A tunnel excavation state detection method, device, equipment and medium

Abstract

This invention discloses a method, apparatus, equipment, and medium for detecting tunnel excavation status. The method includes: during tunnel excavation, acquiring in real-time the longitudinal movement trajectory of the excavating equipment, surface optical information, and visual information of the current cross-section; determining whether the current cross-section is a target cross-section based on the longitudinal movement trajectory of the excavating equipment, the surface optical information, and the visual information of the current cross-section; when the current cross-section is a target cross-section, generating a measured tunnel cross-section profile of the target cross-section based on the distance between the current cross-section and the excavating equipment; and determining the lateral excavation status of the current cross-section based on the difference between the measured tunnel cross-section profile and the desired tunnel cross-section profile. This invention can improve the accuracy of detecting tunnel excavation status and ensure the safety of subsequent construction.

Description

Technical Field

[0001] This invention relates to the field of tunnel engineering inspection, and in particular to a method, apparatus, equipment and medium for detecting the excavation status of tunnels. Background Technology

[0002] Tunnel cross-sectional profile refers to the outline of the tunnel excavation section, also known as the tunnel profile line under different working conditions. Tunnel cross-sectional profile inspection is used to evaluate the quality of tunnel excavation, including the regularity of the excavation section and the amount of over-excavation and under-excavation. Tunnel cross-sectional profile inspection is an important method for detecting the tunnel excavation status.

[0003] In tunnel cross-section profile inspection, target sections are typically selected based on the excavation distance to generate the tunnel cross-section profile for inspection. Selecting target sections at a fixed excavation distance essentially uses a small, discrete number of sections to represent the continuously changing tunnel profile, resulting in a complete blind spot between the two target sections. During actual excavation, changes in geological structure and tunnel damage and deformation caused by blasting occur. These anomalies may happen to occur precisely between the two selected sections and be completely missed. This leads to inaccurate detection of the tunnel excavation status, masks local defects, and leaves safety hazards for subsequent construction. Summary of the Invention

[0004] This invention provides a method, apparatus, equipment, and medium for detecting the excavation status of tunnels, which can improve the accuracy of detecting the excavation status of tunnels and ensure the safety of subsequent construction.

[0005] According to one aspect of the present invention, an embodiment of the present invention provides a method for detecting the state of tunnel excavation, the method comprising: During the tunnel excavation process, the longitudinal movement trajectory of the excavation equipment, as well as the surface optical and visual information of the current cross-section, are acquired in real time. Based on the longitudinal movement trajectory of the excavating equipment, the surface optical information and visual information of the current cross-section, determine whether the current cross-section is the target cross-section; When the current section is the target section, the measured tunnel cross-section profile of the target section is generated based on the distance between the current section and the excavation equipment. The lateral excavation status of the current section is determined based on the difference between the measured tunnel cross-section profile and the desired tunnel cross-section profile.

[0006] According to another aspect of the present invention, embodiments of the present invention also provide a tunnel excavation status detection device, the device comprising: The information acquisition module is used to acquire the longitudinal movement trajectory of the excavation equipment, the surface optical information and visual information of the current section in real time during the tunnel excavation process; The section selection module is used to determine whether the current section is the target section based on the longitudinal movement trajectory of the excavating equipment, the surface optical information and visual information of the current section; The contour construction module is used to generate the measured tunnel cross-section contour of the target cross-section based on the distance between the current cross-section and the excavation equipment when the current cross-section is the target cross-section. The status determination module is used to determine the lateral excavation status of the current section based on the difference between the measured tunnel cross-section profile and the desired tunnel cross-section profile.

[0007] According to another aspect of the present invention, embodiments of the present invention also provide a tunnel excavation status detection device, the tunnel excavation status detection device comprising: At least one processor; and A memory that is communicatively connected to at least one processor; wherein, The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to perform the tunnel excavation status detection method according to any embodiment of the present invention.

[0008] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions for causing a processor to execute and implement the tunnel excavation status detection method of any embodiment of the present invention.

[0009] According to another aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program that, when executed by a processor, implements the tunnel excavation status detection method according to any embodiment of the present invention.

[0010] The technical solution of this invention determines the target cross-section by analyzing the movement trajectory, surface optical information, and visual information of the excavating equipment during the tunnel excavation process. This effectively avoids blind spots in cross-section detection, prevents missed detections due to geological changes and blasting deformation, and generates the measured tunnel cross-section contour of the target cross-section based on distance information. By comparing this contour with the desired tunnel cross-section contour, the current excavation status of the cross-section is obtained, improving the accuracy of tunnel excavation status detection and ensuring the safety of subsequent construction.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart of a tunnel excavation status detection method according to an embodiment of the present invention; Figure 2 This is a flowchart of a tunnel excavation status detection method according to an embodiment of the present invention; Figure 3 This is a structural diagram of a tunnel excavation status detection device according to an embodiment of the present invention; Figure 4 This is a structural schematic diagram of a tunnel excavation status detection device provided in an embodiment of the present invention. Detailed Implementation

[0014] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0015] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0016] The acquisition, storage, and application of surface optical information and visual information involved in the technical solutions of this invention comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0017] Figure 1This is a flowchart illustrating a tunnel excavation status detection method provided by an embodiment of the present invention. This embodiment is applicable to detecting the excavation status of a tunnel currently being excavated. The method can be executed by a tunnel excavation status detection device, which can be implemented in hardware and / or software. This tunnel excavation status detection device can be configured in a server.

[0018] See Figure 1 The tunnel excavation status detection method shown includes: S101. During the tunnel excavation process, the longitudinal movement trajectory of the excavation equipment, as well as the surface optical and visual information of the current section, are acquired in real time.

[0019] Among these, surface optical information can be the characteristic information presented by the reflection or scattering of the object's surface through optical means. Visual information can be intuitively perceived information that can be captured and recognized visually. Excavation equipment can be engineering machinery used for tunnel excavation operations. The current cross-section can be the transverse section under the current working conditions. The longitudinal movement trajectory can be the path trajectory formed by movement in the depth direction.

[0020] During tunnel excavation, data from the inertial measurement unit (IMU) of the excavating equipment is acquired in real time. For example, the acceleration and angular velocity of the excavating equipment during operation are acquired. This data is then used to calculate the motion state and attitude of the excavating equipment, generating its longitudinal trajectory.

