33 results about "Landslide susceptibility" patented technology

The invention discloses a landslide susceptibility assessment method based on domestic GF-1 satellite data. The method comprises the following steps: (1) data collection; (2) data preprocessing; finally, the remote sensing data and other data are superposed and clipped to have uniform latitude and longitude range and grid element; (3) remote sensing extraction of landslide information; (4) candidate evaluation factors extraction and screening; (5) landslide susceptibility assessment. The method solves the problem of extracting landslide information by remote sensing, extraction and screening of landslide susceptibility assessment factors, construction of a landslide susceptibility assessment model, divides zones of landslide hazard susceptibility in the assessment area, provides basic datafor landslide hazard prediction and prediction, and is a scheme with high accuracy to assess landslide hazard susceptibility.

The invention provides a landslide susceptibility evaluation method based on a remote-sensing image of an unmanned aerial vehicle. According to the method, by employing the image acquired by the unmanned aerial vehicle as main characteristic data for evaluation of the landslide susceptibility and with the combination of a regional geological map and meteorological data of an evaluation object region, landslide related parameters for representing different local area landslide related characteristics in a detailed manner are comprehensively extracted as landslide susceptibility influence factors, and landslide disaster occurrence probabilities of different local areas in a to-be-evaluated region are determined by employing a landslide disaster occurrence probability model established by a logistic regression method so that a landslide susceptibility regionalization evaluation result of the to-be-evaluated region is determined, the landslide susceptibility regionalization evaluation can be accurately realized even for the to-be-evaluated regions with small overall area ranges, the evaluation results are high in local area pertinence, high in reliability, and good in practicality, and accurate and scientific data reference information can be provided for the application of landslide disaster prevention and early warning, engineering construction detection, and geological research etc.

The invention discloses a landslide susceptibility evaluation method based on a fractal-machine learning hybrid model. The method comprises the following steps: selecting landslide susceptibility evaluation factors; analyzing a fractal relationship between historical landslide geological disaster points and landslide susceptibility evaluation factors in the experimental area based on a fractal model, and calculating a preliminary landslide susceptibility index on the basis of solving fractal dimensions between the landslide susceptibility evaluation factors and the historical geological disaster points; constructing a multi-scene sample data set: constructing sample data sets of three different scenes by the three non-landslide samples and a unified landslide sample; taking the sample datasets of the three scenes respectively as inputs of the NB model and the SVM model to carry out landslide susceptibility evaluation research. Compared with a negative sample generated from a low-slopearea and a non-landslide area in traditional landslide susceptibility research, the negative sample quantitatively selected based on the fractal model can improve the quality of a landslide susceptibility evaluation sample, and the use of the fractal-machine learning hybrid model can improve the accuracy of landslide susceptibility evaluation.

The invention discloses a landslide susceptibility improvement evaluation method based on InSAR and LR-IOE models, and the method comprises the following steps: S1, collecting the geological and geographic data of an evaluation region, extracting evaluation factors, and building a landslide susceptibility evaluation index system; S2, obtaining the deformation rate of the earth surface of the evaluation area in the radar sight line direction through the SBAS-InSAR technology, screening out reliable deformation points, and converting the deformation rate from the radar sight line direction to the maximum gradient direction to serve as a landslide susceptibility evaluation factor; and S3, constructing an entropy index-logistic regression coupling model, and performing landslide susceptibility evaluation. According to the method, the SBAS-InSAR technology is used for collecting deformation information of the long-time-sequence earth surface, the deformation rate in the maximum gradient direction serves as a landslide susceptibility evaluation factor, evaluation and optimization are conducted on the landslide susceptibility based on the entropy index-logistic regression coupling model, and the prediction precision of the model on the landslide susceptibility is remarkably improved.

