Regional risk monitoring and early warning system based on space-air-ground multi-source data fusion

Abstract

The invention discloses a regional risk monitoring and early warning system based on space-air-ground multi-source data fusion, and relates to the technical field of disaster monitoring and early warning. According to the invention, comprehensive monitoring of regional risks is realized by fusing various data sources such as satellite remote sensing, unmanned aerial vehicle aerial survey and ground-based radar. The system can automatically process data, construct a high-precision three-dimensional real scene model, and identify a potential risk area based on a time sequence InSAR technology and a multi-source data fusion technology. And meanwhile, non-contact high-precision measurement is carried out on key facilities. An interactive monitoring and early warning platform is constructed, functions of deformation graph generation, target detection and risk division, data fusion processing, potential risk surface determination, stress-strain analysis and early warning information release are realized, and support is provided for regional risk prevention. The system has the characteristics of high automation, intelligence and visualization, and is suitable for various application scenes such as earthquakes, floods, landslides and the like.

Description

Technical Field

[0001] The present invention belongs to the technical field of disaster monitoring and early warning, and specifically relates to a regional risk monitoring and early warning system based on multi-source data fusion of air, space, and land. Background Art

[0002] With the rapid development of social economy and the acceleration of urbanization, regional risk problems caused by various natural disasters and human factors have become increasingly prominent. These problems not only threaten the safety of people's lives and property, but also have a serious impact on the sustainable development of the economy and society. Therefore, establishing an effective regional risk monitoring and early warning system is of great significance for preventing and mitigating the impact of natural and man-made disasters.

[0003] The existing regional risk monitoring and early warning system mainly relies on a single data source and technical means, and has the following shortcomings:

[0004] Single data source: Traditional monitoring systems usually rely on a single data source, such as ground sensors or satellite remote sensing data, which cannot fully reflect the complex and changing regional risk conditions.

[0005] Limited monitoring range: Ground sensors have limited monitoring range and cannot cover large areas; although satellite remote sensing data has a wide coverage, its temporal and spatial resolution is low.

[0006] Insufficient data processing capabilities: The existing system has limited data processing capabilities and cannot effectively process multi-source heterogeneous data, resulting in low accuracy and timeliness of monitoring results.

[0007] Single risk assessment model: Existing risk assessment models are often targeted at specific types of disasters, lack versatility and flexibility, and are difficult to adapt to diverse application scenarios.

[0008] Delayed release of early warning information: The release and dissemination mechanism of early warning information is not perfect, resulting in limited timeliness and coverage of early warning information. Summary of the invention

[0009] In order to overcome the shortcomings and deficiencies of the above-mentioned prior art, the present invention adopts the following technical solutions:

[0010] The regional risk monitoring and early warning system based on the fusion of multi-source data from space, air, ground and space includes:

[0011] Global deformation monitoring and risk identification module: used to obtain multi-view satellite images and digital elevation models of the area to be detected, and use InSAR technology to obtain surface deformation monitoring data; accurately align satellite images, generate interferometric image pairs, extract surface deformation information, and remove the influence of atmospheric effects; based on time-series InSAR technology and multi-source data, conduct deformation monitoring and terrain analysis, and identify potential risk areas in combination with risk assessment models in specific scenarios;

[0012] Risk inspection and 3D real-scene model construction module: Use drones equipped with laser radar and visible light cameras to regularly inspect key engineering areas and obtain high-density laser point cloud data and high-resolution images; process the data obtained by drones, build a pixel loss matrix, adjust pixels, and form a 3D real-scene model; study wide-area image control-cloud control technology, build a high-precision control framework, integrate drone high-precision photogrammetry modules, and build a 3D model of the engineering real scene;

[0013] Intelligent sensing technology and equipment module: Develop high-precision, multi-parameter, lightweight monitoring technology and equipment for key facilities such as roadbed slopes, tunnel construction, and bridge structures; use ground-based radar, machine vision and LiDAR combination technology, measurement robots, and laser scanners to achieve non-contact high-precision measurement;

[0014] Multi-source data fusion monitoring and early warning module: Integrate multi-source heterogeneous data, perform data cleaning and feature extraction, and develop intelligent fusion processing and analysis models; build an interactive monitoring and early warning platform that integrates user foundation, multi-source data protocol, real-scene model database and other functions; through deformation map generation, target detection and risk classification, data fusion processing, potential risk surface determination, stress-strain analysis, it realizes the release of safety warning information for high-risk areas.

