How to create a 3D geological model of a tunnel

Abstract

To provide a method for creating a three-dimensional geological model of a tunnel, which enables intuitive understanding of an actual geological structure by connecting geological images of an inner peripheral surface appearing after excavation of the tunnel.SOLUTION: The present invention is a method for creating a three-dimensional geological model of a tunnel whose excavated surface and inner peripheral surface are covered with shot concrete. It comprises steps (S1-S3) of repeating the work of acquiring image data of parallax image and RGB image by photographing the excavated surface including the shot part of the shot concrete after excavation for each unit excavation process, a step S4 of creating a unit three-dimensional model from the image data for each unit excavation process, a step S5 of creating an initial synthetic model by superimposing overlapping portions of unit three-dimensional models of successive unit excavation processes, and a step S6 of creating a three-dimensional geological model by removing the shot part from the initial synthetic model.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method for creating a three-dimensional geological model of a tunnel whose excavation surface and inner circumferential surface are covered with shotcrete. [Background technology]

[0002] Observing the tunnel face immediately after excavation to understand the geological structure around the tunnel is essential for managing the schedule, safety, and quality of mountain tunnel construction. For this reason, construction staff observe the tunnel face (the front of the tunnel at the very tip of the excavation) after each excavation and record geological information about the face (rock type, cracks, hardness, etc.).

[0003] Here, there is a risk of rockfall and collapse directly below the tunnel face, so after excavation, sprayed concrete is applied quickly to ensure safety directly below the face before moving on to the next step.As a result, there is only a short time after excavation when it is possible to directly observe the geological information around the tunnel.

[0004] Currently, photographs of the tunnel face are recorded after each excavation, but since the geological information is collected for each excavation (approximately 1 m), the information is intermittent and it is difficult to understand the geological structure of the entire tunnel continuously at a glance.

[0005] On the other hand, as disclosed in Patent Documents 1 to 3, various methods for measuring the exact shape of the tunnel face and devices for photographing the excavation surface, such as the side wall surface of a tunnel, have been developed. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-200521 [Patent Document 2] Japanese Patent Application Laid-Open No. 2018-109311 [Patent Document 3] Japanese Patent Application Laid-Open No. 2018-204288 Summary of the Invention [Problem to be solved by the invention]

[0007] However, as mentioned above, most of the tunnel's inner surface is quickly covered with shotcrete after excavation to ensure safety, so the excavation surface on which geological information can be observed at one time is limited to a small portion near the tunnel face. It is difficult for inexperienced engineers to correctly understand sudden geological changes such as faults simply by collecting these partially taken photographs.

[0008] Therefore, the present invention aims to provide a method for creating a three-dimensional geological model of a tunnel that allows users to intuitively understand the actual geological structure by connecting together images of the geological features of the inner surface that appear after the tunnel is excavated. [Means for solving the problem]

[0009] In order to achieve the above-mentioned object, the method for creating a three-dimensional geological model of a tunnel of the present invention is a method for creating a three-dimensional geological model of a tunnel whose excavation surface and inner surface are covered with sprayed concrete, and is characterized by comprising the steps of: repeating, for each unit excavation process, the work of obtaining image data of parallax images and RGB images by photographing the excavation surface including the sprayed portion of the sprayed concrete after excavation; creating a unit three-dimensional model from the image data for each unit excavation process; creating an initial composite model by superimposing overlapping portions of the unit three-dimensional models of successive unit excavation processes; and creating a three-dimensional geological model by removing the sprayed portion from the initial composite model.

[0010] Here, the work of acquiring the image data is carried out using a photographing device that is self-propelled to the vicinity of the tunnel face immediately after excavation, and the photographing device can be configured to photograph the inner surface of the tunnel in a circumferential direction.

[0011] The parallax images may be acquired by a stereo camera, and the RGB images may be acquired by a high-resolution camera. The unit 3D model may be created by superimposing point cloud data generated from the parallax images on the RGB images, and the superimposition of overlapping portions when creating the initial composite model may be performed based on the point cloud data. [Effects of the Invention]

[0012] In the method for creating a 3D geological model of a tunnel configured as described above, the excavation surface including the sprayed sections is photographed to obtain image data of parallax images and RGB images.Then, the 3D geological model is created by removing the sprayed sections from an initial composite model created by superimposing unit 3D models created from the image data.

