Tunnel geological parameter rapid acquisition and surrounding rock grading method and system

Abstract

The invention provides a tunnel geological parameter rapid acquisition and surrounding rock grading method and system. The method comprises the following steps: performing multi-source original data acquisition on a target tunnel face or a target side wall planned with a geological survey window based on acquisition equipment adaptive to a tunnel environment; preprocessing the multi-source original data to obtain processed data, and extracting lithology information and structural plane parameter information based on the processed data; calling a preset surrounding rock grading standard algorithm to perform surrounding rock grading calculation based on the lithologic information and the structural plane parameter information to obtain a surrounding rock quality grade; and generating a tunnel grading report according to the surrounding rock quality grade, the lithology information and the structural plane parameter information. Multi-dimensional information is integrated to generate the tunnel grading report, the whole process standardization from data acquisition, processing and analysis to report generation is realized, and the efficiency and precision of tunnel geological parameter acquisition and surrounding rock grading are improved.

Description

Technical Field

[0001] This invention relates to the field of tunnel construction technology in water conservancy and hydropower engineering, specifically to a method and system for rapid acquisition of tunnel geological parameters and classification of surrounding rock. Background Technology

[0002] With the deepening of energy structure transformation, hydropower, as a widely distributed, clean, and renewable energy source, is receiving increasing attention for its development and construction. Against this backdrop, a series of large-scale water conservancy and hydropower projects have been planned and implemented in southwestern my country. In these projects, the classification of surrounding rock in tunnels is a core basis for guiding support design, construction decisions, and risk control; its accuracy and timeliness directly affect the safety and economy of the project.

[0003] The existing technology has the following drawbacks: ① Low efficiency of parameter acquisition: Traditional manual acquisition requires 2-3 testers to carry tools such as compasses and measuring tapes. It takes 40-60 minutes to complete the work of lithology identification and structural surface parameter acquisition for a single tunnel section. Data entry and surrounding rock classification calculation take an additional 30 minutes. For long tunnels exceeding 10km in total length, the data collection cycle for a single tunnel section far exceeds the tunneling cycle (8 hours / cycle), resulting in a lag in geological information feedback. According to statistics from the "Code for Construction of Water Conservancy and Hydropower Tunnels," the incidence of untimely support caused by such lag exceeds 20%. ② Significant errors in manual measurement: The collection of parameters such as structural surface spacing and attitude relies on manual judgment. Among them, the measurement error of spacing can reach ±15cm, and the dip angle error is ±5°. This leads to significant deviations between the classification results of various classification standards (such as the hydropower engineering method, BQ method, and RMR method) and the actual situation, further resulting in unreasonable on-site support design schemes and problems such as excessive costs or insufficient support risks. ③ Disconnect between data flow and visualization: Existing on-site data collection requires manual entry into the system, which is prone to human subjective errors such as typos. Moreover, the synchronization between on-site data collection and cloud database and surrounding rock classification calculation is seriously delayed.

[0004] In other words, how to provide a method for rapid acquisition of tunnel geological parameters and classification of surrounding rock to improve the efficiency of on-site data acquisition and the accuracy of surrounding rock classification is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0005] This invention provides a method and system for rapid acquisition of tunnel geological parameters and classification of surrounding rock to solve at least one of the above-mentioned technical problems.

[0006] Firstly, this application provides a method for rapid acquisition of tunnel geological parameters and classification of surrounding rock, the method comprising: The acquisition equipment, adapted to the tunnel environment, performs multi-source raw data acquisition on the target face or sidewall with planned geological measurement windows. The multi-source raw data includes raw image data, shooting distance, and corresponding inertial sensor data; the geological measurement window covers at least one complete exposed section of the structural surface. The multi-source raw data is preprocessed to obtain processed data, and lithological information and structural surface parameter information are extracted based on the processed data; The surrounding rock quality grade is obtained by calling a preset surrounding rock classification standard algorithm based on the lithological information and the structural surface parameter information to perform surrounding rock classification calculation. A tunnel classification report is generated based on the surrounding rock quality grade, the lithological information, and the structural surface parameter information.

[0007] Optionally, the preprocessing of the multi-source raw data to obtain processed data includes: The original image data is sequentially subjected to dehazing, illumination equalization, and distortion correction to obtain the processed image data. The shooting distance is subjected to distortion correction processing to obtain the processed shooting distance; The inertial sensing data is fused and calculated to obtain the shooting angle and three-dimensional attitude angle corresponding to the current captured image; The processed image data, the processed shooting distance, the shooting angle, and the three-dimensional attitude angle constitute the processed data.

[0008] Optionally, the acquisition device includes: a mobile acquisition device and an auxiliary acquisition device. The mobile acquisition device is equipped with a camera, a laser ranging unit and an inertial navigation unit. The camera is used to acquire image data of the target face or the target sidewall. The laser ranging unit is used to acquire the shooting distance between the camera and the target face or the target sidewall during the current shooting. The inertial navigation unit is used to acquire the raw inertial sensing data corresponding to the current shooting. The auxiliary acquisition equipment includes a supplementary lighting bracket and a pose correction device. The supplementary lighting bracket is placed inside the tunnel to provide illumination compensation for the camera. The pose correction device is located at the bottom center of the geological survey window to provide a reference plane for image acquisition. The reference plane consists of three fluorescent marker discs that are fixedly connected to each other.

[0009] Optionally, the extraction of lithological information and structural surface parameter information based on the processed data includes: The processed image data is input into the trained lithology identification model for lithology identification processing, and the lithology information of the currently collected target face or target sidewall is output. The lithology information includes lithology type and rock mass integrity level. Based on the processed image data, the processed shooting distance, the shooting angle, and the three-dimensional attitude angle, the structural surface parameter information of the target face or the target sidewall is calculated. The structural surface parameter information includes structural surface geometric parameters and structural surface attitude parameters.

