Method for extracting construction information of chamber

Abstract

The embodiment of the invention relates to the field of exploration, in particular to a method for extracting chamber structure information. According to the chamber construction information extraction method provided by the embodiment of the invention, image acquisition is carried out on the chamber wall, the first three-dimensional point cloud model is determined according to the acquired image information, laser scanning is carried out on the chamber wall, and the second three-dimensional point cloud model is determined according to the acquired laser scanning data; and determining a three-dimensional point cloud model of the tunnel wall according to the first three-dimensional point cloud model and the second three-dimensional point cloud model. According to the method, two means of non-contact measurement and extraction of construction information are combined, the measurement efficiency can be improved, and errors caused by a single measurement means are reduced.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of surveying, and in particular to a method for extracting chamber structure information. Background Art

[0002] The statements herein merely provide background information related to the present application and do not necessarily constitute prior art.

[0003] A chamber is a spatial structure excavated and built underground for a specific purpose in a mine. It is often used to install equipment, store materials or for other special purposes. Joints are cracks or fissures in rocks. In a chamber, the development of joints will directly affect the overall stability of the rock mass. If the joints are densely developed or large in scale, it may cause the rock mass to loosen, thereby increasing the risk of collapse. Therefore, in order to ensure the safe use of chambers, detailed joint measurement and analysis are required. Summary of the invention

[0004] A brief overview of the present application is provided below in order to provide a basic understanding of certain aspects of the present application. It should be understood that this overview is not an exhaustive overview of the present application. It is not intended to identify the key or important parts of the present application, nor is it intended to limit the scope of the present application. Its purpose is merely to present certain concepts in a simplified form as a prelude to a more detailed description discussed later.

[0005] An embodiment of the present application provides a method for extracting structural information of a cavern, which includes the following steps: S10: setting multiple spatial positioning control points in the cavern wall; S20: collecting images of the cavern wall; S30: determining a first three-dimensional point cloud model of the cavern wall according to the images; S40: correcting the first three-dimensional point cloud model according to the multiple spatial positioning control points to obtain a corrected first three-dimensional point cloud model of the cavern wall; S50: laser scanning the cavern wall to obtain laser scanning data of the cavern wall; S60: determining a second three-dimensional point cloud model of the cavern wall according to the laser scanning data; S70: determining a three-dimensional point cloud model of the cavern wall according to the corrected first three-dimensional point cloud model and the second three-dimensional point cloud model; S80: determining the cavern structural information according to the three-dimensional point cloud model.

[0006] The embodiment of the present application provides a method for extracting the structural information of a cavern, which captures images of the cavern wall and determines a first three-dimensional point cloud model based on the acquired image information, and performs laser scanning on the cavern wall and determines a second three-dimensional point cloud model based on the acquired laser scanning data, and then determines the three-dimensional point cloud model of the cavern wall based on the first three-dimensional point cloud model and the second three-dimensional point cloud model. It combines two means of non-contact measurement to extract structural information, which can improve measurement efficiency and reduce errors caused by a single measurement method. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to further illustrate the above and other advantages and features of the present application, the specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. The accompanying drawings together with the following detailed description are included in this specification and form a part of this specification. Elements with the same function and structure are represented by the same reference numerals. It should be understood that these drawings only describe typical examples of the present application and should not be regarded as limiting the scope of the present application.

[0008] Figure 1 It is a schematic diagram of the sequence of single-pass acquisition of cave wall images using image tilt photography technology provided in an embodiment of the present application; Figure 2 It is a schematic diagram of the embodiment of the present application when using the three-dimensional laser scanning technology to collect data on the cave wall in a direction parallel to the axis of the cave chamber; Figure 3 It is a schematic side view perpendicular to the axis of the chamber when collecting data on the chamber wall using three-dimensional laser scanning technology provided in an embodiment of the present application; Figure 4 It is a schematic diagram of a top view perpendicular to the axis of the chamber when collecting data on the chamber wall using a three-dimensional laser scanning technology provided in an embodiment of the present application; Figure 5 It is a schematic diagram of the error comparison between the actual measured occurrence of the chamber and the point cloud model interpretation value provided by the embodiment of the present application; Figure 6 is a schematic diagram of matching errors between a first three-dimensional point cloud model and a second three-dimensional point cloud model provided in an embodiment of the present application; Figure 7 is a schematic diagram of maximum point errors between the registered point cloud model and the first three-dimensional point cloud model and the second three-dimensional point cloud model provided in an embodiment of the present application; Figure 8 is a schematic diagram of the overlap rate between the registered point cloud model and the first three-dimensional point cloud model and the second three-dimensional point cloud model provided in an embodiment of the present application; Fig. 9 It is a schematic diagram of the distribution direction of the artificially observed fracture structure provided in the embodiment of the present application; Fig.10 is a schematic diagram of the distribution direction of the fracture structure obtained by interpreting the point cloud model provided in an embodiment of the present application; Fig.11 is a schematic diagram of contour lines of artificially observed fracture structure information provided in an embodiment of the present application; Fig.12 It is a schematic diagram of contour lines of fracture structure information obtained by interpreting a point cloud model according to an embodiment of the present application.

