System and method for monitoring deformation of roadway while drilling

Abstract

A system (100) and method for monitoring deformation of a roadway while drilling. The system (100) for monitoring deformation of a roadway while drilling comprises a plurality of feature points (11, 12, 13), at least one group of deformation monitoring binocular cameras (2), and a computer (3). The plurality of feature points (11, 12, 13) are dispersedly arranged on the surface of the roadway; the deformation monitoring binocular cameras (2) photograph the feature points (11, 12, 13) from different angles; acquired images (201, 202) comprise light of the feature points (11, 12, 13); the deformation monitoring binocular cameras (2) are in signal connection with the computer (3), so as to transmit the acquired images (201, 202) to the computer (3); and the computer (3) calculates three-dimensional spatial coordinates of the feature points (11, 12, 13) in a camera coordinate system in real time on the basis of the parallax of the feature points (11, 12, 13) in the images (201, 202) at different angles acquired by the deformation monitoring binocular cameras (2), and then calculates the deformation of the surface of the roadway in real time. According to the system (100) and method for monitoring deformation of a roadway while drilling, the feature points (11, 12, 13) are provided on the surface of the roadway, the deformation monitoring binocular cameras (2) photograph the feature points (11, 12, 13) from different positions, and a binocular vision measurement principle is used to achieve real-time monitoring of the deformation of the surface of the roadway, thereby achieving real-time continuous monitoring.

Description

Tunnel excavation deformation monitoring system and method Technical Field

[0001] The present disclosure relates to the field of mining technology, and in particular to a system and method for monitoring deformation of a tunnel during excavation. Background Art

[0002] Due to limited space within the tunneling face, current roadway pressure monitoring sensors are located behind the tunneling face. Roadway deformation primarily occurs within the initial three hours of surrounding rock exposure. Therefore, timely monitoring of roadway deformation during excavation is crucial for evaluating surrounding rock stability and optimizing support strategies.

[0003] During existing tunneling, the constant movement of the machine and the on-site workers' continuous work of laying mesh and anchor bolting make it difficult to use traditional tunnel deformation measurement methods, such as the cross-point method. Furthermore, after 3D laser scanning of the tunneling face, dust severely hinders the propagation of the laser light through the air, resulting in a large amount of dust noise in the collected point cloud data. This makes it difficult to acquire tunnel topography point clouds using laser scanning or lidar and calculate real-time deformation of the tunnel surface.

[0004] Public content

[0005] The disclosed embodiment provides a tunnel excavation deformation monitoring system for real-time monitoring of the deformation of the tunnel surface.

[0006] The disclosed embodiment also proposes a method for monitoring deformation of a tunnel during excavation, which realizes real-time monitoring of the deformation of the tunnel surface through the principle of binocular vision measurement.

[0007] The tunnel deformation monitoring system of the disclosed embodiment includes: multiple feature points, multiple feature points are scattered on the tunnel surface, and the feature points can emit light or reflect light; at least one group of deformation monitoring binocular cameras, the deformation monitoring binocular cameras shoot the feature points from different angles, and the captured images include the light of the feature points; the computer, the deformation monitoring binocular cameras are connected to the computer signal to transmit the captured images to the computer, and the computer calculates the three-dimensional spatial coordinates of the feature points in the camera coordinate system in real time based on the parallax of the feature points in the images at different angles captured by the deformation monitoring binocular cameras, and then calculates the deformation of the tunnel surface in real time.

[0008] The tunnel deformation monitoring system of the disclosed embodiment sets feature points on the tunnel surface, and a deformation monitoring binocular camera photographs the feature points from different positions. A computer collects the camera images in real time and performs visual operations, and uses the binocular vision measurement principle to achieve real-time monitoring of tunnel surface deformation.

[0009] In some embodiments, the feature points are reflective auxiliary feature points, and multiple cameras include fill lights for fill light. The reflective auxiliary feature points can reflect the light emitted by the fill lights so that the images captured by the multiple cameras include the reflected light of the feature points.

[0010] In some embodiments, the feature point is a patch reflective auxiliary feature point or a spherical reflective auxiliary feature point.

[0011] In some embodiments, the feature point is a luminous auxiliary feature point, which can emit light so that the images captured by the multiple cameras include the emitted light of the feature point.

[0012] In some embodiments, the feature points include a first feature point, a second feature point and a third feature point, the first feature point is set on the top plate of the tunnel, the second feature point is set on the left side of the tunnel, and the third feature point is set on the right side of the tunnel opposite to the left side.

[0013] In some embodiments, the change in vertical distance from the first feature point to the line connecting the second feature point and the third feature point is used as the amount of top plate sinking, and the change in length of the line connecting the second feature point and the third feature point is used as the amount of the two sides moving closer.

[0014] In some embodiments, the deformation monitoring binocular camera includes a first camera and a second camera spaced apart in a horizontal direction, wherein the first camera is adjacent to a left side of the alley, and the second camera is adjacent to a right side of the alley.

[0015] In some embodiments, the deformation monitoring binocular camera further includes a mounting frame, the mounting frame is horizontally arranged, the first camera is arranged at one end of the mounting frame, and the second camera is arranged at the other end of the mounting frame.

[0016] In some embodiments, the mounting bracket of the deformation monitoring binocular camera is installed on a drilling and anchoring machine or an anchor transfer machine.

[0017] In some embodiments, there are multiple groups of deformation monitoring binocular cameras, and the multiple groups of deformation monitoring binocular cameras shoot from different angles.

[0018] The method for monitoring tunnel deformation during excavation according to the embodiment of the present disclosure includes:

[0019] Setting feature points: setting at least one feature point on the left side, right side and roof of the tunnel respectively; Image acquisition: using a deformation monitoring binocular camera to collect images of the feature points from different angles;

[0020] Coordinate solution: Based on the parallax of feature points in the image collected by the deformation monitoring binocular camera, the three-dimensional spatial coordinates of the feature points in the camera coordinate system are calculated in real time;

[0021] Deformation calculation: Calculate the deformation of the tunnel surface in real time based on the change in the three-dimensional spatial coordinates of the feature points.

[0022] The tunnel deformation monitoring method of the disclosed embodiment sets feature points on the tunnel surface, and a deformation monitoring binocular camera photographs the feature points from different positions. A computer collects camera images in real time and performs visual operations. The binocular vision measurement principle is used to realize real-time monitoring of tunnel surface deformation. The monitoring process is less affected by dust, and real-time continuous monitoring can be achieved.

[0023] In some embodiments, in the step of setting feature points: the feature points are reflective auxiliary feature points, and multiple cameras include fill lights for fill light, and the reflective auxiliary feature points can reflect the light emitted by the fill lights, so that the images captured by the multiple cameras include the reflected light of the feature points; or, the feature points are luminous auxiliary feature points, and the reflective auxiliary feature points can emit light, so that the images captured by the multiple cameras include the emitted light of the feature points.