[0015] Preferably, the workflow of the global deformation monitoring and risk identification module is as follows:

[0016] S11, obtaining multi-view satellite images and digital elevation models of the area to be detected;

[0017] S12. Precisely align multi-view satellite images to ensure data consistency and accuracy; generate interferometric image pairs to extract surface deformation information; remove atmospheric effects and improve interferometric phase quality;

[0018] S13. Based on the time-series InSAR technology, conduct large-scale deformation monitoring in the entire monitoring area; combine DEM, optical images and other multi-source data to extract slope information of the monitoring area, conduct morphological and surface cover type analysis, and identify risk points; according to different application scenarios, use corresponding risk assessment models to identify potential risk areas.

[0019] Preferably, the workflow of the risk inspection and 3D real scene model construction module is as follows:

[0020] S21. Use drones equipped with laser radar and visible light cameras to conduct regular inspections of key project areas and autonomously obtain high-density laser point cloud data and high-resolution images;

[0021] S22. Process the data obtained by the drone, construct a pixel loss matrix, perform pixel adjustment, and form a three-dimensional real-scene model; study wide-area image control-cloud control technology and build a high-precision control framework; integrate the drone's high-precision photogrammetry module, combine geological and other data, and build a high-precision three-dimensional model of the engineering real scene.

[0022] Preferably, the workflow of the intelligent sensing technology and equipment module specifically includes:

[0023] S31. Develop high-precision, multi-parameter, lightweight monitoring technology and equipment for key facilities such as roadbed slopes, tunnel construction, and bridge structures;

[0024] S32, using ground-based radar, machine vision and LiDAR combined technology, measuring robots and laser scanners to achieve non-contact high-precision measurement.

[0025] Preferably, the workflow of the multi-source data fusion monitoring and early warning module is as follows:

[0026] S41. Use the Internet of Things platform as the underlying support to integrate multi-source heterogeneous data, match the interface protocol types of various project equipment, and access air-space-ground multi-dimensional data; study abnormal data cleaning, data noise reduction and feature extraction methods to ensure high data quality; establish an intelligent fusion processing and analysis model for multi-source monitoring data, and develop a multi-source data fusion processing monitoring and early warning module;

[0027] S42. Based on the high-precision real-life 3D model base constructed by UAV aerial survey technology, the visual mapping method of multi-source monitoring data on the platform model base is studied to realize the intuitive display of monitoring data; platform architecture design and interactive product design are carried out to provide a configuration framework that connects the perception device layer, data acquisition layer, communication transmission layer and data analysis and mining layer; the platform covers user basic modules, multi-source data protocol library, real-life model database, BIM information library, geographic information library, data query library, early warning alarm model library and multi-source heterogeneous monitoring data fusion analysis model library content;

[0028] S43. Through the steps of deformation map generation, target detection and risk classification, data fusion processing, potential risk surface determination, and stress-strain analysis, safety warning information for high-risk areas can be released.

[0029] Preferably, removing the influence of atmospheric effects includes removing atmospheric turbulence effects and atmospheric vertical stratification effects.

[0030] Preferably, the specific process of constructing a three-dimensional real scene model includes: marking feature points and standard feature points, constructing a pixel loss function, obtaining a pixel loss matrix through least squares processing, performing pixel adjustment on the inclined image to form a regional map, and then converting the pixels in the regional map into a point cloud based on the elevation data to obtain a point cloud model, and mapping the color information to the point cloud model.

[0031] Preferably, the process of issuing safety warning information for high-risk areas includes: generating a deformation map, performing target detection and risk division, and dividing the deformation map into high-risk areas and low-risk areas; issuing safety warning information for high-risk areas; performing data fusion processing on low-risk areas to obtain a three-dimensional deformation model; determining potential risk surfaces based on stratigraphic structure information, geological structure information and terrain characteristics; performing stress-strain analysis to obtain a safety factor; and issuing safety warning information for areas where the safety factor is less than a safety factor threshold.

[0032] Preferably, the specific process of data fusion processing includes: aligning and density matching the deformation map and the three-dimensional model, performing data fusion processing based on the corresponding weights to obtain a three-dimensional deformation model. The specific operation is to convert the deformation matching map into point cloud data to obtain an initial deformation model, convert the initial deformation model and the three-dimensional matching model into a Bayesian function, obtain the likelihood function based on the corresponding weights and Gaussian distribution function, and obtain the three-dimensional deformation model through maximum a posteriori estimation processing.

