Landslide identification method and device, computer device and storage medium

Abstract

The present application relates to a landslide identification method, device and computer equipment. Among them, the landslide identification method, by comprehensively considering continuous disaster factors and linear disaster factors, and combining the ChiMerge discretization processing method, the contribution of each factor to the occurrence of landslide is quantified, thereby improving the accuracy and reliability of landslide prediction; secondly, by adopting ChiMerge discretization to discretize the Euclidean distance of continuous disaster factors and linear disaster factors respectively, the complexity of the prediction model can be reduced, the accuracy of the model can be improved, and the calculation cost caused by continuous variables can be reduced; finally, by obtaining the current landslide optical identification timing characteristics and historical landslide optical identification timing characteristics of the shape change response of each grid in the study area, and confirming the landslide probability of each grid in the study area according to the above, the landslide probability of each grid in the study area can be obtained more conveniently.

Description

Technical Field

[0001] This invention relates to the field of landslide identification, and more specifically, to a landslide identification method, apparatus, computer equipment, and storage medium. Background Technology

[0002] Landslides are a common geological hazard, referring to the phenomenon where the earth's surface or soil mass loses stability due to various factors, causing the soil or rock mass to move downwards along a certain sliding surface or crack. Landslides not only pose a serious threat to human life and property but can also cause severe environmental damage, affecting land use and economic development. Therefore, timely and accurate identification and monitoring of landslides are of great significance.

[0003] Landslide identification refers to the process of identifying potential landslide bodies and providing monitoring and early warning by analyzing relevant factors such as geology, topography, and climate, using methods such as geological surveys, remote sensing technology, and geographic information systems. Landslide identification typically involves various data and indicators, such as topographic maps, geological maps, remote sensing imagery, vegetation cover indices (e.g., NDVI), and soil moisture content. By comprehensively analyzing these data and indicators, signs such as abnormal changes in the land surface, vegetation loss, and increased soil moisture can be detected, thereby identifying potential landslide risk areas.

[0004] However, the inventors found that existing landslide identification methods suffer from high false positive rates and poor applicability. Summary of the Invention

[0005] Therefore, it is necessary to provide a landslide identification method, device, computer equipment, and storage medium that reduces the false positive rate and improves applicability.

[0006] To achieve the above objectives, one embodiment of this application provides a landslide identification method, including the following steps:

[0007] Obtain landslide-prone environmental factors; landslide-prone environmental factors include continuous and linear landslide-prone factors.

[0008] ChiMerge discretization was performed on the continuous disaster-prone factors to obtain the first discretization interval of each continuous disaster-prone factor.

[0009] Obtain the Euclidean distance of the linear pregnancy factor;

[0010] The ChiMerge discretization method is applied to divide the Euclidean distance of the linear hazard factors into spatial distances, and the second discretization interval of each linear hazard factor is obtained.

[0011] Calculate the landslide susceptibility index for the first and second discretization intervals, and train the prediction model using the first and second discretization intervals and the landslide susceptibility index.

[0012] The landslide susceptibility index of the study area was confirmed using a trained prediction model;

[0013] Obtain temporal deformation and temporal optical characteristics of historical landslide areas in the study area;

[0014] By processing temporal deformation features and temporal optical features, we obtain the historical landslide optical identification temporal features that vary with deformation response;

[0015] The current landslide optical identification time-series characteristics of each grid in the study area with deformation response are obtained, and the landslide probability of each grid in the study area is confirmed based on the landslide susceptibility index, historical landslide optical identification time-series characteristics and current landslide optical identification time-series characteristics.

[0016] In one embodiment, the step further includes:

[0017] The first mutual information between continuous disaster-generating factors and the temporal deformation characteristics of historical landslide areas, and the second mutual information between linear disaster-generating factors and the temporal deformation characteristics of historical landslide areas are obtained.

[0018] Continuous disaster-prone factors whose first mutual information value is less than the preset value are removed to obtain the processed continuous disaster-prone factors.

[0019] Linear disaster-prone factors whose second mutual information value is less than the preset value are removed to obtain the processed linear disaster-prone factors.