[0024] In some embodiments, in the step of setting feature points: the feature points include a first feature point, a second feature point and a third feature point, the first feature point is set on the top plate of the alley, the second feature point is set on the left side of the alley, and the third feature point is set on the right side of the alley opposite to the left side, and the second feature point and the third feature point are opposite in the width direction of the alley.

[0025] In some embodiments, in the image acquisition step: the deformation monitoring binocular camera includes a first camera and a second camera spaced apart in the horizontal direction, the first camera is adjacent to the left side of the alley, and the second camera is adjacent to the right side of the alley.

[0026] In some embodiments, the image acquisition step further includes an adjustment step of the deformation monitoring binocular camera, including:

[0027] activating the first camera and the second camera;

[0028] The first camera and the second camera collect images of feature points in the tunnel in real time from different angles, and the collected images include at least one feature point respectively set on the left side, right side and roof of the tunnel;

[0029] adjusting the positions of the first camera and the second camera in a horizontal direction;

[0030] Calculating the horizontal position of the feature point in the image;

[0031] determining whether the horizontal position of the feature point in the image is within a first preset position range, and if the horizontal position of the feature point in the image is not within the first preset position range, continuing to adjust the positions of the first camera and the second camera in the horizontal direction until the horizontal position of the feature point in the image is within the first preset position range;

[0032] When the horizontal position of the feature point in the image is within a first preset position range, the pitch angles of the first camera and the second camera are adjusted;

[0033] Calculating the longitudinal position of the feature point in the image;

[0034] Determine whether the longitudinal position of the feature point in the image is within a second preset position range; if the longitudinal position of the feature point in the image is not within the second preset position range, continue adjusting the pitch angles of the first camera and the second camera until the longitudinal position of the feature point in the image is within the second preset position range.

[0035] In some embodiments, the image acquisition step further includes an adjustment step of the deformation monitoring binocular camera, including:

[0036] activating the first camera and the second camera;

[0037] The first camera and the second camera collect images of feature points in the tunnel in real time from different angles, and the collected images include at least one feature point respectively set on the left side, right side and roof of the tunnel;

[0038] adjusting the pitch angles of the first camera and the second camera;

[0039] Calculating the longitudinal position of the feature point in the image;

[0040] determining whether the longitudinal position of the feature point in the image is within a third preset position range, and if the longitudinal position of the feature point in the image is not within the third preset position range, continuing to adjust the pitch angles of the first camera and the second camera until the longitudinal position of the feature point in the image is within the third preset position range;

[0041] When the longitudinal position of the feature point in the image is within a third preset position range, the positions of the first camera and the second camera are adjusted in the horizontal direction;

[0042] Calculating the horizontal position of the feature point in the image;

[0043] Determine whether the horizontal position of the feature point in the image is within a fourth preset position range; if the horizontal position of the feature point in the image is not within the fourth preset position range, continue adjusting the positions of the first camera and the second camera in the horizontal direction until the horizontal position of the feature point in the image is within the fourth preset position range.

[0044] In some embodiments, in the step of adjusting the positions of the first camera and the second camera in the horizontal direction, the closer the distance between the feature points is, the smaller the horizontal spacing between the first camera and the second camera is adjusted, and the farther the distance between the feature points is, the larger the horizontal spacing between the first camera and the second camera is adjusted.

[0045] In some embodiments, there are multiple groups of deformation monitoring binocular cameras, and the multiple groups of deformation monitoring binocular cameras shoot from different angles.

[0046] In some embodiments, the coordinate solving step specifically includes:

[0047] Feature point extraction: Extract feature points from the image and calculate the two-dimensional centroid coordinates of the highlighted circular features in the image through visual recognition algorithms;

[0048] Calculating the 3D coordinates of feature points: Based on the pre-calibrated external parameter information of the deformation monitoring binocular camera, using the principle of triangulation, the 3D coordinates of the feature points in the camera coordinate system are calculated based on the different parallaxes of the same feature point in different cameras of the deformation monitoring binocular camera.

[0049] Roof feature classification: Classify the feature points with three-dimensional coordinates that are actually solved, and determine whether the feature points are on the left side, right side or roof of the tunnel.

[0050] In some embodiments, the coordinate solving step further includes, before extracting the feature points:

[0051] Image correction: the deformation monitoring binocular camera transmits the collected image to a computer, the computer obtains the image source file and performs format conversion, converts the image information into matrix data, and the computer corrects the image into a distortion-free image based on pre-calibrated distortion parameters;

[0052] The feature point extraction step is performed based on the feature points in the corrected image.

[0053] In some embodiments, the feature point extraction step also includes: feature point matching, matching the feature points in multiple images captured by the deformation monitoring binocular camera according to position and shape features to ensure that the matched feature points correspond to the same feature point in space.

[0054] In some embodiments, in the deformation calculation step: a line is drawn between the feature points located on the left side of the tunnel and the feature points on the right side, the change in the vertical distance from the feature point located on the top plate of the tunnel to the line is used as the amount of top plate subsidence, and the change in the length of the line is used as the amount of the two sides moving closer. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] FIG1 is a schematic diagram of a tunnel excavation deformation monitoring system according to an embodiment of the present disclosure.

[0056] FIG2 is a schematic diagram of the monitoring principle of the tunnel excavation deformation monitoring system according to an embodiment of the present disclosure.

[0057] FIG3 is a flow chart of a monitoring method of a tunnel excavation deformation monitoring system according to an embodiment of the present disclosure.

[0058] FIG4 is a flow chart of a method for adjusting a deformation monitoring binocular camera according to an embodiment of the present invention.

[0059] Reference numerals:

[0060] Tunnel excavation deformation monitoring system 100,

[0061] Feature point 1, patch reflective auxiliary feature point 101, spherical reflective auxiliary feature point 102, first feature point 11, second feature point 12, third feature point 13, deformation monitoring binocular camera 2, first image 201, second image 202, first camera 21, second camera 22, first guide rail 24, second guide rail 25, mounting bracket 23, computer 3, network cable 4. DETAILED DESCRIPTION

[0062] The embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present disclosure, but should not be understood as limiting the present disclosure.

[0063] The tunnel excavation deformation monitoring system 100 according to an embodiment of the present disclosure will be described below with reference to FIG. 1 and FIG. 2 . The tunnel excavation deformation monitoring system 100 includes a plurality of feature points 1 , at least one set of deformation monitoring binocular cameras 2 , and a computer 3 .

[0064] Multiple feature points 1 are dispersed across the tunnel surface, each capable of emitting or reflecting light. Each set of deformation monitoring binocular cameras 2 captures the feature points 1 from different angles, and the captured images include the light from the feature points 1. The deformation monitoring binocular cameras 2 are connected to a computer 3 for signal transmission, transmitting the captured images to the computer 3. Based on the parallax of the feature points in the images captured from different angles by the deformation monitoring binocular cameras 2, the computer 3 calculates the three-dimensional spatial coordinates of the feature points 1 in the camera coordinate system in real time, thereby calculating the deformation of the tunnel surface in real time.

