A method for obtaining rock mass structure information based on rock mass surface and hole scanning

Abstract

The present invention provides a method for obtaining rock mass structural information based on rock mass surface and borehole scanning, comprising: collecting an original digital image of the exposed surface of the rock mass excavation; collecting an original digital image of the inner wall of the borehole and recording the borehole position information; processing the original digital image to obtain a binary image containing only crack patterns; obtaining the coordinates of all crack patterns in the binary image, and obtaining a function expression of the crack patterns in the binary image of the inner wall of the borehole by fitting; analyzing and calculating to obtain a crack information set in the rock mass at the engineering site; inputting the crack information set into an information induction model for classification and organization to obtain a structural surface information set; constructing a three-dimensional model of the structural surface; and building a rock mass structural model to obtain rock mass structural information. The present invention makes the process of obtaining structural surface parameters more reliable and complete, and provides a more efficient, convenient and reliable method for obtaining rock mass structural information in actual engineering projects, which has a good guiding significance for engineering sites.

Description

Technical Field

[0001] The present invention relates to the technical field of building three-dimensional rock structure models, and in particular to a method for acquiring rock structure information based on rock surface and hole scanning. Background Art

[0002] Rock masses evolve through lengthy geological processes. Tectonic forces create numerous structural planes within the rock mass, which divide it into blocks of varying shapes and sizes. These blocks are called structural bodies. Together, these structural planes and structural bodies form the rock mass structure. The geometric parameters of these structural planes influence its stability, controlling its deformation, failure, and seepage paths. Therefore, constructing a three-dimensional model of rock mass structure is crucial for practical engineering and is key to economical design and safe construction.

[0003] In practical engineering, the main method for obtaining structural surface parameters in rock masses is to measure exposed rock surfaces. Alternatively, structural surface parameters can be inferred from structural surface traces on multiple exposed surfaces. However, these structural surface parameters often suffer from missing information and low reliability, making them ineffective in providing critical guidance at the engineering site. By effectively combining information about exposed rock surfaces with that of the borehole's inner wall, and computing and organizing these information through a neural network, the process of obtaining structural surface parameters is more reliable and comprehensive. This provides a more efficient, convenient, and reliable method for obtaining rock mass structural information in practical engineering, and has significant guiding significance for engineering sites. Summary of the Invention

[0004] The purpose of the present invention is to provide an efficient, convenient and reliable method for obtaining rock structure information based on rock surface and hole scanning, so as to solve the technical problems of information loss and low reliability of existing rock structure information acquisition methods.

[0005] In order to achieve the above object, the present invention provides a method for obtaining rock mass structure information based on rock mass surface and hole scanning, comprising the following steps:

[0006] S1, collects digital images of the original appearance of the exposed surface of the rock mass excavation at the engineering site;

[0007] S2, drilling a hole in the rock mass at the construction site, collecting a digital image of the original appearance of the inner wall of the hole and recording the hole location information;

[0008] S3, processing the original digital images collected in S1 and S2, strengthening and highlighting the cracks in the images as the focus, and obtaining a binary image containing only the crack graphics;

[0009] S4, identifying and extracting crack patterns in the binary image, obtaining coordinates of all crack patterns in the binary image, and fitting the coordinates of the crack patterns in the binary image of the inner wall of the borehole to obtain a function expression of the crack patterns in the binary image of the inner wall of the borehole;

[0010] S5, analyzing and calculating the function expression and coordinates of the crack graph to obtain a set of crack information in the rock mass at the engineering site;

[0011] S6, inputting the fracture information set into the information summarization model, classifying and sorting all fracture information sets to obtain the structural surface information set;

[0012] S7, performing interpolation calculation on the structural surface information set according to an interpolation method to construct a three-dimensional model of the structural surface;

[0013] S8, arranging the three-dimensional structural surface model in the rock mass surface frame model according to the spatial position information, constructing a rock mass structural model, and obtaining rock mass structural information through the rock mass structural model.

[0014] Furthermore, in S1, shooting points are arranged at a preset distance in front of the exposed surface of the rock mass, and a digital camera is used to capture the original digital image of the exposed surface of the excavated rock mass. In S2, several horizontal boreholes are drilled on the exposed surface of the excavated rock mass and the position information of the boreholes is recorded. Then, the inside of the boreholes is repeatedly flushed with water to remove impurities from the boreholes. After the boreholes are dry, image acquisition is started at a preset distance in front of the exposed surface of the rock mass. The straight borehole TV probe is slowly placed into the borehole along the wall of the borehole, and the drill bit is slowly pushed in to the bottom of the borehole to obtain an image of the inner wall of the borehole.