[0020] The steps for ChiMerge discretization of continuous pregnancy factors include:

[0021] The processed continuous pregnancy factors were discretized using ChiMerge.

[0022] The steps to obtain the Euclidean distance of the linear pregnancy factor include:

[0023] Obtain the Euclidean distance of the processed linear pregnancy factor.

[0024] In one embodiment, the step of determining the landslide probability of each grid cell in the study area based on the landslide susceptibility index, historical landslide optical identification time-series characteristics, and current landslide optical identification time-series characteristics includes:

[0025] Based on the landslide susceptibility index, susceptibility grids with high or extremely high susceptibility were identified from the study area.

[0026] Using the current landslide optical identification time-series features as input variables and the historical landslide optical identification time-series features with deformation response as landslide samples, the probability of classifying each landslide-prone grid as a landslide sample is obtained using the extreme random tree algorithm.

[0027] The probability was confirmed as the landslide probability.

[0028] In one embodiment, the step of determining the landslide probability of each grid cell in the study area based on the landslide susceptibility index, historical landslide optical identification time-series characteristics, and current landslide optical identification time-series characteristics includes:

[0029] Build a neural network model;

[0030] The neural network model was trained using the landslide susceptibility index and the time-series characteristics of historical landslide optical identification.

[0031] Input the current landslide optical recognition time-series features into the trained neural network model to obtain the corresponding landslide susceptibility index;

[0032] The probability of a landslide is determined based on the corresponding landslide susceptibility index.

[0033] In one embodiment, the step of processing temporal deformation features and temporal optical features to obtain historical landslide optical identification temporal features that correspond to deformation response includes:

[0034] By employing the Pearson correlation method or the dynamic time warp method, temporal deformation features and temporal optical features are processed to obtain the historical landslide optical identification temporal features that respond to deformation.

[0035] In one embodiment, the continuous disaster-prone factor includes any one or any combination of the following factors: weathering crust thickness, elevation, slope, curvature, annual average rainfall, and normalized vegetation index.

[0036] In one embodiment, the linear disaster-inducing factor includes any one or any combination of the following factors: minor faults, lithological boundaries, rivers, and roads.

[0037] On the other hand, embodiments of the present invention also provide a landslide identification device, including:

[0038] The first acquisition module is used to acquire landslide-prone environmental factors; landslide-prone environmental factors include continuous landslide-prone factors and linear landslide-prone factors.

[0039] The first discretization module is used to perform ChiMerge discretization on the continuous disaster-prone factors to obtain the first discretization interval of each continuous disaster-prone factor.

[0040] The second acquisition module is used to obtain the Euclidean distance of the linear pregnancy factor;

[0041] The second discretization module is used to apply the ChiMerge discretization method to perform spatial distance segmentation on the Euclidean distance of the linear disaster-causing factors, and obtain the second discretization interval of each linear disaster-causing factor.

[0042] The model training module is used to calculate the landslide susceptibility index for the first discretization interval and the second discretization interval, and to train the prediction model using the first discretization interval, the second discretization interval, and the landslide susceptibility index.

[0043] The confirmation module is used to confirm the landslide susceptibility index of the study area using a trained prediction model.

[0044] The third acquisition module is used to acquire the temporal deformation characteristics and temporal optical characteristics of historical landslide areas in the study area;

[0045] The feature processing module is used to process temporal deformation features and temporal optical features to obtain historical landslide optical recognition temporal features that vary with deformation response;

[0046] The probability calculation module is used to obtain the current landslide optical identification time-series characteristics of the deformation response of each grid in the study area, and to determine the landslide probability of each grid in the study area based on the landslide susceptibility index, historical landslide optical identification time-series characteristics and current landslide optical identification time-series characteristics.

[0047] On the one hand, this application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and a computer-readable storage medium stores a computer program, wherein the computer program is configured to execute the above-described method at runtime.

[0048] On the other hand, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method.

