Geological disaster early warning analysis method based on multi-source remote sensing data

Abstract

The invention discloses a geological disaster early warning analysis method based on multi-source remote sensing data, and relates to the technical field of geological disaster monitoring, and the method comprises the following steps: S1, carrying out the unified expression construction through employing the multi-source remote sensing data; s2, performing sensitive element extraction by using a geological environment unified framework; s3, performing coupling evolution analysis by using the potential risk factors; s4, performing space-time mapping by using the critical condition set; and S5, performing consistency check by using the geological disaster risk distribution map in combination with historical disaster occurrence records. By setting sensitive element extraction and coupling evolution analysis, the method can fully excavate feature information closely related to geological disaster development in multi-source remote sensing data, differentiated extraction and joint modeling are performed on the elements, the problem of information loss caused by traditional single-index monitoring is effectively avoided, and the accuracy of geological disaster development is improved. And dynamically fusing the short-term trigger factor and the long-term cumulative effect to form a continuous description of the disaster breeding process.

Description

Technical Field

[0001] This invention relates to the field of geological disaster monitoring technology, specifically a geological disaster early warning analysis method based on multi-source remote sensing data. Background Technology

[0002] Traditional methods for monitoring and early warning of geological disasters mainly rely on a single remote sensing data source or ground-based monitoring, which suffers from insufficient spatial coverage, limited monitoring accuracy, and poor data real-time performance, making it difficult to fully reflect the process of disaster formation and evolution.

[0003] In existing technologies, remote sensing early warning methods for geological disasters often remain at the level of single-source data application, making it difficult to achieve effective fusion of cross-source data, resulting in one-sided monitoring results. In addition, existing time series analysis methods are mostly based on static indicators and lack dynamic characterization of the disaster's gestation and evolution process, thus causing problems such as delayed early warning and insufficient accuracy. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a geological disaster early warning analysis method based on multi-source remote sensing data, in order to solve the problems mentioned in the background section.

[0005] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, embodiments of the present invention provide a geological disaster early warning analysis method based on multi-source remote sensing data, comprising the following steps: S1. Using multi-source remote sensing data to construct a unified representation, a unified framework for the geological environment is obtained; S2. Sensitive elements are extracted using a unified geological environment framework to obtain potential risk factors; S3. Use potential risk factors to perform coupled evolution analysis to obtain the set of critical conditions for disaster occurrence; S4. Use the set of critical conditions for spatiotemporal mapping to obtain a geological hazard risk distribution map; S5. Use the geological hazard risk distribution map in conjunction with historical disaster occurrence records to conduct a consistency check and obtain the geological hazard risk level results.

[0006] To further optimize this technical solution, step S2 involves characterizing data from different dimensions within a unified geological environment framework to extract sensitive elements that have a direct impact on the occurrence of geological disasters. These sensitive elements refer to environmental feature quantities identified from multi-source remote sensing data within a unified geological environment framework that have a significant response or indication effect on the incubation, development, and triggering process of geological disasters. Step S2, in the process of obtaining potential risk factors, includes the following three steps: sensitive element screening, element normalization and weight allocation, and potential risk factor generation.

[0007] To further optimize this technical solution, in step S2, when screening sensitive elements, sensitive element screening is performed based on the multi-source fusion data set output in step S1 within a unified framework: Correlation coefficient analysis was used to calculate the value of each data element. Correlation with historical geological disaster events; Statistical thresholding is used to filter out a subset of events whose correlation with historical geological disaster records exceeds a preset threshold: ; in This indicates the sensitive elements after filtering.

[0008] To further optimize this technical solution, in step S2, when performing element normalization and weight allocation, for each sensitive element... After normalization, we get: ; in This represents the normalized value of the sensitive element; Weight The determination of the risk factors adopts mature entropy weight method or analytic hierarchy process to ensure that different elements contribute reasonably in the calculation of potential risk factors. The weighted sensitive element set is denoted as .