[0021] In some embodiments, the method for obtaining the longitudinal movement trajectory of the excavating equipment is as follows: The absolute distances from the excavation equipment to the cross-section and tunnel wall are measured using a direct time-of-flight sensor. Left and right images are acquired using a binocular vision camera and matched to obtain the relative positions of points in three-dimensional space. The inertial measurement unit (IMU) obtains the relative motion between adjacent time points through pre-integration, including changes in position, velocity, and attitude. A factor graph is constructed based on the aforementioned absolute distances, relative positions, and relative motions. The position and orientation of the excavation equipment at each time point are determined based on the factor graph. Furthermore, the longitudinal trajectory of the excavation equipment is determined.

[0022] The surface optical information of the current cross-section can be acquired by a sensor. In some embodiments, the surface optical information can be the reflection intensity. Reflection intensity refers to the intensity of light reflected back after incident light hits the surface of an object. Generally, reflection intensity can be acquired in laser rangefinders. Visual information can be image information, i.e., an image of the current cross-section.

[0023] In some embodiments, the surface optical and visual information acquisition devices and the inertial measurement unit are all installed in the excavation equipment. In other embodiments, the surface optical and visual information acquisition devices and the inertial measurement unit are integrated into a handheld inspection terminal.

[0024] S102. Based on the longitudinal movement trajectory of the excavating equipment, the surface optical information and visual information of the current cross-section, determine whether the current cross-section is the target cross-section.

[0025] The target cross-section can be a measured cross-section used for comparison with the cross-section in the design drawings. Understandably, during tunnel excavation, it is necessary to frequently compare the actual excavation site with the design drawings to check whether the construction is proceeding according to the design. Cross-section inspection is a commonly used method for this. The target cross-section is the selected cross-section used for comparison with the design drawings.

[0026] The data collection interval can be determined based on the designed length of the tunnel. When the excavation reaches the interval point, the target cross-section is selected. For example, if the data collection interval is set to 5, a target cross-section is determined when the movement trajectory is 5 meters and another target cross-section is determined when the movement trajectory is 10 meters.

[0027] In addition, the target cross-section needs to be determined based on surface optical and visual information. If surface optical information indicates a change in the material or structure of the current cross-section, then the current cross-section can be identified as the target cross-section. If visual information indicates potential deformation or damage to the steel frame of the current cross-section, then the current cross-section can be identified as the target cross-section. If the movement trajectory indicates a sudden change in the tunnel's orientation at the current cross-section, then the current cross-section can be identified as the target cross-section.

[0028] S103. When the current cross-section is the target cross-section, the measured tunnel cross-section profile of the target cross-section is generated based on the distance between the current cross-section and the excavation equipment.

[0029] Among them, the tunnel contour is the shape profile of the tunnel cross-section. The measured tunnel contour is the shape profile of the tunnel cross-section obtained by measurement.

[0030] If the current cross-section is the target cross-section, the measured tunnel cross-section profile of the current cross-section is generated. If the current cross-section is not the target cross-section, the step of generating the measured tunnel cross-section profile of the current cross-section is skipped, and the target cross-section is determined again as the tunnel excavation operation progresses and the longitudinal movement trajectory, surface optical information, and visual information of the excavation equipment are acquired.

[0031] The measured distance between the current cross-section and the excavation equipment is the distance measured by the distance sensor between the current cross-section and the sensor. Generally, the data acquisition equipment includes a distance sensor installed in the excavation equipment. Based on the measured distance and the excavation equipment's pose at that moment, the planar position of each measuring point on the current cross-section is calculated; by fitting all the measuring points, the measured tunnel cross-section profile of the target cross-section can be obtained.

[0032] In an optional embodiment, generating the measured tunnel cross-section profile of the target cross-section based on the distance between the current cross-section and the excavation equipment includes: acquiring a laser point cloud and at least one multi-view image of the current cross-section; mapping the laser point cloud to a target visual coordinate system to obtain the distance between the corresponding pixel of the laser point cloud and the excavation equipment in the target visual coordinate system, and using this distance as the depth of the corresponding pixel of the laser point cloud; calculating the depth of each pixel in the current cross-section based on the depth of the corresponding pixel of the laser point cloud and each of the multi-view images; and generating the measured tunnel cross-section profile of the target cross-section based on the depth of each pixel in the current cross-section and the laser point cloud.

[0033] Among them, the laser point cloud is a collection of massive discrete measurement point ranging information obtained by the lidar emitting laser light towards the current cross-section and calculating the reflected echo. The multi-view vision image can be a cross-section image captured by multiple vision cameras. The target vision coordinate system can be the coordinate system of a certain multi-view vision image.

[0034] In some embodiments, laser point clouds are acquired using a direct time-of-flight sensor, and two multi-view images, referred to as the left-eye visual image and the right-eye visual image, are acquired using a binocular camera.

[0035] After obtaining the laser point cloud and multi-view visual images, the multi-view visual images are first converted into depth images using the laser point cloud as an aid. Then, the measured tunnel cross-section contour is generated based on the laser point cloud and depth image. The depth image can be an image describing the depth of each pixel in the multi-view visual image. Depth can refer to the distance between the multi-view visual image acquisition device and the actual physical location corresponding to the pixel.

[0036] In engineering, the distance between the actual physical locations of each pixel in a multi-view vision image can be calculated based on the coordinate differences between pixels describing the same physical location in multiple multi-view vision images and the parameters of the acquisition device. For example, a binocular vision camera, mimicking the binocular parallax imaging mechanism of the human eye, uses two cameras with identical parameters and fixed positions to synchronously acquire images of the same target cross-section at the same time, obtaining a left-eye vision image and a right-eye vision image. By performing distortion correction and feature matching on the two vision images, the corresponding pixels of the same physical location in each vision image are found, resulting in pixel matching pairs. Using the camera's intrinsic and extrinsic parameters and the coordinate positions of the pixel matching pairs in the two vision images, the depth between the physical location and the camera is calculated. However, the depth obtained using this method is calculated based on multiple captured multi-view vision images, which can lead to inaccurate depth calculations due to inaccurate matching of pixels at the same physical location across multiple multi-view vision images.