[0049] Compared with the prior art, one of the above technical solutions has the following advantages and beneficial effects:

[0050] The aforementioned landslide identification method, by comprehensively considering both continuous and linear landslide-inducing factors and combining them with the ChiMerge discretization method, quantifies the contribution of each factor to landslide occurrence, thereby improving the accuracy and reliability of landslide prediction and enhancing its applicability. Secondly, by using ChiMerge discretization to discretize the Euclidean distances of the continuous and linear landslide-inducing factors respectively, the complexity of the prediction model can be reduced, the accuracy of the model can be improved, and the computational cost caused by continuous variables can be reduced. Finally, by obtaining the current landslide optical identification time-series features and historical landslide optical identification time-series features of the deformation response of each grid in the study area, and based on these features, the landslide probability of each grid in the study area can be confirmed, making it relatively convenient to obtain the landslide probability of each grid in the study area. Attached Figure Description

[0051] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0052] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0053] Figure 1 This is a schematic flowchart of a landslide identification method in one embodiment;

[0054] Figure 2 This is a schematic diagram of a landslide plan in one embodiment;

[0055] Figure 3 This is a first schematic flowchart illustrating the steps of determining the landslide probability of each grid in the study area based on the landslide susceptibility index, historical landslide optical identification time-series characteristics, and current landslide optical identification time-series characteristics in one embodiment.

[0056] Figure 4 This is a first schematic flowchart illustrating the steps of determining the landslide probability of each grid in the study area based on the landslide susceptibility index, historical landslide optical identification time-series characteristics, and current landslide optical identification time-series characteristics, in one embodiment. Detailed Implementation

[0057] To facilitate understanding of this application, a more complete description will be provided below with reference to the accompanying drawings, which illustrate embodiments of the present application. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that the disclosure of this application will be thorough and complete.

[0058] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

[0059] In the following description, the use of suffixes such as "module," "part," or "unit" to denote elements is solely for the purpose of illustration and has no specific meaning in itself. Therefore, "module" and "part" may be used interchangeably.

[0060] It is understood that the term "connection" in the following embodiments should be understood as "electrical connection," "communication connection," etc., if the connected circuits, modules, units, etc., have electrical signal or data transmission with each other.

[0061] When used herein, the singular forms of “a,” “an,” and “the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms “comprising,” “including,” or “having,” etc., specify the presence of the stated feature, whole, step, operation, component, part, or combination thereof, but do not preclude the possibility of the presence or addition of one or more other features, wholes, steps, operations, components, parts, or combinations thereof.

[0062] In one embodiment, such as Figure 1 As shown, a landslide identification method is provided, including the following steps:

[0063] S110, Obtain landslide-prone environmental factors; landslide-prone environmental factors include continuous-prone factors and linear-prone factors.

[0064] Among them, landslide-inducing environmental factors are a set of data that influence or cause landslides. Continuous landslide-inducing factors include any one or any combination of the following factors: weathering crust thickness, elevation, slope, curvature, annual average rainfall, and normalized difference vegetation index. Linear landslide-inducing factors include any one or any combination of the following factors: minor faults, lithological boundaries, rivers, and roads.

[0065] Specifically, landslide-prone environmental factors can be pre-input through methods such as collection and organization, and can be directly obtained when needed.

[0066] S120, ChiMerge discretization is performed on the continuous disaster-prone factors to obtain the first discretization interval of each continuous disaster-prone factor.

[0067] Specifically, ChiMerge discretization is a commonly used data preprocessing method for converting continuous data into discrete data. Compared to conventional equal-interval and naturally discontinuous discretization, the ChiMerge discretization proposed in this patent is beneficial for maximizing the retention of data information. This method reduces data complexity by merging adjacent numerical intervals while retaining important data information. The specific steps for ChiMerge discretization of continuous disaster factors are as follows: Sort the values ​​of the continuous disaster factors according to their magnitude and treat each value as a separate interval; calculate the chi-square value of adjacent interval pairs. The chi-square value can be calculated based on a statistical indicator, such as a classification label; merge adjacent interval pairs according to their chi-square values; recalculate the chi-square value after merging until the merging condition is met or the preset maximum number of intervals is reached; determine the first discretization interval for each continuous disaster factor based on the merging result.

[0068] S130, obtain the Euclidean distance of the linear pregnancy factor;

[0069] Specifically, the Euclidean distance between linear hazard factors refers to the distance between two linear hazard factors in the feature space. Typically, the Euclidean distance formula can be used to calculate the distance between two vectors.

