An automated identification method for risk areas of clustered shallow soil landslide disasters

Abstract

This invention discloses an automated method for identifying risk zones of clustered shallow soil landslides, including methods for identifying landslide risk zones and for identifying houses threatened by landslides within those zones. First, remote sensing technology is used to acquire information on the region's topography, soil structure, vegetation, and rock distribution. Remote sensing imagery, infrared images, lidar data, and synthetic aperture radar data are used for data acquisition and processing to generate a high-precision digital elevation model and geological structure map. Through data fusion and analysis, combined with information on soil layers, topography, and vegetation, potential landslide risk zones are identified, and houses within the landslide threat zones are identified and assessed. This method improves the accuracy of identifying landslide risk zones and threatened houses, providing efficient and precise technical support for landslide disaster early warning and prevention. This invention has the advantages of automation, efficiency, and accuracy, and can be widely applied to real-time monitoring, risk assessment, and emergency management of landslide disasters.

Description

Technical Field

[0001] This invention belongs to the field of natural disaster monitoring and early warning, specifically an automated identification method for risk zones of clustered shallow soil landslide disasters. Background Technology

[0002] Landslides, a typical geological hazard, usually occur in mountainous, hilly, or sloping areas, characterized by their suddenness and destructiveness. With increasing global climate change and human activities, landslides are becoming more frequent and their damage is increasing year by year, seriously affecting people's lives and property and causing substantial economic losses and social impacts. Traditional landslide monitoring and assessment methods mostly rely on geological exploration, field investigations, and manual analysis. While these methods can assess landslide risk to some extent, they suffer from problems such as high workload, high cost, low efficiency, and poor accuracy when dealing with large-scale, complex terrain and high-frequency monitoring needs.

[0003] Currently, with the development of remote sensing, radar, and artificial intelligence technologies, landslide hazard monitoring and assessment methods based on remote sensing data and Geographic Information Systems (GIS) have gradually become a research hotspot. These technologies enable large-scale real-time monitoring through various means such as drones, satellites, and aerial platforms, providing accurate information on topography, soil layers, vegetation, and rock distribution, thus providing strong technical support for landslide hazard identification and early warning. In particular, technologies such as Synthetic Aperture Radar (SAR) and LiDAR can acquire high-precision three-dimensional ground data to identify potential landslide risk areas and hazard sources.

[0004] However, existing technologies have the following main problems:

[0005] Complexity of data acquisition and processing: Existing landslide hazard identification methods usually rely on multiple remote sensing techniques and complex geological analysis. Errors are prone to occur during data processing and fusion, and it is difficult to achieve automated and efficient processing over large areas.

[0006] Limitations of landslide risk assessment: Most existing methods identify landslide areas through traditional geological surveys and remote sensing image analysis. However, when dealing with landslide risks in large-scale and complex terrain, there is a lack of efficient risk assessment models and systems, which cannot accurately reflect the dynamic changes of landslide disasters.

[0007] Insufficient identification of houses threatened by landslides: In the prediction and early warning of landslide disasters, although landslide risk areas can be identified, there is often a lack of assessment of the degree of threat to specific buildings (such as houses) within the landslide threat area, thus failing to provide timely and accurate information support for post-disaster emergency response. Summary of the Invention

[0008] To address the problems mentioned in the background section, this invention provides an automated method for identifying risk zones of clustered shallow soil landslides. This method not only rapidly identifies landslide risk zones by establishing specific identification rules and combining high-precision remote sensing data and digital elevation difference models, but also further identifies buildings within the landslide threat area, providing more accurate disaster warnings and risk assessments. The implementation of this method can significantly improve the accuracy of landslide disaster warnings and response efficiency, providing a more scientific and operable technical means for disaster prevention and control.

[0009] To achieve the above objectives, the present invention provides the following technical solution:

[0010] An automated method for identifying risk zones of clustered shallow soil landslides includes a landslide risk zone identification method and a landslide-threatened housing identification method within the area. The landslide-threatened housing identification method uses a high-precision digital elevation model (DEM) and housing vector data, and implements automated identification using Geographic Information System (GIS) software to identify potentially threatened housing and delineate potential risk areas. The method includes the following steps:

[0011] S1. Acquire high-precision DEM data and prepare building vector data;

[0012] S2. Generate slope, elevation difference, and aspect layers;

[0013] S3. Mark the turning points of the houses, calculate the height difference and slope, and determine the threat to the houses caused by building on the slope.

[0014] S4. Generate the extension lines of the house boundaries, analyze the relationship between the extension lines and the slope and aspect, and identify houses threatened by clustered shallow soil landslides;

[0015] S5. Extract eligible houses as threatened houses files and export them for subsequent on-site investigation and risk zone delineation.

[0016] The landslide-threatened house identification method in the area generates slope and elevation difference layers from a high-precision DEM and extracts inflection points from house vector files. It uses elevation difference and slope values ​​to determine the threat of houses built on slopes and marks the inflection points of houses with elevation differences greater than 10 meters and slopes greater than 8 degrees. Based on this, potentially threatened houses are identified.

[0017] The landslide-threatened house identification method in the area generates outward extension lines of house boundaries and extracts the slope, aspect and elevation difference information of each extension line. Combined with the distance between the slope toe and the house, the direction angle of the extension line and the slope conditions, it identifies the danger lines that pose a landslide threat. When a house has one or more danger lines, it is determined that the house is a threatened house and the corresponding threatened house vector surface data is generated.

[0018] The landslide risk zone identification method includes the following steps:

[0019] S1. Obtain information on the region's topography, geology, slope structure, vegetation distribution, etc., through remote sensing technology and regional geological data;

[0020] S2. Combine optical image analysis technology to analyze the species, distribution, size and age characteristics of vegetation in order to evaluate the anchoring effect of the root system of different vegetation types on shallow soil.

[0021] S3. Obtain relevant local data from the database, including historical landslide data, meteorological data, soil moisture, precipitation, etc.

[0022] S4. The data from S1 to S3 are integrated and analyzed. A risk assessment model is used to comprehensively assess the risk level of landslide disasters and generate landslide risk areas and risk levels.

[0023] The remote sensing technologies in step S1 include visible light remote sensing images, infrared remote sensing images, lidar technology and synthetic aperture radar technology, which are used to acquire high-resolution topographic, soil layer and vegetation distribution information of the region.

[0024] Step S2 uses multispectral or hyperspectral images to assess the anchoring effect of vegetation roots on shallow soil, and calculates vegetation coverage and root strength using image classification and analysis techniques.

[0025] This involves analyzing soil stability and landslide susceptibility using topographic data, calculating soil slope, aspect, and soil moisture factors, and generating landslide risk heat maps by combining historical landslide event data and meteorological data to assess landslide susceptibility in various regions.