The invention relates to earthquake disaster processing, and discloses an earthquake area landslide susceptibility evaluation method. The method comprises the following steps: 1, obtaining an earthquake landslide catalogue according to the difference between earthquake area pre-earthquake information and earthquake post-earthquake information; (2) selecting a landslide sample from the landslide catalogue, based on the landslide sample, adopting a single-class support vector machine model to obtain an initial earthquake landslide susceptibility index, normalizing the initial earthquake landslide susceptibility index into a probability index, and partitioning the landslide according to the probability index; (3) selecting non-landslide samples from the first landslide first easy-to-occur area and the first landslide second easy-to-occur area, and constructing a sample data set; and (4) based on the sample data set, obtaining landslide susceptibility indexes of the seismic region throughtwo support vector machine models, and partitioning the landslide by adopting a natural breakpoint method. The evaluation method can rapidly and accurately realize landslide easy-to-occur evaluation of the earthquake disaster area so as to provide a guidance basis for subsequent disaster emergency rescue, disaster monitoring and land planning.

The invention discloses a landslide susceptibility evaluation method based on GBDT-LR and an information amount model, which combines characteristics of landslide development of a research area, comprehensively considers research results of previous people based on GIS, Python and other software, and finally determines 13 evaluation factors. A landslide susceptibility evaluation model is constructed based on an information amount model (I) and a gradient boosting decision tree-logistic regression (GBDT-LR) algorithm, the problem of asymmetry of landslide samples is solved by adopting a Borderline-Smote algorithm, the evaluation model is evaluated through a frequency ratio, a ten-fold line cross validation and an ROC curve, and finally, a Guiding pool susceptibility zoning map is made. The rationality of the prepared landslide susceptibility zoning map is analyzed in combination with actual conditions; a reference is provided for the possibility of landslide occurrence, and a scientific reference basis is provided for the government to treat landslide disasters.

A landslide susceptibility evaluation method based on a weighted information amount method comprises the following steps: step 1, collecting landslide basic data of a to-be-evaluated area, and obtaining landslide disaster data of the to-be-evaluated area; step 2, based on the data collected and obtained in the step 1, analyzing the landslide disaster distribution rule and the control influence factors of the region to obtain influence factors for controlling the development of the landslide disaster of the region, and obtaining influence conditions of different levels of influence factors on the landslide disaster; and 3, calculating the information amount of the influence factors by using a weighted information amount method based on the obtained main influence factors for controlling thedevelopment of the landslide disasters in the region and the influence conditions of the influence factors of different levels on the landslide disasters, and carrying out weighted superposition calculation on the influence factors based on a GIS platform to obtain a landslide disaster susceptibility evaluation result. And the influence of subjective factors of people is reduced, the principle issimple in actual work, operation is easy, and accuracy and high efficiency are achieved.

The invention relates to a regional landslide susceptibility prediction method based on a semi-supervised random forest model. The method comprises the following steps: S1, carrying out spatial analysis on landslide records and related control factors in a research region, and screening out known landslide samples; s2, based on the frequency ratio and correlation analysis, determining a control factor which can best represent landslide development characteristics, and establishing a random forest model; s3, on the basis of the FR values of the control factors, known landslide grid units and randomly selected non-landslide grid units, performing full-supervised machine learning, namely a random forest model, and according to the five types of landslide susceptibility grades in the step S2, outputting and predicting an initial landslide susceptibility value; s4, expanding the known landslide sample; s5, randomly selecting grid units from the extremely low susceptible areas as non-landslide samples; and S6, establishing a semi-supervised random forest model. According to the invention, the landslide susceptibility prediction modeling performance is further improved on the basis of fully supervised machine learning.

The invention provides a landslide disaster susceptibility space prediction method based on a clustering-information quantity coupling model. Clustering grading is carried out on evaluation indexes ofthe terrain evaluation factors by adopting a K-means algorithm, the internal relation of each terrain evaluation factor in a single evaluation unit is fully considered, and grading is carried out onthe distance type evaluation factors from the aspect of numerical value by adopting a natural breakpoint method; the information amount of each evaluation factor under different levels is calculatedaccording to the information amount model; and according to the sum of the information amounts of the grades of the evaluation factors of each evaluation unit, the landslide information amount of theevaluation unit is obtained so as to obtain the landslide information amounts of all the evaluation units in the map, and information amount grading is carried out in combination with a frequency ratio model so as to finally obtain a landslide susceptibility grade division map. The method has high spatial prediction precision of landslide disaster susceptibility, and can be used as a basis for disaster prevention and control and disaster treatment.