[0010] Optionally, the structural surface geometric parameters include the structural surface trace length, and the calculation process for the structural surface trace length is as follows: The improved YOLOv5 algorithm is used to identify the first and second trace length feature points in the current frame of the processed image data, and the pixel distance P between the first and second trace length feature points is calculated. ab The first trace length feature point and the second trace length feature point are the two endpoints of the exposed trace of the structural surface in the current frame; The length L of the structural surface trace is calculated according to the following formula. t : L t =P ab ×L×cosθ / f; Where f is the camera's calibrated focal length, θ is the shooting angle of the current frame, and L is the processed shooting distance of the current frame; The structural plane geometric parameters also include the structural plane opening, which is calculated as follows: The first and second edge feature points of the current frame in the processed image data are identified by the improved YOLOv5 algorithm. The first and second edge feature points are the two edge points on both sides of the structural surface gap in the current frame. Calculate the pixel distance P between the first edge feature point and the second edge feature point along the vertical direction of the gap. d ; The opening W of the structural surface is calculated according to the following formula: W=P d ×L×cosθ / f.

[0011] Optionally, the structural surface geometric parameters also include the structural surface spacing, and the calculation process for the structural surface spacing is as follows: By improving the YOLOv5 algorithm, the first and second center feature points of the current frame are identified, and the pixel distance P between them is calculated. c The first and second center feature points are the center reference points of two adjacent structural planes in the current frame; the spacing L between the structural planes is calculated according to the following formula. s : L s =P c ×L×cosθ / f.

[0012] Optionally, the structural plane attitude parameters include the structural plane dip, and the calculation process for the structural plane dip is as follows: Obtain the inclination of the reference plane, and calculate the relative offset angle Δθ between the inclination of the reference plane and the trace of the structural surface according to the following formula: Δθ= arctan {(u1-u0) / (v1-v0)}-φ; Where (u0, v0) are the center pixel coordinates of the three fluorescent marker disks, (u1, v1) are the midpoint pixel coordinates of the structure surface trace in the current frame, and φ is the heading angle; The dip direction of the reference plane is corrected based on the relative offset angle Δθ to obtain the strike azimuth angle of the structural plane. The dip direction α of the structural plane is then calculated based on the strike azimuth angle using the following formula: ; Where α0 is the dip of the reference plane.

[0013] Optionally, the structural plane attitude parameters also include the structural plane dip angle, which is calculated as follows: Obtain the pixel height difference ΔP of the structural surface traces in the current frame of the processed image data. v ; According to the formula Δh = ΔP v Calculate the vertical height difference Δh of the structural surface traces in space using the formula: × L × cosθ / f; According to formula L h = arctan (P L Calculate the spatial horizontal length L of the trace line of the structure surface by calculating ×L×cosθ / f) ×cosβ0. h ; Calculate the structural plane dip angle β using the following formula: β = arcsin (Δh / L h ) + β0-90°; If the calculated result of the inclination angle of the structural surface is negative, then the absolute value is taken and the direction is adjusted; Among them, P L The structure surface trace is based on the pixel length of the current frame, and β0 is the spatial tilt angle of the reference plane.

[0014] Optionally, the method further includes: The web interface acquires the tunnel classification report, the surrounding rock quality grade, the lithological information, the structural surface parameter information, and the environmental parameter information in real time, and displays them in three dimensions based on the above information.

[0015] Secondly, this application provides a rapid acquisition system for tunnel geological parameters and a system for classifying surrounding rock, including: The acquisition module is used to acquire multi-source raw data from the target face or sidewall with geological survey windows based on acquisition equipment adapted to the tunnel environment. The multi-source raw data includes raw image data, shooting distance, and corresponding inertial sensor data; the geological survey window covers at least one complete exposed section of a structural surface. The processing module is used to preprocess the multi-source raw data to obtain processed data; The calculation and extraction module is used to extract lithological information and structural surface parameter information based on the processed data; and to call a preset surrounding rock grading standard algorithm to perform surrounding rock grading calculation based on the lithological information and the structural surface parameter information to obtain the surrounding rock quality grade. The generation module is used to generate a tunnel classification report based on the surrounding rock quality grade, the lithological information, and the structural surface parameter information.

[0016] Technical effects: This invention first utilizes data acquisition equipment adapted to the complex environment of tunnels to collect multi-source raw data from target tunnel faces or sidewalls with planned geological survey windows. These geological survey windows ensure coverage of at least one complete exposed section of a structural plane, guaranteeing the integrity and relevance of the raw data from the source and preventing subsequent analysis biases due to missing data or inappropriate collection scope. Through preprocessing and key information extraction of the multi-source raw data, accurate acquisition of lithological and structural plane parameter information is achieved, providing reliable data support for surrounding rock classification. Classification calculations are performed using a pre-defined surrounding rock classification standard algorithm, replacing the traditional subjective judgment mode of manual classification and improving the objectivity and accuracy of surrounding rock quality grade determination. Finally, multi-dimensional information is integrated to generate a tunnel classification report, achieving standardization of the entire process from data acquisition, processing, and analysis to report generation, significantly improving the efficiency and professionalism of tunnel geological parameter acquisition and surrounding rock classification. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the 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.

[0018] Figure 1 A flowchart illustrating a method for rapid acquisition of tunnel geological parameters and classification of surrounding rock provided in this application; Figure 2 This is a schematic diagram of the deployment of the mobile data acquisition terminal provided in this application; Figure 3 The web-based 3D visualization effect provided for this application; Figure 4 This is a schematic diagram of a tunnel geological parameter rapid acquisition and surrounding rock classification system provided in this application.

