Tunnel surrounding rock water seepage precision measurement method and system

Abstract

This invention provides a precise method for measuring seepage in tunnel surrounding rock, relating to the field of tunnel safety monitoring and geological disaster prevention technology. It addresses existing technical problems such as large subjective errors in manual identification of seepage at the tunnel face, low accuracy of infrared image recognition, and inaccurate spatial positioning of seepage areas. This invention employs steps including multimodal sensing data acquisition, surrounding rock feature map creation, infrared image processing and seepage area extraction, image geographic reference establishment, precise seepage measurement, and result output alarms. It combines infrared image enhancement, lightweight convolutional neural network segmentation, and spatial projection correction technologies to achieve automatic identification, spatial visualization, and quantitative measurement of seepage areas. This invention boasts high identification accuracy, precise spatial positioning, high device integration, and strong mobility, making it suitable for rapid deployment and real-time monitoring at tunnel construction sites, providing an efficient and objective technical means for tunnel construction safety monitoring.

Description

Technical Field

[0001] This invention relates to the field of tunnel safety monitoring and geological disaster prevention and control technology, and in particular to a method and system for accurate measurement of seepage in tunnel surrounding rock. Background Technology

[0002] During tunnel construction, seepage at the tunnel face is a core characteristic reflecting the integrity of the surrounding rock structure and the state of groundwater activity. Its detection and identification results directly determine the rationality of tunnel construction safety control strategies and drainage engineering design schemes. Existing methods for detecting seepage at tunnel faces have many technical shortcomings, making it difficult to meet the high-precision, automated monitoring requirements of actual engineering projects. Specific shortcomings are as follows:

[0003] 1. Traditional inspection relies on manual inspection and visual observation. The inspection results are greatly affected by the operator's experience and subjective judgment, resulting in low accuracy. Moreover, in the complex environment of tunnels with high humidity, high temperature and insufficient light, it is easy to miss or misjudge water seepage areas.

[0004] 2. Although infrared thermal imaging technology has been applied to seepage detection, the temperature difference at the tunnel face is small, and infrared images have problems such as low resolution and poor signal-to-noise ratio. The temperature response of the seepage area is close to that of the background, and the automatic recognition accuracy of traditional threshold or edge detection methods is insufficient.

[0005] 3. The application results of deep learning semantic segmentation models in the field of natural imagery are difficult to directly transfer to infrared data, and the scene annotation data for tunnel seepage detection is scarce, resulting in high model training and deployment costs and poor generalization ability.

[0006] 4. Existing detection methods neglect the fusion of infrared images and three-dimensional geometric information, making it impossible to accurately project the seepage identification results onto the three-dimensional spatial model of the tunnel. This makes it difficult to achieve precise spatial positioning and quantitative analysis of the seepage area, and thus cannot provide comprehensive and accurate technical support for engineering decisions.

[0007] Therefore, there is an urgent need for a precise method and system for measuring seepage in tunnel surrounding rock to solve the existing technical problems. Summary of the Invention

[0008] The present invention aims to solve at least one of the technical problems existing in the prior art, and proposes a method and system for accurate measurement of seepage in tunnel surrounding rock.

[0009] In a first aspect, embodiments of the present invention provide a method for accurately measuring seepage in tunnel surrounding rock, comprising:

[0010] Multimodal sensing data of the surrounding rock of the tunnel is collected, including infrared data, visible light data and laser point cloud data;

[0011] The multimodal sensing data are fused to establish a tunnel surrounding rock feature map data;

[0012] Infrared data of the surrounding rock of the tunnel are processed to extract the seepage area;

[0013] Based on the tunnel surrounding rock feature map data, an image geographic reference of infrared data and point cloud data is established for the seepage area;

[0014] Based on the image georeferenced data of the infrared data and point cloud data, the seepage is accurately measured, the measurement results are output, and an alarm is generated according to relevant standards.

[0015] Furthermore, a handheld SLAM device integrating multiple sensors is used to collect multimodal perception data of the tunnel surrounding rock. The specific steps include:

[0016] Infrared camera, visible light camera, lidar, RTK positioning module and interactive touch module are installed on the body of mobile handheld measuring device to ensure that the fields of view of each sensor overlap and the lidar field of view is unobstructed.

