Landslide monitoring method based on remote sensing data

Abstract

The invention discloses a landslide monitoring method based on remote sensing data, and the method comprises the steps: obtaining a high-resolution optical remote sensing image and a synthetic aperture radar image of a research region, carrying out the registration processing of the optical and radar remote sensing images, and obtaining a remote sensing image with an accurate spatial position and an accurate geometric shape; generating a landslide area fusion image with high spectral resolution; the matched high-resolution DEM data is obtained; based on the matched high-resolution DEM data, gradient parameters of the research area are calculated, and a gradient threshold value is determined according to gradient distribution characteristics and prior knowledge; extracting slope direction parameters of the research area by adopting a slope direction algorithm based on grids and vectors, and screening out a landslide-prone area in combination with the slope threshold value; obtaining a landslide risk distribution map; and obtaining the occurrence possibility and danger level of the landslide, and obtaining the landslide condition according to the occurrence possibility and danger level of the landslide. According to the invention, precise identification and dynamic monitoring of landslide disasters are realized.

Description

Technical Field

[0001] The invention belongs to the technical field of landslide monitoring, and in particular relates to a landslide monitoring method based on remote sensing data. Background Art

[0002] Landslide disaster warning is a complex technical issue involving the fusion and analysis of multi-source remote sensing data. The core challenge is how to effectively integrate the advantages of optical and radar remote sensing images to achieve high-precision and timely landslide risk assessment. First, the images obtained by different sensors have geometric distortion and spatial offset. How to ensure the accurate registration of multi-source data is a basic problem. Secondly, optical and radar images have their own characteristics. How to design a reasonable fusion strategy to generate a comprehensive image with high spatial resolution, high spectral resolution and multi-polarization characteristics is the key. Furthermore, the spatial matching of the fused image and DEM data directly affects the accuracy of subsequent terrain parameter extraction. On this basis, how to extract key terrain parameters from massive remote sensing data and determine appropriate threshold conditions to identify potential landslide areas is a technical difficulty. Finally, in the face of dynamically changing surface environments, how to establish a sensitive and reliable landslide warning mechanism, timely capture abnormal changes in terrain parameters and combine other factors to make accurate judgments is a major challenge facing the landslide disaster warning system. This series of technical issues are interrelated and together constitute the core technical contradictions in the field of landslide disaster warning. Summary of the invention

[0003] To solve the above technical problems, the present invention proposes a landslide monitoring method based on remote sensing data, which realizes the accurate identification and dynamic monitoring of landslide disasters, and provides effective technical support for landslide early warning and disaster prevention and mitigation.

[0004] To achieve the above-mentioned object, the present invention provides a landslide monitoring method based on remote sensing data, comprising: obtaining high-resolution optical remote sensing images and synthetic aperture radar images of a study area, performing registration processing on the optical and radar remote sensing images, and obtaining remote sensing images with precise spatial position and accurate geometric shape;

[0005] According to the texture clarity of optical images and the penetration of radar images, pixel-level and feature-level fusion strategies are designed, combined with multi-scale fusion algorithms to generate high-spectral-resolution fused images of landslide areas.

[0006] The high spectral resolution fused image of the landslide area is registered and cropped, and the region of interest that is consistent with the DEM spatial range of the study area is extracted to obtain matched high-resolution DEM data;

[0007] Based on the matched high-resolution DEM data, the slope parameters of the study area are calculated, and the slope threshold is determined according to the slope distribution characteristics and prior knowledge;

[0008] The aspect algorithm based on raster and vector is used to extract the aspect parameters of the study area, and combined with the slope threshold, the landslide-prone areas are screened out;

[0009] Calculate the curvature parameters of the screened landslide-prone areas, extract the high-risk areas prone to landslides by setting the curvature threshold, and obtain the landslide hazard distribution map;

[0010] New remote sensing image data is acquired in real time to detect changes in high-risk landslide areas and combined with the landslide hazard distribution map to obtain the possibility and hazard level of landslide occurrence, and the landslide situation is obtained based on the possibility and hazard level of landslide occurrence.

