A geological disaster and engineering safety monitoring intelligent early warning method and system

Abstract

The present application relates to an intelligent early warning method and system for monitoring geological disasters and engineering safety, and relates to the field of geological disaster early warning technology. It solves the problem that setting a single threshold to judge the safety of tunnel engineering ignores the correlation between different monitoring points and the changing trend of monitoring data, which may lead to false alarms or missed reports. The method includes: obtaining all monitoring data of geological disaster-related parameters of tunnel engineering; analyzing and determining the warning levels of different locations according to the preset data interval range of geological disaster-related parameters at different locations and the correspondence between the data change rate interval range and the warning level; analyzing and determining the linkage response plan inside and outside the tunnel according to the correspondence between the warning level and the linkage response plan inside and outside the tunnel and the analyzed and determined warning level; executing the linkage response plan inside and outside the tunnel. The present application has the following effects: improving the prevention, monitoring and emergency response capabilities of geological disasters and engineering safety accidents.

Description

Technical Field

[0001] The present invention relates to the field of geological disaster early warning technology, and in particular to an intelligent early warning method and system for geological disaster and engineering safety monitoring. Background Art

[0002] The geological disaster monitoring and early warning system is a tool for disaster prevention, mitigation and avoidance. Its purpose is to obtain disaster object information as soon as possible so as to respond quickly to dangerous situations and reduce the threat of disaster objects to human life, health and property.

[0003] At present, the geological disaster monitoring and early warning system mainly collects data on disaster-related parameters for relevant engineering projects by deploying professional instruments and equipment, and then conducts separate analysis based on the data of various disaster-related parameters and displays them to professionals.

[0004] At present, tunnel engineering monitoring mainly focuses on the real-time data of a single monitoring point, such as displacement and stress, and sets a single threshold to judge the safety of the tunnel project. When the monitoring data exceeds the preset threshold, a warning is issued. However, this method ignores the correlation between different monitoring points and the changing trend of monitoring data, which may lead to false alarms or omissions. Summary of the invention

[0005] In order to improve the prevention, monitoring and emergency response capabilities of geological disasters and engineering safety accidents, the present application provides a geological disaster and engineering safety monitoring intelligent early warning method and system.

[0006] In the first aspect, the present application provides an intelligent early warning method for geological disaster and engineering safety monitoring, which adopts the following technical solutions:

[0007] An intelligent early warning method for geological disaster and engineering safety monitoring, comprising:

[0008] Obtain all monitoring data of tunnel engineering geological disaster related parameters and the change rate of the monitoring data, wherein the related parameters are various parameters of the tunnel structure and the surrounding environment, including but not limited to displacement, temperature, humidity, vibration, strain and settlement displacement;

[0009] According to the preset data interval range of geological disaster-related parameters at different locations and the corresponding relationship between the data change rate interval range and the warning level, the warning level at different locations is analyzed and determined;

[0010] Analyze and determine the tunnel internal and external linkage response plan based on the corresponding relationship between the warning level and the tunnel internal and external linkage response plan and the analyzed and determined warning level;

[0011] Execute the joint response plan inside and outside the tunnel.

[0012] By adopting the above-mentioned technical scheme, the warning level is determined by comprehensively analyzing the data range and data change rate of disaster-related parameters of different engineering projects. Dual control indicators are used to determine the warning level, which improves the accuracy and reliability of the determination. In addition, an appropriate internal and external linkage response plan for the tunnel will be adopted according to the warning level determined by the analysis, thereby improving the prevention, monitoring and emergency response capabilities of geological disasters and engineering safety accidents.

[0013] Optionally, all monitoring data of tunnel engineering geological disaster related parameters are obtained, including:

[0014] Obtain the warning levels of the target tunnel project within a preset time range, and analyze the number of times different warning levels occur;

[0015] According to the corresponding relationship between the warning level and the number of indicators and the number of occurrences of different warning levels obtained through analysis, the total number of indicators is obtained through analysis and calculation;

[0016] Analyze and determine the deadline for collecting all monitoring data based on the correspondence between the total indicator number interval range that the total indicator number falls into within the preset time range and the deadline for collecting all monitoring data;

[0017] Based on the analysis, the deadline for collecting all monitoring data is determined, and all monitoring data related to tunnel engineering geological hazards are collected.

[0018] By adopting the above technical solutions, the warning level and index number of geological disasters in tunnel engineering can be intelligently analyzed, and the deadline for monitoring data collection can be accurately determined, thus achieving comprehensive and efficient monitoring of the safety status of the tunnel. This approach not only improves the pertinence and timeliness of data collection, but also can timely discover potential geological disaster risks and provide strong data support for prevention and control measures. At the same time, the intelligent data processing process has also greatly improved work efficiency, reduced labor costs, and ensured the long-term safe operation of tunnel engineering.

[0019] Optionally, all monitoring data of tunnel engineering geological disaster related parameters are obtained, including:

[0020] Analyze whether the monitoring location points of the target tunnel project have received warnings exceeding the preset warning level;

[0021] If not, the collection period is determined based on the corresponding relationship between the monitoring location and the collection period, and all monitoring data of the corresponding monitoring location are collected and acquired according to the collection period;

[0022] If yes, then analyze whether the monitoring location point receives a warning exceeding the preset warning level next time;

[0023] If not, all monitoring data of the corresponding monitoring location point are collected and acquired according to the collection period of the corresponding monitoring location point when no warning of the preset warning level is received;

[0024] If yes, the collection period is determined based on the obtained warnings exceeding the preset warning level, the corresponding relationship between the monitoring location points and the collection period, and all monitoring data of the corresponding monitoring location points are collected and obtained according to the collection period.

[0025] By adopting the above technical solutions, the tunnel engineering geological disaster monitoring data is analyzed in detail, the warning level of the monitoring location is intelligently determined, and the data collection period is flexibly adjusted accordingly. This dynamic adjustment mechanism not only ensures the timeliness and accuracy of data collection, but also responds quickly to potential geological disaster risks. By optimizing the data collection strategy, the solution effectively improves the efficiency and reliability of tunnel engineering safety monitoring and provides strong support for the prevention of geological disasters.

[0026] Optionally, the method further includes the following steps in parallel with obtaining all monitoring data of parameters related to tunnel engineering geological hazards:

[0027] Regularly collect point cloud data and images of the tunnel and its environment within a preset distance;

[0028] The collected point cloud data is preprocessed, followed by data registration, aligning the data sets of multiple perspectives into a unified coordinate system, and then using MeshLab and PCL tools to construct a 3D point cloud model through point cloud triangulation. Data preprocessing includes denoising, filtering, and data pruning.

[0029] Create a 3D geometric model based on the 3D point cloud model;

[0030] Combine the 3D geometric model with the image to form a real scene model;

[0031] The collected images are used to construct a panoramic expansion map through a panoramic expansion map construction method, and a mapping relationship between the real scene and the panorama is established to obtain a transformation function between the three-dimensional space and the expansion plane map. According to the transformation function, an information display space that is a unified and coordinated information display space of the physical space and the apparent plane is constructed, and the previous information display space is replaced by the constructed information display space that is a unified and coordinated information display space of the physical space and the apparent plane.