[0009] To further optimize this technical solution, in step S2, when generating potential risk factors, the weighted sensitive elements are combined and mapped to generate a set of potential risk factors, the formula model of which is: ; in Indicates a unified framework The next generated One potential risk factor; By employing nonlinear mapping, through logistic regression or support vector machines, weighted elements are mapped to risk factor values, ensuring logical compatibility with the results of linear combinations. Step S2 ultimately outputs a set of potential risk factors. , .

[0010] To further optimize this technical solution, step S3 analyzes the coupling effect and dynamic evolution law between potential risk factors, identifies key combination conditions that may trigger geological disasters, and formalizes these conditions into a set of critical conditions. Step S3, in the process of obtaining the set of critical conditions for the occurrence of a disaster, includes the following four steps: calculation of coupling function, construction of evolution matrix, determination of critical condition threshold, and generation of critical condition set.

[0011] To further optimize this technical solution, in step S3, when calculating the coupling function, covariance analysis is used to calculate linear coupling, or logistic regression is used to capture nonlinear interactions, for potential risk factors. and Establish coupling relationships between them: ; in Representation factor and Coupling strength under geological disaster triggering conditions; With coupling strength Using the matrix elements, we obtain a symmetric coupling matrix. .

[0012] To further optimize this technical solution, step S3, when constructing the evolution matrix, includes the following time / space partitioning: The time series data uses historical monitoring intervals; Spatial units are divided into grids to ensure that the matrix corresponds to the actual environment; Arrange the coupling matrix according to time series or spatial units to form the evolution matrix: .

[0013] To further optimize this technical solution, in step S3, when determining the critical condition threshold, a threshold vector is set. Corresponding to each potential risk factor, a coupling threshold is then set. Corresponding factor pairs; In the process of determining the threshold, for individual risk factors The long-term monitoring data were sorted, and their statistical quantiles were used as the critical threshold. .

[0014] To further optimize this technical solution, in step S3, when generating the critical condition set, for each potential risk factor... and its coupling value determination: if and For all If established, then Inclusion of critical conditions middle; Repeat the above process to obtain the complete set of critical conditions, whose expression is: ; Each This indicates the environmental combination state that may trigger a disaster when potential risk factors and their coupling effects reach a threshold.

[0015] In a second aspect, embodiments of the present invention provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and the computer program instructions, when executed by the processor, implement the steps of a geological disaster early warning analysis method based on multi-source remote sensing data as described in the first aspect of the present invention.

[0016] Thirdly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, they implement the steps of a geological disaster early warning analysis method based on multi-source remote sensing data as described in the first aspect of the present invention.

[0017] Compared with existing technologies, this invention provides a geological disaster early warning and analysis method based on multi-source remote sensing data, which has the following beneficial effects: This geological disaster early warning analysis method based on multi-source remote sensing data, through the extraction of sensitive elements and coupled evolutionary analysis, can fully mine characteristic information closely related to the formation of geological disasters from multi-source remote sensing data, such as sensitive elements like surface deformation, vegetation cover changes, and fluctuations in hydrological conditions. It then performs differentiated extraction and joint modeling of these elements, effectively avoiding the information gaps caused by traditional single-indicator monitoring. Simultaneously, based on the coupled evolutionary analysis mechanism, it dynamically integrates short-term triggering factors with long-term cumulative effects, forming a continuous characterization of the disaster formation process, thereby improving the accuracy and stability of early warning analysis. This method can not only identify potentially high-risk areas in the early stages but also achieve real-time updates and risk trend prediction during the disaster evolution process, greatly enhancing the foresight and reliability of the early warning system. Compared to the limitations of existing technologies that rely on single-source data and static indicators, this method significantly improves monitoring accuracy and early warning timeliness, providing strong technical support for geological disaster risk management and emergency decision-making. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating a geological disaster early warning analysis method based on multi-source remote sensing data proposed in this invention. Figure 2 This is a schematic diagram of the sensitive element extraction process for a geological disaster early warning analysis method based on multi-source remote sensing data proposed in this invention; Figure 3This is a schematic diagram of the coupled evolution analysis process of a geological disaster early warning analysis method based on multi-source remote sensing data proposed in this invention; Figure 4 This is a schematic diagram of the consistency verification process for a geological disaster early warning analysis method based on multi-source remote sensing data proposed in this invention. Detailed Implementation