[0011] Optional remote sensing images with precise spatial position and accurate geometric shape include:

[0012] Extract characteristic information of each temporal phase image from optical and radar remote sensing images, and complete spatiotemporal registration of remote sensing images of different temporal phases through multi-temporal image feature matching;

[0013] According to the registered optical remote sensing image and synthetic aperture radar remote sensing image, the imaging parameters of the image are obtained by solving the image imaging geometric model, and a unified spatial coordinate system is established in combination with the ground control point information to complete the geometric correction of the remote sensing image;

[0014] Image resampling technology is used to eliminate the geometric distortion of optical remote sensing images and synthetic aperture radar remote sensing images after geometric correction to obtain remote sensing images with precise spatial position and accurate geometric shape.

[0015] Optionally, generate a high spectral resolution fused image of the landslide area including:

[0016] According to the penetration of the radar image, the threshold segmentation method is used to extract the contour of the object, and the segmentation result is optimized by the region growing algorithm to obtain the penetration feature map of the radar image;

[0017] Normalize the pixel brightness value of the optical image and the backscattering coefficient of the penetration feature map of the radar image, and fuse them using IHS transformation to obtain the initial fusion image of the optical image texture and radar image penetration.

[0018] According to the initial fused image, multi-scale features are extracted by wavelet transform, and feature-level fusion is performed by setting fusion weights of different frequency bands to generate a fused image of the landslide area with high spectral resolution.

[0019] Optionally, the matched high-resolution DEM data includes:

[0020] By registering and clipping the fused image of the landslide area with high spectral resolution, and using image clipping technology, extracting the image and DEM subset of the area of ​​interest that are consistent with the spatial range of the study area;

[0021] Resampling and unifying the spatial resolution of the cropped image of the region of interest and the DEM subset, extracting the terrain parameters in the study area using a terrain analysis algorithm, and obtaining a terrain factor layer corresponding to the remote sensing image space;

[0022] The extracted terrain factor layer is overlaid with the image of the area of ​​interest to obtain the correlation between the terrain factor and the spectral and texture characteristics of the objects;

[0023] By utilizing the correlation between terrain factors and the spectrum and texture characteristics of ground objects, matched high-resolution DEM data are obtained.

[0024] Optionally, based on the matched high-resolution DEM data, calculating the slope parameters of the study area includes:

[0025] Obtain high-resolution DEM data after matching in the study area, pre-process the DEM data, remove outliers and noise data, and ensure data quality;

[0026] The maximum value method is used to process DEM data, calculate the maximum elevation difference of each pixel point in its neighborhood, and obtain preliminary slope parameters;

[0027] The preliminary slope parameters are processed by using the finite difference method to calculate the elevation gradient of each pixel point in the horizontal and vertical directions to obtain accurate slope parameters.

[0028] Optionally, screen out landslide-prone areas including:

[0029] According to the geographic information data of the study area, the slope parameters of each raster or vector unit in the study area are extracted;

[0030] Statistical analysis was performed on the extracted aspect parameters to obtain the distribution characteristics and variation patterns of the aspect parameters in the study area.

[0031] Preset a slope threshold and a slope aspect condition, perform cluster analysis on the raster or vector units that simultaneously meet the slope condition and the slope aspect condition, and aggregate the units that are spatially adjacent and meet the combination condition into one area;

[0032] The aggregated areas are classified and the landslide-prone areas are screened out according to the characteristic parameters of the areas.

[0033] Optionally, obtain a landslide hazard distribution map including:

[0034] According to the digital elevation model data of the screened landslide-prone areas, the plane curvature and profile curvature values ​​of each grid are calculated;

[0035] According to the geological conditions of the selected landslide-prone areas and historical landslide data, the danger thresholds of the plane curvature and profile curvature are determined by statistical analysis methods;

[0036] Determine whether the plane curvature and profile curvature of each grid exceeds the threshold. If so, the grid is classified as a high-risk landslide area.

