Method and device for acquiring hidden body point cloud data of hidden area through base plane mirror

Abstract

The invention discloses a method for acquiring point cloud data of a hidden body in a hidden area based on a plane mirror, and solves the problems that the point cloud data of the hidden body in a hidden area and a narrow zone in a station is difficult to collect and incomplete in collection in three-dimensional scanning of oil and gas field stations, buildings and factories. The method comprises the following steps of: forming a range with an angle of 60-150 degrees by using a plane mirror, a scanner and a hidden body, aligning each surface of the hidden body, and collecting feature points of more than 80% of side line points and vertexes of the hidden body covered by the hidden body; according to the method, the edge lines of the n surfaces are fitted and spliced by using a least square fitting method of CYCLONE software to obtain the point cloud data of the target hidden body, an integral target hidden body is formed, the point cloud data of the hidden body can be quickly and completely obtained, and complete and reliable data are provided for construction of a digital station yard in the later period. The method is applied to the field of hidden body point cloud data acquisition in the hidden area.

Description

Technical Field

[0001] The invention belongs to the field of point cloud data acquisition, and in particular relates to a method and a device for acquiring point cloud data of a hidden body in a hidden area by using a base plane reflector. Background Art

[0002] With the requirements for digital and intelligent construction of oil and gas field stations, the demand for digital station construction in the petroleum and petrochemical industries has been increasing in recent years. Digital stations can not only provide daily measurement and information query, but also play an important role in future three-dimensional design, operation management, fire drills, etc.

[0003] The construction of digitalized stations first acquires data through scanning, and then generates the shape of objects in reverse based on the collected point cloud data. When making digital models of oil and gas fields, it is necessary to collect three-dimensional data of buildings, equipment, pipelines, accessories, etc. in the station. When using conventional means to collect data, the station will encounter equipment obstruction, dense pipelines, and narrow space. Point cloud data in the above environment is often difficult to collect or incomplete, and is later improved through manual measurement, photography, coordinate collection, etc. Therefore, a lot of manpower and material resources are consumed, and the accuracy is poor and the efficiency is low. In order to improve the efficiency of station data acquisition, data integrity, and accuracy, there is an urgent need for a new technology and method to solve the problem of point cloud data collection integrity and efficiency of hidden objects in hidden areas of oil and gas field stations. Summary of the invention

[0004] In order to solve the problems that point cloud data collection is difficult or incomplete due to equipment obstruction, dense pipe networks and narrow space in the station, and subsequent improvement through manual measurement, photography and coordinate collection consumes a lot of manpower and material resources, and has poor accuracy and low efficiency, the present invention provides a method and device for obtaining point cloud data of hidden bodies in hidden areas with a base plane reflector.

[0005] The technical solution of the present invention:

[0006] In one aspect, a method for obtaining point cloud data of a hidden body in a hidden area using a base plane reflector is provided, the method comprising the following steps:

[0007] Step 1: Hidden body point cloud data collection

[0008] For the target hidden body in the area that cannot be scanned, first conduct a site survey and select the location for placing the plane reflector; the scanner, plane reflector and target hidden body form a three-dimensional angle of 60°-150°, and use the scanner to scan the n faces of the target hidden body in the plane reflector to collect point cloud data and obtain the initial point cloud data of the target hidden body;

[0009] Step 2: Point cloud data processing

[0010] According to the initial point cloud data of the target hidden body, the least square fitting method of the CYCLONE software is used to fit the edge lines of n faces and splice them; then the least square fitting method frame selection mode of the CYCLONE software is used to select a target face of the collected hidden body, select the point cloud data on the target face, perform target face domain fitting, and delete redundant point cloud data; then select n-1 faces in turn, complete n face domain fitting to obtain the edge line, use the line projected by the n faces as the boundary, delete the redundant point cloud data outside the boundary, obtain the point cloud data of the target hidden body, and form an overall target hidden body;

[0011] Step 3: Point cloud data conversion

[0012] According to the point cloud data of the plane reflector, at least 3 points are collected on the left side, right side, top side and bottom side of the plane reflector frame, a three-point normal equation is established, and the point cloud data of the target hidden body is input into the three-point normal equation to calculate the point cloud data of the target hidden body.

[0013] Preferably, in step 1, forming a three-dimensional angle of 60°-150° among the scanner, the plane reflector and the target hidden body means that the hidden body in the hidden area is observed through the scanner telescope and presented in the plane reflector; the plane reflector is aimed at one surface of the hidden body, and the feature points of the edge points and vertices covering more than 80% of the hidden body are collected to facilitate the subsequent data stitching.

[0014] The plane reflector is arranged on a cylindrical rod, which is divided into two retractable sections, the lower section of which has three brackets extending outward from the bottom. A connecting device is fixed to the upper section of the cylindrical rod, and the connecting device is connected to the cylindrical rod through a concave member. The concave member allows the mirror frame to rotate freely 360°.

[0015] Preferably, the plane reflector has lock buckles on both sides and the bottom of the frame, and the lock buckles can fix the plane mirror on the frame after being stretched, and the lock buckles automatically return to their original positions after the plane mirror is taken out;

[0016] The plane reflector is rectangular, and white paint is applied to an area at least larger than 5 mm of the plane reflector frame to form a white reflective area; the white reflective area is used for laser scanning and identification of point cloud data mirror conversion positioning point extraction, wherein the thickness of the white paint does not exceed 0.2 mm.

[0017] Preferably, the step 2 of fitting the edges of n faces and splicing them using the least squares fitting method of CYCLONE software also includes the following steps:

[0018] Step 201 uses CYCLONE software to fit the edge lines of n faces to the initial point cloud data of the target hidden body using the least squares fitting method;

[0019] Step 202: According to the fitted edge lines, each face obtains at least three vertices; the other n-1 faces are fitted in turn, and the vertices are marked;

[0020] In step 203, according to the vertices on the n faces, three vertices of the adjacent faces are randomly selected as common points for splicing, and the splicing accuracy is less than 1 mm.

[0021] Among them, deleting redundant point cloud data also includes the following steps:

[0022] a. First, select one of the faces of the target hidden body and use the least square fitting method of CYCLONE software to perform face fitting;

[0023] b. Then select n-1 faces in turn to perform face domain fitting;

[0024] c. Adjust the view of the entire point cloud of the target hidden body to a top view, set the point cloud to a non-selectable mode, use the line of the projection of the n faces of the target hidden body as the boundary, select a box to delete the redundant point cloud data;

[0025] d After deleting the redundant point cloud data, the complete target hidden body point cloud data is obtained.

