Road slope deformation monitoring system and method

Abstract

The invention belongs to the technical field of road deformation monitoring, and discloses a road slope deformation monitoring system and method, and the system comprises a fixed point monitoring mechanism which monitors the displacement, strain, underground water level and inclination parameters of a slope in real time at key points and key positions of the road slope; the route planning module plans the flight route of the unmanned aerial vehicle; the cruise monitoring mechanism is used for monitoring the side slope in all directions; the weather data acquisition module is used for acquiring local weather data; the three-dimensional laser scanner data analysis module is used for identifying abnormal conditions on the side slope; the image analysis module is used for identifying abnormal conditions on the slope; the comprehensive evaluation module is used for comprehensively evaluating the stability of the slope by utilizing a comprehensive evaluation model; and the alarm module comprises an alarm. According to the invention, the fixed-point monitoring mechanism and the cruise monitoring mechanism are combined to monitor the slope in all directions, the comprehensive evaluation module fuses data of various monitoring means, and the comprehensive evaluation model is utilized to comprehensively evaluate the slope stability.

Description

Technical Field

[0001] The present application relates to the technical field of highway slope deformation monitoring, and more specifically, to a highway slope deformation monitoring system and method. Background Art

[0002] Slope instability and landslides have become a major safety hazard for mountain highway traffic. Slope instability and landslides do not occur in an instantaneous period of time, but develop from creep to instability and sliding after a certain period of accumulation of instability factors. Continuous monitoring of slope deformation, deformation rate and deformation development trend at each period is the core of evaluating whether the slope can produce destructive landslides, so as to timely and safely deal with the slope.

[0003] The document with the prior art publication number CN220853506U provides a highway slope deformation monitoring device based on Beidou satellite, which relates to the technical field of highway slope deformation monitoring, including a device body and a trigger device, the two sides of the device body are connected to interfaces, the rear end of the device body is connected to a bearing, the center of the bottom surface of the device body is connected to an indicator light, the back of the indicator light is connected to a wire tube, the two sides of the wire tube are connected to a trigger device, the two sides of the wire tube are connected to one end of a switch, the other end of the switch is connected to one end of a spring rod, and the other end of the spring rod is connected to one end of a trigger rod. The device is used for a highway slope deformation monitoring device based on Beidou satellite, and there is an insulator connecting between the devices. When the highway slope is deformed, the insulator is squeezed and damaged, and the spring outside the spring rod is compressed to eject the trigger rod. The trigger rods between the two devices are connected to form a closed circuit, and the indicator light will light up to determine the deformed section, so that manual protection measures can be taken accurately.

[0004] Although the above-mentioned existing technical solutions can achieve relevant beneficial effects through the structure of existing technologies, they still have the following defects: 1. The existing technology only relies on a single monitoring method, resulting in a limited monitoring range and unable to take into account both macro and micro changes of the slope. 2. The data of multiple monitoring methods are not effectively integrated and comprehensively evaluated, and it is impossible to provide decision makers with a comprehensive and scientific slope stability assessment report, thereby limiting the scientificity and effectiveness of the slope management and maintenance plan.

[0005] In view of this, we propose a highway slope deformation monitoring system and method. Summary of the invention

[0006] 1. Technical issues to be solved

[0007] The purpose of this application is to provide a highway slope deformation monitoring system and method, which solves the technical problems raised in the above-mentioned background technology, realizes all-round monitoring of the slope through the combination of fixed-point monitoring mechanism and cruise monitoring mechanism, and the comprehensive evaluation module integrates the data of various monitoring methods, and uses the comprehensive evaluation model to comprehensively evaluate the stability of the slope.

[0008] 2. Technical solution

[0009] The technical solution of the present application provides a highway slope deformation monitoring system, comprising:

[0010] Data collection module: collects a large amount of data on highway slopes, including environmental data, geological data, and real-time status data of slopes. At the same time, the collected images are annotated as training samples.

[0011] Fixed-point monitoring agencies: Install GNSS displacement monitoring stations, strain gauges, groundwater dynamic monitors, inclinometers and other sensors at key and critical locations on highway slopes to monitor key parameters such as slope displacement, strain, groundwater level and inclination in real time.

[0012] Route planning module: Plan the flight route of the drone based on the terrain, topography and monitoring requirements of the highway slope to ensure that the drone can cover the monitoring area efficiently and comprehensively.

[0013] Cruise monitoring mechanism: including drones, high-definition cameras and 3D laser scanners; cruise monitoring is carried out according to the planned route. The high-definition cameras and 3D laser scanners carried by drones conduct high-altitude, long-distance and all-round monitoring of the slopes to capture subtle changes in the slopes.

[0014] Weather data acquisition module: timely obtain local weather data, focusing on rainfall and strong wind data;

[0015] 3D laser scanner data analysis module: pre-processes the data collected by the 3D laser scanner (such as denoising, registration, splicing, etc.), and then uses algorithms to identify abnormal conditions on the slope, including cracks, collapse, landslides, scours, slope swelling, slope seepage, and slope vegetation damage;

[0016] Image analysis module: pre-processes, extracts features and analyzes the collected images to identify abnormal conditions on the slope, including cracks, collapse, landslide, scour, slope swelling, slope seepage and slope vegetation damage, etc.

[0017] Comprehensive evaluation module: The monitoring results of the fixed-point monitoring agency, the weather monitoring results of the weather data acquisition module, the analysis results of the 3D laser scanner data analysis module and the analysis results of the image analysis module are integrated to conduct a comprehensive evaluation of the stability of the highway slope. The data and results from different monitoring modules are integrated to conduct a comprehensive evaluation of the stability of the slope using a comprehensive evaluation model.

[0018] Alarm module: including an alarm. When monitoring abnormal conditions or potential risks on the slope, the alarm module will issue an alarm in time.

[0019] Central control unit: network connected with alarm module, fixed-point monitoring mechanism, route planning module, cruise monitoring mechanism, 3D laser scanner data analysis module, image analysis module and comprehensive evaluation module.

[0020] Through the above technical solution, comprehensive, real-time and accurate monitoring and evaluation of highway slopes are achieved, providing strong guarantees for the safe operation of highways.

[0021] As an optional solution of the present invention, the fixed-point monitoring mechanism monitors the key parameters of the slope such as displacement, strain, groundwater level and inclination in real time, including the following steps:

[0022] 1. Monitoring point selection and planning: Determine the key and critical monitoring locations based on the topography, geological conditions, historical monitoring data and other factors of the highway slope. Ensure that the monitoring point locations can fully reflect the overall stability and potential risks of the slope. Select high-precision and high-stability GNSS displacement monitoring stations, strain gauges, groundwater dynamic monitors, inclinometers and other sensors.

[0023] 2. On-site survey: Conduct on-site surveys of monitoring points to understand the impact of terrain, landforms, vegetation, traffic and other factors on equipment installation. Determine the specific location and method of equipment installation to ensure that the equipment is stable and can accurately receive satellite signals.

[0024] 3. Equipment installation: Install a GNSS displacement monitoring station and ensure that the antenna is facing an open and unobstructed area to receive satellite signals. Connect the components and perform preliminary debugging to ensure the normal operation of the equipment. Install strain gauges at key locations on the slope, such as potential slip surfaces, near cracks, etc. Ensure that the strain gauges fit tightly against the slope and can accurately measure the strain changes of the slope. Install groundwater dynamic monitoring instruments inside or near the slope to monitor changes in groundwater level and water quality. Install inclinometers at key locations on the slope to monitor changes in the slope's inclination. Ensure that the inclinometer is installed firmly and perpendicular to the slope to accurately measure the inclination angle.

[0025] 4. Equipment debugging and calibration: Debug all installed equipment to ensure that the equipment can receive signals and transmit data normally. Check the stability and accuracy of data transmission.