[0037] Obtain the spatial distribution of high-risk landslide areas and generate a distribution map of high-risk landslide areas;

[0038] According to the distribution map of high-risk landslide areas and the topographic and geomorphological characteristics of the study area, a landslide hazard prediction model was established using a machine learning algorithm.

[0039] Through the landslide hazard prediction model, the landslide hazard of each grid in the landslide-prone area is predicted and evaluated to obtain the landslide hazard index;

[0040] According to the landslide hazard index, a hierarchical mapping method is used to obtain the landslide hazard distribution map.

[0041] Optionally, obtain the probability and hazard level of landslide occurrence including:

[0042] The topographic parameters in the landslide area are extracted through image interpretation technology and compared with the preset threshold to determine whether a mutation has occurred;

[0043] If the terrain parameters exceed the preset threshold, the image is further interpreted to obtain vegetation and soil information in the landslide area, and to analyze the sudden changes in terrain parameters and changes in vegetation and soil factors;

[0044] Based on the sudden changes in terrain parameters and changes in vegetation and soil factors, a comprehensive assessment of landslide hazard is conducted to obtain the possibility and hazard level of landslide occurrence.

[0045] Technical effect of the invention: The present invention discloses a landslide monitoring method based on remote sensing data. First, geometric correction and spatial registration are performed on high-resolution optical and radar remote sensing images, and a multi-scale fusion algorithm is used to generate a fused image of the landslide area with high spatial resolution, high spectral resolution and multi-polarization characteristics. Then, based on the matched DEM data, terrain parameters such as slope, slope aspect, curvature, etc. are extracted, and threshold conditions are set to screen potential landslide-prone areas. When the newly acquired remote sensing data detects a sudden change in the terrain parameters of the high-risk area and exceeds the preset threshold, a landslide warning signal is triggered, and the hazard assessment is dynamically updated in combination with the image interpretation results. The present invention realizes the accurate identification and dynamic monitoring of landslide disasters, and provides effective technical support for landslide warning and disaster prevention and mitigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0047] Figure 1 The figure is a flow chart of a landslide monitoring method based on remote sensing data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0048] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0049] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0050] like Figure 1 As shown, this embodiment provides a landslide monitoring method based on remote sensing data, including: obtaining high-resolution optical remote sensing images and synthetic aperture radar images of the study area, performing registration processing on the optical and radar remote sensing images, and obtaining remote sensing images with precise spatial position and accurate geometric shape;

[0051] According to the texture clarity of optical images and the penetration of radar images, pixel-level and feature-level fusion strategies are designed, combined with multi-scale fusion algorithms to generate high-spectral-resolution fused images of landslide areas.

[0052] The high-spectral-resolution fused images of the landslide area were registered and cropped to extract the area of ​​interest that was consistent with the DEM spatial range of the study area, and the matched high-resolution DEM data was obtained.

[0053] Based on the matched high-resolution DEM data, the slope parameters of the study area are calculated, and the slope threshold is determined according to the slope distribution characteristics and prior knowledge;

[0054] The aspect algorithm based on raster and vector was used to extract the aspect parameters of the study area, and combined with the slope threshold, the landslide-prone areas were screened out;

[0055] Calculate the curvature parameters of the screened landslide-prone areas, extract the high-risk areas prone to landslides by setting the curvature threshold, and obtain the landslide hazard distribution map;

[0056] New remote sensing image data is acquired in real time to detect changes in high-risk landslide areas and combined with landslide hazard distribution maps to obtain the possibility and hazard level of landslides, and the landslide situation is obtained based on the possibility and hazard level of landslides.