[0017] The sensor calibration and time synchronization are performed to complete the calibration of the intrinsic and extrinsic parameters of each sensor, and to achieve timestamp alignment and coordinate system unification among multiple sensors;

[0018] The system collects and fuses multimodal data, including infrared image data, visible light image data, and three-dimensional laser point cloud data of the tunnel face.

[0019] Furthermore, the multimodal sensing data is fused to establish tunnel surrounding rock feature map data. Specific steps include:

[0020] A three-dimensional point cloud map of the tunnel surrounding rock was constructed, and a laser-inertial fusion SLAM mapping algorithm was used to complete the three-dimensional spatial reconstruction of the tunnel.

[0021] Save the acquisition trajectory and pose file, and record the timestamps and corresponding pose parameters of the device acquisition process;

[0022] The data is fused and a tunnel surrounding rock feature map is obtained. The point cloud and camera image data are aligned according to the timestamp, and the image is projected onto the point cloud coordinate system to achieve multimodal data fusion, generating a surrounding rock feature map that includes geometric shape, surface texture and thermal radiation distribution.

[0023] Further, the infrared data of the tunnel surrounding rock is processed to extract the seepage area. Specific steps include:

[0024] The infrared image sequence was preprocessed, and noise reduction was achieved by a combination of median filtering and bilateral filtering. The image was enhanced by limiting contrast adaptive histogram equalization, and the preprocessed image was obtained through pseudo-color mapping and resolution enhancement.

[0025] Edge feature enhancement and multi-channel feature map construction are performed on infrared images. The Laplacian operator and Sobel operator are combined to extract image gradient features, calculate local statistical features, and construct a multi-channel feature tensor.

[0026] Preliminary detection of seepage candidate regions is performed, and the multi-channel feature map is input into a lightweight convolutional neural network segmentation model to output a seepage probability map.

[0027] Temporal consistency screening and result refinement are performed. Temporal consistency analysis is conducted on the continuous frame water seepage probability map. The final water seepage area mask is obtained by combining morphological operations or conditional random field processing.

[0028] Furthermore, bilateral filtering is used for joint noise reduction. The specific methods include: suppressing noise by weighting the spatial domain and the pixel value domain, limiting contrast adaptive histogram equalization to calculate the cumulative distribution function for each local block to achieve image enhancement, and using bicubic interpolation or a super-resolution algorithm based on deep learning to improve resolution.

[0029] Furthermore, the multi-channel feature tensor is constructed by combining the infrared image preprocessing results, the second-order gradient features of the Laplacian operator, the first-order gradient features of the Sobel operator, the local mean features, and the local variance features.

[0030] Furthermore, the lightweight convolutional neural network segmentation model is a DeepLabV3+ network based on MobileNet or ShuffleNet backbone, which learns the difference between the seepage area and the background area through pixel-level binary classification.

[0031] Furthermore, based on the tunnel surrounding rock feature map data, an image georeference of infrared data and point cloud data is established for the seepage area. The specific steps include: based on the extrinsic parameter transformation matrix of the infrared camera and lidar, establishing the projection relationship between the pixel points and the collected point cloud, and mapping the pixel-level seepage area of ​​the seepage area mask onto the point cloud model or three-dimensional mesh model.

[0032] Furthermore, based on the image georeferenced representation of the infrared data and point cloud data, precise water seepage measurement is performed, specifically including the following steps:

[0033] The seepage area mask is processed, and a complete and continuous seepage distribution area is obtained through morphological closing operation, connected component filtering, and merging of adjacent small regions.

[0034] Within the connected domain, quantitative parameters and seepage intensity estimation classification are calculated. Quantitative parameters such as area, perimeter, and shape coefficient of seepage patches are calculated, and seepage intensity classification is completed in combination with preset standards.

[0035] Output pixel-level water seepage mask, geographic annotation results, and statistical reports including water seepage area, intensity level, and spatial distribution;

[0036] The system monitors and warns of water seepage. When the intensity or area of ​​seepage exceeds a preset threshold, it automatically generates alarm information indicating the location and severity of the abnormal area, outputs the measurement results, and generates alarms according to relevant standards.