[0070] S140, the ChiMerge discretization method is applied to divide the Euclidean distance of the linear disaster-causing factors into spatial distances to obtain the second discretization interval of each linear disaster-causing factor.

[0071] Specifically, compared to the conventional multi-buffering method for obtaining the discrete interval of linear factors, this method can accurately obtain the dominant distance range of linear factors. The ChiMerge discretization method can be used to process the Euclidean distance to obtain a second discretized interval. The steps of the ChiMerge discretization method have already been mentioned in S120: the Euclidean distance is divided according to certain conditions to obtain discretized intervals. This transforms continuous Euclidean distances into discrete intervals.

[0072] S150, calculate the landslide susceptibility index for the first discretization interval and the second discretization interval, and train the prediction model using the first discretization interval, the second discretization interval, and the landslide susceptibility index.

[0073] The landslide susceptibility index can include parameters such as area, number of landslides, and landslide density.

[0074] Specifically, after confirming the first and second discretization intervals, the interval area, number of landslides, and landslide density corresponding to each interval can be determined.

[0075] Furthermore, the prediction model can employ a commonly used binary classification model.

[0076] S160, The landslide susceptibility index of the study area was confirmed using a trained prediction model;

[0077] Specifically, the trained prediction model can be used to confirm the landslide susceptibility index of the entire study area.

[0078] S170, to obtain the temporal deformation characteristics and temporal optical characteristics of historical landslide areas in the study area;

[0079] Specifically, the temporal deformation characteristics of historical landslide areas can be obtained through the following methods: Selecting suitable radar satellite data as the data source, such as data from the European Space Agency's Sentinel-1 satellite. This data typically includes SAR images from multiple time points; preprocessing the acquired SAR image data, including radiometric correction, Earth surface elevation correction, and atmospheric correction, to reduce interference factors in the images; selecting a set of small baselines from the preprocessed SAR image data, these baselines typically have small lengths and angles to ensure coherence and accuracy; using the selected small baseline set, interferometric processing is performed on the time-series SAR images to obtain a series of coherence maps; unwrapping the formed coherence maps, and obtaining surface deformation information, including the average deformation rate and cumulative deformation, through a phase unwrapping algorithm; based on the deformation information obtained from the unwrapping process, the long-term deformation characteristics of the surface, also known as the temporal deformation characteristics, can be obtained.

[0080] A landslide is the process by which a mass of rock or other debris slides downwards along one or more fractured sliding surfaces. In terms of planar structure, a landslide has a groove at its rear edge and a bulge at its front edge caused by the downward sliding of the landslide mass; both exhibit opposite deformation characteristics (e.g., ...). Figure 2 (As shown). The deformation patterns and optical characteristics of different parts of a landslide vary. The landslide area is divided into three regions: the rear edge, the step, and the leading edge, for regional deformation pattern analysis. The landslide boundary is the plane boundary between the landslide body and the surrounding stationary slope. Tensile cracks are cracks formed by tensile stress and are mostly distributed in the axial part of anticline structures. Shear cracks refer to cracks generated on the cross-section of a member under the combined action of shear force or shear force and bending moment. From the perspective of average and cumulative deformation over the study period, key deformation points of the landslide rear wall, the step, and the leading edge are captured. The temporal deformation direction, rate, inducing factors, landslide deformation fluctuation range, and pre-instability deformation threshold of typical deformation points in each region are statistically analyzed to obtain the deformation patterns of each region. The obtained deformation data are arranged in chronological order to form a temporal deformation characteristic.

[0081] Furthermore, multiple Sentinel-2 remote sensing images of the region were acquired. Multispectral data (B2, B3, B4, B8, B11, B12), indices (normalized difference vegetation index, normalized difference snow index), and texture features (mean, variance, evenness, contrast, dissimilarity, entropy, quadratic moment, correlation) of the regions of interest were extracted from the images. Specifically, B2: Blue band, wavelength range approximately 490-560 nm; B3: Green band, wavelength range approximately 560-590 nm; B4: Red band, wavelength range approximately 640-680 nm; B8: NIR band, wavelength range approximately 780-900 nm; B11: SWIR1 band, wavelength range approximately 1565-1655 nm. B12: SWIR 2 band, shortwave infrared 2 band, with a wavelength range of approximately 2100-2200 nanometers.