[0057] Furthermore, obtaining remote sensing images with precise spatial position and accurate geometric shape includes:

[0058] Extract characteristic information of each temporal phase image from optical and radar remote sensing images, and complete spatiotemporal registration of remote sensing images of different temporal phases through multi-temporal image feature matching;

[0059] According to the registered optical remote sensing image and synthetic aperture radar remote sensing image, the imaging parameters of the image are obtained by solving the image imaging geometric model, and a unified spatial coordinate system is established in combination with the ground control point information to complete the geometric correction of the remote sensing image;

[0060] Image resampling technology is used to eliminate the geometric distortion of optical remote sensing images and synthetic aperture radar remote sensing images after geometric correction to obtain remote sensing images with precise spatial position and accurate geometric shape.

[0061] Specifically, according to the registered optical and radar remote sensing images, the RFM and Range-Doppler models are used to perform imaging geometry modeling, and the model is optimized and solved with the help of elevation control points to establish a unified spatial coordinate system, and the remote sensing images are geometrically corrected. The average deviation between the corrected images and the actual ground position is less than 1 meter. The corrected remote sensing images are resampled using bilinear interpolation resampling technology to remove geometric distortion and make the spatial position of the image accurately correspond to the actual ground objects. On this basis, the spectrum, texture, geometry and other information of the optical and radar remote sensing images are fused using image fusion algorithms such as IHS and wavelet transform to generate remote sensing images with both high spatial resolution and high spectral resolution. The fused remote sensing images are orthorectified, and DEM data is used to remove the influence of terrain undulations so that the geometric projection of the image is consistent with the map coordinate system. The generated orthorectified remote sensing images can be directly used for GIS analysis and mapping.

[0062] Furthermore, the generation of high spectral resolution fused images of landslide areas includes:

[0063] According to the penetration of radar images, the threshold segmentation method is used to extract the contour of the object, and the segmentation result is optimized by the region growing algorithm to obtain the penetration feature map of the radar image;

[0064] Normalize the pixel brightness value of the optical image and the backscattering coefficient of the penetration feature map of the radar image, and fuse them using IHS transformation to obtain the initial fusion image of the optical image texture and radar image penetration.

[0065] Based on the initial fused image, wavelet transform is used to extract multi-scale features. By setting the fusion weights of different frequency bands, feature-level fusion is performed to generate a fused image of the landslide area with high spectral resolution.

[0066] Specifically, according to the penetration of the radar image, the threshold segmentation method is used to extract the contour of the object, and the segmentation result is optimized by the region growing algorithm to obtain the penetration feature map of the radar image; the pixel brightness value of the optical image and the backscattering coefficient of the radar image are normalized, and the two are fused by IHS transformation to obtain an initial fused image that retains the texture of the optical image and the penetration of the radar image; for the initial fused image, wavelet transform is used to extract multi-scale features, and by setting the fusion weights of different frequency bands, feature-level fusion is achieved to generate a fused image of the landslide area with multi-scale features. ; If the spatial resolution of the fused image of the landslide area does not meet the requirements, the interpolation algorithm is used to improve the spatial resolution; if the spectral resolution of the fused image of the landslide area is insufficient, the hyperspectral data is introduced for spectral enhancement; according to the polarization scattering characteristics of the landslide area objects, the polarization channel combination is selected to generate the multi-polarization characteristics of the fused image of the landslide area; through the combination of visual interpretation and quantitative evaluation, it is judged whether the fused image of the landslide area achieves the expected effect. If it is not ideal, the wavelet transform step is returned to adjust the fusion strategy until the target landslide area fused image with high spatial, hyperspectral and multi-polarization characteristics is generated.

[0067] Furthermore, the matched high-resolution DEM data includes:

[0068] By registering and cropping the fused images of the landslide area with high spectral resolution, and using image cropping technology, the images and DEM subsets of the area of ​​interest that are consistent with the spatial range of the study area are extracted;

[0069] Resample and unify the spatial resolution of the cropped area of ​​interest images and DEM subsets, extract the terrain parameters in the study area using terrain analysis algorithms, and obtain the terrain factor layer corresponding to the remote sensing image space;

[0070] The extracted terrain factor layer is overlaid with the image of the area of ​​interest to obtain the correlation between the terrain factor and the spectral and texture characteristics of the objects;

[0071] By utilizing the correlation between terrain factors and the spectrum and texture characteristics of ground objects, matched high-resolution DEM data are obtained.