[0026] 5. Data monitoring and analysis: Data from all monitoring equipment is transmitted to the data processing center or cloud platform. A database and data processing platform for the monitoring system is established to store, process and analyze the data.

[0027] The monitoring system monitors the displacement, strain, groundwater level, inclination and other key parameters of the slope in real time, detects abnormal conditions in time and automatically triggers the early warning mechanism.

[0028] Process and analyze the monitoring data to evaluate the stability of the slope. Perform a comprehensive evaluation according to the following formula:

[0029] I assessment =w1*I displacement +w2*I strain +w3*I water_level +w4*I inclination ; w1+w2+w3+w4=1;

[0030] Among them, I assessment It is a comprehensive evaluation index used to quantify the overall condition of slope stability.

[0031] I displacement It is the displacement influence index, which indicates the monitoring result of slope displacement and is usually a standardized value. strain It is the strain influence index, which indicates the monitoring result of slope strain and reflects the deformation degree of the material. It is usually a standardized value. water_level It is the water level impact index, which indicates the monitoring result of groundwater level, usually a standardized value, reflecting the change of groundwater level. inclination is the tilt influence index, which indicates the monitoring result of slope tilt, usually a standardized value, reflecting the tilt degree of the slope. w1, w2, w3 and w4 are weight coefficients, corresponding to the weights of displacement, strain, water level and tilt, respectively, and are used to indicate the relative importance of each monitoring parameter in the comprehensive evaluation index.

[0032] Through the above technical solution, it can be ensured that the GNSS displacement monitoring stations, strain gauges, groundwater dynamic monitors, inclinometers and other sensors installed at key and critical locations on the highway slopes can work stably and accurately, providing a direct basis for the assessment of slope stability.

[0033] As an optional solution of the present invention, the route planning module plans the flight route of the drone, including the following steps:

[0034] 1. Data collection: Determine the specific content of monitoring, such as slope stability, vegetation cover, crack development, etc. Obtain accurate topographic maps and elevation models (DEM) of the slope through satellite images, preliminary drone reconnaissance or ground mapping. Identify the slope's slope, uneven terrain, and obstacles.

[0035] 2. Set flight parameters: Set the appropriate flight altitude according to monitoring requirements and camera performance to ensure adequate resolution and avoid collisions. Determine the appropriate flight speed based on the camera's shutter speed, shooting range, and the level of detail required for monitoring.

[0036] 3. Route planning: Use drone route planning software (such as DJI GS Pro, Mission Planner, etc.) to input terrain data and set the route based on the above parameters. Manually set waypoints in key monitoring areas or complex terrain to ensure that the drone can fly accurately and capture important data. Set safety buffers on both sides and top of the route to avoid collisions between the drone and obstacles (such as trees and power lines). Combined with weather forecast data, adjust the route to minimize the impact of wind resistance on flight stability and safety.

[0037] 4. Simulation adjustment: simulate the flight path of the drone in the planning software to check whether there are uncovered areas or potential safety issues. Adjust the route parameters such as altitude, speed, heading angle, etc. according to the simulation results until the optimal monitoring effect and flight efficiency are achieved.

[0038] 5. Field verification and adjustment: Conduct field test flights before formal monitoring to check whether the route is executed as expected and whether the shooting data meets the requirements. If necessary, adjust the route to further optimize the subsequent flight route planning.

[0039] As an optional solution of the present invention, the 3D laser scanner data analysis module pre-processes the data collected by the 3D laser scanner and then uses an algorithm to identify abnormal conditions on the slope, including the following steps:

[0040] 1. Data preprocessing: De-noise the data to remove noise points introduced by environmental factors (such as weather changes, air dust), equipment errors or reflective surface characteristics (such as multiple reflections from smooth surfaces). Use statistical filtering, morphological filtering and other methods to automatically or manually remove abnormal points based on the spatial distribution and attribute characteristics of point cloud data.

[0041] 2. Registration: Use the iterative closest point (ICP) algorithm to achieve data registration by minimizing the geometric differences between different point clouds. Convert point cloud data from different scanning positions or different time points into a unified coordinate system to achieve accurate alignment of multi-site data.

[0042] 3. Stitching: According to the overlapping areas and registration parameters between point clouds, the parts are stitched together to form a continuous and seamless 3D surface. Multiple registered point cloud data sets are merged into a complete 3D model.

[0043] 4. Point cloud segmentation: Based on the spatial position, geometric shape, color and other attributes of the points, point cloud data is segmented into regions or objects with similar attributes or geometric features using clustering algorithms, region growing and other methods to facilitate subsequent analysis and processing.

[0044] 5. Feature extraction: Use point cloud processing technology to identify geometric features such as planes, edges, corners, and morphological features such as slope and curvature, extract the geometric and morphological features of the slope from the point cloud data, and provide basic data for abnormal situation identification.

[0045] 6. Abnormal situation identification:

[0046] Crack identification: Use edge detection algorithms to identify cracks by detecting discontinuous areas or abnormal change areas in the point cloud. The formula for calculating the crack depth is as follows:

[0047] D crack =max i∈crack (d i -dˉ)+W crack *Θcrack; where D crack is the crack depth, which indicates the maximum depth of the crack. i is the number of crack instances in the traversal crack set. i is the depth measurement of point i or the vertical distance relative to some reference plane. dˉ is the average depth measurement of all points in the crack. W crack is the crack width, which indicates the lateral size of the crack. Θcrack is the crack strike angle, which indicates the inclination angle of the crack relative to a certain reference direction.

[0048] Collapse and landslide identification: Monitor changes by comparing scan data at different time points, analyze the height changes and morphological distortions of the point cloud, and identify the collapse areas and landslide bodies of the slope.

[0049] The volume change calculation formula for collapse and landslide identification is:

[0050] V change =V new -V old +λ*Δt; where λ is the volume change rate and Δt is the time difference between two scans. change is the volume change, which represents the difference in volume between the two scans. V new is the volume of the new scan data, which represents the volume measurement value obtained from the most recent scan. V oldis the volume of the old scan data, which represents the volume measurement obtained from a previous scan.

[0051] Scour identification: Observe the point cloud density and morphological changes at the bottom or edge area of ​​the slope to identify scour phenomena.

[0052] Slope bulging identification: Analyze the curvature changes and geometric characteristics of the slope surface to identify slope bulging phenomena.

[0053] Slope seepage identification: Indirect judgment by observing the changes in slope surface moisture (such as with the assistance of multi-spectral or infrared scanning).

[0054] Slope vegetation damage identification: Vegetation damage can be identified by comparing the point cloud density and color changes of vegetation coverage areas.

[0055] 7. Result output: The identified abnormalities are presented to relevant departments in a visual form (such as 3D models, 2D drawings, reports, etc.).

[0056] Through the above technical solutions, the 3D laser scanner data analysis module can efficiently and accurately identify abnormal conditions on the slope, providing a scientific basis for slope stability assessment and management.

[0057] As an optional solution of the present invention, the image analysis module performs preprocessing, feature extraction and image analysis on the collected image to identify abnormal conditions on the slope, including the following steps:

[0058] 1. Image preprocessing: including noise removal, image enhancement and image segmentation;

[0059] 2. Feature extraction: Extract features from the preprocessed image, including geometric features, texture features, and color features;

[0060] 3. Image analysis: including crack detection, collapse and landslide detection, scour detection, slope swelling detection, slope seepage detection and slope vegetation damage detection;

[0061] Crack detection: Use edge detection algorithms (such as Canny edge detector) and morphological operations (such as erosion and dilation) to identify cracks in the image. Analyze parameters such as length, width, and direction of the cracks to assess their severity.

[0062] Collapse and landslide detection: By comparing images at different time points, analyzing the height changes and morphological distortions of the slope surface, and identifying the collapse and landslide areas. Use optical flow or feature point tracking algorithms to monitor the dynamic changes of the slope.

