A geological disaster monitoring method, device, medium and product

Abstract

The application discloses a geological disaster monitoring method and device, medium and product, and relates to the technical field of geological monitoring. The method comprises the following steps: determining a micro-deformation parameter of a to-be-monitored region according to remote sensing observation data corresponding to a slope body of the to-be-monitored region; determining a macro-deformation parameter of the to-be-monitored region according to optical remote sensing data and terrain data of the to-be-monitored region; and determining a landslide remote sensing geomechanical deformation type of the to-be-monitored region according to a material composition, a movement mode, a slope body structure, the micro-deformation parameter and the macro-deformation parameter of the to-be-monitored region. The application improves the accuracy of geological disaster monitoring.

Description

Technical Field

[0001] This invention relates to the field of geological monitoring technology, and in particular to a geological disaster monitoring method, equipment, medium and product. Background Technology

[0002] Large-scale geological disasters are often located in high-altitude, difficult-to-access, and sparsely populated areas. Traditional on-site investigations and mass monitoring and prevention methods are insufficient to study their causal patterns and deformation mechanisms. It is necessary to use modern high-precision Earth observation technologies (such as high-resolution optical remote sensing and interferometric synthetic aperture radar (InSAR)) to study the deformation process of landslides over long periods of time, and to correlate the deformation characteristics with their causal patterns, thereby assisting in the monitoring and early warning of high-altitude, remote landslides.

[0003] Macroscopic deformation of geological hazard risks mainly manifests as dynamic changes in topography and geomorphology, as well as dynamic changes in significant cracks and small-scale preceding collapses on the hazard body. These significant deformations are incoherent in SAR images and therefore cannot be dynamically monitored using InSAR technology. However, the development of macroscopic deformation is, to some extent, an external manifestation of the accumulation of microscopic deformation to a certain degree. Therefore, tracking macroscopic deformation is key to revealing the coupling process and mechanism between macroscopic and microscopic deformation. Current advancements in high spatial resolution and high temporal resolution optical remote sensing technologies have provided an opportunity to track the macroscopic deformation characteristics of geological hazard risks, enabling dynamic tracking and analysis of features such as the hazard-inducing background, topography, and significant cracks, including monitoring and extracting geomorphic unit boundaries and dynamically monitoring slope cracks.

[0004] InSAR technology is a hot topic in microscopic deformation detection and precise measurement applications, and is also a commonly used technology for geological hazard detection and early warning. With the rapid development of computer hardware and software technology, InSAR technology has continued to innovate, giving rise to methods such as D-InSAR, PSI, SBAS, and MAI, which have achieved significant results in removing atmospheric effects and improving measurement accuracy. However, many key problems still need to be solved, such as uncertainties caused by decoherence, atmospheric delay, and orbital errors, and insensitivity to north-south deformation. In order to change the large amount of phase unwrapping calculations in InSAR algorithms and the inability of D-InSAR to obtain large-scale deformations, Michel et al. first proposed the pixel offset tracking (POT) technique in 1999. This method does not require unwrapping and is not affected by image coherence, and can obtain good deformation information. It has good applicability and reliability in deformation / displacement monitoring of earthquakes, glaciers, and landslides. POT (Positive Occlusion) technology is divided into coherence tracking and intensity tracking. The latter uses grayscale information from optical images to register SAR images, and can also perform sub-pixel-level registration for multiple optical images. Therefore, it can be used for deformation analysis of SAR images as well as horizontal deformation analysis of multiple optical images. Thus, POT-based optical image deformation analysis can compensate for the insensitivity to north-south deformation in radar image deformation analysis based on InSAR technology.

[0005] Optical remote sensing images can effectively identify areas with obvious signs of deformation, but they are easily affected by cloud cover and vegetation cover. When deformation is not obvious in the early stages, optical images may not reflect the changes clearly. InSAR technology can effectively identify large areas that are slowly deforming, but this technology is easily limited by factors such as observation angle, vegetation cover, water vapor, and data processing technology. Therefore, the accuracy of geological disaster monitoring needs to be improved. Summary of the Invention

[0006] The purpose of this invention is to provide a geological disaster monitoring method, equipment, medium, and product to improve the accuracy of geological disaster monitoring.

[0007] To achieve the above objectives, the present invention provides the following solution:

[0008] A method for monitoring geological hazards, comprising:

[0009] Based on the remote sensing observation data corresponding to the slope body of the area to be monitored, the micro deformation parameters of the area to be monitored are determined;

[0010] Based on the optical remote sensing data and topographic data of the area to be monitored, determine the macroscopic deformation parameters of the area to be monitored;

[0011] Based on the material composition, movement mode, slope structure, micro-deformation parameters, and macro-deformation parameters of the area to be monitored, the landslide remote sensing geomechanical deformation type of the area to be monitored is determined.