[0072] Specifically, in order to obtain high-resolution DEM data in the study area, this embodiment adopts aerial photogrammetry technology, uses a drone equipped with a high-precision camera, and takes aerial photos with 80% heading overlap and 60% lateral overlap to obtain 1-meter resolution orthophotos, and generates 1-meter resolution DEMs through stereo matching. For the fused 5-meter resolution remote sensing image, the SIFT algorithm is used to extract feature points, and the nearest neighbor distance ratio matching strategy is used to achieve accurate registration with the reference DEM, and the root mean square error after registration is controlled within 5 pixels. Through mask clipping technology, the image of the area of ​​interest and the DEM subset within the scope of the study area are extracted and resampled to 5-meter resolution. Using terrain analysis algorithms such as slope, aspect, curvature, etc., the terrain parameters in the study area are extracted to generate a terrain factor layer that is consistent with the image space of the area of ​​interest. Through correlation analysis, it is found that the vegetation index is negatively correlated with the slope, with a correlation coefficient of -67; the distribution of bare rocks is positively correlated with the aspect, with a correlation coefficient of 58. Finally, the images, DEM and terrain parameter layers of the area of ​​interest were superimposed to construct a geographic dataset containing multi-source information such as spectrum, texture, elevation and terrain, providing important data support for subsequent landslide identification and risk assessment.

[0073] Furthermore, based on the matched high-resolution DEM data, the slope parameters of the study area are calculated, including:

[0074] Obtain high-resolution DEM data after matching in the study area, pre-process the DEM data, remove outliers and noise data, and ensure data quality;

[0075] The maximum value method is used to process DEM data, calculate the maximum elevation difference of each pixel point in its neighborhood, and obtain preliminary slope parameters;

[0076] The finite difference method is used to process the preliminary slope parameters, calculate the elevation gradient of each pixel point in the horizontal and vertical directions, and obtain accurate slope parameters.

[0077] Specifically, the DEM data is preprocessed to remove outliers and noise data to ensure data quality. The maximum value method is used to process the DEM data, and the maximum elevation difference of each pixel point in its neighborhood is calculated to obtain preliminary slope parameters. The finite difference method is used to process the DEM data, and the elevation gradient of each pixel point in the horizontal and vertical directions is calculated to obtain accurate slope parameters. The slope parameters calculated by the maximum value method and the finite difference method are fused, and the advantages and disadvantages of the two methods are comprehensively considered to obtain more accurate and reliable slope distribution information. According to the topographic and geomorphic characteristics of the study area and the known landslide distribution, the slope distribution law of the potential landslide area is summarized to form prior knowledge. The slope distribution information is compared and analyzed with the prior knowledge to determine the slope threshold conditions of the potential landslide area, that is, the area that meets certain slope conditions is determined as a potential landslide area. The slope threshold conditions of the potential landslide area are verified and optimized, and the threshold conditions are continuously adjusted through field investigations and comparisons with historical landslide data to improve the accuracy of the determination of the potential landslide area.

[0078] Furthermore, the areas prone to landslides are screened out, including:

[0079] According to the geographic information data of the study area, the slope parameters of each raster or vector unit in the study area are extracted;

[0080] Statistical analysis was performed on the extracted aspect parameters to obtain the distribution characteristics and variation patterns of the aspect parameters in the study area.

[0081] Preset the slope threshold and aspect conditions, perform cluster analysis on the raster or vector units that meet both the slope and aspect conditions, and aggregate the spatially adjacent units that meet the combined conditions into one area;

[0082] The aggregated areas are classified and the landslide-prone areas are screened out according to the characteristic parameters of the areas.