[0026] Preferably, in step 3, a three-point normal equation is established, and the point cloud data of the target hidden body is input into the three-point normal equation to calculate the point cloud data of the target hidden body, which means that the formula of the three-point normal equation is established as follows:

[0027] Take the coordinates of the white areas R1, R2, and R3 of the plane reflector frame as (a1, b1, c1), (a2, b2, c2), and (a3, b3, c3) respectively.

[0028] The plane equation of the plane where the plane reflector is located is expressed as Ax+By+Cz+D=0, where A, B, and C are the three components of the normal vector, and D is the intercept of the plane, all of which are parameters to be determined. The vector on the plane is

[0029] Let the normal vector of the plane mirror be Since the normal vector of a plane should be perpendicular to the two non-collinear vectors in the plane, that is and From the definition of vector cross product, we know that At the same time perpendicular to the vector and Therefore, the normal vector of the plane mirror can be obtained by the following formula: Where i, j, and k represent the unit vectors in the positive direction of the x-axis, the positive direction of the y-axis, and the positive direction of the z-axis, respectively.

[0030] Where A = (b2-b1)(c3-c1)-(b3-b1)(c2-c1) (1)

[0031] B=(c2-c1)(a3-a1)-(c3-c1)(a2-a1) (2)

[0032] C=(a2-a1)(b3-b1)-(a3-a1)(b2-b1) (3)

[0033] Substituting the coordinates of point R1 (a1, b1, c1) into the plane equation Ax+By+Cz+D=0, we can get D=-A·a1+B·b1+C·c1

[0034] Assume that the coordinates of any known point in the collected target hidden body point cloud data are (X1, Y1, Z1), and the corresponding mirror point coordinates are (X1′, Y1′, Z1′). Then, from the symmetric formula, we can know that

[0035] X1′=X1-(2A(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (4)

[0036] Y1′=Y1-(2B(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (5)

[0037] Z1′=Z1-(2C(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (6)

[0038] Substitute the target hidden body point cloud data into formulas (4), (5), and (6) to calculate the target hidden body point cloud data. The target hidden body point cloud data is the point cloud data of the actual position coordinates of the target hidden body.

[0039] In another aspect, a device for collecting point cloud data of a hidden body in a hidden area is provided, the device comprising a scanner, a plane reflector unit, and a point cloud data processing and conversion unit;

[0040] A scanner, used for observing the hidden body in the hidden area through a scanner telescope, and collecting initial point cloud data of n faces of the hidden body;

[0041] Plane reflector unit, the plane reflector is set on a cylindrical rod, the cylindrical rod is divided into two sections, the lower section is retractable, the bottom of the lower section of the cylindrical rod extends three brackets, the upper section of the cylindrical rod is fixed with a connecting device, the connecting device is connected to the cylindrical rod through a concave member, and the concave member allows the mirror frame to rotate 360 ​​degrees freely;

[0042] The plane reflector forms an angle of 60°-150° with the scanner and the concealed object, and is aimed at each surface of the concealed object to collect feature points of the edge points and vertices of the concealed object that cover more than 80% of the concealed object;

[0043] The point cloud data processing and conversion unit is used for the initial point cloud data of the target hidden body, and uses the least squares fitting method of the CYCLONE software to fit the edge lines of n faces and splice them; uses the least squares fitting method of the CYCLONE software to select a target face of the hidden body, selects the point cloud data on the target face, performs face domain fitting, and deletes redundant point cloud data; then selects n-1 faces in turn, completes n face domain fitting to obtain the edge line, uses the line projected by the n faces as the boundary, deletes the redundant point cloud data outside the boundary, obtains the point cloud data of the target hidden body, and forms an overall target hidden body;

[0044] According to the point cloud data of the plane reflector, at least 3 points are collected on the left side, right side, top side and bottom side of the plane reflector frame, a three-point normal equation is established, and the point cloud data of the target hidden body is input into the three-point normal equation to calculate the point cloud data of the target hidden body.

[0045] Preferably, the plane reflector has lock buckles on both sides and the bottom of the frame, and the lock buckles can fix the plane mirror on the frame after being stretched, and the lock buckles automatically return to their original positions after the plane mirror is taken out;

[0046] The plane reflector is rectangular, and the plane reflector frame area is at least larger than 5mm and painted with white paint to form a white reflective area; the white reflective area is used for laser scanning and recognition of point cloud data mirror conversion positioning point extraction, wherein the thickness of the white paint does not exceed 0.2mm;

[0047] Preferably, fitting the edges of n faces using the least squares fitting method of CYCLONE software and then splicing them also includes the following steps:

[0048] Step 201 uses CYCLONE software to fit the edge lines of n faces to the initial point cloud data of the target hidden body using the least squares fitting method;

[0049] Step 202: According to the fitted edge lines, each face obtains at least three vertices; the other n-1 faces are fitted in turn, and the vertices are marked;

[0050] Step 203: based on the vertices on the n faces, three vertices of the adjacent faces are randomly selected as common points for splicing, and the splicing accuracy is less than 1 mm;

[0051] Step 204 first selects one of the faces of the target hidden body and performs face domain fitting using the least square fitting method of the CYCLONE software; then selects n-1 faces in turn and performs face domain fitting;

[0052] Step 205 adjusts the view of the entire point cloud of the target hidden body to a top view, sets the point cloud to a non-selectable mode, uses the lines projected by each surface as boundaries, and deletes redundant point cloud data outside the boundaries; after deleting the redundant point cloud, the complete point cloud data of the target hidden body is obtained.

[0053] Preferably, establishing a three-point normal equation, inputting the point cloud data of the target hidden body into the three-point normal equation to calculate the point cloud data of the target hidden body means establishing the three-point normal equation formula as follows:

[0054] On the point cloud of the plane reflector, collect three points R1, R2, and R3 on the left and right sides or the upper and lower sides of the mirror frame, establish the normal equation based on these three points, and write a program.