[0032] By adopting the above technical solutions, point cloud data and images of the tunnel and its environment within a preset distance are regularly collected to construct a three-dimensional real-life model and panoramic display space, thus achieving all-round and high-precision monitoring of the tunnel project. Combining advanced data processing technology and panoramic expansion map construction methods, it not only improves the visualization of data, but also strengthens the collaborative display capability of information. This innovative method significantly improves the efficiency and accuracy of tunnel project monitoring, and provides more intuitive and comprehensive data support for the prevention of geological disasters.

[0033] Optionally, the method further includes a step after constructing the information display space in which the entity space and the apparent plane are coordinated in a unified manner according to the transformation function, which is as follows:

[0034] Obtaining defect-related information of the tunnel project obtained by non-destructive testing technology, wherein the defect-related information includes shape, location, length, spacing and area;

[0035] When displaying the panoramic view, various defects are outlined by lines and polygons, and specific data annotations are made for the parameters involved in the defect-related information.

[0036] By adopting the above technical solutions, combined with non-destructive testing technology and panoramic display, the intuitive and efficient presentation of tunnel engineering defect information is achieved. In the panoramic view, various defects are clearly outlined by lines and polygons, and detailed data annotations are attached, so that the shape, location, length, spacing and area of ​​the defects are clear at a glance. This innovative display method not only improves the readability and analysis efficiency of the data, but also helps to timely discover and deal with potential safety hazards in tunnel engineering, providing a strong guarantee for the long-term safe operation of the tunnel.

[0037] Optionally, it also includes steps of outlining various defects by lines and polygons when displaying the panoramic image, and simultaneously annotating specific data of parameters involved in the defect-related information, as follows:

[0038] Obtain the location of the monitoring point where a warning exceeding a preset warning level occurs;

[0039] According to the acquired location of the monitoring point, the relevant parameter data of the corresponding monitoring point is synchronously displayed and presented in the real-life model and the panoramic expansion map as the marked location point. When the panoramic map is displayed, various defects within a preset distance from the monitoring point are outlined by lines and polygons, and specific data annotations are made for the parameters involved in the defect-related information.

[0040] By adopting the above technical solutions, through the combination of real-time monitoring and panoramic display, efficient monitoring of tunnel engineering defects and geological disaster risks is achieved. In the panoramic view, not only the monitoring points and their surrounding defects that exceed the preset warning level are clearly outlined, but also the relevant parameter data are displayed simultaneously. This innovative display method greatly improves the intuitiveness and analysis efficiency of the data, helps to timely discover and deal with potential safety hazards, and provides more comprehensive and accurate protection for the safe operation of tunnel engineering.

[0041] Optionally, it also includes the steps of outlining various defects within a preset distance from the monitoring point by lines and polygons when displaying the panoramic image, and annotating specific data of parameters involved in the defect-related information simultaneously, as follows:

[0042] Obtain an annotated image constructed based on the point cloud data and images of the tunnel and its environment within a preset distance range collected regularly each time in history. The annotated image includes an information display space and defect annotation information within a preset distance range from the monitoring point. The defect annotation information within a preset distance range from the monitoring point includes various defects within the preset distance range from the monitoring point outlined by lines and polygons, as well as specific data annotations for parameters involved in defect-related information;

[0043] Using feature extraction algorithms, stable feature points are extracted from the earlier and updated images and accurately matched to ensure the consistency of the two images in space and scale;

[0044] Based on the matching results of feature points, the updated image is accurately aligned with the earlier image using a preset image transformation model;

[0045] Based on the aligned images, a pixel-by-pixel difference detection algorithm is calculated to generate a difference image that highlights areas of structural changes or potential disease.

[0046] Segment the difference image using a preset threshold to clearly distinguish the changed area from the background area, and further process the segmented changed area using morphological operations, including but not limited to dilation and erosion;

[0047] The processed changed area is fused with the original image for display, and the specific location and development trend of the structural changes or potential diseases are indicated by preset marks or colors.

[0048] By adopting the above technical solutions and refined image processing technology, efficient detection and visualization of tunnel structure changes and potential diseases are achieved. Through feature point matching and image transformation, accurate alignment of images at different time points is ensured, and then the change area is accurately identified through the difference detection algorithm. Combining morphological operations and fusion display, it not only improves the accuracy of disease identification, but also more intuitively presents the location and development trend of the disease, providing strong support for tunnel maintenance and safety management.

[0049] In the second aspect, the present application provides an intelligent early warning system for geological disaster and engineering safety monitoring, which adopts the following technical solutions:

[0050] An intelligent early warning system for geological disaster and engineering safety monitoring, comprising:

[0051] An acquisition module is used to obtain all monitoring data of tunnel engineering geological disaster related parameters and the change rate of the monitoring data;

[0052] The analysis module analyzes and determines the warning levels at different locations based on the preset data interval ranges of geological disaster-related parameters at different locations and the corresponding relationship between the data change rate interval ranges and the warning levels; analyzes and determines the linkage response plans inside and outside the tunnel based on the corresponding relationship between the warning level and the linkage response plans inside and outside the tunnel and the analyzed and determined warning level.

[0053] Execution module, executes the linkage response plan inside and outside the tunnel.

[0054] By adopting the above technical solutions and integrating the three modules of acquisition, analysis and execution, efficient monitoring and response of geological disasters in tunnel engineering can be achieved. The acquisition module comprehensively captures monitoring data and change rates to ensure real-time and accurate data; the analysis module intelligently determines the warning level and determines the response plan to improve decision-making efficiency; the execution module quickly executes the response plan to ensure engineering safety. Overall, the system uses intelligent means to improve the monitoring and response capabilities of geological disasters, providing solid protection for the safety of tunnel engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a flow chart of an intelligent early warning method for geological disaster and engineering safety monitoring according to an embodiment of the present application.

[0056] Figure 2 It is a flowchart of another embodiment of the present application for obtaining all monitoring data of parameters related to geological disasters in tunnel engineering.

[0057] Figure 3 It is a flowchart of another embodiment of the present application for obtaining all monitoring data of parameters related to geological disasters in tunnel engineering.

[0058] Figure 4It is a schematic diagram of the device involved in obtaining tunnel deformation parameters in an embodiment of the present application.

[0059] Figure 5 This is a system block diagram of a geological disaster and engineering safety monitoring intelligent early warning system according to another embodiment of the present application. DETAILED DESCRIPTION

[0060] The present application is further described in detail below in conjunction with the accompanying drawings.

[0061] Reference Figure 1 , is a geological disaster and engineering safety monitoring intelligent early warning method disclosed in this application, comprising:

[0062] Step S100, obtaining all monitoring data of tunnel engineering geological disaster related parameters.