[0083] Specifically, the DEM data and land use data of the study area were used, and the grid-based D8 algorithm was used to extract the slope parameters of each 30m×30m grid cell. Statistical analysis showed that the slope directions in the study area were mainly concentrated in the south and southeast, accounting for about 65% of the total area. According to the existing landslide distribution data, the slope threshold was set to 20°, and the slope of each grid cell was judged. Those exceeding the threshold were marked as meeting the slope condition. In the grid cells that met the slope condition, it was further judged whether the slope direction was between the south and the southeast. If so, it was marked as meeting the slope condition. For grid cells that met both conditions, the DBSCAN clustering algorithm was used for spatial aggregation, and the clustered areas were used as candidate landslide-prone areas. Using the known 100 landslide point data, the characteristic parameters such as slope, slope direction, elevation, and lithology of each landslide point were extracted as training samples, and the support vector machine algorithm was used to construct a landslide-prone area discrimination model. The characteristic parameters of the candidate areas are input into the discriminant model for classification, and the distribution of landslide-prone areas in the study area is obtained. Through the visualization function of ArcGIS software, a distribution map of landslide-prone areas is generated, and compared with known landslide points for verification. It is found that 85% of the landslide points are located in the predicted prone area, which proves the effectiveness of the prediction results and can provide an important reference for local landslide disaster prevention and control. Further, the landslide hazard distribution map is obtained, including:

[0084] According to the digital elevation model data of the screened landslide-prone areas, the plane curvature and profile curvature values ​​of each grid are calculated;

[0085] According to the geological conditions of the selected landslide-prone areas and historical landslide data, the danger thresholds of the plane curvature and profile curvature are determined by statistical analysis methods;

[0086] Determine whether the plane curvature and profile curvature of each grid exceeds the threshold. If so, the grid is classified as a high-risk landslide area.

[0087] Obtain the spatial distribution of high-risk landslide areas and generate a distribution map of high-risk landslide areas;

[0088] According to the distribution map of high-risk landslide areas and the topographic and geomorphological characteristics of the study area, a landslide hazard prediction model was established using a machine learning algorithm.

[0089] Through the landslide hazard prediction model, the landslide hazard of each grid in the landslide-prone area is predicted and evaluated to obtain the landslide hazard index;

[0090] According to the landslide hazard index, a hierarchical mapping method is used to obtain the landslide hazard distribution map.

[0091] Specifically, according to the digital elevation model data of the study area, the curvature calculation tool of ArcGIS software was used to calculate the plane curvature and profile curvature values ​​of each 30m×30m grid. By statistically analyzing the plane curvature and profile curvature of 527 landslide points that have occurred in the study area, it was determined that the grids with a plane curvature greater than 15 or a profile curvature greater than 2 are high-risk areas for landslides. Using the raster calculator, the plane curvature and profile curvature of each grid are threshold judged. The grids exceeding the threshold are assigned a value of 1, indicating that the grid is a high-risk area for landslides, and the remaining grids are assigned a value of 0. The raster to vector tool is used to convert the raster of the high-risk area of ​​landslides into a vector surface, obtain the spatial distribution of the high-risk area of ​​landslides, and generate a distribution map of the high-risk area of ​​landslides.

[0092] Furthermore, the possibility and danger level of landslide occurrence are obtained including:

[0093] The topographic parameters in the landslide area are extracted through image interpretation technology and compared with the preset threshold to determine whether a mutation has occurred;

[0094] If the terrain parameters exceed the preset threshold, the image is further interpreted to obtain vegetation and soil information in the landslide area, and to analyze the sudden changes in terrain parameters and changes in vegetation and soil factors;

[0095] Based on the sudden changes in terrain parameters and changes in vegetation and soil factors, a comprehensive assessment of landslide hazard is conducted to obtain the possibility and hazard level of landslide occurrence.