[0055] Assume that the spatial coordinates of the three points R1, R2, and R3 on the plane mirror are (a1, b1, c1), (a2, b2, c2), and (a3, b3, c3), respectively. The plane equation of the plane where the mirror is located is expressed as Ax+By+Cz+D=0, where A, B, and C are the three components of the plane normal vector, and D is the intercept of the plane, all of which are parameters to be determined. The vector on the plane

[0056] Let the normal vector of the plane mirror be Since the normal vector of a plane should be perpendicular to the two non-collinear vectors in the plane, that is Head From the definition of vector cross product, we know that At the same time perpendicular to the vector and Therefore, the normal vector of the plane mirror can be obtained by the following formula: Among them, i, j, and k represent the unit vectors in the positive direction of the x-axis, the positive direction of the y-axis, and the positive direction of the z-axis respectively;

[0057] Where A = (b2-b1)(c3-c1)-(b3-b1)(c2-c1)

[0058] B=(c2-c1)(a3-a1)-(c3-c1)(a2-a1)

[0059] C=(a2-a1)(b3-b1)-(a3-a1)(b2-b1)

[0060] Substituting the coordinates of point R1 (a1, b1, c1) into the plane equation Ax+By+Cz+D=0, we can get D=-A·a1+B·b1+C·c1

[0061] Assume that the coordinates of any known point in the collected target hidden body point cloud data are (X1, Y1, Z1), and the corresponding mirror point coordinates are (X1′, Y1′, Z1′). Then, from the symmetric formula, we can know that

[0062] X1′=X1-(2A(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (4)

[0063] Y1′=Y1-(2B(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (5)

[0064] Z1′=Z1-(2C(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (6)

[0065] Substitute the target hidden body point cloud data into formulas (4), (5), and (6) to calculate the target hidden body point cloud data, which is the point cloud data of the actual position coordinates of the target hidden body.

[0066] The present invention provides a method for acquiring point cloud data of hidden bodies in hidden areas by using a plane reflector, which solves the problem of difficult and incomplete acquisition of point cloud data of hidden bodies in hidden areas and narrow areas within oil and gas field stations, buildings and factories in three-dimensional scanning. The present invention adopts point cloud data acquisition, forms an angle of 60°-150° with the plane reflector, the scanner and the hidden body, and aligns with each face of the hidden body to acquire feature points of edge points and vertices covering more than 80% of the hidden body; uses the least squares fitting method of CYCLONE software to fit the edge lines of n faces and splice them to obtain point cloud data of the target hidden body, and forms an overall target hidden body. The point cloud data of the above-mentioned hidden body can be acquired quickly and completely, and complete and reliable data can be provided for the construction of digital stations in the later stage. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a flow chart of a method for obtaining point cloud data of hidden bodies in hidden areas with a base plane reflector;

[0068] Figure 2A method for obtaining point cloud data of a hidden body in a hidden area using a base plane reflector; a schematic diagram of a hidden body;

[0069] Figure 3 A method for obtaining point cloud data of hidden bodies in hidden areas using a base plane reflector is a schematic diagram of the position of the plane mirror;

[0070] Figure 4 It is a method for obtaining point cloud data of hidden bodies in hidden areas with a base plane reflector. The characteristic schematic diagram of the mirror when placed;

[0071] Figure 5 A method for obtaining point cloud data of hidden bodies in hidden areas based on a plane reflector, and a plane reflector placement bracket;

[0072] Figure 6 A method for obtaining point cloud data of hidden bodies in hidden areas with a base plane reflector is a schematic diagram of a cylindrical rod fixing connection device of a mirror placement bracket;

[0073] Figure 7 A method for obtaining point cloud data of hidden bodies in hidden areas based on a plane reflector is a schematic diagram of the automatic return of the plane reflector lock;

[0074] Figure 8 The present invention is a method for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector. The present invention is a schematic diagram of a plane mirror design style;

[0075] Fig. 9 It is a method for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector. Scanner data acquisition diagram;

[0076] Fig.10 It is a method for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector. The hidden body scans 4 surfaces and collects the point cloud data for fitting edge line diagram;

[0077] Fig.11 It is a method for obtaining point cloud data of hidden bodies in hidden areas with a base plane reflector, and a schematic diagram of the four vertices of the fitted edge line;

[0078] Fig.12 It is a method for obtaining point cloud data of hidden bodies in hidden areas with a base plane reflector, and a schematic diagram of redundant point data;

[0079] Fig.13 It is a method for obtaining point cloud data of hidden body in hidden area by using base plane reflector. The hidden body scans four surface fitting diagrams;

[0080] Fig.14 A schematic diagram of deleting redundant point clouds of four surfaces by selecting a frame in an embodiment of a method for obtaining point cloud data of a hidden body in a hidden area with a base plane reflector;

[0081] Fig.15 A method for obtaining point cloud data of a hidden body in a hidden area with a base plane reflector is provided, in which a point cloud top view after deleting redundant point cloud data is provided;

[0082] Fig.16 The invention is a method for obtaining point cloud data of a hidden body in a hidden area by a base plane reflector, wherein the redundant point cloud data is deleted and the point cloud oblique view is shown;

[0083] Fig.17 A method for obtaining point cloud data of a hidden body in a hidden area using a plane reflector is provided. In the embodiment, a schematic diagram of collecting three points on the point cloud of the plane reflector is provided;

[0084] Fig.18 It is a method for obtaining point cloud data of hidden bodies in concealed areas through base plane reflectors. Mirror transformation Input normal line to calculate mirror coordinates. Run interface Input normal line to calculate reference point coordinates (mirror coordinates). Screenshot;

[0085] Fig.19 It is a method for obtaining point cloud data of hidden bodies in hidden areas with a base plane reflector. Click on the normal and constant to calculate the four parameters of the normal equation.

[0086] Fig. 20 It is a method for obtaining point cloud data of hidden bodies in hidden areas with a base plane reflector. Single point calculation input original coordinate screenshot;

[0087] Fig.21 It is a method for obtaining point cloud data of hidden bodies in hidden areas with a base plane reflector. Click Convert to get a screenshot of the converted coordinates;

[0088] Fig. 22 It is a method for obtaining point cloud data of hidden bodies in hidden areas with a base plane reflector. Batch calculation screenshot;

[0089] Fig.23 A method for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector is implemented by importing coordinate data files and selecting a path map saved after conversion;

[0090] Fig.24 A method for obtaining point cloud data of hidden bodies in hidden areas with a base plane reflector. Embodiment: Click on the pop-up window prompt after the conversion is completed;

[0091] Fig.25 The invention is a method for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector, and test data diagram of an embodiment;

[0092] Fig.26 The invention discloses a converted data diagram of an embodiment of a method for obtaining point cloud data of a hidden body in a hidden area by using a base plane reflector. DETAILED DESCRIPTION

[0093] Here, exemplary embodiments will be described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention.