[0037] After obtaining the laser point cloud, its coordinate system can be transformed to the target visual coordinate system based on the parameters of the laser point cloud acquisition device and the multi-view vision image acquisition device. For example, the laser point cloud can be transformed to the left-eye visual coordinate system. In this case, the measured points of the laser point cloud are mapped to pixels representing the same physical location in the left-eye visual image. The depth of this pixel includes not only the depth obtained by matching and calculating pixels in the left and right visual images, but also the depth of the laser point cloud. Combining these two depths yields a depth map. The depth map describes the depth of each pixel in the current cross-section. The depth measured by the laser point cloud is more accurate than the depth obtained by matching and calculating pixels in the left and right visual images. The depth map obtained by combining the two depths is more accurate.

[0038] It is evident that by integrating the high-precision ranging advantage of laser point clouds with the rich texture matching characteristics of multi-view visual images, the laser point cloud is mapped to the target visual coordinate system and the pixel depth is calibrated. Using this as a reference depth, the multi-view visual image is converted into a depth image, obtaining the depth of each pixel in the front section. This compensates for the sparse sampling of laser point clouds and makes the generated depth image more accurate, thereby improving the accuracy of the measured tunnel cross-section contour generated from the target cross-section.

[0039] In an optional embodiment, calculating the depth of each pixel in the current cross-section based on the depth of the pixel corresponding to the laser point cloud and each of the multi-view visual images includes: calculating the point cloud disparity value of the pixel corresponding to the laser point cloud based on the depth of the pixel corresponding to the laser point cloud and the acquisition device parameters of the acquisition device of the target visual coordinate system; determining the disparity matching range of each pixel in the target visual coordinate system in the multi-view visual image based on the point cloud disparity value; for each pixel in the target visual coordinate system, within the disparity matching range of the pixel in the multi-view visual image, detecting matching points at the same position as the cross-section represented by the pixel, forming a matching point group corresponding to the pixel; calculating the disparity value between the pixel and each matching point in the corresponding matching point group, and determining it as the disparity value corresponding to the pixel; converting the disparity value of each pixel in the target visual coordinate system into the depth value of each pixel in the target visual coordinate system based on the acquisition device parameters, to obtain the depth of each pixel in the current cross-section.

[0040] In this context, disparity refers to the lateral coordinate difference between the imaged positions of the same physical location in different multi-view visual images. Disparity reflects depth; a larger disparity value indicates a closer object and a smaller depth, while a smaller disparity value indicates a farther object and a greater depth. Point cloud disparity refers to the disparity value obtained by converting laser point clouds. The disparity matching range can be the range within which matching is performed using disparity values. A matching point group can be a set of pixels representing the same actual physical region in multiple multi-view visual images.

[0041] After mapping the laser point cloud to the target's visual coordinate system to obtain the depth of the corresponding pixels in the laser point cloud, the depth value can be converted into a disparity value based on the camera parameters. For example, the laser point cloud provides the precise depth of a physical location in the lidar coordinate system. Mapping this laser point cloud to the target's visual coordinate system yields the depth value of the corresponding pixel location. Then, using the camera's intrinsic parameters, the depth value is converted into a disparity value.

[0042] When performing pixel matching on multiple multi-view visual images, a semi-global matching (SGM) method can be used. In SGM, the point cloud disparity value can be used to narrow down the matching range.

[0043] For example, the known data includes: a left-eye visual image, a right-eye visual image, and point cloud disparity values ​​corresponding to several discrete pixels mapped from the laser point cloud to the left-eye visual coordinate system and converted. For a pixel in the left-eye visual image whose point cloud disparity value has been obtained, the row and column coordinates of the pixel are first determined in the left-eye visual image. According to the geometric constraints of binocular vision, the matching point of this pixel in the right-eye visual image must be located at the same row coordinate. Therefore, it is only necessary to determine the column coordinate of the matching pixel. Using the point cloud disparity value of this pixel, the search range of the column coordinates of the matching pixel is narrowed. Specifically, the point cloud disparity value is subtracted from the column coordinate of the pixel in the left-eye visual image, and the result is the estimated column coordinate of the matching point in the right-eye visual image. Then, with the estimated column coordinate as the center, a preset neighborhood radius is extended to the left and right to form a continuous column coordinate search range, which is the disparity matching range. Within the disparity matching range, for each candidate column coordinate, the matching cost between the pixel in the left visual image and the corresponding candidate pixel in the right visual image is calculated. The matching cost measures the similarity between two pixels. The matching costs are then aggregated along multiple directions to obtain the aggregated cost for each candidate pixel. The candidate pixel with the smallest aggregated cost is selected as the matched pixel in the right visual image. At this point, the matching between the pixel in the left visual image and the pixel at the corresponding column coordinate in the right visual image is complete. For pixels without point cloud disparity values, the conventional method of binocular semi-global matching is used, performing cost calculation, cost aggregation, and optimal selection within the disparity range from minimum to maximum disparity to complete the matching.

[0044] After finding the matching point of a pixel in the left visual image for a given pixel using a binocular semi-global matching method, the coordinate difference between these two pixels is calculated. Specifically, the column coordinates of the matching point in the right visual image are subtracted from the column coordinates of the pixel in the left visual image. The difference is the disparity value corresponding to that pixel. This process is repeated to obtain the disparity value distribution of all pixels in the entire image.

[0045] Using pre-calibrated acquisition device parameters, i.e. camera parameters, and based on the geometric principle that parallax is inversely proportional to depth, the parallax value of each pixel is converted into the actual distance from the physical position represented by that pixel to the camera plane, i.e., the depth value.

[0046] Optionally, when there are more than two multi-view images, the multi-view images are grouped in pairs and the above process is performed. This yields multiple depth maps describing the depth of each pixel in the current cross-section. These multiple depth maps are then weighted and fused to obtain the final depth map.

[0047] It is evident that by utilizing the depth obtained from laser point clouds and combining it with the parameters of the acquisition device to solve for the precise point cloud disparity value of each pixel, the disparity matching search range of each pixel in the multi-view visual image is limited, significantly narrowing the matching interval and reducing invalid matching operations. This effectively avoids matching errors caused by blind global matching, as well as the problems of large computational load and inaccurate solutions, thereby improving the accuracy and computational efficiency of depth calculation for all pixels in the current cross-section.