[0013] This allows the geological features of the inner surface that appear after tunnel excavation to be seamlessly linked together, allowing even inexperienced engineers to intuitively understand the actual geological structure. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a flowchart illustrating the process flow of a method for creating a three-dimensional geological model of a tunnel according to the present embodiment. [Figure 2] FIG. 1 is an explanatory diagram showing an overview of a photographing situation using a self-propelled photographing device. [Figure 3] 1A and 1B are diagrams illustrating a camera unit of a self-propelled photographing device, in which (a) is an external view and (b) is an explanatory view of the inside with the protective case removed. [Figure 4] FIG. 1 is an explanatory diagram illustrating an outline of the principle of a stereo camera. [Figure 5] FIG. 2 is an explanatory diagram illustrating a parallax image obtained by photographing with a stereo camera. [Figure 6] FIG. 10 is an explanatory diagram illustrating processing of point cloud data. [Figure 7]FIG. 10 is an explanatory diagram showing an outline of a process for creating a unit 3D model from a parallax image and an RGB image. [Figure 8] FIG. 10 is an explanatory diagram showing an outline of the process of superimposing unit 3D models that are continuous in the excavation direction. [Figure 9] FIG. 1 is an explanatory diagram illustrating a three-dimensional geological model. DETAILED DESCRIPTION OF THE INVENTION

[0015] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Fig. 1 is a flowchart illustrating the process flow of a method for creating a 3D geological model of a tunnel according to this embodiment. Fig. 2 is an explanatory diagram showing an overview of the photographing situation by a self-propelled photographing device 1, which serves as a photographing device.

[0016] In the construction of a tunnel T such as a mountain tunnel, the ground is excavated using an excavator, and the exposed excavated surface is quickly covered with shotcrete. Figure 2 shows the state in which the excavator (not shown) has retreated toward the tunnel entrance, and shotcrete has been applied to form the shotcrete section F1.

[0017] Tunnel T is excavated by digging face T1, the most extreme part of the excavation, which is shown as a semicircle in Figure 2. Behind face T1 (towards the tunnel entrance), the inner circumferential surface of tunnel T, which is formed in a semi-cylindrical shape, appears.

[0018] Immediately after excavation, the inner surface of this tunnel T is exposed as the excavated surface T2, but because there is a risk of rockfall or collapse if left as is, sprayed concrete is gradually applied to replace the excavated section F1.

[0019] For example, in a construction project where the unit excavation step is approximately 1 m, the excavation peripheral surface T2 is extended by 1 m every time the face T1 is excavated by 1 m, and a new spraying section F1 is installed on the inner peripheral surface for 1 m behind it. Note that the spraying section F1 can also be installed for multiple steps at once.

[0020] In the method for creating a three-dimensional geological model of a tunnel in this embodiment, the unlined excavated surface T2 and the sprayed section F1 are photographed together, so when photographing is done after excavation, the unlined excavated surface T2 is always exposed on the inner surface of the tunnel T between the face T1 and the sprayed section F1.

[0021] The excavation surface can be photographed using, for example, a self-propelled camera 1 as shown in Figure 2. This self-propelled camera 1 is a device for safely photographing the inner surface of the tunnel T near the face T1 by remote control.

[0022] The self-propelled photographing device 1 comprises a running unit 11 that serves as a traveling means, an arm unit 12 that extends upward from above the running unit 11, and a camera unit 2 that is attached to the tip of the arm unit 12. The running unit 11 is preferably provided as a crawler type so that it can travel on uneven ground T3.

[0023] Although not shown, the self-propelled photographing device 1 also includes a receiving unit that receives operation signals, a control unit that operates a drive unit in response to operation signals, an arithmetic processing unit and memory unit for photographing and recording using the camera unit 2, and a battery. Furthermore, an inertial measurement unit (IMU) that measures angular velocity and acceleration with high precision can also be attached to control autonomous driving.