[0063] Scour detection: Observe the image changes at the bottom or edge of the slope to identify erosion caused by scour. Analyze the area, depth and other parameters of the scour area to assess its impact range.

[0064] Slope bulge detection: Analyze the curvature changes and geometric features of the slope surface to identify slope bulge phenomena. Use curvature-based methods or machine learning algorithms to detect slope bulge.

[0065] Slope seepage detection: By analyzing the color changes, humidity distribution and other characteristics of the slope surface, the slope seepage phenomenon can be indirectly identified.

[0066] Slope vegetation damage detection: Compare the image changes of vegetation coverage areas to identify vegetation damage. Analyze the area, degree and other parameters of vegetation damage to evaluate its impact on slope stability.

[0067] 4. Result output: The identified abnormalities are presented to relevant departments in a visual form (such as annotated images, reports, etc.).

[0068] As an optional solution of the present invention, the comprehensive evaluation module performs a comprehensive evaluation on the stability of the highway slope, including the following steps:

[0069] 1. Data collection and collation: Collect groundwater level monitoring data, displacement monitoring data, and deformation monitoring data. Ensure that the time series of the data is complete and the accuracy meets the assessment requirements. Obtain the three-dimensional point cloud data, DEM (digital elevation model) data, slope morphology, and abnormal area identification results of the slope from the three-dimensional laser scanner data analysis module. Verify the accuracy and completeness of the data to ensure that there are no omissions or errors. Obtain high-definition images of the slope, crack identification results, vegetation damage, scour and landslide signs, etc. from the image analysis module. Collect weather monitoring results from the weather data acquisition module; classify and annotate image data for subsequent fusion processing.

[0070] 2. Data preprocessing: All collected data are preprocessed, including denoising, correction, format unification, etc. Ensure that data from different sources can match and correspond to each other in time and space.

[0071] 3. Data fusion: Before data fusion, calibrate data from different sources to ensure their consistency in space and time. Establish a unified data framework, define a unified coordinate system and data format, and convert all data sources to this framework.

[0072] The weighted average multi-source data fusion technology is used to fuse different types of data. Considering the correlation, conflict and complementarity between data, the weights are reasonably allocated to obtain more accurate comprehensive evaluation results. The comprehensive stability evaluation formula is as follows:

[0073] ∑ m i＝1 (c i )=1; where c i is the weight of the ith data source. ST i is the quantitative value of the ith data source, including the monitoring results of the fixed-point monitoring agency, the weather monitoring results of the weather data acquisition module, the analysis results of the 3D laser scanner data analysis module, and the quantitative value corresponding to the analysis results of the image analysis module. 综合 is the quantitative value of the fused data, the slope stability assessment value. m is the total number of data sources. If ST 综合 A higher value of indicates that the slope is more stable after integrating all data sources. 综合 A lower value indicates that the slope stability is weak and may require further investigation or action.

[0074] 4. Comprehensive evaluation: Use independent data sets to verify the model to ensure the accuracy and reliability of its prediction results. The evaluation model is dynamically adjusted and optimized based on new monitoring data and actual conditions to adapt to changes in slope stability. Set various parameters of the evaluation model according to actual conditions, such as mechanical parameters, geological parameters, environmental factors, etc. Input the preprocessed and fused data into the comprehensive evaluation model. Calculate the stability index or comprehensive evaluation score of the slope. These indicators include safety factor, stability level, potential risk level, etc.

[0075] 5. Result output: Output the assessment results, including stability level, warning information, recommended measures, etc. Interpret and analyze the assessment results to clarify the stability status, potential risks and possible impact range of the slope.

[0076] 6. Visualization of results: Visualize the evaluation results in the form of charts, maps, etc. Write a detailed evaluation report, including the purpose, methods, process, results and recommendations of the evaluation.

[0077] Through the above technical solutions, the comprehensive evaluation module can make full use of the data and results from different monitoring modules to comprehensively evaluate the stability of highway slopes and provide strong support for slope management and maintenance.

[0078] The present invention provides a method for monitoring highway slope deformation, comprising the following steps:

[0079] S1. Fixed-point monitoring agencies install GNSS displacement monitoring stations, strain gauges, groundwater dynamic monitoring instruments, inclinometers and other sensors at key and critical locations on highway slopes to monitor key parameters such as slope displacement, strain, groundwater level and inclination in real time;

[0080] S2, the route planning module plans the flight route of the drone according to the topography, landform and monitoring requirements of the highway slope, ensuring that the drone can cover the monitoring area efficiently and comprehensively;

[0081] S3. The patrol monitoring agency conducts patrol monitoring according to the planned route. The high-definition camera and three-dimensional laser scanner carried by the drone conduct high-altitude, long-distance and all-round monitoring of the slope;

[0082] S4, weather data acquisition module timely acquires local weather data, focusing on rainfall and strong wind data;

[0083] S5. The 3D laser scanner data analysis module pre-processes the data collected by the 3D laser scanner (such as denoising, registration, splicing, etc.), and then uses algorithms to identify abnormal conditions on the slope. Identify abnormal conditions on the slope, including cracks, collapse, landslides, scours, slope swelling, slope seepage, and slope vegetation damage;

[0084] S6, the image analysis module performs preprocessing, feature extraction and image analysis on the collected images, and identifies abnormal conditions on the slope, including cracks, collapse, landslide, scour, slope swelling, slope seepage and slope vegetation damage, etc.;

[0085] S7. The comprehensive evaluation module integrates the monitoring results of the fixed-point monitoring agency (including groundwater level, displacement monitoring results and deformation monitoring results), the weather monitoring results of the weather data acquisition module, the analysis results of the 3D laser scanner data analysis module and the analysis results of the image analysis module, and uses the comprehensive evaluation model to comprehensively evaluate the stability of the slope.

[0086] S8. When monitoring of slope abnormalities or potential risks, the alarm module will issue an alarm in time.

[0087] 3. Beneficial effects

[0088] One or more technical solutions provided in the technical solution of this application have at least the following technical effects or advantages:

[0089] 1. The present invention combines a fixed-point monitoring mechanism with a cruise monitoring mechanism to conduct all-round monitoring of the slope. The fixed-point monitoring mechanism can continuously and stably monitor the key parameters of the slope, such as displacement, strain, groundwater level and inclination, by installing high-precision sensors, and provide accurate basic data. The cruise monitoring mechanism, especially the high-definition camera and three-dimensional laser scanner carried by the drone, can capture the macro changes and subtle features of the slope from a high-altitude perspective. The combination of the two realizes all-round monitoring from macro to micro, from static to dynamic, greatly improving the accuracy and comprehensiveness of monitoring.

[0090] 2. The present invention uses automated and intelligent monitoring methods, so that the fixed-point monitoring mechanism and the cruise monitoring mechanism can continuously and stably carry out monitoring work, reducing the frequency and intensity of manual inspections. At the same time, the combination of the two realizes the complementarity and coordination of monitoring work, further improves monitoring efficiency, and reduces labor costs.

[0091] 3. The 3D laser scanner data analysis module and image analysis module pre-process and deeply analyze the collected data through advanced algorithms, and can accurately identify various abnormal conditions on the slope, such as cracks, collapse, landslides, scours, slope swelling, slope seepage and vegetation damage, thereby improving the accuracy and reliability of monitoring.

[0092] 4. The present invention integrates the data of various monitoring methods through a comprehensive evaluation module, and uses a comprehensive evaluation model to comprehensively evaluate the stability of the slope, providing scientific and comprehensive information support for decision makers, which helps to formulate more reasonable and effective slope management and maintenance plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Figure 1 A schematic diagram of the flow of a highway slope deformation monitoring method disclosed in a preferred embodiment of the present application;

[0094] Figure 2 The figure is an overall schematic diagram of a highway slope deformation monitoring system disclosed in a preferred embodiment of the present application. DETAILED DESCRIPTION

[0095] The present application is further described in detail below in conjunction with the accompanying drawings.

