Geological landslide identification method and system

Abstract

The invention discloses a geological landslide identification method and system, and relates to the technical field of geological disaster monitoring. The method comprises the following steps: after acquiring a remote sensing image and laser radar scanning data of an area to which a target slope belongs, preprocessing based on the remote sensing image to obtain a stratum lithology label value, a vegetation coverage label value and a nearest water body distance value of each target pixel point; the method comprises the following steps: scanning data of a target slope, obtaining an elevation value, a slope value and a slope direction value of each target pixel point based on preprocessing of the scanning data, then importing the values into a landslide mass recognition model which completes pre-training based on an artificial intelligence algorithm to obtain landslide mass recognition results of each target pixel point, and summarizing to generate a landslide mass distribution binary image of the target slope; and finally, edge contour extraction processing, connected domain area calculation and area comparison are performed on a landslide mass connected domain in the binary image, and a landslide area surrounded by an edge contour is used as a geological landslide identification result when a comparison condition is satisfied, so that environmental protection and disaster reduction and prevention are facilitated.

Description

Technical Field

[0001] The invention belongs to the technical field of geological disaster monitoring, and in particular relates to a method and system for identifying geological landslides. Background Art

[0002] Under the influence of human activities and natural conditions, the rock or soil on the slope will lose its stability and then slide down along the damage surface, which is called a landslide disaster. Landslide geological disasters are very common, with the characteristics of mass, multiple, sudden and harmful. They have become a global environmental problem, seriously threatening the safety of people's lives and property. Its impact on human society is getting bigger and bigger and can no longer be ignored. Sorted by the degree of harm caused by natural disasters, landslide geological disasters have become the largest geological disaster besides earthquakes. The landslide disaster problem is closely related to people's living environment. Improving the level of research on landslide disaster prevention and control has important theoretical and engineering practical significance for environmental protection and disaster reduction and prevention.

[0003] At present, there are three main existing geological landslide identification schemes: (1) field surveys. The advantage is that the morphology of the landslide can be observed at close range and the various landslide elements can be delineated. However, the disadvantage is that the work efficiency is low and landslide surveys cannot be completed in difficult-to-reach areas. (2) Satellite remote sensing technology (i.e., "digital landslide"), which is divided into optical satellite remote sensing technology and imaging radar satellite remote sensing technology. Its disadvantages are that data collection is greatly restricted by weather conditions, it cannot penetrate vegetation, and its ability to identify landslides under vegetation cover is limited. In addition, it cannot simultaneously obtain meter-level precision DEM (Digital Elevation Model). (3) Aerial remote sensing technology, also known as aerial photogrammetry remote sensing technology, can achieve decimeter-level resolution in aerial images and generate DEM models from stereo images. It has achieved remarkable results in the field of landslide hazard identification, especially with the help of DEM models generated by stereo images. In a three-dimensional system, the relevant parameters of the landslide can be calculated more accurately. However, aerial images still cannot penetrate vegetation, and the recognition effect of landslide elements under vegetation cover or shadow is poor, especially for small deformations under such landform conditions.

[0004] Therefore, in the face of the above-mentioned problems existing in the existing geological landslide identification schemes, how to provide a new geological landslide identification scheme to improve work efficiency and the accuracy of landslide identification results, and thus benefit environmental protection and disaster reduction and prevention, is a topic that technical personnel in this field urgently need to study. Summary of the invention

[0005] The purpose of the present invention is to provide a method, system, computer device, computer readable storage medium and computer program product for identifying geological landslides, so as to solve the problems of low working efficiency, limited landslide identification ability and poor identification effect existing in existing geological landslide identification schemes.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, a method for identifying a geological landslide is provided, comprising:

[0008] Obtain remote sensing images and lidar scanning data of the area where the target slope belongs;

[0009] For each pixel point in the remote sensing image, the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels are imported together into a stratigraphic lithology classification model pre-trained based on the first artificial intelligence algorithm, and the corresponding stratigraphic lithology classification result is output, wherein the stratigraphic lithology classification result is used to indicate a stratigraphic lithology label value of the corresponding pixel point and one of multiple stratigraphic lithology label values, and the multiple stratigraphic lithology label values ​​correspond one to one to multiple stratigraphic types;