[0012] Optionally, based on remote sensing data corresponding to the slope body of the area to be monitored, the micro-deformation parameters of the area to be monitored are determined, specifically including:

[0013] If the slope of the area to be monitored is an east-west slope, then the temporal micro-deformation process of the landslide body is inverted using InSAR technology based on the time-series radar satellite imagery data, and the micro-deformation data is determined based on the temporal micro-deformation process. The micro-deformation data includes the micro-deformation level and the deformation region.

[0014] If the slope of the area to be monitored is a north-south slope, then the micro-deformation data of the area to be monitored is obtained by using POT technology based on optical image time series data.

[0015] Optionally, micro-deformation data are determined based on the time-series micro-deformation process, specifically including:

[0016] The average deformation rate and cumulative deformation are extracted from the time-series micro-deformation process;

[0017] The average deformation rate is determined based on the deformation rate and the cumulative deformation.

[0018] The micro-deformation level is determined based on the average deformation rate.

[0019] Optionally, when the average deformation rate is greater than 100 mm / a, it is considered a large microdeformation; when the average deformation rate is greater than 50 mm / a and less than or equal to 100 mm / a, it is considered a medium microdeformation; and when the average deformation rate is less than or equal to 50 mm / a, it is considered a small microdeformation.

[0020] Optionally, based on the optical remote sensing data and topographic data of the area to be monitored, the macroscopic deformation parameters of the area to be monitored are determined, specifically including:

[0021] Based on optical remote sensing time-series data, the random forest classification method is used to extract local landslide areas in the area to be monitored;

[0022] Based on the topographic data and optical remote sensing time-series data, cracks and gullies in the area to be monitored are extracted using edge detection and random forest classification; the topographic data includes multi-period digital elevation models.

[0023] The macroscopic deformation level in the area to be monitored is determined based on the ratio of the area of ​​the local collapse zone to the area to be monitored.

[0024] Optionally, when the area ratio is greater than 25%, it is a large macroscopic deformation; when the area ratio is less than or equal to 25% but greater than 10%, it is a medium macroscopic deformation; and when the area ratio is less than or equal to 10%, it is a small macroscopic deformation.

[0025] Optionally, the landslide remote sensing geomechanical deformation types include soil shoveling landslides, soil traction landslides, rock reverse-dip landslides, blocky rock mass landslides, rock flat-push landslides, and rock bedding-parallel landslides.

[0026] In the aforementioned soil-driven landslide, the material composition is soil, the movement mode is shoveling, the slope structure is soil, the micro-deformation parameters are large-scale micro-deformation or medium-scale micro-deformation in the upper part of the landslide, the macro-deformation parameters include medium-scale or small-scale landslides in the local collapse area, and the macro-deformation parameters also include cracks with a length of more than 10 meters.

[0027] In the aforementioned soil-induced landslide, the material composition is soil, the movement mode is traction sliding, the slope structure is soil, the micro-deformation parameters are large-scale or medium-scale micro-deformation in the lower part of the landslide, the macro-deformation parameters include whether the local collapse area is a medium-scale or small-scale collapse, and the macro-deformation parameters also include crack lengths exceeding 10 meters; wherein, the local collapse area is located in the lower part of the landslide body;

[0028] In the aforementioned rock-based reverse-dip landslide, the material composition is rock, the movement mode is rotational sliding, the slope structure is a reverse-dip slope, the micro-deformation parameters are large-scale or medium-scale micro-deformation in the lower part of the landslide, and the macro-deformation parameters include large-scale or medium-scale landslides in the local collapse area, and the macro-deformation parameters also include the presence of cracks with a length of more than 10 meters and a width of more than 1 meter; wherein, the local collapse area is located in the middle region of the landslide body;

[0029] In the blocky rock landslide, the material is composed of rock, the movement is rotational sliding, the slope structure is a blocky slope, the micro-deformation parameters are large-scale micro-deformation or medium-scale micro-deformation in the upper part of the landslide, the macro-deformation parameters include large-scale or medium-scale landslides in the local collapse area, and the macro-deformation parameters also include cracks with a length of more than 10 meters; wherein, the local collapse area is located in the lower part of the landslide body;

[0030] In the rock-type flat-push landslide, the material composition is rock, the movement mode is flat-push, the slope structure is nearly horizontal, the micro-deformation parameters are none, and the macro-deformation parameters include crack widths of more than 10 meters.

[0031] In the aforementioned bedding rock landslide, the material composition is rock, the movement mode is planar sliding, the slope structure is a bedding slope, the micro-deformation parameters are none, and the macro-deformation parameters include local collapse areas being small collapses, and macro-deformation parameters also including crack lengths exceeding 10 meters; wherein, the local collapse areas are located in the lower part of the landslide body.