[0024] Figure 3 is a diagram illustrating the camera unit 2 of the self-propelled photographing device 1. Figure 3(a) shows an external view, and Figure 3(b) shows the inside of the camera unit 2. The camera unit 2 is covered by a protective case 21 in the shape of a rectangular parallelepiped.

[0025] When protective case 21 is removed, as shown in Fig. 3(b), stereo camera 22, high-resolution camera 23, and line laser 24 are incorporated. Protective case 21 has windows made of transparent plates or the like for protecting stereo camera 22, high-resolution camera 23, and line laser 24.

[0026] 4 is an explanatory diagram showing an outline of the principle of the stereo camera 22. The stereo camera 22 has a pair of built-in cameras separated by an inter-camera distance B. The left camera captures a left image, and the right camera captures a right image.

[0027] When the object P is photographed by the stereo camera 22, the coordinate positions (P L (x L ,y L ), P R (x R ,y R )), the object P is photographed. Then, the lens distance between the left and right cameras (the distance between the cameras B), the focal length f of the left and right cameras, and the coordinate position (P L (x L ,y L ), P R (x R ,y R )), the three-dimensional coordinates (X, Y, Z) of the object P can be calculated. In a situation where it is difficult to photograph the object P, the object P can be visualized by irradiating it with a laser output from the line laser 24.

[0028] 5 is an explanatory diagram illustrating a parallax image obtained by photographing with the stereo camera 22. In short, a parallax image showing the distance from the stereo camera 22 can be created from the left and right images photographed by the stereo camera 22.

[0029] For example, if there are irregularities on the inner surface of the tunnel T, such as the excavation surface, by extracting only the range where the distance from the stereo camera 22 is shorter than a predetermined value, only the convex parts of the inner surface can be output as point cloud data, as described below.

[0030] Next, the process flow of the method for creating a three-dimensional geological model of a tunnel according to this embodiment will be described with reference to the flowchart shown in FIG. First, at the construction site of tunnel T, excavation work is carried out for each unit excavation process (step S1). For example, after 1 m of excavation is carried out using blasting or an excavator, the excavated soil is removed, and the inner surface of tunnel T exposed in the previous unit excavation process is covered with shotcrete.

[0031] Next, the self-propelled camera 1 is driven toward the tunnel face T1 immediately after excavation. At this stage, the excavator, excavated soil, and spraying vehicle have retreated toward the tunnel entrance, and there are no obstacles to photography near the tunnel face T1.

[0032] The self-propelled camera 1 is remotely controlled to move to a position suitable for taking pictures near the working face T1. If the self-propelled camera 1 is under automatic driving control, it will move by itself on the ground T3 to the position of the set coordinates.

[0033] For example, as shown in Figure 2, the self-propelled camera device 1 is stopped on the axis of the tunnel T in front of the unexcavated peripheral surface T2, which is the inner peripheral surface of the tunnel T newly exposed in the current process. The self-propelled camera device 1 can take pictures by operating the camera unit 2 at the tip of the arm part 12 in the circumferential direction R.

[0034] Therefore, continuous photographs are taken by pointing the lens of the camera unit 2 toward the leg of the arch-shaped excavated peripheral surface T2, and then rotating the camera unit 2 in the circumferential direction R from there to the top of the tunnel T and the leg on the opposite side (step S2).

[0035] The photographing area 20 is set so that it includes the newly exposed 1m of unlined excavated surface T2 in the direction of tunnel T excavation, and the 2m of sprayed section F1 adjacent to and behind it. Note that the photographing area 20 may include areas other than the aforementioned 3m width, as well as the ground surface T3 and the working face T1. Unnecessary images can be removed in a process described below.

[0036] When photographing with the camera unit 2, photographing is performed simultaneously with the stereo camera 22 and the high-resolution camera 23. When photographing a location with few features, such as a flat surface, photographing can also be performed while irradiating the area with a line laser 24 as auxiliary light.