[0094] In one aspect, an embodiment of the present invention provides a method for obtaining point cloud data of a hidden body in a hidden area using a base plane reflector. Figure 1 As shown, the method comprises the following steps:

[0095] Step 1: Hidden body point cloud data collection

[0096] For target hidden bodies in areas that cannot be scanned, first conduct an on-site survey and select a location to place the plane reflector; the scanner, plane reflector, and target hidden body form a three-dimensional angle of 60°-150°, and use a scanner to scan the n faces of the target hidden body in the plane reflector to collect point cloud data and obtain the initial point cloud data of the target hidden body; thus, complete point cloud collection of the hidden body is achieved.

[0097] Step 2: Point cloud data processing

[0098] According to the initial point cloud data of the target hidden body, the least square fitting method of the CYCLONE software is used to fit the edge lines of n faces and splice them; then the least square fitting method frame selection mode of the CYCLONE software is used to select a target face of the collected hidden body, select the point cloud data on the target face, perform target face domain fitting, and delete redundant point cloud data; then select n-1 faces in turn, complete n face domain fitting to obtain the edge line, use the line projected by the n faces as the boundary, delete the redundant point cloud data outside the boundary, obtain the point cloud data of the target hidden body, and form an overall target hidden body;

[0099] Step 3: Point cloud data conversion

[0100] According to the point cloud data of the plane reflector, at least 3 points are collected on the left side, right side, top side and bottom side of the plane reflector frame, a three-point normal equation is established, and the point cloud data of the target hidden body is input into the three-point normal equation to calculate the point cloud data of the target hidden body.

[0101] In one embodiment, forming a three-dimensional angle of 60°-150° among the scanner, plane reflector and target hidden body in step 1 means observing the hidden body in the hidden area through the scanner telescope and presenting it in the plane reflector; aligning the plane reflector with one surface of the hidden body and adjusting the acquisition to cover more than 80% of the edge points and vertex feature points of the hidden body for later data stitching.

[0102] The plane reflector is arranged on a cylindrical rod, which is divided into two retractable sections, the lower section of which has three brackets extending outward from the bottom. A connecting device is fixed to the upper section of the cylindrical rod, and the connecting device is connected to the cylindrical rod through a concave member. The concave member allows the mirror frame to rotate freely 360°.

[0103] In one embodiment, the plane reflector frame is provided with lock buckles on both sides and the bottom side, and the lock buckles can fix the plane reflector on the frame after being stretched, and the lock buckles automatically return to their original positions after the plane reflector is taken out;

[0104] The plane reflector is rectangular, and the plane reflector frame area is at least larger than 5mm area painted with white paint to form a white reflective area; the white reflective area is used for laser scanning and identification of point cloud data mirror conversion positioning point extraction, where the thickness of the white paint does not exceed 0.2mm; if the oil and gas thickness is greater than 0.2mm, the error of the point cloud coordinates collected at the paint later will be greater than 0.5mm, affecting the overall solution accuracy in the later stage) for the precise solution of the mirror equation in the later stage.

[0105] In one embodiment, the least squares fitting method using CYCLONE software in step 2 to fit the edges of n faces and then splice them further includes the following steps:

[0106] Step 201 uses CYCLONE software to fit the edge lines of n faces to the initial point cloud data of the target hidden body using the least squares fitting method;

[0107] Step 202: According to the fitted edge lines, each face obtains at least three vertices; the other n-1 faces are fitted in turn, and the vertices are marked;

[0108] In step 203, according to the vertices on the n faces, three vertices of the adjacent faces are randomly selected as common points for splicing, and the splicing accuracy is less than 1 mm.

[0109] Among them, deleting redundant point cloud data also includes the following steps:

[0110] a. First, select one of the faces of the target hidden body and use the least square fitting method of CYCLONE software to perform face fitting;

[0111] b. Then select n-1 faces in turn to perform face domain fitting;

[0112] c. Adjust the view of the entire point cloud of the target hidden body to a top view, set the point cloud to a non-selectable mode, use the line of the projection of the n faces of the target hidden body as the boundary, select a box to delete the redundant point cloud data;

[0113] d After deleting the redundant point cloud data, the complete target hidden body point cloud data is obtained.

[0114] In one embodiment, in step 3, a three-point normal equation is established, and the point cloud data of the target hidden body is input into the three-point normal equation to calculate the point cloud data of the target hidden body, which means that the three-point normal equation formula is established as follows:

[0115] On the point cloud of the plane reflector, collect three points R1, R2, and R3 on the left and right sides or the upper and lower sides of the mirror frame. (Change the names of the three points collected on the plane mirror from A, B, and C to R1, R2, and R3 so as not to be confused with the constant term of the plane equation Ax+By+Cz+D=0)

[0116] ② Establish the normal equation based on these three points and write a program.

[0117] Assume that the spatial coordinates of the three points R1, R2, and R3 on the plane mirror are (a1, b1, c1), (a2, b2, c2), and (a3, b3, c3), respectively. The plane equation of the plane where the mirror is located is expressed as Ax+By+Cz+D=0, where A, B, and C are the three components of the plane normal vector, and D is the intercept of the plane, all of which are parameters to be determined. The vector on the plane

[0118] Let the normal vector of the plane mirror be Since the normal vector of a plane should be perpendicular to the two non-collinear vectors in the plane, that is Head From the definition of vector cross product, we know that At the same time perpendicular to the vector and Therefore, the normal vector of the plane mirror can be obtained by the following formula: Where i, j, and k represent the unit vectors in the positive direction of the x-axis, the positive direction of the y-axis, and the positive direction of the z-axis, respectively.

[0119] Where A = (b2-b1)(c3-c1)-(b3-b1)(c2-c1) (1)

[0120] B=(c2-c1)(a3-a1)-(c3-c1)(a2-a1) (2)

[0121] C=(a2-a1)(b3-b1)-(a3-a1)(b2-b1) (3)

[0122] Substituting the coordinates of point R1 (a1, b1, c1) into the plane equation Ax+By+Cz+D=0, we can get D=-A·a1+B·b1+C·c1

[0123] Assume that the coordinates of any known point in the collected target hidden body point cloud data are (X1, Y1, Z1), and the corresponding mirror point coordinates are (X1′, Y1′, Z1′). Then, from the symmetric formula, we can know that

[0124] X1′=X1-(2A(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (4)

[0125] Y1′=Y1-(2B(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (5)

[0126] Z1′=Z1-(2C(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (6)

[0127] Substitute the target hidden body point cloud data into formulas (4), (5), and (6) to calculate the target hidden body point cloud data. The target hidden body point cloud data is the point cloud data of the actual position coordinates of the target hidden body.