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0021] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0023] Example 1: Reference Figures 1-4 This is the first embodiment of the present invention, which provides a geological disaster early warning analysis method based on multi-source remote sensing data, including the following steps: S1. Using multi-source remote sensing data to construct a unified representation, a unified framework for the geological environment is obtained; Step S1, as the initial step in the method, is used to unify and represent data from different remote sensing platforms so that coupling analysis and parameter extraction can be performed in subsequent steps. It includes the following four steps: Data source confirmed: Multi-source remote sensing data includes optical remote sensing images (to acquire surface texture and vegetation information), synthetic aperture radar (SAR, used to monitor surface deformation), lidar (LiDAR, used to build high-precision three-dimensional terrain models), and multispectral or hyperspectral images (used to reflect surface material composition and humidity characteristics).

[0024] The unification of space and time: Because different remote sensing data have different spatial and temporal resolutions, they need to be standardized: By using mature multi-source remote sensing image registration technology, data acquired by different sensors are spatially aligned to make them comparable under the same geographic coordinate system. By using time series resampling methods, the time scale of data collected at different time points is unified, ensuring that various types of data are synchronized in subsequent analyses.

[0025] Data feature transformation: To address the differences in physical quantities among different remote sensing data, it is necessary to transform them into comparable and uniform indicators: Through mature normalization processing technology, the gray values ​​in optical images are converted into reflectance indicators. The surface displacement information obtained by interferometric SAR technology is converted into deformation rate parameters. By using digital elevation model (DEM) construction technology, LiDAR point clouds are transformed into continuous terrain slope and elevation information.

[0026] The formation of a unified framework: The registered, resampled, and feature-transformed data is uniformly stored and represented using a geospatial grid, resulting in a unified framework for the geological environment. express; The framework includes basic parameters such as topography, deformation, vegetation cover, and soil moisture. All parameters are organized within the same coordinate system and time scale, forming a multi-source fused dataset, represented by the following expression: ; Each element This corresponds to a pre-processed and standardized remote sensing data type, such as reflectance layers of optical images, SAR deformation rate layers, slope or elevation information constructed from LiDAR, and surface moisture or vegetation cover indicators derived from multispectral or hyperspectral images. The collection is organized under a unified geographic coordinate system and time scale to ensure spatial and temporal consistency of different data sources, so that the screening of sensitive elements and the generation of potential risk factors in subsequent steps can be carried out on the same basis. Multi-source fusion dataset It is to construct a unified framework for the geological environment. The original data foundation, and the unified framework of geological environment Yes A unified representation of space, time, and features.

[0027] S2. Sensitive elements are extracted using a unified geological environment framework to obtain potential risk factors; Step S2 serves as a framework for understanding the geological environment in this method. The data from different dimensions (topography, geomorphology, hydrology, geological structure, vegetation cover, etc.) are characterized to extract sensitive elements that have a direct impact on the occurrence of geological disasters. These sensitive elements refer to environmental feature quantities identified from multi-source remote sensing data within a unified geological environment framework that have a significant response or indication effect on the incubation, development, and triggering process of geological disasters. These elements are analyzable quantities formed by characterizing and extracting information such as topography, geomorphology, hydrology, geological structure, and surface cover status. When the values ​​or states of these analyzable quantities change, they are accompanied by significant changes in the probability or degree of danger of geological disasters.

[0028] By normalizing, weighting, and combining these sensitive elements, potential risk factors that can be directly used for risk assessment are generated. .

[0029] Specifically, step S2 includes the following three steps: Sensitive element screening: Under a unified framework The following is based on the multi-source fusion data set output in step S1. Sensitive elements are screened, and mature technical means are used to ensure feasibility, including: Correlation coefficient analysis was used to calculate the value of each data element. Correlation with historical geological disaster events; Using statistical thresholds (such as significance level) Select a subset whose correlation with historical geological disaster records exceeds a preset threshold: ; in This indicates the sensitive elements after filtering.