[0096] Specifically, in ArcGIS, the landslide high-risk area distribution map is used, combined with the terrain and geomorphological features of the study area such as terrain relief and terrain roughness, and the support vector machine algorithm is used to establish a landslide hazard prediction model. The model is applied to each grid in the study area to obtain the landslide hazard index of each grid, with a value range of 0 to 1. The landslide hazard index is divided into 5 levels using the natural breakpoint method, representing very low, low, medium, high, and very high hazard, and a landslide hazard distribution map of the study area is generated to provide a reference for landslide prevention and risk management. According to the sudden change of terrain parameters and the changes in vegetation and soil factors, the support vector machine algorithm is used to comprehensively evaluate the landslide hazard and obtain the possibility and hazard level of landslide occurrence. If the assessment result shows that the landslide hazard is high, an early warning signal is triggered, and relevant departments and personnel are notified in time through SMS, application push, etc., prompting them to take preventive measures. In the new image data obtained later, the changes of various parameters and factors in the landslide area are continuously monitored, and the incremental learning method is used to dynamically update and optimize the hazard assessment model. Based on the dynamically updated assessment results, the level and frequency of early warning signals are adjusted to provide continuous and effective decision-making support for landslide prevention and control.

[0097] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A landslide monitoring method based on remote sensing data, characterized in that: include: Obtain high-resolution optical remote sensing images and synthetic aperture radar images of the study area, perform registration processing on the optical and radar remote sensing images, and obtain remote sensing images with precise spatial position and accurate geometric shape; According to the texture clarity of optical images and the penetration of radar images, pixel-level and feature-level fusion strategies are designed, combined with multi-scale fusion algorithms to generate high-spectral-resolution fused images of landslide areas. The high spectral resolution fused image of the landslide area is registered and cropped, and the region of interest that is consistent with the DEM spatial range of the study area is extracted to obtain matched high-resolution DEM data; Based on the matched high-resolution DEM data, the slope parameters of the study area are calculated, and the slope threshold is determined according to the slope distribution characteristics and prior knowledge; The aspect algorithm based on raster and vector is used to extract the aspect parameters of the study area, and combined with the slope threshold, the landslide-prone areas are screened out; Calculate the curvature parameters of the screened landslide-prone areas, extract the high-risk areas prone to landslides by setting the curvature threshold, and obtain the landslide hazard distribution map; New remote sensing image data is acquired in real time to detect changes in high-risk landslide areas and combined with the landslide hazard distribution map to obtain the possibility and hazard level of landslide occurrence, and the landslide situation is obtained based on the possibility and hazard level of landslide occurrence.

2. The landslide monitoring method based on remote sensing data according to claim 1, characterized in that: Remote sensing images with precise spatial position and accurate geometric shape are obtained by: Extract characteristic information of each temporal phase image from optical and radar remote sensing images, and complete spatiotemporal registration of remote sensing images of different temporal phases through multi-temporal image feature matching; According to the registered optical remote sensing image and synthetic aperture radar remote sensing image, the imaging parameters of the image are obtained by solving the image imaging geometric model, and a unified spatial coordinate system is established in combination with the ground control point information to complete the geometric correction of the remote sensing image; Image resampling technology is used to eliminate the geometric distortion of optical remote sensing images and synthetic aperture radar remote sensing images after geometric correction to obtain remote sensing images with precise spatial position and accurate geometric shape.

3. The landslide monitoring method based on remote sensing data according to claim 1, characterized in that: Generating high spectral resolution fused images of landslide areas includes: According to the penetration of the radar image, the threshold segmentation method is used to extract the contour of the object, and the segmentation result is optimized by the region growing algorithm to obtain the penetration feature map of the radar image; Normalize the pixel brightness value of the optical image and the backscattering coefficient of the penetration feature map of the radar image, and fuse them using IHS transformation to obtain the initial fusion image of the optical image texture and radar image penetration. According to the initial fused image, multi-scale features are extracted by wavelet transform, and feature-level fusion is performed by setting fusion weights of different frequency bands to generate a fused image of the landslide area with high spectral resolution.