[0063] Among them, the relevant parameters are various parameters of the tunnel structure and the surrounding environment, including but not limited to displacement, temperature, humidity, vibration, strain and settlement displacement. Specifically, for displacement monitoring, high-precision rangefinders or laser scanners can be used to capture tiny movements of the tunnel structure. In terms of temperature and humidity, temperature and humidity sensors can be deployed, which can transmit data in real time to reflect the environmental conditions in the tunnel. Vibration monitoring can be achieved by installing seismic detectors or accelerometers, which can keenly capture any vibrations around the tunnel. As for settlement displacement, regular observations can be made using a level or satellite positioning system to promptly detect surface settlement. Strain monitoring relies on strain gauges, which can accurately measure the deformation of the tunnel structure under stress.

[0064] In addition, the relevant parameter can also be structural durability. Specifically, a full set of durability sensors are placed in the pipe segment by pre-embedded sensors. The full set of durability sensors includes Ag / AgCl electrodes, MnO2 solid reference electrodes, anode ladders, temperature and humidity sensors, and strain gauges.

[0065] Reference Figure 4 The relevant parameters may also be tunnel deformation, which includes segment joint deformation and segment displacement. Segment joint deformation can be monitored by collecting the deformation state of the joint through strain gauges set at the joint position, and segment displacement can be collected and acquired through a multi-point image displacement system.

[0066] Reference Figure 4 The multi-point image displacement system is generally equipped with a fill light, an infrared camera, and a target. The specific description is as follows:

[0067] Fill lights play a vital role in multi-point image displacement systems. As the light inside the tunnel may be insufficient, fill lights can provide stable and uniform lighting conditions to ensure that the camera can capture clear and accurate images.

[0068] The infrared camera is one of the core components of the multi-point image displacement system. It can capture the image of the target in various lighting conditions, including at night or in low-light environments. Through infrared technology, the camera can penetrate certain obstacles, such as smoke or mist, ensuring that the position change of the target can be accurately captured in any situation.

[0069] Targets act as key reference points in the multi-point image displacement system. They are usually markers with specific shapes, colors or patterns, and are placed at key locations inside the tunnel. The camera captures the images of these targets and uses image processing technology to analyze their position changes, thereby automatically collecting the displacement of the pipe segments.

[0070] Step S200, analyzing and determining the warning levels of different locations based on the preset data interval ranges of geological disaster-related parameters at different locations and the corresponding relationship between the data change rate interval ranges and the warning levels.

[0071] Among them, the warning level can be divided into four levels according to the type, impact range and degree of harm of geological disasters, as follows: Level 1 (particularly serious), Level 2 (serious), Level 3 (relatively serious) and Level 4 (general). For geological disaster-related parameters in different locations, we preset the data interval range and data change rate interval range. These parameters include but are not limited to: displacement, temperature, humidity, seismic wave velocity, soil displacement, groundwater level, and rainfall. Each parameter has its corresponding safety range, warning range, and danger range.

[0072] When analyzing and determining the warning level of different locations, a comprehensive evaluation method can be used. Specifically, the system will score each parameter based on the real-time monitored parameter data and its preset range. The higher the score, the more dangerous the current state of the parameter. Then, the system will weight the scores of each parameter and get the comprehensive score of the location.

[0073] Based on the comprehensive score, we can determine the warning level of the location. For example, if the comprehensive score exceeds a certain set threshold, it is judged as a level 1 warning, which means that there is a particularly serious geological disaster risk at the location and emergency measures need to be taken immediately.

[0074] Step S300: Analyze and determine the tunnel internal and external linkage response scheme according to the corresponding relationship between the warning level and the tunnel internal and external linkage response scheme and the analyzed and determined warning level.

[0075] Specifically, according to the four levels of warning levels mentioned in the description of step S200, a linkage emergency mechanism can be formed. The platform will display red, orange, yellow and blue indicator lights corresponding to the four warning levels. The warning level and content, response measures, supporting facilities (base station indicator lights, entrance whistles, information signs), personnel card prompts (vibration, instructions, sounds, text messages), and personnel or unit responsibilities are integrated together to build a linkage response plan inside and outside the tunnel.

[0076] The specific tunnel internal and external linkage response plan can be constructed as follows: When the warning level is level 4 (general), that is, when the platform displays a blue indicator light, it indicates that the tunnel is in good working condition. When the warning level is level 3 (heavy), that is, when the platform displays a yellow indicator light, it indicates that the tunnel is in general working condition. The manager can further analyze the cause of the abnormality by adjusting the monitoring frequency, and organize personnel to manually check the source of danger. When the warning level is level 2 (serious), that is, when the platform displays an orange indicator light, it indicates that the tunnel is in poor working condition. In order to avoid further danger, the manager should organize a professional inspection team to inspect the tunnel and check the risk sources one by one. When the warning level is level 1 (particularly serious), that is, when the platform displays a red indicator light, it indicates that the tunnel is in very poor working condition and is very likely to cause a safety accident. At this time, the area indicated by the red indicator light should be closed, and a professional team should be sent to evaluate and select appropriate methods for reinforcement.

[0077] Step S400, executing the tunnel internal and external linkage response plan.

[0078] exist Figure 1 In step S300, it is also necessary to further consider the period for obtaining all monitoring data of tunnel engineering geological disaster related parameters. The warning level and index number of tunnel engineering geological disasters should be comprehensively considered. Figure 2 The illustrated embodiment is described in detail.

[0079] Reference Figure 2 , obtain all monitoring data of tunnel engineering geological disaster related parameters including:

[0080] Step S310, obtaining the warning level of the target tunnel project within a preset time range, and analyzing the number of times different warning levels occur.

[0081] Specifically, the system records and stores each warning level issued and its corresponding timestamp. By setting a specific time range, all warning records within that time period can be filtered out, and then different warning levels can be counted and analyzed. For example, suppose we set the most recent week as the preset time range, and the system issued a total of 10 warnings during that week. Among them, level 1 warnings (the highest level) appeared 3 times, level 2 warnings appeared 5 times, and level 3 warnings (lower levels) appeared 2 times. Through such data analysis, we can clearly understand the risk status of tunnel engineering geological disasters and the triggering frequency of warnings at various levels during this week.

[0082] Step S320, analyzing and calculating the total number of indicators according to the corresponding relationship between the warning level and the number of indicators and the number of occurrences of different warning levels obtained through analysis.

[0083] First, identify the number of indicators corresponding to each warning level, multiply the number of times each warning level occurs by its corresponding indicator number, and then add these products to get the total indicator number.

[0084] Step S330, analyzing and determining the deadline for collecting all monitoring data according to the correspondence between the total indicator number interval range that the total indicator number falls into within the preset time range and the deadline for collecting all monitoring data.

[0085] Specifically, we determine which interval the total indicator falls into based on its actual value within the preset time range. Finally, based on the interval the total indicator falls into, we will consult the preset correspondence table to determine the deadline for collecting all monitoring data. The setting of different total indicator intervals is based on historical data, expert evaluation, and actual project needs.

[0086] Step S340, determining the time limit for collecting all monitoring data based on the analysis, and collecting all monitoring data of parameters related to tunnel engineering geological hazards.