[0032] A computer device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the geological disaster monitoring method.

[0033] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the geological disaster monitoring method described above.

[0034] A computer program product includes a computer program that, when executed by a processor, implements the steps of the geological hazard monitoring method described above.

[0035] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0036] This invention determines the micro-deformation parameters and macro-deformation parameters of the area to be monitored based on multi-source data, and determines the landslide remote sensing geomechanical deformation type of the area to be monitored based on the micro-deformation parameters and macro-deformation parameters. The accuracy of landslide monitoring and early warning can be improved based on the landslide remote sensing geomechanical deformation type. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a schematic diagram of a geological disaster monitoring method provided in an embodiment of the present invention;

[0039] Figure 2 This is a schematic diagram of the landslide remote sensing geomechanical deformation model construction process provided in an embodiment of the present invention;

[0040] Figure 3 This is a schematic diagram of the landslide micro-deformation acquisition process based on radar image data provided in an embodiment of the present invention;

[0041] Figure 4 This is a schematic diagram of the landslide deformation acquisition process based on Gaofen-2 optical remote sensing data provided in an embodiment of the present invention;

[0042] Figure 5 A schematic diagram showing the east-west deformation results of the Baige landslide from 2015 to 2018 under different windows based on POT technology, provided for an embodiment of the present invention;

[0043] Figure 6 This is a schematic diagram of the process for extracting macroscopic deformation of landslides using the random forest method provided in an embodiment of the present invention;

[0044] Figure 7 A schematic diagram of the landslide boundary provided in an embodiment of the present invention;

[0045] Figure 8 This is a schematic diagram of a local landslide sample provided in an embodiment of the present invention;

[0046] Figure 9 This is a schematic diagram of a local landslide spectral sample provided in an embodiment of the present invention;

[0047] Figure 10 This is a schematic diagram of vegetation samples provided in an embodiment of the present invention;

[0048] Figure 11 This is a schematic diagram of a bare land sample provided in an embodiment of the present invention;

[0049] Figure 12 A schematic diagram of a water sample provided in an embodiment of the present invention;

[0050] Figure 13 This is a schematic diagram of a building sample provided in an embodiment of the present invention;

[0051] Figure 14 This is a schematic diagram of vegetation spectral samples provided in an embodiment of the present invention;

[0052] Figure 15 This is a schematic diagram of bare ground spectral samples provided in an embodiment of the present invention;

[0053] Figure 16 A schematic diagram of a water spectral sample provided in an embodiment of the present invention;

[0054] Figure 17 This is a schematic diagram of building spectral samples provided in an embodiment of the present invention;

[0055] Figure 18 This is a schematic diagram of the results of extracting local collapse and landslide information of the Baige landslide provided in an embodiment of the present invention;

[0056] Figure 19 This is a diagram of the internal structure of a computer device. Detailed Implementation

[0057] 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.

[0058] The purpose of this invention is to provide a geological disaster monitoring method, equipment, medium, and product to improve the accuracy of geological disaster monitoring.

[0059] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0060] Example 1

[0061] This embodiment, supported by long-term sequence spaceborne high-resolution optical and radar imagery data, utilizes temporal interferometric synthetic aperture radar (InSAR) analysis and pixel offset tracking (POT) analysis techniques to dynamically acquire microscopic surface deformation parameters of landslide hazards. Based on landslide formation models, long-term sequence spaceborne high-resolution optical imagery, and digital terrain model (DEM) data, it uses deep learning information extraction algorithms and geographic information spatial analysis and modeling techniques to dynamically acquire macroscopic deformation characteristics of landslides, including the changes in disaster-prone background, disaster-prone micro-topography, and significant cracks. Based on the spatiotemporal matching model of landslide micro- and macroscopic deformation parameters, a landslide remote sensing deformation model is constructed. Based on existing... This invention presents 19 landslide geomechanical models (constructed from slope material composition, movement patterns, and slope structure, using traditional methods, specifically modeling based on information from field surveys). From a remote sensing perspective, it analyzes and summarizes the surface deformation trajectories of these 19 traditional geological models, explores the mapping relationship between landslide remote sensing deformation patterns and landslide geomechanical models, and establishes the geomechanical deformation types of landslides from a remote sensing perspective. This invention utilizes remote sensing information acquisition to solve the mechanical deformation patterns of remote, high-altitude landslides that are inaccessible to humans. It achieves accurate data acquisition in early warning and forecasting of remote, high-altitude landslides while saving manpower and resources, and assists in the research of the mechanism of mechanical deformation patterns in remote, high-altitude landslides.

