Geological disaster detection method and monitoring system based on unmanned aerial vehicle scanning

Abstract

The invention relates to the technical field of geological disaster detection, in particular to a geological disaster detection method and monitoring system based on unmanned aerial vehicle scanning. Comprising the following steps: S1, configuring an unmanned aerial vehicle-mounted tilt camera, a multispectral laser radar and a high-precision positioning module, planning a route, carrying out multi-angle scanning on a target area, and obtaining earth surface three-dimensional point cloud data, a multispectral image and terrain elevation information; s2, preprocessing the collected original image data, including point cloud denoising, image distortion correction and multi-source data registration, and generating a high-resolution live-action model fused with three-dimensional geographic information data; s3, extracting various data based on the live-action three-dimensional model to construct a geological disaster hidden danger analysis model, calculating a risk index through multi-factor weighted fusion, training a transfer learning model in combination with historical disaster data, and outputting a hidden danger type and probability; according to the invention, the geological disaster type can be analyzed based on the geological condition and the targeted processing strategy can be specified.

Description

Technical Field

[0001] The present invention relates to the technical field of geological disaster detection, and in particular to a geological disaster detection method and monitoring system based on drone scanning. Background Art

[0002] Geological disasters (such as landslides, collapses, and mudslides) pose a serious threat to people's lives and property due to their sudden and destructive nature. Traditional geological disaster monitoring relies primarily on manual inspections and remote sensing satellite technology. However, manual surveys are inefficient and highly risky, while satellite remote sensing is limited by resolution, weather conditions, and cost, making it difficult to obtain high-precision data in a timely manner. In recent years, drone technology has significantly improved data collection efficiency and accuracy through low-altitude remote sensing and lidar scanning. However, existing research has primarily focused on hazard identification and monitoring, lacking a closed-loop technology system for disaster response strategies.

[0003] Current geological disaster response measures rely heavily on manual decision-making based on experience, which can be subject to lags and subjectivity. For example, reinforcement scheme designs often employ static parameters, failing to dynamically adjust key parameters like anchor length and monitoring network density based on hazard probability. Emergency evacuation route planning also lacks support from real-world 3D models, resulting in inefficient responses. Furthermore, existing technologies lack an automated mechanism for matching hazard types with response strategies, preventing closed-loop management from data collection to project implementation.

[0004] Therefore, based on the above problems, there is an urgent need for a technical solution that integrates multi-source data collection, dynamic assessment of hidden dangers and generation of intelligent response strategies to solve the core problems of the lack of response measures and low decision-making efficiency in existing technologies. Summary of the Invention

[0005] In view of the lag and subjectivity of current measures to deal with geological disasters, this application provides a geological disaster detection method based on drone scanning, which is characterized by including: S1: The drone is equipped with an oblique camera, a multispectral lidar, and a high-precision positioning module. It plans a route to scan the target area from multiple angles, acquiring three-dimensional point cloud data, multispectral images, and terrain elevation information. S2: Preprocess the collected raw image data, including point cloud denoising, image distortion correction, and multi-source data registration, and generate a high-resolution real-scene model that integrates 3D geographic information data; S3: Based on the real-life 3D model, a geological hazard analysis model is constructed by extracting coverage, slope factor, terrain relief, fracture distance factor, lithology distribution parameters, and soil moisture index. The risk index is calculated through multi-factor weighted fusion, and a transfer learning model is trained in combination with historical disaster data to output the hazard type and probability. S4: For rock collapse hazards, analyze the correlation between rock mass stability coefficient and slope factor and generate dynamic warning values; S5: Match the response strategy library according to the emergency type, and output strategies including reinforcement project parameters, monitoring network layout plan and emergency evacuation route.

[0006] Adopt the above technical solution: The above solution provides a geological disaster detection method, which can realize image acquisition of the area to be detected based on the drone camera, analyze the image based on the analysis of different types of data, confirm the risk index, and generate corresponding response strategies.

