Automatic fracture parameter acquisition method based on digital twinborn model

Abstract

The invention relates to an automatic fracture parameter acquisition method based on a digital twinborn model, and belongs to the technical field of intelligent statistics of rock mass parameters. According to the technical scheme, a static digital twinborn model of a rock mass is established; projecting the panoramic image into a plane image for the crack area; identifying a crack on the plane image, and marking a crack point; and establishing a corresponding relationship between the plane image and the point cloud, and converting the crack point into the corresponding point cloud. Connecting fracture points on the digital twinborn model to carry out fracture visualization; and based on the three-dimensional point cloud of the fracture, calculating the trend, inclination angle and the like of the fracture, the grouping of the fracture, the length of the fracture and other occurrences. The method has the beneficial effects that the spatial distribution condition and occurrence information of rock mass fractures can be quickly obtained, and manual work can be replaced to complete geological sketch or geological record; the method has the beneficial effects of high precision, high flexibility, high speed and the like.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent statistical analysis of rock mass parameters, and relates to a method for automatically acquiring fracture parameters, and in particular to a method for automatically acquiring fracture parameters based on a digital twin model. Background Art

[0002] Rock mass cracks are of great significance in rock engineering, and the rock mass crack joints are important indicators for evaluating rock mass quality. Joints and cracks may lead to rock mass destruction, deformation, and sliding, which will have a negative impact on the stability of rock structures and the safety of rock engineering. Therefore, accurate identification and extraction of cracks in rock mass is of great significance for the design, construction, and evaluation of rock engineering.

[0003] The traditional method of identifying rock cracks is mainly direct visual method. First, the surface of the geological body is manually visually inspected to find cracks, then the rock mass is recorded by sketching, and finally the on-site experts evaluate the quality of the surrounding rock based on their experience. Therefore, more and more engineers and scientific researchers use machine vision and image recognition to detect cracks and joints in the surrounding rock. This method can quickly and accurately identify and measure cracks through computer automated processing, with high efficiency and accuracy, saving a lot of manpower and material resources, and improving the safety of workers.

[0004] However, the existing recognition methods based on machine vision have certain problems. It is difficult to obtain the parameter information of the cracks identified in the image in the real world. Since the image does not have size information, this method determines that the acquisition of crack parameters is limited to obtaining certain quantitative information. It is powerless to obtain important crack development parameters such as crack length, crack group spacing, crack inclination, and direction.

[0005] Therefore, a method is needed to identify cracks in rock mass and simultaneously obtain crack development parameters such as their occurrence. Summary of the invention

[0006] In order to solve the above-mentioned technical problems existing in the prior art, the present invention provides a method for automatically acquiring fracture parameters based on a digital twin model, which can obtain rock fracture parameters with high accuracy, high flexibility and high speed.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for automatically acquiring fracture parameters based on a digital twin model, the specific steps are as follows:

[0009] S1. Establish a static digital twin model of the rock mass;

[0010] S2, for the crack area, project the panoramic image into a plane image;

[0011] S3, identifying cracks on the plane image and marking the crack points;

[0012] S4. Establish the correspondence between the plane image and the point cloud, convert the crack points into corresponding point clouds; connect the crack points on the digital twin model to visualize the cracks;

[0013] S5. Based on the three-dimensional point cloud of the fracture, calculate the fracture strike, inclination, fracture grouping, fracture length and other characteristics.

[0014] Furthermore, the specific steps of step S1 are: use a three-dimensional laser scanner (TLS) equipped with a synchronous camera to scan the rock mass, and simultaneously obtain rock mass images and three-dimensional point cloud data; pre-process the point cloud data and image data respectively, wherein the point cloud data needs to be denoised, and the image data is seamlessly fused and stored as a panoramic image. The point cloud data and image data are fused to establish a virtual digital model of the rock mass with high precision and high fidelity.

[0015] Furthermore, the specific steps of step S2 are to establish the connection between the coordinates of the plan view and the panoramic view and the corresponding coordinate system, convert the panoramic view into the plan view, and perform crack recognition on the plan view.