[0037] Secondly, embodiments of the present invention provide a precise measurement system for seepage in tunnel surrounding rock, comprising a multimodal sensing data acquisition and processing unit, a tunnel surrounding rock feature map data establishment unit, a deep-water area extraction unit, an image geographic reference establishment unit, and a precise seepage measurement unit; wherein:

[0038] A multimodal sensing data acquisition and processing unit is used to acquire multimodal sensing data of the tunnel surrounding rock, including infrared data, visible light data and laser point cloud data.

[0039] The tunnel surrounding rock feature map data establishment unit is used to fuse the multimodal sensing data to establish tunnel surrounding rock feature map data;

[0040] The deep-water area extraction unit is used to process infrared data of the tunnel surrounding rock to complete the extraction of seepage areas;

[0041] The image georeference establishment unit is used to establish an image georeference of infrared data and point cloud data for the seepage area based on the tunnel surrounding rock feature map data.

[0042] The seepage precision measurement unit is used to perform precise seepage measurement based on the image georeference of the infrared data and point cloud data, output the measurement results, and generate alarms according to relevant standards.

[0043] This invention provides a precise method for measuring seepage in tunnel surrounding rock, addressing the technical problems of large subjective errors in manual identification of seepage detection at the tunnel face, low accuracy of infrared image recognition, and inaccurate spatial positioning of seepage areas. This invention achieves automatic identification, spatial visualization, and quantitative measurement of seepage areas through steps including multimodal sensing data acquisition, surrounding rock feature map creation, infrared image processing and seepage area extraction, image geographic reference establishment, precise seepage measurement, and result output alarms. It combines infrared image enhancement, lightweight convolutional neural network segmentation, and spatial projection correction technologies. This invention boasts high identification accuracy, precise spatial positioning, high device integration, and strong mobility, making it suitable for rapid deployment and real-time monitoring at tunnel construction sites, providing an efficient and objective technical means for tunnel construction safety monitoring.

[0044] Compared with the prior art, the present invention has the following significant advantages:

[0045] 1. This invention enables automated identification and location of seepage areas, replacing traditional manual visual inspection, completely avoiding subjective errors in manual inspection, improving the objectivity and reliability of inspection results, and significantly increasing inspection efficiency;

[0046] 2. This invention proposes a complete infrared image processing workflow. By combining noise reduction, image enhancement, and edge feature extraction to construct a multi-channel feature map, and combining it with a lightweight convolutional neural network segmentation model, it effectively solves the problems of low resolution and weak temperature difference contrast in infrared images of tunnel face, and significantly improves the recognition accuracy of weak water seepage features in low-resolution infrared images.

[0047] 3. This invention achieves deep fusion of infrared images and three-dimensional geometric information, accurately projecting the seepage identification results onto the tunnel point cloud model or three-dimensional mesh model, completing the spatial positioning and quantitative measurement of the seepage area, realizing accurate characterization of the seepage area, shape, and intensity level, and meeting the actual engineering needs for spatial positioning and quantitative analysis. Attached Figure Description

[0048] Figure 1 A flowchart illustrating a method for accurately measuring seepage in tunnel surrounding rock according to an embodiment of the present invention;

[0049] Figure 2 A structural block diagram of a tunnel surrounding rock seepage precision measurement system provided in an embodiment of the present invention;

[0050] Figure 3 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0051] To enable those skilled in the art to better understand the technical solutions of the present invention, exemplary embodiments of the present invention are described below in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0052] Where there is no conflict, the various embodiments of the present invention and the features thereof may be combined with each other.

[0053] As used herein, the term “and / or” includes any and all combinations of one or more related enumerated entries.

[0054] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used herein, the singular forms “a” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that when the terms “comprising” and / or “made of” are used in this specification, the presence of the stated feature, integral, step, operation, element, and / or component is specified, but the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof is not excluded. Terms such as “connected” or “linked” are not limited to physical or mechanical connections but can include electrical connections, whether direct or indirect.

[0055] Unless otherwise specified, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having the meaning consistent with their meaning in the context of the relevant art and the invention, and will not be interpreted as having an idealized or overly formal meaning unless expressly so defined herein.