[0019] Figure label: 1. Geological survey window; 2. Structural planes; 3. Posture correction device; 4. Lighting stand; 5. Mobile data acquisition devices. Detailed Implementation

[0020] This application provides a method and system for rapid acquisition of tunnel geological parameters and classification of surrounding rock, in order to solve at least one of the above-mentioned technical problems.

[0021] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application 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 described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules is not necessarily limited to those explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices. The naming or numbering of steps appearing in this application does not imply that the steps in the method flow must be performed in the chronological / logical order indicated by the naming or numbering. The execution order of named or numbered process steps can be changed according to the desired technical purpose, as long as the same or similar technical effect is achieved.

[0023] The module division described in this application is a logical division. In practical applications, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the coupling or direct coupling or communication connection between modules shown or discussed may be through some interfaces, and the indirect coupling or communication connection between modules may be electrical or other similar forms, none of which are limited in this application. Furthermore, the modules or sub-modules described as separate components may or may not be physically separated, may or may not be physical modules, or may be distributed in multiple circuit modules. Some or all of the modules may be selected to achieve the purpose of the solution in this application according to actual needs.

[0024] Next, please refer to Figure 1-2 , Figure 1 This is a flowchart illustrating a method for rapid acquisition of tunnel geological parameters and classification of surrounding rock in an embodiment of the present invention. As an embodiment of the method for rapid acquisition of tunnel geological parameters and classification of surrounding rock provided by the present invention, the method includes the following steps S110 to S140: Step S110: Based on the acquisition equipment adapted to the tunnel environment, multi-source raw data is acquired for the target face or target sidewall with planned geological measurement windows. The multi-source raw data includes raw image data, shooting distance and corresponding inertial sensor data; the geological measurement window covers at least one complete exposed section of the structural surface. As one feasible approach, the data acquisition equipment configured at the tunnel site in this application includes: a mobile data acquisition device and an auxiliary data acquisition device. The mobile data acquisition device is an industrial tablet equipped with an IP67 waterproof and dustproof shell, and is equipped with a ≥16-megapixel camera (supporting ISO100-6400 and low-light enhancement mode), a laser ranging unit (range 0.5-50m, error ≤±2mm), and an inertial navigation unit (IMU, angle measurement accuracy ±0.1°). The camera is used to acquire image data of the target face or target sidewall, the laser ranging unit is used to acquire the shooting distance between the camera and the target face or target sidewall during the current shooting, and the inertial navigation unit is used to acquire the raw inertial sensing data corresponding to the current shooting. The raw inertial sensing data includes three-axis acceleration and three-axis angular velocity. The auxiliary acquisition equipment includes a supplementary lighting bracket and a pose correction device. The supplementary lighting bracket can be implemented using an existing foldable bracket structure with 3 sets of LED lights. The illuminance adjustment range can be set to 50-500 lux, and it is placed inside the tunnel to provide illumination compensation for the camera (adapting to low-light environments with illuminance <50 lux). The pose correction device is located at the bottom center of the geological survey window, such as... Figure 2As shown, a reference plane is provided for image acquisition. This reference plane consists of three fluorescent marker disks that are fixedly connected to each other. The diameter of the fluorescent marker disks can be set to 5cm and the reflectivity ≥80%.

[0025] In step S110, which involves acquiring multi-source raw data at the tunnel site, only one set of data for the target face or target sidewall is considered. To acquire complete geological data for the tunnel, step S1 needs to be repeated until data acquisition for different sections along the tunnel is completed. Furthermore, when acquiring data for the target sidewall, if the tunnel is long, geological survey windows need to be planned on the sidewall at preset intervals to present the quality levels of different tunnel lengths as completely as possible.

[0026] Step S120: Preprocess the multi-source raw data to obtain processed data, and extract lithological information and structural surface parameter information based on the processed data; As one possible approach, the preprocessing of multi-source raw data involved in step S120 above to obtain processed data includes: The original image data is sequentially subjected to dehazing, illumination equalization, and distortion correction to obtain the processed image data. The shooting distance is subjected to distortion correction processing to obtain the processed shooting distance; The inertial sensing data is fused and calculated to obtain the shooting angle and three-dimensional attitude angle corresponding to the current captured image; The processed image data, the processed shooting distance, the shooting angle, and the three-dimensional attitude angle constitute the processed data.

[0027] Specifically, the original image data is sequentially subjected to dehazing processing based on dark channel prior, illumination equalization processing based on the Retinex algorithm, and distortion correction processing based on camera intrinsic parameters to obtain processed image data. The original inertial sensor data can then be fused using Kalman filtering or complementary filtering algorithms to obtain the shooting angle and three-dimensional attitude angle corresponding to the currently captured image. The three-dimensional attitude angle includes the heading angle, pitch angle, and roll angle. The shooting angle is the angle between the camera and the normal direction of the structural surface, and can be derived from the three-dimensional attitude angle combined with the orientation of the reference plane.

[0028] Specifically, addressing the issues of insufficient lighting, fog interference, and distortion caused by shooting distance in tunnel environments, the raw image data underwent defogging and illumination equalization processing, effectively improving image clarity and contrast. This laid a high-quality image foundation for subsequent lithology identification and structural surface parameter extraction. Distortion correction of the shooting distance eliminated distance measurement errors, ensuring the accuracy of spatial dimension data. Through the fusion calculation of inertial sensor data, the shooting angle and three-dimensional attitude angle were accurately obtained, providing a unified spatial reference benchmark for the fusion analysis of multi-source data. This multi-dimensional preprocessing process comprehensively optimized the quality of the raw data, enabling the processed data to more realistically and comprehensively reflect the geological characteristics of the target tunnel face or sidewall, avoiding the adverse effects of raw data defects on subsequent analysis results, and further improving the reliability and accuracy of the entire method.

