data obtained by the scanner to obtain three-dimensional laser point clouds with correct spatial coordinates and POS data of panoramic photos; The specific method of formatting the POS data of the panoramic photos is: perform formatting processing on the POS data of the panoramic photos and save it as a csv file. The csv file includes at least the following columns: Id, Image, Time, X, Y, Z, Heading, Roll, Pitch; Among them, Id is the encoding of each row of data, incrementing from 1; Image is the file name of the panoramic photo; Time is the shooting time of the panoramic photo, recorded in the Unix timestamp format; X, Y, and Z are the spatial coordinates of the camera after registration when the panoramic photo is taken; Heading, Roll, and Pitch are the angles of the camera rotating around each axis when the panoramic photo is taken, that is, the three angles of the Euler angle; for devices that record camera POS data using quaternions, convert the quaternion to the Euler angle through the following formula: ， ， ， Among them, ψ, θ, and φ correspond to the heading angle, pitch angle, and roll angle of the Euler angle respectively, and q0, q1, q2, and q3 are the four components of the quaternion; Step 3, geological logging; Import the point cloud data in the las format, panoramic photos in the jpg format, and the csv file into three-dimensional logging software, project the panoramic photos into a panoramic sphere, perform the fusion of the three-dimensional point cloud and the panoramic sphere, then carry out geological logging work under the panoramic view, record key geological information, and finally set the projection plane and export the logging results; the logging results include logging data and logging maps; The underground chamber includes an exploration adit, a water conveyance tunnel, a traffic tunnel, and an underground powerhouse.

2. The three-dimensional laser scanning data processing and geological logging method for underground chambers according to claim 1, characterized in that: In the said Step 1, the control points are fixed by targets and marked with paint; the arrangement of the control points is determined according to the length of the underground chamber, and the determination method is as follows: (a) If the length of the underground chamber does not exceed 100m, only arrange control points outside the chamber; (b) If the length of the underground chamber is greater than 100m, arrange control points both inside and outside the chamber; Among them, the control points outside the cave use RTK equipment or total station to measure the coordinates, and the distance between the control points is not less than 10m, and the number of control points is 4-6; the control points inside the cave use total station to measure the coordinates, and the distance between the control points is 50-100m.

3. The three-dimensional laser scanning data processing and geological logging method for underground chambers according to claim 2, wherein: The distance between the control points outside the cave is 20-30m, preferably 20m, and the number of control points is 5.

4. The three-dimensional laser scanning data processing and geological logging method for underground chambers according to claim 1, characterized in that: In the step 1, The specific scanning method is as follows: hold the 3D laser scanner in your right hand, keep the device 20cm away from your body, rotate the scanner so that the front lens faces the direction where you need to take photos; the body is located on the perpendicular midline of the front and rear lenses; when walking to the control point, align the scanner's crosshairs with the control point and wait for 10 seconds to allow the scanner to recognize the control point; The specific scanning route is: walk, pass the control points outside the cavern in turn, continue walking, enter the underground cavern, keep the scanner stable in the underground cavern, start scanning along the right wall, reach the end of the cavern, turn slowly to ensure that the scanner can collect relevant data on the upper face, then continue walking to the right and scanning until you return to the starting position and end the scan.

5. The 3D laser scanning data processing and geological logging method for underground caverns according to claim 4, characterized in that: The walking speed is 0.5 m / s.