[0007] Preferably, it also includes: Real-time data transmission and dynamic model updates are achieved through the cloud data platform, and risk prediction results are dynamically adjusted in combination with meteorological data. The effectiveness of the response strategy is verified, and engineering parameters are optimized by simulating collapse trajectories and comparing them with stability after reinforcement.

[0008] Adopting the above technical solution: the above solution can analyze the strategies adopted based on the acquisition and analysis of geological hazards, and further analyze the effectiveness of the response strategies.

[0009] Preferably, the risk index is calculated as follows: ; Among them, K t is the risk index, a, 、 、 as well as is the risk weight coefficient, which is dynamically allocated through the entropy weight method; S is the normalized slope value; is the terrain relief; is the distance to the fault zone; is the vegetation cover; is the soil moisture index.

[0010] Adopting the above technical solution: the above solution further refines and discloses the calculation formula of the risk index, which can realize the calculation of the geological disaster risk index based on slope value, terrain undulation, fault zone distance, vegetation coverage and soil moisture index.

[0011] Preferably, the calculation formula of the rock mass stability coefficient is: ; Where: P s is the rock mass stability coefficient, is the compressive strength of rock mass, is the slope inclination, is the rock mass self-weight stress, is the acceleration due to gravity, and h is the depth of the potential sliding surface.

[0012] Adopting the above technical solution: The above solution further optimizes the calculation formula of the rock stability coefficient, and can realize the calculation of the rock stability coefficient based on the rock compressive strength, slope inclination, rock self-weight stress and potential sliding surface depth.

[0013] Preferably, the reinforcement engineering parameters act on a reinforcement support device, the reinforcement support device includes an anchor rod, the reinforcement engineering parameters include the length and spacing of the anchor rod, and the calculation formula of the reinforcement engineering parameters is: ; ; in, is the anchor rod length, is the anchor spacing, F p is the slope factor, H is the slope height, is the friction angle in the rock mass, d is the diameter of the anchor rod, and f y is the anchor yield strength, Design tensile strength.

[0014] Adopting the above technical solution: the above solution can realize the calculation of the length and spacing factors of the reinforcement support equipment based on factors such as slope data, so as to realize the reinforcement and support of the rock mass.

[0015] Preferably, the coping strategy library includes: Rock collapse treatment: Based on the dynamic warning value generated by the correlation between the analyzed rock mass stability coefficient and the slope shape factor, the corresponding anchor cable equipment size is analyzed, and prestressed anchor cables are used for reinforcement. Flexible protective nets are used to receive the broken rocks and reduce the gravity of the rocks. Landslide hazard treatment: underground water level sensors are deployed to monitor the groundwater level, and surface displacement sensors are set up to survey landslide conditions, and treatment is carried out by intercepting and photographing ditches; Debris flow risk management: Based on the debris flow risk assessment, retaining dams are constructed and vegetation restoration data is analyzed. Development is prohibited in areas with slopes greater than 28°.

[0016] Adopting the above technical solution: the above solution can realize the matching of corresponding anchor equipment devices based on the analysis and calculation of the rock mass; the detection of groundwater level and landslide data can realize the effective analysis of landslide hazards, and set corresponding treatment based on the debris flow analysis.

[0017] Preferably, the triggering condition for dynamically adjusting the risk prediction is: When continuous rainfall exceeds 50 mm / 24 h, the risk index weight increases by 30%; when the vegetation coverage rate decreases by 10%, the fracture distance factor increases to 10%.

[0018] Adopting the above technical solution: the above solution can further optimize the risk index weight in the formula based on the rainfall situation, and optimize the fracture distance factor based on the vegetation coverage situation.

[0019] Preferably, the collapse trajectory is simulated using a discrete element method, and the inter-particle force model of the discrete element method is: ; ;

[0020] Among them, K n is the normal stiffness coefficient, K t is the tangential stiffness coefficient, C n is the damping coefficient, is the friction coefficient.