[0033] Preferably, the safety factor acquisition process is as follows:

[0034] S11. Calculate stress safety factor D: Among them, M y is the allowable stress; M s is the actual stress;

[0035] S11. Calculate the strain safety factor V: Among them, N y is the allowable strain; N s is the actual strain;

[0036] S11. Calculate the final safety factor E: E=αD+βV; wherein D is the stress safety factor; V is the strain safety factor; α and β are the weights corresponding to the stress safety factor D and the strain safety factor, respectively.

[0037] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0038] 1. The present invention integrates multiple data sources such as satellite remote sensing, drone aerial survey, ground-based radar, LiDAR, machine vision, etc., realizes all-round and multi-level data collection from the sky to the ground, and improves the accuracy and comprehensiveness of risk monitoring; through intelligent fusion processing and analysis models, it effectively integrates multi-source heterogeneous data, solves data inconsistency and information island problems, and improves data processing efficiency and warning accuracy.

[0039] 2. The present invention uses a drone equipped with a laser radar and a visible light camera to construct a high-precision three-dimensional real-scene model, providing an intuitive and visual display method for risk monitoring; through the visual mapping method of multi-source monitoring data on the platform model base, the intuitive display and interactive analysis of monitoring data are realized.

[0040] 3. The present invention realizes real-time monitoring of multiple physical quantities of key facilities (such as roadbed slopes, tunnel construction, bridge structures, etc.) through high-precision displacement monitoring equipment based on micro-electromechanical systems (MEMS) technology and wireless composite sensing technology, thereby improving the accuracy and reliability of monitoring; it adopts non-contact measurement technology, such as ground-based radar, vision and LiDAR combined measurement, etc., to reduce manual intervention and improve measurement efficiency and safety.

[0041] 4. The present invention constructs an intelligent risk assessment model through deformation map generation, target detection and risk classification, data fusion processing, potential risk surface determination, stress and strain analysis, etc., and realizes the release of safety warning information for high-risk areas. The warning mechanism not only takes into account the surface deformation information, but also combines multiple factors such as building distribution, vegetation distribution, stratum structure, geological structure, etc., to improve the accuracy and pertinence of the warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 The module diagram of the regional risk monitoring and early warning system based on the fusion of multi-source data of air, space and ground of the present invention is shown;

[0044] Figure 2 The working flow chart of the system of the present invention is shown. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] In addition, the described features, structures or characteristics may be combined in one or more example embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the example embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. may be adopted. In other cases, well-known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0047] Embodiment 1:

[0048] See also Figure 1 As shown, the regional risk monitoring and early warning system based on the fusion of multi-source data of air, space, and ground in this embodiment includes a global deformation monitoring and risk identification module, a risk inspection and three-dimensional real-scene model construction module, an intelligent perception technology and equipment module, and a multi-source data fusion monitoring and early warning module.

[0049] Global deformation monitoring and risk identification module: used to obtain multi-view satellite images and digital elevation models of the area to be inspected, and use InSAR technology to obtain surface deformation monitoring data; accurately align satellite images, generate interferometric image pairs, extract surface deformation information, and remove the influence of atmospheric effects; based on time-series InSAR technology and multi-source data, perform deformation monitoring and terrain analysis, and identify potential risk areas in combination with risk assessment models in specific scenarios.

[0050] Risk inspection and three-dimensional real-scene model construction module: Use drones equipped with lidar and visible light cameras to regularly inspect key engineering areas and obtain high-density laser point cloud data and high-resolution images; process the data obtained by the drone, build a pixel loss matrix, perform pixel adjustments, and form a three-dimensional real-scene model; study wide-area image control-cloud control technology, build a high-precision control framework, integrate drone high-precision photogrammetry modules, and build a three-dimensional model of the engineering real scene.

[0051] Intelligent sensing technology and equipment module: Develop high-precision, multi-parameter, lightweight monitoring technology and equipment for key facilities such as roadbed slopes, tunnel construction, and bridge structures; use ground-based radar, machine vision and LiDAR combination technology, measurement robots, and laser scanners to achieve non-contact high-precision measurement.