The invention relates to a local scale landslide susceptibility prediction method based on hierarchical Bayesian. The method mainly comprises four steps of extracting a basic evaluation unit; calculating the regional scale weight of the evaluation factor; establishing a spatial local logistic regression model, fitting spatial heterogeneity of the evaluation factor in a local scale, realizing information fusion of two scales under a hierarchical Bayesian framework, and verifying the model; and carrying out local-scale landslide susceptibility evaluation by utilizing the spatial local logistic regression model. According to the invention, the geographical detector is used to calculate the overall trend of each evaluation factor in the regional scale; fitting spatial heterogeneity of the evaluation factor contribution rate in a local scale by applying a spatial local regression model; and based on a hierarchical Bayesian framework, information fusion is carried out on the overall trend and spatial heterogeneity of the evaluation factor, so that the problem that the stability of the overall trend and heterogeneity of the local scale of the evaluation factor cannot be considered at thelocal scale is solved, and regional local scale landslide susceptibility evaluation is realized.

The invention discloses a landslide susceptibility evaluation method based on sample automatic selection and earth surface deformation rate, comprising the following steps: extracting evaluation factors to obtain earth surface deformation rate; automatically selecting landslide samples and non-landslide samples in different periods; performing normalization processing on the evaluation factors by using the selected landslide sample; training an sDAE-RF model by using the landslide sample and the non-landslide sample to obtain a landslide susceptibility initial probability; and correcting the initial probability of the landslide susceptibility by using the surface deformation rate to obtain a landslide susceptibility evaluation partition map. According to the method, the surface deformation rate reflecting the landslide motion state is utilized, so that the evaluation result is more consistent with the actual distribution condition of the regional landslide, the accuracy and practicability of the evaluation method are improved, and accurate decision data are provided for subsequent landslide disaster risk management.

The invention relates to a landslide susceptibility prediction method and system based on a semi-supervised support vector machine model. The method comprises the following steps of: acquiring an evaluation factor of a research area and a known landslide sample; calculating a deterministic coefficient value of each evaluation factor; carrying out susceptibility partitioning on the research area according to the deterministic coefficient value, and dividing the research area into five types of landslide susceptibility levels; selecting a non-landslide sample and known landslide sample to jointly form a first training test data set to perform training test on the support vector machine model; and adopting a trained and tested support vector machine model to predict an initial landslide susceptibility value of the research area, thereby determining an expanded landslide sample and a secondarily selected non-landslide sample to carry out training test on the support vector machine model, and further generating a semi-supervised support vector machine model to carry out landslide susceptibility prediction on the research area. According to the method, the landslide sample data acquisition process is simplified, and the precision of landslide susceptibility prediction is improved.

The invention relates to an earthquake disaster risk assessment method, and discloses an earthquake region landslide susceptibility assessment method based on multi-modal classification, which comprises the following steps: (1) collecting and preprocessing landslide disaster-causing factor characteristic data, and combining related landslide disaster-causing factor characteristic data according to the intrinsic attribute difference of the landslide disaster-causing factor characteristic data to obtain a landslide risk assessment result; and obtaining different types of landslide disaster-causing factor modal data; (2) obtaining similarity graphs of various landslide disaster-causing factor modal data by adopting a random forest principle; (3) fusing similarity graphs of the various landslide disaster-causing factor modal data by adopting a nonlinear fusion method so as to generate a similarity fusion graph; and (4) obtaining a landslide sensitivity graph through a related classification algorithm on the basis of the similarity fusion graph so as to evaluate the earthquake region landslide susceptibility. According to the earthquake region landslide susceptibility evaluation method based on multi-modal classification, earthquake landslide risk evaluation can be rapidly and accurately realized.

The embodiment of the invention discloses a landslide susceptibility prediction method and device, equipment and a storage medium. The method comprises the steps that at least one influence factor graph layer of landslide data of a to-be-predicted landslide area is acquired, wherein each influence factor graph layer comprises a grading result of influence factors corresponding to the to-be-predicted landslide area; and the at least one influence factor layer is input into a trained landslide susceptibility prediction model to obtain the probability of occurrence of landslide at each landslidesite in the to-be-predicted landslide area, wherein the landslide susceptibility prediction model is trained based on at least one historical landslide data. Therefore, the effect of quickly predicting the landslide susceptibility based on the plurality of influence factors in the whole landslide area is achieved.