[0065] The tunnel deformation monitoring system of the disclosed embodiment sets feature points on the tunnel surface, and a deformation monitoring binocular camera photographs the feature points from different positions. A computer collects camera images in real time and performs visual operations, and uses binocular vision measurement principles to achieve real-time monitoring of tunnel surface deformation. The monitoring process is less affected by dust, and real-time continuous monitoring can be achieved.

[0066] In some embodiments, as shown in Figure 1, the feature point 1 is a reflective auxiliary feature point, and the deformation monitoring binocular camera 2 includes a fill light for fill light. The reflective auxiliary feature point can reflect the light emitted by the fill light so that the image captured by the deformation monitoring binocular camera 2 includes the reflected light of the feature point.

[0067] Optionally, the feature point 1 is made of a highly reflective material and does not require a power source, thereby reducing the cost of the feature point 1 .

[0068] In some optional embodiments, the feature point 1 is a patch reflective auxiliary feature point 101, and the patch reflective auxiliary feature point 101 is attached to the surface of the lane, and the installation method is simple and convenient.

[0069] In some optional embodiments, the feature point 1 is a spherical reflective auxiliary feature point 102. The spherical reflective auxiliary feature point 102 can reflect light in different directions, so that the deformation monitoring binocular cameras 2 located at different positions can capture the light of the fill light reflected by the feature point 1.

[0070] In other embodiments, feature point 1 is a light-emitting auxiliary feature point that emits light, so that the image captured by the deformation monitoring binocular camera 2 includes the light emitted by feature point 1. In other words, feature point 1 actively emits light, making it easier for the deformation monitoring binocular camera 2 to capture it.

[0071] In some embodiments, as shown in FIG2 , the characteristic points 1 include a first characteristic point 11, a second characteristic point 12, and a third characteristic point 13. The first characteristic point 11 is located on the roof of the laneway, the second characteristic point 12 is located on the left side of the laneway, and the third characteristic point 13 is located on the right side of the laneway opposite the left side. The left side of the laneway is the left side wall of the laneway, and the right side of the laneway is the right side wall of the laneway.

[0072] Images captured from different angles by the deformation monitoring binocular camera 2 all include first feature points 11, second feature points 12, and third feature points 13. Computer 3 uses binocular vision measurement principles to calculate the three-dimensional spatial coordinates of each of the first, second, and third feature points 11, 12, and 13 based on their parallax in different images. Based on the changes in the three-dimensional spatial coordinates of the first, second, and third feature points 11, 12, and 13, the deformation of the roadway surface can be calculated in real time. This requires fewer feature points 1, facilitating their arrangement.

[0073] When calculating the deformation of the tunnel, the change in the vertical distance from the first feature point 11 to the line connecting the second feature point 12 and the third feature point 13 is used as the roof sinking amount, and the change in the length of the line connecting the second feature point 12 and the third feature point 13 is used as the amount of movement of the two sides.

[0074] In other words, a line is connected between the second feature point 12 and the third feature point 13, and the change in the vertical distance between the first feature point 11 and the line is the amount of subsidence of the tunnel roof, and the change in the distance between the second feature point 12 and the third feature point 13 is the amount of movement between the two sides of the tunnel, that is, the change in the width of the tunnel.

[0075] In order to better characterize the roof subsidence and the amount of movement of the two sides of the tunnel, the second characteristic point 12 and the third characteristic point 13 are opposite in the width direction of the tunnel, and the line between the second characteristic point 12 and the third characteristic point 13 is a horizontal line and is orthogonal to the length direction of the tunnel.

[0076] Furthermore, a perpendicular line is drawn from the first feature point 11 to the ground, and the perpendicular line can be connected to the line between the second feature point 12 and the third feature point 13 and be perpendicular to each other, that is, the first feature point 11, the second feature point 12 and the third feature point 13 are on the same vertical plane.

[0077] In some embodiments, as shown in Figures 1 and 2, the deformation monitoring binocular camera 2 includes a first camera 21 and a second camera 22 spaced apart horizontally. A roadway has opposing left and right sides, with the first camera 21 positioned adjacent to the left side of the roadway and the second camera 22 positioned adjacent to the right side. In other words, the first camera 21, acting as the left camera of the deformation monitoring binocular camera 2, photographs each feature point 1 from a position adjacent to the left side of the roadway, generating a first image 201. The second camera 22, acting as the right camera of the deformation monitoring binocular camera 2, photographs each feature point 1 from a position adjacent to the right side of the roadway, generating a second image 202.

[0078] Since the positions of the first camera 21 and the second camera 22 are different, the shooting angles of the first camera 21 and the second camera 22 are different. Therefore, there is parallax between the feature points 1 in the obtained first image 201 and the second image 202, so that the computer 3 can realize real-time monitoring of the deformation of the tunnel surface based on the image information of the first image 201 and the second image 202 based on the binocular vision measurement principle.

[0079] As shown in Figures 1 and 2, the deformation monitoring binocular camera 2 also includes a horizontal mounting bracket 23, with the first camera 21 and the second camera 22 mounted on its upper surface. The mounting bracket 23 is mounted on a drill or anchoring machine or bolting loader, and moves with the drill or anchoring machine to enable ongoing monitoring.

[0080] In some embodiments, the first camera 21 and the second camera 22 are movable in the horizontal direction, and the first camera 21 and the second camera 22 are pivotally mounted on the mounting bracket 23 so that their pitch angles are adjustable.

[0081] Specifically, as shown in Figure 3, the deformation monitoring binocular camera 2 includes a first guide rail 24 and a second guide rail 25, which are arranged on the upper surface of the mounting frame 23 and both extend along the horizontal direction, wherein the first camera 21 is pivotally arranged on the first guide rail 24 and can slide along the first guide rail 24, thereby adjusting the horizontal position and vertical position of the feature point 1 in the image captured by the first camera 21, and the second camera 22 is pivotally arranged on the second guide rail 25 and can slide along the second guide rail 25, thereby adjusting the horizontal position and vertical position of the feature point 1 in the image captured by the second camera 21.

[0082] The deformation monitoring binocular camera 2 also includes a first guide rail motor and a second guide rail motor (not shown). The first guide rail motor is used to drive the first camera 21 to slide along the first guide rail 24, and the second guide rail motor is used to drive the second camera 22 to slide along the second guide rail 25. The deformation monitoring binocular camera 2 also includes a first rotary motor and a second rotary motor (not shown). The first rotary motor acts on the first camera 21 to drive the adjustment of the first camera 21's pitch angle, and the second rotary motor acts on the second camera 22 to drive the adjustment of the second camera 22's pitch angle.