[0030] Element normalization and weight allocation: For each sensitive element After normalization, we get: ; in This represents the normalized value of the sensitive element; Weight The determination of the risk factors adopts mature entropy weight method or analytic hierarchy process (AHP) to ensure that different elements contribute reasonably in the calculation of potential risk factors; The weighted sensitive element set is denoted as .

[0031] Generation of potential risk factors: The set of potential risk factors is generated by combining and mapping weighted sensitive elements. The formula model is as follows: ; in Indicates a unified framework The next generated One potential risk factor; If a nonlinear mapping is used, it can be achieved through mature models, such as logistic regression or support vector machine (SVM), to map the weighted elements into risk factor values, ensuring that they are logically compatible with the results of linear combinations.

[0032] Step S2 ultimately outputs a set of potential risk factors. , .

[0033] Existing technologies typically employ a single data source or regionalized empirical models for sensitive factor screening, lacking a unified framework to support multi-source geological environmental data. Step S2 utilizes the unified framework constructed in S1, significantly improving the systematicness and consistency of sensitive element extraction and potential risk factor generation. It solves the problems of scattered element selection and arbitrary weight determination in traditional methods, while providing a quantifiable and reproducible calculation method.

[0034] S3. Use potential risk factors to perform coupled evolution analysis to obtain the set of critical conditions for disaster occurrence; Step S3 in this method takes the risk factor set output from step S2 and uses it to analyze the coupling effect and dynamic evolution law between potential risk factors, identify the key combination conditions that may trigger geological disasters, and formalize these conditions into a set of critical conditions.

[0035] Specifically, step S3 includes the following steps: Coupled function calculation: Covariance analysis can be used to calculate linear coupling, or logistic regression functions can be used to capture nonlinear interactions, for potential risk factors. and Establish coupling relationships between them: ; in Representation factor and Coupling strength under geological disaster triggering conditions; With coupling strength Using the matrix elements, we obtain a symmetric coupling matrix. .

[0036] Evolutionary matrix construction: The time / space partitioning during matrix construction includes: The time series can use historical monitoring intervals, such as monthly or weekly records; Spatial units can be divided into grids, for example, each grid covers a certain geographical area (e.g., 1 km²), ensuring that the matrix corresponds to the actual environment; Arrange the coupling matrix according to time series or spatial units to form the evolution matrix: .

[0037] Critical condition threshold determination: Set threshold vector Corresponding to each potential risk factor, a coupling threshold is then set. Corresponding factor pairs; In the process of determining the threshold, for individual risk factors The long-term monitoring data are sorted, and their statistical quantiles (such as the 90th quantile) are used as the critical threshold. This method can reflect the potential for triggering disasters when factors reach extreme states.

[0038] Critical condition set generation: For each potential risk factor and its coupling value determination: if and For all If established, then Inclusion of critical conditions middle; Repeat the above process to obtain the complete set of critical conditions, whose expression is: ; Each This indicates the environmental combination state that may trigger a disaster when potential risk factors and their coupling effects reach a threshold.

[0039] Existing technologies mostly rely on single-factor threshold determination or a small number of factor combinations, lacking multi-factor coupling and dynamic evolution analysis. This step quantifies factor interactions through potential risk factor coupling functions and evolution matrices, and generates a set of multi-factor critical conditions, enabling a systematic and quantifiable analysis of disaster triggering conditions. Compared with traditional methods, this improves the comprehensiveness and reproducibility of the analysis.

[0040] S4. Use the set of critical conditions for spatiotemporal mapping to obtain a geological hazard risk distribution map; Step S4 in this method is used to combine the set of critical conditions with geospatial information and map them according to spatial units and time series, so as to intuitively display the distribution of geological disaster risks in different regions and time periods.

[0041] Specifically, step S4 includes the following steps: Spatial unit division: Based on the unified geological environment framework constructed in step S1, the study area is decomposed into several spatial units (such as regular grids or partitions based on terrain features) using common geographic information system (GIS) grid division techniques. Each spatial unit serves as the basic computational unit for mapping, ensuring that the results correspond to the actual geographical location.