[0082] S180 processes temporal deformation features and temporal optical features to obtain historical landslide optical identification temporal features that respond to deformation.

[0083] Specifically, deformation points in various regions of historical landslides are correlated with corresponding multispectral, exponential, and textural features in Sentinel-2 imagery to obtain the temporal characteristics of historical landslide optical identification in response to deformation. In one example, Pearson Correlation (PC) analysis is used to calculate the correlation between deformation features and spectral features of deformation points. The PC value reflects the degree of linear correlation between the two variables, ranging from -1 to 1, where -1 indicates a perfect negative correlation, 1 indicates a perfect positive correlation, and 0 indicates no correlation. In another example, Dynamic Time Warping (DTW) analysis is used to calculate the similarity between deformation features and spectral features of deformation points. DTW can measure the similarity between two sequences, allowing for correct matching even if they are time-lags or have different velocities.

[0084] S190: Obtain the current landslide optical identification time-series characteristics of each grid in the study area as a function of deformation, and confirm the landslide probability of each grid in the study area based on the landslide susceptibility index, historical landslide optical identification time-series characteristics and current landslide optical identification time-series characteristics.

[0085] Specifically, the study is divided into individual grids. In one example, such as Figure 3As shown, the steps for determining the landslide probability of each grid in the study area based on the landslide susceptibility index, historical landslide optical identification time-series characteristics, and current landslide optical identification time-series characteristics include: S310, identifying high- or extremely high-susceptibility grids in the study area based on the landslide susceptibility index; specifically, evaluating all grids in the study area based on the landslide susceptibility index to determine high- or extremely high-susceptibility grids. The evaluation criteria for high or extremely high susceptibility can refer to commonly used standards in this field. S320, using the current landslide optical identification time-series characteristics as input variables and the historical landslide optical identification time-series characteristics that respond to deformation as landslide samples, using an extreme random tree algorithm to obtain the probability of classifying each susceptible grid as a landslide sample; specifically, classifying the current landslide optical identification time-series characteristics using an extreme random tree algorithm and obtaining the probability of it being classified as a landslide sample. S330, confirming the probability as the landslide probability. Landslide susceptibility assessment can narrow the scope of landslide identification and reduce interference from complex environmental backgrounds. Furthermore, combining radar deformation and optical information to formulate time-series multi-source characterization indicators can achieve early identification of landslides from multiple angles, avoiding missed landslides due to landslide range, activity time, and subsequent vegetation.

[0086] In another embodiment, such as Figure 4 As shown, the steps for determining the landslide probability of each grid in the study area based on the landslide susceptibility index, historical landslide optical identification time-series features, and current landslide optical identification time-series features include: S410, constructing a neural network model; S420, training the neural network model using the landslide susceptibility index and historical landslide optical identification time-series features; S430, inputting the current landslide optical identification time-series features into the trained neural network model to obtain the corresponding landslide susceptibility index; and S440, confirming the landslide probability based on the corresponding landslide susceptibility index. Specifically, the structure of the neural network is determined, including the number of nodes in the input layer, hidden layer, and output layer. Then, appropriate activation functions, loss functions, and optimization algorithms are selected to complete the construction of the neural network model. Training the neural network model using the landslide susceptibility index and historical landslide optical identification time-series features: inputting the current landslide optical identification time-series features into the trained neural network model to obtain the corresponding landslide susceptibility index; and converting the landslide susceptibility index output by the neural network model into a landslide probability. The landslide susceptibility index is defined as the landslide density under different environmental disaster-causing factors. Different weights are assigned to these factors, and the landslide probability is obtained based on these weights and the corresponding landslide densities. This step involves inputting the current landslide optical recognition time-series features into a trained neural network model to identify the corresponding landslide susceptibility index, and then deriving the landslide probability from this index. This method can obtain the landslide probability directly from the optical recognition time-series features, demonstrating strong applicability and operability.