[0015] Furthermore, the image processing methods in S3 include scaling, grayscale, filtering, edge detection, thinning, and iterative connection.

[0016] Furthermore, in S4, an image plane coordinate system is established with the upper left corner endpoint of the image as the origin, the horizontal right direction as the positive direction of the u-axis, the vertical downward direction as the positive direction of the v-axis, and pixels as units; the crack patterns in the image are numbered, and the coordinates of all crack patterns in the image plane coordinate system are obtained, and each crack pattern corresponds to a crack coordinate set; the coordinates of the crack patterns in the borehole inner wall image are fitted and calculated using a sine function model to obtain the function expressions of all crack patterns in the borehole inner wall image.

[0017] Furthermore, the sine function model is:

[0018]

[0019] Where A represents the amplitude of the crack graphic function in the borehole inner wall image, u and v represent the two coordinate variables of the image plane coordinate system, represents the initial phase of the crack graphic function in the borehole inner wall image, and c represents the upward translation distance of the crack graphic function in the borehole inner wall image.

[0020] Furthermore, in S5, a cylindrical coordinate system is established with the center of the borehole as the origin, vertical downward as the polar axis direction, counterclockwise as the positive angle direction, and horizontal inside the borehole as the positive z-axis direction. The inclination angle α of the structural surface in the cylindrical coordinate system is obtained by the amplitude A of the crack graphic function in the borehole inner wall image. The initial phase of the function is obtained by the initial phase of the function. Obtain the inclination β of the structural surface in the cylindrical coordinate system, and obtain the occurrence information set of the structural surface in the cylindrical coordinate system. Each crack pattern corresponds to one occurrence information set. The calculation formula of the structural surface inclination angle in the cylindrical coordinate system is:

[0021]

[0022] Wherein, α represents the inclination angle of the structural surface, A represents the amplitude of the crack graphic function in the borehole inner wall image, and M represents the size width of the borehole inner wall image;

[0023] The calculation formula of the structural surface inclination in the cylindrical coordinate system is:

[0024]

[0025] Among them, β represents the angle between the inclination of the structural surface and the section direction of the borehole inner wall image, Represents the initial phase of the crack graphic function in the borehole inner wall image;

[0026] The fracture coordinate set in the borehole inner wall image plane coordinate system is converted into the fracture coordinate set in the cylindrical coordinate system through coordinate transformation. The formula for converting the image plane coordinate system into the cylindrical coordinate system is:

[0027]

[0028] Wherein, θ, ρ, and z represent the three coordinate variables of the cylindrical coordinate system, u and v represent the two coordinate variables of the image plane coordinate system, M represents the width of the borehole inner wall image, r represents the actual borehole radius, and k represents the ratio of the actual length to the pixel length in the borehole inner wall image.

[0029] A three-dimensional rectangular coordinate system is established with the upper left end point of the shooting range of the rock mass excavation exposure surface as the origin, the horizontal right direction as the positive direction of the x-axis, the vertical downward direction as the positive direction of the y-axis, and the horizontal direction inside the rock mass as the positive direction of the z-axis. The unit is meter. The fracture coordinate set and the occurrence information set in the cylindrical coordinate system are converted into the fracture coordinate set and the occurrence information set in the three-dimensional rectangular coordinate system through coordinate transformation. The formula for converting the cylindrical coordinate system to the three-dimensional rectangular coordinate system is:

[0030]

[0031] Where x, y, and z represent the three coordinate variables of the three-dimensional rectangular coordinate system, θ, ρ, and z represent the three coordinate variables of the cylindrical coordinate system, and C1 and C2 are constants.

[0032] The fracture coordinate set in the exposure surface image plane coordinate system is converted into the fracture coordinate set in the three-dimensional rectangular coordinate system through coordinate transformation. The formula for converting the image plane coordinate system into the three-dimensional rectangular coordinate system is:

[0033]

[0034] Among them, x, y, and z represent the three coordinate variables of the three-dimensional rectangular coordinate system, u and v represent the two coordinate variables of the image plane coordinate system, and l represents the ratio of the actual length to the pixel length in the rock exposure surface image;

[0035] After sorting, a fracture information set consisting of occurrence information and coordinate information is obtained. The fracture information set corresponding to the fractures on the inner wall of the borehole contains occurrence information and coordinate information, while the fracture information set corresponding to the fractures on the excavation exposed surface only contains coordinate information.