[0083] In some embodiments, as shown in FIG1 , there are multiple groups of deformation monitoring binocular cameras 2 , and the multiple groups of deformation monitoring binocular cameras 2 are combined to shoot from different angles so that the three-dimensional spatial coordinates of the obtained feature points 1 in the camera coordinate system are more accurate, and thus the calculated deformation of the tunnel surface is more accurate.

[0084] As an example, the deformation monitoring binocular cameras 2 are three groups.

[0085] Optionally, the computer 3 is a mine-use explosion-proof and intrinsically safe computer.

[0086] Optionally, the deformation monitoring binocular camera 2 and the computer 3 are connected by a network cable 4 signal, or the deformation monitoring binocular camera 2 and the computer 3 are connected by a wireless signal.

[0087] The following describes a method for monitoring deformation of a tunnel during excavation according to an embodiment of the present disclosure with reference to FIG. 3 . The method for monitoring deformation of a tunnel during excavation includes the following steps:

[0088] Set feature points: set at least one feature point 1 on the left side, right side and roof of the tunnel respectively;

[0089] Image acquisition: The deformation monitoring binocular camera 2 is used to collect images of the feature points 1 from different angles, and the collected images include at least one feature point set on the left side, right side and roof of the tunnel respectively;

[0090] Coordinate solution: Based on the parallax of feature point 1 in the image captured by the deformation monitoring binocular camera 2, the three-dimensional spatial coordinates of feature point 1 in the camera coordinate system are calculated in real time;

[0091] Deformation calculation: Based on the change in the three-dimensional spatial coordinates of feature point 1, the deformation of the tunnel surface is calculated in real time.

[0092] It should be understood that the left side of the tunnel is the left side wall of the tunnel, and the right side is the right side wall of the tunnel.

[0093] Among them, in the image acquisition step, the feature points 1 displayed in the images captured by the deformation monitoring binocular camera 2 from different angles include at least one feature point 1 on the left side of the tunnel, at least one feature point 1 on the right side of the tunnel, and at least one feature point 1 on the top plate of the tunnel, and the feature points 1 in the images captured from different angles are the same feature points 2. In the coordinate solution step, the coordinates of these feature points 1 are solved respectively.

[0094] The tunnel deformation monitoring method of an embodiment of the present invention sets feature points on the tunnel surface, and a deformation monitoring binocular camera photographs the feature points from different positions. A computer collects camera images in real time and performs visual operations. The binocular vision measurement principle is used to realize real-time monitoring of tunnel surface deformation. In addition, the monitoring process is less affected by dust, and real-time continuous monitoring can be achieved.

[0095] In some embodiments, as shown in Figure 1, in the step of setting feature points: feature point 1 is a reflective auxiliary feature point, the deformation monitoring binocular camera 2 includes a fill light for fill light, and the reflective auxiliary feature point can reflect the light emitted by the fill light, so that the image captured by the deformation monitoring binocular camera 2 includes the reflected light of the feature point.

[0096] Optionally, the feature point 1 is made of a highly reflective material and does not require a power source, thereby reducing the cost of the feature point 1 .

[0097] In some optional embodiments, as shown in FIG1 , the feature point 1 is a patch reflective auxiliary feature point 101 , and the patch reflective auxiliary feature point 101 is attached to the surface of the lane, and the installation method is simple and convenient.

[0098] In some optional embodiments, as shown in Figure 1, the feature point 1 is a spherical reflective auxiliary feature point 102. The spherical reflective auxiliary feature point 102 can reflect light in different directions, so that the deformation monitoring binocular cameras 2 located at different positions can capture the light of the fill light reflected by the feature point 1.

[0099] In other embodiments, feature point 1 is a light-emitting auxiliary feature point that emits light, so that the image captured by the deformation monitoring binocular camera 2 includes the light emitted by feature point 1. In other words, feature point 1 actively emits light, making it easier for the deformation monitoring binocular camera 2 to capture it.

[0100] In some embodiments, as shown in Figure 2, the feature point 1 includes a first feature point 11, a second feature point 12 and a third feature point 13. The first feature point 11 is set on the top plate of the tunnel, the second feature point 12 is set on the left side of the tunnel, and the third feature point 13 is set on the right side of the tunnel opposite to the left side.

[0101] During the image acquisition step, images captured from different angles by the deformation monitoring binocular camera 2 each include the first feature point 11, the second feature point 12, and the third feature point 13. During the coordinate calculation step, utilizing binocular vision measurement principles, the three-dimensional coordinates of the first, second, and third feature points 11, 12, and 13 are calculated based on the parallax between the different images. During the deformation calculation step, the roadway surface deformation is calculated in real time based on the change in the three-dimensional coordinates of the first, second, and third feature points 11, 12, and 13. This requires fewer feature points 1, facilitating their placement.

[0102] In order to better characterize the roof subsidence and the amount of movement of the two sides of the tunnel, the second characteristic point 12 and the third characteristic point 13 are opposite in the width direction of the tunnel, and the line between the second characteristic point 12 and the third characteristic point 13 is a horizontal line and is orthogonal to the length direction of the tunnel.

[0103] Furthermore, a perpendicular line is drawn from the first feature point 11 to the ground, and the perpendicular line can be connected to the line between the second feature point 12 and the third feature point 13 and be perpendicular to each other, that is, the first feature point 11, the second feature point 12 and the third feature point 13 are on the same vertical plane.

[0104] In some embodiments, as shown in Figures 1 and 2 , during the image acquisition step, the deformation monitoring binocular camera 2 includes a first camera 21 and a second camera 22 spaced apart horizontally. The first camera 21 is positioned adjacent to the left side of the laneway, while the second camera 22 is positioned adjacent to the right side of the laneway. In other words, the first camera 21, acting as the left camera of the deformation monitoring binocular camera 2, photographs each feature point 1 from a position adjacent to the left side of the laneway to obtain a first image 201. The second camera 22, acting as the right camera of the deformation monitoring binocular camera 2, photographs each feature point 1 from a position adjacent to the right side of the laneway to obtain a second image 202.

[0105] Since the positions of the first camera 21 and the second camera 22 are different, the shooting angles of the first camera 21 and the second camera 22 are different. Therefore, there is parallax between the feature points 1 in the obtained first image 201 and the second image 202, so that the computer 3 can realize real-time monitoring of the deformation of the tunnel surface based on the image information of the first image 201 and the second image 202 based on the binocular vision measurement principle.

[0106] As shown in Figures 1 and 2, the deformation monitoring binocular camera 2 also includes a horizontal mounting bracket 23, with the first camera 21 and the second camera 22 mounted on its upper surface. The mounting bracket 23 can be installed on a driller or anchoring machine, allowing it to move with the machine to enable ongoing monitoring.