[0042] Condition determination and risk level assignment: The set of critical conditions obtained in step S3 Distribution of potential risk factors applied to each spatial unit; By employing mature spatial overlay analysis methods, the risk factors of each spatial unit are compared with... If a certain critical condition is met, the spatial unit is determined to be a risk unit. Within a risk unit, different risk levels are assigned based on the number and severity of the conditions met, using existing mature grading methods (such as threshold grading).

[0043] Time series mapping: The time dimension of the evolution matrix in step S3 is introduced into the spatial determination process. Mature time series analysis technology is used to superimpose the risk levels of different time points onto the spatial units. The result is a disaster risk distribution dataset that evolves dynamically over time and is visualized spatially.

[0044] Visualization and Output: Using a mature GIS visualization module, the risk level results are transformed into a geological disaster risk distribution map; The output, presented as a layer, shows the risk level distribution across different times and spaces. This data serves as input for the next step, and its expression is: This formula is a function that varies with spatial position. and time A set of changing, multidimensional results is used to represent the distribution of disaster risk levels in different regions and time periods; Furthermore, step S4, based on existing disaster event archives, monitoring reports, or remote sensing interpretation results, derives historical disaster occurrence records according to spatial location and time nodes, the expression of which is: In addition, historical disaster records Geological disaster risk distribution map They are unified under the same coordinate system and time scale.

[0045] S5. Use the geological hazard risk distribution map in conjunction with historical disaster occurrence records to conduct a consistency check and obtain the geological hazard risk level results; Step S5 in this method is used to process the geological hazard risk distribution map obtained in step S4. Records of historical disasters Perform a consistency check.

[0046] Specifically, step S5 includes the following process: Consistency measure: Set a consistency check function, with the function expression as follows: ; The global consistency index is calculated using the following formula. : ; in, This indicates the number of all spatial units and time nodes; This indicator is used to measure the degree of alignment between risk distribution and historical disaster records as a whole.

[0047] Regional consistency comparison: Based on consistency function According to the statistical results, when a certain area At multiple time points superior, and If the matching rate is lower than a preset threshold (e.g., less than 60%), the region is considered to have poor consistency. Based on global consistency index If the local consistency value of a certain region Significantly below the overall average level (For example, than) If the percentage is more than 20% lower, then the region is also defined as a region with poor consistency. Risk level revision: If there is poor consistency in certain regions, the risk level will be revised by combining the intensity and frequency of historical disaster events. The revision methods include adjusting the weighting factors or downgrading / upgrading the level to make the results more in line with the actual situation. The final risk level result is recorded as follows: Its presentation can be in the form of a raster level map or a regional level matrix, which can intuitively reflect the risk level distribution of different regions.

[0048] The final result obtained in step S5 is the geological hazard risk level result, which is a spatially distributed hierarchical result, including the level interval, the regional range and the corresponding hazard risk level.

[0049] Most existing mature technologies rely on single indicators or statistical comparisons, making it difficult to achieve comprehensive verification across spatiotemporal scales. Step S5, however, compares the risk distribution results generated by the model with real historical disaster data within a unified spatial and temporal framework, and introduces consistency measurement and correction mechanisms to make the risk level results more consistent with the actual disaster distribution patterns, thus possessing higher accuracy and practicality.

[0050] Example 2: This embodiment also provides a computer device applicable to a geological disaster early warning analysis method based on multi-source remote sensing data, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the geological disaster early warning analysis method based on multi-source remote sensing data proposed in the above embodiment.

[0051] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements a geological disaster early warning analysis method based on multi-source remote sensing data as proposed in the above embodiment.