[0016] Furthermore, the specific steps of step S3 are to obtain the crack points in the plane view by using the confidence method, connect the crack points on the same crack one by one to obtain a relatively complete crack, and mark the crack points.

[0017] Furthermore, the specific steps of step S4 are to establish a corresponding relationship between the plane image and the point cloud, convert the crack points into corresponding point cloud coordinates; connect the crack points on the digital twin model and draw them on the three-dimensional model to visualize the cracks.

[0018] Furthermore, the specific steps of step S5 are to use the coordinates of the three-dimensional point cloud corresponding to the crack point obtained from S4 to calculate the length of the crack, fit the same crack as a spatial straight line, and obtain the parameters of the spatial straight line; calculate the direction and tendency of the spatial straight line and statistically analyze the grouping information between the straight lines to obtain complete crack parameters.

[0019] Beneficial effects of the present invention:

[0020] Compared with the prior art, the method for automatically acquiring fracture parameters based on the digital twin model of the present invention has the following technical characteristics or beneficial effects:

[0021] (1) High precision and authenticity: By combining a 3D laser scanner (TLS) and a synchronous camera, the present invention can simultaneously obtain images and 3D point cloud data of the rock mass, thereby establishing a high-precision and high-fidelity virtual digital model of the rock mass. This model not only reflects the geometric shape of the rock mass, but also retains its real appearance characteristics, providing a solid foundation for the identification of cracks and parameter calculation.

[0022] (2) Comprehensive acquisition of crack parameters: Traditional machine vision methods are limited by the lack of image size information and cannot accurately obtain a variety of important crack parameters. The present invention uses a digital twin model to convert crack points from a plane image into a three-dimensional point cloud, so that key parameters such as the length, direction, inclination, and grouping of the crack can be calculated. This not only improves the accuracy of crack identification, but also provides more comprehensive data support for the design, construction, and evaluation of rock mass engineering.

[0023] (3) Efficiency and automation: The crack parameter acquisition method of the present invention realizes an automated process from image recognition to parameter calculation. The crack points are quickly identified through machine vision technology, and three-dimensional visualization is performed using a digital twin model, which greatly reduces the time for manual intervention and data processing. This not only improves work efficiency, but also reduces the risk of human error.

[0024] (4) Flexibility and applicability: The method of the present invention is not only applicable to different types of rock fracture identification, but can also be customized according to actual needs. For example, the accuracy of fracture identification and the level of detail of parameter calculation can be adjusted according to engineering needs to meet application requirements in different scenarios.

[0025] (5) Improved safety: Compared with the traditional manual visual method, the present invention utilizes machine vision and image recognition technology to avoid direct contact of workers in dangerous environments, thereby improving the safety of rock engineering operations.

[0026] The present invention provides a high-precision, high-efficiency, comprehensive and flexible method for acquiring rock fracture parameters, which not only solves the problem of incomplete parameter acquisition in traditional methods, but also improves the safety and reliability of engineering operations. It has broad application prospects and important practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the present invention will be described in detail below in combination with the accompanying drawings and detailed embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0028] Figure 1 It is the technical roadmap of the present invention;

[0029] Figure 2 A coordinate system diagram for converting a plan view and a panoramic view established by the present invention;

[0030] Figure 3 This is a visualization result diagram of rock mass fracture space in Example 1;

[0031] Figure 4 The trend rose diagram and the dip histogram of the fracture in Example 1 are shown in FIG. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Figure 1-4 The automatic acquisition method of fracture parameters based on the digital twin model is further explained, and the technical scheme in the embodiment of the present application is clearly and completely described; obviously, the described embodiment is only a part of the embodiments of the present application, not all the embodiments, and based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0033] The present invention provides a method for automatically acquiring fracture parameters based on a digital twin model. The technical route is as follows: Figure 1 As shown, the specific steps include:

[0034] S1, establish a static digital twin model of the rock mass.

[0035] S2, for the crack area, the panoramic image is projected into a plane image.

[0036] S3, identifying cracks on the plane image and marking the crack points.

[0037] S4, establish the correspondence between the plane image and the point cloud, convert the crack points into the corresponding point cloud, and connect the crack points on the digital twin model to visualize the cracks.