[0021] Adopting the above technical solution: the above solution further refines the model analysis formula of the discrete element method for simulating collapse trajectory, and can realize the generation of analysis model based on the algorithm formula.

[0022] A monitoring system, applied to any one of the above detection methods, comprising: UAV image data acquisition unit: The drone is equipped with an oblique camera, spectral lidar, and a high-precision positioning module. It plans a route to scan the target area from multiple angles to obtain three-dimensional point cloud data, multispectral images, and terrain elevation information. Image processing unit: pre-processes the collected raw image data, including point cloud denoising, image distortion correction, and multi-source data registration, and generates a high-resolution real-scene model that integrates 3D geographic information data; Hazard type analysis unit: Builds a geological hazard analysis model based on various analysis factors of the real-scene 3D model, calculates the risk index through multi-factor weighted fusion, and trains a transfer learning model based on historical disaster data to output the hazard type and probability; Dynamic warning value analysis unit: for rock collapse hazards, analyzes the correlation between rock mass stability coefficient and slope factor, and generates dynamic warning values; Emergency strategy generation unit: matches the response strategy library according to the emergency type, and outputs strategies including reinforcement project parameters, monitoring network layout plan and emergency evacuation route.

[0023] Preferably, the UAV is configured as follows: oblique camera resolution ≥ 20 million pixels, frame rate 30fps; laser radar scanning accuracy ± 3 cm / 100 m; real-time differential GPS positioning accuracy ≤ 2 cm. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 This is a flow chart of the geological disaster detection method based on drone scanning in this application.

[0026] In the picture: 1. UAV image data acquisition unit; 2. Image processing unit; 3. Emergency type analysis unit; 4. Dynamic warning value analysis unit; 5. Emergency strategy generation unit. DETAILED DESCRIPTION

[0027] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0028] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0029] See also Figure 1 In the existing technology, the analysis of geological disasters is often obtained through corresponding data analysis, but it does not have the ability to analyze the occurrence period of geological disasters, nor can it scientifically analyze the corresponding response strategies based on the geological disaster situation. Based on the above problems, this application provides a geological disaster detection method based on drone scanning, which is characterized by including: S1: Configure a drone equipped with an oblique camera, a multispectral lidar, and a high-precision positioning module, plan a route to scan the target area from multiple angles, and obtain surface three-dimensional point cloud data, multispectral images, and terrain elevation information. This solution uses drones and corresponding camera modules to scan the target area from multiple angles, obtaining the desired three-dimensional point cloud data, multispectral image data, and terrain elevation information for subsequent image processing units to perform corresponding processing. Prior to data collection, corresponding route planning will be performed, and the overlap rate will be set according to the complexity of the terrain in the target area. The heading overlap rate is >80%, and the lateral overlap rate is >70%. A zigzag route is used to cover the entire area, and the lidar scanning frequency is set to 300kHz13.

[0030] S2: Preprocess the collected raw image data, including point cloud denoising, image distortion correction, and multi-source data registration, and generate a high-resolution real-scene model that integrates 3D geographic information data; S3: Based on the real-life 3D model, a geological hazard analysis model is constructed by extracting coverage, slope factor, terrain relief, fracture distance factor, lithology distribution parameters, and soil moisture index. The risk index is calculated through multi-factor weighted fusion, and a transfer learning model is trained in combination with historical disaster data to output the hazard type and probability. S4: For rock collapse hazards, analyze the correlation between rock mass stability coefficient and slope factor and generate dynamic warning values; S5: Match the response strategy library according to the emergency type, and output strategies including reinforcement project parameters, monitoring network layout plan and emergency evacuation route.

[0031] It is worth mentioning that the above scheme provides a geological disaster detection method, which can realize image acquisition of the inspection area based on the drone camera, analyze the images based on different types of data, confirm the risk index, and generate corresponding response strategies.