[0010] For each pixel point, the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels are imported together into a vegetation cover recognition model pre-trained based on a second artificial intelligence algorithm, and a corresponding vegetation cover recognition result is output, wherein the vegetation cover recognition result is used to indicate a vegetation cover label value of the corresponding pixel point, and the vegetation cover label value uses a value of 1 to indicate that it is covered by vegetation, and uses a value of 0 to indicate that it is not covered by vegetation;

[0011] For each pixel point, the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels are imported into a surface water body recognition model pre-trained based on a third artificial intelligence algorithm, and the corresponding surface water body recognition result is output, wherein the surface water body recognition result is used to indicate whether the corresponding pixel point is a water body pixel point;

[0012] For each target pixel point in the remote sensing image and located within the target slope, determine the water body pixel point closest to the corresponding pixel point according to the surface water body recognition result of each pixel point, and calculate the distance value between the water body pixel point and the corresponding pixel point;

[0013] Generating a digital elevation model of the area to which the target slope belongs based on the laser radar scanning data;

[0014] According to the digital elevation model, an elevation distribution image, a slope distribution image and a slope aspect distribution image of the area to which the target slope belongs are obtained, wherein the pixel value of each pixel point in the elevation distribution image is an elevation value, the pixel value of each pixel point in the slope distribution image is a slope value, and the pixel value of each pixel point in the slope aspect distribution image is a slope aspect value;

[0015] For each target pixel point, the corresponding stratum lithology label value, vegetation coverage label value, distance value, elevation value, slope value and aspect value as well as the stratum lithology label value, vegetation coverage label value, distance value, elevation value, slope value and aspect value of the corresponding surrounding target pixel points are imported into the landslide body recognition model pre-trained based on the fourth artificial intelligence algorithm, and the corresponding landslide body recognition result is output, wherein the landslide body recognition result is used to indicate whether the corresponding pixel point is a landslide body pixel point;

[0016] Generate a landslide body distribution binary image of the target slope according to the landslide body identification results of each target pixel point, wherein the landslide body distribution binary image uses a first pixel value to indicate that the corresponding pixel point is a landslide body pixel point, and uses a second pixel value to indicate that the corresponding pixel point is not a landslide body pixel point;

[0017] Performing edge contour extraction processing on the landslide body connected domain in the landslide body distribution binary image to obtain the edge contour of the landslide body connected domain;

[0018] According to the edge contour of the landslide body connected domain, the area of ​​the landslide body connected domain is calculated;

[0019] If the area of ​​the connected domain of the landslide body reaches a preset area threshold, the landslide area surrounded by the edge contour of the connected domain of the landslide body is taken as a geological landslide identification result.

[0020] Based on the above invention, a new scheme for geological landslide identification based on the hybrid of remote sensing images and lidar scanning data is provided, that is, after obtaining the remote sensing images and lidar scanning data of the area to which the target slope belongs, the stratigraphic lithology label value, vegetation coverage label value and the nearest water body distance value of each target pixel point are first obtained based on the remote sensing image preprocessing, and the elevation value, slope value and slope aspect value of each target pixel point are obtained based on the scanning data preprocessing, and then these values ​​are imported into the landslide body identification model pre-trained based on the artificial intelligence algorithm to obtain the landslide body identification value of each target pixel point. The identification results are summarized and a binary image of the landslide body distribution of the target slope is generated. Finally, the edge contour extraction, connected domain area calculation and area comparison are performed on the connected domain of the landslide body in the binary image. When the comparison conditions are met, the landslide area surrounded by the edge contour is used as the geological landslide identification result. This can not only improve work efficiency, but also make full use of the implicit information of landslide impact in remote sensing images and lidar scanning data by leveraging their strengths and avoiding their weaknesses, effectively improve the landslide identification ability, recognition accuracy and recognition effect, benefit environmental protection and disaster reduction and prevention, and facilitate practical application and promotion.

[0021] In one possible design, the first artificial intelligence algorithm, the second artificial intelligence algorithm, the third artificial intelligence algorithm or the fourth artificial intelligence algorithm adopts a machine learning algorithm based on support vector machine, stochastic gradient descent, multivariate linear regression, multilayer perceptron, decision tree, back propagation neural network or radial basis function network.