[0009] Description of reference numerals: 100, chamber; 101, floor; 110, high-density joint area; 201, scanning equipment; 202, target; 203, standard color card; 204, lighting lamp; 501, fracture dip error curve including the estimated occurrence in the manual catalog; 502, fracture dip error curve excluding the estimated occurrence in the manual catalog; 503, fracture dip error curve including the estimated occurrence in the manual catalog; 504, fracture dip error curve excluding the estimated occurrence in the manual catalog Curve; 601, maximum point error curve; 602, average point error curve; 603, mismatch curve; 701, maximum point error curve between the registered point cloud model and the second three-dimensional point cloud model; 702, maximum point error curve between the registered point cloud model and the first three-dimensional point cloud model; 801, overlap rate curve between the registered point cloud model and the first three-dimensional point cloud model; 802, overlap rate curve between the registered point cloud model and the second three-dimensional point cloud model. DETAILED DESCRIPTION

[0010] Exemplary embodiments of the present application will be described below in conjunction with the accompanying drawings. For the sake of clarity and conciseness, not all features of the actual implementation are described in the specification. However, it should be understood that many implementation-specific decisions must be made in the process of developing any such actual implementation in order to achieve the developer's specific goals, such as meeting those constraints related to the system and business, and these constraints may vary from implementation to implementation. In addition, it should be understood that although the development work may be very complex and time-consuming, it is only a routine task for those skilled in the art who benefit from the content of this application.

[0011] It is also necessary to explain here that, in order to avoid obscuring the present application due to unnecessary details, only the device structure and / or processing steps closely related to the scheme according to the present application are shown in the accompanying drawings, while other details that are not very relevant to the present application are omitted.

[0012] The disclosure below provides a plurality of different embodiments or examples to implement the present application. In order to simplify the disclosure of the present application, the parts and methods of specific examples are described below. Of course, they are merely examples, and the purpose is not to limit the present application. In the description of the embodiments of the present application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically limited.

[0013] In the process of joint measurement, the acquisition of structural information of traditional drill-and-blast chambers is usually done manually on site with the help of tools such as compasses, which is not only time-consuming and labor-intensive, but also inefficient. When a full-face tunnel boring machine is used to excavate a chamber, the chamber wall obtained by this method is smooth, and most of the fracture joint surfaces are not exposed, making it more difficult to directly measure these structural information manually. Therefore, a method that can fully identify the unexposed joints is needed to measure the joints of the smooth chamber wall.

[0014] An embodiment of the present application provides a method for extracting structural information of a cavern, which includes the following steps: S10: setting multiple spatial positioning control points in the cavern wall; S20: collecting images of the cavern wall; S30: determining a first three-dimensional point cloud model of the cavern wall according to the images; S40: correcting the first three-dimensional point cloud model according to the multiple spatial positioning control points to obtain a corrected first three-dimensional point cloud model of the cavern wall; S50: laser scanning the cavern wall to obtain laser scanning data of the cavern wall; S60: determining a second three-dimensional point cloud model of the cavern wall according to the laser scanning data; S70: determining a three-dimensional point cloud model of the cavern wall according to the corrected first three-dimensional point cloud model and the second three-dimensional point cloud model; S80: determining the cavern structural information according to the three-dimensional point cloud model.

[0015] The embodiment of the present application provides a method for extracting the structural information of a cavern, which captures images of the cavern wall and determines a first three-dimensional point cloud model based on the acquired image information, and performs laser scanning on the cavern wall and determines a second three-dimensional point cloud model based on the acquired laser scanning data, and then determines the three-dimensional point cloud model of the cavern wall based on the first three-dimensional point cloud model and the second three-dimensional point cloud model. It combines two means of non-contact measurement to extract structural information, which can improve measurement efficiency and reduce errors caused by a single measurement method.