[0107] In some embodiments, the first camera 21 and the second camera 22 are movable in the horizontal direction, and the first camera 21 and the second camera 22 are pivotally mounted on the mounting bracket 23 so that their pitch angles are adjustable.

[0108] Specifically, as shown in Figure 1, the deformation monitoring binocular camera 2 includes a first guide rail 24 and a second guide rail 25, which are arranged on the upper surface of the mounting frame 23 and both extend along the horizontal direction, wherein the first camera 21 is pivotally arranged on the first guide rail 24 and can slide along the first guide rail 24, thereby adjusting the horizontal position and vertical position of the feature point 1 in the image captured by the first camera 21, and the second camera 22 is pivotally arranged on the second guide rail 25 and can slide along the second guide rail 25, thereby adjusting the horizontal position and vertical position of the feature point 1 in the image captured by the second camera 21.

[0109] The deformation monitoring binocular camera 2 also includes a first guide rail motor and a second guide rail motor (not shown). The first guide rail motor is used to drive the first camera 21 to slide along the first guide rail 24, and the second guide rail motor is used to drive the second camera 22 to slide along the second guide rail 25. The deformation monitoring binocular camera 2 also includes a first rotary motor and a second rotary motor (not shown). The first rotary motor acts on the first camera 21 to drive the adjustment of the first camera 21's pitch angle, and the second rotary motor acts on the second camera 22 to drive the adjustment of the second camera 22's pitch angle.

[0110] Optionally, before the image acquisition step, a step of adjusting the deformation monitoring binocular camera 2 is further included. As shown in FIG4 , the adjustment method of the deformation monitoring binocular camera 2 specifically includes:

[0111] Step S101: Start the first camera 21 and the second camera 22;

[0112] Step S102: The first camera 21 and the second camera 22 collect images of the feature points 1 in the tunnel from different angles in real time, and the collected images include at least one feature point 1 set on the left side, right side, and roof of the tunnel respectively;

[0113] Step S103: adjusting the positions of the first camera 21 and the second camera 22 in the horizontal direction;

[0114] Step S104: Calculate the horizontal position of feature point 1 in the image;

[0115] Step S105: determining whether the horizontal position of the feature point 1 in the image is within a first preset position range; if the horizontal position of the feature point 1 in the image is not within the first preset position range, continuing to adjust the positions of the first camera 21 and the second camera 22 in the horizontal direction until the horizontal position of the feature point 1 in the image is within the first preset position range;

[0116] Step S106: If the horizontal position of the feature point 1 in the image is within the first preset position range, adjusting the pitch angles of the first camera 21 and the second camera 22;

[0117] Step S107: Calculate the vertical position of feature point 1 in the image;

[0118] Step S108: Determine whether the longitudinal position of feature point 1 in the image is within the second preset position range. If the longitudinal position of feature point 1 in the image is not within the second preset position range, continue to adjust the pitch angles of the first camera 21 and the second camera 22 until the longitudinal position of feature point 1 in the image is within the second preset position range.

[0119] Taking Figures 1 and 2 as an example, in step S102, the image captured by the first camera 21 contains the first feature point 11, the second feature point 12, and the third feature point 13. The image captured by the second camera 22 shows the first feature point 11, the second feature point 12, and the third feature point 13. In step S103, the first camera 21 is slid along the first guide rail 24, and the second camera 22 is slid along the second guide rail 24 to adjust the positions of the first camera 21 and the second camera 22 in the horizontal direction.

[0120] According to the above-mentioned adjustment method of the deformation monitoring binocular camera 2, the lateral translation and pitch angles of the first camera 21 and the second camera 22 are adjusted, and the computer calculates and determines the lateral position and longitudinal position of the feature point 1 in the image in real time. According to the difference between the lateral position and longitudinal position of the feature point 1 in the image and the target set position, the translation position and pitch angle of the first camera 21 and the second camera 22 can be adjusted in real time. Then, based on the image of the feature point 1 obtained by the binocular camera, the deformation amount of the tunnel can be calculated more accurately. The above-mentioned adjustment method of the deformation monitoring binocular camera 2 can also make the deformation monitoring binocular camera 2 more applicable and have a wider range of application methods.

[0121] Adjusting the positions of the first camera 21 and the second camera 22 in the horizontal direction in step S103 may specifically include: adjusting the positions of the first camera 21 and the second camera 22 in the horizontal direction to reduce the distance between the first camera 21 and the second camera 22 as the distance between the feature points 1 is closer. Adjusting the positions of the first camera 21 and the second camera 22 in the horizontal direction to increase the distance between the first camera 21 and the second camera 22 as the distance between the feature points 1 is farther.

[0122] According to the position of the feature point 1 in the depth, width and length directions of the tunnel, the spacing between the first camera 21 and the second camera 22 is adjusted, and then the parallax of the feature point 1 in the images captured by the first camera 21 and the second camera 22 is adjusted to more accurately identify each feature point 1 and accurately calculate the three-dimensional spatial coordinates of the feature point 1 in the camera coordinate system, thereby achieving a higher accuracy in the calculated tunnel deformation result.

[0123] It should be noted that the adjustment method of the deformation monitoring binocular camera 2 is not limited to this. In other optional examples, in the adjustment method of the deformation monitoring binocular camera 2, the longitudinal position of the feature point 1 in the image can be adjusted first, and then the lateral position of the feature point 1 in the image can be adjusted. That is, the pitch angles of the first camera 21 and the second camera 21 can be adjusted first, and then the positions of the first camera 21 and the second camera 22 in the horizontal direction can be adjusted. The specific steps are as follows:

[0124] Step S201: Start the first camera 110 and the second camera 120;

[0125] Step S202: The first camera 110 and the second camera 120 collect images of the feature points 200 in the lane from different angles in real time, and the collected images include at least one feature point 200 respectively set on the left side, right side, and roof of the lane;

[0126] Step S203: adjusting the pitch angles of the first camera 110 and the second camera 120;

[0127] Step S204: calculating the longitudinal position of the feature point 200 in the image;

[0128] Step S205: determining whether the longitudinal position of the feature point 200 in the image is within a third preset position range; if the longitudinal position of the feature point 200 in the image is not within the third preset position range, continuing to adjust the pitch angles of the first camera 110 and the second camera 120 until the longitudinal position of the feature point 200 in the image is within the third preset position range;

[0129] Step S206: When the longitudinal position of the feature point 200 in the image is within a third preset position range, the positions of the first camera 110 and the second camera 120 are adjusted in the horizontal direction;

[0130] Step S207: Calculate the horizontal position of the feature point 200 in the image;

[0131] Step S208: Determine whether the horizontal position of the feature point 200 in the image is within a fourth preset position range. If the horizontal position of the feature point 200 in the image is not within the fourth preset position range, continue to adjust the positions of the first camera 110 and the second camera 120 in the horizontal direction until the horizontal position of the feature point 200 in the image is within the fourth preset position range.