[0048] In an optional embodiment, generating the measured tunnel cross-sectional profile of the target cross-section based on the depth of each pixel in the current cross-section and the laser point cloud includes: calculating the first radial distance from the reference point of the current cross-section to the inner wall of the tunnel in multiple directions based on the pose of the excavating equipment and the depth of each pixel in the current cross-section, obtaining the first radial distance in each direction; calculating the second radial distance from the reference point in the current cross-section to the inner wall of the tunnel in each direction based on the pose of the excavating equipment and the laser point cloud, obtaining the second radial distance in each orientation; weightedly fusing the first radial distance and the second radial distance in the same direction to obtain the radial distance in that direction; and generating the measured tunnel cross-sectional profile of the target cross-section based on the radial distance in each direction.

[0049] The reference point in the current cross-section can be a fixed reference point used when generating the measured tunnel cross-sectional profile of the target cross-section. The first radial distance can be the radial distance calculated based on the depth of each pixel in the current cross-section. The second radial distance can be the radial distance obtained from laser point cloud computing. The radial distance refers to the distance from the reference point to the interior of the tunnel.

[0050] In a specific embodiment, the reference point can be selected as the intersection of the tunnel axis and the current cross-section. The tunnel axis can be obtained by using the longitudinal movement trajectory of the excavating equipment, fitting the main direction of the tunnel using the Random Sample Consensus (RANSAC) algorithm, and estimating the direction and position of the tunnel axis using the projection density variance minimization method.

[0051] When generating the measured tunnel cross-sectional profile of the target section, the radial distance from the same reference point on the current section to the tunnel inner wall is calculated based on two data sources, according to the current pose parameters of the excavating equipment: the first is based on the depth of each pixel on the current section, calculating the first radial distance of the reference point in multiple directions; the second is based on high-precision laser point cloud data, calculating the second radial distance of the reference point in multiple directions. Subsequently, for each direction, the first and second radial distances in the same direction are weighted and fused according to preset weights to obtain the radial distance in that direction. Optionally, the weights of the first and second radial distances are determined based on the confidence level of the acquisition equipment. For example, the weight of the second radial distance is set based on the confidence level of the laser ranging sensor. The weight of the first radial distance is set based on the confidence level of the binocular vision camera. Finally, the radial distances fused from all directions are connected in angular order to generate the measured tunnel cross-sectional profile of the current target section.

[0052] As can be seen, by calculating the first radial distance in each direction based on the pixel depth and the excavation equipment pose, and calculating the second radial distance in each direction based on the laser point cloud and the excavation equipment pose, and then weighting and fusing the two types of radial distances in the same direction, the system can take into account both the integrity of the global pixel depth of multi-view vision and the high precision advantage of laser point cloud ranging, while mitigating the measurement error and local missing problems of a single data source. This effectively improves the integrity and measurement accuracy of tunnel cross-section contour forming, and meets the application requirements for accurate reconstruction of tunnel cross-sections in complex excavation environments.

[0053] S104. Determine the lateral excavation status of the current section based on the difference between the measured tunnel cross-section profile and the desired tunnel cross-section profile.

[0054] Among them, the desired tunnel cross-sectional profile is the design profile, which is the ideal cross-sectional profile defined on the drawings. The lateral excavation state can be the excavation state of the lateral cross-section of the tunnel orientation.

[0055] In tunnel engineering, the desired tunnel cross-sectional profile is the ideal cross-section defined on the drawings, while the measured tunnel cross-sectional profile is the actual cross-sectional profile acquired by sensors. The coordinate systems of the two are often inconsistent. Therefore, it is necessary to align the coordinate systems of the measured tunnel cross-sectional profile with the desired tunnel cross-sectional profile before comparison.

[0056] Optionally, the measured tunnel cross-sectional profile can be aligned with the desired tunnel cross-sectional profile using the longitudinal movement trajectory of the excavating equipment, the axis information obtained from the longitudinal movement trajectory of the excavating equipment, and the spatial orientation information of the tunnel design axis.

[0057] Under the same coordinate system, calculate the difference between the measured tunnel cross-sectional profile and the expected tunnel cross-sectional profile, and determine the lateral excavation status of the current cross-section based on the difference.

[0058] In an optional embodiment, determining the lateral excavation state of the current section based on the difference between the measured tunnel cross-section profile and the desired tunnel cross-section profile includes: dividing the measured tunnel cross-section profile into multiple profile units; calculating the difference value between the profile unit and the corresponding desired profile for each profile unit to obtain the difference value of the profile unit; determining that the lateral excavation state on the profile unit is over-excavated when the difference value of the profile unit is positive; determining that the lateral excavation state on the profile unit is under-excavated when the difference value of the profile unit is negative; determining that the lateral excavation state on the profile unit is normal when the difference value of the profile unit is zero; and determining the lateral excavation state of each profile unit as the lateral excavation state of the current section.

[0059] The contour element can be a unit used to divide the measured tunnel cross-section contour. The desired contour can be a unit used to divide the desired tunnel cross-section contour.

[0060] The measured tunnel cross-sectional profile and the desired tunnel cross-sectional profile are divided into several profile units and several desired profiles using the same method. The profile units are compared with their corresponding desired profiles, and the differences are calculated.

[0061] Optionally, for ease of comparison and analysis, the measured tunnel cross-sectional profile and the desired tunnel cross-sectional profile can be simultaneously divided using equal angles or equal arc lengths. Specifically, taking the tunnel design center point as a reference, the entire measured tunnel cross-sectional profile is divided into multiple continuous profile units along the profile line at equal central angle intervals or equal arc lengths. The desired tunnel cross-sectional profile is then divided using a completely consistent method and node positions to obtain a desired profile that corresponds one-to-one with each profile unit.

[0062] Subtract the desired profile from the current profile unit to obtain the difference value. If the difference value is positive, it means that the current profile unit is larger than the desired profile, which can be understood as the actual excavation exceeding the design boundary. Therefore, the lateral excavation state of the current profile unit is over-excavation. If the difference value is negative, it means that the current profile unit is smaller than the desired profile, which can be understood as the actual excavation within the design boundary. Therefore, the lateral excavation state of the current profile unit is under-excavation. If the difference value is zero, it means that the current profile unit is equal to the desired profile, which can be understood as the actual excavation exactly on the design boundary. Therefore, the lateral excavation state of the current profile unit is normal.