[0026] Step S4 involves fusing remote sensing images, radar data, and optical images, and combining geographic information system technology and machine learning methods to train a model using relevant local data in the database, thereby automatically assessing the landslide disaster risk index of the region.

[0027] The risk assessment model described therein is based on machine learning methods such as random forest, support vector machine or convolutional neural network. It is trained with historical landslide data and real-time data, and combined with meteorological data, geological data and historical landslide records in the database to predict the probability of landslide occurrence.

[0028] The landslide risk assessment results include a probability distribution map or heat map of landslide occurrence, and output the landslide risk level, which includes four levels: low risk, medium risk, high risk, and extremely high risk. The risk assessment results are output after comprehensive analysis of historical landslide data, precipitation data, meteorological forecast data, etc.

[0029] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0030] 1. This invention combines a high-precision digital elevation model (DEM), building vector data, and advanced GIS technology. This method can accurately analyze factors such as terrain changes, slope, aspect, and elevation differences within a region, effectively identifying potential landslide risk areas and threatened buildings. Compared to traditional methods relying on manual field surveys, this method has significant advantages in spatial coverage and accuracy, avoiding the high error rate that may result from traditional methods. The automated process for generating identification results greatly improves data processing efficiency, especially in applications covering large areas. It reduces manual intervention and time consumption, increases identification speed, and thus identifies more potential threat areas and affected buildings in a shorter time.

[0031] By identifying threatened houses and landslide risk areas in advance, precise data support can be provided for subsequent disaster early warning and risk assessment. Based on this, management departments can develop more accurate emergency plans and evacuation plans, reducing casualties and property losses caused by landslides. The method does not rely on historical disaster sites or hazard reports; it can automatically identify potential risk areas based on data such as house location and surrounding terrain, avoiding potential omissions or underestimations of risk areas in traditional methods. Precisely labeling the threat to each house improves the accuracy and targeting of disaster management.

[0032] Through GIS-based automated analysis, this method can identify and delineate potential disaster areas in advance, reducing the workload of on-site investigations. Especially when dealing with large mountainous areas or multiple disaster zones, traditional manual investigation methods can be extremely cumbersome and inefficient, while this method can significantly improve investigation efficiency and reduce personnel costs. By identifying threatened buildings in advance, this method can optimize disaster early warning processes, avoiding ineffective evacuations or emergency response operations, thereby reducing costs and resource waste for management departments.

[0033] This method integrates high-precision DEM data, building vector data, and the analysis of multiple factors such as slope and aspect to comprehensively assess landslide potential without relying on historical data or a single assessment dimension. This not only enhances the understanding of landslide mechanisms but also improves the accuracy of risk assessment, avoiding important geological factors that traditional methods may overlook. By identifying different hazard types, such as clustered shallow landslides and slope-cutting construction, this method can analyze different types of landslide threats separately, avoiding bias caused by treating all risks as the same hazard, and ensuring the comprehensiveness and accuracy of the assessment results.

[0034] 2. Traditional methods typically rely on static geological survey data and historical records for landslide risk assessment. This approach does not adequately consider the spatiotemporal changes in environmental factors. Especially under the influence of dynamic factors such as precipitation, traditional methods cannot respond to environmental changes in real time.

[0035] This invention utilizes high-resolution remote sensing imagery and radar data, along with dynamic analysis of real-time meteorological data, to perform refined spatiotemporal predictions, capturing changes across different time periods and regions. This makes the assessment results not only more accurate but also reflects the dynamic characteristics of landslide risk changing over time and with the environment. It can predict landslide occurrences under varying conditions of precipitation, temperature fluctuations, and other factors. This dynamic prediction method possesses strong real-time performance and responsiveness, enabling proactive preparedness. Particularly under extreme weather conditions, this invention can accurately predict high-risk periods for landslides, providing specific timeframes and spatial areas, offering precise guidance for disaster emergency response.

[0036] 3. Traditional landslide risk assessment methods often neglect the impact of vegetation root systems on soil stability and landslide occurrence, or simply consider vegetation as surface cover. They lack in-depth analysis of the anchoring effect of vegetation on soil stability.

[0037] This invention analyzes vegetation types, distribution, size, and age using optical imaging, and assesses the strength and stability of plant root systems. It also analyzes landslide risk by considering the interaction between vegetation roots and the soil layer. This approach not only considers the surface cover effect of vegetation but also deeply analyzes the soil-stabilizing effect of plant roots. This technology can effectively distinguish the different impacts of different vegetation types (such as herbaceous plants and trees) on soil stability, which is particularly important for risk assessment in mountainous and hilly areas, helping to accurately identify potential landslide zones.

[0038] 4. Traditional methods often assess landslide risk through coarse regional divisions (such as large-scale high-risk, medium-risk, and low-risk areas), lacking detailed and personalized analysis and failing to meet the risk management needs of different regions and environments. This invention, through comprehensive analysis of multiple dimensions (topography, slope structure, vegetation, meteorology, etc.), establishes a personalized risk assessment model for each specific region, enabling refined classification and management of landslide risk. This refined risk assessment provides more specific and operational decision support.

[0039] 5. Traditional landslide monitoring and risk assessment methods often require the deployment of numerous sensors, underground detection equipment, or monitoring systems. This not only necessitates substantial hardware but also involves complex installation and maintenance, especially in inaccessible or dangerous mountainous environments where sensor deployment is challenging and risky. This invention achieves high efficiency and accuracy in landslide risk assessment through non-invasive data acquisition methods such as remote sensing technology, radar data, optical imagery, and meteorological data. These technologies can acquire regional environmental and geological feature information without requiring extensive sensor deployment. Remote sensing methods such as drones and satellite imagery can simultaneously acquire data over a large area, avoiding the need for extensive on-site equipment and sensor deployment, reducing project implementation costs, and avoiding complex construction and equipment maintenance issues. It can quickly cover large areas, especially in mountainous regions, canyons, and other areas difficult to access manually. There is no need to worry about sensor malfunctions or errors.

[0040] 6. In traditional landslide monitoring, especially when relying on sensor deployment and manual surveys, human interference is significant. For example, sensor installation deviations, human error, and misjudgments of the environment can all affect the accuracy and reliability of the data. This invention reduces human interference, providing more objective and stable data. Most of this data is processed automatically, minimizing the possibility of human error, especially in large-scale data collection where the degree of automation is higher. Attached Figure Description

[0041] Figure 1 This is a flowchart illustrating the method for identifying houses threatened by landslides within the area described in this invention.