[0056] In the technical solution of this invention, the collection, storage, use, processing, transmission, provision, and disclosure of user personal information all comply with relevant laws and regulations and do not violate public order and good morals. The use of user data in this technical solution follows relevant national laws and regulations (e.g., the "Information Security Technology - Personal Information Security Specification"). For example: appropriate measures are taken for personal information access control; restrictions are imposed on the display of personal information; the purpose of using personal information does not exceed the scope of direct or reasonable association; and explicit identity targeting is eliminated when using personal information to avoid precisely locating a specific individual.

[0057] To address at least one of the technical problems existing in the aforementioned related technologies, the present invention provides a method and system for accurate measurement of seepage in tunnel surrounding rock.

[0058] This embodiment discloses a method for accurately measuring seepage in tunnel surrounding rock, such as... Figure 1 ,include:

[0059] S100. Collect multimodal sensing data of the tunnel surrounding rock, wherein the multimodal sensing data includes infrared data, visible light data and laser point cloud data;

[0060] In this embodiment, a handheld SLAM device integrating multiple sensors is used to collect multimodal perception data of tunnel surrounding rock. The specific steps include:

[0061] S101. Infrared camera, visible light camera, LiDAR, RTK positioning module, and interactive touch module are installed onto the mobile handheld measuring device body, ensuring overlapping fields of view for each sensor and unobstructed LiDAR field of view. Specifically, the infrared camera, visible light camera, LiDAR, RTK positioning module, and interactive touch module are installed onto the mobile handheld measuring device body, which includes a support frame, power module, and data acquisition and storage module, ensuring that each sensor is stably fixed and has a reasonable installation angle to achieve comprehensive coverage of the tunnel face area. It should be noted that the reasonable installation angle here refers to the LiDAR having a suitable installation angle to avoid other components or devices on the device body or users occupying a large portion of the LiDAR's field of view. Furthermore, there should be sufficient overlap in the fields of view between each pair of the visible light camera, infrared camera, and LiDAR. Through the above integration, multimodal integrated imaging of infrared thermal imaging, visible light imaging, and laser point cloud acquisition is achieved.

[0062] S102. Sensor calibration and time synchronization are performed, completing the calibration of intrinsic and extrinsic parameters of each sensor, and achieving timestamp alignment and coordinate system unification among multiple sensors; specifically, joint calibration is performed on the infrared camera, visible light camera, and lidar. This includes: firstly, intrinsic parameter calibration is performed, based on a checkerboard pattern or target, calibrating the focal length, principal point, and distortion coefficients of the infrared and visible light cameras; secondly, extrinsic parameter calibration is performed, establishing the spatial pose relationship between different sensors; thirdly, the time synchronization module is used to achieve time alignment of multi-sensor acquisitions; finally, through a unified coordinate system, multimodal data is converted to the same spatial coordinate frame, thereby ensuring the consistency and fusionability of different modal data.

[0063] S103. Collect and fuse multimodal data to generate results, including infrared image data, visible light image data, and three-dimensional laser point cloud data of the tunnel surrounding rock face.

[0064] S200. The multimodal sensing data is fused to establish tunnel surrounding rock feature map data; in this embodiment, the specific steps for fusing the multimodal sensing data to establish tunnel surrounding rock feature map data include:

[0065] S201. Construct a 3D point cloud map of the tunnel surrounding rock and use a laser-inertial fusion-based SLAM mapping algorithm to complete the 3D reconstruction of the tunnel space. Specifically, perform time synchronization and formatted parsing on multi-source data to extract point cloud data and equipment attitude data. Employ a laser-inertial fusion-based SLAM mapping algorithm (such as LIO-SAM, FAST-LIO2, or VINS-Fusion) to complete the 3D reconstruction of the tunnel space after eliminating sensor drift and motion distortion. This algorithm can estimate the pose changes of the equipment inside the tunnel in real time and construct a dense point cloud map containing the geometric features of the tunnel surrounding rock.