[0063] In practice, warning thresholds are typically set for over-excavation. For example, the over-excavation warning threshold can be set to 150mm. When the difference between the outline unit and the desired outline is positive and does not exceed 150mm, no alarm is triggered. Alternatively, the under-excavation warning threshold can be set to 50mm. When the difference between the outline unit and the desired outline is negative and does not exceed 50mm, no alarm is triggered. Or, the over-excavation area alarm threshold can be set to 0.5m. 2 .

[0064] As can be seen, by dividing the measured tunnel cross-section profile into multiple independent profile units, calculating the difference value between each unit and the expected profile, and using positive and negative zero values ​​as the judgment criteria, the three transverse excavation states of over-excavation, under-excavation, and normal corresponding to each profile unit can be accurately identified. Then, the status results of all profile units are integrated to obtain the overall current transverse excavation status of the cross-section. This avoids the problem of ambiguous status judgment caused by general comparison of the overall profile, and can accurately locate the over-excavation and under-excavation areas at different locations of the cross-section, providing accurate data basis for real-time attitude adjustment of excavation equipment and correction of excavation path.

[0065] The technical solution of this invention determines the target cross-section by analyzing the movement trajectory, surface optical information, and visual information of the excavating equipment during the tunnel excavation process. This effectively avoids blind spots in cross-section detection, prevents missed detections due to geological changes and blasting deformation, and generates the measured tunnel cross-section contour of the target cross-section based on distance information. By comparing this contour with the desired tunnel cross-section contour, the current excavation status of the cross-section is obtained, improving the accuracy of tunnel excavation status detection and ensuring the safety of subsequent construction.

[0066] Figure 2 This is a flowchart of a tunnel excavation status detection method provided by an embodiment of the present invention. Based on the above embodiments, this embodiment of the present invention further defines the step of determining whether the current cross-section is a target cross-section based on the longitudinal movement trajectory of the excavating equipment, the surface optical information of the current cross-section, and visual information as follows: acquiring the laser reflection intensity of the current cross-section and using it as the surface optical information of the current cross-section; calculating the surface optical difference between the surface optical information of the current cross-section and the surface optical information of the previous cross-section, and determining the surface abrupt change detection result; acquiring multiple scene images of the current cross-section obtained based on multi-view vision acquisition; performing consistency detection on each scene image to obtain a consistency detection result; calculating the longitudinal trajectory curvature of the longitudinal movement trajectory and determining the trajectory detection result; and determining whether the current cross-section is a target cross-section based on the surface abrupt change detection result, the consistency detection result, and the trajectory detection result.

[0067] It should be noted that for parts not described in detail in the embodiments of the present invention, please refer to the descriptions in other embodiments.

[0068] See Figure 2 The tunnel excavation status detection method shown includes: S201. During the tunnel excavation process, the longitudinal movement trajectory of the excavation equipment, as well as the surface optical and visual information of the current section, are acquired in real time.

[0069] S202. Obtain the laser reflection intensity of the current cross-section and use it as the surface optical information of the current cross-section.

[0070] Among them, laser reflection intensity can refer to the intensity of laser light reflected back from the cross-section.

[0071] Laser reflection intensity can reflect the surface material of the current cross-section. Different materials reflect laser light at different intensities; therefore, laser reflection intensity can be used as surface optical information of the current cross-section.

[0072] In some embodiments, a laser point cloud is a massive set of discrete measurement points obtained by a lidar emitting laser light into a target scene and calculating the reflected echo, consisting of spatial three-dimensional coordinates and reflection intensity. The information in the laser point cloud includes the laser reflection intensity.

[0073] S203. Calculate the surface optical difference between the surface optical information of the current section and the surface optical information of the previous section, and determine the surface abrupt change detection result.

[0074] Here, "previous section" can refer to the section excavated at the previous moment. Surface abrupt change detection results can be obtained by detecting whether the surface material has changed.

[0075] Calculate the intensity difference between the laser reflection intensity of the current cross-section and the laser reflection intensity of the previous cross-section, and determine whether this intensity difference is greater than an intensity difference threshold. If the intensity difference is greater than the intensity difference threshold, it indicates that the material of the current cross-section has changed. If the intensity difference is not greater than the intensity difference threshold, it indicates that the material of the current cross-section is the same as the material of the previous cross-section.

[0076] For example, to detect abrupt changes in material properties such as pipe segment joints, the threshold for the average change in laser reflection intensity can be set to 15%. When the average change in laser reflection intensity is greater than 15%, it indicates that the material has changed.

[0077] S204. Acquire multiple scene images of the current cross section based on multi-view vision acquisition.

[0078] The scene image can be an image of the on-site environment acquired through multi-view vision.

[0079] Scene images are acquired using multi-view vision devices. For example, a binocular vision camera is used to acquire images of the surrounding environment, resulting in two scene images.

[0080] S205. Perform consistency detection on each of the scene images to obtain consistency detection results.

[0081] Consistency detection can refer to the detection of whether the same feature regions are consistent across multiple scene images.

[0082] If the tunnel lining surface is continuous, smooth, and has a uniform texture, then pixel areas representing the same physical location in different scene images will have similar gray levels and smooth parallax. In this case, the matching algorithm can reliably find the correct matching point with high confidence. However, if deformation or damage occurs in a certain cross-section of the tunnel, the geometric shape and optical properties of the surface at that point will be disrupted, resulting in abrupt parallax changes, texture loss, occlusion, or strong reflections. The matching algorithm will then be unable to find a correct match, leading to low confidence.

[0083] Therefore, feature matching can be performed on each scene image, and the matching confidence score of each region in the scene image can be obtained. If the matching confidence score is lower than the matching confidence score threshold, it indicates that the feature matching confidence is low and the features are inconsistent, and the consistency detection result for that region is failed; if the matching confidence score is not lower than the matching confidence score threshold, it indicates that the feature matching confidence is high and the features are consistent, and the consistency detection result for that region is passed.

[0084] Optionally, if there are regions where the consistency check result is not passed, then the consistency check result is determined to be not passed.

[0085] S206. Calculate the longitudinal trajectory curvature of the longitudinal motion trajectory and determine the trajectory detection result.

[0086] The longitudinal trajectory curvature is the curvature of the longitudinal motion trajectory. It is used to represent the tunnel's orientation. The trajectory detection result is obtained by detecting the tunnel's orientation.