[0016] In some embodiments, step S70 also includes the following steps: S701: respectively determine multiple overlapping points in the first three-dimensional point cloud model and the second three-dimensional point cloud model, a point on the wall of the chamber has a corresponding point in the first three-dimensional point cloud model and a corresponding point in the second three-dimensional point cloud model, and the overlapping point is the overlap of the corresponding point in the first three-dimensional point cloud model and the corresponding point in the second three-dimensional point cloud model; S702: determine the normal vectors of multiple overlapping points; S703: according to the normal vector, align the first three-dimensional point cloud model and the second three-dimensional point cloud model; S704: according to the alignment, determine the three-dimensional point cloud model of the wall of the chamber.

[0017] In some embodiments, the determined multiple overlapping points are a predetermined number (for example, not less than 5), and the multiple overlapping points are not on the same straight line, so that each overlapping point and its surrounding adjacent points can form a plane, so that the multiple overlapping points that are not on the same straight line can respectively form different planes with their adjacent points, and the multiple overlapping points have normal vectors perpendicular to the plane on their respective planes. Therefore, the planes they belong to are different and the normal vectors have different directions, thereby achieving the distinction of the normal vectors of the multiple overlapping points and better aligning the first three-dimensional point cloud model and the second three-dimensional point cloud model.

[0018] In some embodiments, step S70 also includes: S705: determining the distance between corresponding points in the first three-dimensional point cloud model and the second three-dimensional point cloud model, the corresponding point being a point on the wall of the chamber having a corresponding point in the first three-dimensional point cloud model and a corresponding point in the second three-dimensional point cloud model; S706: determining the corresponding point whose distance is greater than a predetermined value; S707: deleting the point determined in S706 from the three-dimensional point cloud model of the wall of the chamber determined in step S704.

[0019] In some embodiments, the distances between all corresponding points in the first three-dimensional point cloud model and the second three-dimensional point cloud model determined in step S705 are calculated, and points whose spacing is not within a predetermined range are removed from the three-dimensional point cloud model, so that the average value of the spacing of all points in the finally obtained three-dimensional point cloud model is a predetermined value (for example, the average value of the spacing of all points is 0.5 mm). After extracting some points from the three-dimensional point cloud model, the density of the point cloud can be optimized so that the points in the point cloud are distributed within a reasonable range, and reducing the total number of points in the point cloud can improve the calculation efficiency.

[0020] In some embodiments, in step S70, the following steps are also included: S710: determining the distance between corresponding points in the first three-dimensional point cloud model and the second three-dimensional point cloud model, the corresponding point is a point on the wall of the chamber that has a corresponding point in the first three-dimensional point cloud model and a corresponding point in the second three-dimensional point cloud model; S720: determining the points in the first three-dimensional point cloud model and the second three-dimensional point cloud model based on the distance and threshold determined in step S710; S730: determining the normal vector corresponding to the point determined in step S720; S740: determining the smoothness of the normal vector; S750: determining the three-dimensional point cloud model of the wall of the chamber based on the smoothness.

[0021] In some embodiments, a predetermined value of the distance between corresponding points in the first three-dimensional point cloud model and the second three-dimensional point cloud model is set as a distance threshold between corresponding points (for example, the distance threshold between corresponding points is set to 0.1 mm), and the points in the first three-dimensional point cloud model and the second three-dimensional point cloud model are divided according to the determined threshold of the distance between corresponding points, wherein points whose distance between corresponding points is less than the threshold are called Class I points, points whose distance between corresponding points is greater than the threshold and points that have no corresponding points in the other three-dimensional point cloud model are called Class II points.

[0022] In some embodiments, step S730 further includes the following steps: S731: determining the angle between adjacent normal vectors; S732: determining the smoothness of the normal vector based on the angle.

[0023] In some embodiments, on a smooth surface, the direction of the normal vectors of adjacent points changes less, so the angle between the normal vectors is also smaller; on the contrary, on an uneven or rough surface, the direction of the normal vectors of adjacent points changes more, and the angle between the normal vectors increases accordingly. Therefore, the smallest angle between adjacent normal vectors, that is, the smallest angle difference between the normal vectors, indicates that the three-dimensional point cloud model obtained by registration is the smoothest.