[0029] As one feasible approach, lithological information and structural parameters are extracted from processed data, specifically including the following: The processed image data is input into the trained lithology identification model for lithology identification processing, and the lithology information of the target face or target sidewall currently collected is output. The lithology information includes lithology type and rock mass integrity level. Based on the processed image data, processed shooting distance, shooting angle, and three-dimensional attitude angle, the structural surface parameter information of the target face or target sidewall is calculated. The structural surface parameter information includes the structural surface geometric parameters and the structural surface attitude parameters.

[0030] Among them, the lithology identification model is an improved MobileNetV3 deep learning model with an identification accuracy of ≥92%.

[0031] Specifically, the processed image data is input into a trained lithology identification model, enabling automated identification of lithology types and rock mass integrity levels. This replaces the subjectivity and inefficiency of traditional manual visual identification, significantly improving the speed and accuracy of lithology information extraction. By combining the processed image data with corrected shooting distance, shooting angle, and three-dimensional attitude angles, structural surface parameters are calculated, achieving precise quantification of structural surface geometric and attitude parameters, avoiding parameter extraction errors caused by single-dimensional data analysis. Through multi-dimensional extraction of lithology and structural surface parameter information, the comprehensiveness and relevance of key geological parameters are ensured, providing rich and accurate core data support for subsequent surrounding rock classification calculations, further enhancing the scientific validity and reliability of the surrounding rock classification results.

[0032] As one feasible approach, the structural surface geometric parameters include the structural surface trace length, and the calculation process for the structural surface trace length is as follows: By improving the YOLOv5 algorithm, the first and second trace length feature points of the current frame in the processed image data are identified, and the pixel distance P between the first and second trace length feature points is calculated. ab The first and second trace length feature points are the two endpoints of the exposed trace of the structural surface in the current frame. If the trace of the structural surface is occluded, the algorithm identifies the endpoints of the visible segment and marks them as "incomplete traces". At the same time, it extrapolates the pixel length of the complete trace based on the joint orientation trend. The length L of the structural surface trace is calculated using the following formula. t : L t =P ab ×L×cosθ / f; Where f is the camera's calibrated focal length, θ is the shooting angle of the current frame, and L is the processed shooting distance of the current frame.

[0033] As an feasible approach, the structural plane geometric parameters also include the structural plane opening, which is calculated as follows: The algorithm improves the YOLOv5 algorithm to identify the first and second edge feature points of the current frame in the processed image data. The first and second edge feature points are the two edge points on both sides of the gap in the structural surface in the current frame. For structural surfaces with an opening of <1mm, the algorithm automatically enlarges the local image area before performing pixel measurement. Calculate the pixel distance P between the first edge feature point and the second edge feature point along the vertical direction of the gap. d ; The structural plane opening W is calculated using the following formula: W=P d ×L×cosθ / f.

[0034] Specifically, an improved YOLOv5 algorithm is used to accurately identify the endpoints of structural surface traces and feature points at the edges of gaps. Compared with traditional feature point recognition methods, it has higher recognition accuracy and anti-interference capabilities, effectively capturing the detailed features of structural surfaces in complex geological environments within tunnels. By establishing a conversion formula between pixel distance and actual spatial distance, the measured values ​​of image pixel dimensions are combined with parameters such as shooting distance, shooting angle, and camera focal length, enabling precise quantitative calculation of the length and opening of structural surface traces. This calculation method fully considers various influencing factors during on-site tunnel imaging, eliminates measurement errors caused by differences in shooting conditions, ensures the accuracy and comparability of structural surface geometric parameters, and provides accurate quantitative basis for surrounding rock stability analysis and classification evaluation.

[0035] As an feasible approach, the structural surface geometric parameters also include the spacing between structural surfaces. The calculation process for the spacing between structural surfaces is as follows: By improving the YOLOv5 algorithm, the first and second center feature points of the current frame are identified, and the pixel distance P between them is calculated. c The first and second center feature points are the center reference points of two adjacent structural planes in the current frame; The spacing L between structural surfaces is calculated using the following formula. s : L s =P c ×L×cosθ / f.

[0036] Specifically, by improving the YOLOv5 algorithm to identify the center reference point of adjacent structural surfaces, the pixel distance between them can be quickly obtained, simplifying the operation process of measuring structural surface spacing. This avoids the tedious steps of point-by-point marking and manual measurement required in traditional measurement methods, thus improving the efficiency of on-site data collection. Using a unified spatial conversion formula, the pixel distance is converted into the actual structural surface spacing. The formula incorporates corrected parameters such as shooting distance, shooting angle, and camera focal length, effectively offsetting the influence of factors such as shooting angle and distance on the measurement results, ensuring the accuracy of structural surface spacing calculation. This method achieves automated and accurate extraction of structural surface spacing, improves the geometric parameter system of structural surfaces, and provides crucial data support for comprehensively assessing rock mass integrity and surrounding rock stability.

[0037] As one feasible approach, the structural plane attitude parameters include the structural plane dip direction, and the calculation process for the structural plane dip direction is as follows: Obtain the inclination of the reference plane, and calculate the relative offset angle Δθ between the inclination of the reference plane and the trace of the structural surface according to the following formula: Δθ= arctan {(u1-u0) / (v1-v0)}-φ; Where (u0, v0) are the center pixel coordinates of the reference plane formed by the three fluorescent marker disks, (u1, v1) are the midpoint pixel coordinates of the structure surface trace in the current frame, and φ is the heading angle in the three-dimensional attitude angle. The dip direction of the reference plane is corrected based on the relative offset angle Δθ to obtain the strike azimuth of the structural plane. The dip direction α of the structural plane is then calculated based on the strike azimuth. The calculation formula is as follows: ; Where α0 is the dip direction of the reference plane, + To determine the azimuth angle.

[0038] The inclination of the reference plane and its spatial tilt angle can be obtained based on manual measurements by on-site personnel or by monitoring corresponding sensors configured on the pose correction device.