[0087] exist Figure 1 In step S300, it is necessary to further analyze the tunnel engineering geological disaster monitoring data, intelligently determine the warning level of the monitoring location point, and flexibly adjust the data collection period accordingly. Figure 2 The illustrated embodiment is described in detail.

[0088] Reference Figure 3 , obtain all monitoring data of tunnel engineering geological disaster related parameters including:

[0089] Step S3a0, analyzing whether the monitoring location of the target tunnel project has received a warning exceeding a preset warning level. If not, execute step S3b0; if yes, execute step S3c0.

[0090] Specifically, first, the system sets a preset warning level for each monitoring location. This level is usually determined based on factors such as the geological conditions, historical data, and expert evaluation of the location. When the monitoring data reaches or exceeds this preset level, the system will issue a corresponding warning. For example, suppose we preset a level 2 warning level at a key monitoring location in a tunnel project. This means that when the monitoring data at the location reaches the threshold of the level 2 warning, the system will issue a warning.

[0091] Step S3b0: Analyze and determine the collection period according to the correspondence between the monitoring location points and the collection period, and collect and obtain all monitoring data of the corresponding monitoring location points according to the collection period.

[0092] Different monitoring locations have different collection periods, mainly considering that different monitoring locations have different requirements for data collection frequency due to their geological environment, structural characteristics, historical data performance, etc. For example, some key nodes or vulnerable areas may require more intensive monitoring.

[0093] Step S3c0, analyzing whether the monitoring location receives a warning exceeding the preset warning level next time. If not, execute step S3d0; if yes, execute step S3e0.

[0094] Step S3d0, collecting and acquiring all monitoring data of the corresponding monitoring location point according to the collection period of the corresponding monitoring location point when no warning of a preset warning level is received.

[0095] The setting of step S3d0 is the same as that of step S3b0, which will not be described in detail here.

[0096] Step S3e0: Analyze and determine the collection period based on the obtained correspondence between the warning exceeding the preset warning level, the monitoring location point and the collection period, and collect and obtain all monitoring data of the corresponding monitoring location point according to the collection period.

[0097] The analysis and determination of the collection period is as follows: the warning level and monitoring location corresponding to the obtained warning exceeding the preset warning level are used as query objects, and the collection period is queried from a preset database storing the corresponding relationship between the obtained warning exceeding the preset warning level, the monitoring location and the collection period. For example, suppose that at a key monitoring location of a tunnel project, the system detects data exceeding the second-level preset warning level. We adjusted the collection period from once a week to once a day.

[0098] exist Figure 1In step S100, while obtaining all monitoring data of tunnel engineering geological disaster related parameters, it is also necessary to consider displaying a three-dimensional real-scene model and a panoramic display space to facilitate the staff to intuitively control the internal conditions of the tunnel. Specific details are described in detail with reference to the following embodiment.

[0099] A method for intelligent early warning of geological disasters and engineering safety monitoring also includes the steps of obtaining all monitoring data of tunnel engineering geological disaster related parameters in parallel, as follows:

[0100] Step Sa00, regularly collect point cloud data and images of the tunnel and its environment within a preset distance range.

[0101] The period for regularly collecting point cloud data and images of the tunnel and its environment within the preset distance range can be 1 day or 3 days, or other set periods. The preset distance range is to set a reasonable radius range with the tunnel as the center, such as 500 meters, 1 kilometer, etc., and include all environmental factors within this range in the definition of "environment within the preset distance range".

[0102] The regular collection of point cloud data can adopt advanced 3D laser scanning technology. By setting up scanners at key locations inside and outside the tunnel, a full range of scanning operations can be performed to accurately capture detailed 3D coordinate data of the tunnel surface and its surrounding environment. The collection of images of the tunnel and its environment within a preset distance range can be achieved by combining high-definition cameras and drone technology. The specific methods are as follows: First, by installing high-definition cameras at key locations inside and outside the tunnel, the image information inside the tunnel can be monitored and captured in real time. These cameras can have night vision functions to ensure that clear images can be obtained even in low light conditions. At the same time, by setting a reasonable shooting interval and storage mechanism, these image data can be saved regularly for subsequent analysis and comparison. Secondly, drone technology can be used to obtain images of the outside of the tunnel and the surrounding environment from an aerial perspective. Drones can be equipped with high-definition cameras or video cameras to fly and shoot on a preset route, thereby capturing key information such as the facade of the outside of the tunnel, the surrounding terrain and landforms, and potential signs of geological disasters. These aerial image data combined with the images captured by ground cameras can form a more three-dimensional and comprehensive image database of the tunnel and its surrounding environment.

[0103] In step Sb00, the collected point cloud data is preprocessed, followed by data registration, where the data sets of multiple perspectives are aligned into a unified coordinate system, and then MeshLab and PCL tools are used to construct a three-dimensional point cloud model through point cloud triangulation. Data preprocessing includes denoising, filtering, and data pruning.

[0104] Data registration is to accurately align point cloud data sets from different perspectives or different time points into a unified coordinate system. This involves finding the correspondence between the data sets and achieving accurate alignment between them by calculating the transformation matrix. Specifically, after data preprocessing, you can consider using the Structure from Motion (SFM) method to reconstruct the sparse point cloud and realize the conversion between the actual detection coordinates and the reconstructed coordinates. The conversion formula is:

[0105] ;

[0106] Among them, R is the rotation matrix, t is the translation vector, s is the scaling factor, To control the measured value, is the coordinate of the i-th control point in the detection coordinate system after transformation, is the transformation error of the i-th control point, that is, the coordinates of the control point in the reconstructed coordinate system are transformed to the detection coordinate system through the three parameters R, t, and s and the control measurement value The distance between them, n is the number of control points, n ≥ 3.

[0107] Among them, MeshLab is an open source system for processing and editing unstructured 3D triangular meshes. PCL is an independent, large-scale, open source project for 2D / 3D image and point cloud processing.

[0108] Regarding denoising, denoising removes noise points generated by measurement errors or environmental factors from point cloud data to improve data quality. Regarding data pruning, point cloud data is trimmed according to needs to remove irrelevant areas or highlight key parts. The process of constructing a 3D point cloud model through point cloud triangulation is as follows: After completing data preprocessing and alignment, use professional tools such as MeshLab or PCL to convert discrete point cloud data into a continuous 3D mesh model. This process is called point cloud triangulation.

[0109] Step Sc00, creating a three-dimensional geometric model based on the three-dimensional point cloud model.

[0110] Specifically, the process of creating a 3D geometric model is as follows: 1. Point cloud data processing: First, pre-process the acquired 3D point cloud model, including removing noise, filling missing data, smoothing the surface, etc. These operations can improve the quality of point cloud data and provide an accurate basis for subsequent geometric model creation. 2. Geometric shape extraction: Next, use professional 3D modeling software or algorithms to extract basic geometric shapes from point cloud data. These shapes may include planes, surfaces, cylinders, spheres, etc., which together constitute the basic structure of the object. 3. Model construction and optimization: After extracting the basic geometric shapes, further build and optimize the model. This includes adjusting the size, position and orientation of the shapes to ensure that they can be accurately spliced ​​together to form a complete 3D geometric model.