[0024] In some embodiments, step S70 also includes: S760: re-determining the distance between corresponding points in the first three-dimensional point cloud model and the second three-dimensional point cloud model, the corresponding point being a point on the wall of the chamber that has a corresponding point in the first three-dimensional point cloud model and a corresponding point in the second three-dimensional point cloud model; S770: determining the point corresponding to the distance greater than a predetermined value; S780: deleting the point determined in S770 from the three-dimensional point cloud model of the wall of the chamber determined in step S750.

[0025] In some embodiments, the distance between corresponding points in the first three-dimensional point cloud model and the second three-dimensional point cloud model is calculated, and the corresponding points in the first three-dimensional point cloud model and the second three-dimensional point cloud model are divided according to predetermined values, wherein points whose distance between corresponding points is less than a threshold are called Class I points, points whose distance between corresponding points is greater than the threshold and points that have no corresponding points in the other three-dimensional point cloud model are called Class II points, and Class II points are deleted from the three-dimensional point cloud model of the chamber wall determined in step S750 to correct the alignment results of the first three-dimensional point cloud model and the second three-dimensional point cloud model. At the same time, reducing the total number of points can make the points in the point cloud distributed within a reasonable range, and reducing the total number of points in the point cloud can improve computing efficiency.

[0026] In some embodiments, step S50 also includes the following steps: S51: setting multiple scanning starting positions in the chamber; S52: setting multiple targets on or between the multiple scanning starting positions so that there are multiple identical targets between the multiple scanning starting positions; S53: the scanning device completes the scan from one scanning starting position to the next scanning starting position, and then completes the scan from the next scanning starting position to the next scanning starting position, until the scan of the entire chamber wall is completed.

[0027] Figure 2 A schematic diagram showing the use of a three-dimensional laser scanning technology to collect data on a cave wall in a direction parallel to the cave chamber axis provided in an embodiment of the present application. In some embodiments, for example, Figure 2 As shown, the scanning device 201 is placed in the middle of the cross section of the chamber as the scanning starting position, and the distances between the multiple scanning positions are equal, and the distances between the multiple scanning positions are determined according to the size of the chamber, so that in the direction of the axis of the chamber, the distance between the scanned chamber wall and the scanning device 201 is maintained at an appropriate distance to ensure the accuracy of the acquired laser scanning data. The lens of the scanning device 201 maintains a predetermined height (for example, the lens height is not less than 1.5m) to ensure that the scanning device 201 is not too far from the top of the chamber, so that the top of the chamber can be scanned to obtain comprehensive laser scanning data. For example, when the maximum height of the chamber is 8m and the minimum height of the scanning device 201 is 1.5m, the maximum distance between the scanned chamber wall and the scanning device 201 is 6.8m, and the three-dimensional laser scanning device can effectively ensure the point cloud quality within a range of 10m, thereby reducing the probability of point loss in the point cloud.

[0028] In some embodiments, the distances between multiple scanning positions are equal, for example, the spacing is 4m. This can ensure that after the scanning point clouds obtained by scanning at each scanning position are spliced, the distance between the data points obtained by scanning at each scanning position and the scanning device 201 along the axis of the chamber is not greater than 2m, and can ensure that the data overlap rate of the scanning point clouds obtained by scanning at adjacent scanning positions is above 80%.

[0029] Figure 3 1 shows a schematic side view perpendicular to the axis of the chamber when data is collected on the chamber wall using a three-dimensional laser scanning technology according to an embodiment of the present application. In some embodiments, Figure 3As shown, a target 202 is respectively set at a predetermined height of the two side walls at 2m on both sides of the scanning device 201 along the axis direction of the chamber; in some embodiments, in step S50, a standard color card 203 is set in the chamber for coloring the second three-dimensional point cloud model, and the standard color card 203 is set at the bottom of the chamber side wall in the same plane as the scanning device 201 that is perpendicular to the axis of the chamber to improve the accuracy of the scanning device 201 in acquiring the color of the chamber wall and restore the true color of the chamber wall; an illumination lamp 204 is set 6m behind the lens of the scanning device 201 to provide sufficient light for the scanning operation to ensure the accuracy of the acquired scanning data.

[0030] Figure 4 FIG. 1 shows a schematic diagram of a top view perpendicular to the axis of a cave chamber when a three-dimensional laser scanning technology is used to collect data on a cave wall according to an embodiment of the present application. In some embodiments, for example, Figure 4 As shown, a target 202 is set on each of the two side walls 4m in front of the lens of the scanning device 201, that is, according to Figure 3 and Figure 4 As shown, a total of 6 targets 202 are arranged on the side wall of the chamber.