[0022] In one possible design, the plurality of strata types include Cambrian strata, Ordovician strata, Silurian strata, Devonian strata, Carboniferous strata, Permian strata and / or Triassic strata.

[0023] In a possible design, generating a digital elevation model of the area to which the target slope belongs based on the laser radar scanning data includes:

[0024] Using differential GPS technology, combined with dynamic GPS data of the airborne platform and data from a ground GPS reference station, the three-dimensional coordinates of the flight trajectory of the airborne platform are obtained by joint solution, wherein the airborne platform refers to an aircraft platform used to carry a laser radar scanning device to obtain the laser radar scanning data;

[0025] Integrate the inertial navigation data of the airborne platform into the three-dimensional coordinates of the flight trajectory, and then solve to obtain a new flight trajectory with the instantaneous position and attitude information of the laser radar scanning device;

[0026] Combining the new flight trajectory and the laser radar scanning data, and adding system error correction parameters and coordinate projection parameters, to obtain laser point cloud data with three-dimensional coordinates in the WGS-84 coordinate system;

[0027] Performing ground point extraction processing on the laser point cloud data to obtain ground point cloud data;

[0028] A digital elevation model of the area to which the target slope belongs is generated according to the ground point cloud data.

[0029] In a possible design, edge contour extraction processing is performed on the landslide body connected domain in the landslide body distribution binary image to obtain the edge contour of the landslide body connected domain, including:

[0030] The region growing method is used to remove holes from the landslide body connected domain in the landslide body distribution binary image to obtain a new landslide body connected domain;

[0031] The edge contour extraction process of the connected new domain of the landslide body is performed based on the Canny algorithm to obtain the edge contour of the connected domain of the landslide body.

[0032] In a possible design, after taking the landslide area defined by the edge contour of the landslide body connected domain as the geological landslide identification result, the method further includes:

[0033] Obtain a precipitation distribution image of the area to which the target slope belongs in the current most recent K historical unit time periods and the current next future unit time period, wherein the pixel value of each pixel point in the precipitation distribution image is the precipitation, and K represents a positive integer;

[0034] According to the precipitation distribution image, the average precipitation of all pixel points located in the landslide area in the current recent multiple historical unit time periods and the current next future unit time period is obtained by statistics;

[0035] According to the formation lithology classification results of the respective pixel points, the proportions of the various pixel points in the total number of pixel points are statistically obtained, wherein the various pixel points correspond one to one to the various formation types;

[0036] The connected domain area of ​​the landslide body, the average precipitation, the proportion of the multiple pixel points, and the average vegetation coverage label value, average distance value, average elevation value, average slope value, and average slope aspect value of all pixel points are imported into a landslide event prediction model pre-trained based on the fifth artificial intelligence algorithm, and the probability value of the landslide event occurring in the next future unit period is output;

[0037] If the probability value reaches a preset probability threshold, a landslide event warning action is triggered for the landslide area.

[0038] In a possible design, according to the formation lithology classification result of each pixel point, the proportion of multiple pixel points in all the pixel points is statistically obtained, including:

[0039] For each formation type among the multiple formation types, according to the formation lithology classification results of each pixel point, the number of pixel points with corresponding formation lithology label values ​​among all the pixel points is statistically obtained, and the ratio of this number to the total number of pixel points of all the pixel points is taken as the proportion of the corresponding pixel points among all the pixel points.

[0040] In a possible design, the landslide event prediction model is pre-trained as follows:

[0041] Acquire multiple negative sample data and multiple positive sample data corresponding to multiple historical landslide events, wherein the negative sample data contains a first model input item and a first model output item with a value of 0, the first model input item includes the landslide body connected domain area of ​​the first known landslide area, the average precipitation in the continuous K+1 historical unit time period when no landslide event occurs, the proportion of multiple pixel points in all pixel points located in the first known landslide area, and the average vegetation coverage label value, average distance value, average elevation value, average vegetation coverage label ... The positive sample data contains a second model input item and a second model output item with a value of 1, wherein the second model input item includes the landslide body connected domain area of ​​the second known landslide area, the average precipitation in the most recent K historical unit time periods before the corresponding landslide event and the historical unit time period when the corresponding landslide event occurred, the proportion of multiple pixel points in all pixel points located in the second known landslide area, and the average vegetation coverage label value, average distance value, average elevation value, average slope value and average aspect value of all pixel points located in the second known landslide area;

[0042] The multiple negative sample data and the multiple positive sample data are applied to calibrate and verify the machine learning model based on the fifth artificial intelligence algorithm to obtain the landslide event occurrence prediction model, wherein the landslide event occurrence prediction model and the confidence level with an output value of 1 are used as the probability value of a landslide event occurring in the next unit period.