[0111] Step Sd00, combining the three-dimensional geometric model with the image to form a real scene model;

[0112] Specifically, the process of forming a real scene model is as follows: Use professional 3D modeling software to accurately align the geometric model with the image. This usually involves adjusting the position, rotation and scaling of the model to ensure that it fits perfectly with the image data. Next, perform texture mapping. This process attaches image data (such as photos, texture maps, etc.) to the surface of the geometric model to give the model a realistic visual appearance. By fine-tuning the coordinates and mapping method of the texture, you can ensure that the texture blends perfectly with the surface of the model. Detail optimization: After the real scene model is generated, further detail optimization is performed. This includes adjusting the lighting and shadow effects to enhance the realism of the model; fixing possible texture seams or distortion problems to ensure the visual continuity of the model. Finally, export the generated real scene model to common 3D file formats (such as OBJ, FBX, etc.) for viewing, editing and application in other software or platforms.

[0113] Step Se00, construct a panoramic unfolded map of the collected images through a panoramic unfolded map construction method, and establish a mapping relationship between the real scene and the panorama, obtain the transformation function of the three-dimensional space and the unfolded plane map, and construct an information display space that is a unified and coordinated information display space of the physical space and the apparent plane according to the transformation function, and replace the previous information display space with the constructed information display space that is a unified and coordinated information display space of the physical space and the apparent plane.

[0114] Specifically, the process of constructing and updating the information display space that is unified and coordinated between the physical space and the apparent plane is as follows: 1. Construction of a panoramic expansion map: First, the panoramic expansion map construction method is used to process the collected images to generate a panoramic expansion map, wherein the panoramic expansion map construction method can be the method of Zhu et al., and the following methods can also be used: a panoramic expansion map construction method based on spherical mapping, a panoramic expansion map construction method based on cube mapping, a panoramic expansion map construction method based on cylindrical mapping, and a panoramic expansion map construction based on multi-view geometry. 2. Establishing the mapping relationship between the real scene and the panorama: Then, through precise algorithms and technical means, a mapping relationship between the real scene and the panorama is established. 3. Obtaining the transformation function: On the basis of the establishment of the mapping relationship, the transformation function between the three-dimensional space and the expansion plane is further derived. This function describes how to convert from a three-dimensional space to a two-dimensional plane, and how to restore from a two-dimensional plane to a three-dimensional space. 4. Constructing a unified and coordinated information display space: Using the above transformation function, we can construct an information display space that is unified and coordinated between the physical space and the apparent plane. In this space, the real scene and the panorama can be seamlessly connected to jointly present rich spatial information. 5. Update the information display space: Finally, replace the previous information display space with the newly constructed information display space to ensure the timeliness and accuracy of the information.

[0115] The transformation function from three-dimensional space to the unfolded plane graph is as follows:

[0116] ;

[0117] in, and η represent the two-dimensional coordinate system on the panorama. In this system, It usually represents the position in the horizontal direction, while η represents the position in the vertical direction. i and j are the discrete coordinate systems of the unfolded plane graph, where i represents the row coordinate, j represents the column coordinate, X, Y, and Z are coordinate parameters in the three-dimensional space, which respectively represent the position information in the three directions in the three-dimensional space. X usually represents the displacement in the horizontal direction, Y represents the displacement in the vertical direction, and Z represents the displacement in the depth direction.

[0118] When browsing a panorama, the panorama coordinates corresponding to the cursor point can be calculated through the cursor coordinate position in the panorama browsing window and the current offset and zoom ratio ( , η ) and its row and column coordinates ( i, j ). Using the unfolded graph coordinates of the four adjacent points on the row and column and their corresponding spatial grid coordinates, the transformation function from the unfolded plane graph to the three-dimensional space can be obtained, as follows:

[0119] .

[0120] Through the derived transformation function, we can convert between the three-dimensional tunnel space and the two-dimensional panorama. Finally, we construct an information display space that unifies and coordinates the physical tunnel space and the apparent plane.

[0121] After step Se00, it is also necessary to consider that while displaying the information display space in a unified and coordinated manner between the physical space and the apparent plane, it is also necessary to consider presenting the tunnel engineering defect information intuitively and efficiently, which is described in detail with reference to the embodiment shown below.

[0122] A method for intelligent early warning of geological disasters and engineering safety monitoring also includes steps after constructing an information display space that is unified and coordinated between a physical space and an apparent plane according to a transformation function, which are as follows:

[0123] Step Sf00, obtaining defect-related information of the tunnel engineering through non-destructive testing technology.

[0124] Among them, the non-destructive testing technology can be ground penetrating radar technology, and its detection method is an electromagnetic technology that accurately identifies and locates invisible targets or interfaces between different media inside the detected object based on the electrical property differences of the detected medium. Ground penetrating radar follows the principle of geometric optics to locate the target or medium interface by electromagnetic wave reflection, and determines the distance of the target or medium interface based on the two-way travel time of the echo and its propagation speed in the corresponding medium, and then comprehensively determines the physical characteristics of the target through the electrical and physical property differences of the medium. The propagation speed in the medium and the distance to the target or medium interface are calculated as follows:

[0125] ;

[0126] ;

[0127] In the formula, is the distance between the target object and the contact surface; t is the time interval from the emission to the reception of electromagnetic waves; v is the propagation speed of electromagnetic waves in rock and soil; x is the distance between the transmitting and receiving antennas; c is the propagation speed of electromagnetic waves in vacuum; is the relative dielectric constant of the medium.

[0128] Defect-related information for tunnel projects Defect-related information includes shape, location, length, spacing and area.

[0129] The process of obtaining defect-related information of tunnel projects through ground penetrating radar technology is as follows: 1. Use ground penetrating radar to scan the tunnel in all directions to ensure that potential defects are captured without blind spots. Preprocess the collected raw data, including denoising and filtering, to improve data quality; 2. Convert the preprocessed data into two-dimensional or three-dimensional images for intuitive analysis and identification of defects. Use image recognition technology to automatically mark suspected defect areas to reduce manual intervention. 3. Extract key features such as shape, position, length, spacing and area of ​​defects through image processing technology. Use machine learning or deep learning algorithms to classify the extracted features and accurately identify various types of defects. 4. Combine the CAD drawings or three-dimensional models of the tunnel project to achieve accurate positioning of defects. Quantitatively analyze the defects, including calculating the area and volume of the defects, to provide data support for maintenance and reinforcement. 5. Automatically generate a detailed defect detection report, including key information such as defect type, location, and size.

[0130] Step Sg00, when displaying the panoramic image, various defects are outlined by lines and polygons, and specific data are simultaneously marked on the parameters involved in the defect-related information.