[0087] The aforementioned landslide identification method, by comprehensively considering both continuous and linear landslide-inducing factors and combining them with the ChiMerge discretization method, quantifies the contribution of each factor to landslide occurrence, thereby improving the accuracy and reliability of landslide prediction. Secondly, by employing ChiMerge discretization to discretize the Euclidean distances of both continuous and linear landslide-inducing factors, the complexity of the prediction model can be reduced, the model's accuracy improved, and the computational cost due to continuous variables reduced. Finally, by obtaining the current and historical landslide optical identification time-series features of the deformation response of each grid in the study area, and based on this, the landslide probability of each grid in the study area can be confirmed, improving the reliability and accuracy of landslide identification for each grid, and providing a relatively convenient way to obtain the landslide probability of each grid in the study area.

[0088] In one embodiment, the step further includes:

[0089] The first mutual information between continuous disaster-generating factors and the temporal deformation characteristics of historical landslide areas, and the second mutual information between linear disaster-generating factors and the temporal deformation characteristics of historical landslide areas are obtained.

[0090] Continuous disaster-prone factors whose first mutual information value is less than the preset value are removed to obtain the processed continuous disaster-prone factors.

[0091] Linear disaster-prone factors whose second mutual information value is less than the preset value are removed to obtain the processed linear disaster-prone factors.

[0092] The steps for ChiMerge discretization of continuous pregnancy factors include:

[0093] The processed continuous pregnancy factors were discretized using ChiMerge.

[0094] The steps to obtain the Euclidean distance of the linear pregnancy factor include:

[0095] Obtain the Euclidean distance of the processed linear pregnancy factor.

[0096] Specifically, mutual information is a measure of the correlation between two random variables; it represents the amount of information contained in one random variable about the other. In landslide analysis, we can use mutual information to quantify the correlation between landslide-causing factors (such as topography, geology, and climate) and the temporal deformation characteristics of landslides. In this step, landslide-causing factors with poor correlation are removed, and the removed factors are used in subsequent data processing calculations.

[0097] In one embodiment, a landslide identification device is provided, comprising:

[0098] The first acquisition module is used to acquire landslide-prone environmental factors; landslide-prone environmental factors include continuous landslide-prone factors and linear landslide-prone factors.

[0099] The first discretization module is used to perform ChiMerge discretization on the continuous disaster-prone factors to obtain the first discretization interval of each continuous disaster-prone factor.

[0100] The second acquisition module is used to obtain the Euclidean distance of the linear pregnancy factor;

[0101] The second discretization module is used to apply the ChiMerge discretization method to perform spatial distance segmentation on the Euclidean distance of the linear disaster-causing factors, and obtain the second discretization interval of each linear disaster-causing factor.

[0102] The model training module is used to calculate the landslide susceptibility index for the first discretization interval and the second discretization interval, and to train the prediction model using the first discretization interval, the second discretization interval, and the landslide susceptibility index.

[0103] The confirmation module is used to confirm the landslide susceptibility index of the study area using a trained prediction model.

[0104] The third acquisition module is used to acquire the temporal deformation characteristics and temporal optical characteristics of historical landslide areas in the study area;

[0105] The feature processing module is used to process temporal deformation features and temporal optical features to obtain historical landslide optical recognition temporal features that vary with deformation response;

[0106] The probability calculation module is used to obtain the current landslide optical identification time-series characteristics of the deformation response of each grid in the study area, and to determine the landslide probability of each grid in the study area based on the landslide susceptibility index, historical landslide optical identification time-series characteristics and current landslide optical identification time-series characteristics.

[0107] Specific limitations regarding the landslide identification device can be found in the limitations of the landslide identification method described above, and will not be repeated here. Each module in the aforementioned landslide identification device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module. It should be noted that the module division in this embodiment is illustrative and only represents a logical functional division; other division methods may be used in actual implementation.

[0108] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0109] Obtain landslide-prone environmental factors; landslide-prone environmental factors include continuous and linear landslide-prone factors.

[0110] ChiMerge discretization was performed on the continuous disaster-prone factors to obtain the first discretization interval of each continuous disaster-prone factor.