[0031] In some embodiments, after scanning is completed at the scanning starting position, when the scanning device 201 is moved to the next scanning position, the target 202 behind the lens of the scanning device 201 is moved at the same time. Figure 4 As shown, when the scanning device 201 moves 4m in front of the lens to the next scanning position, the two targets 202 2m behind it move forward 8m along the axial direction of the chamber, so that at the new scanning position, the relative positions of the two targets 202 whose scanning starting positions are 2m in front of the lens of the scanning device 201 become 2m behind the lens of the scanning device 201. Such movement can make the six targets 202 recycled during the laser scanning process, while avoiding the overall movement of all the targets 202. Each time the scanning device 201 is moved, only two targets 202 need to be moved at the same time, which saves the operating steps of the laser scanning and improves the scanning efficiency. At the same time, the lighting lamp 204 and the standard color card 203 move the same distance as the scanning device 201 to keep the relative positions unchanged.

[0032] In some embodiments, in step S60, a second three-dimensional point cloud model of the wall of the chamber is determined according to the scanning starting position and the target.

[0033] In some embodiments, in step S60, the data acquired by laser scanning can be imported into data processing software to create a scanning point cloud, and the position of the target 202 set during scanning is determined in the scanning point cloud, and the data of the points at the corresponding positions are marked. Since adjacent scanning positions share a part of the target 202, the target points that overlap between two adjacent scanning positions can be determined based on the data of the points marked corresponding to the target 202. The scanning point clouds acquired from the two adjacent scanning positions are spliced ​​using the target points as connecting points, and finally the second three-dimensional point cloud model of the entire chamber is obtained by splicing.

[0034] In some embodiments, the shared partial targets 202 may be arranged to be located in different planes to ensure the accuracy of stitching the scanning point clouds obtained at different scanning positions.

[0035] In some embodiments, step S60 further includes performing coordinate transformation on the second three-dimensional point cloud model according to the relationship between the relative coordinates of the plurality of positioning control points in the second three-dimensional point cloud model and the absolute coordinates thereof in the chamber.

[0036] In some embodiments, performing coordinate transformation on the second three-dimensional point cloud model facilitates further correction of the second three-dimensional point cloud model through the coordinates to improve the accuracy of the point cloud model finally acquired.

[0037] In some embodiments, in step S20, the image of the wall of the chamber can be captured using image tilt photography technology, wherein the image capture device can be a SLR camera. Figure 1 FIG. 1 is a schematic diagram showing a sequence of using image tilt photography technology to collect images of a cave wall in a single pass according to an embodiment of the present application. In some embodiments, for example, Figure 1 As shown, a bottom plate 101 is provided at the bottom of the chamber 100 for the image collector to stand on, and the image collector stays at the midline position of the bottom plate 101 during the image collection operation, holds the image collection device and faces the inner wall of one side of the chamber 100, starts shooting from the bottom of the chamber wall, and follows Figure 1 In the shooting sequence shown, 360° images are collected on the inner wall of the chamber 100, and after completing a single round of collection, the image collection is continued by moving along the center line of the bottom plate 101 to the next measurement point. The image overlap rate between adjacent images taken in the same round needs to be maintained at a predetermined value, for example, the image overlap rate is 60%-80%, to ensure that the images obtained in each round of shooting can fully reflect the structural image at the measurement point.

[0038] In some embodiments, in step S30, the images obtained by a single shot are input into the data processing software, and the overlapping and consistent image elements in the image are identified to perform image stitching. The model obtained by stitching is the real-life three-dimensional model of the geological entity photographed at a single measuring point. After completing the overall image acquisition of the chamber 100, the real-life three-dimensional models obtained at each measuring point are spliced ​​together to obtain the first three-dimensional point cloud model of the entire chamber 100.

[0039] In some embodiments, in step S40, since multiple spatial positioning control points are set in the wall of the chamber in step S10, after the first three-dimensional point cloud model is constructed, multiple spatial positioning control points are marked in the model, and their real space coordinate information is input, thereby realizing the correction of the three-dimensional spatial form of the first three-dimensional point cloud model, and the fracture parameters extracted from the corrected first three-dimensional point cloud model are the true values ​​of the real geographic space position.