[0039] Specifically, using the dip of a reference plane as a benchmark, a scientific calculation logic for the azimuth angle of the structural surface is established by calculating the relative offset angle between the reference plane and the trace of the structural surface, and then correcting it with the heading angle. This avoids errors caused by directly measuring the dip of the structural surface. The segmented formula for calculating the dip of the structural surface ensures the rationality and accuracy of the calculation results within a 360° azimuth range, effectively resolving potential issues of exceeding the range or logical contradictions during angle calculation. This calculation method fully utilizes the precise benchmark role of the reference plane, combining image pixel coordinates and spatial attitude parameters to achieve precise quantification of the dip of the structural surface, providing reliable attitude parameters for analyzing the spatial distribution characteristics of the structural surface and the stability of the surrounding rock.

[0040] As an achievable method, the structural plane attitude parameters also include the structural plane dip angle, which is calculated as follows: Obtain the pixel height difference ΔP of the structure surface traces in the current frame of the processed image data. v ; According to the formula Δh = ΔP v Calculate the vertical height difference Δh of the structural surface traces in space using the formula: × L × cosθ / f; According to formula L h = arctan (P L Calculate the spatial horizontal length L of the trace line of the structure surface by calculating ×L×cosθ / f) ×cosβ0. h ; Calculate the structural plane dip angle β using the following formula: β = arcsin (Δh / L h ) + β0-90°; If the calculated result of the inclination angle of the structural surface is negative, then the absolute value is taken and the direction is adjusted; Among them, P L The structure surface trace is based on the pixel length of the current frame, and β0 is the spatial tilt angle of the reference plane.

[0041] Specifically, by extracting the pixel height difference of structural surface traces in the image and combining it with parameters such as shooting distance, angle, and focal length to calculate the spatial vertical height difference and horizontal length, a bridge is established between the pixel dimension and the spatial dimension, enabling indirect and accurate calculation of the structural surface dip angle. This method fully considers the influence of the reference plane's spatial dip angle and eliminates measurement deviations caused by the inclination of the reference plane through formula correction, ensuring the accuracy of the dip angle calculation. Compared with the traditional, coarse-grained method of manually measuring dip angles, this calculation process is more scientific and repeatable, accurately capturing the dip characteristics of structural surfaces and providing precise quantitative indicators for the stability evaluation of structural surfaces in surrounding rock classification.

[0042] Step S130: Call the preset surrounding rock classification standard algorithm to calculate the surrounding rock classification based on lithological information and structural surface parameter information, and obtain the surrounding rock quality grade; Among them, the surrounding rock classification standard algorithm includes one or more of the following: hydropower engineering (HC method), BQ method and improved RMR method. All three algorithms can be built into the mobile terminal acquisition device and can be selected and called accordingly when used. Users can choose the target classification algorithm on the mobile terminal according to the engineering design requirements, geological exploration specifications or established project standards. After the algorithm is started, the complete index system and scoring rules of the corresponding standard are automatically loaded without manual configuration.

[0043] The extracted lithological information (including lithological type and rock mass integrity grade) and structural surface parameter information (including structural surface geometric parameters: trace length, opening, and spacing; structural surface attitude parameters: dip direction and dip angle) are automatically mapped and matched with the index system of the selected surrounding rock grading standard algorithm. For example, in the improved RMR method, the rock mass integrity grade "intact" corresponds to a rock mass integrity score of 30 points, and a structural surface spacing of 0.8m corresponds to a structural surface condition score of 22 points. For non-automatically extracted parameters such as groundwater condition, geostress, and structural surface infill type required by the grading algorithm, the mobile terminal acquisition device will guide the on-site testing personnel to supplement and enter the parameters through a visual form. The input items are accompanied by standard descriptions and option prompts (such as groundwater condition providing options such as "no seepage", "drip seepage", and "linear flow") to ensure the standardization of the supplemented parameters. Then, the parameters are manually supplemented and scored based on the supplemented parameters.

[0044] Then, the surrounding rock grading standard algorithm, based on the mapped parameter scores and manually completed parameter scores, completes the total score calculation according to the calculation logic of the selected grading standard (e.g., the improved RMR method sums the scores of five indicators: uniaxial compressive strength of rock mass, rock mass integrity, structural surface condition, groundwater condition, and geostress correction). For the improved RMR algorithm, if the maximum principal stress σ1 in the input geostress data is greater than 50MPa, the geostress correction mechanism will be automatically triggered, deducting 5-10 points to ensure that the calculation results are consistent with the actual situation of complex stress environments such as deep-buried tunnels. The calculation process takes ≤10 seconds, and the results are automatically matched with the grade thresholds built into the algorithm (e.g., in the improved RMR method, a total score of 81-100 points is Grade I, 61-80 points is Grade II, etc.), and the corresponding surrounding rock quality grade (Grade I-V) is output.

[0045] After the grading calculation is completed, the mobile data acquisition device synchronously displays the surrounding rock quality grade, the sub-scores of each indicator, and the total score, generating a scoring radar chart to intuitively present the performance of each dimension. If there are missing parameters, parameters exceeding the reasonable range, or the calculation results deviate too much from the historical data of the same period (the deviation threshold can be customized), the system will automatically pop up an abnormal prompt, mark the abnormal items, and suggest a review to ensure the reliability of the surrounding rock quality grade.

[0046] Specifically, by incorporating multiple standardized grading algorithms adapted to tunnel engineering, the traditional method of manual calculation based on grading standard manuals is replaced. This avoids inconsistencies in grading caused by different personnel's misunderstandings of the standards and significantly reduces the time required for grading calculations. Combined with the automatic parameter mapping function, it achieves high efficiency and standardization in grading calculations. The collaborative design of parameter mapping and completion guidance ensures the full utilization of automatically extracted data and reduces the operational threshold and error rate of manual parameter supplementation through standardized forms, ensuring that the data source for grading calculations is comprehensive and accurate.