[0132] In step S202 , the feature points 200 displayed in the image captured by the first camera 110 include at least one feature point 200 on the left side of the tunnel, at least one feature point 200 on the right side of the tunnel, and at least one feature point 200 on the tunnel roof. The same applies to the second camera 120 .

[0133] In some embodiments, as shown in FIG1 , in the image acquisition step: there are multiple groups of deformation monitoring binocular cameras 2, and the multiple groups of deformation monitoring binocular cameras 2 are combined to shoot from different angles, so that the three-dimensional spatial coordinates of the obtained feature points 1 in the camera coordinate system are more accurate, and thus the calculated deformation of the tunnel surface is more accurate.

[0134] As an example, the deformation monitoring binocular cameras 2 are three groups.

[0135] In some embodiments, in the deformation calculation step: a line is drawn between feature point 1 located on the left side of the tunnel and feature point 1 located on the right side, the vertical distance change from feature point 1 located on the top plate of the tunnel to the line is taken as the amount of top plate subsidence, and the length change of the line is taken as the amount of the two sides moving closer.

[0136] As an example, a line is connected between the second feature point 12 and the third feature point 13. The change in the vertical distance between the first feature point 11 and the line is the amount of subsidence of the tunnel roof. The change in the distance between the second feature point 12 and the third feature point 13 is the amount of movement between the two sides of the tunnel, that is, the change in the width of the tunnel.

[0137] Optionally, the computer 3 is a mine-use explosion-proof and intrinsically safe computer.

[0138] Optionally, as shown in FIG1 , the deformation monitoring binocular camera 2 and the computer 3 are connected via a network cable 4 .

[0139] Optionally, the deformation monitoring binocular camera 2 and the computer 3 are connected by wireless signals.

[0140] In some embodiments, the coordinate solving step specifically includes:

[0141] Feature point extraction: Extract feature points 1 from the image and calculate the two-dimensional centroid coordinates of the highlighted circular features in the image through a visual recognition algorithm;

[0142] Calculate the 3D coordinates of feature points: Based on the pre-calibrated external parameter information of the deformation monitoring binocular camera 2, the triangulation principle is used to calculate the 3D coordinates of the feature points in the camera coordinate system based on the different parallaxes of the same feature point 1 in different cameras of the deformation monitoring binocular camera 2.

[0143] Roof feature classification: classify the feature point 1 with three-dimensional coordinates actually solved, and determine whether the feature point 1 is on the left side, right side or roof of the tunnel.

[0144] In some embodiments, the coordinate solving step further includes, before extracting the feature points:

[0145] Image correction: The deformation monitoring binocular camera 2 transmits the collected image to the computer 3. The computer 3 obtains the image source file and performs format conversion, converts the image information into matrix data, and corrects the image to a distortion-free image based on pre-calibrated distortion parameters;

[0146] In the feature point extraction step, feature points 1 in the corrected image are extracted.

[0147] In some embodiments, the feature point extraction step also includes: feature point matching, matching the feature points 1 in multiple images captured by the deformation monitoring binocular camera 2 according to position and shape features to ensure that the matched feature points 1 correspond to the same feature point in space.

[0148] In some embodiments, in the deformation calculation step: the change in the vertical distance from the feature point located on the tunnel roof to the line connecting the feature points located on the two sides of the tunnel is used as the roof sinking amount, and the change in the length of the line connecting the feature points located on the two sides of the tunnel is used as the amount of movement of the two sides.

[0149] The following describes a tunnel deformation monitoring method in a specific embodiment of the present disclosure using FIG. 1 to FIG. 3 as an example. The tunnel deformation monitoring method is based on a tunnel deformation monitoring system 100 for monitoring.

[0150] In some examples, the tunnel excavation deformation monitoring system 100 includes a first feature point 11 , a second feature point 12 , and a third feature point 13 , at least one set of deformation monitoring binocular cameras 2 , and a computer 3 .

[0151] As shown in Figure 3, the tunnel excavation deformation monitoring method includes the following steps:

[0152] Step 1: Set feature points. Set a first feature point 11 on the top plate of the lane, set a second feature point 12 on the left side of the lane, and set a third feature point 13 on the right side of the lane.

[0153] Step 2: Start the program and start the tunnel deformation monitoring program through the host computer software.

[0154] Step 3: Image acquisition: The first camera 21 and the second camera 22 of the deformation monitoring binocular camera 2 acquire images of the feature point 1 from different angles. The first camera 21 obtains a first image 201 , and the second camera 22 obtains a second image 202 .

[0155] Step 4: Coordinate solution, including the following steps:

[0156] Step 401: Image correction: The deformation monitoring binocular camera 2 transmits the captured image to the computer 3. The computer 3 obtains the image source file and performs format conversion, converting the image information into matrix data. The computer 3 corrects the image to a distortion-free image based on pre-calibrated distortion parameters.

[0157] Step 402: Feature point extraction: extract feature point 1 from the corrected image and calculate the two-dimensional centroid coordinates of the highlighted circular feature in the image using a visual recognition algorithm.

[0158] Step 403: Feature point matching: matching the feature point 1 in the first image 201 and the second image 202 based on position and shape features to ensure that the matched feature point 1 corresponds to the same feature point 1 in space;

[0159] Step 404: Determine the three-dimensional coordinates of the feature point. Using the pre-calibrated external parameter information of the deformation monitoring binocular camera 2 and the principle of triangulation, the three-dimensional coordinates of the feature point 1 in the camera coordinate system are calculated based on the difference in parallax between the first camera 21 and the second camera 22 of the deformation monitoring binocular camera 2.

[0160] Step 405: top wall feature classification, classifying the feature point 1 with three-dimensional coordinates actually solved, and determining whether the feature point 1 is on the left wall, right wall or top plate of the tunnel.

[0161] Step 5: Calculate the deformation amount: take the vertical distance change from the first feature point 11 to the line connecting the second feature point 12 and the third feature point 13 as the top plate sinking amount, and take the length change of the line connecting the second feature point 12 and the third feature point 13 as the two sides moving closer.

[0162] Step 6: Data transmission: The calculated tunnel deformation is transmitted.

[0163] Optionally, before step 3, the above-mentioned adjustment method for the deformation monitoring binocular camera 2 may be used to adjust the deformation monitoring binocular camera 2 .