[0066] S202. Save the acquisition trajectory and pose files, recording the timestamps and corresponding pose parameters of the device's acquisition process; specifically, during the mapping process, the system simultaneously saves the device's acquisition movement trajectory files, which contain the acquisition timestamps and corresponding device poses (position and attitude parameters) at each moment. These data constitute the basis for subsequent spatiotemporal registration of multimodal fusion;

[0067] S203. Data fusion and tunnel surrounding rock feature map acquisition: Point cloud and camera image data are aligned based on timestamps. The images are projected onto the point cloud coordinate system to achieve multimodal data fusion, generating a surrounding rock feature map containing geometric morphology, surface texture, and thermal radiation distribution. Specifically, after completing the 3D map construction, point cloud and camera image data are aligned based on timestamp information. Infrared or visible light images are projected onto the point cloud coordinate system using an extrinsic parameter matrix to achieve multimodal data fusion. The fusion result forms a high-precision tunnel surrounding rock feature map, containing multidimensional information such as geometric morphology, surface texture, and thermal radiation distribution, providing a data foundation for subsequent surrounding rock structure analysis and seepage identification.

[0068] S300. Process the infrared data of the tunnel surrounding rock to extract the seepage area; In this embodiment, the specific steps for processing the infrared data of the tunnel surrounding rock to extract the seepage area include:

[0069] S301. Preprocess the infrared image sequence, use median filtering and bilateral filtering to reduce noise, enhance the image by limiting contrast adaptive histogram equalization, and obtain the preprocessed image through pseudo-color mapping and resolution enhancement.

[0070] Specifically, to address the limitations of imaging device noise, thermal drift, and external environmental interference in tunnel face infrared images, image preprocessing is performed on consecutive frame infrared image sequences. Median filtering and bilateral filtering are used for combined noise reduction. Let the original image be... The mathematical expression for the median filter output is:

[0071]

[0072] in The neighborhood window is used. Bilateral filtering further suppresses noise through weighted summaries of the spatial and pixel value domains; its mathematical expression is:

[0073]

[0074] in , These are the Gaussian space kernel and the pixel similarity Gaussian kernel, respectively. This is the normalization factor. Subsequently, contrast-limited adaptive histogram equalization (CLAHE) is used to enhance the infrared image, improving local contrast and suppressing over-amplified noise. The cumulative distribution function (CDF) is calculated for each local block, and its mathematical expression is:

[0075]

[0076] Next, the single-channel infrared image is pseudo-color mapped, and the temperature value is mapped to the RGB space using standard color tables (such as JET and TURBO). In order to adapt to the subsequent network input requirements, bicubic interpolation or a super-resolution algorithm based on deep learning is used to improve the resolution of the image.

[0077] S302. Enhance the edge features and construct multi-channel feature maps for infrared images. Combine the Laplacian operator and Sobel operator to extract image gradient features, calculate local statistical features, and construct multi-channel feature tensors.

[0078] Specifically, based on the preprocessing results, the edge and local structural features of the infrared image are enhanced. First, the Laplacian operator response is calculated, and the second-order gradient information is extracted. Its mathematical expression is:

[0079]

[0080] And by combining the Sobel operator to obtain the first-order gradient features, its mathematical expression is:

[0081]

[0082] in and Use a Sobel convolution kernel. Further calculate local statistical features, including local means. With variance Its mathematical expression is:

[0083]

[0084] Finally, the above features are combined to construct a multi-channel feature tensor, the mathematical expression of which is:

[0085]

[0086] This multi-channel feature tensor provides enhanced input for the subsequent segmentation network;

[0087] S303. Perform preliminary detection on candidate seepage regions, inputting the multi-channel feature map into a lightweight convolutional neural network segmentation model, and outputting a seepage probability map; specifically, the multi-channel feature map... Input to a lightweight segmentation network For example, a lightweight DeepLabV3+ network based on MobileNet or ShuffleNet backbones outputs a preliminary water seepage probability map, the mathematical expression of which is:

[0088]

[0089] The network learns the difference between the water seepage area and the background area through pixel-level binary classification, and outputs the probability that each pixel belongs to the water seepage area.

[0090] S304. Perform temporal consistency screening and result refinement, conduct temporal consistency analysis on the continuous frame water seepage probability map, and obtain the final water seepage area mask by combining morphological operations or conditional random field processing.

[0091] Specifically, based on the above results, the continuous frame segmentation results are... Perform time-series consistency analysis to remove transient noise. The mathematical expression for the time connectivity weight is:

[0092]

[0093] in The time-weighted coefficient, This is a normalization constant. When When the pixel is identified as a persistent water seepage area, further refinement of the segmentation result is achieved using conditional random fields or morphological operations to obtain the final water seepage area mask. The mathematical expression for mask refinement is as follows:

[0094]

[0095] This enables accurate and stable identification of water seepage areas at the tunnel face.