[0047] Step S140: Generate a tunnel classification report based on the surrounding rock quality grade, lithological information, and structural surface parameter information.

[0048] As one possible approach, the method also includes: The web interface provides real-time access to tunnel classification reports, surrounding rock quality grades, lithological information, structural surface parameters, and environmental parameters, and displays these information in a 3D visualization.

[0049] Specifically, the tunnel classification report, surrounding rock quality grade, lithological information, structural surface parameter information, and data acquisition location (GPS / BeiDou positioning, error ≤5m) and upload time are synchronized to the cloud database and the web-based digital platform via 5G or BeiDou short message service. The web platform receives this information in real time and marks the surrounding rock quality grade at the corresponding mileage in the tunnel's 3D model. Different colors can be used to distinguish different grades, and the information is linked to structural surface parameter images and lithological photographs for 3D visualization. Figure 3 As shown.

[0050] Specifically, the web-based system enables real-time access to tunnel classification reports and various core data, achieving timely and shared data transmission. This allows relevant personnel to remotely monitor tunnel geological conditions in real time, obtaining crucial information without needing to be physically present on-site, thus improving work efficiency. The 3D visualization function transforms abstract lithological information, structural parameters, and surrounding rock quality grades into intuitive 3D models, making geological features and classification results clearer and easier to understand. This facilitates quick understanding of core information and aids in decision-making. This function not only achieves efficient data transmission and intuitive presentation but also provides convenient data support and visualization tools for subsequent tunnel construction plan optimization and risk warning, further expanding the application value of this method.

[0051] Compared with the prior art, the present invention has the following significant advantages: (1) Breakthrough in efficiency: The time for single-section data collection and classification has been shortened from 70-90 minutes to 12-15 minutes, with an efficiency improvement of 82%, meeting the geological information and surrounding rock stability feedback requirements of the "8-hour tunneling cycle" for long tunnels and avoiding support risks caused by delays.

[0052] (2) Significantly improved accuracy: By replacing manual on-site measurements with image intelligent recognition and other algorithm models integrated into mobile devices, the measurement errors of parameters such as structural surface spacing, trace length, and lithology type are effectively reduced, resulting in a significant improvement in the accuracy of surrounding rock classification results.

[0053] (3) Strong environmental adaptability: For tunnels with low light and high dust, supplementary lighting adjustment, defogging algorithm and dedicated hardware protection are used to achieve high adaptability even with light intensity of 50 lux and PM2.5 of 1000 μg / m³. 3 It maintains stable performance (recognition accuracy ≥85%) even under certain conditions, solving the problem of laboratory models being difficult to implement.

[0054] (4) Full-process automation: Based on the collaborative work of the hardware layer, algorithm layer, application layer and data interaction layer of the mobile device, end-to-end automation of "image acquisition - parameter extraction - hierarchical calculation - data upload" can be realized, and the data entry error is reduced from 8% to less than 1%.

[0055] (5) Close data linkage: Real-time synchronization with cloud database and Web digital platform (delay ≤30 seconds) to form a closed-loop management of "on-site collection - background analysis - three-dimensional display", providing real-time data support for dynamic adjustment of support scheme.

[0056] The following describes an embodiment of the tunnel geological parameter rapid acquisition and surrounding rock classification system in this invention.

[0057] Please see Figure 4 , Figure 4 This is a schematic diagram of an embodiment of the tunnel geological parameter rapid acquisition and surrounding rock classification system 400 of the present invention. The tunnel geological parameter rapid acquisition and surrounding rock classification system 400 includes: The acquisition module 401 is used to acquire multi-source raw data of the target face or sidewall with geological measurement windows based on the acquisition equipment adapted to the tunnel environment. The multi-source raw data includes raw image data, shooting distance and corresponding inertial sensor data; the geological measurement window covers at least one complete exposed section of the structural surface. Processing module 402 is used to preprocess the multi-source raw data to obtain processed data; The calculation and extraction module 403 is used to extract lithological information and structural surface parameter information based on the processed data; And call the preset surrounding rock classification standard algorithm to perform surrounding rock classification calculation based on lithological information and structural surface parameter information to obtain the surrounding rock quality grade; Module 404 is used to generate a tunnel classification report based on the surrounding rock quality grade, lithological information, and structural surface parameter information.

[0058] In one possible manner, the acquisition module 401 is also configured to perform the following steps: The original image data is subjected to dehazing and illumination equalization to obtain the processed image data; The shooting distance is subjected to distortion correction to obtain the processed shooting distance; The inertial sensing data is fused and calculated to obtain the shooting angle and three-dimensional attitude angle corresponding to the current captured image; The processed data consists of the processed image data, the processed shooting distance, the shooting angle, and the three-dimensional pose angle.

[0059] As one possible approach, the acquisition device includes: a mobile acquisition device and an auxiliary acquisition device. The mobile acquisition device is equipped with a camera, a laser ranging unit, and an inertial navigation unit. The camera is used to acquire image data of the target face or the target sidewall. The laser ranging unit is used to acquire the shooting distance between the camera and the target face or the target sidewall during the current shooting. The inertial navigation unit is used to acquire the raw inertial sensing data corresponding to the current shooting. The auxiliary acquisition equipment includes a lighting bracket and a posture correction device. The lighting bracket is placed inside the tunnel to provide illumination compensation for the camera. The inertial navigation unit is located at the bottom center of the geological survey window to provide a reference plane for image acquisition. The reference plane consists of three fluorescent marker disks that are fixedly connected to each other.