[0131] Specifically, the process is as follows: 1. Defect outline: Use lines (such as straight lines, curves, etc.) and polygons (such as rectangles, circles, custom polygons, etc.) to accurately outline the outline of the defect on the panorama. The color, thickness and style of the lines and polygons can be customized according to the type or severity of the defect. 2. Data annotation: Add specific data annotations next to or above the outlined defects, including key parameters such as the defect type, size, and location. Data annotations can be in the form of text, numbers, or symbols. 3. Synchronous update: When the panorama is updated or the defect status changes, the line and polygon outlines and data annotations should be updated synchronously.

[0132] For example, let's take a panoramic view of the interior of a tunnel. Suppose a crack defect is found on the top of the tunnel. First, use the line or polygon tool to outline the crack on the panoramic view. You can draw it accurately according to the direction and shape of the crack. Then, add data annotations next to the crack outline.

[0133] After step Sg00, it should be considered that in the panoramic view, not only the monitoring points exceeding the preset warning level and their surrounding defects are clearly outlined, but also the relevant parameter data are simultaneously displayed, which is described in detail with reference to the embodiment shown below.

[0134] A method for intelligent early warning of geological disasters and engineering safety monitoring also includes steps of outlining various defects by lines and polygons when displaying a panoramic image, and simultaneously marking specific data on parameters involved in defect-related information, which are as follows:

[0135] Step Sh00, obtaining the location of the monitoring point where a warning exceeding a preset warning level occurs.

[0136] Step Si00, based on the acquired location of the monitoring point as the marked location point, the relevant parameter data of the corresponding monitoring point is synchronously displayed in the real-life model and the panoramic expansion map, and when displaying the panoramic map, various defects within a preset distance from the monitoring point are outlined by lines and polygons, and specific data annotations are made for the parameters involved in the defect-related information.

[0137] The specific process is as follows: 1. Synchronous display of monitoring point positions: Multiple monitoring points are arranged inside the tunnel to obtain accurate location information of each monitoring point. These location information are synchronized to the real-life model and the panoramic expansion map, and presented in the form of marked points. 2. Display of relevant parameter data: The parameter data related to each monitoring point is displayed next to the marked point, such as temperature, humidity, pressure, gas concentration, structural stress, etc. Ensure that the data is updated in real time to reflect the latest situation inside the tunnel in a timely manner. 3. Defect delineation and data annotation: According to the preset distance range (such as XX meters in radius with the monitoring point as the center), various defects within the range are automatically detected and outlined on the panoramic map, such as cracks, water leakage, structural deformation, etc. Lines and polygons are used to accurately depict the outline and range of the defects. Specific data annotation is performed next to the outlined defects, including defect type, size, severity, and possible impact.

[0138] For example, if a crack is found near a monitoring point, the system will use lines to accurately depict the direction and length of the crack, and mark information such as the type, width and possible hazard level of the crack next to it.

[0139] In step Si00, when displaying the panoramic view, various defects within a preset distance from the monitoring point are outlined by lines and polygons, and the parameters involved in the defect-related information are simultaneously annotated with specific data. The specific location and development trend of structural changes or potential diseases should also be considered and indicated. Specific description is given with reference to the embodiment shown below.

[0140] A method for intelligent early warning of geological disasters and engineering safety monitoring also includes the steps of simultaneously outlining various defects within a preset distance range from a monitoring point through lines and polygons when displaying a panoramic view, and simultaneously annotating specific data of parameters involved in defect-related information, as follows:

[0141] Step SA00, obtain the annotated image constructed based on the point cloud data and images of the tunnel and its environment within a preset distance range collected regularly in history, the annotated image includes the information display space and the defect annotation information within the preset distance range from the monitoring point, the defect annotation information within the preset distance range from the monitoring point includes the various defects within the preset distance range from the monitoring point outlined by lines and polygons, and the specific data annotations of the parameters involved in the defect-related information.

[0142] Among them, the means of implementing image annotation are as follows: 1. Use lidar and cameras to regularly collect point cloud data and images of the tunnel and its surrounding environment; 2. Use image processing technology to pre-process the collected data, including denoising and enhancement, to construct a clear annotated image; 3. Mark the tunnel structure, defect location and related information on the image, such as crack length and width.

[0143] In step SB00, a feature extraction algorithm is used to extract stable feature points from the early and updated images, and accurate matching is performed to ensure the consistency of the two images in space and scale.

[0144] Specifically, the feature extraction algorithm may be SIFT (Scale Invariant Feature Transform), which extracts a large number of feature points from two images through the SIFT algorithm, and uses a matching algorithm to find the correspondence between these feature points, thereby achieving accurate alignment of the images.

[0145] Step SC00, based on the matching results of the feature points, a preset image transformation model is used to accurately align the updated image with the earlier image.

[0146] The means to achieve precise image alignment are as follows: based on the matching results of feature points, an image transformation model of perspective transformation is used to accurately align the updated image with the earlier image, and the transformation parameters are adjusted through an optimization algorithm to minimize the alignment error.

[0147] Step SD00, based on the aligned images, a difference image is generated by calculating the pixel difference detection algorithm to highlight the areas of structural changes or potential diseases.

[0148] Specifically, based on the aligned images, the difference values ​​between the two images are calculated pixel by pixel, such as grayscale difference, color difference, etc. A difference image is generated based on the difference values ​​to highlight areas of structural changes or potential diseases.

[0149] Step SE00, segmenting the difference image using a preset threshold, clearly distinguishing the changed area from the background area, and further processing the segmented changed area using morphological operations, including but not limited to dilation and erosion.

[0150] Specifically, by setting a threshold, the changed area in the difference image is presented in white, while the background area is presented in black. Then, the white area is smoothed using morphological operations to obtain a more accurate contour of the changed area.

[0151] Step SF00, the processed changed area is fused with the original image for display, and the specific location and development trend of the structural change or potential disease are indicated by preset marks or colors.

[0152] Specifically, the processed changed area is fused with the original image to intuitively display the specific location and development trend of structural changes or potential diseases. The importance and urgency of the changed area are indicated by preset marks (such as arrows, circles, etc.) or colors.

[0153] For example, in the fused image, you can see that the original tunnel image is superimposed with the processed contour lines of the changed area and the corresponding marks and colors.

[0154] exist Figure 1 In step S300, the process of analyzing and determining the warning levels of different locations should also take into account the impact of subsequent rainfall changes, which will be described in detail with reference to the following embodiment.

[0155] According to the preset data interval range of geological disaster-related parameters at different locations and the corresponding relationship between the data change rate interval range and the warning level, the warning levels at different locations are analyzed and determined as follows:

[0156] Step S3A0, obtaining the rainfall intensity within the current time range and the future preset time range in the area where the target tunnel project is located, and analyzing and obtaining the rainfall intensity difference.

[0157] The specific process is as follows: 1. Use meteorological monitoring equipment or third-party meteorological services to obtain real-time rainfall intensity data in the area where the target tunnel project is located. 2. Integrate the acquired rainfall intensity data to ensure the accuracy and completeness of the data. 3. Compare the rainfall intensity within the current time range with the rainfall intensity within the future preset time range, and calculate the difference between the two, that is, the rainfall intensity difference.