[0111] Obtain the Euclidean distance of the linear pregnancy factor;

[0112] The ChiMerge discretization method is applied to divide the Euclidean distance of the linear hazard factors into spatial distances, and the second discretization interval of each linear hazard factor is obtained.

[0113] Calculate the landslide susceptibility index for the first and second discretization intervals, and train the prediction model using the first and second discretization intervals and the landslide susceptibility index.

[0114] The landslide susceptibility index of the study area was confirmed using a trained prediction model;

[0115] Obtain temporal deformation and temporal optical characteristics of historical landslide areas in the study area;

[0116] By processing temporal deformation features and temporal optical features, we obtain the historical landslide optical identification temporal features that vary with deformation response;

[0117] The current landslide optical identification time-series characteristics of each grid in the study area with deformation response are obtained, and the landslide probability of each grid in the study area is confirmed based on the landslide susceptibility index, historical landslide optical identification time-series characteristics and current landslide optical identification time-series characteristics.

[0118] In one embodiment, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, performs the following steps:

[0119] Obtain landslide-prone environmental factors; landslide-prone environmental factors include continuous and linear landslide-prone factors.

[0120] ChiMerge discretization was performed on the continuous disaster-prone factors to obtain the first discretization interval of each continuous disaster-prone factor.

[0121] Obtain the Euclidean distance of the linear pregnancy factor;

[0122] The ChiMerge discretization method is applied to divide the Euclidean distance of the linear hazard factors into spatial distances, and the second discretization interval of each linear hazard factor is obtained.

[0123] Calculate the landslide susceptibility index for the first and second discretization intervals, and train the prediction model using the first and second discretization intervals and the landslide susceptibility index.

[0124] The landslide susceptibility index of the study area was confirmed using a trained prediction model;

[0125] Obtain temporal deformation and temporal optical characteristics of historical landslide areas in the study area;

[0126] By processing temporal deformation features and temporal optical features, we obtain the historical landslide optical identification temporal features that vary with deformation response;

[0127] The current landslide optical identification time-series characteristics of each grid in the study area with deformation response are obtained, and the landslide probability of each grid in the study area is confirmed based on the landslide susceptibility index, historical landslide optical identification time-series characteristics and current landslide optical identification time-series characteristics.

[0128] In specific implementation, the embodiments of this application can be referred to the above embodiments and have corresponding technical effects.

[0129] It is understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit can be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers, microprocessors, other electronic units for performing the functions described herein, or combinations thereof.

[0130] For software implementation, the techniques described herein can be implemented by units that perform the functions described herein. The software code can be stored in memory and executed by a processor. The memory can be implemented in the processor or external to the processor.

[0131] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0132] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0133] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0134] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0135] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0136] If the aforementioned function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks. It should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. In the absence of further restrictions, an element defined by the phrase "comprising a..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0137] The above description is merely a specific embodiment of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A landslide identification method, characterized in that, include: Acquire landslide-prone environmental factors; the landslide-prone environmental factors include continuous landslide-prone factors and linear landslide-prone factors; The continuous pregnancy factors are discretized using ChiMerge to obtain the first discretization interval of each continuous pregnancy factor. Obtain the Euclidean distance of the linear pregnancy factor; The ChiMerge discretization method is applied to perform spatial distance segmentation on the Euclidean distance of the linear hazard factors to obtain the second discretization interval of each linear hazard factor; Calculate the landslide susceptibility index for the first discretization interval and the second discretization interval, and train a prediction model using the first discretization interval, the second discretization interval, and the landslide susceptibility index; The landslide susceptibility index of the study area was confirmed using a trained prediction model; Obtain the temporal deformation and temporal optical characteristics of historical landslide areas in the study area; By processing the temporal deformation features and temporal optical features, the historical landslide optical identification temporal features that respond to deformation are obtained; The current landslide optical identification time-series features of the deformation response of each grid in the study area are obtained, and the landslide probability of each grid in the study area is confirmed based on the landslide susceptibility index, the historical landslide optical identification time-series features and the current landslide optical identification time-series features.