[0060] In one possible manner, the calculation extraction module 403 is also configured to perform the following steps: The processed image data is input into the trained lithology identification model for lithology identification processing, and the lithology information of the target face or target sidewall currently collected is output. The lithology information includes lithology type and rock mass integrity level. Based on the processed image data, processed shooting distance, shooting angle, and three-dimensional attitude angle, the structural surface parameter information of the target face or target sidewall is calculated. The structural surface parameter information includes the structural surface geometric parameters and the structural surface attitude parameters.

[0061] As one feasible approach, the structural surface geometric parameters include the structural surface trace length, and the calculation process for the structural surface trace length is as follows: By improving the YOLOv5 algorithm, the first and second trace length feature points of the current frame in the processed image data are identified, and the pixel distance P between the first and second trace length feature points is calculated.ab The first trace length feature point and the second trace length feature point are the two endpoints of the exposed trace of the structural surface in the current frame; The length L of the structural surface trace is calculated using the following formula. t : L t =P ab ×L×cosθ / f; Where f is the focal length of the camera, θ is the shooting angle of the current frame, and L is the processed shooting distance of the current frame; The structural plane geometric parameters also include the structural plane opening, which is calculated as follows: The first and second edge feature points of the current frame in the image data after processing are identified by improving the YOLOv5 algorithm. The first and second edge feature points are the two edge points on both sides of the structural surface gap in the current frame. Calculate the pixel distance P between the first edge feature point and the second edge feature point along the vertical direction of the gap. d ; The structural plane opening W is calculated using the following formula: W=P d ×L×cosθ / f.

[0062] As an feasible approach, the structural surface geometric parameters also include the spacing between structural surfaces. The calculation process for the spacing between structural surfaces is as follows: By improving the YOLOv5 algorithm, the first and second center feature points of the current frame are identified, and the pixel distance P between them is calculated. c The first and second center feature points are the center reference points of two adjacent structural planes in the current frame; The spacing L between structural surfaces is calculated using the following formula. s : L s =P c ×L×cosθ / f.

[0063] As one feasible approach, the structural plane attitude parameters include the structural plane dip direction, and the calculation process for the structural plane dip direction is as follows: Obtain the inclination of the reference plane, and calculate the relative offset angle Δθ between the inclination of the reference plane and the trace of the structural surface according to the following formula: Δθ= arctan {(u1-u0) / (v1-v0)}-φ; Where (u0, v0) are the center pixel coordinates of the three fluorescent marker disks, (u1, v1) are the midpoint pixel coordinates of the structure surface trace in the current frame, and φ is the camera heading angle; The dip direction of the reference plane is corrected based on the relative offset angle Δθ to obtain the strike azimuth of the structural plane. The dip direction α of the structural plane is then calculated based on the strike azimuth. The calculation formula is as follows: ; Where α0 is the dip of the reference plane.

[0064] As an achievable method, the structural plane attitude parameters also include the structural plane dip angle, which is calculated as follows: Obtain the pixel height difference ΔP of the structure surface traces in the current frame of the processed image data. v ; Calculate the structural plane dip angle β using the following formula: β = arcsin (Δh / L h ) + β0-90°; Where Δh is the vertical height difference of the structural surface traces in space, according to the formula Δh = ΔP v Calculate L × cosθ / f, where L h Let L be the spatial horizontal length of the structural surface trace, and the formula L h = arctan (P L Calculate P using (×L×cosθ / f) ×cosβ0. L The structure surface trace is based on the pixel length of the current frame, and β0 is the spatial tilt angle of the reference plane.

[0065] As one possible implementation, the system further includes: The visualization module is used to obtain tunnel classification reports, surrounding rock quality grades, lithological information, structural surface parameter information and environmental parameter information in real time through the web interface, and to perform three-dimensional visualization display based on the above information.

[0066] This invention first utilizes a data acquisition module 401 with an acquisition device adapted to the tunnel environment to collect multi-source raw data covering the exposed section of the complete structural surface, ensuring the relevance and completeness of data acquisition and providing a high-quality data foundation for subsequent processing and analysis. The processing module 402, calculation and extraction module 403, and generation module 404 work collaboratively to automate the entire process from raw data preprocessing, key information extraction, surrounding rock classification calculation to report generation, replacing the cumbersome process of traditional manual operation and significantly improving the efficiency of tunnel geological parameter acquisition and surrounding rock classification. The specialized design of each module ensures the processing quality at every stage, effectively reducing errors caused by human intervention, improving the accuracy and reliability of surrounding rock classification results, and providing strong system support for the safe construction and scientific management of tunnel projects.

[0067] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or units may be electrical, mechanical, or other forms.

[0068] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0069] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0070] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0071] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for rapid acquisition of tunnel geological parameters and classification of surrounding rock, characterized in that, The method includes: The acquisition equipment, adapted to the tunnel environment, performs multi-source raw data acquisition on the target face or sidewall with planned geological measurement windows. The multi-source raw data includes raw image data, shooting distance, and corresponding inertial sensor data; the geological measurement window covers at least one complete exposed section of the structural surface. The multi-source raw data is preprocessed to obtain processed data, and lithological information and structural surface parameter information are extracted based on the processed data; The surrounding rock quality grade is obtained by calling a preset surrounding rock classification standard algorithm based on the lithological information and the structural surface parameter information to perform surrounding rock classification calculation. A tunnel classification report is generated based on the surrounding rock quality grade, the lithological information, and the structural surface parameter information.

2. The method according to claim 1, characterized in that, The preprocessing of the multi-source raw data to obtain processed data includes: The original image data is sequentially subjected to dehazing, illumination equalization, and distortion correction to obtain the processed image data. The shooting distance is subjected to distortion correction processing to obtain the processed shooting distance; The inertial sensing data is fused and calculated to obtain the shooting angle and three-dimensional attitude angle corresponding to the current captured image; The processed image data, the processed shooting distance, the shooting angle, and the three-dimensional attitude angle constitute the processed data.