The invention provides a landslide susceptibility evaluation model training method, evaluation method and device, and a medium. The training method comprises the following steps: collecting landslide evaluation factor data and landslide distribution data of a plurality of grid units in a specified area; marking the landslide evaluation factor data of each grid unit according to the landslide distribution data to obtain landslide sample marking data of each grid unit; for each grid unit, combining the landslide sample labeling data of the grid unit with the landslide sample labeling data of the adjacent grid unit to generate a landslide data matrix corresponding to the grid unit; and constructing a convolutional neural network, and training the convolutional neural network by adopting a data set formed by all the landslide data matrixes to obtain a landslide susceptibility evaluation model. According to the technical scheme of the invention, the comprehensiveness and accuracy of landslide susceptibility evaluation can be improved.

The invention discloses a landslide susceptibility evaluation method based on earth surface deformation and disaster-gestational environmental conditions, which utilizes dynamic factors to evaluate landslide susceptibility, and specifically comprises the following steps: acquiring landslide static factors and dynamic factors of a landslide susceptibility evaluation target area; according to the landslide static factor, establishing a random forest model to perform landslide susceptibility evaluation of the target area to obtain a first susceptibility evaluation result; establishing a concatenation matrix of the first susceptibility evaluation result and the dynamic factor to obtain a second susceptibility evaluation result; and comparing and integrating the first susceptibility evaluation result and the second susceptibility evaluation result through a GIS (Geographic Information System). The landslide dynamic factors are added into the landslide susceptibility evaluation model, so that the evaluation result is more accurate.

The invention discloses a prediction method for rock landslide susceptibility in a frozen-thawed zone of a high-altitude mountainous area. The prediction method solves the problem that in the prior art, long-term and real-time deformation data is obtained by relying on sensors additionally installed at specific sites of a slope to realize landslide prediction. According to the prediction method for rock landslide susceptibility in the frozen-thawed zone of the high-altitude mountainous area, based on surface air temperature data of a prediction place and an SAR image, a surface air temperaturechange period of the prediction place is divided according to the surface air temperature data, the time sequence-temperature data and time sequence-rock deformation quantity data at each stage of the change period are extracted, then the identified critical temperature index is used for determining a function relationship between the surface temperature change and the rock mass deformation, theaverage freezing temperature-rock deformation rate of the perdition place is calculated, finally, based on critical conditions of landslide generation, the degree of landslide susceptibility of the prediction place is judged, and the prediction of landslide susceptibility is realized. The prediction method solves the problem of landslide susceptibility prediction of a slope which does not have thecondition to implement sensor monitoring and early warning. The implementation cost is economic, and the perdition method is suitable for high-altitude areas.

The invention provides a landslide susceptibility evaluation method and device, equipment and a readable storage medium, and relates to the technical field of geological disaster prevention and control, and the method comprises the following steps: screening out landslide susceptibility disaster-inducing factors of a sample region; combining the landslide susceptibility disaster-inducing factors with a landslide sample set collected from the sample area to obtain a disaster-inducing factor-sample set; and taking the disaster-inducing factor-sample set as input data for training, and adopting a deep learning mode to obtain an ant colony optimization algorithm-deep belief network model for generating a landslide susceptibility result of the to-be-evaluated region. According to the invention, the deep belief network model and the ant colony optimization algorithm are combined and applied to the landslide susceptibility evaluation, and the landslide susceptibility disaster-inducing factors with weak correlation characteristics are optimized by screening out the landslide susceptibility disaster-inducing factors of the sample area, so that more values of the deep learning model in disaster prevention and reduction applications such as landslide susceptibility and the like are realized.