[0042] Figure 2 This is a flowchart illustrating the landslide risk zone identification method of the present invention. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] See Figure 1 and 2 An automated identification method for risk zones of clustered shallow soil landslides, including a landslide risk zone identification method and a landslide-threatened house identification method within the area.

[0045] See Figure 2 Specifically, the method for identifying landslide risk zones includes the following steps:

[0046] S1. Obtain information on regional topography, soil structure, vegetation, and rock and soil distribution using remote sensing technology.

[0047] 1. Remote sensing data acquisition

[0048] Remote sensing technology collects various information about the Earth's surface and subsurface through remote detection methods. Data acquisition can be performed using sensors mounted on satellites, drones, and aerial platforms. For information on topography, soil structure, vegetation, and rock and soil distribution, the following methods are employed for specific implementation:

[0049] 1.1 Acquiring Visible Light Remote Sensing Imagery

[0050] Use high-resolution satellite remote sensing imagery or RGB optical cameras mounted on drones. For even higher precision, aerial imagery acquisition platforms (such as high-resolution aerial cameras) can be used.

[0051] Data processing and analysis: Preprocessing steps such as radiometric correction and atmospheric correction of remote sensing images are used to eliminate the influence of atmospheric and lighting conditions. Image classification algorithms (maximum likelihood classification, support vector machine, etc.) are used to extract soil structure, vegetation cover, and topographic features from the images.

[0052] 1.2 Acquiring Infrared Remote Sensing Imagery

[0053] Use satellites, drones (equipped with infrared sensors), or airborne platforms.

[0054] Data processing and analysis: Areas with high reflectivity in infrared images are extracted, as these areas typically represent important information such as vegetation, soil moisture, and soil temperature. Soil moisture and vegetation health are analyzed in conjunction with vegetation indices (NDVI, EVI, etc.), thereby indirectly assessing soil stability.

[0055] 1.3 LiDAR technology for acquiring high-resolution terrain data

[0056] LiDAR equipment (such as airborne LiDAR, UAV LiDAR systems, etc.) is used to acquire three-dimensional information about the ground, vegetation, and buildings. LiDAR can accurately obtain the height and shape data of target objects by emitting laser beams and measuring reflection time, and can effectively extract details such as terrain, slope, and surface features.

[0057] Data processing and analysis: High-precision digital elevation models (DEM) and digital surface models (DSM) are generated using point cloud data processing software. Parameters such as slope, aspect, and surface roughness are extracted from LiDAR data; these parameters are crucial basic data for assessing landslide risk.

[0058] 1.4 Synthetic Aperture Radar (SAR) Technology

[0059] SAR satellites are used for continuous ground monitoring. SAR can penetrate clouds and rain / fog, and can still acquire ground information under various adverse weather conditions. SAR technology is particularly suitable for monitoring soil subsidence, surface deformation, landslide activity, and changes in water systems.

[0060] Data Processing and Analysis: Interferometric Synthetic Aperture Radar (InSAR) technology extracts ground deformation information through differential analysis of multi-temporal SAR images. It is particularly effective for monitoring phenomena such as landslides and soil subsidence. Radial basis function interpolation (Kriging) and multi-temporal analysis methods can help identify subtle changes in soil layers and further assess landslide risk areas.

[0061] 2. Data Processing and Fusion

[0062] By working in concert with multiple remote sensing technologies, comprehensive information on soil layers, topography, rock and soil distribution, and vegetation can be obtained. Further data fusion and processing are then required.

[0063] 2.1 Data Fusion

[0064] Spatial registration and fusion of data from different sources (satellite remote sensing images, LiDAR point clouds, SAR images, etc.) are performed on a GIS platform. After fusion, the information is stored in raster data formats, vector data formats, etc., to facilitate subsequent analysis and display. For example, combining topographic data obtained from LiDAR with soil and vegetation information from remote sensing images can better identify potential landslide-prone areas.

[0065] 2.2 Soil Structure Analysis

[0066] By combining SAR data, infrared remote sensing data, and high-precision topographic maps using LiDAR technology, the soil structure of the region is analyzed, particularly the thickness, density, and type of soil layers. By comparing different geological data, geological faults and soil-rock structures are identified using remote sensing data, generating soil-rock type maps, which provide crucial information for subsequent landslide risk assessment.

[0067] 2.3 Vegetation distribution and soil stability assessment

[0068] Using remote sensing images and spectral data, we analyze vegetation cover and root distribution within the region, and combine this with surface undulation information obtained from lidar or SAR technology to analyze whether vegetation can effectively anchor the soil and prevent soil slippage. By combining vegetation indices (NDVI, EVI, etc.) with indicators such as relative soil stability and soil moisture, we determine the stabilizing effect of vegetation within the region.

[0069] 3. Data Output

[0070] Topographic data: slope, aspect, surface deformation, etc.

[0071] Soil data: soil structure, soil thickness, moisture, etc.

[0072] Vegetation data: vegetation coverage, NDVI value, root density, etc.

[0073] These data will serve as the basis for subsequent steps, supporting the analysis and prediction of landslide risks.

[0074] S2. Utilize radar technology to obtain information on soil faults, soil density distribution, underground cavities, and water systems.

[0075] Radar technology (including ground-based, airborne, or satellite radar) is used to detect target areas and obtain information such as soil faults, soil density distribution, underground cavities, and water systems. This data is crucial for assessing soil stability, landslide risk, and groundwater flow. The process involves adjusting radar parameters such as frequency and power to ensure accurate detection data is obtained at different target depths.

[0076] 1. Select appropriate radar technology

[0077] Select the appropriate radar technology based on the size of the detection area and the depth of the soil layer:

[0078] Ground-based radar: suitable for smaller areas, with a detection depth typically ranging from 0 to 30 meters. It has high spatial resolution and is suitable for detecting soil faults and underground cavities.

[0079] Airborne radar: Suitable for detection over a wide area, especially for obtaining information on soil density distribution and water systems over large areas. Airborne radar operates at a low frequency and can detect depths exceeding 50 meters.

[0080] Satellite radar: Suitable for deep exploration over a wide area, capable of detecting depths from 10 meters to hundreds of meters, and is suitable for large-scale monitoring of soil density distribution and water systems.

[0081] 2. Adjustment of radar technical parameters

[0082] Based on the target depth and specific needs, the radar's frequency, power, and other parameters are adjusted to ensure effective detection of soil layer information at different depths.

[0083] Frequency selection:

[0084] High frequency (500 MHz - 1 GHz): Suitable for shallow exploration (0 - 10 meters), with high resolution, providing clearer details such as faults and underground cavities.

[0085] Low frequency (50 MHz - 500 MHz): Suitable for deeper exploration (10 - 50 meters). Although the resolution is lower, it has strong penetration ability and is suitable for obtaining information on soil density distribution and deep water system.