[0096] Reference Figure 1 and Figure 2 , the embodiment of the present application provides a highway slope deformation monitoring system, comprising:

[0097] Data collection module: collects a large amount of data on highway slopes, including environmental data (such as rainfall, temperature, humidity, etc.), geological data (such as soil type, rock layer distribution, etc.) and real-time status data of slopes. At the same time, the collected images are annotated as training samples for the subsequent machine learning or deep learning algorithm training of the image analysis module.

[0098] Fixed-point monitoring institutions: Install GNSS displacement monitoring stations, strain gauges, groundwater dynamic monitoring instruments, inclinometers and other sensors at key and critical locations on highway slopes to ensure that the equipment is stable and can accurately receive satellite signals. Strain gauges and inclinometers and other sensors are used to measure deformation parameters such as slope strain and inclination. They can monitor key parameters such as slope displacement, strain, groundwater level and inclination in real time, providing a direct basis for slope stability assessment.

[0099] Route planning module: Plan the flight route of the drone; plan the flight route of the drone according to the terrain, topography and monitoring requirements of the highway slope to ensure that the drone can cover the monitoring area efficiently and comprehensively. Consider factors such as flight altitude, speed, heading angle, etc. to optimize monitoring effects and flight efficiency.

[0100] Cruise monitoring mechanism: including drones, high-definition cameras and 3D laser scanners; conduct all-round monitoring of highway slopes; conduct cruise monitoring according to the planned route, collect high-definition images through high-definition cameras, and conduct all-round scanning through 3D laser scanners; the high-definition cameras and 3D laser scanners carried by drones conduct high-altitude, long-distance, and all-round monitoring of slopes to capture subtle changes in slopes. Through non-contact monitoring, interference and damage to slopes are reduced; it is efficient, flexible, and can quickly respond to monitoring needs.

[0101] Weather data acquisition module: timely obtain local weather data, focusing on rainfall and strong wind data; timely and accurate acquisition of local weather data;

[0102] 3D laser scanner data analysis module: pre-process the data collected by the 3D laser scanner (such as denoising, registration, splicing, etc.), and then use the algorithm to identify abnormal conditions on the slope. Pre-process the collected laser scanning data; analyze the pre-processed data to identify abnormal conditions on the slope, including cracks, collapse, landslides, scours, slope swelling, slope seepage, and slope vegetation damage;

[0103] Image analysis module: pre-processes, extracts features and analyzes the collected images to identify abnormal conditions on the slope, including cracks, collapse, landslide, scour, slope swelling, slope seepage and slope vegetation damage, etc.

[0104] Comprehensive evaluation module: The monitoring results of the fixed-point monitoring agency (including groundwater level, displacement monitoring results and deformation monitoring results), the weather monitoring results of the weather data acquisition module, the analysis results of the 3D laser scanner data analysis module and the analysis results of the image analysis module are integrated to conduct a comprehensive evaluation of the stability of the highway slope. The data and results from different monitoring modules are integrated to conduct a comprehensive evaluation of the stability of the slope using a comprehensive evaluation model.

[0105] Alarm module: including an alarm. When monitoring abnormal conditions or potential risks on the slope, the alarm module will issue an alarm in time.

[0106] Central control unit: network connected with alarm module, fixed-point monitoring mechanism, route planning module, cruise monitoring mechanism, 3D laser scanner data analysis module, image analysis module and comprehensive evaluation module.

[0107] In this technical solution, through the collaborative work of various modules, comprehensive, real-time and accurate monitoring and evaluation of highway slopes are achieved, providing strong guarantees for the safe operation of highways.

[0108] Furthermore, the fixed-point monitoring mechanism monitors the displacement, strain, groundwater level and inclination of the slope in real time, including the following steps:

[0109] 1. Monitoring point selection planning: Determine the key and critical monitoring locations based on the topography, geological conditions, historical monitoring data and other factors of the highway slope, ensuring that the location of the monitoring points can fully reflect the overall stability and potential risks of the slope.

[0110] Select high-precision and high-stability GNSS displacement monitoring stations, strain gauges, groundwater dynamic monitors, inclinometers and other sensors. Ensure that the equipment meets the requirements of relevant standards and specifications and can adapt to the needs of the monitoring environment.

[0111] 2. On-site survey: Conduct on-site surveys of monitoring points to understand the impact of terrain, landforms, vegetation, traffic and other factors on equipment installation. Determine the specific location and method of equipment installation to ensure that the equipment is stable and can accurately receive satellite signals.

[0112] 3. Equipment installation: Install the GNSS displacement monitoring station and ensure that the antenna is facing an open and unobstructed area to receive satellite signals. Ensure that the entire monitoring station is stable and does not shake. Connect the components and perform preliminary debugging to ensure the normal operation of the equipment. Install strain gauges at key locations on the slope, such as potential slip surfaces, near cracks, etc. Ensure that the strain gauges fit tightly against the slope and can accurately measure the strain changes of the slope. Install groundwater dynamic monitoring instruments inside or near the slope to monitor changes in groundwater level and water quality. Install inclinometers at key locations on the slope to monitor changes in the slope's inclination. Ensure that the inclinometer is installed firmly and perpendicular to the slope to accurately measure the inclination angle.

[0113] 4. Equipment debugging and calibration: Debug all installed equipment to ensure that the equipment can receive signals and transmit data normally. Check the stability and accuracy of data transmission to ensure the reliability of monitoring data. Calibrate the GNSS displacement monitoring station to ensure that the positioning accuracy meets the requirements. Calibrate and verify the strain gauges, groundwater dynamic monitors and inclinometers to ensure the accuracy of the measurement data.

[0114] 5. Data monitoring and analysis: Data from all monitoring equipment is transmitted to the data processing center or cloud platform. A database and data processing platform for the monitoring system is established to store, process and analyze the data.

[0115] The monitoring system monitors the displacement, strain, groundwater level, inclination and other key parameters of the slope in real time, detects abnormal conditions in time and automatically triggers the early warning mechanism.

[0116] Process and analyze the monitoring data to evaluate the stability of the slope. Perform a comprehensive evaluation according to the following formula:

[0117] I assessment =w1*I displacement +w2*I strain +w3*I water_level +w4*I inclination ; w1+w2+w3+w4=1;

[0118] Among them, I assessment It is a comprehensive evaluation index used to quantify the overall condition of slope stability.

[0119] I displacement It is the displacement influence index, which indicates the monitoring result of slope displacement and is usually a standardized value. strain It is the strain influence index, which indicates the monitoring result of slope strain and reflects the deformation degree of the material. It is usually a standardized value. water_level It is the water level impact index, which indicates the monitoring result of groundwater level, usually a standardized value, reflecting the change of groundwater level. inclination is the tilt influence index, which indicates the monitoring result of slope tilt, usually a standardized value, reflecting the tilt degree of the slope. w1, w2, w3 and w4 are weight coefficients, corresponding to the weights of displacement, strain, water level and tilt, respectively, and are used to indicate the relative importance of each monitoring parameter in the comprehensive evaluation index.

[0120] In this technical solution, it can be ensured that sensors such as GNSS displacement monitoring stations, strain gauges, groundwater dynamic monitors and inclinometers installed at key and critical locations on highway slopes can work stably and accurately, providing a direct basis for the assessment of slope stability.

[0121] Furthermore, the route planning module plans the flight route of the drone, including the following steps:

[0122] 1. Data collection: Determine the specific content of monitoring, such as slope stability, vegetation cover, crack development, etc. Obtain accurate topographic maps and elevation models (DEM) of the slope through satellite images, preliminary drone reconnaissance or ground mapping. Identify the slope's slope, uneven terrain, obstacles (such as trees, poles, buildings), etc.