[0032] It also includes: achieving real-time data transmission and dynamic model updates through a cloud data platform, dynamically adjusting risk prediction results in combination with meteorological data; verifying the effectiveness of response strategies, and optimizing engineering parameters by comparing simulated collapse trajectories with stability after reinforcement.

[0033] The above scheme can analyze the strategies adopted based on the acquisition and analysis of geological hazards, and further analyze the effectiveness of the response strategies.

[0034] Preferably, the risk index is calculated as follows: ; The weight coefficient is dynamically adjusted by the entropy weight method, and the normalized factor value is input. The normalized factor value range is 0-1. Among them, K t is the risk index, a, 、 、 as well as is the risk weight coefficient, which is dynamically allocated through the entropy weight method; S is the normalized slope value; is the terrain relief; is the distance to the fault zone; is the vegetation cover; is the soil moisture index.

[0035] In the above scheme, when K t >0.7 and the lithology distribution parameter is hard rock, the hidden danger type is determined to be rock collapse. t When the soil moisture index is >0.6 and >0.8, it is determined that there is a landslide hazard. When the calculated terrain relief is >50m / km 2 And when the slope factor is greater than 25°, a debris flow risk warning will be issued.

[0036] In K t >0.6 area is calculated as 50m GNSS monitoring stations are deployed in a 50m grid for monitoring network deployment, with data return frequency > 1 time / min; Based on the real 3D model, the normalized slope value S>30° and the fault zone are avoided. Emergency evacuation route planning is carried out for paths less than 100m.

[0037] It is worth mentioning that the above scheme further refines and discloses the calculation formula of the risk index, which can realize the calculation of the geological disaster risk index based on slope value, terrain undulation, fault zone distance, vegetation coverage and soil moisture index.

[0038] The calculation formula of the rock mass stability coefficient is: ; Where: P s is the rock mass stability coefficient, is the compressive strength of rock mass, is the slope inclination, is the rock mass self-weight stress, is the acceleration due to gravity, and h is the depth of the potential sliding surface.

[0039] The above scheme further optimizes the calculation formula of the rock mass stability coefficient, and can calculate the rock mass stability coefficient based on the rock mass compressive strength, slope inclination, rock mass self-weight stress and potential sliding surface depth.

[0040] The reinforcement engineering parameters act on the reinforcement support device, the reinforcement support device includes anchor rods, the reinforcement engineering parameters include the length and spacing of the anchor rods, and the calculation formula of the reinforcement engineering parameters is: ; ; in, is the anchor rod length, is the anchor spacing, F p is the slope factor, H is the slope height, is the friction angle in the rock mass, d is the diameter of the anchor rod, and f y is the anchor yield strength, Design tensile strength.

[0041] The above scheme can calculate the length and spacing of the reinforcement support equipment based on factors such as slope data, so as to achieve reinforcement and support of the rock mass.

[0042] The response strategy library includes: Rock collapse treatment: Based on the dynamic warning value generated by the correlation between the analyzed rock mass stability coefficient and the slope shape factor, the corresponding anchor cable equipment size is analyzed, and prestressed anchor cables are used for reinforcement. Flexible protective nets are used to receive the broken rocks and reduce the gravity of the rocks. Landslide hazard treatment: underground water level sensors are deployed to monitor the groundwater level, and surface displacement sensors are set up to survey landslide conditions, and treatment is carried out by intercepting and photographing ditches; Debris flow risk management: Based on the debris flow risk assessment, retaining dams are constructed and vegetation restoration data is analyzed. Development is prohibited in areas with slopes greater than 28°.

[0043] The above scheme can realize the matching of corresponding anchor equipment devices based on rock mass analysis and calculation; the detection of groundwater level and landslide data can effectively analyze landslide hazards and set corresponding treatment based on mud-rock flow analysis.

[0044] The triggering conditions for dynamically adjusting risk prediction are: When continuous rainfall exceeds 50 mm / 24 h, the risk index weight increases by 30%; when the vegetation coverage rate decreases by 10%, the fracture distance factor increases to 10%.