[0052] Multi-source data fusion monitoring and early warning module: Integrate multi-source heterogeneous data, perform data cleaning and feature extraction, and develop intelligent fusion processing and analysis models; build an interactive monitoring and early warning platform that integrates user foundation, multi-source data protocol, and real-scene model database;

[0053] Through the steps of deformation map generation, target detection and risk classification, data fusion processing, potential risk surface determination, stress-strain analysis, etc., safety warning information for high-risk areas can be issued.

[0054] See also Figure 2 As shown in the figure, the regional risk monitoring and early warning method based on the fusion of multi-source data of air, space and ground has the following workflow:

[0055] Step 1: Global deformation monitoring and risk identification based on long-term satellite remote sensing.

[0056] S11. Obtain multi-view satellite images and a digital elevation model (DEM) of the area to be detected.

[0057] The synthetic aperture radar interferometry (InSAR) technology is used to obtain large-scale surface deformation monitoring data and detect millimeter-level surface deformation.

[0058] S12. Data processing.

[0059] Precisely align multi-view satellite images to ensure data consistency and accuracy.

[0060] Generate interferometric image pairs and extract surface deformation information.

[0061] Remove the atmospheric turbulence effect and the atmospheric vertical stratification effect to improve the interferometric phase quality.

[0062] S13. Risk identification.

[0063] Based on the time-series InSAR technology, large-scale deformation monitoring is carried out in the entire monitoring area.

[0064] By combining multi-source data such as DEM and optical images, the slope information of the monitoring area is extracted, and the morphology, surface cover type and other analyses are conducted to identify risk points.

[0065] According to different application scenarios (such as earthquakes, floods, landslides, etc.), corresponding risk assessment models are used to identify potential risk areas.

[0066] Step 2: Risk inspection and three-dimensional real-scene model construction based on drone aerial survey.

[0067] S21. Data collection.

[0068] Use drones equipped with lidar and visible light cameras to conduct regular inspections of key project areas (such as high slopes, tunnel entrances, bridges, etc.).

[0069] Autonomously acquire high-density laser point cloud data and high-resolution images.

[0070] S22. Data processing.

[0071] The pixels in the current oblique image whose pixel values ​​are extreme values ​​after Gaussian blur processing at different scales and whose neighborhood contrast is less than the contrast threshold are marked as feature points.

[0072] The pixels at the corresponding positions of these feature points in the standard image are marked as standard feature points.

[0073] Based on the pixel values ​​of feature points and standard feature points, a pixel loss function is constructed, and the pixel loss matrix is ​​obtained by least squares processing.

[0074] Based on the pixel loss matrix, the current tilted image is pixel-adjusted to obtain a current adjusted image.

[0075] Each adjusted image is sequentially connected with the adjusted images corresponding to adjacent acquisition times by mixing pixels in the same area to obtain a region map.

[0076] Based on the elevation data, the pixels in the area map are converted into point clouds to obtain a point cloud model.

[0077] The color information in the area map is mapped to the point cloud model to obtain a three-dimensional model.

[0078] Research wide-area image control-cloud control technology and build a high-precision control framework.

[0079] Integrate the high-precision photogrammetry module of drones and combine it with geological and other data to build a high-precision 3D model of the real engineering scene.

[0080] Step 3: Low-cost, high-precision, multi-parameter, lightweight intelligent sensing technology and equipment.

[0081] S31. Monitoring of key facilities.

[0082] Roadbed slope monitoring: Develop high-precision displacement ropes and ultra-large-range magnetostrictive displacement meters based on micro-electromechanical systems (MEMS) technology to achieve high-precision perception of long-period slope matrix surface settlement and deep horizontal displacement of soil.

[0083] Tunnel construction monitoring: Develop tunnel vault settlement monitoring technology and displacement rope pre-embedding method to achieve full-process monitoring of tunnel construction.

[0084] Bridge structure monitoring: Develop wireless composite sensing technology to achieve real-time monitoring of multiple physical quantities such as bridge deflection, displacement, settlement, tilt and temperature.

[0085] Monitoring of other facilities: According to different application scenarios (such as dams, reservoirs, mines, etc.), develop corresponding high-precision, multi-parameter, lightweight intelligent sensing technologies and equipment.

[0086] S32, non-contact measurement.

[0087] Ground-based radar measurement: Use ground-based radar to perform non-contact measurement of static and dynamic deformation of a wide range of targets.