[0123] 2. Set flight parameters: Set the appropriate flight altitude according to monitoring requirements and camera performance to ensure adequate resolution and avoid collisions. Determine the appropriate flight speed based on the camera's shutter speed, shooting range, and the level of detail required for monitoring. Consider terrain undulations when planning routes, avoid blind spots caused by straight-line flight, and use zigzag, spiral, or grid-shaped routes for coverage. Set the horizontal and vertical overlap rates between photo or video frames to ensure the continuity and accuracy of subsequent data processing.

[0124] 3. Route planning: Use drone route planning software (such as DJI GS Pro, Mission Planner, etc.) to input terrain data and set the route based on the above parameters. Manually set waypoints in key monitoring areas or complex terrain to ensure that the drone can fly accurately and capture important data. Set safety buffers on both sides and top of the route to avoid collisions between the drone and obstacles (such as trees and power lines). Combined with weather forecast data, adjust the route to minimize the impact of wind resistance on flight stability and safety.

[0125] 4. Simulation adjustment: simulate the flight path of the drone in the planning software to check whether there are uncovered areas or potential safety issues. Adjust the route parameters such as altitude, speed, heading angle, etc. according to the simulation results until the optimal monitoring effect and flight efficiency are achieved.

[0126] 5. Field verification and adjustment: Conduct field test flights before formal monitoring to check whether the route is executed as expected and whether the shooting data meets the requirements. During the flight, the GPS and sensor data carried by the drone can be transmitted to the ground station in real time, and the route can be adjusted if necessary. Further optimize the subsequent flight route planning.

[0127] Further, the 3D laser scanner data analysis module pre-processes the data collected by the 3D laser scanner and then uses an algorithm to identify abnormal conditions on the slope, including the following steps;

[0128] 1. Data preprocessing: De-noise the data to remove noise points introduced by environmental factors (such as weather changes, air dust), equipment errors or reflective surface characteristics (such as multiple reflections from smooth surfaces). Use statistical filtering, morphological filtering and other methods to automatically or manually remove abnormal points based on the spatial distribution and attribute characteristics of point cloud data.

[0129] 2. Registration: Use the iterative closest point (ICP) algorithm to achieve data registration by minimizing the geometric differences between different point clouds. Convert point cloud data from different scanning positions or different time points into a unified coordinate system to achieve accurate alignment of multi-site data.

[0130] The registration is performed using the following formula:

[0131] In the formula, E ICP represents the registration error, i.e., the goodness-of-fit measure between point clouds calculated by the ICP algorithm. i is the index of the summation, representing a point pair. N is the total number of point pairs in the point cloud. w i is the weight factor of the i-th pair of points, used to indicate the importance or confidence of the points, which may be determined based on the quality, reliability or other criteria of the points. i is the i-th point in the source point cloud. i is the same as p in the target point cloud i The nearest point. |||p i -q i ∣∣ is point p i and q i The Euclidean distance between .

[0132] 3. Stitching: According to the overlapping areas and registration parameters between point clouds, stitch the parts together to form a continuous and seamless 3D surface. Merge multiple registered point cloud data sets into a complete 3D model. Evaluate the point cloud density at different scales according to the following formula to capture features at different scales:

[0133] In the formula, O overlap represents the splicing overlap metric, that is, the percentage of the point cloud density of the overlapping area considered at different scales to the entire point cloud density. S is the number of scales, which refers to the total number of different spatial scales used to evaluate the point cloud density. s N is the overlapping area at scale s, which refers to the area considered to be overlapping in the point cloud dataset at a specific scale. s is the number of points at scale s, indicating the total number of points in the point cloud dataset at scale s. i,s is the density value of point i at scale s, which represents the local density estimate of point i at a specific scale. s is a specific scale level used to index different scales, with values ​​ranging from 1 to S.

[0134] 4. Point cloud segmentation: Segmentation is based on the spatial position, geometric shape, color and other attributes of the points. Clustering algorithms, region growing and other methods can be used to segment the point cloud data into regions or objects with similar attributes or geometric features for subsequent analysis and processing.

[0135] The following formula is used to calculate the similarity of the clustering algorithm for point cloud segmentation:

[0136]

[0137] In the formula, S cluster It is the cluster similarity, which represents the consistency measure of the attribute distribution of the points in the cluster and the probability model of the cluster center.cluster is the number of points in the cluster, that is, the total number of points in a particular cluster. i )∣μ c ,Σ c ] is point p i The attribute vector I(p i ) is the probability density function under the probability model of the cluster center c. It describes the probability density function of the observed point p under the given cluster center parameters. i The probability of the attribute. I(p i ) is point p i The attribute vector of μ includes multi-dimensional features such as color, intensity, and height. c Is the mean vector of the cluster center c, representing the central position of the cluster center on each attribute. c is the covariance matrix of the cluster center c, which describes the changes in the attribute vector of the cluster center, including variance and correlation coefficient. i is the i-th point in the cluster.

[0138] 5. Feature extraction: Use point cloud processing technology to identify geometric features such as planes, edges, corners, and morphological features such as slope and curvature, extract the geometric and morphological features of the slope from the point cloud data, and provide basic data for abnormal situation identification.

[0139] 6. Abnormal situation identification:

[0140] Crack Identification: Use edge detection algorithms to identify cracks by detecting discontinuous areas or areas of abnormal changes in the point cloud.

[0141] The formula for calculating the crack depth is as follows:

[0142] D crack =max i∈crack (d i -dˉ)+W crack *Θcrack; where D crack is the crack depth, which indicates the maximum depth of the crack. i is the number of crack instances in the traversal crack set. i is the depth measurement of point i or the vertical distance relative to some reference plane. dˉ is the average depth measurement of all points in the crack. W crack is the crack width, which indicates the lateral size of the crack. Θcrack is the crack strike angle, which indicates the inclination angle of the crack relative to a certain reference direction.

[0143] Collapse and landslide identification: Monitor changes by comparing scan data at different time points, analyze the height changes and morphological distortions of the point cloud, and identify the collapse areas and landslide bodies of the slope.

[0144] The volume change calculation formula for collapse and landslide identification is:

[0145] V change =V new -V old +λ*Δt; where λ is the volume change rate and Δt is the time difference between two scans. change is the volume change, which represents the difference in volume between the two scans. V new is the volume of the new scan data, which represents the volume measurement value obtained from the most recent scan. V old is the volume of the old scan data, which represents the volume measurement obtained from a previous scan.

[0146] Scour identification: Observe the point cloud density and morphological changes at the bottom or edge area of ​​the slope to identify scour phenomena.

[0147] Slope bulging identification: Analyze the curvature changes and geometric characteristics of the slope surface to identify slope bulging phenomena.

[0148] Slope seepage identification: Indirect judgment by observing the changes in slope surface moisture (such as with the assistance of multi-spectral or infrared scanning).

[0149] Slope vegetation damage identification: Vegetation damage can be identified by comparing the point cloud density and color changes of vegetation coverage areas.

[0150] 7. Result output: The identified abnormalities are presented to relevant departments in a visual form (such as 3D models, 2D drawings, reports, etc.).

[0151] In this technical solution, the 3D laser scanner data analysis module can efficiently and accurately identify abnormal conditions on the slope, providing a scientific basis for slope stability assessment and management.

[0152] Furthermore, the image analysis module performs preprocessing, feature extraction and image analysis on the collected images to identify abnormal conditions on the slope, including the following steps:

[0153] 1. Image preprocessing: including noise removal, image enhancement and image segmentation;

[0154] Noise removal: Remove random noise from the image, which may be caused by sensor noise, transmission errors or environmental factors. Common denoising methods include median filtering, Gaussian filtering, etc.