[0040] Figure 6 A schematic diagram showing the matching error between the first three-dimensional point cloud model and the second three-dimensional point cloud model provided in an embodiment of the present application is shown. In some embodiments, for example, Figure 6 As shown, the point errors of corresponding points of the first three-dimensional point cloud model and the second three-dimensional point cloud model are calculated to obtain the corresponding maximum point error curve 601, average point error curve 602 and mismatch curve 603 at different mileages of the chamber 100, wherein the maximum point error range is 4mm-13mm, and the average point error range is 2mm-4mm.

[0041] Figure 7 A schematic diagram showing the maximum point error between the registered point cloud model and the first three-dimensional point cloud model and the second three-dimensional point cloud model provided by an embodiment of the present application is shown. In some embodiments, for example Figure 7 As shown, the distance between the point in the registered point cloud model and the corresponding point in the second three-dimensional point cloud model is calculated to obtain the maximum point error curve 701 between the registered point cloud model and the second three-dimensional point cloud model, where the maximum point error range is 1.5mm-4mm; the distance between the point in the registered point cloud model and the corresponding point in the first three-dimensional point cloud model is calculated to obtain the maximum point error curve 702 between the registered point cloud model and the first three-dimensional point cloud model, where the maximum point error range is 1mm-4.5mm. Therefore, compared with the registered point cloud model Figure 6 The maximum point error of the first three-dimensional point cloud model and the second three-dimensional point cloud model obtained by a single method is shown to be significantly reduced, indicating that the point cloud model obtained after aligning the first three-dimensional point cloud model and the second three-dimensional point cloud model is closer to the real scene model of the cave wall.

[0042] Figure 8A schematic diagram showing the overlap ratio between the registered point cloud model and the first three-dimensional point cloud model and the second three-dimensional point cloud model provided in an embodiment of the present application is shown. In some embodiments, for example Figure 8 As shown, the overlap rate between the points in the registered point cloud model and the points in the first three-dimensional point cloud model is calculated to obtain the overlap rate curve 801 between the registered point cloud model and the first three-dimensional point cloud model; the overlap rate between the points in the registered point cloud model and the points in the second three-dimensional point cloud model is calculated to obtain the overlap rate curve 802 between the registered point cloud model and the second three-dimensional point cloud model. Most of the overlap rates are above 90%, and some overlap rates are as high as 97%. Therefore, the registered point cloud model retains most of the points in the first three-dimensional point cloud model and the second three-dimensional point cloud model, effectively making up for the data of points lost in the first three-dimensional point cloud model and the second three-dimensional point cloud model obtained by a single method due to other factors such as the environment, making the obtained point cloud data more accurate and comprehensive.

[0043] In some embodiments, in step S50, the scanning parameters may be set to: resolution 1 / 4, quality parameter ×4, horizon weighted photometry, and HDR or night mode not turned on. This can optimize the point cloud details and colors by controlling the lighting conditions and resolution, making the data acquired by laser scanning more accurate.

[0044] In some embodiments, in step S80, data processing software can be used to further analyze the point cloud model obtained in step S70 and extract characteristic parameters, wherein the cracks revealed in the point cloud model can be identified in a predetermined manner. For example, the experimenter observes the cracks revealed in the point cloud model in the data processing software and manually draws the traces of the cracks in the point cloud model to generate a crack surface and obtain the characteristic parameters of the crack to reduce the error of automatic identification of the traces and improve the utilization rate of the point cloud model.

[0045] In some embodiments, the chamber whose structural information needs to be extracted can be first manually measured and cataloged, that is, the cracks on the walls of the chamber can be numbered, measured and drawn, and a point cloud model of the walls of the chamber can be constructed through image acquisition methods, and the cracks on the walls of the chamber can be identified and extracted from the point cloud model, and the cracks obtained by manual measurement and cataloging can be compared with the cracks with the same number obtained by image acquisition, and the error values ​​of the same crack under different measurement methods can be calculated, and the obtained error value can be compared with a predetermined error threshold. If the error value is within the threshold range, it indicates that the method of constructing a point cloud model to extract structural information can control the error value within a reasonable range, that is to say, the feasibility and accuracy of extracting structural information by constructing a point cloud model can be guaranteed.

[0046] The feasibility and accuracy of extracting structural information from a point cloud model is verified through a specific example below.