[0087] When the trajectory detection result shows that the longitudinal trajectory curvature is greater than the preset curvature threshold, it indicates that the tunnel's orientation has changed significantly. When the trajectory detection result shows that the longitudinal trajectory curvature is not greater than the preset curvature threshold, it indicates that the tunnel's orientation is stable and there are no significant changes.

[0088] S207. Based on the surface mutation detection results, consistency detection results, and trajectory detection results, determine whether the current cross-section is the target cross-section.

[0089] Among them, the surface abrupt change detection result can reflect whether the surface material of the current section has undergone abrupt changes. The consistency detection result can reflect whether there is deformation or damage to the current section. The trajectory detection result can reflect whether there has been a significant change in the tunnel's orientation.

[0090] If the surface mutation detection result reflects a sudden change in the surface material of the current section, or the consistency detection result reflects deformation or damage to the current section, or the trajectory detection result reflects a significant change in the tunnel orientation, then the current section is determined as the target section.

[0091] In an optional embodiment, determining whether the current cross-section is a target cross-section based on the surface mutation detection result, consistency detection result, and trajectory detection result includes: determining the current cross-section as a feature cross-section when the surface mutation detection result indicates the presence of a mutation, the consistency detection result indicates failure, or the trajectory detection result indicates that the trajectory curvature is greater than a curvature threshold; calculating the distance interval between the position of the feature cross-section and the position of the previous target cross-section; determining the feature cross-section as a target cross-section when the distance interval is greater than an interval threshold; and determining that the current cross-section is not a target cross-section when the distance interval is less than or equal to the interval threshold.

[0092] Among them, the characteristic section can be a section with special circumstances.

[0093] To avoid selecting target sections too frequently, which could lead to strain on computing resources and reduced equipment efficiency, when the surface mutation detection results indicate a sudden change in the surface material of the current section, or the consistency detection results indicate deformation or damage to the current section, or the trajectory detection results indicate a significant change in the tunnel's orientation, the current section can be identified as the characteristic section.

[0094] Then, the distance between the feature section and the previous target section is calculated. If the distance is short, that is, less than or equal to the interval threshold, the feature section is not identified as a target section. If the distance is long, that is, greater than the interval threshold, the feature section is identified as a target section.

[0095] It is evident that by only designating a cross-section as a characteristic cross-section when there are sudden changes in cross-sectional material, deformation or damage, or significant changes in the tunnel's orientation, and by combining the distance between the characteristic cross-section and the previous target cross-section with the interval threshold for screening, it is possible to accurately capture key abnormal locations during tunnel excavation while avoiding the problem of excessive computing resources and reduced equipment operating efficiency caused by frequent selection of target cross-sections. This ensures that no changes in key operating conditions are missed while reasonably controlling the selection frequency of target cross-sections, thus achieving a balance between detection accuracy and computing resource consumption.

[0096] S208. When the current cross-section is the target cross-section, the measured tunnel cross-section profile of the target cross-section is generated based on the distance between the current cross-section and the excavation equipment.

[0097] S209. Determine the lateral excavation status of the current section based on the difference between the measured tunnel cross-section profile and the desired tunnel cross-section profile.

[0098] The technical solution of this invention uses laser reflection intensity as the surface optical information of the current cross-section. By comparing the differences in surface optical information of adjacent cross-sections, the surface abrupt change detection result is obtained. Consistency detection is performed on the scene images acquired by multi-view vision to obtain the corresponding detection result. At the same time, the trajectory curvature of the longitudinal movement trajectory of the excavating equipment is calculated to obtain the trajectory detection result. The three types of detection information are combined to participate in the determination of whether the current cross-section is the target cross-section. The comprehensive consideration is made from multiple aspects such as the surface material characteristics of the cross-section, the integrity of the cross-section structure, and the trend of the tunnel travel trajectory. This enriches the information dimensions of the target cross-section determination, effectively improves the scientificity and rationality of the target cross-section selection, and lays a reliable foundation for subsequent tunnel cross-section contour measurement and excavation status analysis.

[0099] Figure 3 This is a schematic diagram of a tunnel excavation status detection device provided in an embodiment of the present invention. The present invention is applicable to the detection of the excavation status of a tunnel under excavation. The device can execute a tunnel excavation status detection method and can be implemented in hardware and / or software.

[0100] See Figure 3 The tunnel excavation status detection device shown includes: The information acquisition module 301 is used to acquire the longitudinal movement trajectory of the excavation equipment, the surface optical information and visual information of the current section in real time during the tunnel excavation process. The section selection module 302 is used to determine whether the current section is a target section based on the longitudinal movement trajectory of the excavating equipment, the surface optical information and visual information of the current section; The contour construction module 303 is used to generate the measured tunnel cross-section contour of the target cross-section based on the distance between the current cross-section and the excavation equipment when the current cross-section is the target cross-section. The state determination module 304 is used to determine the lateral excavation state of the current section based on the difference between the measured tunnel cross-section profile and the desired tunnel cross-section profile.

[0101] The technical solution of this invention determines the target cross-section by analyzing the movement trajectory, surface optical information, and visual information of the excavating equipment during the tunnel excavation process. This effectively avoids blind spots in cross-section detection, prevents missed detections due to geological changes and blasting deformation, and generates the measured tunnel cross-section contour of the target cross-section based on distance information. By comparing this contour with the desired tunnel cross-section contour, the current excavation status of the cross-section is obtained, improving the accuracy of tunnel excavation status detection and ensuring the safety of subsequent construction.

[0102] Optionally, the section selection module 302 includes: A surface optical information determination unit is used to obtain the laser reflection intensity of the current cross-section and use it as the surface optical information of the current cross-section; The mutation detection unit is used to calculate the surface optical difference between the surface optical information of the current cross section and the surface optical information of the previous cross section, and to determine the surface mutation detection result. The scene image acquisition unit is used to acquire multiple scene images of the current section obtained based on multi-view vision acquisition; A consistency detection unit is used to perform consistency detection on each of the scene images and obtain a consistency detection result. The trajectory detection unit is used to calculate the longitudinal trajectory curvature of the longitudinal motion trajectory and determine the trajectory detection result; The cross-section determination unit is used to determine whether the current cross-section is the target cross-section based on the surface mutation detection result, consistency detection result, and trajectory detection result.