[0155] Image enhancement: Improve the contrast and brightness of the image to make the details of the slope clearer. This includes techniques such as histogram equalization and contrast stretching.

[0156] Image segmentation: Divide an image into multiple regions or objects to facilitate subsequent feature extraction and analysis. Methods such as threshold segmentation, edge detection, and region growing can be used.

[0157] 2. Feature extraction: Extract features from the preprocessed image, including geometric features, texture features, and color features;

[0158] Geometric feature extraction: Extract the geometric features of the slope from the image, such as edges, corners, straight lines, curves, etc.

[0159] Texture feature extraction: Analyze the texture changes on the slope surface and identify phenomena such as vegetation destruction and scour. Commonly used texture feature extraction methods include gray level co-occurrence matrix (GLCM), local binary pattern (LBP), etc.

[0160] Color feature extraction: For color images, color features of the slope can be extracted, such as color histogram, color moment, etc. These features help to identify color changes caused by slope seepage, etc.

[0161] 3. Image analysis:

[0162] Crack detection: Use edge detection algorithms (such as Canny edge detector) and morphological operations (such as erosion and dilation) to identify cracks in the image. Analyze parameters such as length, width, and direction of the cracks to assess their severity.

[0163] Collapse and landslide detection: By comparing images at different time points, analyzing the height changes and morphological distortions of the slope surface, and identifying the collapse and landslide areas. Use optical flow or feature point tracking algorithms to monitor the dynamic changes of the slope.

[0164] Scour detection: Observe the image changes at the bottom or edge of the slope to identify erosion caused by scour. Analyze the area, depth and other parameters of the scour area to assess its impact range.

[0165] Slope bulge detection: Analyze the curvature changes and geometric features of the slope surface to identify slope bulge phenomena. Use curvature-based methods or machine learning algorithms to detect slope bulge.

[0166] Slope seepage detection: By analyzing the color changes, humidity distribution and other characteristics of the slope surface, the slope seepage phenomenon can be indirectly identified.

[0167] Slope vegetation damage detection: Compare the image changes of vegetation coverage areas to identify vegetation damage. Analyze the area, degree and other parameters of vegetation damage to evaluate its impact on slope stability.

[0168] 4. Result output: The identified abnormalities are presented to relevant departments in a visual form (such as annotated images, reports, etc.).

[0169] Furthermore, the comprehensive assessment module conducts a comprehensive assessment on the stability of the highway slope, including the following steps:

[0170] 1. Data collection and collation:

[0171] Collect groundwater level monitoring data, displacement monitoring data and deformation monitoring data. Ensure that the time series of the data is complete and the accuracy meets the assessment requirements.

[0172] The 3D point cloud data, DEM (digital elevation model) data, slope morphology and abnormal area identification results of the slope are obtained from the 3D laser scanner data analysis module. The accuracy and completeness of the data are verified to ensure that there are no omissions or errors.

[0173] The image analysis module obtains high-definition images of the slope, crack identification results, vegetation damage, scour and landslide signs, etc.

[0174] Collect weather monitoring results from the weather data acquisition module;

[0175] Classify and label image data to facilitate subsequent fusion processing.

[0176] 2. Data preprocessing: All collected data are preprocessed, including denoising, correction, format unification, etc. Ensure that data from different sources can match and correspond to each other in time and space.

[0177] 3. Data fusion: Before data fusion, calibrate data from different sources to ensure their consistency in space and time. Establish a unified data framework, define a unified coordinate system and data format, and convert all data sources to this framework.

[0178] The weighted average multi-source data fusion technology is used to fuse different types of data. Considering the correlation, conflict and complementarity between data, the weights are reasonably allocated to obtain more accurate comprehensive evaluation results. The comprehensive stability evaluation formula is as follows:

[0179] ∑ m i＝1 (c i )=1; where c i is the weight of the ith data source. ST i is the quantitative value of the ith data source, including the monitoring results of the fixed-point monitoring agency, the weather monitoring results of the weather data acquisition module, the analysis results of the 3D laser scanner data analysis module, and the quantitative value corresponding to the analysis results of the image analysis module. 综合 is the quantitative value of the fused data, the slope stability assessment value. m is the total number of data sources. If ST 综合 A higher value of indicates that the slope is more stable after integrating all data sources. 综合A lower value indicates that the slope stability is weak and may require further investigation or action.

[0180] 4. Comprehensive evaluation:

[0181] The model was validated using an independent dataset to ensure the accuracy and reliability of its prediction results.

[0182] The assessment model is dynamically adjusted and optimized according to the new monitoring data and actual conditions to adapt to changes in slope stability. The various parameters of the assessment model are set according to the actual conditions, such as mechanical parameters, geological parameters, environmental factors, etc. These parameters should be determined based on geological survey reports, historical monitoring data, expert opinions, etc. The preprocessed and fused data are input into the comprehensive assessment model. The stability index or comprehensive assessment score of the slope is calculated. These indicators include safety factor, stability level, potential risk level, etc.

[0183] 5. Result output: Output the assessment results, including stability level, warning information, recommended measures, etc. Interpret and analyze the assessment results to clarify the stability status, potential risks and possible impact range of the slope.

[0184] 6. Result visualization: Visualize the evaluation results in the form of charts, maps, etc. For example, you can draw a slope stability distribution map, a potential risk area map, etc. Write a detailed evaluation report, including the purpose, method, process, results and suggestions of the evaluation. The report clearly, accurately and objectively reflects the stability status of the slope and the comprehensive evaluation results, providing a scientific basis for slope management and maintenance.

[0185] In this technical solution, the comprehensive assessment module can make full use of the data and results from different monitoring modules to conduct a comprehensive assessment of the stability of highway slopes and provide strong support for slope management and maintenance.

[0186] Reference Figure 1 The present invention provides a method for monitoring highway slope deformation, comprising the following steps:

[0187] S1. Fixed-point monitoring agencies install GNSS displacement monitoring stations, strain gauges, groundwater dynamic monitoring instruments, inclinometers and other sensors at key and critical locations on highway slopes to monitor key parameters such as slope displacement, strain, groundwater level and inclination in real time;

[0188] S2, the route planning module plans the flight route of the drone according to the topography, landform and monitoring requirements of the highway slope, ensuring that the drone can cover the monitoring area efficiently and comprehensively;

[0189] S3. The patrol monitoring agency conducts patrol monitoring according to the planned route. The high-definition camera and 3D laser scanner carried by the drone conducts high-altitude, long-distance and all-round monitoring of the slope to capture subtle changes in the slope;

[0190] S4, weather data acquisition module timely acquires local weather data, focusing on rainfall and strong wind data; timely and accurately acquires local weather data;

[0191] S5. The 3D laser scanner data analysis module pre-processes the data collected by the 3D laser scanner (such as denoising, registration, splicing, etc.), and then uses algorithms to identify abnormal conditions on the slope. Identify abnormal conditions on the slope, including cracks, collapse, landslides, scours, slope swelling, slope seepage, and slope vegetation damage;

[0192] S6, the image analysis module performs preprocessing, feature extraction and image analysis on the collected images, and identifies abnormal conditions on the slope, including cracks, collapse, landslide, scour, slope swelling, slope seepage and slope vegetation damage, etc.;

[0193] S7. The comprehensive evaluation module integrates the monitoring results of the fixed-point monitoring agency (including groundwater level, displacement monitoring results and deformation monitoring results), the analysis results of the 3D laser scanner data analysis module and the analysis results of the image analysis module, and uses the comprehensive evaluation model to comprehensively evaluate the stability of the slope.

[0194] S8. When monitoring of slope abnormalities or potential risks, the alarm module will issue an alarm in time.