[0062] Landslides typically undergo a long and slow deformation and evolution process before collapsing as a whole. From a geomechanical perspective, landslides with different formation models experience varying degrees of concentrated stress (such as tensile, compressive, and shear stress) at different locations during their deformation and evolution. Corresponding to these different stress concentration areas, corresponding regions within the landslide body undergo deformation commensurate with their mechanical properties. As deformation accumulates, cracks appear, the deformation slowly expands, and eventually evolves into a localized collapse, altering the slope's geomorphological characteristics. The slow, continuous microscopic deformation of the slope, the continuous macroscopic expansion / increase in cracks, and the emergence of localized collapse characteristics are related to the landslide's formation and its geomechanical model, and are key indicators for landslide remote sensing identification and detection. Therefore, various remote sensing observation methods can be used to obtain spatiotemporal deformation parameters at different scales during different deformation processes of landslides. Based on the combination patterns of spatiotemporal deformation parameters and their correspondence with the landslide geomechanical evolution process, landslide remote sensing geomechanical deformation types can be constructed, solving the problem of difficulty in obtaining deformation parameters and studying geological models of high-altitude, remote landslides.

[0063] The technical approach of this embodiment is as follows: First, based on 19 existing landslide geomechanical models, existing landslide data are classified; second, based on microwave remote sensing data and optical remote sensing data, temporal interferometric synthetic aperture radar (InSAR) analysis technology and pixel offset tracking (POT) technology are used to obtain the microscopic deformation characteristics of various landslides; simultaneously, based on optical remote sensing data and topographic data, various automatic information extraction and analysis methods are used to obtain the macroscopic deformation information of various landslides, and geographic information spatial analysis and modeling are used to reveal the macroscopic deformation characteristics of different landslide types; then, based on the spatiotemporal matching pattern of microscopic deformation parameters and macroscopic deformation parameters of different types of landslides, the spatial matching pattern of microscopic-macroscopic deformation parameters of various types of landslides is analyzed from the perspective of surface deformation signs during the movement process of various types of landslides, constructing landslide remote sensing geomechanical deformation types, solving the problem of geomechanical deformation patterns of remote high-altitude landslides, and the specific technical route is as follows. Figure 2 As shown.

[0064] Based on slope deformation and failure models, and focusing on key factors controlling and influencing landslide formation, this study analyzes slope movement patterns and material composition (rock and soil). Through the analysis of landslide formation conditions and basic deformation and failure laws, landslide disasters in western mountainous areas are categorized into 19 causal models:

[0065] Tilting: ① Block tilting, ② Shallow tilting (rock, soil), ③ Compression-tilting, and ④ Deep tilting.

[0066] Slip: Divided into a) Rotational slip: ⑤ Creep-tensile cracking (soil), ⑥ Creep-tensile cracking-shear, ⑦ Compression-tensile cracking-shear, ⑧ Collapse-tensile cracking-shear, ⑨ Slip-shear; b) Planar slip: ⑩ Slip-tensile cracking (soil). Bedding slip-tension cracking Rotational slip-tear wedge slip, Horizontal sliding, See the tendency of slip-shear, c. Irregular slip: Stepped slip Slip-supported arch-shear (soil) Slide-bending-shear. This scheme considers slope movement patterns, material composition, and key disaster-causing factors, but the classification is too detailed to identify every type of landslide from a remote sensing perspective.

[0067] Based on 19 existing traditional geomechanical models, all collected landslide data were classified, and each type of landslide will be used for the next step of calculating micro-optical deformation parameters and macro-optical deformation parameters and conducting statistical analysis.

[0068] like Figure 1 As shown, a geological disaster monitoring method in this embodiment includes the following steps.

[0069] Step 101: Determine the micro-deformation parameters of the area to be monitored based on the remote sensing observation data corresponding to the slope body of the area to be monitored.

[0070] Step 102: Determine the macroscopic deformation parameters of the area to be monitored based on the optical remote sensing data and topographic data of the area to be monitored.

[0071] Step 103: Determine the landslide remote sensing geomechanical deformation type of the area to be monitored based on the material composition, movement mode, slope structure, micro-deformation parameters, and macro-deformation parameters of the area to be monitored.

[0072] Step 101 specifically includes:

[0073] If the slope of the area to be monitored is an east-west oriented slope, then the temporal micro-deformation process of the landslide body is inverted using InSAR technology based on the time-series radar satellite imagery data. The micro-deformation data is determined based on the temporal micro-deformation process, and the micro-deformation data includes the micro-deformation level and the deformation region.

[0074] Micro-deformation data is determined based on the time-series micro-deformation process, specifically including:

[0075] The average deformation rate and cumulative deformation amount are extracted from the time-series micro-deformation process.

[0076] The micro-deformation level is determined based on the average deformation rate.