[0036] Furthermore, the information induction model in S6 is obtained by training the convolutional neural network to summarize the crack information set and integrate the crack information sets belonging to the same structural surface into one structural surface information set. The specific training process is as follows:

[0037] Construct a convolutional neural network model; mark the crack information sets corresponding to the cracks on the excavation exposed surface and the inner wall of the borehole, and input them into the convolutional neural network as training samples; continuously adjust the model parameters through the loss function and gradient descent algorithm; repeatedly iterate until the model converges, and establish an information induction model.

[0038] Furthermore, the Kriging interpolation method is used in S7 to obtain the value of the unknown point using the values ​​of all known points. The specific process is: all known structural surface coordinate information is calculated to obtain the horizontal distance d between the two coordinate points. ij And the semivariance r of the elevation value between two coordinate points ij , and obtain the array (d ij ,r ij ); obtain the functional relationship between distance and semi-variance by fitting the theoretical model. The fitting theoretical model adopts three models: spherical model, Gaussian model and exponential model, and find the optimal functional relationship through optimization algorithm; obtain the semi-variance of the elevation values ​​between all two points through the optimal functional relationship; obtain the weight coefficients corresponding to all coordinate points by solving the equation group of weight coefficients; substitute all weight coefficients into the Kriging interpolation formula to obtain the elevation values ​​of all unknown points.

[0039] Furthermore, the horizontal distance d between the coordinate points ij :

[0040]

[0041] Among them, d ij represents the horizontal distance between the i-th known point and the j-th known point, (x i ,y i ) is the horizontal coordinate of the i-th known point, (x j ,y j ) is the horizontal azimuth coordinate of the jth known point;

[0042] Semivariance r of elevation values ​​between coordinate points ij :

[0043]

[0044] Among them, r ij represents the semivariance of the elevation value between the i-th known point and the j-th known point, z i and z i are the elevation values ​​of the i-th known point and the j-th known point respectively;

[0045] Spherical model:

[0046]

[0047] Gaussian model:

[0048]

[0049] Exponential Model:

[0050]

[0051] Where r represents semivariance, d represents horizontal azimuth distance, C0 represents nugget value, C represents partial sill value, and a represents range;

[0052] Solve the system of equations for the weight coefficients:

[0053]

[0054] Among them, r ij represents the semivariance of the elevation value between the i-th known point and the j-th known point, r it represents the semivariance of the elevation value between the i-th known point and the t-th unknown point, μ is the Lagrange multiplier, λ i Indicates the weight coefficient corresponding to the i-th known point when obtaining the elevation value of the t-th unknown point;

[0055] Kriging interpolation formula:

[0056]

[0057] in, represents the estimated value of the elevation of the tth unknown point, λ i Indicates the weight coefficient corresponding to the i-th known point, z i Represents the elevation value of the i-th known point.

[0058] Furthermore, the rock mass structure model in S8 is composed of a rock mass surface framework model and a structural surface three-dimensional model. When constructing the rock mass surface framework model, it is necessary to determine it according to the artificially defined rock mass research scope.

[0059] The above solution of the present invention has the following beneficial effects:

[0060] The rock structure information acquisition method based on rock surface and in-hole scanning provided by the present invention combines the information extracted from the rock exposure surface and the borehole television, and calculates and organizes it through a neural network, making the acquisition process of structural surface parameters more reliable and complete. This provides a more efficient, convenient and reliable way to obtain rock structure information in actual engineering, and has great guiding significance for engineering sites.

[0061] Other beneficial effects of the present invention will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a flowchart of the steps of the present invention;

[0063] Figure 2 This is a schematic diagram of the process of capturing images of the exposed surface of a rock mass excavation according to the present invention;

[0064] Figure 3 Schematic diagram of the process of collecting borehole inner wall images according to the present invention;

[0065] Figure 4 Schematic diagram of the three-dimensional model of the structural surface of the present invention;

[0066] Figure 5 Schematic diagram of the rock mass structure model of the present invention. DETAILED DESCRIPTION

[0067] The following describes the embodiments of the present disclosure through specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present disclosure.

[0068] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on this disclosure, it should be understood by those skilled in the art that an aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects described herein can be used to implement an apparatus and / or practice a method. In addition, other structures and / or functionalities other than one or more of the aspects described herein can be used to implement this apparatus and / or practice this method.