[0045] The above scheme can further optimize the risk index weight in the formula based on rainfall conditions and optimize the fracture distance factor based on vegetation coverage.

[0046] The collapse trajectory is simulated using the discrete element method, and the inter-particle force model of the discrete element method is: ; ; Among them, K n is the normal stiffness coefficient, K t is the tangential stiffness coefficient, C n is the damping coefficient, is the friction coefficient.

[0047] The above scheme further refines the model analysis formula of the discrete element method for simulating collapse trajectories, and can generate an analysis model based on the algorithm formula.

[0048] A monitoring system, applied to any one of the above detection methods, comprising: UAV image data acquisition unit: The drone is equipped with an oblique camera, spectral lidar, and a high-precision positioning module. It plans a route to scan the target area from multiple angles to obtain three-dimensional point cloud data, multispectral images, and terrain elevation information. Image processing unit: pre-processes the collected raw image data, including point cloud denoising, image distortion correction, and multi-source data registration, and generates a high-resolution real-scene model that integrates 3D geographic information data; Hazard type analysis unit: Builds a geological hazard analysis model based on various analysis factors of the real-scene 3D model, calculates the risk index through multi-factor weighted fusion, and trains a transfer learning model based on historical disaster data to output the hazard type and probability; Dynamic warning value analysis unit: for rock collapse hazards, analyzes the correlation between rock mass stability coefficient and slope factor, and generates dynamic warning values; Emergency strategy generation unit: matches the response strategy library according to the emergency type, and outputs strategies including reinforcement project parameters, monitoring network layout plan and emergency evacuation route.

[0049] Preferably, the UAV is configured as follows: oblique camera resolution ≥ 20 million pixels, frame rate 30fps; laser radar scanning accuracy ± 3 cm / 100 m; real-time differential GPS positioning accuracy ≤ 2 cm.

[0050] Unless otherwise specified, the device components involved in the above embodiments are all conventional device components, and the connection methods and control methods involved are all conventional connection methods and control methods unless otherwise specified.

[0051] The present invention has been described in detail above with reference to the embodiments. However, those skilled in the art will appreciate that, without departing from the spirit of the present invention, the specific parameters in the above embodiments may be modified to form multiple specific embodiments, which are all within the common variation range of the present invention and will not be described in detail here.

Claims

1. A geological disaster detection method based on drone scanning, characterized in that: include: S1: The drone is equipped with an oblique camera, a multispectral lidar, and a high-precision positioning module. It plans a route to scan the target area from multiple angles, acquiring three-dimensional point cloud data, multispectral images, and terrain elevation information. S2: Preprocess the collected raw image data, including point cloud denoising, image distortion correction, and multi-source data registration, and generate a high-resolution real-scene model that integrates 3D geographic information data; S3: Based on the real-life 3D model, a geological hazard analysis model is constructed by extracting coverage, slope factor, terrain relief, fracture distance factor, lithology distribution parameters, and soil moisture index. The risk index is calculated through multi-factor weighted fusion, and a transfer learning model is trained in combination with historical disaster data to output the hazard type and probability. S4: For rock collapse hazards, analyze the correlation between rock mass stability coefficient and slope factor and generate dynamic warning values; S5: Match the response strategy library according to the emergency type, and output strategies including reinforcement project parameters, monitoring network layout plan and emergency evacuation route.

2. The method for detecting geological hazards based on drone scanning according to claim 1, characterized in that: Also includes: Realize real-time data transmission and dynamic model updates through the cloud data platform, and dynamically adjust risk prediction results in combination with meteorological data; The effectiveness of the response strategy is verified, and the engineering parameters are optimized by simulating the collapse trajectory and comparing it with the stability after reinforcement.