[0096] S400. Based on the tunnel surrounding rock feature map data, establish an image georeference of infrared data and point cloud data for the seepage area; In this embodiment, based on the tunnel surrounding rock feature map data, establish an image georeference of infrared data and point cloud data for the seepage area. The specific steps include: based on the extrinsic parameter transformation matrix of the infrared camera and lidar, establish the projection relationship between the pixel points and the collected point cloud, and map the pixel-level seepage area of ​​the seepage area mask onto the point cloud model or three-dimensional mesh model.

[0097] Specifically, based on the collected data, a geographic reference is established for the infrared imagery of water seepage. Based on the corresponding extrinsic parameter transformation matrix of the infrared camera and lidar, a projection relationship between pixels and the collected point cloud is established. Based on this projection relationship, pixel-level water seepage areas are masked within the infrared image. Mapping to point cloud model or 3D mesh model In this process, a georeferenced image of water seepage infrared imagery was established.

[0098] S500. Based on the image georeference of the infrared data and point cloud data, perform precise water seepage measurement, output the measurement results, and generate an alarm according to relevant standards.

[0099] In this embodiment, based on the image georeferenced data of the infrared data and point cloud data, precise water seepage measurement is performed. Specific steps include:

[0100] S501. Process the seepage area mask by performing morphological closing operations, connected component filtering, and merging of adjacent small regions to obtain a complete and continuous seepage distribution area. Specifically, this stage mainly processes the seepage area mask. First, morphological closing operations are used to remove noise and fill small voids to smooth the region boundaries. Then, connected component filtering is performed to remove invalid regions that are too small or isolated. Finally, adjacent small regions are merged based on a distance threshold to form a complete and continuous seepage distribution area.

[0101] S502. Calculate quantified parameters and estimate and classify seepage intensity within connected domains. Calculate quantified parameters such as area, perimeter, and shape coefficient of seepage patches, and classify seepage intensity according to preset standards. Specifically, this stage calculates quantified parameters such as area, perimeter, and shape coefficient of each seepage patch based on the connected domain. Combined with preset seepage intensity classification standards, intensity is estimated and graded for each connected domain, realizing the quantification and classification of seepage degree.

[0102] S503. Output pixel-level water seepage mask, geographic annotation results, and statistical reports including water seepage area, intensity level, and spatial distribution. Specifically, this stage outputs multi-level result data, including pixel-level mask of the water seepage area, geographic annotation results registered with laser point cloud or 3D mesh, and statistical reports containing information such as water seepage area, intensity level, and spatial distribution, for subsequent analysis and evaluation.

[0103] S504. Perform seepage monitoring and early warning. When the seepage intensity or area exceeds a preset threshold, automatically generate alarm information indicating the location and severity of the abnormal area, output the measurement results, and generate alarms according to relevant standards. Specifically, this stage performs automated monitoring and early warning based on the estimated seepage intensity. When seepage intensity or area is detected to exceed a preset threshold, the system automatically generates alarm information, indicating the location and severity of the abnormal area, providing decision support for tunnel safety monitoring and maintenance.

[0104] This embodiment provides a precise method for measuring seepage in tunnel surrounding rock, addressing the technical problems of large subjective errors in manual identification of seepage detection at the tunnel face, low accuracy of infrared image recognition, and inaccurate spatial positioning of seepage areas. This invention achieves automatic identification, spatial visualization, and quantitative measurement of seepage areas through steps including multimodal sensing data acquisition, surrounding rock feature map creation, infrared image processing and seepage area extraction, image geographic reference establishment, precise seepage measurement, and result output alarms. It combines infrared image enhancement, lightweight convolutional neural network segmentation, and spatial projection correction technologies. This invention boasts high identification accuracy, precise spatial positioning, high device integration, and strong mobility, making it suitable for rapid deployment and real-time monitoring at tunnel construction sites, providing an efficient and objective technical means for tunnel construction safety monitoring.