[0088] Combined measurement of vision and LiDAR: Using machine vision and 3D laser scanning technology to achieve high-precision non-contact measurement of tunnel openings, bridge structures, etc.

[0089] Measuring robots and laser scanners: 3D deformation measurement using measuring robots and high-precision laser scanners.

[0090] Step 4: Design and development of a monitoring and early warning platform that integrates air, space and ground multi-source data.

[0091] S41. Data fusion.

[0092] Multi-source data integration: Using the Internet of Things platform as the underlying support, integrate multi-source heterogeneous data, match the interface protocol types of various project equipment, and access air-space-ground multi-dimensional data.

[0093] Data cleaning and feature extraction: Research on abnormal data cleaning, data noise reduction and feature extraction methods to ensure high data quality.

[0094] Intelligent fusion processing: Establish intelligent fusion processing and analysis models for multi-source monitoring data, and develop multi-source data fusion processing monitoring and early warning modules.

[0095] S42. Platform design and development.

[0096] Visual mapping technology: Based on the high-precision real-life three-dimensional model base constructed using drone aerial survey technology, we study the visual mapping method of multi-source monitoring data on the platform model base to achieve intuitive display of monitoring data.

[0097] Interactive product design: Combine the research content of the subject to carry out platform architecture design and interactive product design, and provide a configuration framework that connects the perception device layer, data collection layer, communication transmission layer and data analysis and mining layer.

[0098] Functional modules: The platform covers user basic modules, multi-source data protocol library, real-scene model database, BIM information library, geographic information library, data query library, early warning and alarm model library, and multi-source heterogeneous monitoring data fusion analysis model library.

[0099] S43. Risk assessment and early warning.

[0100] S431, generating a deformation map.

[0101] The radar data of the area to be inspected at the current and historical neighborhood moments are obtained respectively, and after image registration processing, the phase difference of the corresponding position points is obtained to obtain a phase difference map.

[0102] The phase difference map is converted into a surface deformation map after least squares processing to obtain a deformation map.

[0103] S432. Target detection and risk classification.

[0104] The deformation map is processed by target detection to obtain building distribution information and vegetation distribution information.

[0105] Based on building distribution information, vegetation distribution information and surface deformation threshold, the deformation map is divided into high-risk areas and low-risk areas.

[0106] High-risk areas are: areas in the deformation map where the average surface deformation is greater than the first surface deformation threshold and the surface deformation variance is less than the variance threshold; or areas in the deformation map where the average surface deformation is not greater than the first surface deformation threshold but greater than the second surface deformation threshold and the building distribution area is greater than the building distribution area threshold. The first surface deformation threshold is greater than the second surface deformation threshold.

[0107] S433. Early warning information release.

[0108] Issue safety warning information for high-risk areas.

[0109] Perform the next step for low-risk areas.

[0110] S434: Data fusion processing.

[0111] After the deformation map and the 3D model are registered and density matched, data fusion processing is performed based on the corresponding weights to obtain a 3D deformation model.

[0112] The operation of data fusion processing is specifically as follows: based on the coordinate data and elevation data of the area to be inspected, the deformation matching map is converted into point cloud data to obtain an initial deformation model.

[0113] The initial deformation model and the three-dimensional matching model are respectively converted into Bayesian functions to obtain an initial deformation Bayesian model and a three-dimensional Bayesian model.

[0114] Based on the corresponding weights and the Gaussian distribution function, the likelihood functions of the initial deformation Bayesian model and the three-dimensional Bayesian model are obtained to obtain the initial deformation likelihood model and the three-dimensional likelihood model.

[0115] The initial deformation likelihood model and the three-dimensional likelihood model are processed by maximum a posteriori estimation to obtain a three-dimensional deformation model.

[0116] S435. Identify potential risks.

[0117] Based on the stratigraphic structure information, geological structure information, and terrain characteristics (such as slope direction, slope gradient, slope angle, etc.) of the area to be inspected, the potential risk surface of the three-dimensional deformation model is determined.

[0118] S436. Stress-strain analysis.

[0119] Based on the environmental parameters of the future neighborhood time (such as rainfall duration and rainfall, seismic activity, etc.) and the geological information corresponding to the potential risk surface, stress-strain analysis is performed to obtain the safety factor.