2. The landslide identification method according to claim 1, characterized in that, It also includes the following steps: Obtain the first mutual information between the continuous disaster-causing factor and the temporal deformation characteristics of the historical landslide area, and the second mutual information between the linear disaster-causing factor and the temporal deformation characteristics of the historical landslide area; Continuous disaster-prone factors whose first mutual information value is less than a preset value are removed to obtain processed continuous disaster-prone factors. Linear disaster-prone factors whose second mutual information value is less than a preset value are removed to obtain the processed linear disaster-prone factors. The step of performing ChiMerge discretization on the continuous pregnancy factors includes: The processed continuous pregnancy factors are discretized using ChiMerge. The step of obtaining the Euclidean distance of the linear pregnancy factor includes: Obtain the Euclidean distance of the processed linear pregnancy factor.

3. The landslide identification method according to claim 1, characterized in that, The step of determining the landslide probability of each grid cell in the study area based on the landslide susceptibility index, the historical landslide optical identification time-series characteristics, and the current landslide optical identification time-series characteristics includes: Based on the landslide susceptibility index, susceptibility grids with high or extremely high susceptibility are identified from the study area; Using the current landslide optical identification time-series features as input variables and the historical landslide optical identification time-series features that vary with deformation response as landslide samples, the probability of classifying each of the landslide-prone grids as a landslide sample is obtained using an extreme random tree algorithm. The probability was confirmed as the landslide probability.

4. The landslide identification method according to claim 1, characterized in that, The step of determining the landslide probability of each grid cell in the study area based on the landslide susceptibility index, the historical landslide optical identification time-series characteristics, and the current landslide optical identification time-series characteristics includes: Build a neural network model; The neural network model is trained using the landslide susceptibility index and the time-series optical identification features of historical landslides; Input the current landslide optical recognition time-series features into the trained neural network model to obtain the corresponding landslide susceptibility index; The landslide probability is determined based on the corresponding landslide susceptibility index.

5. The landslide identification method according to claim 1, characterized in that, The steps of processing the temporal deformation features and temporal optical features to obtain the historical landslide optical identification temporal features with deformation response include: The temporal deformation features and temporal optical features are processed using the Pearson correlation method or the dynamic time warp method to obtain the historical landslide optical identification temporal features that respond to deformation.

6. The landslide identification method according to any one of claims 1-5, characterized in that, The continuous disaster-inducing factors include any one or any combination of the following factors: weathering crust thickness, elevation, slope, curvature, annual average rainfall, and normalized vegetation index.

7. The landslide identification method according to any one of claims 1-5, characterized in that, The linear disaster-inducing factors include any one or any combination of the following factors: minor faults, lithological boundaries, rivers, and roads.

8. A landslide identification device, characterized in that, include: The first acquisition module is used to acquire landslide-prone environmental factors; the landslide-prone environmental factors include continuous landslide-prone factors and linear landslide-prone factors. The first discretization processing module is used to perform ChiMerge discretization processing on the continuous pregnancy factors to obtain the first discretization interval of each continuous pregnancy factor. The second acquisition module is used to acquire the Euclidean distance of the linear pregnancy factor; The second discretization module is used to apply the ChiMerge discretization method to perform spatial distance segmentation on the Euclidean distance of the linear disaster-causing factors, so as to obtain the second discretization interval of each linear disaster-causing factor. The model training module is used to calculate the landslide susceptibility index of the first discretization interval and the second discretization interval, and to train the prediction model using the first discretization interval, the second discretization interval, and the landslide susceptibility index. The confirmation module is used to confirm the landslide susceptibility index of the study area using a trained prediction model. The third acquisition module is used to acquire the temporal deformation characteristics and temporal optical characteristics of the historical landslide areas in the study area; The feature processing module is used to process the temporal deformation features and temporal optical features to obtain the historical landslide optical recognition temporal features that respond to deformation. The probability calculation module is used to obtain the current landslide optical identification time-series characteristics of the deformation response of each grid in the study area, and to determine the landslide probability of each grid in the study area based on the landslide susceptibility index, the historical landslide optical identification time-series characteristics and the current landslide optical identification time-series characteristics.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images