[0105] Based on the same inventive concept, embodiments of the present invention also provide a precise measurement system for seepage in tunnel surrounding rock, such as... Figure 2 It includes a multimodal sensing data acquisition and processing unit, a tunnel surrounding rock feature map data establishment unit, a deep-water area extraction unit, an image geographic reference establishment unit, and a seepage precision measurement unit; among which:

[0106] A multimodal sensing data acquisition and processing unit is used to acquire multimodal sensing data of the tunnel surrounding rock, including infrared data, visible light data and laser point cloud data.

[0107] The tunnel surrounding rock feature map data establishment unit is used to fuse the multimodal sensing data to establish tunnel surrounding rock feature map data;

[0108] The deep-water area extraction unit is used to process infrared data of the tunnel surrounding rock to complete the extraction of seepage areas;

[0109] The image georeference establishment unit is used to establish an image georeference of infrared data and point cloud data for the seepage area based on the tunnel surrounding rock feature map data.

[0110] The seepage precision measurement unit is used to perform precise seepage measurement based on the image georeference of the infrared data and point cloud data, output the measurement results, and generate alarms according to relevant standards.

[0111] The specific working methods of the multimodal sensing data acquisition and processing unit, the tunnel surrounding rock feature map data establishment unit, the deep water area extraction unit, the image geographic reference establishment unit, and the seepage precision measurement unit have been described in detail in the above-mentioned tunnel surrounding rock seepage precision measurement method, and will not be repeated here in this embodiment.

[0112] Those skilled in the art will understand that all or some of the steps, systems, and apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software can be distributed on a computer-readable storage medium, which may include computer storage media (or non-transitory media) and communication media (or transient media).

[0113] As is known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable program instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technologies, portable compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, it is known to those skilled in the art that communication media typically contain computer-readable program instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0114] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0115] The computer program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of the invention.

[0116] The computer program product described herein can be implemented specifically through hardware, software, or a combination thereof. In one alternative embodiment, the computer program product is specifically embodied in a computer storage medium; in another alternative embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.

[0117] Various aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0118] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0119] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0120] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0121] Example embodiments have been disclosed herein, and while specific terminology has been used, it is for illustrative purposes only and should be construed as such, and is not intended to be limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of the invention as set forth in the appended claims.

Claims

1. A method for accurately measuring seepage water in tunnel surrounding rock, characterized in that, include: Multimodal sensing data of the surrounding rock of the tunnel is collected, including infrared data, visible light data and laser point cloud data; The multimodal sensing data is fused to establish a tunnel surrounding rock feature map data; Infrared data of the surrounding rock of the tunnel are processed to extract the seepage area; Based on the tunnel surrounding rock feature map data, an image geographic reference of infrared data and point cloud data is established for the seepage area; Based on the image georeferenced data of the infrared data and point cloud data, the seepage is accurately measured, the measurement results are output, and an alarm is generated according to relevant standards.

2. The measurement method according to claim 1, characterized in that, The following steps are involved in acquiring multimodal sensing data of tunnel surrounding rock using a handheld SLAM device with integrated multiple sensors: Infrared camera, visible light camera, lidar, RTK positioning module and interactive touch module are installed on the body of mobile handheld measuring device to ensure that the fields of view of each sensor overlap and the lidar field of view is unobstructed. The sensor calibration and time synchronization are performed to complete the calibration of the intrinsic and extrinsic parameters of each sensor, and to achieve timestamp alignment and coordinate system unification among multiple sensors; The system collects and fuses multimodal data, including infrared image data, visible light image data, and three-dimensional laser point cloud data of the tunnel face.

3. The measurement method according to claim 1, characterized in that, The multimodal sensing data is fused to establish a tunnel surrounding rock feature map. Specific steps include: A three-dimensional point cloud map of the tunnel surrounding rock was constructed, and a laser-inertial fusion SLAM mapping algorithm was used to complete the three-dimensional spatial reconstruction of the tunnel. Save the acquisition trajectory and pose file, and record the timestamps and corresponding pose parameters of the device acquisition process; The data is fused and a tunnel surrounding rock feature map is obtained. The point cloud and camera image data are aligned according to the timestamp, and the image is projected onto the point cloud coordinate system to achieve multimodal data fusion, generating a surrounding rock feature map that includes geometric shape, surface texture and thermal radiation distribution.