[0195] The working principle of a highway slope deformation monitoring system of the present invention is as follows: a fixed-point monitoring mechanism installs GNSS displacement monitoring stations, strain gauges, groundwater dynamic monitoring instruments, inclinometers and other sensors at key and critical positions of the highway slope to monitor key parameters such as displacement, strain, groundwater level and inclination of the slope in real time; a route planning module plans the flight route of an unmanned aerial vehicle according to the topography, landform and monitoring requirements of the highway slope to ensure that the unmanned aerial vehicle can efficiently and comprehensively cover the monitoring area; a cruise monitoring mechanism performs cruise monitoring according to the planned route, and a high-definition camera and a three-dimensional laser scanner carried by the unmanned aerial vehicle perform high-altitude, long-distance and all-round monitoring of the slope to capture subtle changes in the slope; a weather data acquisition module timely acquires local weather data, focusing on rainfall and strong wind data; and timely and accurately acquires local weather data; a three-dimensional laser scanner data analysis module pre-processes the data collected by the three-dimensional laser scanner (such as denoising, registration, splicing, etc.), and then uses an algorithm to identify abnormal conditions on the slope. Identify abnormal conditions on the slope, including cracks, collapse, landslide, scour, slope swelling, slope seepage and slope vegetation damage; the image analysis module preprocesses, extracts features and analyzes the collected images to identify abnormal conditions on the slope, including cracks, collapse, landslide, scour, slope swelling, slope seepage and slope vegetation damage; the comprehensive evaluation module integrates the monitoring results of the fixed-point monitoring agency (including groundwater level, displacement monitoring results and deformation monitoring results), the analysis results of the 3D laser scanner data analysis module and the analysis results of the image analysis module, and uses the comprehensive evaluation model to conduct a comprehensive assessment of the stability of the slope. When the slope is monitored to have abnormal conditions or potential risks, the alarm module will issue an alarm in time.

[0196] The present invention combines a fixed-point monitoring mechanism with a cruise monitoring mechanism to conduct all-round monitoring of the slope. The fixed-point monitoring mechanism can continuously and stably monitor the key parameters of the slope, such as displacement, strain, groundwater level and inclination, by installing high-precision sensors, and provide accurate basic data. The cruise monitoring mechanism, especially the high-definition camera and three-dimensional laser scanner carried by the drone, can capture the macro changes and subtle features of the slope from a high-altitude perspective. The combination of the two realizes all-round monitoring from macro to micro, from static to dynamic, greatly improving the accuracy and comprehensiveness of monitoring.

[0197] Through automated and intelligent monitoring methods, fixed-point monitoring agencies and patrol monitoring agencies can carry out monitoring work continuously and stably, reducing the frequency and intensity of manual inspections. At the same time, the combination of the two realizes the complementarity and coordination of monitoring work, further improves monitoring efficiency, and reduces labor costs. The 3D laser scanner data analysis module and image analysis module use advanced algorithms to pre-process and deeply analyze the collected data, and can accurately identify various abnormal conditions such as cracks, collapses, landslides, scours, slope swelling, slope seepage, and vegetation damage on the slope, thereby improving the accuracy and reliability of monitoring. The comprehensive evaluation module integrates data from multiple monitoring methods and uses a comprehensive evaluation model to conduct a comprehensive assessment of the stability of the slope, providing scientific and comprehensive information support for decision makers, which helps to formulate more reasonable and effective slope management and maintenance plans.

[0198] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring highway slope deformation, characterized in that: The following steps are involved: S1. Fixed-point monitoring agencies install GNSS displacement monitoring stations, strain gauges, groundwater dynamic monitoring instruments and inclinometers at key and critical locations on highway slopes to monitor the displacement, strain, groundwater level and inclination parameters of the slopes in real time; S2, the route planning module plans the flight route of the UAV according to the topography, landform and monitoring requirements of the highway slope; S3. The patrol monitoring agency conducts patrol monitoring according to the planned route, and the high-definition camera and 3D laser scanner carried by the drone conduct all-round monitoring of the slope; S4, weather data acquisition module timely acquires local weather data, focusing on rainfall and strong wind data; S5, the 3D laser scanner data analysis module pre-processes the data collected by the 3D laser scanner, and then uses an algorithm to identify abnormal conditions on the slope; S6, the image analysis module performs preprocessing, feature extraction and image analysis on the collected images, and identifies abnormal conditions on the slope; S7, the comprehensive evaluation module integrates the monitoring results of the fixed-point monitoring agency, the weather monitoring results of the weather data acquisition module, the analysis results of the 3D laser scanner data analysis module and the analysis results of the image analysis module, and uses the comprehensive evaluation model to comprehensively evaluate the stability of the slope; S8. When monitoring of slope abnormalities or potential risks, the alarm module will issue an alarm in time.

2. The highway slope deformation monitoring method according to claim 1, characterized in that: Step S1 includes the following steps: S11. Monitoring point selection planning: Determine the key and critical monitoring locations based on the topography, geological conditions, and historical monitoring data of the highway slope; ensure that the location of the monitoring points can fully reflect the overall stability and potential risks of the slope; S12. On-site investigation: Conduct on-site investigation of monitoring points to determine the specific location and method of equipment installation; S13, Equipment installation: Install GNSS displacement monitoring stations, connect various components, conduct preliminary debugging, and ensure the normal operation of the equipment; install strain gauges at key locations on the slope, install groundwater dynamic monitoring instruments inside the slope, and install inclinometers at key locations on the slope; S14, Equipment debugging and calibration: debug all installed equipment and check the stability and accuracy of data transmission; S15. Data monitoring and analysis: Data from all monitoring equipment is transmitted to the data processing center; a database and data processing platform for the monitoring system is established to store, process and analyze the data; the displacement, strain, groundwater level and tilt parameters of the slope are monitored in real time through the monitoring system, abnormal conditions are discovered in time and the early warning mechanism is automatically triggered; the monitoring data is processed and analyzed to evaluate the stability of the slope, and a comprehensive evaluation is performed according to the following formula: I assessment =w1*I displacement +w2*I strain +w3*I water_level +w4*I inclination ;w1+w2+w3+w4=1; Among them, I assessment is the comprehensive evaluation index, I displacement Indicates the monitoring result of slope displacement; I strain Represents the monitoring result of slope strain; I water_level Indicates the monitoring result of groundwater level; I inclination Represents the monitoring result of slope inclination; w1, w2, w3 and w4 are weight coefficients, corresponding to the weights of displacement, strain, water level and inclination respectively.

3. The highway slope deformation monitoring method according to claim 1, characterized in that: Step S5 includes the following steps: S51, data preprocessing: denoising the data to remove noise points introduced by environmental factors, equipment errors or reflective surface characteristics; S52, Registration: Use the iterative closest point ICP algorithm to achieve data registration by minimizing the geometric differences between different point clouds, converting point cloud data from different scanning positions or different time points into a unified coordinate system, and achieving accurate alignment of multi-site data; S53, stitching: stitching together the parts according to the overlapping areas and registration parameters between the point clouds to form a continuous and seamless 3D surface; merging the registered multiple point cloud data sets into a complete 3D model; S54, point cloud segmentation: based on the spatial position, geometric shape and color attributes of the points, the point cloud data is segmented into regions or objects with similar attributes or geometric features using a clustering algorithm to facilitate subsequent analysis and processing; S55, Feature extraction: Use point cloud processing technology to identify plane, edge and corner geometric features, as well as slope and curvature morphological features; S56. Abnormal situation identification: Use edge detection algorithm to identify cracks by detecting discontinuous areas or abnormal change areas in the point cloud; monitor changes by comparing scan data at different time points, analyze the height change and morphological distortion of the point cloud, and identify the collapse area and landslide body of the slope; observe the point cloud density and morphological changes at the bottom or edge of the slope to identify scour phenomena; analyze the curvature changes and geometric features of the slope surface to identify slope swelling phenomena; indirectly judge slope seepage by observing the changes in the humidity of the slope surface; identify vegetation damage phenomena by comparing the point cloud density and color changes in the vegetation coverage area; S57. Result output: The identified abnormalities are presented to the relevant departments in a visual form.