[0043] In a second aspect, a geological landslide identification system is provided, including an image data acquisition unit, a stratigraphic lithology classification unit, a vegetation cover identification unit, a surface water body identification unit, a near-water distance calculation unit, a digital model generation unit, a distribution image acquisition unit, a landslide body identification unit, a binary image generation unit, an edge contour extraction unit, a connected domain area calculation unit, and a landslide area determination unit;

[0044] The image data acquisition unit is used to acquire remote sensing images and laser radar scanning data of the area to which the target slope belongs;

[0045] The stratigraphic lithology classification unit is communicatively connected to the image data acquisition unit, and is used for, for each pixel point in the remote sensing image, importing the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels into a stratigraphic lithology classification model pre-trained based on the first artificial intelligence algorithm, and outputting a corresponding stratigraphic lithology classification result, wherein the stratigraphic lithology classification result is used to indicate a stratigraphic lithology label value of the corresponding pixel point and one of the multiple stratigraphic lithology label values, and the multiple stratigraphic lithology label values ​​correspond one to one to multiple stratigraphic types;

[0046] The vegetation coverage identification unit is communicatively connected to the image data acquisition unit, and is used for importing the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels into the vegetation coverage identification model pre-trained based on the second artificial intelligence algorithm for each pixel point, and outputting the corresponding vegetation coverage identification result, wherein the vegetation coverage identification result is used to indicate the vegetation coverage label value of the corresponding pixel point, and the vegetation coverage label value uses a value of 1 to indicate that it is covered by vegetation, and uses a value of 0 to indicate that it is not covered by vegetation;

[0047] The surface water body recognition unit is communicatively connected to the image data acquisition unit, and is used for importing the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels into the surface water body recognition model pre-trained based on the third artificial intelligence algorithm for each pixel point, and outputting the corresponding surface water body recognition result, wherein the surface water body recognition result is used to indicate whether the corresponding pixel point is a water body pixel point;

[0048] The near-water distance calculation unit is communicatively connected to the surface water body identification unit, and is used to determine the water body pixel point closest to the corresponding pixel point according to the surface water body identification result of each pixel point in the remote sensing image and located in the target slope, and calculate the distance value between the water body pixel point and the corresponding pixel point;

[0049] The digital model generating unit is communicatively connected to the image data acquiring unit and is used to generate a digital elevation model of the area to which the target slope belongs according to the laser radar scanning data;

[0050] The distribution image acquisition unit is communicatively connected to the digital model generation unit, and is used to acquire the elevation distribution image, the slope distribution image and the slope aspect distribution image of the area to which the target slope belongs according to the digital elevation model, wherein the pixel value of each pixel point in the elevation distribution image is the elevation value, the pixel value of each pixel point in the slope distribution image is the slope value, and the pixel value of each pixel point in the slope aspect distribution image is the slope aspect value;

[0051] The landslide body identification unit is respectively communicatively connected to the stratum lithology classification unit, the vegetation cover identification unit, the near-water distance calculation unit and the distribution image acquisition unit, and is used for importing the corresponding stratum lithology label value, vegetation cover label value, distance value, elevation value, slope value and slope aspect value of each target pixel point and the stratum lithology label value, vegetation cover label value, distance value, elevation value, slope value and slope aspect value of the corresponding surrounding target pixel points into the landslide body identification model pre-trained based on the fourth artificial intelligence algorithm, and outputting the corresponding landslide body identification result, wherein the landslide body identification result is used to indicate whether the corresponding pixel point is a landslide body pixel point;

[0052] The binary image generating unit is communicatively connected to the landslide body identifying unit, and is used to generate a binary image of the landslide body distribution of the target slope according to the landslide body identification results of each target pixel point, wherein the binary image of the landslide body distribution uses a first pixel value to indicate that the corresponding pixel point is a landslide body pixel point, and uses a second pixel value to indicate that the corresponding pixel point is not a landslide body pixel point;