[0047] In this embodiment, in a certain tunnel drilling and blasting section, there are 1119 cracks identified and extracted by the image acquisition method, and 509 cracks cataloged by the manual observation method. By comparing the occurrence of the same numbered cracks in the cataloging results of the two methods, it is found that the average error of the occurrence of the cracks cataloged by the image acquisition method and the manual observation method is 10.9°, and the average error of the inclination is 6.8°. Due to the large height of the chamber, during the manual observation process, the cataloging personnel cannot accurately measure the cracks at the top of the tunnel, and mainly rely on naked eye observation and estimation. Therefore, when comparing the occurrence of the cracks cataloged by image acquisition and manual observation, it is necessary to remove such estimated cracks, and the remaining manually cataloged cracks after removal are the actual measurements of the cataloging personnel.

[0048] Figure 5 A schematic diagram showing the error comparison between the actual measured occurrence of the chamber and the point cloud model interpretation value provided by the embodiment of the present application is shown as follows: Figure 5 As shown, the trends of the fracture dip error curve 501 including the estimated occurrence in the manual catalog and the fracture dip error curve 502 not including the estimated occurrence in the manual catalog are roughly the same, the average dip error between the fracture occurrence extracted by the point cloud model interpretation and the measured fracture occurrence is 9.8°, and the dip error when the estimated occurrence in the manual catalog is removed is generally smaller than the dip error when the estimated occurrence in the manual catalog is included, and the error value is significantly reduced compared with the value before removal; similarly, the trends of the fracture dip angle error curve 503 including the estimated occurrence in the manual catalog and the fracture dip angle error curve 504 not including the estimated occurrence in the manual catalog are roughly the same, the average dip angle error between the fracture occurrence extracted by the point cloud model interpretation and the measured fracture occurrence is 6.1°, and the dip angle error when the estimated occurrence in the manual catalog is removed is generally smaller than the dip angle error when the estimated occurrence in the manual catalog is included, and the error value is significantly reduced compared with the value before removal.

[0049] Fig. 9 A schematic diagram showing the distribution direction of the artificially observed fracture structure provided by the embodiment of the present application is shown as follows: Fig. 9 As shown, the cracks obtained by manual observation are mostly distributed in the 10°-20° direction and the 190°-200° direction, indicating that the cracks are mainly developed in these two directions; Fig.10 A schematic diagram showing the distribution direction of the fracture structure obtained by interpreting the point cloud model provided in an embodiment of the present application is shown as follows: Fig.10 As shown, the cracks obtained by point cloud model interpretation are mostly distributed in the 20°-30° direction and the 200°-210° direction, indicating that the cracks are mainly developed in these two directions. Therefore, the distribution error between the crack occurrence extracted by point cloud model interpretation and the manually recorded measured crack occurrence is within 10°, indicating that the accuracy of crack extraction by point cloud model interpretation is controlled within the threshold.

[0050] Fig.11 A schematic diagram of contour lines of artificially observed fracture structure information provided by an embodiment of the present application is shown, such as Fig.11 As shown, the contour lines in the directions of 90°-110° and 270°-290° are closed, indicating that the joints in these two directions are densely distributed and are high-density joint areas 110; Fig.12 The schematic diagram of the contour lines of the fracture structure information obtained by interpreting the point cloud model provided by the embodiment of the present application is shown as follows: Fig.12 As shown, the contour lines in the directions of 90°-100° and 280°-300° are closed, indicating that the joints in these two directions are densely distributed, which is a high-density joint area 110. Therefore, the distribution error between the fracture occurrence extracted by point cloud model interpretation and the manually recorded measured fracture occurrence is within 10°, indicating that the accuracy of fracture extraction by point cloud model interpretation is controlled within the threshold.

[0051] The above statistics show that the average error between the fracture occurrence extracted by point cloud model interpretation and the manually catalogued measured fracture occurrence is within a controllable range, and the maximum error angle does not exceed 10°. This indicates that for objects that cannot be measured manually, the structural information of joints or faults can be quickly extracted by constructing and interpreting point cloud models. For some fractures that can be measured manually, the measured data can be used to verify the accuracy of the structural information obtained based on the point cloud model.

[0052] Regarding the embodiments of the present application, it should also be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other to obtain new embodiments.