[0038] S5, based on the three-dimensional point cloud of the fracture, calculate the fracture strike, inclination, fracture grouping, fracture length and other occurrences.

[0039] Example 1

[0040] S1, use a 3D laser scanner (TLS) equipped with a synchronous camera to scan the rock mass, and obtain rock mass images and 3D point cloud data at the same time. The obtained point cloud data is used to remove noise and irrelevant points using data dimensionality reduction combined with image processing. The denoised point cloud is used to reconstruct the 3D surface and obtain the 3D geometric grid. The image data is fused and saved as a panorama, and the correspondence between the point cloud and the pixel is established. For each 3D grid, a matching image patch is calculated. The coordinates and grid and mapping information are stored in the data exchange format of the 3D model, and visualized using the corresponding open source software.

[0041] S2, before crack identification, the panoramic image needs to be converted into a local plane view, and crack identification is performed on the plane view. The conversion of the panoramic image into a plane view requires the establishment of a connection between the plane view and the panoramic view coordinates, and the establishment of the conversion coordinate system is as follows: Figure 2 As shown, the XY plane is parallel to the view plane, and the Z axis is perpendicular to the view plane and points outward. Similarly, a two-dimensional coordinate system is established for the plane view, with the origin at the center of the view and the coordinate range being x∈ Among them, w and h represent the width and height of the plane view respectively. The direction vector of a point (X, Y, Z) on the plane view is:

[0042]

[0043] Where f represents the focal length of the perspective model, which is calculated as:

[0044]

[0045] What is the meaning of FOV? Figure 2 As shown, it represents the horizontal range angle of the plane view. Normalize the direction vector of this point for easy calculation:

[0046]

[0047] There are three parameters representing a certain viewing angle (FOV, yaw, pitch), which represent the field of view angle, Y-axis yaw angle and X-axis yaw angle respectively. The rotation matrix rotating around these two axes is

[0048]

[0049] So the total rotation matrix is:

[0050] R=R pitch (φ 0 )·R yaw (θ 0 ) (6)

[0051] The direction vector after rotation is:

[0052] v′=R·v (7)

[0053] Convert the direction vector to spherical coordinates:

[0054]

[0055] Here we get the correspondence between the spherical coordinates and the coordinates of the pixel points on the plane view.

[0056] S3, first calculate the eigenvalues ​​of the Hessian matrix of the crack plane graph, give a score to the points whose eigenvalues ​​conform to the crack, and comprehensively evaluate the grayscale value and grayscale symmetry of these points to give these candidate points a crack confidence score. Filter out the points with low confidence scores, and the remaining ones are called crack points. Connect the candidate points on the same crack one by one to obtain a more complete crack, and mark the crack points. The detailed operation method can be found in patent CN116883373A and will not be repeated here.

[0057] S4, the corresponding calculation formula between point cloud and pixel is:

[0058]

[0059] where x ip ,y ip Represents the coordinates of the pixel corresponding to the i-th point in the point cloud, x i ,y i ,z i represents the world coordinates of the point, W and H represent the resolution width and height of the panorama respectively. One is subtracted at the end of the formula because the pixel coordinates in the computer start from 0.

[0060] Through formula (9), we can get the corresponding relationship between the panoramic image and the image pixels:

[0061]

[0062] The corresponding relationship between the pixels in the point cloud plane view can be obtained by formulas (1)-(10). The crack points identified by S3 are marked on the digital twin model established by S1, and the crack points of the same crack are connected to visualize the cracks on the rock mass model. The crack visualization result in this embodiment is as follows: Figure 3 shown.

[0063] S5, using the coordinates of the three-dimensional point cloud corresponding to the crack point obtained from S4, calculates the length of the crack and fits the same crack as a spatial straight line. The specific method is:

[0064] Assume that the spatial coordinates of the crack point are (x, y, z), and the spatial coordinates of the crack point cloud are an N×3 matrix. First, calculate the centroid of the crack point cloud:

[0065]

[0066] For each point (x i ,y i ,z i ), centralized by:

[0067] P i =(x i -centroid x ,y i -centroid y ,z i -centroid z ) (12)

[0068] Compute the covariance matrix of the point cloud using the centering method:

[0069]

[0070] Perform eigenvalue decomposition on the covariance matrix C to obtain eigenvalues ​​and eigenvectors. The eigenvector corresponding to the eigenvalue can represent the main direction of the point cloud. The eigenvector with the largest eigenvalue is selected as the direction of the fitted line. The parametric equation of the line can be expressed as:

[0071] r(t)=centroid+t·d (14)

[0072] where d is the eigenvector of the principal direction and t is the parameter.