[0052] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0053] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0054] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0055] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0056] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0057] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1.A geological disaster early warning analysis method based on multi-source remote sensing data, characterized in that, The method comprises the following steps: S1, unified expression construction using multi-source remote sensing data to obtain a unified geological environment framework; S2, sensitive element extraction using the unified geological environment framework to obtain potential risk factors; S3, coupling evolution analysis using the potential risk factors to obtain a critical condition set of disaster occurrence; S4, spatio-temporal mapping using the critical condition set to obtain a geological disaster risk distribution map; S5, consistency test using the geological disaster risk distribution map combined with historical disaster occurrence records to obtain a geological disaster risk grade result. 2.The geological disaster early warning analysis method based on multi-source remote sensing data according to claim 1, characterized in that, The step S2 characterizes different dimensions of data under the unified geological environment framework, extracts sensitive elements that have a direct impact on geological disaster occurrence, and the sensitive elements refer to environmental characteristic quantities that are identified from multi-source remote sensing data under the unified geological environment framework and have a significant response or indication effect on the gestation, development and triggering process of geological disasters. The step S2 includes the following three links in the process of obtaining potential risk factors: sensitive element screening, element normalization and weight distribution, and potential risk factor generation. 3.The geological disaster early warning analysis method based on multi-source remote sensing data according to claim 2, characterized in that, The step S2 performs sensitive element screening under the unified framework based on the multi-source fusion data set output by step S1: Through correlation coefficient analysis, calculate each data element Correlation with historical geologic hazard events; The statistical threshold is used to screen a subset whose correlation with historical geological disaster occurrence records exceeds a preset threshold: ； wherein represents a sensitive element after screening. 4.The geological disaster early warning analysis method based on multi-source remote sensing data according to claim 2, characterized in that, The step S2, when performing element normalization and weight distribution, obtains the normalized value of each sensitive element The normalized value is obtained by performing normalization processing ； wherein denotes the normalized sensitive element value; Weight The determination adopts mature entropy weight method or analytic hierarchy process to ensure that different elements reasonably contribute to the calculation of potential risk factors. The set of weighted sensitive elements is denoted by . 5.The geological disaster early warning analysis method based on multi-source remote sensing data according to claim 2, characterized in that, The step S2 generates a potential risk factor set by combining and mapping the weighted sensitive elements, and the formula model is: ； wherein represents a first potential risk factor generated under a unified framework ​​ Nonlinear mapping is adopted to map the weighted elements into risk factor values through logistic regression or support vector machine, ensuring that the linear combination results can be logically connected; Step S2 finally outputs the set of potential risk factors , . 6.The geological disaster early warning analysis method based on multi-source remote sensing data according to claim 1, characterized in that, The step S3 analyzes the coupling effect and dynamic evolution law between potential risk factors, identifies key combination conditions that may trigger geological disasters, and formalizes these conditions into a critical condition set; Step S3 includes the following four links in the process of obtaining the critical condition set of disaster occurrence: coupling function calculation, evolution matrix construction, critical condition threshold determination, and critical condition set generation. 7.The geological disaster early warning analysis method based on multi-source remote sensing data according to claim 6, characterized in that, The step S3 establishes a coupling relationship between the potential risk factors and the risk events by using covariance analysis to calculate linear coupling or using a logistic regression function to capture nonlinear interaction when performing coupling function calculation: With ​ ； wherein represents a factor with coupling strength under the triggering condition of geological disasters; with the coupling strength as matrix elements, resulting in a symmetric coupling matrix . 8.The geological disaster early warning analysis method based on multi-source remote sensing data according to claim 6, characterized in that, The step S3 performs evolution matrix construction, and the time / space division includes: The time sequence selects historical monitoring intervals; The spatial unit is divided according to the grid to ensure that the matrix corresponds to the actual environment; The coupling matrix is arranged according to the time sequence or the spatial unit to form an evolution matrix: 。 9.The geological disaster early warning analysis method based on multi-source remote sensing data according to claim 6, characterized in that, The step S3 sets a threshold vector when performing the critical condition threshold determination Further, a coupling threshold is set for each potential risk factor for each factor pair; In the determination of the threshold, the long-term monitoring data of the individual risk factor are sorted and the statistical quantile is taken as the critical threshold . 10.The geological disaster early warning analysis method based on multi-source remote sensing data according to claim 6, characterized in that, The step S3, when performing the critical condition set generation, determines for each potential risk factor and its coupling value decision: If and for all then is included in the critical condition ; The above process is repeated to obtain a complete critical condition set, and the expression is: ； each represents the environmental combination state that can trigger a disaster when the potential risk factor and its coupling reach a threshold value.

Images