3. The method according to claim 2, characterized in that: The acquisition device includes a mobile acquisition device and an auxiliary acquisition device. The mobile acquisition device is equipped with a camera, a laser ranging unit, and an inertial navigation unit. The camera is used to acquire image data of the target face or the target sidewall. The laser ranging unit is used to acquire the shooting distance between the camera and the target face or the target sidewall during the current shooting. The inertial navigation unit is used to acquire the raw inertial sensing data corresponding to the current shooting. The auxiliary acquisition device includes a supplementary lighting bracket and a pose correction device. The supplementary lighting bracket is used to provide illumination compensation for the camera. The pose correction device is located at the bottom center of the geological window and is used to provide a reference plane for image acquisition. The reference plane consists of three fluorescent marker discs that are fixedly connected to each other.

4. The method according to claim 3, characterized in that, The extraction of lithological information and structural surface parameter information based on the processed data includes: The processed image data is input into the trained lithology identification model for lithology identification processing, and the lithology information of the currently collected target face or target sidewall is output. The lithology information includes lithology type and rock mass integrity level. Based on the processed image data, the processed shooting distance, the shooting angle, and the three-dimensional attitude angle, the structural surface parameter information of the target face or the target sidewall is calculated. The structural surface parameter information includes structural surface geometric parameters and structural surface attitude parameters.

5. The method according to claim 4, characterized in that, The geometric parameters of the structural surface include the trace length of the structural surface, and the calculation process for the trace length of the structural surface is as follows: The improved YOLOv5 algorithm is used to identify the first and second trace length feature points in the current frame of the processed image data, and the pixel distance P between the first and second trace length feature points is calculated. ab The first trace length feature point and the second trace length feature point are the two endpoints of the exposed trace of the structural surface in the current frame; The length L of the structural surface trace is calculated according to the following formula. t : L t =P ab ×L×cosθ / f; Where f is the camera's calibrated focal length, θ is the shooting angle of the current frame, and L is the processed shooting distance of the current frame; The structural plane geometric parameters also include the structural plane opening, which is calculated as follows: The first and second edge feature points of the current frame in the processed image data are identified by the improved YOLOv5 algorithm. The first and second edge feature points are the two edge points on both sides of the structural surface gap in the current frame. Calculate the pixel distance P between the first edge feature point and the second edge feature point along the vertical direction of the gap. d ; The opening W of the structural surface is calculated according to the following formula: W=P d ×L×cosθ / f。 6. The method according to claim 5, characterized in that, The structural surface geometric parameters also include the structural surface spacing, and the calculation process for the structural surface spacing is as follows: By improving the YOLOv5 algorithm, the first and second center feature points of the current frame are identified, and the pixel distance P between them is calculated. c The first central feature point and the second central feature point are the central reference points of two adjacent structural planes in the current frame or the corresponding points of two adjacent sets of traces within the same structural plane. The spacing L between the structural surfaces is calculated using the following formula. s : L s =P c ×L×cosθ / f。 7. The method according to claim 6, characterized in that, The structural plane attitude parameters include the structural plane dip direction, and the calculation process for the structural plane dip direction is as follows: Obtain the inclination of the reference plane, and calculate the relative offset angle Δθ between the inclination of the reference plane and the trace of the structural surface according to the following formula: Δθ= arctan {(u1-u0) / (v1-v0)}-φ; Where (u0, v0) are the center pixel coordinates of the three fluorescent marker disks, (u1, v1) are the midpoint pixel coordinates of the structure surface trace in the current frame, and φ is the heading angle; The dip direction of the reference plane is corrected based on the relative offset angle Δθ to obtain the strike azimuth angle of the structural plane. The dip direction α of the structural plane is then calculated based on the strike azimuth angle using the following formula: ； Where α0 is the dip of the reference plane.

8. The method according to claim 1, characterized in that, The structural plane attitude parameters also include the structural plane dip angle, which is calculated as follows: Obtain the pixel height difference ΔP of the structural surface traces in the current frame of the processed image data. v ; According to the formula Δh = ΔP v Calculate the vertical height difference Δh of the structural surface traces in space using the formula: × L × cosθ / f; According to formula L h = arctan (P L Calculate the spatial horizontal length L of the trace line of the structure surface by calculating ×L×cosθ / f) ×cosβ0. h ; Calculate the structural plane dip angle β using the following formula: β = arcsin (Δh / L h ) + β0 -90°; If the calculated result of the inclination angle of the structural surface is negative, then the absolute value is taken and the direction is adjusted; Among them, P L The structure surface trace is based on the pixel length of the current frame, and β0 is the spatial tilt angle of the reference plane.

9. The method according to claim 1, characterized in that, The method further includes: The web interface acquires the tunnel classification report, the surrounding rock quality grade, the lithological information, the structural surface parameter information, and the environmental parameter information in real time, and displays them in three dimensions based on the above information.

10. A rapid acquisition system for tunnel geological parameters and a system for classifying surrounding rock, characterized in that, include: The acquisition module is used to acquire multi-source raw data from the target face or sidewall with geological survey windows based on acquisition equipment adapted to the tunnel environment. The multi-source raw data includes raw image data, shooting distance, and corresponding inertial sensor data; the geological survey window covers at least one complete exposed section of a structural surface. The processing module is used to preprocess the multi-source raw data to obtain processed data; The calculation and extraction module is used to extract lithological information and structural surface parameter information based on the processed data; It is used to call a preset surrounding rock classification standard algorithm to perform surrounding rock classification calculation based on the lithology information and the structural surface parameter information, and obtain the surrounding rock quality grade; The generation module is used to generate a tunnel classification report based on the surrounding rock quality grade, the lithological information, and the structural surface parameter information.

Images