[0086] Power adjustment:

[0087] High power (30 W - 50 W): Increases signal strength, which can improve penetration depth and is suitable for deep-layer detection.

[0088] Low power (5 W - 20 W): Suitable for shallower areas, reducing interference signals and noise, and improving signal quality.

[0089] Scanning mode:

[0090] Single scan: Used for rapid scanning of small areas, such as detecting specific soil faults or underground cavities.

[0091] Continuous scanning: Suitable for monitoring large areas, generating comprehensive soil density distribution maps and water system distribution maps.

[0092] 3. Data Collection

[0093] Ground detection and data acquisition are performed using radar at appropriate frequency and power settings. The main data collected includes:

[0094] Soil fault information: The distribution and depth of soil faults can be determined by analyzing the time and intensity of radar wave reflections. Radar waves exhibit different reflection characteristics when encountering different media; by analyzing the delay time and intensity of the echoes, the boundaries and locations of faults can be determined.

[0095] Soil density distribution: The intensity of radar wave reflection is related to the dielectric constant of the soil layer. Soil density and moisture content significantly affect radar wave propagation. Based on the intensity of the reflected signal, the density distribution of the soil layer can be estimated, and thus the stability of the soil layer can be analyzed.

[0096] Underground cavities: The presence of underground cavities can cause abnormal reflections of radar waves, manifesting as weak signal echoes or abnormal echo times. By performing multiple scans and comparing different frequencies, the distribution and depth of underground cavities can be accurately located.

[0097] Water system information: Radar waves will produce obvious reflections in water bodies. By analyzing the reflected signals of water systems, we can obtain information such as the distribution of groundwater flow, water level, and flow direction.

[0098] 4. Data Processing and Analysis

[0099] The data acquired by radar needs to be processed and analyzed, including the following steps:

[0100] Data preprocessing: The raw radar data collected is processed by denoising, smoothing, filtering and other methods to remove unnecessary interference signals.

[0101] Time-domain and frequency-domain analysis:

[0102] Time-domain analysis: Analyze the arrival time and duration of echo signals to estimate the depth, thickness, and fault conditions of the soil layer.

[0103] Frequency domain analysis: By analyzing the spectrum of the reflected wave, the electromagnetic properties of the soil layer are extracted, and then the density distribution and water system conditions of the soil layer are analyzed.

[0104] Inversion and Modeling: Using inversion algorithms, two-dimensional or three-dimensional models of soil layers are generated based on radar wave reflection data, accurately depicting the distribution of underground structures. Common inversion methods are employed, including least squares and genetic algorithms.

[0105] Underground cavity detection: Underground cavities are identified using the reflection difference method. The reflection intensity and signal propagation time differ significantly in cavity areas; through comparative analysis, the boundaries and location of the cavities can be accurately determined.

[0106] Soil density analysis: Soil density can be estimated by analyzing the relative propagation speed and reflection intensity of radar waves. Areas with higher reflection intensity generally indicate denser soil layers, while those with lower reflection intensity may be looser soil layers.

[0107] 5. Result Output

[0108] Location and orientation of soil faults: to provide fault information for subsequent landslide risk assessment.

[0109] Soil density distribution: reflects soil stability and guides subsequent soil stability analysis.

[0110] Distribution of underground cavities: Identify potential landslide triggering factors, especially the geological instability that underground cavities may cause.

[0111] Water system information: Provides data support for further analysis of hydrological and precipitation impacts.

[0112] This data will serve as input for subsequent landslide risk assessments, helping to develop accurate landslide risk models.

[0113] S3. Analyze the species, distribution, size, and age characteristics of vegetation using optical image analysis techniques to assess the anchoring effect of root systems on shallow soil for different vegetation types.

[0114] Optical image analysis technology (multispectral or hyperspectral image analysis) is used to classify and analyze the vegetation in the area, extract information such as vegetation type, distribution, size and age, and assess the anchoring effect of vegetation roots on shallow soil based on these characteristics, providing auxiliary support for landslide risk assessment.

[0115] 1. Acquisition and processing of optical images

[0116] Optical images (including multispectral and hyperspectral images) can be acquired through remote sensing platforms such as satellites and drones. Different spectral bands correspond to different vegetation characteristics, such as the red band and the near-infrared band.

[0117] Multispectral images typically include red, green, blue, and near-infrared bands. They are commonly used for vegetation analysis and classification.

[0118] Hyperspectral imaging: Acquires data across multiple consecutive spectral bands, providing more detailed information than multispectral imaging, and is suitable for precise vegetation analysis.

[0119] The acquired optical images are preprocessed before further analysis. Preprocessing includes radiometric correction, atmospheric correction, and geometric correction.

[0120] NDVI (Normalized Difference Vegetation Index):

[0121] Where NIR: reflectance in the near-infrared band; Red: reflectance in the red band;

[0122] The NDVI value reflects vegetation growth, ranging from -1 to 1, with values ​​close to 1 indicating well-grown vegetation. Vegetation cover calculation: The NDVI index is used to estimate vegetation cover in different areas. Vegetation cover. It is calculated based on the pixel classification results in the image.

[0123] ;in, Vegetation coverage (%); Pixels in vegetation areas: Number of pixels classified as vegetation; Total pixels: Total number of pixels in the image;

[0124] 2. Tree Age and Species Analysis

[0125] Tree age estimation generally relies on the tree's growth characteristics and the chlorophyll content of the vegetation. By analyzing the reflectance of different bands in hyperspectral images, information such as chlorophyll content, leaf area index (LAI), and leaf thickness can be extracted; these parameters are closely related to tree age. The following methods are applied using the spectral reflectance characteristics of vegetation:

[0126] Formula for estimating chlorophyll content (based on hyperspectral reflectance data):

[0127] ;in, Red band reflectivity; Near-infrared reflectivity; The regression coefficient, obtained by fitting experimental data, shows that chlorophyll content is closely related to factors such as tree age, tree health, and soil moisture. Therefore, the age of a tree can be inverted by the chlorophyll content in the image.