[0128] In one embodiment, in another aspect, there is provided a device for collecting point cloud data of a hidden body in a hidden area, the device comprising a scanner, a plane reflector unit, and a point cloud data processing and conversion unit;

[0129] A scanner, used for observing the hidden body in the hidden area through a scanner telescope, and collecting initial point cloud data of n faces of the hidden body;

[0130] Plane reflector unit, the plane reflector is set on a cylindrical rod, the cylindrical rod is divided into two sections, the lower section is retractable, the bottom of the lower section of the cylindrical rod extends three brackets, the upper section of the cylindrical rod is fixed with a connecting device, the connecting device is connected to the cylindrical rod through a concave member, and the concave member allows the mirror frame to rotate 360 ​​degrees freely;

[0131] The plane reflector forms an angle of 60°-150° with the scanner and the concealed object, and is aimed at each surface of the concealed object to collect feature points of the edge points and vertices of the concealed object that cover more than 80% of the concealed object;

[0132] The point cloud data processing and conversion unit is used for the initial point cloud data of the target hidden body, and the edge lines of n faces are fitted and spliced ​​using the least squares fitting method of the CYCLONE software; the least squares fitting method frame selection mode of the CYCLONE software is used to select a target face of the collected hidden body, select the point cloud data on the target face, perform face domain fitting, and delete redundant point cloud data; then select n-1 faces in turn, complete n face domain fitting to obtain the edge line, use the line projected by the n faces as the boundary, delete the redundant point cloud data outside the boundary, obtain the point cloud data of the target hidden body, and form an overall target hidden body;

[0133] According to the point cloud data of the plane reflector, at least 3 points are collected on the left, right, top and bottom of the plane reflector frame, and a three-point normal equation is established. The point cloud data of the target hidden body is input into the three-point normal equation to calculate the point cloud data of the target hidden body. The point cloud data of the target hidden body mentioned here refers to the coordinate point cloud data of the actual position of the target hidden body.

[0134] In one embodiment, the plane reflector frame is provided with lock buckles on both sides and the bottom side, and the lock buckles can fix the plane reflector on the frame after being stretched, and the lock buckles automatically return to their original positions after the plane reflector is taken out;

[0135] The plane reflector is rectangular, and white paint is applied to the plane reflector frame area at least larger than 5 mm to form a white reflective area; the white reflective area is used for laser scanning and identification of point cloud data mirror conversion positioning point extraction, wherein the thickness of the white paint does not exceed 0.2 mm; and is used for solving the mirror equation.

[0136] In one embodiment, using the least squares fitting method of CYCLONE software to fit the edges of n faces and then splicing them also includes the following steps:

[0137] Step 201 uses CYCLONE software to fit the edge lines of n faces to the initial point cloud data of the target hidden body using the least squares fitting method;

[0138] Step 202: According to the fitted edge lines, each face obtains at least three vertices; the other n-1 faces are fitted in turn, and the vertices are marked;

[0139] Step 203: based on the vertices on the n faces, three vertices of the adjacent faces are randomly selected as common points for splicing, and the splicing accuracy is less than 1 mm;

[0140] Step 204 first selects one of the faces of the target hidden body and performs face domain fitting using the least square fitting method of the CYCLONE software; then selects n-1 faces in turn and performs face domain fitting;

[0141] Step 205 adjusts the view of the entire point cloud of the target hidden body to a top view, sets the point cloud to a non-selectable mode, uses the lines projected by each surface as boundaries, and deletes redundant point cloud data outside the boundaries; after deleting the redundant point cloud, the complete point cloud data of the target hidden body is obtained.

[0142] In one embodiment, establishing a three-point normal equation and inputting the point cloud data of the target hidden body into the three-point normal equation to calculate the point cloud data of the target hidden body means establishing the three-point normal equation formula as follows:

[0143] On the point cloud of the plane reflector, collect three points R1, R2, and R3 on the left and right sides or the upper and lower sides of the mirror frame. (Change the names of the three points collected on the plane mirror from A, B, and C to R1, R2, and R3 so as not to be confused with the constant term of the plane equation Ax+By+Cz+D=0)

[0144] ② Establish the normal equation based on these three points and write a program.

[0145] Assume that the spatial coordinates of the three points R1, R2, and R3 on the plane mirror are (a1, b1, c1), (a2, b2, c2), and (a3, b3, c3), respectively. The plane equation of the plane where the mirror is located is expressed as Ax+By+Cz+D=0, where A, B, and C are the three components of the plane normal vector, and D is the intercept of the plane, all of which are parameters to be determined. The vector on the plane

[0146] Let the normal vector of the plane mirror be Since the normal vector of a plane should be perpendicular to the two non-collinear vectors in the plane, that is Head From the definition of vector cross product, we know that At the same time perpendicular to the vector and Therefore, the normal vector of the plane mirror can be obtained by the following formula: Where i, j, and k represent the unit vectors in the positive direction of the x-axis, the positive direction of the y-axis, and the positive direction of the z-axis, respectively.

[0147] Where A = (b2-b1)(c3-c1)-(b3-b1)(c2-c1) (1)

[0148] B=(c2-c1)(a3-a1)-(c3-c1)(a2-a1) (2)

[0149] C=(a2-a1)(b3-b1)-(a3-a1)(b2-b1) (3)

[0150] Substituting the coordinates of point R1 (a1, b1, c1) into the plane equation Ax+By+Cz+D=0, we can get D=-A·a1+B·b1+C·c1

[0151] Assume that the coordinates of any known point in the collected target hidden body point cloud data are (X1, Y1, Z1), and the corresponding mirror point coordinates are (X1′, Y1′, Z1′). Then, from the symmetric formula, we can know that

[0152] X1′=X1-(2A(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (4)

[0153] Y1′=Y1-(2B(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (5)

[0154] Z1′=Z1-(2C(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (6)

[0155] Substitute the target hidden body point cloud data into formulas (4), (5), and (6) to calculate the target hidden body point cloud data. The target hidden body point cloud data is the point cloud data of the actual position coordinates of the target hidden body.

[0156] The embodiment of the present invention provides a method for obtaining point cloud data of hidden bodies in hidden areas based on a plane reflector, which solves the problem of difficult and incomplete point cloud data collection of hidden bodies in hidden areas and narrow areas within oil and gas field stations, buildings, and factories in three-dimensional scanning. The present invention adopts point cloud data collection, forms an angle of 60°-150° with the plane reflector, the scanner, and the hidden body, and aligns with each face of the hidden body to collect feature points of edge points and vertices covering more than 80% of the hidden body; uses the least squares fitting method of CYCLONE software to fit the edges of n faces and splice them to obtain point cloud data of the target hidden body, forming an overall target hidden body, and can quickly and completely obtain the point cloud data of the above-mentioned hidden body, achieving the effect of complete data collection within the oil and gas field station, improving the efficiency of point cloud data collection in the oil and gas field station, and providing complete and reliable data for the construction of digital stations in the later stage.