[0103] Optional, the section-determining element includes: The feature section determination unit is used to determine the current section as a feature section when the surface mutation detection result is that a mutation exists, the consistency detection result is that it fails, or the trajectory detection result is that the trajectory curvature is greater than the curvature threshold. An interval determination unit is used to calculate the distance interval between the position of the feature section and the position of the previous target section; The first interval determination unit is used to determine the feature section as the target section when the distance interval is greater than the interval threshold. The second interval determination unit is used to determine that the current cross section is not the target cross section when the distance interval is less than or equal to the interval threshold.

[0104] Optionally, the contour building module 303 includes: The distance information acquisition unit is used to acquire the laser point cloud and at least one multi-view visual image of the current cross section; A laser mapping unit is used to map the laser point cloud onto the target visual coordinate system to obtain the distance between the corresponding pixel of the laser point cloud and the excavation device in the target visual coordinate system, and use it as the depth of the corresponding pixel of the laser point cloud. The image depth calculation unit is used to calculate the depth of each pixel in the current cross-section based on the depth of the pixel corresponding to the laser point cloud and each of the multi-view vision images; The contour generation unit is used to generate the measured tunnel cross-sectional contour of the target cross-section based on the depth of each pixel in the current cross-section and the laser point cloud.

[0105] Optionally, the image depth calculation unit includes: The point cloud disparity calculation unit is used to calculate the point cloud disparity value of the pixel corresponding to the laser point cloud based on the depth of the pixel corresponding to the laser point cloud and the acquisition device parameters of the acquisition device of the target visual coordinate system. The matching range determination unit is used to determine the disparity matching range of each pixel in the target visual coordinate system in the multi-view visual image based on the point cloud disparity value. The feature matching unit is used to detect matching points at the same position as the cross section represented by the pixel point within the disparity matching range of the pixel point in the multi-view vision image for each pixel point in the target visual coordinate system, and form a matching point group corresponding to the pixel point. The disparity determination unit is used to calculate the disparity value between the pixel and each matching point in the corresponding matching point group, and determine it as the disparity value corresponding to the pixel. The pixel depth calculation unit is used to convert the disparity value of each pixel in the target visual coordinate system into the depth value of each pixel in the target visual coordinate system according to the parameters of the acquisition device, so as to obtain the depth of each pixel in the current section.

[0106] Optionally, the contour generation unit includes: The first radial distance determination unit is used to calculate the first radial distance from the reference point of the current section to the inner wall of the tunnel in multiple directions based on the pose of the excavation equipment and the depth of each pixel in the current section, and obtain the first radial distance in each direction. The second radial distance determination unit is used to calculate the second radial distance from the reference point in the current cross-section to the inner wall of the tunnel in each of the directions based on the pose of the excavation equipment and the laser point cloud, so as to obtain the second radial distance of each of the directions; A radial distance fusion unit is used to weightedly fuse the first radial distance and the second radial distance in the same direction to obtain the radial distance in the direction. The cross-section profile generation unit is used to generate the measured tunnel cross-section profile of the target cross-section based on the radial distance of each of the directions.

[0107] Optionally, the status determination module 304 includes: A contour division unit is used to divide the measured tunnel cross-section contour into multiple contour units. The difference calculation unit is used to calculate the difference value between the contour unit and the corresponding desired contour for each contour unit, and obtain the difference value of the contour unit; The first state determination unit is used to determine that the lateral excavation state on the contour unit is an over-excavation state when the difference value of the contour unit is positive. The second state determination unit is used to determine that the lateral excavation state on the contour unit is an under-excavation state when the difference value of the contour unit is negative. The third state determination unit is used to determine that the lateral excavation state on the contour unit is a normal state when the difference value of the contour unit is zero. The overall state determination unit is used to determine the lateral excavation state of each of the contour units as the lateral excavation state of the current section.

[0108] The tunnel excavation status detection device provided in this embodiment of the invention can execute the tunnel excavation status detection method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the tunnel excavation status detection method.

[0109] Figure 4 A schematic diagram of the structure of a tunnel excavation status detection device 400 that can be used to implement an embodiment of the present invention is shown.

[0110] like Figure 4 As shown, the tunnel excavation status detection device 400 includes at least one processor 401 and a memory, such as a read-only memory 402 or a random access memory 403, communicatively connected to the at least one processor 401. The memory stores computer programs executable by the at least one processor. The processor 401 can perform various appropriate actions and processes based on the computer program stored in the read-only memory 402 or loaded from the storage unit 408 into the random access memory 403. The random access memory 403 can also store various programs and data required for the operation of the tunnel excavation status detection device 400. The processor 401, read-only memory 402, and random access memory 403 are interconnected via a bus 404. An input / output interface 405 is also connected to the bus 404.

[0111] Multiple components in the tunnel excavation status monitoring device 400 are connected to an input / output interface 405, including: an input unit 406, such as a keyboard, mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a disk, optical disk, etc.; and a communication unit 409, such as a network card, modem, wireless transceiver, etc. The communication unit 409 allows the tunnel excavation status monitoring device 400 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0112] Processor 401 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 401 include, but are not limited to, central processing units, graphics processing units, various special-purpose artificial intelligence computing chips, various processors running machine learning model algorithms, digital signal processors, and any suitable processor, controller, microcontroller, etc. Processor 401 performs the various methods and processes described above, such as tunnel excavation status detection methods.

[0113] In some embodiments, the tunnel excavation status detection method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 408. In some embodiments, part or all of the computer program may be loaded and / or installed on the tunnel excavation status detection device 400 via read-only memory 402 and / or communication unit 409. When the computer program is loaded into random access memory 403 and executed by processor 401, one or more steps of the tunnel excavation status detection method described above may be performed. Alternatively, in other embodiments, processor 401 may be configured to perform the tunnel excavation status detection method by any other suitable means (e.g., by means of firmware).

[0114] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays, application-specific integrated circuits (ASICs), application-specific standard products (ASICs), systems-on-a-chip (SoCs), complex programmable logic devices, computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0115] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0116] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, flash memory, optical fiber, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0117] To provide interaction with the user, the systems and techniques described herein can be implemented on the operational monitoring equipment, which includes: a display device (e.g., a cathode ray tube or liquid crystal monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the tunnel excavation status monitoring equipment. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0118] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0119] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product within the cloud computing service system. This addresses the shortcomings of traditional physical hosts and virtual private servers, such as high management difficulty and weak business scalability.