[0077] The process of obtaining landslide micro-deformation parameters using InSAR technology is as follows: Figure 3 As shown, the specific steps include the following.

[0078] 1) Image acquisition: Collect long-term series of radar satellite remote sensing images and digital terrain (DEM) data.

[0079] 2) Generate interference pairs: Based on the time series t0, t1…t n A total of N+1 Single Look Complex (SLC) images were acquired. One of these images was selected as the master image and registered with the other images. An appropriate spatiotemporal baseline constraint threshold was chosen to generate interferometric pairs. N represents the number of images, and t... n This is the nth SLC image.

[0080] 3) Unwrapping: Using orbital information and external DEM data, the interferometric pairs are differentially processed one by one to remove flat-ground and topographic effects, resulting in a multi-look differential interferogram. The Minimum Cost Flow (MCF) method is used to complete phase unwrapping. Orbital information includes instrument parameters, calibration parameters, orbital parameters, etc., specifically referring to precise orbit determination ephemeris data in this invention.

[0081] 4) Optimization: After removing interferometric pairs containing phase errors and low coherence, stable ground control points (GCPs) are selected for optimization. Residual constant phase and phase slope existing after unwrapping are removed through optimization and re-flattening.

[0082] 5) First inversion: The deformation rate and residual terrain are estimated using a linear model, and a second unwrapping is performed to optimize the input interferogram.

[0083] 6) Second inversion: Based on the existing deformation rate, atmospheric filtering is performed to estimate and remove the atmospheric phase. Finally, the final deformation rate is obtained through least squares (LS) or singular value decomposition (SVD) methods, and the displacement in the time series is calculated.

[0084] 7) Geocoding: Before geocoding, the deformation results are in slant distance coordinates. After geocoding, the output is in geographic coordinates, obtaining the deformation rate and deformation level in the line of sight. The landslide microdeformation is classified according to the deformation rate and cumulative deformation. When the average deformation rate is greater than 100 mm / a, it is a large microdeformation; when the average deformation rate is greater than 50 mm / a and less than or equal to 100 mm / a, it is a medium microdeformation; and when the average deformation rate is less than or equal to 50 mm / a, it is a small microdeformation.

[0085] 8) Microscopic deformation locations: The microscopic deformation locations of the landslide are obtained by overlaying deformation data with spatial data.

[0086] Because of the imaging mechanism of radar satellites, they are only sensitive to east-west slopes. For north-south slopes, the method of inverting landslide bodies using InSAR technology is powerless in extracting micro-deformation.

[0087] If the slope in the monitored area is a north-south trending slope, then based on the time-series optical image data, the POT (Positive Targeting) technique is used to acquire the microscopic deformation data of the monitored area. The POT pixel shifting technique is used to dynamically track the deformation process, obtaining quantitative data such as the landslide's deformation magnitude, deformation rate, and deformation area. The process for acquiring landslide deformation using POT technology based on high-resolution variable-rate optical remote sensing images (taking Gaofen-2 as an example) is described below. Figure 4 As shown. Specifically, it includes the following steps.

[0088] 1) Image selection: Select multi-phase and multi-temporal remote sensing images from the same sensor (same satellite remote sensing data) to ensure that optical remote sensing images from the same sensor within the research object area (such as Gaofen-2, etc.) meet 100% overlap.

[0089] 2) Image preprocessing: Preprocessing remote sensing images (optical images) includes radiometric calibration, atmospheric correction and image registration.

[0090] 3) Image enhancement: Principal component transformation is performed on the preprocessed remote sensing image to enhance it. The first principal component feature band after the principal component transformation is extracted for deformation detection. The information in this band accounts for about 85% of the total information in the remote sensing image.

[0091] 4) Deformation detection: Utilizing the Co-registration of Optically Sensed Images and Correlation (COSI-Corr) plugin, the calculation window is adjusted ( Figure 5 Parameters such as window size (from 32-8 to 512-512) were used to calculate surface deformation data between images after principal component transformation at different time phases based on the Fourier algorithm. (See...) Figure 5 (Taking the Baige landslide from 2015 to 2018 as an example), the landslide deformation is classified according to the surface deformation data. Deformation less than 10 meters is classified as medium deformation, and deformation greater than 10 meters is classified as large deformation.

[0092] Step 102 specifically includes:

[0093] Based on optical remote sensing time-series data, a random forest classification method is used to extract local landslide areas in the area to be monitored.

[0094] Based on the topographic data and optical remote sensing time-series data, cracks and gullies in the area to be monitored are extracted using edge detection and random forest classification; the topographic data includes multi-period digital elevation models.

[0095] The macroscopic deformation level in the area to be monitored is determined based on the ratio of the area of ​​the local collapse zone to the area to be monitored.