[0128] 3. Assessment of the anchoring effect of vegetation roots on shallow soil

[0129] The anchoring effect of vegetation roots on shallow soil is a crucial factor in preventing landslides and soil erosion. The anchoring force of roots is influenced by various factors, including tree species, tree age, root depth, and root density. While optical imaging cannot directly observe root distribution, it can indirectly infer root growth characteristics and soil stability. The anchoring effect of roots is assessed through the following aspects:

[0130] Root strength assessment: Root strength of vegetation Root strength is closely related to root density and tree health. It can be estimated by assessing vegetation growth using the following formula:

[0131] ;in, Root strength (dimensionless); LAI: Leaf area index (m²) 2 / m 2 Tree age: Tree age (in years); : Empirical constants, adjusted according to different tree species and soil types;

[0132] Vegetation root density and soil stability: Root density It is usually positively correlated with soil stability. It can be assessed using the following formula:

[0133] ;in, Root density (kg / m 2 ); Root mass: Total mass of vegetation roots (kg); Area: Area of ​​the analysis region (m²) 2 );

[0134] The greater the root density, the stronger the anchoring effect of vegetation on the soil, the higher the soil stability, and the lower the probability of landslides.

[0135] Soil stability calculation: Root density and soil shear strength ( There is a close relationship between them. It can be assessed using the soil shear strength formula:

[0136] ;in, Soil shear strength ; Soil cohesion (Pa); Normal pressure of the soil (Pa); The internal friction angle of the soil (°); : The adhesion of roots to soil (dimensionless);

[0137] 4 Output

[0138] Vegetation Cover Table: Records the vegetation cover ratio in each study area, and can be used to generate heat maps or coverage maps using GIS technology.

[0139] Root strength data table: Records the root strength of each type of vegetation, and calculates it by weighting factors such as plant species and tree age.

[0140] Plant species and tree age data: Based on remote sensing image analysis and plant classification data, output the species and tree age of the main plants in the region.

[0141] Vegetation health status map: Based on NDVI values, a distribution map of vegetation health status is drawn.

[0142] This data will provide important input for subsequent landslide risk assessments, helping the system identify areas that may have a higher risk of landslides due to vegetation factors.

[0143] S4. Utilize relevant local data in the database, including historical landslide data, meteorological data, soil moisture, precipitation, and seismic activity.

[0144] The goal of this process is to calculate the landslide risk level of different regions by combining historical records with real-time data, and to provide a basis for early warning and prevention of landslide disasters.

[0145] 1. Data Sources and Data Preprocessing

[0146] 1.1 Data Source:

[0147] Historical landslide data includes detailed information such as the time, location, cause, affected area, landslide type, and volume of historical landslide events. This data typically comes from government departments, research institutions, and regional disaster record systems.

[0148] Meteorological data includes meteorological conditions such as precipitation, temperature, wind speed, and humidity, and is usually provided by meteorological stations, satellite remote sensing data, and weather forecasting systems.

[0149] Soil moisture data: Soil moisture data can be obtained through ground weather stations, remote sensing satellites, or soil moisture sensors.

[0150] Precipitation data: acquired through meteorological stations and remote sensing technology, and typically includes the spatiotemporal variations of monthly, quarterly, and annual precipitation.

[0151] 1.2 Data Preprocessing: To ensure the validity and consistency of the data, the following preprocessing is required:

[0152] Data cleaning: removing outlier data, duplicate data, and filling or removing missing values.

[0153] Data standardization: When different data have different units of measurement, standardization is required to facilitate comparison and model training.

[0154] Time synchronization: Time-aligning data from different time scales (such as meteorological data, historical landslide events, real-time earthquake data, etc.) to enable joint analysis.

[0155] 2. Feature extraction and landslide risk factor construction

[0156] After acquiring and cleaning the data, it is necessary to extract the key factors influencing landslide occurrence and construct appropriate features for landslide risk assessment. These factors include, but are not limited to:

[0157] Slope: Topographic slope is an important factor in landslide occurrence. The steeper the slope, the greater the likelihood of a landslide. Slope can be calculated from remote sensing imagery or topographic data (such as digital elevation models).

[0158] Soil moisture: The higher the soil moisture, the weaker the soil's shear strength, and the greater the likelihood of landslides. Soil moisture can be obtained through meteorological data, remote sensing data, and soil sensors.

[0159] Precipitation: The impact of precipitation on landslides is obvious. Excessive precipitation saturates the soil, making it more prone to soil movement and triggering landslides. Precipitation data can be analyzed using historical and real-time meteorological data.

[0160] Historical landslide event data: Historical landslide events can help assess the landslide susceptibility of a region, understand the patterns of landslide occurrence, their spatial and temporal distribution, and their relationship with factors such as meteorology, earthquakes, and soil.

[0161] 3. Output and Decision Support

[0162] Meteorological data table: includes fields such as time, location, precipitation, temperature, and humidity.

[0163] Historical landslide event records: a database containing information such as the specific time, location, scale, and type of landslides.

[0164] This data provides detailed environmental context and potential triggering factors for subsequent steps, which can help identify which areas are most vulnerable to landslides and support the prediction and prevention of landslide risks.

[0165] S5. Data Fusion and Final Risk Assessment

[0166] Data Fusion: The key to S5 lies in the effective fusion of data from different sources (topographic data, radar data, vegetation data, meteorological data, etc.). Each data type provides a different perspective on landslide risk factors, such as terrain slope, soil structure stability, the impact of vegetation roots on the soil, and the triggering effect of meteorological factors. By fusing this data, we can obtain a more comprehensive and accurate landslide risk assessment model.

[0167] Construction of the risk assessment model: By using machine learning or statistical methods and combining the weights and influences of different data sources, a multi-dimensional landslide risk prediction model is constructed. This model can calculate the probability of landslide occurrence based on the combination of various factors and classify them according to risk level.

[0168] The specific steps are as follows:

[0169] 1. Data preprocessing and standardization: Ensure that data from different sources are comparable and avoid data differences affecting the final results.

[0170] Missing value handling: For missing geographical, meteorological or other data, interpolation, mean imputation or other methods are used to handle them.

[0171] Data standardization: Standardize data of different scales and units to bring them to the same scale and avoid the influence of bias from a particular data source on the final model.

[0172] For example, remote sensing image data (image data) and meteorological data (numerical data) are standardized to ensure that they have the same influence in the model.

[0173] Data smoothing: Utilizing filtering methods to smooth data, remove outliers and noise, and improve data quality and accuracy.

[0174] 2. Data Fusion: Combining multiple data sources (remote sensing images, radar data, optical images, meteorological data, etc.) into a unified, multi-dimensional dataset.

[0175] This includes spatial data fusion: fusing spatial data such as terrain, radar data, and vegetation distribution, and integrating these data into a multi-level spatial model through a geographic information system (GIS).

[0176] For example, data such as vegetation cover, slope, soil type, humidity, and precipitation can be combined according to geographical regions to generate comprehensive risk characteristics for each region.

[0177] Time-series data fusion: For meteorological data and historical landslide data, time-series analysis can be used to combine the time nodes of landslide occurrence to identify the spatiotemporal distribution patterns of landslides.

[0178] 3. Feature engineering: Extracting features related to landslide risk from different datasets.