4. The highway slope deformation monitoring method according to claim 3 is characterized in that: In step S52, the registration is performed using the following formula: In the formula, E ICP represents the registration error; i is the index of the summation, representing the point pair; N is the total number of point pairs in the point cloud; w i is the weight factor of the i-th pair of points; p i is the i-th point in the source point cloud; q i is the same as p in the target point cloud i The nearest point; |||p i -q i ∣∣ is point p i and q i The Euclidean distance between .

5. The method for monitoring highway slope deformation according to claim 3, characterized in that: In step S52, the point cloud density is evaluated at different scales according to the following formula to capture the features of different scales: In the formula, O overlap represents the splicing overlap metric; S is the number of scales; overlap s is the overlapping area at scale s; N s represents the total number of points in the point cloud dataset at scale s; d i,s is the density value of point i at scale s; s is a specific scale level used to index different scales, with values ​​ranging from 1 to S.

6. The highway slope deformation monitoring method according to claim 3 is characterized by: In step S52, the clustering algorithm similarity calculation for point cloud segmentation is performed using the following formula: In the formula, S cluster is the cluster similarity; N cluster is the number of points in the cluster; P[I(p i )∣μ c ,Σ c ] is point p i The attribute vector I(p i ) is the probability density function under the probability model of the cluster center c; under the parameters of the given cluster center, the point p is observed i The probability of the attribute; I(p i ) is point p i The attribute vector, μ c is the mean vector of cluster center c; Σ c is the covariance matrix of cluster center c; p i is the i-th point in the cluster.

7. The highway slope deformation monitoring method according to claim 1, characterized in that: Step S6 includes the following steps: S61, Image preprocessing: including noise removal, image enhancement and image segmentation; S62, feature extraction: extracting features from the preprocessed image, including geometric features, texture features, and color features; S63, Image analysis: including crack detection, collapse and landslide detection, scour detection, slope swelling detection, slope seepage detection and slope vegetation damage detection; Crack detection: Use the Canny edge detection algorithm and morphological operations to identify cracks in the image; analyze the length, width and direction parameters of the cracks to assess their severity; Collapse and landslide detection: By comparing images at different time points, analyzing the height changes and morphological distortions of the slope surface, the collapse and landslide areas can be identified; Scour detection: observe the image changes at the bottom or edge of the slope, identify the erosion caused by scour, analyze the area and depth parameters of the scour area, and evaluate its impact range; Slope bulging detection: Analyze the curvature changes and geometric features of the slope surface to identify slope bulging phenomena; Slope seepage detection: Identify slope seepage by analyzing the color changes and humidity distribution characteristics of the slope surface; Slope vegetation damage detection: compare image changes in vegetation coverage areas to identify vegetation damage; analyze the area and extent of vegetation damage to assess its impact on slope stability; S64. Result output: The identified abnormalities are presented to the relevant departments in a visual form.

8. The highway slope deformation monitoring method according to claim 2, characterized in that: Step S7 includes the following steps: S71, data collection and collation: collecting monitoring results of the fixed-point monitoring agency, weather monitoring results of the weather data acquisition module, analysis results of the 3D laser scanner data analysis module, and analysis results of the image analysis module; S72. Data preprocessing: preprocess all collected data, including noise removal, correction and format unification; ensure that data from different sources can match and correspond to each other in time and space; S73. Data fusion: Calibrate data from different sources to ensure their consistency in space and time; establish a unified data framework, define a unified coordinate system and data format, and convert all data sources to this framework; use weighted average multi-source data fusion technology to fuse different types of data. The comprehensive stability evaluation formula is as follows: ∑ m i＝1 (c i )=1; where c i is the weight of the ith data source; ST i is the quantitative value of the i-th data source; ST 综合 is the quantified value of the fused data, the slope stability assessment value; m is the total number of data sources; S74. Comprehensive evaluation: Use independent data sets to validate the model to ensure the accuracy and reliability of its prediction results; dynamically adjust and optimize the evaluation model based on new monitoring data and actual conditions to adapt to changes in slope stability; input the preprocessed and fused data into the comprehensive evaluation model; calculate the slope stability index; S75, Result output: Output the evaluation results, including stability level, warning information and recommended measures; S76. Result visualization: Visualize the evaluation results in the form of charts and maps.

9. The highway slope deformation monitoring method according to claim 1, characterized in that: Step S2 includes the following steps: S21. Data collection: determine the specific content of monitoring, obtain accurate topographic map and elevation model DEM of the slope; identify the slope gradient, concave and convex terrain and obstacles; S22, setting flight parameters: setting a suitable flight altitude according to monitoring requirements and camera performance to ensure adequate resolution and avoid collisions; determining a suitable flight speed according to the camera's shutter speed, shooting range, and the level of detail required for monitoring; S23. Route planning: Use DJI GS Pro drone route planning software, input terrain data, set the route according to the parameters, and manually set waypoints in key monitoring areas or complex terrain to ensure that the drone can fly accurately and capture important data; set safety buffers on both sides and the top of the route to avoid collisions between the drone and obstacles; S24, simulation adjustment: simulate the flight path of the drone in the planning software, check whether there are uncovered areas or potential safety issues, and adjust the route parameters according to the simulation results until the optimal monitoring effect and flight efficiency are achieved; S25. Field verification and adjustment: Conduct field test flights before formal monitoring to check whether the route is executed as expected and whether the shooting data meets the requirements. If necessary, adjust the route to further optimize subsequent flight route planning.

10. A highway slope deformation monitoring system, comprising: Data collection module, fixed-point monitoring mechanism, route planning module, cruise monitoring mechanism, weather data acquisition module, three-dimensional laser scanner data analysis module, image analysis module, comprehensive evaluation module, alarm module and central control unit; characterized in that: Data collection module: collects a large amount of highway slope data, including environmental data, geological data, and real-time slope status data, and annotates the collected images into training samples; Fixed-point monitoring institutions: Install GNSS displacement monitoring stations, strain gauges, groundwater dynamic monitoring instruments and inclinometers at key and critical locations on highway slopes to monitor slope displacement, strain, groundwater level and inclination parameters in real time; Route planning module: Plan the flight route of the drone according to the terrain, topography and monitoring requirements of the highway slope to ensure that the drone can cover the monitoring area efficiently and comprehensively; Cruise monitoring mechanism: including drones, high-definition cameras and 3D laser scanners; cruise monitoring is carried out according to the planned route, and the high-definition cameras and 3D laser scanners carried by drones conduct all-round monitoring of the slopes; Weather data acquisition module: timely obtain local weather data, focusing on rainfall and strong wind data; 3D laser scanner data analysis module: pre-processes the data collected by the 3D laser scanner, and then uses algorithms to identify abnormal conditions on the slope; Image analysis module: pre-processing, feature extraction and image analysis of the collected images to identify abnormal conditions on the slope; Comprehensive evaluation module: integrates the monitoring results of the fixed-point monitoring agency, the weather monitoring results of the weather data acquisition module, the analysis results of the 3D laser scanner data analysis module and the analysis results of the image analysis module, and uses the comprehensive evaluation model to conduct a comprehensive evaluation of the stability of the slope; Alarm module: including an alarm. When monitoring the slope for abnormal conditions or potential risks, the alarm module will issue an alarm in time. Central control unit: network connected with alarm module, fixed-point monitoring mechanism, route planning module, cruise monitoring mechanism, 3D laser scanner data analysis module, image analysis module and comprehensive evaluation module.

Images