[0120] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0121] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for detecting the excavation status of a tunnel, characterized in that, The method includes: During the tunnel excavation process, the longitudinal movement trajectory of the excavation equipment, as well as the surface optical and visual information of the current cross-section, are acquired in real time. Based on the longitudinal movement trajectory of the excavating equipment, the surface optical information and visual information of the current cross-section, determine whether the current cross-section is the target cross-section; When the current section is the target section, the measured tunnel cross-section profile of the target section is generated based on the distance between the current section and the excavation equipment. The lateral excavation status of the current section is determined based on the difference between the measured tunnel cross-section profile and the desired tunnel cross-section profile.

2. The method according to claim 1, characterized in that, The step of determining whether the current cross-section is the target cross-section based on the longitudinal movement trajectory of the excavating equipment, the surface optical information and visual information of the current cross-section includes: The laser reflection intensity of the current cross-section is obtained and used as the surface optical information of the current cross-section; Calculate the surface optical difference between the surface optical information of the current cross section and the surface optical information of the previous cross section, and determine the surface abrupt change detection result; Acquire multiple scene images of the current cross section based on multi-view vision acquisition; Perform consistency detection on each of the scene images to obtain consistency detection results; Calculate the longitudinal trajectory curvature of the longitudinal motion trajectory and determine the trajectory detection result; Based on the surface mutation detection results, consistency detection results, and trajectory detection results, it is determined whether the current cross-section is the target cross-section.

3. The method according to claim 2, characterized in that, The step of determining whether the current cross-section is the target cross-section based on the surface mutation detection results, consistency detection results, and trajectory detection results includes: When the surface mutation detection result is that a mutation exists, the consistency detection result is that it fails, or the trajectory detection result is that the trajectory curvature is greater than the curvature threshold, the current cross-section is determined as a feature cross-section; Calculate the distance interval between the position of the feature section and the position of the previous target section; When the distance interval is greater than the interval threshold, the feature section is determined as the target section; When the distance interval is less than or equal to the interval threshold, it is determined that the current cross section is not the target cross section.

4. The method according to claim 1, characterized in that, The step of generating the measured tunnel cross-sectional profile of the target cross-section based on the distance between the current cross-section and the excavation equipment includes: Acquire the laser point cloud and at least one multi-view visual image of the current cross-section; The laser point cloud is mapped onto the target visual coordinate system to obtain the distance between the corresponding pixel of the laser point cloud and the excavation equipment in the target visual coordinate system, and this distance is used as the depth of the corresponding pixel of the laser point cloud. Based on the depth of the pixels corresponding to the laser point cloud and each of the multi-view visual images, calculate the depth of each pixel in the current cross-section; Based on the depth of each pixel in the current cross-section and the laser point cloud, the measured tunnel cross-section profile of the target cross-section is generated.

5. The method according to claim 4, characterized in that, The step of calculating the depth of each pixel in the current cross-section based on the depth of the corresponding pixel in the laser point cloud and each of the multi-view visual images includes: The point cloud disparity value of the pixel corresponding to the laser point cloud is calculated based on the depth of the pixel corresponding to the laser point cloud and the acquisition device parameters of the acquisition device of the target visual coordinate system. Based on the point cloud disparity values, determine the disparity matching range of each pixel in the target visual coordinate system in the multi-view visual image; For each pixel in the target visual coordinate system, within the disparity matching range of the pixel in the multi-view visual image, a matching point at the same position as the cross section represented by the pixel is detected to form a matching point group corresponding to the pixel. Calculate the disparity value between the pixel and each matching point in the corresponding matching point group, and determine it as the disparity value corresponding to the pixel; Based on the parameters of the acquisition device, the disparity value of each pixel in the target visual coordinate system is converted into the depth value of each pixel in the target visual coordinate system to obtain the depth of each pixel in the current cross-section.

6. The method according to claim 4, characterized in that, The step of generating the measured tunnel cross-sectional profile of the target cross-section based on the depth of each pixel in the current cross-section and the laser point cloud includes: Based on the pose of the excavation equipment and the depth of each pixel in the current cross-section, the first radial distance from the reference point of the current cross-section to the inner wall of the tunnel in multiple directions is calculated, and the first radial distance in each direction is obtained. Based on the pose of the excavation equipment and the laser point cloud, the second radial distance from the reference point in the current cross-section to the inner wall of the tunnel in each of the directions is calculated to obtain the second radial distance of each of the directions. The first radial distance and the second radial distance in the same direction are weighted and fused to obtain the radial distance in the direction. Based on the radial distances in each of the aforementioned directions, the measured tunnel cross-sectional profile of the target section is generated.

7. The method according to claim 1, characterized in that, Determining the lateral excavation state of the current tunnel section based on the difference between the measured tunnel cross-section profile and the desired tunnel cross-section profile includes: The measured tunnel cross-sectional profile is divided into multiple profile units; For each of the contour units, the difference value between the contour unit and the corresponding desired contour is calculated to obtain the difference value of the contour unit; When the difference value of the contour unit is positive, the lateral excavation state on the contour unit is determined to be an over-excavation state. When the difference value of the contour unit is negative, the lateral excavation state on the contour unit is determined to be under-excavation state. When the difference value of the contour unit is zero, the lateral excavation state on the contour unit is determined to be a normal state. The lateral excavation state of each of the contour units is determined as the lateral excavation state of the current section.

8. A tunnel excavation status detection device, characterized in that, The device includes: The information acquisition module is used to acquire the longitudinal movement trajectory of the excavation equipment, the surface optical information and visual information of the current section in real time during the tunnel excavation process; The section selection module is used to determine whether the current section is the target section based on the longitudinal movement trajectory of the excavating equipment, the surface optical information and visual information of the current section; The contour construction module is used to generate the measured tunnel cross-section contour of the target cross-section based on the distance between the current cross-section and the excavation equipment when the current cross-section is the target cross-section. The status determination module is used to determine the lateral excavation status of the current section based on the difference between the measured tunnel cross-section profile and the desired tunnel cross-section profile.

9. A tunnel excavation status detection device, characterized in that, The tunnel excavation status detection equipment includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the tunnel excavation status detection method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the tunnel excavation status detection method according to any one of claims 1-7.

Images