[0096] When the area ratio is greater than 25%, it is considered a large macroscopic deformation; when the area ratio is less than or equal to 25% but greater than 10%, it is considered a medium macroscopic deformation; and when the area ratio is less than or equal to 10%, it is considered a small macroscopic deformation.

[0097] For obtaining macroscopic deformation parameters, based on long-term high-resolution optical remote sensing images, the random forest classification method is used to automatically extract macroscopic deformation information of the landslide body, i.e., local collapse information, such as... Figure 6 As shown.

[0098] 1) Data preprocessing: Perform data preprocessing on high-resolution optical remote sensing images (such as Gaofen-2, Gaofen-7, etc.) including radiometric correction, atmospheric correction, image fusion, orthorectification, and image enhancement.

[0099] 2) Sample Construction: Based on the image features of remote sensing images, such as hue, brightness, texture, and spectrum, and the image features of local landslides, such as shape, hue, texture, and spectrum, local landslide identification markers are established. Based on the established local landslide markers, local collapses and landslides of the landslide body in the remote sensing images are selected to establish landslide samples. At the same time, non-landslide samples are constructed by selecting background features such as roads, bare land, buildings, vegetation, and water bodies. Figures 7-17 As shown.

[0100] Macro-level local landslide information extraction: Import the constructed local landslide samples and background feature samples, set the classification parameters of random forest for information extraction (taking the Baige landslide as an example), such as... Figure 18 The results of extracting local collapse and landslide information for the Baige landslide.

[0101] 3) Based on multi-phase digital elevation models and long-term high-resolution optical images, edge detection and random forest classification methods are used to automatically extract crack and gully information of landslide bodies;

[0102] 4) Use spatial analysis methods to statistically analyze the macroscopic deformation characteristics of the slope where the landslide body is located, including the location of macroscopic deformation and the classification of deformation. The macroscopic deformation classification is based on the ratio of the area of ​​the local landslide body to the total area of ​​the landslide.

[0103] Determining the geomechanical deformation type of landslides using remote sensing.

[0104] 1) Calculation and analysis of micro and macro deformation parameters of landslide remote sensing: The macro deformation parameters and micro deformation parameters of each type of large-scale landslide are calculated through steps 101 and 102 respectively.

[0105] 2) Analysis and summary of the spatiotemporal coupling forms of landslide remote sensing micro-macro.

[0106] The spatiotemporal patterns of microscopic and macroscopic deformation parameters for each type of landslide were analyzed, including characteristics such as deformation level, deformation sequence, and spatial distribution. Similarities and differences in remote sensing deformation characteristics of 19 traditional landslide geomechanical models were explored. By integrating geological models and remote sensing deformation characteristics, landslide types with similar characteristics were summarized and merged. The landslide remote sensing deformation patterns are shown in Table 1.

[0107] Table 1 Remote Sensing Deformation Patterns of Landslides

[0108]

[0109] 3) Construct landslide remote sensing geomechanical deformation types.

[0110] Based on the spatiotemporal coupling of remote sensing micro- and macro-deformation of landslides, the geomechanical deformation process, and the deformation and failure mode, this study analyzes and summarizes the results of the spatiotemporal coupling of remote sensing micro- and macro-deformation of landslides, and constructs six types of landslide remote sensing geomechanical deformation.

[0111] This invention proposes two categories of landslide remote sensing geomechanical deformation types. Starting from the limits and indicators of remote sensing detection, and considering the material composition and failure deformation process of slopes, it integrates remote sensing detection indicators and surface deformation signs of slope deformation and failure. The final result is a list of six types of landslide remote sensing geomechanical deformation types:

[0112] 1) Soil landslides: ① Push-moving landslides, ② Traction-moving landslides

[0113] 2) Rock landslides: ③ Thrust landslides, ④ Bedding landslides, ⑤ Reverse-dip landslides, ⑥ Blocky rock mass landslides

[0114] The remote sensing deformation mode of each landslide and its corresponding micro-macro deformation parameters are shown in Table 2.

[0115] Table 2 Geomechanical Types of Landslides and Their Remote Sensing Detection Characteristics

[0116]

[0117] Among them, the landslide remote sensing geomechanical deformation types mentioned in step 103 include soil shoveling landslides, soil traction landslides, rock reverse dipping landslides, blocky rock mass landslides, rock flat pushing landslides, and rock bedding landslides.

[0118] In practical applications, geological maps and other data can be used to gradually determine the material composition and slope structure, and the movement mode can be determined based on the combination of macroscopic and microscopic data, ultimately determining the remote sensing geomechanical type.

[0119] In the aforementioned soil-driven landslide, the material composition is soil, the movement mode is shove-type sliding, the slope structure is soil, the micro-deformation parameters are large-scale micro-deformation in the upper part of the landslide or medium-scale micro-deformation in the upper part of the landslide, and the macro-deformation parameters include medium-scale or small-scale landslides in local collapse areas and significant cracks at the rear edge.