[0069] It should also be noted that the diagrams provided in the following embodiments are merely schematic illustrations of the basic concepts of the present disclosure. The diagrams only show components relevant to the present disclosure and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the configuration, quantity, and proportion of each component may be varied at will, and the component layout may be more complex. Furthermore, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, those skilled in the art will appreciate that the described aspects may be practiced without these specific details.

[0070] like Figure 1 As shown, an embodiment of the present invention provides a method for obtaining rock mass structure information based on rock mass surface and hole scanning, comprising the following steps:

[0071] S1, collects digital images of the original appearance of the exposed surface of the rock excavation at the construction site.

[0072] In this embodiment, a shooting point is arranged 4 m in front of the rock mass exposure surface, and a digital camera is used to shoot the original digital image of the rock mass excavation exposure surface to obtain the excavation exposure surface image set {exP1, exP2, exP3…exP a}, the process of taking images of rock excavation exposed surface is as follows Figure 2 shown.

[0073] S2, drill a hole in the rock mass at the project site, collect the original digital image of the inner wall of the hole and record the hole location information.

[0074] In this embodiment, nine horizontal boreholes are drilled on the exposed surface of the rock mass and their location information is recorded. The borehole diameter is 60 mm and the drilling depth is 2 m. The inside of the borehole is then repeatedly flushed with water to remove impurities such as rock debris. After the borehole is dry, the operator begins image acquisition about 30 cm in front of the exposed surface of the rock mass. A 52 mm diameter borehole television probe is slowly placed into the borehole along the borehole wall, and the drill bit is slowly pushed in to the bottom of the borehole to obtain an image of the borehole inner wall. The borehole inner wall image set {bhP1, bhP2, bhP3…bhP b}, the process of collecting borehole inner wall images is as follows Figure 3 shown.

[0075] S3, processing the original digital images collected in S1 and S2, strengthening and highlighting the crack patterns in the images as the focus, and obtaining a binary image containing only the crack patterns.

[0076] In this embodiment, the crack pattern in the original digital image is enhanced by an image processing method, and a binary image is obtained after processing, and an excavation exposure surface image set {exBP1, exBP2, exBP3…exBP a} and the binary image set of the borehole inner wall {bhBP1, bhBP2, bhBP3…bhBP b}, providing favorable prerequisites for subsequent recognition operations.

[0077] Specifically, the image processing method includes scaling, grayscale, filtering, edge detection, thinning, and iterative connection. The method itself is an existing technology and will not be described in detail here.

[0078] S4, identifying and extracting the crack patterns in the binary image obtained in S3, obtaining the coordinates of all crack patterns in the binary image, and fitting the coordinates of the crack patterns in the binary image of the inner wall of the borehole to obtain a function expression of the crack patterns in the binary image of the inner wall of the borehole.

[0079] The specific steps of crack pattern recognition and extraction in this embodiment are as follows: first, with the upper left corner endpoint of the image as the origin, the horizontal right direction as the positive direction of the u-axis, and the vertical downward direction as the positive direction of the v-axis, an image plane coordinate system is established in pixels, the crack patterns in the image are numbered, and the coordinates of all crack patterns in the image plane coordinate system are obtained. Each crack pattern corresponds to a crack coordinate set; finally, a sine function model is used to fit the coordinates of the crack patterns in the borehole inner wall image to obtain a function expression of all crack patterns in the borehole inner wall image. The above-mentioned sine function model is:

[0080]

[0081] Where A represents the amplitude of the crack graphic function in the borehole inner wall image, u and v represent the two coordinate variables of the image plane coordinate system, represents the initial phase of the crack graphic function in the borehole inner wall image, and c represents the upward translation distance of the crack graphic function in the borehole inner wall image.

[0082] In this embodiment, in an independently established image plane coordinate system, the pixel coordinates of the crack graphics in the binary image are extracted by a pre-set algorithm to obtain the crack coordinate set exPC(1,2…a) of the excavation exposure surface image group a and the crack coordinate set bhPC(1,2…b) of the borehole inner wall image group b. Finally, the function expression of all the crack graphics in the borehole inner wall image is obtained by fitting calculation.

[0083] S5, analyzing and calculating the function expression and coordinates of the crack graph obtained in S4 to obtain a crack information set in the rock mass at the engineering site.