[0158] Step S3B0, based on the mapping relationship between the rainfall intensity difference and the influence degree values ​​of different parameters and the current monitoring data of different parameters, predict and analyze to obtain the data and change rate of different parameters within a preset time range in the future.

[0159] Specifically, by combining the predicted data of rainfall intensity difference with the established mapping relationship, a mathematical model is used to predict the possible values ​​of each parameter within a preset time range in the future (e.g., the next 24 hours, 48 ​​hours, etc.).

[0160] Step S3C0, analyze and determine the warning levels for different locations within a future preset time range based on the correspondence between the preset data interval range and data change rate interval range of geological disaster-related parameters at different locations and the warning levels, as well as the data and change rates of different parameters obtained through analysis within a future preset time range.

[0161] Specifically, firstly, the real-time monitoring system and prediction model are used to obtain the geological disaster-related parameter data at different locations of the tunnel within the future preset time range (such as the next 24 hours, 48 ​​hours). These data are analyzed to calculate the change rate of each parameter. Then the analyzed future data is compared with the preset parameter interval range to determine the interval of each parameter. Combined with the parameter change rate, according to the preset corresponding relationship, the warning level of different locations of the tunnel within the future preset time range is determined.

[0162] Step S3D0, if the warning level of different locations within the future preset time range is different from the current warning level of different locations, the warning level of different locations within the future preset time range is used as the warning level of the corresponding location.

[0163] Based on the same inventive concept, an embodiment of the present invention also provides an intelligent early warning system for geological disaster and engineering safety monitoring, including an acquisition module, an analysis module, and an execution module.

[0164] Among them, the acquisition module is used to obtain all monitoring data of tunnel engineering geological disaster related parameters and the change rate of the monitoring data. The analysis module is used to analyze and determine the warning level of different locations according to the preset data interval range of geological disaster related parameters at different locations and the corresponding relationship between the data change rate interval range and the warning level; according to the corresponding relationship between the warning level and the tunnel internal and external linkage response plan and the analyzed and determined warning level, the tunnel internal and external linkage response plan is analyzed and determined. The execution module is used to execute the tunnel internal and external linkage response plan.

[0165] In addition, a geological disaster and engineering safety monitoring intelligent early warning system also includes a monitoring map module, a video image module, a video image module, a data management module, an early warning management module and an inspection module.

[0166] The monitoring map module collects multi-source data through satellites and drones, and combines it with the geographic information system to achieve high-precision, multi-angle real-time display of the monitoring area.

[0167] Specifically, satellites can provide macroscopic data on the surrounding environment of tunnels, including key information such as topography and landforms, providing basic support for tunnel construction and operation. Drones can go deep into the tunnel and flexibly capture real-time data from every corner of the tunnel. The collected multi-source data is integrated into the geographic information system, providing a powerful data analysis and visualization platform for tunnel monitoring. Through this system, we can conduct all-round and multi-angle real-time monitoring and display of the tunnel, including the structural status, environmental parameters and external geological environment of the tunnel. In addition, the high-precision and multi-angle display functions of the monitoring map module have significant practical significance in tunnel scenarios. Through high-precision data, we can accurately grasp every detail change inside the tunnel, such as the expansion of cracks, the location of water leakage, etc., to provide a basis for timely maintenance measures.

[0168] The video image module uses video surveillance to achieve comprehensive monitoring of key areas and key areas. Among them, the key areas in this application can be the tunnel entrance and exit areas, key nodes inside the tunnel, and the key areas can be areas with a high frequency of high-level warnings; video surveillance technology mainly relies on cameras to capture video images and send these images to the central control room or designated monitoring equipment through a transmission system. The video image module is responsible for receiving, processing and displaying these images so that monitoring personnel can clearly view the real-time situation of each area.

[0169] The data management module is used for project management, data collection, data analysis, and output of results. Specifically, in tunnel projects, the data management module is first used for comprehensive project management. Through this module, the project team can monitor the construction progress, resource allocation, and safety status of the tunnel in real time. For example, using smart sensors and data analysis technology, the progress of each construction section of the tunnel can be accurately grasped to ensure that the project proceeds as planned. Data collection is one of the core functions of the data management module. In tunnel projects, this link involves collecting data from various sensors, monitoring equipment, and manual input. These data include geological information, environmental parameters, various indicators during the construction process, etc.

[0170] Data analysis is a key part of the data management module. In this phase, the project team uses advanced data analysis techniques to conduct in-depth mining and processing of the collected data. For example, by analyzing the geological data during the tunnel construction process, potential geological risks can be predicted and corresponding countermeasures can be formulated; the output of results is the final manifestation of the data management module. In tunnel engineering, this link mainly involves presenting the results of data analysis in an intuitive and easy-to-understand way, such as reports, charts or visualization interfaces.

[0171] The early warning management module is used for early warning management, expert database management, emergency resources and early warning model management. Specifically, early warning management is the core function of the early warning management module. It monitors the key indicators in the tunnel in real time, detects abnormal situations in time and issues early warnings. For example, when the concentration of harmful gases in the tunnel exceeds the standard or small cracks appear in the structure, the early warning system can immediately capture these abnormalities and remind managers to take corresponding measures through sound and light alarms. Expert database management provides strong intellectual support for the early warning management module. By establishing a complete expert database, the module can gather the wisdom and experience of experts from various fields and provide professional guidance and suggestions for tunnel safety operation. When an early warning occurs, experts in the expert database can quickly intervene, analyze and judge the abnormal situation, and propose targeted solutions to ensure that tunnel safety is handled in a timely and effective manner. Emergency resource management is an indispensable part of the early warning management module. It is mainly responsible for the comprehensive management and deployment of emergency resources in the tunnel, including emergency equipment, rescue personnel, etc. When an early warning occurs, emergency resource management can quickly start the emergency response mechanism, mobilize resources from all parties, and ensure the efficient implementation of rescue work. Emergency resource management is an indispensable part of the early warning management module. It is mainly responsible for the comprehensive management and deployment of emergency resources in the tunnel, including emergency equipment, rescue personnel, etc. When an early warning occurs, the emergency resource management can quickly activate the emergency response mechanism, mobilize resources from all parties, and ensure the efficient implementation of the rescue work.

[0172] The inspection module is used to issue tasks, record inspections, and conduct group testing and prevention. Specifically, the inspection module can accurately push tunnel inspection tasks to inspectors through an intelligent task release system. These tasks can include regular inspections of tunnel structural integrity and monitoring of environmental parameters in tunnels. During the inspection process, the inspection module provides a convenient inspection recording function. Inspectors can use smart terminal devices to record various tunnel data in real time, including structural conditions, abnormal environmental parameters, etc.

[0173] The inspection module also introduces the innovative concept of group monitoring and group prevention. By building a collaborative platform with the participation of multiple parties, it integrates the forces of tunnel operators, professional inspection teams and the public to jointly participate in the inspection and protection of tunnels.