The invention discloses a landslide susceptibility evaluation method and system. The method comprises the steps of selecting evaluation factors based on basic data of a to-be-researched region; giving evaluation factor weights through a frequency ratio model; using SPSS software to carry out multi-collinearity diagnosis on the evaluation factors, and removing factors with multi-collinearity; based on the weight values of the residual evaluation factors, grading the geological disaster susceptibility through a logistic regression model; generating an average deformation rate map based on the SAR data; establishing an incidence matrix based on the average deformation rate diagram and the geological disaster susceptibility grade; and correcting the geological disaster susceptibility grade based on the relation matrix. On one hand, the frequency ratio-logistic regression coupling model is utilized to solve the defect that the precision of evaluating the landslide susceptibility by a single model is low, on the other hand, the InSAR is utilized to solve the limitation that a real historical landslide data set is difficult to obtain due to complex topographic conditions, and meanwhile, the InSAR result is utilized to optimize the grade of the landslide susceptibility.

The invention provides a landslide susceptibility evaluation model training method, evaluation method, device and medium. The training method includes: obtaining landslide evaluation factor data and landslide distribution data of multiple grid units in a designated area; data to label the landslide evaluation factor data of each of the grid units to obtain the landslide sample label data of each of the grid units; for each of the grid units, the landslide of the grid unit The sample annotation data is combined with the landslide sample annotation data of adjacent grid units to generate a landslide data matrix corresponding to the grid unit; a convolutional neural network is constructed using a data set composed of all the landslide data matrices Train the convolutional neural network to obtain a landslide susceptibility assessment model. The technical scheme of the invention can improve the comprehensiveness and accuracy of landslide susceptibility assessment.

The invention relates to the technical field of landslide disaster display, in particular to a landslide susceptibility model comparative analysis system based on big data and a display device.The landslide susceptibility model comparative analysis system comprises a bottom plate, a supporting base is fixedly connected to the top face of the bottom plate, a positioning shaft is rotatably inserted into the top of the supporting base, and rotary display mechanisms are arranged on the two sides of the supporting base; the rotary display mechanism comprises a transmission ring rotationally connected with the supporting seat, and the outer side of the transmission ring is fixedly connected with a plurality of first connecting shafts annularly at equal angles. According to the landslide susceptibility model test device, the rotary display mechanism drives the object source bodies to rotate, so that the object source bodies cut with different slope surfaces are contrasted and displayed in a multi-angle and all-directional manner, and the display effect of a landslide susceptibility model test is improved; the material source body sliding onto the material receiving hopper in the test process can be conveniently and directionally conveyed and collected, so that the cleanness and tidiness of a test display environment are ensured.

The invention discloses a landslide susceptibility distribution map determination method and device and a storage medium, and relates to the technical field of geology. The method screens geographic index parameters through relevance among various geographic index parameters so as to improve the accuracy of the determined landslide susceptibility distribution diagram. The method comprises the following steps: acquiring parameter values of at least two geographic index parameters of each grid unit of a to-be-predicted region; acquiring a correlation degree parameter between every two geographic index parameters in the at least two geographic index parameters through a preset correlation coefficient matrix model; screening out a target geographic index parameter from the at least two geographic index parameters based on a correlation degree parameter between every two geographic index parameters; and determining a landslide susceptibility distribution map of the to-be-predicted area according to the parameter value of the target geographic index parameter.

The invention discloses a landslide susceptibility evaluation method based on spatial logistic regression and a geographical detector. The method mainly comprises the following steps: extracting a basic evaluation unit; screening evaluation factors; establishing a spatial logistic regression model; evaluating a spatial logistic regression model. Moreover, the contribution degree of the influence factors in the slope units to the spatial distribution of the landslide is calculated through a geographic detector, the influence factors with the significant contribution degree to the spatial distribution of the landslide are selected as independent variables of the spatial logistic regression model, and then the regression coefficient of the spatial logistic regression model is solved by usingthe test data set. According to the invention, the spatial logistic regression model is established; according to the method, an influence factor with significant contribution degree of landslide spatial distribution is selected as an independent variable, attribute information and spatial structure information of spatial data are utilized, and a spatial autocorrelation effect is used as a potential information source to improve the model, so that the fitting degree and prediction precision of the model are significantly improved.