[0037] When taking a photo, the date and time of the photo are also recorded along with the image. Furthermore, if the self-propelled camera 1 has position information, the position information is also recorded along with the image. For example, when the self-propelled camera 1 equipped with an inertial measurement unit (IMU) is driven, the angular velocity and acceleration detected during driving are integrated, and position information of the position where the device stopped after driving from a position of known coordinates can be obtained.

[0038] This is the unit excavation process and one step of photography, after which tunnel construction resumes (step S31). Then, in step S32, it is determined whether or not photography should be continued, and if photography is to be continued, the process proceeds to step S1 in accordance with "YES". On the other hand, if photography is to be stopped once and the process proceeds to the creation of a 3D geological model, the process proceeds to step S4 (NO in step S32).

[0039] As described above, image data of a parallax image can be obtained from the image captured by the stereo camera 22. Also, image data of an RGB image can be obtained from the image captured by the high-resolution camera 23.

[0040] 6 is an explanatory diagram illustrating the processing of point cloud data obtained from parallax images. Since the parallax images provide data on the distance from the stereo camera 22, point cloud data can be generated by extracting convex portions that are closer to the stereo camera 22 than a threshold value as points.

[0041] The upper image in Figure 6 shows an example of point cloud data generated from a parallax image containing a 1-m section of the excavation surface T2 and a 2-m section of the sprayed section F1. The most important data for creating a 3D geological model is the image data of the excavation surface T2.

[0042] On the other hand, point cloud data of the ground T3, the unexcavated face T1, and other unnecessary locations that will not be used for overlay can be removed at this stage. The point cloud data and RGB image data obtained in this way are expanded onto a plane for the subsequent overlay process.

[0043] FIG. 7 is an explanatory diagram showing an outline of the process (step S4) of creating a unit 3D model from the parallax image and the RGB image. The point cloud data obtained from the parallax image and the image data of the RGB image are extracted from a 3m wide range (the process range for one shooting) at the same position.

[0044] Therefore, a unit 3D model is created by overlaying the point cloud data developed on a plane with the image data of the RGB image and returning it to the three-dimensional shape of the inner surface of the tunnel T. This unit 3D model includes 1 m of the unlined excavated surface T2 (rock) and 2 m of the sprayed portion F1 (concrete).

[0045] Such a unit 3D model is created for each photographing session. Then, in step S5, a plurality of unit 3D models that are continuous in the excavation direction of the tunnel T are superimposed to create a 3D initial composite model.

[0046] Figure 8 is an explanatory diagram showing an overview of the process of superimposing three consecutive unit 3D models in the excavation direction. Here, the unlined excavation surface T2 present in each unit 3D model is hidden by the shotcrete in the next photograph, so it cannot be used as the basis for superimposition.

[0047] Therefore, the sprayed area F1 is used for overlaying. That is, the 1 m sprayed area F1 adjacent to the unpainted excavated surface T2 photographed the first time is also present in the unit 3D model created by photographing the second time.

[0048] Therefore, the first and second unit 3D models, which model the same area, are connected by overlapping the overlapping parts of the first and second unit 3D models.

[0049] When overlaying these overlapping parts, the point cloud data of the sprayed area F1 is used. That is, point cloud data of the second unit 3D model that matches the characteristics of the point cloud data of the first unit 3D model is extracted, and these are identified as the same location and overlay processing is performed.

[0050] The same location can be identified not only by the convex part of the sprayed part F1, but also by the position of the support and rock bolts. Image segmentation technology can also be applied to identify the excavation surface T2 and the sprayed part F1.

[0051] Then, when superimposing the unit 3D model for the third time, the same process as the first and second superimpositions is performed. That is, the third unit 3D model is superimposed on the 3D model that is a composite of the first and second models.

[0052] The initial composite model created in this way is a three-dimensional model that has three rounds (3 m) of uncut excavation surface T2 and the third round of 1 m of sprayed section F1. Therefore, in step S6, the 1 m of sprayed section F1 is removed from the initial composite model to create a three-dimensional geological model (see the bottom diagram in Figure 8).