[0073] Calculation of fracture occurrence: Fit the three-dimensional coordinates of the fracture to obtain the fitted spatial straight line. The calculation method of fracture inclination and dip angle is:

[0074] Dip dir = arctan(Ax / Ay) (15)

[0075]

[0076] Among them, Ax, Ay, and Az represent the unit direction vectors of the space line respectively. The components on the three coordinate axes of X, Y, and Z are calculated. The inclination and dip of each crack are counted to automatically calculate the occurrence of the crack. The trend rose diagram and dip of the crack parameters in this embodiment are as follows: Figure 4 shown.

[0077] For a crack that has been fitted as a straight line, the direction vector angle θ can be calculated using the following formula:

[0078]

[0079] By clustering method, the cracks with θ<ε are divided into a group of cracks, and the number of groups of crack inclination angles is counted.

[0080] The formula for calculating the length of the crack is:

[0081] In the i-th crack, let the distance between its two nodes be L ij By calculating:

[0082]

[0083] The length of each crack can be obtained through this formula.

[0084] The foregoing description only discloses and describes exemplary embodiments of the present invention. From the above description and the accompanying drawings and claims, those skilled in the art will easily appreciate that various changes, modifications and variations may be made to the present invention without departing from the spirit and scope of the present invention as defined by the appended claims.

[0085] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for automatically acquiring crack parameters based on a digital twin model, characterized in that: Here are the steps: S1. Establish a static digital twin model of the rock mass; S2, for the crack area, project the panoramic image into a plane image; S3, identifying cracks on the plane image and marking the crack points; S4, establishing a corresponding relationship between the plane image and the point cloud, and converting the crack points into corresponding point clouds; Connect fracture points on the digital twin model for fracture visualization; S5. Based on the three-dimensional point cloud of the crack, calculate the direction, inclination, grouping and length of the crack.

2. The method for automatically acquiring crack parameters based on a digital twin model according to claim 1, characterized in that: The specific steps of step S1 are: use a three-dimensional laser scanner equipped with a synchronous camera to scan the rock mass, and simultaneously obtain a rock mass image and three-dimensional point cloud data; pre-process the point cloud data and the image data respectively, wherein the point cloud data is denoised, and the image data is seamlessly fused and stored as a panoramic image; the point cloud data and the image data are fused to establish a virtual digital model of the rock mass with high precision and high fidelity.

3. The method for automatically acquiring crack parameters based on a digital twin model according to claim 1, characterized in that: The specific steps of step S2 are: establishing the connection between the coordinates of the plane view and the panoramic view and the corresponding coordinate system, converting the panoramic view into the plane view, and performing crack recognition on the plane view.

4. The method for automatically acquiring crack parameters based on a digital twin model according to claim 1, characterized in that: The specific steps of step S3 are: obtaining the crack points in the plane view by using the confidence method, connecting the crack points on the same crack one by one to obtain a complete crack, and marking the crack points.

5. The method for automatically acquiring crack parameters based on a digital twin model according to claim 1, characterized in that: The specific steps of step S4 are: establishing a corresponding relationship between the plane image and the point cloud, converting the crack points into corresponding point cloud coordinates; connecting the crack points on the digital twin model and drawing them on the three-dimensional model to visualize the cracks.

6. The method for automatically acquiring crack parameters based on a digital twin model according to claim 1, characterized in that: The specific steps of step S5 are: using the coordinates of the three-dimensional point cloud corresponding to the crack point obtained from S4, calculating the length of the crack, fitting the same crack as a spatial straight line, and obtaining the parameters of the spatial straight line; The direction and inclination of the spatial straight line are calculated, and the grouping information between the statistical straight lines is obtained to obtain the complete fracture parameters.

Images