[0179] Topographic features: slope, aspect, topographic curvature, and other data.

[0180] Soil characteristics: soil moisture, soil structure, soil type, etc.

[0181] Vegetation characteristics: vegetation coverage, root strength, vegetation health status, root distribution depth, etc.

[0182] Meteorological characteristics: precipitation, temperature, humidity, air pressure, etc.

[0183] Characteristics of historical landslides: frequency, location, and scope of impact of historical landslides.

[0184] 4. Model selection and training: Use appropriate machine learning models to predict landslide risks and determine risk levels.

[0185] Optional,

[0186] Supervised learning algorithms: These algorithms use historical data on known landslide occurrences and non-occurrences to train the model. Commonly used models include:

[0187] Random Forest: Random forests are constructed using decision trees and can effectively handle multi-dimensional, non-linear data, making them suitable for risk prediction of geological and meteorological data.

[0188] Support Vector Machine: Support Vector Machine can effectively classify complex boundary conditions and is suitable for delineating landslide risk areas.

[0189] Neural networks: especially convolutional neural networks, can be used to process image data (such as remote sensing images) and extract complex spatial features.

[0190] Gradient Boosting Machine (GBM): GBM can optimize feature selection and is suitable for accurate landslide risk prediction.

[0191] Unsupervised learning algorithms: For unlabeled data, clustering methods can be used to analyze the risk distribution within a region and classify the risk level.

[0192] 5. Risk Level Assessment: Based on the model output, determine the risk level of the landslide.

[0193] Probability threshold method: Based on the probability of landslide occurrence output by the model, different thresholds are set to classify the risk level. For example:

[0194] Low risk: The probability of a landslide occurring is less than 30%.

[0195] Medium risk: The probability of a landslide is between 30% and 60%.

[0196] High risk: The probability of a landslide is between 60% and 90%.

[0197] Extremely high risk: The probability of a landslide exceeds 90%.

[0198] Multi-factor weighted method: Different features (such as topography, vegetation, and weather) are assigned different weights, and a comprehensive landslide risk value is obtained by weighted summation. Then, the risk level is determined based on this comprehensive value.

[0199] 6. Risk area map generation: The risk assessment results are presented in map form, showing the landslide risk level of each area.

[0200] Using GIS technology, the risk levels of different areas are visualized as layers, and different risk levels are distinguished by color.

[0201] The spatial distribution and intensity of risks are displayed through methods such as heat maps and contour maps.

[0202] Output area maps with different risk levels for decision-makers to use as a reference in planning, architectural design, disaster prevention, and other aspects.

[0203] S5 output results:

[0204] Landslide risk levels (low risk, medium risk, high risk, extremely high risk): Different risk levels are classified based on the probability predicted by the model and risk indicators.

[0205] Landslide Risk Zone Map: Based on model-based risk assessment, this map outputs a distribution map of landslide risk in various regions. This map can serve as a reference for landslide monitoring, land use planning, and disaster prevention.

[0206] Risk assessment report: A detailed record of the risk assessment results for each area, including the risk level, main influencing factors, and predicted probability of landslide occurrence, for decision-makers' reference.

[0207] S5 relies on data collected in phases S1 through S4. Through data fusion and machine learning algorithms, and by comprehensively considering multiple factors such as topography, soil, vegetation, meteorology, and historical landslides, S5 can accurately assess landslide risk and generate a risk level map. The final output risk area map will provide important reference for engineering projects, disaster prevention, and emergency response.

[0208] See Figure 1 Furthermore, after identifying landslide risk areas, if there are houses within the risk area, the houses threatened by landslides within the risk area are further identified, as follows:

[0209] 1. Data Collection and Preprocessing

[0210] First, relevant foundational data needs to be collected and processed, including a high-precision digital elevation model (DEM) and building vector data. The quality and accuracy of this data directly affect the identification results.

[0211] 1.1 High-precision DEM data

[0212] Data source: Obtain elevation data of the target area through remote sensing technology or existing digital elevation data (such as LiDAR data, SRTM data, etc.).

[0213] Accuracy requirements: The DEM accuracy should be no less than 2 meters or 5 meters, depending on the terrain features of the target area.

[0214] Data preprocessing: DEM data is preprocessed using GIS software, including noise removal, filling missing values, and correcting terrain anomalies.

[0215] 1.2 House Vector Data

[0216] Data source: Obtain building vector data from satellite remote sensing imagery, drone aerial photography, or existing geographic information databases.

[0217] Data preprocessing: Convert the building vector data into a format suitable for analysis in conjunction with DEM data, ensuring accurate building boundaries.

[0218] 2. DEM Data Analysis and Layer Generation

[0219] GIS software (such as ArcGIS, QGIS, etc.) is used to analyze high-precision DEMs and generate slope, elevation difference, and aspect layers related to building vector data.

[0220] 2.1 Generate slope and elevation difference layers

[0221] Slope layer: A slope layer is generated by calculating the slope (i.e., the slope of elevation change) of each raster cell in the DEM data.

[0222] Elevation layer: Calculate the elevation difference between each raster cell and its adjacent raster cells in the DEM to obtain the elevation difference distribution within the region.

[0223] 2.2 Generate slope aspect layer

[0224] Slope aspect layer: Based on DEM data, the slope aspect of each raster cell is calculated, i.e., the direction of the slope surface, to help identify areas where landslides may occur.

[0225] 3. Identification of houses threatened by hillside construction

[0226] 3.1 Extraction of House Turning Points

[0227] Extract the inflection points (i.e., the corners of the building boundaries) from the building vector data, and extract the elevation difference and slope information within a 20-meter radius around each inflection point.

[0228] 3.2 Analysis of Elevation Difference and Slope

[0229] For each house inflection point, calculate the elevation difference at that point (i.e., the maximum elevation within a 20-meter radius of that point minus the elevation at the inflection point), as well as the slope at that point.

[0230] Threshold setting: If the slope is greater than 8 degrees and the elevation difference is greater than 10 meters, then mark the point as a threatened point.

[0231] House screening: If a house has one or more markers at its inflection point, it indicates that the house is threatened by hillside construction, and the house is identified as a threatened house.

[0232] 4. Identification of houses threatened by clustered shallow soil landslides

[0233] 4.1 Generation of Extension Lines

[0234] House boundary extension lines: Generate straight lines extending outward based on the boundaries of the house vector data. One extension line is generated every 10 meters. Four extension lines are generated at the inflection points of the house. Two lines are perpendicular to the two side boundaries, plus two other lines that divide the boundary into thirds.

[0235] Extension line length: Set the extension line length to 50 meters.