[0084] The specific steps of the function expression and coordinate analysis calculation of the crack pattern in this embodiment are as follows: first, a cylindrical coordinate system is established with the center of the borehole as the origin, vertical downward as the polar axis direction, counterclockwise as the positive angle direction, and horizontal inside the borehole as the positive z-axis direction, with meters as the unit. The inclination angle α of the structural surface in the cylindrical coordinate system is obtained by the amplitude A of the crack pattern function in the borehole inner wall image. The initial phase of the function is obtained by the initial phase of the function. Obtain the inclination β of the structural surface in the cylindrical coordinate system, and obtain the occurrence information set of the structural surface in the cylindrical coordinate system. Each crack pattern corresponds to one occurrence information set. The calculation formula of the structural surface inclination angle in the cylindrical coordinate system is:

[0085]

[0086] Wherein, α represents the inclination angle of the structural surface, A represents the amplitude of the crack graphic function in the borehole inner wall image, and M represents the size width of the borehole inner wall image;

[0087] The calculation formula of the structural surface inclination in the cylindrical coordinate system is:

[0088]

[0089] Among them, β represents the angle between the inclination of the structural surface and the section direction of the borehole inner wall image, Represents the initial phase of the crack graphic function in the borehole inner wall image;

[0090] The fracture coordinate set in the borehole inner wall image plane coordinate system is converted into the fracture coordinate set in the cylindrical coordinate system through coordinate transformation. The formula for converting the image plane coordinate system into the cylindrical coordinate system is:

[0091]

[0092] Wherein, θ, ρ, and z represent the three coordinate variables of the cylindrical coordinate system, u and v represent the two coordinate variables of the image plane coordinate system, M represents the width of the borehole inner wall image, r represents the actual borehole radius, and k represents the ratio of the actual length to the pixel length in the borehole inner wall image.

[0093] A three-dimensional rectangular coordinate system is established with the upper left end point of the shooting range of the rock mass excavation exposure surface as the origin, the horizontal right direction as the positive direction of the x-axis, the vertical downward direction as the positive direction of the y-axis, and the horizontal direction inside the rock mass as the positive direction of the z-axis. The unit is meter. The fracture coordinate set and the occurrence information set in the cylindrical coordinate system are converted into the fracture coordinate set and the occurrence information set in the three-dimensional rectangular coordinate system through coordinate transformation. The formula for converting the cylindrical coordinate system to the three-dimensional rectangular coordinate system is:

[0094]

[0095] Where x, y, and z represent the three coordinate variables of the three-dimensional rectangular coordinate system, θ, ρ, and z represent the three coordinate variables of the cylindrical coordinate system, and C1 and C2 are constants.

[0096] The fracture coordinate set in the exposure surface image plane coordinate system is converted into the fracture coordinate set in the three-dimensional rectangular coordinate system through coordinate transformation. The formula for converting the image plane coordinate system into the three-dimensional rectangular coordinate system is:

[0097]

[0098] Among them, x, y, and z represent the three coordinate variables of the three-dimensional rectangular coordinate system, u and v represent the two coordinate variables of the image plane coordinate system, and l represents the ratio of the actual length to the pixel length in the rock exposure surface image;

[0099] After sorting, a fracture information set consisting of occurrence information and coordinate information is obtained. The fracture information set corresponding to the fractures on the inner wall of the borehole contains occurrence information and coordinate information, while the fracture information set corresponding to the fractures on the excavation exposed surface only contains coordinate information.

[0100] In this embodiment, the fracture coordinate set in the image plane coordinate system is converted into the fracture coordinate set exTC(1,2…a) of the excavation exposure surface and the fracture coordinate set bhTC(1,2…b) of the inner wall surface of the borehole in the three-dimensional rectangular coordinate system through the coordinate transformation method, and the attitude information in the cylindrical coordinate system is analyzed through the parameters of the function expression, and then the attitude information set TO(1,2…b) in the three-dimensional rectangular coordinate system is obtained through the coordinate transformation. Finally, the fracture coordinate set and the attitude information set of the fractures on the inner wall surface of the borehole are integrated to obtain the fracture information set exFI(1,2…a) of the excavation exposure surface and the fracture information set bhFI(1,2…b) of the inner wall surface of the borehole.

[0101] S6, inputting the crack information set obtained in S5 into the information summarization model, classifying and sorting all the crack information sets to obtain the structural surface information set.

[0102] The information summarization model in this embodiment is obtained by training a convolutional neural network. The model can summarize the crack information set and integrate the crack information sets belonging to the same structural surface into one structural surface information set. The specific training process is as follows:

[0103] Build a convolutional neural network model;

[0104] The crack information sets corresponding to the cracks on the excavation exposed surface and the inner wall of the borehole are marked and input into the convolutional neural network as training samples;

[0105] Continuously adjust model parameters through loss function and gradient descent algorithm;

[0106] Iterate repeatedly until the model converges, and finally establish an information induction model.