4. The landslide monitoring method based on remote sensing data according to claim 1, characterized in that: The matched high-resolution DEM data includes: By registering and clipping the fused image of the landslide area with high spectral resolution, and using image clipping technology, extracting the image and DEM subset of the area of ​​interest that are consistent with the spatial range of the study area; Resampling and unifying the spatial resolution of the cropped image of the region of interest and the DEM subset, extracting the terrain parameters in the study area using a terrain analysis algorithm, and obtaining a terrain factor layer corresponding to the remote sensing image space; The extracted terrain factor layer is overlaid with the image of the area of ​​interest to obtain the correlation between the terrain factor and the spectral and texture characteristics of the objects; By utilizing the correlation between terrain factors and the spectrum and texture characteristics of ground objects, matched high-resolution DEM data are obtained.

5. The landslide monitoring method based on remote sensing data according to claim 1, characterized in that: Based on the matched high-resolution DEM data, the slope parameters of the study area are calculated including: Obtain high-resolution DEM data after matching in the study area, pre-process the DEM data, remove outliers and noise data, and ensure data quality; The maximum value method is used to process DEM data, calculate the maximum elevation difference of each pixel point in its neighborhood, and obtain preliminary slope parameters; The preliminary slope parameters are processed by using the finite difference method to calculate the elevation gradient of each pixel point in the horizontal and vertical directions to obtain accurate slope parameters.

6. The landslide monitoring method based on remote sensing data according to claim 1, characterized in that: The areas that are prone to landslides include: According to the geographic information data of the study area, the slope parameters of each raster or vector unit in the study area are extracted; Statistical analysis was performed on the extracted aspect parameters to obtain the distribution characteristics and variation patterns of the aspect parameters in the study area. Preset a slope threshold and a slope aspect condition, perform cluster analysis on the raster or vector units that simultaneously meet the slope condition and the slope aspect condition, and aggregate the units that are spatially adjacent and meet the combination condition into one area; The aggregated areas are classified and the landslide-prone areas are screened out according to the characteristic parameters of the areas.

7. The landslide monitoring method based on remote sensing data according to claim 1, characterized in that: Obtaining landslide hazard distribution map includes: According to the digital elevation model data of the screened landslide-prone areas, the plane curvature and profile curvature values ​​of each grid are calculated; According to the geological conditions of the selected landslide-prone areas and historical landslide data, the danger thresholds of the plane curvature and profile curvature are determined by statistical analysis methods; Determine whether the plane curvature and profile curvature of each grid exceeds the threshold. If so, the grid is classified as a high-risk landslide area. Obtain the spatial distribution of high-risk landslide areas and generate a distribution map of high-risk landslide areas; According to the distribution map of high-risk landslide areas and the topographic and geomorphological characteristics of the study area, a landslide hazard prediction model was established using a machine learning algorithm. Through the landslide hazard prediction model, the landslide hazard of each grid in the landslide-prone area is predicted and evaluated to obtain the landslide hazard index; According to the landslide hazard index, a hierarchical mapping method is used to obtain the landslide hazard distribution map.

8. The landslide monitoring method based on remote sensing data according to claim 1, characterized in that: Obtaining the probability and hazard level of landslide occurrence includes: The topographic parameters in the landslide area are extracted through image interpretation technology and compared with the preset threshold to determine whether a mutation has occurred; If the terrain parameters exceed the preset threshold, the image is further interpreted to obtain vegetation and soil information in the landslide area, and to analyze the sudden changes in terrain parameters and changes in vegetation and soil factors; Based on the sudden changes in terrain parameters and changes in vegetation and soil factors, a comprehensive assessment of landslide hazard is conducted to obtain the possibility and hazard level of landslide occurrence.

Images