[0236] 4.2 Slope and Aspect Analysis of the Extension Line

[0237] For each extension line, calculate the slope, aspect, and elevation difference of each grid it passes through. In particular, identify potential landslide hazard zones by analyzing the slope variations in the areas traversed by the extension lines.

[0238] 4.3 Determination of Danger Line

[0239] Slope toe calculation: Based on the extension line, determine whether there is a steep slope (i.e., slope toe) with a slope greater than 25 degrees in the area near the house, and calculate the distance from the slope toe to the house.

[0240] Calculation of the angle between the direction angle and the slope aspect: Calculate the direction angle of the extension line and the angle between the direction angle and the slope aspect of the grid through which the extension line passes, and determine whether the conditions for landslide occurrence are met.

[0241] Landslide condition determination:

[0242] The distance between the toe of the slope and the house is less than 10 meters;

[0243] The angle between the direction angle of the extension line and the slope aspect is less than 20 degrees;

[0244] The extension line passes through areas with a slope greater than 25 degrees for more than 40 meters and through areas with a slope greater than 35 degrees for more than 20 meters.

[0245] The total elevation difference of the extension line is greater than 30 meters.

[0246] Danger line marking: When an extension line meets all of the above conditions, it is determined to be a danger line.

[0247] 4.4 Evacuation of threatened houses

[0248] Danger line screening: If a house is crossed by at least one danger line, the house is determined to be a threatened house, and vector data of threatened houses is generated.

[0249] 5. Outputs and Risk Assessment

[0250] 5.1 Recognition Result Output

[0251] The identification results are output as two files: a vector surface file of the threatened building and a corresponding hazard line file, for use in subsequent on-site investigations and risk zone delineation.

[0252] 5.2 Risk Assessment and Further Analysis

[0253] Based on the data of the identified threatened houses, further disaster risk assessments are conducted. By combining historical landslide data, meteorological data, etc., the risk level of each area is calculated, and a landslide risk heat map or probability distribution map is generated to guide disaster prevention and mitigation work.

[0254] 6. System Integration and Automation

[0255] Integrating the above analysis process into an automated system and using a GIS platform (such as ArcGIS or QGIS) to automate operations reduces manual intervention and improves processing efficiency and accuracy.

[0256] The system automates data loading, layer generation, analysis and calculation, and result output, supports batch processing, and is suitable for identifying landslide-threatened houses over a wide area.

[0257] The following are representative application cases and processes of the technology developed based on this invention. These cases can effectively verify the actual effect of this invention in landslide risk assessment.

[0258] Case 1: Automated Identification of Landslide Risk Zones in Mountainous Areas

[0259] A certain mountainous area has long been threatened by landslides and urgently needs to conduct risk assessments using remote sensing technology.

[0260] step:

[0261] Remote sensing data acquisition: High-resolution satellite remote sensing imagery is used to acquire mountainous terrain information. Vegetation cover and soil moisture data are obtained using drones equipped with RGB optical cameras and infrared sensors. LiDAR technology is used to acquire 3D terrain data, generating digital elevation models (DEMs) and digital surface models (DSMs) to identify slope and terrain features in the mountainous areas.

[0262] Data processing and fusion: A GIS platform is used to spatially register and fuse satellite remote sensing images, LiDAR point clouds, and SAR images. Potential landslide-prone areas are identified by analyzing soil structure, vegetation cover, and soil moisture.

[0263] Landslide-threatened housing identification: Based on the identified landslide risk areas, combined with existing housing vector data in the Geographic Information System (GIS), determine which houses are located in high-risk landslide areas.

[0264] Results: High-risk landslide areas in the mountains were successfully identified, and threatened houses were marked, helping the government to deploy disaster prevention measures in a timely manner and avoid potential losses.

[0265] Case 2: Landslide Risk Identification in Urban Surrounding Areas

[0266] The area surrounding a city has hilly terrain, and soil cracks have recently appeared in some areas, necessitating an assessment of landslide risk.

[0267] step:

[0268] Remote sensing data acquisition: High-resolution satellite imagery is used to acquire topographic information about the city's surroundings. Combined with infrared remote sensing imagery, soil moisture is analyzed to identify potential landslide areas. Synthetic aperture radar (SAR) technology is employed to monitor minute changes in the soil layer and identify any subsidence or deformation.

[0269] Data fusion and processing: By combining LiDAR data, SAR images, and infrared remote sensing imagery, the soil structure of the area is analyzed to assess soil stability. The vegetation index (NDVI) is used to analyze whether the vegetation is sufficient to cover the soil and prevent landslides.

[0270] Landslide-threatened housing identification: Based on remote sensing data and geographic information systems, determine whether surrounding houses are in potential landslide areas and assess the likelihood of damage to the houses.

[0271] Results: The method accurately identified high-risk landslide areas around the city, assessed threatened houses, and issued timely warnings, ensuring the safety of people and property.

[0272] Case 3: Landslide Risk Assessment for Engineering Construction in Mountainous Areas

[0273] Infrastructure construction is underway in a mountainous area, which poses a risk of landslides and requires a landslide threat assessment.

[0274] step:

[0275] Remote sensing data acquisition: High-precision terrain scanning was conducted using a drone equipped with a LiDAR system to obtain a digital elevation model (DEM), and the slope, aspect, and surface features of the area were analyzed. Infrared remote sensing imagery was used to assess soil moisture, and SAR technology was used to monitor ground deformation and soil settlement.

[0276] Data fusion and analysis: Remote sensing images, LiDAR data, and SAR data are fused to identify landslide-prone areas within the region. Based on soil structure and vegetation distribution, and considering the location of the construction project, the landslide risk in the construction area is assessed.

[0277] Landslide-threatened building identification: Using a Geographic Information System (GIS), known building information is overlaid with risk areas to identify potentially affected buildings.

[0278] Results: This method successfully assessed the landslide risk in the construction area and provided design adjustment suggestions for the construction unit, thus avoiding potential landslide disasters.

[0279] Case 4: Landslide Risk Assessment in Rural Mountainous Areas

[0280] Small-scale landslides have occurred repeatedly in a rural mountainous area, requiring an assessment of the landslide risk in the region and the development of preventative measures.

[0281] step:

[0282] Remote sensing data acquisition: High-resolution satellite remote sensing imagery and UAV remote sensing are used to acquire data on topography, soil structure, and vegetation distribution. SAR technology is used to detect minute deformations on the land surface and analyze soil layer changes.

[0283] Data fusion and analysis: Remote sensing imagery, LiDAR data, and SAR data are integrated to conduct geological fault analysis and assess soil stability. Factors such as soil moisture, vegetation health, and slope are analyzed, and combined with historical landslide data for the region, to determine the landslide risk level.