[0164] Specifically, in step 4, based on the binocular vision measurement principle, the first camera 21 and the second camera 22 of the deformation monitoring binocular camera 2 simultaneously observe the first feature point 11 (P in FIG. 2 ) located on the top plate. m ), the first image 201 obtained by the first camera 21 has an image point P1, the second image 202 obtained by the second camera 22 has an image point P2, and the first feature point 11 (P in FIG. 2 m ) is the intersection of the line between the optical center of the first camera 21 and the image point P1 and the line between the optical center of the second camera 22 and the image point P2. m )'s three-dimensional coordinates can be expressed as:

[0165] Where s1 is the projection scale factor of the first camera 21, s2 is the projection scale factor of the second camera 22, P1' is the homogeneous coordinate of the image point of the first camera 21, P2' is the homogeneous coordinate of the image point of the second camera 22, K1 is the intrinsic parameter matrix of the first camera 21, K2 is the intrinsic parameter matrix of the second camera 22, [R1|t1] is the extrinsic parameter matrix of the first camera 21, [R2|t2] is the extrinsic parameter matrix of the second camera 22, and P m ' is the homogeneous coordinate of the first feature point 11 in the global coordinate system.

[0166] The coordinate system of the first camera 21 is set as the global coordinate system, and considering the distortion, the distortion formula and the three-dimensional coordinate formula in the coordinate system of the first camera 21 are:

[0167] Where P1″ is the homogeneous coordinate of the image point after distortion correction of the first camera 21, P2″ is the homogeneous coordinate of the image point after distortion correction of the second camera 22, d1 is the distortion function of the first camera 21, d2 is the distortion function of the second camera 22, I is the unit matrix, [R S |t S ] is the extrinsic parameter matrix between the first camera 21 and the second camera 22, and the following formula exists:

[0168] Nonlinear optimization is used to solve the internal and external parameter matrix and distortion parameters of the deformation monitoring binocular camera 2. The formula is as follows:

[0169] Where P1 is the estimated value of point P1 in the reprojected image coordinate system. S and t S After that, the image coordinate P1 is known, and the first feature point 11 in the global coordinate system (P in FIG2 ) can be obtained. m )'s three-dimensional coordinates: P m =RP1+T

[0170] Referring to the above steps, the second feature point 12 (P in FIG2 ) in the global coordinate system is calculated respectively. l )'s three-dimensional coordinates P l and the second characteristic point 13 (P in FIG. 2 r )'s three-dimensional coordinates P r .

[0171] The deformation monitoring during excavation is based on the premise that the relative distance between the characteristic points in the moving coordinate system remains unchanged. The distance between the second characteristic point 12 and the third characteristic point 13 in the same section is approximated as the distance between the two sides of the tunnel. The solution formula for the distance between the two sides is as follows: α=Δ||P l -P r ||

[0172] The change in the distance from the first characteristic point 11 to the connecting line between the second characteristic point 12 and the third characteristic point 13 in the same section is approximated as the roof subsidence, and the solution formula is as follows:

[0173] The tunnel deformation monitoring method provided in the embodiment of the present invention is used to monitor the tunnel deformation. By simply installing three reflective auxiliary feature points on the surface of the tunnel, the tunnel surface deformation monitoring during the excavation process can be realized. The monitoring process is less affected by dust, and real-time continuous monitoring can be achieved.

[0174] In the description of the present disclosure, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present disclosure and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation to the present disclosure.

[0175] In this disclosure, unless otherwise expressly specified or limited, terms such as "mounted," "connected," "connect," and "fixed" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integration; mechanical connections, electrical connections, or communication between them; direct connections or indirect connections through an intermediate medium; and internal communication between two elements or interaction between two elements, unless otherwise expressly limited. Those skilled in the art will understand the specific meanings of the above terms in this disclosure based on specific circumstances.

[0176] In the present disclosure, unless otherwise expressly specified or limited, when a first feature is "above" or "below" a second feature, it may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Furthermore, when a first feature is "above," "above," or "above" a second feature, it may mean that the first feature is directly above or diagonally above the second feature, or simply means that the first feature is at a higher level than the second feature. When a first feature is "below," "below," or "below" a second feature, it may mean that the first feature is directly below or diagonally below the second feature, or simply means that the first feature is at a lower level than the second feature.

[0177] In the present disclosure, the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0178] Although the embodiments of the present disclosure have been shown and described above, it is understood that the above embodiments are illustrative and are not to be construed as limitations on the present disclosure. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present disclosure.

Claims

1. A roadway deformation monitoring system during excavation, characterized in that Comprising: Multiple feature points, with multiple of the feature points dispersedly arranged on the roadway surface, and the feature points being capable of emitting light or reflecting light; At least one set of deformation monitoring binocular cameras, which photograph the feature points from different angles, and the light of the feature points is included in the collected images; A computer, with the deformation monitoring binocular cameras being signal-connected to the computer to transmit the collected images to the computer, and the computer calculates the three-dimensional space coordinates of the feature points in the camera coordinate system in real time according to the parallax of the feature points in the images collected from different angles by the deformation monitoring binocular cameras, and further calculates the deformation condition of the roadway surface in real time.

2. The roadway deformation monitoring system during tunneling according to claim 1, wherein: The feature points are reflective auxiliary feature points, and multiple of the cameras include fill lights for supplementary lighting. The reflective auxiliary feature points are capable of reflecting the light emitted by the fill lights, so that the reflected light of the feature points is included in the images collected by multiple of the cameras.

3. The roadway deformation monitoring system during tunneling according to claim 2, wherein: The feature points are patch reflective auxiliary feature points or spherical reflective auxiliary feature points.

4. The roadway deformation monitoring system during tunneling according to claim 1, wherein: The feature points are light-emitting auxiliary feature points, and the light-emitting auxiliary feature points are capable of emitting light, so that the emitted light of the feature points is included in the images collected by multiple of the cameras.

5. The roadway deformation monitoring system during tunneling according to any one of claims 1-4, wherein: The feature points include a first feature point, a second feature point, and a third feature point. The first feature point is arranged on the roof of the roadway, the second feature point is arranged on the left sidewall of the roadway, and the third feature point is arranged on the right sidewall of the roadway opposite to the left sidewall.

6. The roadway deformation monitoring system during tunneling according to claim 5, wherein: The change amount of the vertical distance from the first feature point to the connection lines between the second feature point and the third feature point is taken as the roof subsidence amount, and the change amount of the length of the connection line between the second feature point and the third feature point is taken as the approaching amount of the two sidewalls.

7. The roadway deformation monitoring system during tunneling according to claim 1, wherein: The deformation monitoring binocular cameras include a first camera and a second camera spaced apart in the horizontal direction. The first camera is adjacent to the left sidewall of the roadway, and the second camera is adjacent to the right sidewall of the roadway.

8. The roadway deformation monitoring system during tunneling according to claim 7, wherein: The deformation monitoring binocular cameras further include a mounting frame, a first guide rail, and a second guide rail. The first guide rail and the second guide rail are both arranged on the mounting frame and both extend along the horizontal direction. The first camera is pivotally arranged on the first guide rail and can slide along the first guide rail, and the second camera is pivotally arranged on the second guide rail and can slide along the second guide rail.

9. The roadway deformation monitoring system during tunneling according to claim 8, wherein: The mounting frame of the deformation monitoring binocular cameras is mounted on a roadheader-anchoring machine or a bolt transfer machine.