[0053] 9 is an explanatory diagram illustrating a 3D geological model created by the method for creating a 3D geological model of a tunnel according to this embodiment. That is, by connecting the unit 3D models created for each photographing session in the excavation direction, a 3D geological model can be created covering the period from the start of excavation of tunnel T to the present.

[0054] Next, the operation of the method for creating a three-dimensional geological model of a tunnel according to this embodiment will be described. In the method for creating a three-dimensional geological model of a tunnel of this embodiment, which is configured as described above, image data of parallax images and RGB images are obtained by photographing the excavation surface, such as the excavation surface T2 including the sprayed portion F1.

[0055] Then, a 3D geological model of only the excavation surface is created by removing the unnecessary sprayed portion F1 from the initial composite model, which is created by overlaying the unit 3D models created from the image data in the excavation direction.

[0056] In this way, geological information about the inner surface that appears after tunnel excavation can be continuously pieced together, allowing even inexperienced engineers to intuitively understand the actual geological structure.

[0057] In other words, it will be possible to confirm geological information such as rock type, cracks, hardness, and softness, which are important for process management, safety management, and quality control in mountain tunnel construction, by visualizing the geological structure in 3D, making it easier to understand changes in the geological condition.

[0058] Furthermore, if photographs are taken using a self-propelled camera device 1 that has been driven to the vicinity of the tunnel face T1 immediately after excavation, the excavation surface can be observed immediately after excavation without the need for a person to approach the face T1, which is at risk of falling rocks or collapses, making it possible to perform highly accurate ground evaluation based on cracks, etc.

[0059] Furthermore, by creating a 3D geological model of the rock surface, it will be possible to observe the ground at any time later. Also, a 3D geological model will allow us to grasp the unevenness of the excavation surface, which can be used for fault and key block analysis.

[0060] The embodiments of the present invention have been described above in detail with reference to the drawings, but the specific configuration is not limited to this embodiment, and design changes that do not deviate from the gist of the present invention are included in the present invention.

[0061] For example, in the above embodiment, the camera unit 2 is described as being equipped with a high-resolution camera 23, but this is not limited to this, and it is sufficient if it is equipped with a digital camera that can acquire image data of an RGB image.

[0062] In addition, in the above embodiment, the unit excavation step was described as being 1 m in the excavation direction, but this is not limited to this, and the unit of the unit excavation step can be set arbitrarily to suit the conditions at the excavation site of tunnel T. [Explanation of symbols]

[0063] 1: Self-propelled camera (camera) 11: Running part 22: Stereo camera 23: High-resolution camera T: Tunnel T1: Face T2: Unexcavated surface (inner surface, excavation surface) F1: Spraying section R: Circumferential direction

Claims

1. A method for creating a three-dimensional geological model of a tunnel whose excavation surface and inner circumferential surface are covered with shotcrete, comprising: a step of repeating, for each unit excavation step, an operation of acquiring image data of a parallax image and an RGB image by photographing the excavation surface including the sprayed portion of the shotcrete after excavation; creating a unit three-dimensional model from the image data for each unit excavation process; For each of the unit 3D models, identifying data corresponding to the spraying portion using an image segmentation technique; creating an initial composite model by superimposing data corresponding to the sprayed portion that is an overlapping portion of the unit three-dimensional models of the successive unit excavation steps; creating a three-dimensional geological model by removing data corresponding to the sprayed portion from the initial composite model; the unit 3D model is created by superimposing point cloud data generated from the parallax image and the RGB image; A method for creating a three-dimensional geological model of a tunnel, characterized in that the overlapping parts when creating the initial composite model are superimposed based on the point cloud data of the data corresponding to the sprayed part.

2. The method for creating a three-dimensional geological model of a tunnel as described in claim 1, characterized in that the work of acquiring the image data is performed by an imaging device that is self-propelled to the vicinity of the tunnel face immediately after excavation, and the imaging device photographs the inner surface of the tunnel in a circumferential direction.

3. 3. The method for creating a three-dimensional geological model of a tunnel according to claim 1, wherein the parallax images are acquired by photographing with a stereo camera, and the RGB images are acquired by a high-resolution camera.

Images