[0284] Landslide-threatened housing identification: Using house vector data, combined with landslide risk areas, threatened farmhouses and residential areas are identified.

[0285] Results: High-risk landslide areas were successfully identified in the region, and the threat of landslides to houses and people was reduced through land use planning and engineering measures.

[0286] Case 5: Landslide Disaster Early Warning Near Coastline

[0287] A low-lying area near a coastline is at risk of landslides, necessitating the implementation of landslide disaster early warning systems.

[0288] step:

[0289] Remote sensing data acquisition: Topographic information of the coastal area was acquired using SAR and LiDAR technologies to analyze soil structure and soil moisture. Infrared remote sensing imagery was combined to assess soil moisture and vegetation distribution.

[0290] Data processing and fusion: SAR images, LiDAR point cloud data, and infrared remote sensing images are fused to analyze the soil stability of the coastline. Landslide risk in the area is comprehensively assessed based on topographic features, soil moisture, and vegetation cover.

[0291] Landslide-threatened housing identification: By overlaying housing vector data with high-risk landslide areas, residential houses located within the risk zone are identified, and emergency evacuation recommendations are provided based on the data.

[0292] Results: This method effectively predicted landslide risk areas near the coastline and successfully identified threatened houses, providing strong decision-making support for local governments.

[0293] These real-world examples demonstrate the effectiveness of this automated identification method in different geographical environments and application scenarios, providing precise data support for landslide disaster prevention and emergency management.

[0294] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An automated method for identifying risk zones of clustered shallow soil landslides, characterized in that, This includes landslide risk zone identification and landslide-threatened housing identification within the area. The identification of landslide-threatened housing within the area utilizes a high-precision digital elevation model (DEM) and housing vector data, and is automated using Geographic Information System (GIS) software to identify potentially threatened housing and delineate potential risk areas. The process includes the following steps: S1. Acquire high-precision DEM data and prepare building vector data; S2. Generate slope, elevation difference, and aspect layers; S3. Mark the turning points of the houses, calculate the height difference and slope, and determine the threat to the houses caused by building on the slope. S4. Generate the extension lines of the house boundaries, analyze the relationship between the extension lines and the slope and aspect, and identify houses threatened by clustered shallow soil landslides; S5. Extract eligible houses as threatened houses documents and export them for subsequent on-site investigation and risk zone delineation; Step S3 includes: S31. House Inflection Point Extraction: Extract the inflection points of houses from the house vector data, and extract the surrounding elevation difference and slope information for each inflection point; S32. Elevation and Slope Analysis: For each house inflection point, calculate the elevation difference and slope at that point; if a house inflection point has one or more marker points, it indicates that the house is threatened by slope-cutting construction, and the house is identified as a threatened house; Step S4 includes: S41. Extension Line Generation: Generates straight lines extending outwards based on the boundaries of the building vector data; S42. Slope and Aspect Analysis of Extension Lines: Calculate the slope, aspect, elevation difference, and slope variation of each grid cell traversed by each extension line, as well as the slope variation of the area traversed by the extension line, to identify potential landslide hazard zones; S43. Danger Line Determination: Based on the extension line, determine whether there is a steep slope with an inclination of more than 25 degrees in the area near the house, i.e., the toe of the slope, and calculate the distance from the toe of the slope to the house; calculate the direction angle of the extension line and calculate the angle between the extension line and the slope direction of the grid through which the extension line passes, and determine whether the conditions for landslide occurrence are met. When an extension line meets the judgment criteria, it is determined to be a danger line. S44. Threatened House Extraction: If a house is crossed by at least one danger line, it is determined to be a threatened house, and threatened house vector data is generated.

2. The method according to claim 1, characterized in that, Landslide-threatened houses in the area are identified by generating slope and elevation difference layers from a high-precision DEM and extracting inflection points from house vector files. The threat of houses built on slopes is determined by using elevation difference and slope values, and inflection points of houses with elevation differences greater than 10 meters and slopes greater than 8 degrees are marked. Based on this, potentially threatened houses are identified.

3. The method according to claim 1, characterized in that, Landslide-threatened houses within the area are identified by generating outward extension lines of the house boundaries and extracting the slope, aspect, and elevation difference information of each extension line. Combined with the distance between the toe of the slope and the house, the direction angle of the extension line, and the slope conditions, danger lines with landslide threats are identified. When a house has one or more danger lines, the house is determined to be a threatened house, and corresponding threatened house vector surface data is generated.

4. The method according to claim 1, characterized in that, The identification of landslide risk zones includes the following steps: ①. Obtain information on the region's topography, geology, slope structure, and vegetation distribution through remote sensing technology and regional geological data; ②. Combine optical image analysis technology to analyze the species, distribution, size and age characteristics of vegetation in order to evaluate the anchoring effect of the root system of different vegetation types on shallow soil; ③. Obtain relevant local data from the database, including historical landslide data, meteorological data, soil moisture, and precipitation; ④. Integrate and analyze the data from ① to ③, use a risk assessment model to comprehensively assess the risk level of landslide disasters, and generate landslide risk areas and risk levels.

5. The method according to claim 4, characterized in that, The remote sensing technologies mentioned therein include visible light remote sensing images, infrared remote sensing images, lidar technology and synthetic aperture radar technology, used to acquire high-resolution topographic, soil layer and vegetation distribution information of the region; the radar technologies mentioned include ground radar, airborne radar or satellite radar, used to detect soil and rock features. Optical image analysis technology uses multispectral or hyperspectral images to assess the anchoring effect of vegetation roots on shallow soil, and calculates vegetation coverage and root strength through image classification and analysis techniques. Data fusion includes the fusion of remote sensing images, radar data and optical images, and combines geographic information system technology and machine learning methods to train models using relevant local data in the database, and automatically assesses the landslide disaster risk index of the region.

6. The method according to claim 4, characterized in that, This involves analyzing soil stability and landslide susceptibility using topographic data, calculating soil slope, aspect, and soil moisture factors, and generating landslide risk heat maps by combining historical landslide event data and meteorological data to assess landslide susceptibility in various regions.

7. The method according to claim 4, characterized in that, The risk assessment model described therein is based on machine learning methods such as random forest, support vector machine or convolutional neural network. It is trained with historical landslide data and real-time data, and combined with meteorological data, geological data and historical landslide records in the database to predict the probability of landslide occurrence.

8. The method according to claim 4, characterized in that, The landslide risk assessment results include a probability distribution map or heat map of landslide occurrence, and output the landslide risk level, which includes four levels: low risk, medium risk, high risk, and extremely high risk. The risk assessment results are output after comprehensive analysis of historical landslide data, precipitation data, and meteorological forecast data.

Images