The landslide susceptibility evaluation method based on terrain unit solves the problem of poor reliability of the existing landslide susceptibility evaluation methods, and belongs to the field of regional geological disaster evaluation. The invention includes: determining the location of the landslide point and the relevant disaster-causing factor data; obtaining discrete intervals of the disaster-causing factor data, each discrete interval corresponds to a classification category, and using the intersection points and inflection points of the area classification ratio curve and the weight ratio index curve to conduct the interval analysis. Optimization; count the area of ​​each discrete interval after optimization and the number of corresponding landslide points to obtain the slip index; use the terrain surface model DEM to build a river network; build a terrain unit according to the river network; The degree index is assigned to each terrain unit; the random forest model is constructed, whose input is the data of each terrain unit, and the output is the landslide susceptibility probability value of each terrain unit.

The invention relates to the field of landslide disaster prediction, and discloses a shallow landslide susceptibility evaluation method based on a data model. According to the method, shallow landslide susceptibility evaluation is carried out by respectively simulating a damage area and a sliding area, shallow sliding damage is evaluated by using logistic regression (LR), and the LR also has the advantage of allowing integration of variables with abnormal distribution. Furthermore, the LR is insensitive to conditional dependencies of the evoked factors, which may result in an overestimated spatial probability value. For sliding evaluation, a simple CA model is provided. The main purpose of the invention is to construct a final shallow landslide sensitivity map including a failure area and a jumping area. Finally, the invention aims to realize the combination of a low-cost method and limited input data, so that the sensitivity evaluation has good performance and can be easily applied to other research fields.

The invention provides a landslide susceptibility evaluation method and device, electronic equipment, a storage medium and a product. The method comprises the following steps: acquiring a to-be-tested remote sensing image; inputting the to-be-tested remote sensing image into a landslide susceptibility evaluation model to obtain a landslide susceptibility score of the to-be-tested remote sensing image output by the landslide susceptibility evaluation model; wherein the landslide susceptibility evaluation model is obtained by training based on a landslide sample set and landslide susceptibility influence factors marked on the landslide sample set, and the landslide sample set is obtained based on a Gaussian thermodynamic diagram of the landslide. Smooth samples are added through the Gaussian thermodynamic diagram of the landslide, so that sample distribution is optimized, the performance and efficiency of the evaluation model are further improved, and a more accurate prediction result is obtained.

The invention discloses a landslide susceptibility evaluation method based on terrain units, solves the problem of poor reliability of an existing landslide susceptibility evaluation method, and belongs to the field of regional geological disaster evaluation. The method comprises the following steps: determining a landslide point position and related disaster-inducing factor data; acquiring discrete intervals of the disaster-inducing factor data, wherein each discrete interval corresponds to a classification category, and the intervals are optimized by adopting intersection points and inflection points of an area classification ratio curve and a weight ratio index curve; carrying out statistics on the optimized discrete interval area and the corresponding landslide point number, and obtaining a slip inclination index; constructing a river network by using a terrain surface model DEM; constructing a terrain unit according to the river network; assigning the slip degree index of each optimization interval of each disaster-inducing factor to each terrain unit; and establishing a random forest model, in which the input is the data of each terrain unit, and the output is the landslide susceptibility probability value of each terrain unit.

The invention relates to the technical field of geological disaster prediction, in particular to a landslide susceptibility prediction model based on principal component analysis and an extreme learning machine, and the model comprises the following steps: S1, obtaining landslide catalog and landslide susceptibility modeling related environmental factors of a research area; s2, carrying out dimensionality reduction on the environmental factors by utilizing principal component analysis, calculating a principal component score as an initial landslide susceptibility value, and dividing different susceptibility intervals; s3, superposing the extremely high incidence area and the remote sensing image, determining a landslide hidden danger point as an expanded landslide sample through visual interpretation, and forming a landslide sample by the landslide record and the expanded landslide sample; s4, grid units are randomly selected from the extremely low susceptible areas to serve as non-landslide samples; and S5, establishing an extreme learning machine prediction model. The correlation between environmental factors and repeated information reflected by the environmental factors during comprehensive evaluation can be eliminated. The redundancy of the data subjected to dimension reduction is greatly reduced through principal component analysis, and time is saved for subsequent calculation.