10. The roadway deformation monitoring system according to claim 1, characterized in that there are multiple groups of the deformation monitoring binocular cameras, and the multiple groups of the deformation monitoring binocular cameras take pictures from different angles.

11. A method for monitoring the deformation of a roadway during excavation, characterized in that, It includes: Setting feature points: At least one feature point is respectively set on the left sidewall, right sidewall and roof of the roadway; Image acquisition: Images of the feature points are acquired from different angles through the deformation monitoring binocular cameras; Coordinate solution: According to the parallax of the feature points in the images acquired by the deformation monitoring binocular cameras, the three-dimensional space coordinates of the feature points in the camera coordinate system are calculated in real time; Deformation amount calculation: According to the change amount of the three-dimensional space coordinates of the feature points, the deformation condition of the roadway surface is calculated in real time.

12. The roadway deformation monitoring method during tunneling according to claim 11, wherein In the step of setting feature points: The feature points are reflective auxiliary feature points, and multiple cameras include fill lights for supplementary lighting. The reflective auxiliary feature points can reflect the light emitted by the fill lights, so that the reflected light of the feature points is included in the images acquired by the multiple cameras; Or, the feature points are light-emitting auxiliary feature points, and the reflective auxiliary feature points can emit light, so that the emitted light of the feature points is included in the images acquired by the multiple cameras.

13. The roadway deformation monitoring method during excavation according to claim 11 or 12, characterized in that In the step of setting feature points: The feature points include a first feature point, a second feature point and a third feature point. The first feature point is set on the roof of the roadway, the second feature point is set on the left sidewall of the roadway, and the third feature point is set on the right sidewall of the roadway opposite to the left sidewall. The second feature point and the third feature point are opposite in the width direction of the roadway.

14. The roadway deformation monitoring method during tunneling according to claim 11, characterized in that, In the image acquisition step: The deformation monitoring binocular cameras include a first camera and a second camera that are spaced apart in the horizontal direction. The first camera is close to the left sidewall of the roadway, and the second camera is close to the right sidewall of the roadway.

15. The roadway deformation monitoring method during tunneling according to claim 14, wherein Before the image acquisition step, there is also an adjustment step for the deformation monitoring binocular cameras, including: Starting the first camera and the second camera; The first camera and the second camera respectively acquire images of the feature points in the roadway from different angles in real time, and the acquired images include at least one feature point respectively set on the left sidewall, right sidewall and roof of the roadway; Adjusting the positions of the first camera and the second camera in the horizontal direction; Calculating the horizontal position of the feature points in the image; Judging whether the horizontal position of the feature points in the image is within a first preset position range. If the horizontal position of the feature points in the image is not within the first preset position range, continue to adjust the positions of the first camera and the second camera in the horizontal direction until the horizontal position of the feature points in the image is within the first preset position range; When the horizontal position of the feature points in the image is within the first preset position range, adjusting the pitching angles of the first camera and the second camera; Calculating the vertical position of the feature points in the image; Judging whether the vertical position of the feature points in the image is within a second preset position range. If the vertical position of the feature points in the image is not within the second preset position range, continue to adjust the pitching angles of the first camera and the second camera until the vertical position of the feature points in the image is within the second preset position range.

16. The roadway deformation monitoring method during tunneling according to claim 14, characterized in that Before the image acquisition step, there is also an adjustment step for the deformation monitoring binocular camera, including: Start the first camera and the second camera; The first camera and the second camera respectively collect images of feature points in the roadway from different angles, and the collected images include at least one feature point respectively arranged on the left sidewall, right sidewall and roof of the roadway; Adjust the pitching angles of the first camera and the second camera; Calculate the longitudinal position of the feature point in the image; Judge whether the longitudinal position of the feature point in the image is within the third preset position range. If the longitudinal position of the feature point in the image is not within the third preset position range, continue to adjust the pitching angles of the first camera and the second camera until the longitudinal position of the feature point in the image is within the third preset position range; When the longitudinal position of the feature point in the image is within the third preset position range, adjust the positions of the first camera and the second camera in the horizontal direction; Calculate the lateral position of the feature point in the image; Judge whether the lateral position of the feature point in the image is within the fourth preset position range. If the lateral position of the feature point in the image is not within the fourth preset position range, continue to adjust the positions of the first camera and the second camera in the horizontal direction until the lateral position of the feature point in the image is within the fourth preset position range.

17. The roadway deformation monitoring method during tunneling according to claim 15 or 16, characterized in that, In the step of adjusting the positions of the first camera and the second camera in the horizontal direction, the closer the distance between the feature points is, the smaller the spacing between the first camera and the second camera in the horizontal direction is adjusted. The farther the spacing between the feature points is, the larger the spacing between the first camera and the second camera in the horizontal direction is adjusted.

18. The roadway deformation monitoring method during tunneling according to claim 11, characterized in that There are multiple groups of the deformation monitoring binocular cameras, and the multiple groups of the deformation monitoring binocular cameras take pictures from different angles.

19. The roadway deformation monitoring method during excavation according to claim 11, characterized in that, The coordinate solving step specifically includes: Feature point extraction: Extract feature points in the image, and calculate the two-dimensional centroid coordinates of the highlighted circular features in the image through a visual recognition algorithm; Feature point three-dimensional coordinate solving: According to the pre-calibrated external parameter information of the deformation monitoring binocular camera, using the principle of triangulation, calculate the three-dimensional space coordinates of the feature point in the camera coordinate system according to the different disparities of the same feature point in different cameras of the deformation monitoring binocular camera; Top and sidewall feature classification: Classify the feature points with actually solved three-dimensional coordinates, and judge whether the feature point is on the left sidewall, right sidewall or roof of the roadway.

20. The roadway deformation monitoring method during tunneling according to claim 19, wherein, Before feature point extraction in the coordinate solving step, there is also: Image correction. The deformation monitoring binocular camera transmits the collected image to the computer. The computer obtains the image source file and performs format conversion, converts the image information into matrix data, and the computer corrects the image into a non-distorted image according to the pre-calibrated distortion parameters; Feature point extraction is based on the feature points in the corrected image.

21. The roadway deformation monitoring method during tunneling according to claim 19, wherein After the feature point extraction step, there is also: Feature point matching: Feature points in multiple images collected by the binocular cameras for deformation monitoring are matched according to their position and shape features to ensure that the matched feature points correspond to the same feature point in space.

22. The roadway deformation monitoring method during excavation according to claim 11, wherein In the deformation amount calculation step: Connect the feature points on the left sidewall of the roadway and the feature points on the right sidewall. The change amount of the vertical distance from the feature points on the roof of the roadway to the connection line is used as the roof subsidence amount, and the change amount of the length of the connection line is used as the approach amount of the two sidewalls.

Images