[0053] The above are only specific implementation methods of the present application, but the protection scope of the present application is not limited thereto. The protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for extracting chamber structure information, characterized in that: It includes the following steps: S10: setting a plurality of spatial positioning control points in the wall of the chamber; S20: collecting an image of the wall of the chamber; S30: Determine a first three-dimensional point cloud model of the wall of the chamber according to the image; S40: Correcting the first three-dimensional point cloud model according to the multiple spatial positioning control points to obtain a corrected first three-dimensional point cloud model of the wall of the chamber; S50: performing laser scanning on the wall of the chamber to obtain laser scanning data of the wall of the chamber; S60: Determine a second three-dimensional point cloud model of the wall of the chamber according to the laser scanning data; S70: determining a three-dimensional point cloud model of the wall of the chamber according to the corrected first three-dimensional point cloud model and the second three-dimensional point cloud model; S80: Determine the chamber structure information according to the three-dimensional point cloud model determined in step S70.

2. The extraction method according to claim 1, characterized in that In step S70, the following steps are also included: S701: Determine a plurality of coincident points in the first three-dimensional point cloud model and the second three-dimensional point cloud model respectively, a certain point of the wall of the chamber has a corresponding point in the first three-dimensional point cloud model and a corresponding point in the second three-dimensional point cloud model, and the coincident point is a point where a corresponding point in the first three-dimensional point cloud model coincides with a corresponding point in the second three-dimensional point cloud model; S702: Determine the normal vectors of the multiple coincident points; S703: Registering the first three-dimensional point cloud model and the second three-dimensional point cloud model according to the normal vector; S704: Determine a three-dimensional point cloud model of the wall of the chamber according to the registration.

3. The extraction method according to claim 1, characterized in that In step S70, the following steps are also included: S710: Determine the distance between corresponding points in the first three-dimensional point cloud model and the second three-dimensional point cloud model, wherein the corresponding point is a point on the wall of the chamber that has a corresponding point in the first three-dimensional point cloud model and a point that has a corresponding point in the second three-dimensional point cloud model; S720: Determine points in the first three-dimensional point cloud model and the second three-dimensional point cloud model according to the distance and threshold determined in step S710; S730: Determine the normal vector corresponding to the point determined in step S720; S740: Determine the smoothness of the normal vector; S750: Determine a three-dimensional point cloud model of the wall of the chamber according to the degree of smoothness.

4. The extraction method according to claim 3, characterized in that In step S730, the following steps are also included: S731: Determine the angle between adjacent normal vectors; S732: Determine the smoothness of the normal vector according to the angle.

5. The extraction method according to claim 2, characterized in that The step S70 also includes: S705: Determine the distance between corresponding points in the first three-dimensional point cloud model and the second three-dimensional point cloud model, wherein the corresponding point is a point on the wall of the chamber that has a corresponding point in the first three-dimensional point cloud model and a point that has a corresponding point in the second three-dimensional point cloud model; S706: Determine the point corresponding to the distance being greater than a predetermined value; S707: Delete the points determined in step S706 from the three-dimensional point cloud model of the chamber wall determined in step S704.

6. The extraction method according to claim 3 or 4, characterized in that The step S70 also includes: S760: Determine again the distance between corresponding points in the first three-dimensional point cloud model and the second three-dimensional point cloud model, wherein the corresponding point is a point on the wall of the chamber that has a corresponding point in the first three-dimensional point cloud model and a point that has a corresponding point in the second three-dimensional point cloud model; S770: Determine a point corresponding to a distance greater than a predetermined value; S780: Delete the points determined in step S770 from the three-dimensional point cloud model of the wall of the chamber determined in step S750.

7. The extraction method according to claim 1, characterized in that In step S50, the following steps are also included: S51: setting a plurality of scanning starting positions in the chamber; S52: setting a plurality of targets on or between the plurality of scanning starting positions, so that there are a plurality of identical targets between the plurality of scanning starting positions; S53: The scanning device completes scanning from one scanning starting position to the next scanning starting position, and then completes scanning from the next scanning starting position to the next scanning starting position, until the scanning of the entire wall of the chamber is completed.

8. The extraction method according to claim 7, characterized in that In step S60, a second three-dimensional point cloud model of the wall of the chamber is determined according to the scanning start position and the target.

9. The extraction method according to claim 7, characterized in that In step S50, a standard color card is set in the chamber for coloring the second three-dimensional point cloud model.

10. The extraction method according to claim 1, characterized in that In step S60, it also includes performing coordinate transformation on the second three-dimensional point cloud model according to the relationship between the relative coordinates of the multiple positioning control points in the second three-dimensional point cloud model and the absolute coordinates in the chamber.

Images