[0120] In the aforementioned soil-induced landslide, the material composition is soil, the movement mode is traction sliding, the slope structure is soil, the micro-deformation parameters are large-scale micro-deformation or medium-scale micro-deformation in the lower part of the landslide, and the macro-deformation parameters include medium-scale or small-scale landslides in the local collapse area and significant rear-edge cracks. Significant rear-edge cracks specifically refer to cracks with a length of more than 10 meters; wherein, the local collapse area is located in the lower part of the landslide body.

[0121] In the aforementioned rock-type reverse-dip landslide, the material composition is rock, the movement mode is rotational sliding, the slope structure is a reverse-dip slope, the micro-deformation parameters are large-scale micro-deformation or medium-scale micro-deformation in the lower part of the landslide, and the macro-deformation parameters include the local collapse area being a large-scale or medium-scale collapse area and the development of rear-edge cracks. The development of rear-edge cracks specifically refers to cracks with a length of more than 10 meters and a width of more than 1 meter; wherein, the local collapse area is located in the middle part of the landslide body.

[0122] In the blocky rock landslide, the material is composed of rock, the movement mode is rotational sliding, the slope structure is a blocky slope, the micro deformation parameters are large-scale micro deformation or medium-scale micro deformation in the upper part of the landslide, and the macro deformation parameters include large-scale or medium-scale landslides in local collapse areas and significant cracks at the rear edge; wherein, the local collapse area is located in the lower part of the landslide body.

[0123] In the rock-type push-type landslide, the material is composed of rock, the movement mode is push-type, the slope structure is nearly horizontal, the micro-deformation parameters are none, and the macro-deformation parameters include significant wide cracks at the rear edge, which means that the crack width reaches more than 10 meters.

[0124] In the aforementioned bedding rock landslide, the material composition is rock, the movement mode is planar sliding, the slope structure is a bedding slope, the micro-deformation parameters are none, and the macro-deformation parameters include small-scale landslides in the local collapse area, and significant landslide boundaries and cracks; wherein, the local collapse area is located in the lower part of the landslide body.

[0125] Remote sensing images can be used to determine the rear and front edges of a landslide. The part closer to the rear edge is the upper or upper-middle part, and the part closer to the front edge is the lower or lower-middle part.

[0126] Currently, domestic and international research focuses on methods for extracting thematic information such as landslide morphology, land cover changes, and surface deformation using technologies such as high-resolution optical remote sensing and spaceborne synthetic aperture radar interferometry (InSAR). However, a correlation has not yet been established between the intrinsic structure, external manifestations, deformation mechanisms, and the morphology and deformation observations of integrated remote sensing images. Studying the evolution of geological hazard risks (landslides) from the quantitative accumulation stage of microscopic deformation to the qualitative change stage of instability and sliding—that is, the micro-macroscopic deformation coupling model—and constructing remote sensing deformation parameters and geomechanical processes are key issues that urgently need to be addressed for accurate identification of geological hazards using integrated remote sensing, and are particularly crucial for the study of long-distance high-altitude landslides. The comprehensive application of remote sensing technology from this invention provides technical support for tracking the macro-microscopic deformation processes of geological hazard risks and quantifying deformation data, which is beneficial for improving the accuracy of geological hazard monitoring.

[0127] Example 2

[0128] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the geological disaster monitoring method in Embodiment 1.

[0129] Example 4

[0130] A computer program product includes a computer program that, when executed by a processor, implements the steps of the geological disaster monitoring method in Embodiment 1.

[0131] Example 4

[0132] A computer device, the internal structure of which can be shown in the diagram below. Figure 19 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The database stores pending transactions. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements the geological disaster monitoring method in Embodiment 1.