[0120] The process of obtaining the safety factor is as follows:

[0121] S11. Calculate stress safety factor D: Among them, M y is the allowable stress; M s is the actual stress;

[0122] S11. Calculate the strain safety factor V: Among them, N y is the allowable strain; N s is the actual strain;

[0123] S11. Calculate the final safety factor E: E=αD+βV; wherein D is the stress safety factor; V is the strain safety factor; α and β are the weights corresponding to the stress safety factor D and the strain safety factor, respectively.

[0124] S437. Early warning information release.

[0125] Safety warning information will be issued for areas where the safety factor is less than the safety factor threshold.

[0126] The beneficial effects of this embodiment are: by integrating multiple data sources such as satellite remote sensing, drone aerial survey, and ground sensors, comprehensive monitoring and timely warning of regional risks are achieved. The key technologies of the system include long-term satellite remote sensing, drone aerial survey, low-cost and high-precision intelligent perception technology, and multi-source data fusion platform. The system is suitable for a variety of application scenarios such as natural disasters (such as earthquakes, floods, landslides, etc.), engineering construction (such as dams, tunnels, bridges, etc.), and environmental protection (such as mines, reservoirs, etc.), providing strong support for regional security and sustainable development.

[0127] The weights of the present invention are used to measure the degree of influence of different factors or variables on a certain result or decision. The definition of weight refers to the value assigned to each factor when comparing and evaluating multiple factors to reflect its importance or priority. These weights can be determined according to specific circumstances and needs, and are usually jointly formulated and confirmed by professionals or relevant stakeholders. By setting weights reasonably, it can help programs or systems make decisions or predictions more accurately.

[0128] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0129] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. Regional risk monitoring and early warning system based on multi-source data fusion of air, space and ground, characterized by: The system comprises: Global deformation monitoring and risk identification module: used to obtain multi-view satellite images and digital elevation models of the area to be detected, and use InSAR technology to obtain surface deformation monitoring data; accurately align satellite images, generate interferometric image pairs, extract surface deformation information, and remove the influence of atmospheric effects; based on time-series InSAR technology and multi-source data, conduct deformation monitoring and terrain analysis, and identify potential risk areas in combination with risk assessment models in specific scenarios; Risk inspection and 3D real-scene model construction module: Use drones equipped with laser radar and visible light cameras to regularly inspect key engineering areas and obtain high-density laser point cloud data and high-resolution images; process the data obtained by drones, build a pixel loss matrix, adjust pixels, and form a 3D real-scene model; study wide-area image control-cloud control technology, build a high-precision control framework, integrate drone high-precision photogrammetry modules, and build a 3D model of the engineering real scene; Intelligent sensing technology and equipment module: For key facilities such as roadbed slopes, tunnel construction, and bridge structures, non-contact high-precision measurement is achieved using ground-based radar, machine vision and LiDAR combined technology, measurement robots, and laser scanners; Multi-source data fusion monitoring and early warning module: Integrate multi-source heterogeneous data, perform data cleaning and feature extraction, and develop intelligent fusion processing and analysis models; build an interactive monitoring and early warning platform that integrates user foundation, multi-source data protocol, and real-scene model database functions; through deformation map generation, target detection and risk classification, data fusion processing, potential risk surface determination, and stress-strain analysis, it realizes the release of safety warning information for high-risk areas.

2. The regional risk monitoring and early warning system based on air-ground-ground multi-source data fusion according to claim 1 is characterized in that: The workflow of the global deformation monitoring and risk identification module is as follows: S11, obtaining multi-view satellite images and digital elevation models of the area to be detected; S12. Perform precise registration of multi-view satellite images to ensure data consistency and accuracy; Generate interferometric image pairs to extract surface deformation information; remove atmospheric effects and improve interferometric phase quality; S13. Based on the time-series InSAR technology, conduct large-scale deformation monitoring in the entire monitoring area; combine DEM, optical images and other multi-source data to extract slope information of the monitoring area, conduct morphological and surface cover type analysis, and identify risk points; according to different application scenarios, use corresponding risk assessment models to identify potential risk areas.

3. The regional risk monitoring and early warning system based on air-ground-ground multi-source data fusion according to claim 1 is characterized in that: The workflow of the risk inspection and 3D real scene model construction module is as follows: S21. Use drones equipped with laser radar and visible light cameras to conduct regular inspections of key project areas and autonomously obtain high-density laser point cloud data and high-resolution images; S22. Process the data obtained by the drone, construct a pixel loss matrix, perform pixel adjustment, and form a three-dimensional real-scene model; study wide-area image control-cloud control technology and build a high-precision control framework; integrate the drone's high-precision photogrammetry module, combine geological and other data, and build a high-precision three-dimensional model of the engineering real scene.