[0107] In this embodiment, the fracture information set is input into the trained neural network model, and the occurrence information and coordinate information are combined for judgment, the fracture information set is classified, and the fracture information sets cut out from the same structural surface are merged into one set, and finally the structural surface information set SI (1, 2…c) is obtained.

[0108] S7, performing interpolation calculation on the structural surface information set obtained in S6 according to the interpolation method to construct a three-dimensional model of the structural surface.

[0109] The interpolation method in this embodiment adopts the Kriging interpolation method, which uses the values ​​of all known points to obtain the values ​​of unknown points. The specific process is: first, all known structural surface coordinate information is calculated to obtain the horizontal distance d between the two coordinate points. ij And the semivariance r of the elevation value between two coordinate points ij , and obtain the array (d ij ,r ij ), and then obtain the functional relationship between distance and semi-variance by fitting the theoretical model. The fitting theoretical model adopts three models: spherical model, Gaussian model and exponential model, and finds the optimal functional relationship through the optimization algorithm. Then, the semi-variance of the elevation values ​​between all points is obtained through the optimal functional relationship, and then the weight coefficients corresponding to all coordinate points are obtained by solving the equation group of weight coefficients. Finally, all weight coefficients are substituted into the Kriging interpolation formula to obtain the elevation values ​​of all unknown points.

[0110] in,

[0111] The horizontal distance d between coordinate points ij :

[0112]

[0113] Among them, d ij represents the horizontal distance between the i-th known point and the j-th known point, (x i ,y i ) is the horizontal coordinate of the i-th known point, (x j ,y j ) is the horizontal azimuth coordinate of the jth known point.

[0114] Semivariance r of elevation values ​​between coordinate points ij :

[0115]

[0116] Among them, r ij represents the semivariance of the elevation value between the i-th known point and the j-th known point, z i and z i are the elevation values ​​of the i-th known point and the j-th known point respectively.

[0117] Spherical model:

[0118]

[0119] Gaussian model:

[0120]

[0121] Exponential Model:

[0122]

[0123] Among them, r represents semivariance, d represents horizontal azimuth distance, C0 represents nugget value, C represents partial sill value, and a represents range.

[0124] Solve the system of equations for the weight coefficients:

[0125]

[0126] Among them, r ij represents the semivariance of the elevation value between the i-th known point and the j-th known point, r it represents the semivariance of the elevation value between the i-th known point and the t-th unknown point, μ is the Lagrange multiplier, λ i Indicates the weight coefficient corresponding to the i-th known point when calculating the elevation value of the t-th unknown point.

[0127] Kriging interpolation formula:

[0128]

[0129] in, represents the estimated value of the elevation of the tth unknown point, Indicates the weight coefficient corresponding to the i-th known point, z i Represents the elevation value of the i-th known point.

[0130] In this embodiment, the known structural surface coordinate information is calculated by interpolation to obtain complete structural surface coordinate information, and the structural surface is three-dimensionally modeled by using these coordinates. The obtained structural surface three-dimensional model is as follows: Figure 4 shown.

[0131] S8, constructing a rock mass surface framework model, arranging the three-dimensional structural surface model obtained in S7 in the rock mass surface framework model according to the spatial position information, constructing a rock mass structure model, and obtaining rock mass structure information through the rock mass structure model.

[0132] The rock mass structure model is composed of a rock mass surface framework model and a structural surface three-dimensional model. When constructing the rock mass surface framework model, it is necessary to determine it according to the artificially defined rock mass research scope. In this embodiment, the rock mass surface framework model is first constructed according to the artificially determined research scope. The rock mass research scope is determined to be a cube with a length, width and height of 1m. Then, the constructed structural surface three-dimensional model is arranged according to the structural surface coordinate information. The final rock mass structure model is as follows: Figure 5 shown.