[0157] Embodiment 1 A specific embodiment of the present invention provides a method for obtaining a point cloud of a hidden body in a hidden area based on plane mirror reflection. The mirror reflection technical solution of this method can solve the problem of difficulty in collecting point clouds of hidden bodies in a hidden area. The specific steps are as follows: Step 1: During the scanning of oil and gas stations, the scanner often cannot scan some areas due to the obstruction of buildings and equipment. Therefore, the point cloud coordinates of the hidden objects are missing, which affects the integrity of the point cloud of the station. Figure 1 As shown; Due to the obstruction of walls and houses, the point cloud of the hidden object cannot be scanned. Figure 2 As shown: Step 2: For areas that cannot be scanned, first conduct a site survey and select the location for the plane mirror. Figure 3 As shown, the angle between the mirror reflected light and the mirror parallel scanner is 2, the angle between the hidden body at the mirror incident line and the mirror parallel scanner is 1, and the angle between the reflected light and the incident light is 3 Step 3: The relationship between the plane reflector and the walls, houses, and hidden objects, and the characteristics of the plane reflector when placed, such as Figure 4 As shown; the characteristics of the plane reflector and its auxiliary facilities are as follows: ①As follows Figure 5 The plane mirror is placed on a cylindrical rod, which has two sections, upper and lower, which are retractable, and three brackets extend outward from the bottom of the lower part of the cylindrical rod; ② The cylindrical rod of the mirror placement bracket is fixed with a connecting device. The upper cylindrical rod is fixed with a connecting device, which is connected to the cylindrical rod through a concave structure. With the help of the concave component, the mirror frame can rotate 360 ​​degrees freely. Figure 6 As shown: ③ There are lock buckles on both sides and the bottom of the frame. The lock buckles can be stretched to fix the plane mirror on the frame. After the plane mirror is taken out, the lock buckles will automatically return to their original positions; see Figure 7 shown.

[0168] ④The design style of the plane mirror is as follows Figure 8 As shown:

[0169] The specific features are as follows:

[0170] a) The size of the plane mirror is 500mm×300mm, and the size of the black area border is 10mm.

[0171] b) Apply white paint to the black border area. This white reflective agent helps with laser scanning and recognition, and facilitates later software recognition and extraction.

[0172] c) The thickness of white paint does not exceed 0.2 mm, which ensures the solution of the subsequent mirror equation.

[0173] Step 5: Align the reflector and use the scanner to collect data. The data should be collected in a window selection mode and at a resolution greater than 2 mm × 2 mm.

[0174] Note: Window selection scanning can be repeated according to the shape of the scanned object until all the hidden objects are collected. The data collected by this scan will not be processed. Fig. 9 As shown; keep adjusting the reflector until all objects in the hidden area are collected.

[0175] Note: After collecting a station, you can view it in the scanner. The collection of feature points in the collection area of ​​this station reaches 80%. After scanning at multiple angles for one week, the point cloud collection of hidden objects can be completed. Hidden objects refer to objects that cannot be collected by conventional stations.

[0176] For hidden objects, it is necessary to scan 4 times. After the 4 surfaces are scanned, the point cloud collection of the hidden object is completed. Fig.10 shown.

[0177] Step 2: The specific implementation steps are as follows:

[0178] ① After completing the collection of each surface, first check the integrity of the data to ensure that the integrity of the previous collection reaches 100%.

[0179] ② For the collected point cloud data, the least squares fitting method of CYCLONE software is used to fit the edges of n faces and automatically fit each edge, such as Fig.10 As shown,

[0180] ③According to the fitted edge lines, four vertices are obtained, such as Fig.11 Schematic diagram of the four vertices of the fitted edge line

[0181] ④Use the least squares fitting method of CYCLONE software to fit the edges of n faces, fit other faces in turn, and mark the vertices.

[0182] ⑤ According to the vertices on each face, arbitrarily use three vertices of adjacent faces as common points for splicing. The splicing accuracy is required to be less than 1mm.

[0183] Step 6 After the point cloud data collection is completed, the data is managed separately. The least squares fitting method of CYCLONE software is used in the box selection mode to retain the point cloud of the hidden area object and delete the redundant point cloud data.

[0184] Fig.12 This is the box selection mode. The black dashed line is the redundant point cloud data. The black line selection features are as follows:

[0185] Select a face of the hidden body collected, click on the point cloud data on the face, and use the least squares fitting method of CYCLONE software to fit the edges of n faces for surface fitting. To delete redundant point cloud data, follow the steps below:

[0186] ① First click on one of the faces to perform surface fitting. Fig. 20 .

[0187] ② Select other faces one by one, see Fig.16 , Fig.17 , Fig.18 until you have made your selection.

[0188] ③ Adjust the image to a top view, set the point cloud to non-selectable mode, use the lines projected by each surface as boundaries, and delete the redundant point cloud data outside the boundaries; see Fig.12 As shown; the four surface regions are fitted as Fig.13 As shown; select the redundant point cloud to delete Fig.14 shown.

[0189] ④ After deleting the redundant point cloud, the complete hidden body point cloud data is obtained. After deleting the redundant point cloud data, the point cloud image (top view) is as follows Fig.15 As shown in the figure, after deleting the redundant point cloud data, the point cloud image (oblique view) is as follows Fig.16 shown.

[0190] Step 7: Although the point cloud data processed in step 6 has completely captured the shape of the object, there is a mirror-symmetric pipeline between the actual object and the plane mirror, so the above point cloud needs to be processed. The data processing is as follows:

[0191] ① Such as Fig.17 As shown, on the point cloud of the plane reflector, three points R1, R2, and R3 are collected on the left and right sides or the upper and lower sides of the mirror frame.

[0192] ② Establish the normal equation based on these three points and write a program.

[0193] The coordinates of R1, R2, and R3 are (a1, b1, c1), (a2, b2, c2), and (a3, b3, c3) respectively.

[0194] The plane equation of the mirror is expressed as Ax+By+Cz+D=0, where A, B, and C are the three components of the normal vector.