3. The method for detecting geological disasters based on drone scanning according to claim 1, characterized in that: The calculation formula of the risk index is: ； Among them, K t is the risk index, a, 、 、 as well as is the risk weight coefficient, which is dynamically allocated through the entropy weight method; S is the normalized slope value; is the terrain relief; is the distance to the fault zone; is the vegetation cover; is the soil moisture index.

4. The method for detecting geological hazards based on drone scanning according to claim 1, characterized in that: The calculation formula of the rock mass stability coefficient is: ； Where: P s is the rock mass stability coefficient, is the compressive strength of rock mass, is the slope inclination, is the rock mass self-weight stress, is the acceleration due to gravity, and h is the depth of the potential sliding surface.

5. The method for detecting geological disasters based on drone scanning according to claim 4, characterized in that: The reinforcement engineering parameters act on the reinforcement support device, the reinforcement support device includes anchor rods, the reinforcement engineering parameters include the length and spacing of the anchor rods, and the calculation formula of the reinforcement engineering parameters is: ； ； in, is the anchor rod length, is the anchor spacing, F p is the slope factor, H is the slope height, is the friction angle in the rock mass, d is the diameter of the anchor rod, and f y is the anchor yield strength, Design tensile strength.

6. The method for detecting geological hazards based on drone scanning according to claim 1, characterized in that: The response strategy library includes: Rock collapse treatment: Based on the dynamic warning value generated by the correlation between the analyzed rock mass stability coefficient and the slope shape factor, the corresponding anchor cable equipment size is analyzed, and prestressed anchor cables are used for reinforcement. Flexible protective nets are used to receive the broken rocks and reduce the gravity of the rocks. Landslide hazard treatment: underground water level sensors are deployed to monitor the groundwater level, and surface displacement sensors are set up to survey landslide conditions, and treatment is carried out by intercepting and photographing ditches; Debris flow risk management: Based on the debris flow risk assessment, retaining dams are constructed and vegetation restoration data is analyzed. Development is prohibited in areas with slopes greater than 28°.

7. The method for detecting geological hazards based on drone scanning according to claim 1, characterized in that: The triggering conditions for dynamically adjusting risk prediction are: When continuous rainfall exceeds 50 mm / 24 h, the risk index weight increases by 30%; when the vegetation coverage rate decreases by 10%, the fracture distance factor increases to 10%.

8. The method for detecting geological disasters based on drone scanning according to claim 2, characterized in that: The collapse trajectory is simulated using the discrete element method, and the inter-particle force model of the discrete element method is: ； ； Among them, K n is the normal stiffness coefficient, K t is the tangential stiffness coefficient, C n is the damping coefficient, is the friction coefficient.

9. A monitoring system, applied to the detection method according to any one of claims 1 to 8, characterized in that: include: UAV image data acquisition unit: The drone is equipped with an oblique camera, spectral lidar, and a high-precision positioning module. It plans a route to scan the target area from multiple angles to obtain three-dimensional point cloud data, multispectral images, and terrain elevation information. Image processing unit: pre-processes the collected raw image data, including point cloud denoising, image distortion correction, and multi-source data registration, and generates a high-resolution real-scene model that integrates 3D geographic information data; Hazard type analysis unit: Builds a geological hazard analysis model based on various analysis factors of the real-scene 3D model, calculates the risk index through multi-factor weighted fusion, and trains a transfer learning model based on historical disaster data to output the hazard type and probability; Dynamic warning value analysis unit: for rock collapse hazards, analyzes the correlation between rock mass stability coefficient and slope factor, and generates dynamic warning values; Emergency strategy generation unit: matches the response strategy library according to the emergency type, and outputs strategies including reinforcement project parameters, monitoring network layout plan and emergency evacuation route.

10. A monitoring system according to claim 1, characterized in that: The drone is configured as follows: oblique camera with a resolution of ≥ 20 million pixels and a frame rate of 30 fps; lidar scanning accuracy of ±3 cm / 100 m; and real-time differential GPS positioning accuracy of ≤2 cm.

Images