[0133] It should be noted that the object information (including but not limited to object device information, object personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this invention are all information and data authorized by the object or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0134] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided by this invention may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided by this invention may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0135] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0136] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for monitoring geological hazards, characterized in that, The method includes: Based on the remote sensing observation data corresponding to the slope body of the area to be monitored, the micro deformation parameters of the area to be monitored are determined; Based on the optical remote sensing data and topographic data of the area to be monitored, determine the macroscopic deformation parameters of the area to be monitored; Based on the material composition, movement mode, slope structure, micro-deformation parameters, and macro-deformation parameters of the area to be monitored, the landslide remote sensing geomechanical deformation type of the area to be monitored is determined; The landslide remote sensing geomechanical deformation types include soil shoving landslides, soil traction landslides, rock reverse dipping landslides, blocky rock mass landslides, rock horizontal pushing landslides, and rock bedding landslides. In the aforementioned soil-driven landslide, the material composition is soil, the movement mode is shoveling, the slope structure is soil, the micro-deformation parameters are large-scale micro-deformation or medium-scale micro-deformation in the upper part of the landslide, the macro-deformation parameters include medium-scale or small-scale landslides in the local collapse area, and the macro-deformation parameters also include cracks with a length of more than 10 meters. In the aforementioned soil-induced landslide, the material composition is soil, the movement mode is traction sliding, the slope structure is soil, the micro-deformation parameters are large-scale or medium-scale micro-deformation in the lower part of the landslide, the macro-deformation parameters include whether the local collapse area is a medium-scale or small-scale collapse, and the macro-deformation parameters also include crack lengths exceeding 10 meters; wherein, the local collapse area is located in the lower part of the landslide body; In the aforementioned rock-based reverse-dip landslide, the material composition is rock, the movement mode is rotational sliding, the slope structure is a reverse-dip slope, the micro-deformation parameters are large-scale or medium-scale micro-deformation in the lower part of the landslide, and the macro-deformation parameters include large-scale or medium-scale landslides in the local collapse area, and the macro-deformation parameters also include the presence of cracks with a length of more than 10 meters and a width of more than 1 meter; wherein, the local collapse area is located in the middle region of the landslide body; In the blocky rock landslide, the material is composed of rock, the movement is rotational sliding, the slope structure is a blocky slope, the micro-deformation parameters are large-scale micro-deformation or medium-scale micro-deformation in the upper part of the landslide, the macro-deformation parameters include large-scale or medium-scale landslides in the local collapse area, and the macro-deformation parameters also include cracks with a length of more than 10 meters; wherein, the local collapse area is located in the lower part of the landslide body; In the rock-type flat-push landslide, the material composition is rock, the movement mode is flat-push, the slope structure is nearly horizontal, the micro-deformation parameters are none, and the macro-deformation parameters include crack widths of more than 10 meters. In the aforementioned bedding rock landslide, the material composition is rock, the movement mode is planar sliding, the slope structure is a bedding slope, the micro-deformation parameters are none, and the macro-deformation parameters include local collapse areas being small collapses, and macro-deformation parameters also including crack lengths exceeding 10 meters; wherein, the local collapse areas are located in the lower part of the landslide body.

2. The geological disaster monitoring method according to claim 1, characterized in that, Based on remote sensing data corresponding to the slope body in the area to be monitored, the micro-deformation parameters of the area to be monitored are determined, specifically including: If the slope of the area to be monitored is an east-west slope, then the temporal micro-deformation process of the landslide body is inverted using InSAR technology based on the time-series radar satellite imagery data, and the micro-deformation data is determined based on the temporal micro-deformation process. The micro-deformation data includes the micro-deformation level and the deformation region. If the slope of the area to be monitored is a north-south slope, then the micro-deformation data of the area to be monitored is obtained by using POT technology based on optical image time series data.

3. The geological disaster monitoring method according to claim 2, characterized in that, Micro-deformation data is determined based on the time-series micro-deformation process, specifically including: The average deformation rate and cumulative deformation are extracted from the time-series micro-deformation process; The average deformation rate is determined based on the deformation rate and the cumulative deformation. The micro-deformation level is determined based on the average deformation rate.

4. The geological disaster monitoring method according to claim 2, characterized in that, When the average deformation rate is greater than 100 mm / a, it is considered a large micro-deformation; when the average deformation rate is greater than 50 mm / a and less than or equal to 100 mm / a, it is considered a medium-sized micro-deformation; and when the average deformation rate is less than or equal to 50 mm / a, it is considered a small micro-deformation.

5. The geological disaster monitoring method according to claim 1, characterized in that, Based on the optical remote sensing data and topographic data of the area to be monitored, the macroscopic deformation parameters of the area to be monitored are determined, specifically including: Based on optical remote sensing time-series data, the random forest classification method is used to extract local landslide areas in the area to be monitored; Based on the topographic data and optical remote sensing time-series data, cracks and gullies in the area to be monitored are extracted using edge detection and random forest classification; the topographic data includes multi-period digital elevation models. The macroscopic deformation level in the area to be monitored is determined based on the ratio of the area of ​​the local collapse zone to the area to be monitored.

6. The geological disaster monitoring method according to claim 5, characterized in that, When the area ratio is greater than 25%, it is considered a large macroscopic deformation; when the area ratio is less than or equal to 25% but greater than 10%, it is considered a medium macroscopic deformation; and when the area ratio is less than or equal to 10%, it is considered a small macroscopic deformation.

7. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the geological hazard monitoring method according to any one of claims 1-6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the geological disaster monitoring method according to any one of claims 1-6.

9. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the geological disaster monitoring method according to any one of claims 1-6.

Images