[0195] A=(b2-b1)(c3-c1)-(b3-b1)(c2-c1)

[0196] B=(c2-c1)(a3-a1)-(c3-c1)(a2-a1)

[0197] C=(a2-a1)(b3-b1)-(a3-a1)(b2-b1)

[0198] Substituting the coordinates of point R1 into the plane equation, we get D = -A·a1+B·b1+C·c1

[0199] Assume that any point (X1, Y1, Z1) of the target hidden body point cloud data is known, and the corresponding mirror point coordinates are (X1′, Y1′, Z1′), then the symmetric formula shows that

[0200] X1′=X1-(2A(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (4)

[0201] Y1′=Y1-(2B(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (5)

[0202] Z1′=Z1-(2C(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (6)

[0203] Substitute the target hidden body point cloud data into formulas (4), (5), and (6) to calculate the target hidden body point cloud data. The target hidden body point cloud data is the point cloud data of the actual position coordinates of the target hidden body.

[0204] ③ Import the point cloud data from step 6 into the program and calculate the point cloud coordinates of the object.

[0205] The program refers to the above formula. The specific program operation is as follows Fig.18 , 19 , 20, 21, 22, 23, 24, 25, and 26. The program can be used to convert the point cloud data collected by the mirror into point cloud data of the actual position, and can truly restore the original properties of the hidden object.

[0206] The embodiment of the present invention provides a method for obtaining point cloud data of hidden bodies in hidden areas based on a plane reflector, which solves the problem of difficult and incomplete point cloud data collection of hidden bodies in hidden areas and narrow areas within oil and gas field stations, buildings, and factories in three-dimensional scanning. The present invention adopts point cloud data collection, forms an angle of 60°-150° with the plane reflector, the scanner, and the hidden body, and aligns with each face of the hidden body to collect feature points of edge points and vertices covering more than 80% of the hidden body; uses the least squares fitting method of CYCLONE software to fit the edges of n faces and splice them to obtain point cloud data of the target hidden body, forming an overall target hidden body, and can quickly and completely obtain the point cloud data of the above-mentioned hidden body, achieving the effect of complete data collection within the oil and gas field station, improving the efficiency of point cloud data collection in the oil and gas field station, and providing complete and reliable data for the construction of digital stations in the later stage.

Claims

1. A method for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector, characterized in that: The method comprises the following steps: Step 1: Hidden body point cloud data collection For the target hidden body in the area that cannot be scanned, first conduct a site survey and select the location for placing the plane reflector; the scanner, plane reflector and target hidden body form a three-dimensional angle of 60°-150°, and use the scanner to scan the n faces of the target hidden body in the plane reflector to collect point cloud data and obtain the initial point cloud data of the target hidden body; Step 2: Point cloud data processing According to the initial point cloud data of the target hidden body, the least square fitting method of the CYCLONE software is used to fit the edge lines of n faces and splice them; then the least square fitting method of the CYCLONE software is used to select a target face of the collected hidden body, select the point cloud data on the target face, perform target face domain fitting, and delete redundant point cloud data; then select n-1 faces in turn, complete n face domain fitting to obtain the edge line, use the line projected by the n faces as the boundary, delete the redundant point cloud outside the boundary, obtain the point cloud data of the target hidden body, and form an overall target hidden body; Step 3: Point cloud data conversion According to the point cloud data of the plane reflector, at least 3 points are collected on the left side, right side, top side and bottom side of the plane reflector frame, a three-point normal equation is established, and the point cloud data of the target hidden body is input into the three-point normal equation to calculate the point cloud data of the target hidden body.

2. The method for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector according to claim 1, characterized in that: In step 1, the three-dimensional angle of 60°-150° formed by the scanner, the plane reflector, and the target hidden body refers to the hidden body in the hidden area being observed by the scanner telescope and presented in the plane reflector; the plane reflector is aimed at one surface of the hidden body, and the feature points of the edge points and vertices covering more than 80% of the hidden body are collected; The plane reflector is arranged on a cylindrical rod, which is divided into two retractable sections, the lower section of which has three brackets extending outward from the bottom. A connecting device is fixed to the upper cylindrical rod, and the connecting device is connected to the cylindrical rod through a concave component. The concave component allows the mirror frame to rotate freely 360°.

3. The method for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector according to claim 2, characterized in that: There are lock buckles on both sides and the bottom of the plane reflector frame. The lock buckles can fix the plane mirror on the frame after being stretched. After the plane mirror is taken out, the lock buckles automatically return to their original positions. The plane reflector is rectangular, and white paint is applied to an area at least larger than 5 mm of the plane reflector frame to form a white reflective area; the white reflective area is used for laser scanning and identification of point cloud data mirror conversion positioning point extraction, wherein the thickness of the white paint does not exceed 0.2 mm.

4. The method for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector according to claim 3, characterized in that: In step 2, the least square fitting method of CYCLONE software is used to fit the edges of n faces and then splice them, which also includes the following steps: Step 201 uses CYCLONE software to fit the edge lines of n faces to the initial point cloud data of the target hidden body using the least squares fitting method; Step 202: According to the fitted edge lines, each face obtains at least three vertices; the other n-1 faces are fitted in turn, and the vertices are marked; In step 203, according to the vertices on the n faces, three vertices of the adjacent faces are randomly selected as common points for splicing, and the splicing accuracy is less than 1 mm.

5. The method for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector according to claim 3, characterized in that: The deleting of redundant point cloud data also includes the following steps: a. First, select one of the faces of the target hidden body and use the least square fitting method of CYCLONE software to perform face fitting; b. Then select n-1 faces in turn to perform face domain fitting; c. Adjust the view of the entire point cloud of the target hidden body to a top view, set the point cloud to a non-selectable mode, use the line of the projection of the n faces of the target hidden body as the boundary, select a box to delete the redundant point cloud data; d After deleting the redundant point cloud data, the complete target hidden body point cloud data is obtained.