4. The regional risk monitoring and early warning system based on air-ground multi-source data fusion according to claim 1 is characterized in that: The workflow of the intelligent sensing technology and equipment module specifically includes: S31. Develop high-precision, multi-parameter, lightweight monitoring technology and equipment for key facilities such as roadbed slopes, tunnel construction, and bridge structures; S32, using ground-based radar, machine vision and LiDAR combined technology, measuring robots and laser scanners to achieve non-contact high-precision measurement.

5. The regional risk monitoring and early warning system based on air-ground multi-source data fusion according to claim 1 is characterized in that: The working process of the multi-source data fusion monitoring and early warning module is as follows: S41. Use the Internet of Things platform as the underlying support to integrate multi-source heterogeneous data, match the interface protocol types of various project equipment, and access air-space-ground multi-dimensional data; study abnormal data cleaning, data noise reduction and feature extraction methods to ensure high data quality; establish an intelligent fusion processing and analysis model for multi-source monitoring data, and develop a multi-source data fusion processing monitoring and early warning module; S42. Based on the high-precision real-life 3D model base constructed by UAV aerial survey technology, the visual mapping method of multi-source monitoring data on the platform model base is studied to realize the intuitive display of monitoring data; platform architecture design and interactive product design are carried out to provide a configuration framework that connects the perception device layer, data acquisition layer, communication transmission layer and data analysis and mining layer; the platform covers user basic modules, multi-source data protocol library, real-life model database, BIM information library, geographic information library, data query library, early warning alarm model library and multi-source heterogeneous monitoring data fusion analysis model library content; S43. Through the steps of deformation map generation, target detection and risk classification, data fusion processing, potential risk surface determination, and stress-strain analysis, safety warning information for high-risk areas can be released.

6. The regional risk monitoring and early warning system based on multi-source data fusion of air, space and ground according to claim 1 is characterized in that: The removal of atmospheric effects includes removing atmospheric turbulence effects and atmospheric vertical stratification effects.

7. The regional risk monitoring and early warning system based on air-ground-ground multi-source data fusion according to claim 1 is characterized in that: The specific process of constructing the three-dimensional real scene model includes: marking feature points and standard feature points, constructing a pixel loss function, obtaining a pixel loss matrix through least squares processing, performing pixel adjustment on the inclined image to form a regional map, and then converting the pixel points in the regional map into a point cloud based on elevation data to obtain a point cloud model, and mapping color information to the point cloud model.

8. The regional risk monitoring and early warning system based on air-ground-ground multi-source data fusion according to claim 1 is characterized in that: The process of issuing safety warning information for high-risk areas includes: generating a deformation map, performing target detection and risk division, and dividing the deformation map into high-risk areas and low-risk areas; issuing safety warning information for high-risk areas; performing data fusion processing on low-risk areas to obtain a three-dimensional deformation model; determining potential risk surfaces based on stratigraphic structure information, geological structure information and terrain characteristics; performing stress-strain analysis to obtain a safety factor; and issuing safety warning information for areas where the safety factor is less than a safety factor threshold.

9. The regional risk monitoring and early warning system based on air-ground-ground multi-source data fusion according to claim 1 is characterized in that: The specific process of the data fusion processing includes: aligning and density matching the deformation map and the three-dimensional model, performing data fusion processing based on corresponding weights to obtain a three-dimensional deformation model. The specific operation is to convert the deformation matching map into point cloud data to obtain an initial deformation model, convert the initial deformation model and the three-dimensional matching model into a Bayesian function, obtain the likelihood function based on the corresponding weights and Gaussian distribution function, and obtain the three-dimensional deformation model through maximum a posteriori estimation processing.

10. The regional risk monitoring and early warning system based on air-ground multi-source data fusion according to claim 8 is characterized in that: The safety factor acquisition process is as follows: S11. Calculate stress safety factor D: Among them, M y is the allowable stress; M s is the actual stress; S11. Calculate the strain safety factor V: Among them, N y is the allowable strain; N s is the actual strain; S11. Calculate the final safety factor E: E=αD+βV; wherein D is the stress safety factor; V is the strain safety factor; α and β are the weights corresponding to the stress safety factor D and the strain safety factor, respectively.

Images