[0133] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for obtaining rock mass structure information based on rock mass surface and hole scanning, characterized in that: The steps include: S1, collects digital images of the original appearance of the exposed surface of the rock mass excavation at the engineering site; S2, drilling a hole in the rock mass at the construction site, collecting a digital image of the original appearance of the inner wall of the hole and recording the hole location information; S3, processing the original digital images collected in S1 and S2, strengthening and highlighting the cracks in the images as the focus, and obtaining a binary image containing only the crack graphics; S4, identifying and extracting crack patterns in the binary image, obtaining coordinates of all crack patterns in the binary image, and fitting the coordinates of the crack patterns in the binary image of the inner wall of the borehole to obtain a function expression of the crack patterns in the binary image of the inner wall of the borehole; In S4, the image plane coordinate system is established with the upper left corner endpoint of the image as the origin, the horizontal right direction as the positive direction of the u axis, the vertical downward direction as the positive direction of the v axis, and the pixel as the unit; the crack patterns in the image are numbered, and the coordinates of all crack patterns in the image plane coordinate system are obtained. Each crack pattern corresponds to a crack coordinate set; the coordinates of the crack patterns in the borehole inner wall image are fitted and calculated using the sine function model to obtain the function expression of all crack patterns in the borehole inner wall image; the sine function model is: ； Where A represents the amplitude of the crack graphic function in the borehole inner wall image, u and v represent the two coordinate variables of the image plane coordinate system, represents the initial phase of the crack graphic function in the borehole inner wall image, and c represents the upward translation distance of the crack graphic function in the borehole inner wall image; S5, analyzing and calculating the function expression and coordinates of the crack graph to obtain a set of crack information in the rock mass at the engineering site; S6, inputting the fracture information set into the information summarization model, classifying and sorting all fracture information sets to obtain the structural surface information set; S7, performing interpolation calculation on the structural surface information set according to an interpolation method to construct a three-dimensional model of the structural surface; S8, arranging the three-dimensional structural surface model in the rock mass surface frame model according to the spatial position information, constructing a rock mass structural model, and obtaining rock mass structural information through the rock mass structural model.

2. The method for obtaining rock mass structure information based on rock mass surface and hole scanning according to claim 1, characterized in that: In S1, shooting points are arranged at a preset distance in front of the exposed surface of the rock mass, and a digital camera is used to capture the original digital image of the exposed surface of the rock mass excavation. In S2, several horizontal boreholes are drilled on the exposed surface of the rock mass excavation and the position information of the boreholes is recorded. Then, the inside of the boreholes is repeatedly flushed with water to remove impurities. After the boreholes are dry, image acquisition is started at a preset distance in front of the exposed surface of the rock mass. The straight borehole TV probe is slowly placed into the borehole along the borehole wall, and the drill bit is slowly pushed in to the bottom of the borehole to obtain an image of the inner wall of the borehole.

3. The method for obtaining rock mass structure information based on rock mass surface and hole scanning according to claim 1, characterized in that: The image processing methods in S3 include scaling, grayscale, filtering, edge detection, thinning, and iterative connection.

4. The method for obtaining rock mass structure information based on rock mass surface and hole scanning according to claim 1, characterized in that: In S5, a cylindrical coordinate system is established with the center of the borehole as the origin, vertical downward as the polar axis direction, counterclockwise as the positive angle direction, and horizontal inside the borehole as the positive z-axis direction. The inclination angle of the structural surface in the cylindrical coordinate system is obtained by the amplitude A of the crack graphic function in the borehole inner wall image. , through the initial phase of the function Obtain the inclination of the structural surface in the cylindrical coordinate system , we get the occurrence information set of the structural surface in cylindrical coordinates. Each crack pattern corresponds to one occurrence information set. The calculation formula of the structural surface inclination angle in the cylindrical coordinate system is: ； in, represents the inclination angle of the structural surface, A represents the amplitude of the crack graphic function in the borehole inner wall image, and M represents the size width of the borehole inner wall image; The calculation formula of the structural surface inclination in the cylindrical coordinate system is: ； in, Indicates the angle between the inclination of the structural surface and the section direction of the borehole inner wall image, Represents the initial phase of the crack graphic function in the borehole inner wall image; The fracture coordinate set in the borehole inner wall image plane coordinate system is converted into the fracture coordinate set in the cylindrical coordinate system through coordinate transformation. The formula for converting the image plane coordinate system into the cylindrical coordinate system is: ； in, 、 、 They represent the three coordinate variables of the cylindrical coordinate system, u and v represent the two coordinate variables of the image plane coordinate system, M represents the size width of the borehole inner wall image, r represents the actual borehole radius, Indicates the ratio of the actual length to the pixel length in the borehole inner wall image; A three-dimensional rectangular coordinate system is established with the upper left end point of the shooting range of the rock mass excavation exposure surface as the origin, the horizontal right direction as the positive direction of the x-axis, the vertical downward direction as the positive direction of the y-axis, and the horizontal direction inside the rock mass as the positive direction of the z-axis. The unit is meter. The fracture coordinate set and the occurrence information set in the cylindrical coordinate system are converted into the fracture coordinate set and the occurrence information set in the three-dimensional rectangular coordinate system through coordinate transformation. The formula for converting the cylindrical coordinate system to the three-dimensional rectangular coordinate system is: ； Among them, x, y, and z represent the three coordinate variables of the three-dimensional rectangular coordinate system. 、 、 are the three coordinate variables of the cylindrical coordinate system, 、 are all constants; The fracture coordinate set in the exposure surface image plane coordinate system is converted into the fracture coordinate set in the three-dimensional rectangular coordinate system through coordinate transformation. The formula for converting the image plane coordinate system into the three-dimensional rectangular coordinate system is: ； Among them, x, y, and z represent the three coordinate variables of the three-dimensional rectangular coordinate system, and u and v represent the two coordinate variables of the image plane coordinate system. Indicates the ratio of the actual length to the pixel length in the rock exposure surface image; After sorting, a fracture information set consisting of occurrence information and coordinate information is obtained. The fracture information set corresponding to the fractures on the inner wall of the borehole contains occurrence information and coordinate information, while the fracture information set corresponding to the fractures on the excavation exposed surface only contains coordinate information.