6. The method for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector according to claim 1, characterized in that: In step 3, the three-point normal equation is established, and the point cloud data of the target hidden body is input into the three-point normal equation to calculate the point cloud data of the target hidden body. The formula for establishing the three-point normal equation is as follows: The coordinates of the white areas R1, R2, and R3 of the plane reflector frame are (a1, b1, c1), (a2, b2, c2), and (a3, b3, c3), respectively. The plane equation of the plane where the plane reflector is located is expressed as Ax+By+Cz+D=0, where A, B, and C are the three components of the normal vector, and D is the intercept of the plane, all of which are parameters to be determined. The vector on the plane is Let the normal vector of the plane mirror be Since the normal vector of a plane should be perpendicular to the two non-collinear vectors in the plane, that is and From the definition of vector cross product, we know that At the same time perpendicular to the vector and Therefore, the normal vector of the plane mirror can be obtained by the following formula: Among them, i, j, and k represent the unit vectors in the positive direction of the x-axis, the positive direction of the y-axis, and the positive direction of the z-axis respectively; Where A = (b2-b1)(c3-c1)-(b3-b1)(c2-c1) (1) B=(c2-c1)(a3-a1)-(c3-c1)(a2-a1) (2) C=(a2-a1)(b3-b1)-(a3-a1)(b2-b1) (3) Substitute the coordinates of point R1 (a1, b1, c1) into the plane equation Ax+By+Cz+D=0, and we get D=-A.a1+B.b1,+C.c1 Assume that the coordinates of any known point in the collected target hidden body point cloud data are (X1, Y1, Z1), and the corresponding mirror point coordinates are (X1′, Y1′, Z1′). Then, the symmetric formula shows that X1′=X1-(2A(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (4) Y1′=Y1-(2B(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (5) Z1′=Z1-(2C(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (6) Substitute the target hidden body point cloud data into formulas (4), (5), and (6) to calculate the target hidden body point cloud data, which is the point cloud data of the actual position coordinates of the target hidden body.

7. A device for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector, characterized in that: The device includes a scanner, a plane reflector unit, and a point cloud data processing and conversion unit; The scanner is used to observe the hidden body in the hidden area through the scanner telescope and present it in the reflector, and collect the initial point cloud data of n faces of the hidden body; The plane reflector unit is a unit in which the plane reflector is arranged on a cylindrical rod. The cylindrical rod is divided into two retractable sections, the bottom of the lower part of the cylindrical rod extends three brackets, and a connecting device is fixed on the upper part of the cylindrical rod. The connecting device is connected to the cylindrical rod through a concave component, and the concave component allows the mirror frame to rotate freely 360 degrees. The plane reflector forms an angle of 60 degrees to 150 degrees with the scanner and the hidden body, and is aligned with each surface of the hidden body to collect feature points of the edge points and vertices of the hidden body that cover more than 80% of the hidden body. The point cloud data processing and conversion unit is used for the initial point cloud data of the target hidden body, and uses the least squares fitting method of the CYCLONE software to fit the edge lines of n faces and splice them; uses the least squares fitting method of the CYCLONE software to select a target face of the hidden body, selects the point cloud data on the target face, performs face domain fitting, and deletes redundant point cloud data; then selects n-1 faces in turn, completes n face domain fitting to obtain the edge line, uses the line projected by the n faces as the boundary, deletes the redundant point cloud data outside the boundary, obtains the point cloud data of the target hidden body, and forms an overall target hidden body; According to the point cloud data of the plane reflector, at least 3 points are collected on the left side, right side, top side and bottom side of the plane reflector frame, a three-point normal equation is established, and the point cloud data of the target hidden body is input into the three-point normal equation to calculate the point cloud data of the target hidden body.

8. The device for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector according to claim 7, characterized in that: There are lock buckles on both sides and the bottom of the plane reflector frame. The lock buckles can fix the plane mirror on the frame after being stretched. After the plane mirror is taken out, the lock buckles automatically return to their original positions. The plane reflector is rectangular, and white paint is applied to an area at least larger than 5 mm of the plane reflector frame to form a white reflective area; the white reflective area is used for laser scanning and identification of point cloud data mirror conversion positioning point extraction, wherein the thickness of the white paint does not exceed 0.2 mm.

9. The device for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector according to claim 7, characterized in that: Using the least squares fitting method of CYCLONE software to fit the edges of n faces and splice them also includes the following steps: Step 201 uses CYCLONE software to fit the edge lines of n faces to the initial point cloud data of the target hidden body using the least squares fitting method; Step 202: According to the fitted edge lines, each face obtains at least three vertices; the other n-1 faces are fitted in turn, and the vertices are marked; Step 203: based on the vertices on the n faces, three vertices of the adjacent faces are randomly selected as common points for splicing, and the splicing accuracy is less than 1 mm; Step 204 first selects one of the faces of the target hidden body and performs face domain fitting using the least square fitting method of the CYCLONE software; then selects n-1 faces in turn and performs face domain fitting; Step 205 adjusts the view of the entire point cloud of the target hidden body to a top view, sets the point cloud to a non-selectable mode, uses the lines projected by each surface as boundaries, and deletes redundant point cloud data outside the boundaries; after deleting the redundant point cloud, the complete point cloud data of the target hidden body is obtained.

10. The device for obtaining point cloud data of hidden bodies in hidden areas by using a base plane reflector according to claim 7, characterized in that: The three-point normal equation is established, and the point cloud data of the target hidden body is input into the three-point normal equation to calculate the point cloud data of the target hidden body, which means that the three-point normal equation formula is established as follows: On the point cloud of the plane reflector, collect three points R1, R2, and R3 on the left and right sides or the upper and lower sides of the mirror frame, establish the normal equation based on these three points, and write a program. Assume that the spatial coordinates of the three points R1, R2, and R3 on the plane mirror are (a1, b1, c1), (a2, b2, c2), and (a3, b3, c3), respectively. The plane equation of the plane where the mirror is located is expressed as Ax+By+Cz+D=0, where A, B, and C are the three components of the plane normal vector, and D is the intercept of the plane, all of which are parameters to be determined. The vector on the plane Let the normal vector of the plane mirror be Since the normal vector of a plane should be perpendicular to the two non-collinear vectors in the plane, that is and From the definition of vector cross product, we know that At the same time perpendicular to the vector and Therefore, the normal vector of the plane mirror can be obtained by the following formula: Among them, i, j, and k represent the unit vectors in the positive direction of the x-axis, the positive direction of the y-axis, and the positive direction of the z-axis respectively; Where A = (b2-b1)(c3-c1)-(b3-b1)(c2-c1) B=(c2-c1)(a3-a1)-(c3-c1)(a2-a1) C=(a2-a1)(b3-b1)-(a3-a1)(b2-b1) Substituting the coordinates of point R1 (a1, b1, c1) into the plane equation Ax+By+Cz+D=0, we can get D=-A·a1+B·b1+C·c1 Assume that the coordinates of any known point in the collected target hidden body point cloud data are (X1, Y1, Z1), and the corresponding mirror point coordinates are (X1′, Y1′, Z1′). Then, from the symmetric formula, we can know that X1′=X1-(2A(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (4) Y1′=Y1-(2B(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (5) Z1′=Z1-(2C(AX1+BY1+CZ1+D) / A 2 +B 2 +C 2 ) (6) Substitute the target hidden body point cloud data into formulas (4), (5), and (6) to calculate the target hidden body point cloud data, which is the point cloud data of the actual position coordinates of the target hidden body.

Images