[0174] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. An intelligent early warning method for geological disaster and engineering safety monitoring, characterized in that: include: Obtain all monitoring data of tunnel engineering geological disaster related parameters and the change rate of the monitoring data, where the related parameters are various parameters of the tunnel structure and the surrounding environment, including displacement, temperature, humidity, vibration, strain and settlement displacement; According to the preset data interval range of geological disaster-related parameters at different locations and the corresponding relationship between the data change rate interval range and the warning level, the warning level at different locations is analyzed and determined; Analyze and determine the tunnel internal and external linkage response plan based on the corresponding relationship between the warning level and the tunnel internal and external linkage response plan and the analyzed and determined warning level; Execute the coordinated response plan inside and outside the tunnel; All monitoring data of tunnel engineering geological disaster related parameters are obtained, including: Analyze whether the monitoring location points of the target tunnel project have received warnings exceeding the preset warning level; If not, the collection period is determined based on the corresponding relationship between the monitoring location and the collection period, and all monitoring data of the corresponding monitoring location are collected and acquired according to the collection period; If yes, then analyze whether the monitoring location point receives a warning exceeding the preset warning level next time; If not, all monitoring data of the corresponding monitoring location point are collected and acquired according to the collection period of the corresponding monitoring location point when no warning of the preset warning level is received; If yes, then the collection period is determined based on the obtained warnings exceeding the preset warning level, the corresponding relationship between the monitoring location points and the collection period, and all monitoring data of the corresponding monitoring location points are collected and obtained according to the collection period; Alternatively, obtain all monitoring data of tunnel engineering geological hazards related parameters including: Obtain the warning levels of the target tunnel project within a preset time range, and analyze the number of times different warning levels occur; According to the corresponding relationship between the warning level and the number of indicators and the number of occurrences of different warning levels obtained through analysis, the total number of indicators is obtained through analysis and calculation. The specific calculation method is as follows: for each warning level, the corresponding number of indicators is determined, the number of occurrences of different warning levels is multiplied by the corresponding number of indicators, and then the products are summed up to obtain the total number of indicators. Analyze and determine the deadline for collecting all monitoring data based on the correspondence between the total indicator number interval range that the total indicator number falls into within the preset time range and the deadline for collecting all monitoring data; Based on the analysis, the deadline for collecting all monitoring data is determined, and all monitoring data related to tunnel engineering geological hazards are collected.

2. The intelligent early warning method for geological disaster and engineering safety monitoring according to claim 1 is characterized in that: It also includes steps in parallel with obtaining all monitoring data of parameters related to tunnel engineering geological hazards, as follows: Regularly collect point cloud data and images of the tunnel and its environment within a preset distance; The collected point cloud data is preprocessed, followed by data registration, aligning the data sets of multiple perspectives into a unified coordinate system, and then using MeshLab and PCL tools to construct a 3D point cloud model through point cloud triangulation. Data preprocessing includes denoising, filtering, and data pruning. Create a 3D geometric model based on the 3D point cloud model; Combine the 3D geometric model with the image to form a real scene model; The collected images are used to construct a panoramic expansion map through a panoramic expansion map construction method, and a mapping relationship between the real scene and the panorama is established to obtain a transformation function between the three-dimensional space and the expansion plane map. According to the transformation function, an information display space that is a unified and coordinated information display space of the physical space and the apparent plane is constructed, and the previous information display space is replaced by the constructed information display space that is a unified and coordinated information display space of the physical space and the apparent plane.

3. The intelligent early warning method for geological disaster and engineering safety monitoring according to claim 2 is characterized in that: The steps after constructing the information display space that unifies and coordinates the physical space and the apparent plane according to the transformation function are also included, which are as follows: Obtaining defect-related information of the tunnel project obtained by non-destructive testing technology, wherein the defect-related information includes shape, location, length, spacing and area; When displaying the panoramic view, various defects are outlined by lines and polygons, and specific data annotations are made for the parameters involved in the defect-related information.

4. The intelligent early warning method for geological disaster and engineering safety monitoring according to claim 3 is characterized in that: It also includes steps after outlining various defects by lines and polygons when displaying the panoramic image, and simultaneously annotating specific data of parameters involved in defect-related information, as follows: Obtain the location of the monitoring point where a warning exceeding a preset warning level occurs; According to the acquired location of the monitoring point, the relevant parameter data of the corresponding monitoring point is synchronously displayed and presented in the real-life model and the panoramic expansion map as the marked location point. When the panoramic map is displayed, various defects within a preset distance from the monitoring point are outlined by lines and polygons, and specific data annotations are made for the parameters involved in the defect-related information.

5. The intelligent early warning method for geological disaster and engineering safety monitoring according to claim 4 is characterized in that: It also includes the steps of outlining various defects within a preset distance from the monitoring point through lines and polygons when displaying the panoramic view, and marking specific data of parameters involved in defect-related information simultaneously, as follows: Obtain an annotated image constructed based on the point cloud data and images of the tunnel and its environment within a preset distance range collected regularly each time in history. The annotated image includes an information display space and defect annotation information within a preset distance range from the monitoring point. The defect annotation information within a preset distance range from the monitoring point includes various defects within the preset distance range from the monitoring point outlined by lines and polygons, as well as specific data annotations for parameters involved in defect-related information; Using feature extraction algorithms, stable feature points are extracted from the earlier and updated images and accurately matched to ensure the consistency of the two images in space and scale; Based on the matching results of feature points, the updated image is accurately aligned with the earlier image using a preset image transformation model; Based on the aligned images, a pixel-by-pixel difference detection algorithm is calculated to generate a difference image that highlights areas of structural changes or potential disease. The difference image is segmented by a preset threshold to clearly distinguish the changed area from the background area, and morphological operations are applied to further process the segmented changed area. Morphological operations include dilation and erosion. The processed changed area is fused with the original image for display, and the specific location and development trend of the structural changes or potential diseases are indicated by preset marks or colors.

6. An intelligent early warning system for geological disaster and engineering safety monitoring, characterized in that: Executing a geological disaster and engineering safety monitoring intelligent early warning method as claimed in claim 1, comprising: An acquisition module is used to obtain all monitoring data of tunnel engineering geological disaster related parameters and the change rate of the monitoring data; The analysis module analyzes and determines the warning levels at different locations based on the preset data interval ranges of geological disaster-related parameters at different locations and the corresponding relationship between the data change rate interval ranges and the warning levels; analyzes and determines the linkage response plan inside and outside the tunnel based on the corresponding relationship between the warning level and the linkage response plan inside and outside the tunnel and the analyzed and determined warning level; Execution module, executes the linkage response plan inside and outside the tunnel.

7. The intelligent early warning system for geological disaster and engineering safety monitoring according to claim 6 is characterized in that: Also includes: The monitoring map module collects multi-source data through satellites and drones, and combines it with the geographic information system to achieve high-precision, multi-angle real-time display of the monitoring area; Video image module, through video monitoring, can realize comprehensive monitoring of key areas and key regions; Data management module, for project management, data collection, data analysis, and results output; Early warning management module, which manages early warning, expert database, emergency resources and early warning models; The inspection module is used to issue tasks, record inspections, and conduct group testing and prevention.

Images