5. The method for obtaining rock mass structure information based on rock mass surface and hole scanning according to claim 4, characterized in that: The information induction model in S6 is obtained by training the convolutional neural network to summarize the crack information set and integrate the crack information sets belonging to the same structural surface into one structural surface information set. The specific training process is as follows: Construct a convolutional neural network model; mark the crack information sets corresponding to the cracks on the excavation exposed surface and the inner wall of the borehole, and input them into the convolutional neural network as training samples; continuously adjust the model parameters through the loss function and gradient descent algorithm; repeatedly iterate until the model converges, and establish an information induction model.

6. The method for obtaining rock mass structure information based on rock mass surface and hole scanning according to claim 5, characterized in that: S7 uses the Kriging interpolation method to obtain the value of the unknown point using the values ​​of all known points. The specific process is: calculate all known structural surface coordinate information to obtain the horizontal distance d between the two coordinate points ij And the semivariance r of the elevation value between two coordinate points ij , and obtain the array (d ij ,r ij ); obtain the functional relationship between distance and semi-variance by fitting the theoretical model. The fitting theoretical model adopts three models: spherical model, Gaussian model and exponential model, and find the optimal functional relationship through optimization algorithm; obtain the semi-variance of the elevation values ​​between all two points through the optimal functional relationship; obtain the weight coefficients corresponding to all coordinate points by solving the equation group of weight coefficients; substitute all weight coefficients into the Kriging interpolation formula to obtain the elevation values ​​of all unknown points.

7. The method for obtaining rock mass structure information based on rock mass surface and hole scanning according to claim 6, characterized in that: The horizontal distance d between coordinate points ij : ； in, represents the horizontal azimuth distance between the i-th known point and the j-th known point, is the horizontal azimuth coordinate of the i-th known point, is the horizontal azimuth coordinate of the jth known point; Semivariance r of elevation values ​​between coordinate points ij : ； in, represents the semivariance of the elevation value between the i-th known point and the j-th known point, and are the elevation values ​​of the i-th known point and the j-th known point respectively; Spherical model: ； Gaussian model: ； Exponential Model: ； Where r represents semivariance, d represents horizontal azimuth distance, C0 represents nugget value, C represents partial sill value, and a represents range; Solve the system of equations for the weight coefficients: ； in, represents the semivariance of the elevation value between the i-th known point and the j-th known point, represents the semivariance of the elevation value between the i-th known point and the t-th unknown point, is the Lagrange multiplier, Indicates the weight coefficient corresponding to the i-th known point when obtaining the elevation value of the t-th unknown point; Kriging interpolation formula: ； in, represents the estimated value of the elevation of the tth unknown point, Represents the weight coefficient corresponding to the i-th known point, Represents the elevation value of the i-th known point.

8. The method for obtaining rock mass structure information based on rock mass surface and hole scanning according to claim 7, characterized in that: The rock mass structure model in S8 is composed of a rock mass surface framework model and a structural surface three-dimensional model. When constructing the rock mass surface framework model, it is necessary to determine it according to the artificially defined rock mass research scope.

Images