6. The three-dimensional laser scanning data processing and geological logging method for underground chambers according to claim 1, characterized in that: In the step 3, geological cataloging includes matching point cloud with panoramic image, geological cataloging, and exporting results; The specific method of matching the point cloud with the panoramic image is: Each pixel of the panoramic photo is projected into three-dimensional space by the following formula to form a panoramic sphere: ， ， ， Where m and n are the vertical and horizontal pixel positions of the panoramic photo, R is the radius of the panoramic sphere, and X', Y', and Z' are the coordinates P' of the pixel projected onto the panoramic sphere; Use the registration file data processed in step 2 to construct the matrices respectively: ， ， ， ； Among them, (T x , T y , T z ) are the spatial coordinates after registration of the coordinates where the camera is located during panoramic photo shooting. a, b, and c are the angles of rotation around the three axes, that is, the three angles of heading angle, pitch angle, and roll angle; Multiply them in sequence to obtain the transformation matrix , and use this matrix to transform the pixel coordinates of the panoramic sphere ', where P' and P are the coordinates of each pixel of the panoramic sphere before and after transformation, respectively; Load and display the 3D point cloud and the transformed panoramic sphere at the same time, and lock the viewing angle at the center of the panoramic sphere to match the point cloud with the panoramic image; The geological logging includes drawing of structural surface traces, measuring of structural surface attitude and description of structural surface; the specific method is as follows: In the panoramic view, draw the structural surface trace according to the morphological features of the structural surface displayed by the panoramic ball: use the graphic drawing tool to click the starting point and end point of the structural surface in turn, and the software will automatically capture the cloud point closest to the cursor, and use the three-dimensional coordinates of the cloud point as the coordinates of the starting point and end point of the structural surface trace to draw the structural surface trace; for different types of structural surfaces, use different colors and numbers to distinguish them; if the structural surface is long or has turning deformation, add nodes at the turning point and in the middle to make the drawn trace more consistent with the shape reflected in the panoramic photo; In the panoramic view, select multiple points belonging to the structural surface to fit the plane for measuring the structural surface occurrence; for the structural surface that has more than 3 nodes selected in the process of drawing the trace, a plane can be determined, and the nodes of the structural surface trace can be directly read to fit the plane for measuring the structural surface occurrence; Describe the structural surface according to its characteristics in the panoramic sphere, including the type, number, inclination, dip, location in the underground cavern, roughness, filling condition, opening, and remarks of the structural surface; The above results are exported by projecting the three-dimensional structural plane traces onto a regular plane, and then unfolding and splicing the plane into a two-dimensional geological record display map of the underground cavern, including projection plane definition, structural plane trace projection, and result export. The specific method is as follows: The projection plane is defined by defining the shape, size, and axis of the cross-section. Different projection methods are used for the structural plane trace projection according to different types of cross-section shapes. The projected structural plane traces are unfolded along the top axis together with the projection plane to obtain a two-dimensional structural plane display map, including the top of the cavern, the left and right side walls. The structural plane display map is exported in the dxf format supported by AutoCAD, and the exported structural plane display map retains the colors of different types of structural planes. The structural plane description information in the structural plane description step is exported in the csv format supported by Excel.

7. The method for processing three-dimensional laser scanning data and geological logging of underground chambers according to claim 6, wherein: If the point cloud density is high, blocking the display of the panoramic photo or affecting the computer performance, the point cloud can be thinned. Use the point cloud resampling function to resample through the spatial relationship between points and specify the minimum distance between two points; the minimum distance is 0.01 - 0.05 m.

8. The method for processing three-dimensional laser scanning data and geological logging of an underground cavern according to claim 6, characterized in that: The specific measurement method of the attitude of the structural plane is as follows: Use the least squares method for fitting to obtain the structural plane expression in the form of AX + BY + CZ + D = 0; (a) Calculate the dip angle: The dip angle is defined as the angle between the plane along the inclined direction and the horizontal plane, and numerically it is equal to the angle between the normal vector and the Z-axis; Obtain the normal vector N of the structural plane according to the structural plane expression, and then obtain the dip angle of the structural plane through the following formula: ， where d is the dip angle of the structural plane, N is the normal vector of the structural plane, and N z is the component of the normal vector in the Z-axis direction; (b) Calculate the strike: The strike is defined as the azimuth angle between the inclined direction of the plane on the horizontal plane and the due north direction; Take the xy components of vector N, project the normal vector onto the XY plane to obtain vector N xy , the Y-axis vector is Y(0, 1), and the angle between it and the Y-axis can be calculated by the following formula: ， Since the calculation result of the above formula is 0 to 180 degrees, it is also necessary to judge whether N is in the clockwise direction of the Y-axis according to the sign of the cross product result of the two vectors. If it is in the counterclockwise direction, the inclination is 360 - t. xy is in the clockwise direction of the Y-axis. If it is in the counterclockwise direction, the inclination is 360 - t.

9. The 3D laser scanning data processing and geological logging method for underground chambers according to claim 6, characterized in that: The structural plane trace projection method is as follows: For an underground cavern with a circular cross-section, polar projection is used. Rays are emitted from the center of the circle to the endpoints and nodes of the structural plane traces, and the intersections of the rays and the circle are the projected points. For an underground cavern with a rectangular cross-section, orthographic projection is used for the side walls and the top surface respectively, projecting perpendicular to the plane to obtain the projected points. For an underground cavern with a horseshoe-shaped cross-section, orthographic projection is used for the side walls and polar projection is used for the arched top to obtain the projected points.

Images