4. The measurement method according to claim 1, characterized in that, The infrared data of the tunnel surrounding rock is processed to extract the seepage area. The specific steps include: The infrared image sequence was preprocessed, and noise reduction was achieved by a combination of median filtering and bilateral filtering. The image was enhanced by limiting contrast adaptive histogram equalization, and the preprocessed image was obtained through pseudo-color mapping and resolution enhancement. Edge feature enhancement and multi-channel feature map construction are performed on infrared images. The Laplacian operator and Sobel operator are combined to extract image gradient features, calculate local statistical features, and construct a multi-channel feature tensor. Preliminary detection of seepage candidate regions is performed. The multi-channel feature map is input into a lightweight convolutional neural network segmentation model, which outputs a seepage probability map. Temporal consistency screening and result refinement are performed. Temporal consistency analysis is conducted on the continuous frame water seepage probability map. The final water seepage area mask is obtained by combining morphological operations or conditional random field processing.

5. The measurement method according to claim 4, characterized in that, The method employs bilateral filtering for joint noise reduction, including: suppressing noise through weighted average of spatial and pixel values; limiting contrast adaptive histogram equalization to calculate the cumulative distribution function for each local block to achieve image enhancement; and improving resolution by using bicubic interpolation or a super-resolution algorithm based on deep learning.

6. The measurement method according to claim 4, characterized in that, The multi-channel feature tensor is constructed by combining the infrared image preprocessing results, the second-order gradient features of the Laplacian operator, the first-order gradient features of the Sobel operator, the local mean features, and the local variance features.

7. The measurement method according to claim 4, characterized in that, The lightweight convolutional neural network segmentation model is a DeepLabV3+ network based on MobileNet or ShuffleNet backbone, which learns the difference between the seepage area and the background area through pixel-level binary classification.

8. The measurement method according to claim 1, characterized in that, Based on the tunnel surrounding rock feature map data, an image georeference for the seepage area is established using infrared data and point cloud data. The specific steps include: establishing the projection relationship between pixels and the collected point cloud based on the extrinsic transformation matrix of the infrared camera and lidar, and mapping the pixel-level seepage area of ​​the seepage area mask onto the point cloud model or 3D mesh model.

9. The measurement method according to claim 1, characterized in that, Based on the image georeferenced data of the infrared data and point cloud data, precise water seepage measurement is performed, and the specific steps include: The seepage area mask is processed, and a complete and continuous seepage distribution area is obtained through morphological closing operation, connected component filtering, and merging of adjacent small regions. Within the connected domain, quantitative parameters and seepage intensity estimation classification are calculated. Quantitative parameters such as area, perimeter, and shape coefficient of seepage patches are calculated, and seepage intensity classification is completed in combination with preset standards. Output pixel-level water seepage mask, geographic annotation results, and statistical reports including water seepage area, intensity level, and spatial distribution; The system monitors and warns of water seepage. When the intensity or area of ​​seepage exceeds a preset threshold, it automatically generates alarm information indicating the location and severity of the abnormal area, outputs the measurement results, and generates alarms according to relevant standards.

10. A precise measurement system for seepage in tunnel surrounding rock, employing the measurement method described in any one of claims 1-9, characterized in that, It includes a multimodal sensing data acquisition and processing unit, a tunnel surrounding rock feature map data establishment unit, a deep-water area extraction unit, an image georeference establishment unit, and a seepage precision measurement unit; among which: A multimodal sensing data acquisition and processing unit is used to acquire multimodal sensing data of the tunnel surrounding rock, including infrared data, visible light data and laser point cloud data. The tunnel surrounding rock feature map data establishment unit is used to fuse the multimodal sensing data to establish tunnel surrounding rock feature map data; The deep-water area extraction unit is used to process infrared data of the tunnel surrounding rock to complete the extraction of seepage areas; The image georeference establishment unit is used to establish an image georeference of infrared data and point cloud data for the seepage area based on the tunnel surrounding rock feature map data. The seepage precision measurement unit is used to perform precise seepage measurement based on the image georeference of the infrared data and point cloud data, output the measurement results, and generate alarms according to relevant standards.

Images