[0053] The edge contour extraction unit is communicatively connected to the binary image generation unit and is used to perform edge contour extraction processing on the landslide body connected domain in the landslide body distribution binary image to obtain the edge contour of the landslide body connected domain;

[0054] The connected domain area calculation unit is communicatively connected to the edge contour extraction unit and is used to calculate the connected domain area of ​​the landslide body according to the edge contour of the connected domain of the landslide body;

[0055] The landslide area determination unit is respectively communicatively connected to the edge contour extraction unit and the connected domain area calculation unit, and is used to use the landslide area surrounded by the edge contour of the connected domain of the landslide body as a geological landslide identification result if the area of ​​the connected domain of the landslide body reaches a preset area threshold.

[0056] In a possible design, it also includes a precipitation acquisition unit, a precipitation averaging unit, a lithology proportion statistics unit, a landslide event prediction unit and a warning action execution unit;

[0057] The precipitation acquisition unit is communicatively connected to the landslide area determination unit, and is used to acquire the precipitation distribution image of the area to which the target slope belongs in the current most recent K historical unit time periods and the current next future unit time period after taking the landslide area defined by the edge contour of the connected domain of the landslide body as the geological landslide identification result, wherein the pixel value of each pixel point in the precipitation distribution image is the precipitation, and K represents a positive integer;

[0058] The precipitation averaging unit is communicatively connected to the precipitation acquisition unit, and is used to obtain, based on the precipitation distribution image, the average precipitation of all pixels located in the landslide area in the current recent multiple historical unit time periods and the current next future unit time period;

[0059] The lithology proportion statistics unit is communicatively connected to the stratum lithology classification unit, and is used to obtain the proportion of multiple pixel points among all the pixel points according to the stratum lithology classification results of each pixel point, wherein the multiple pixel points correspond to multiple stratum types one by one;

[0060] The landslide event prediction unit is respectively communicatively connected to the connected domain area calculation unit, the precipitation averaging unit, the lithology proportion statistics unit, the vegetation coverage identification unit, the near-water distance calculation unit and the distribution image acquisition unit, and is used to import the connected domain area of ​​the landslide body, the average precipitation, the proportion of the multiple pixel points, and the average vegetation coverage label value, average distance value, average elevation value, average slope value and average slope aspect value of all pixel points into the landslide event occurrence prediction model pre-trained based on the fifth artificial intelligence algorithm, and output a probability value of a landslide event occurring in the current next future unit time period;

[0061] The early warning action execution unit is communicatively connected to the landslide event prediction unit, and is used to trigger the execution of a landslide event early warning action for the landslide area if the probability value reaches a preset probability threshold.

[0062] In a third aspect, the present invention provides a computer device comprising a memory, a processor and a transceiver which are communicatively connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the method for identifying a geological landslide as described in the first aspect or any possible design of the first aspect.

[0063] In a fourth aspect, the present invention provides a computer-readable storage medium having instructions stored thereon, which, when executed on a computer, executes the method for identifying a geological landslide as described in the first aspect or any possible design of the first aspect.

[0064] In a fifth aspect, the present invention provides a computer program product, comprising a computer program or instructions, which, when executed by a computer, implements the method for identifying a geological landslide as described in the first aspect or any possible design of the first aspect.

[0065] Beneficial effects of the above scheme:

[0066] (1) The invention provides a new scheme for identifying geological landslides based on the combination of remote sensing images and laser radar scanning data, that is, after obtaining remote sensing images and laser radar scanning data of the area to which the target slope belongs, firstly, the stratigraphic lithology label value, vegetation coverage label value and the nearest water body distance value of each target pixel point are obtained based on remote sensing image preprocessing, and the elevation value, slope value and slope aspect value of each target pixel point are obtained based on scanning data preprocessing, and then these values ​​are imported into the landslide body recognition model pre-trained based on artificial intelligence algorithm to obtain the landslide body recognition result of each target pixel point, and the landslide body distribution binary image of the target slope is generated by summarizing, and finally, the edge contour extraction processing and the connected domain area calculation and area comparison of the landslide body in the binary image are performed, and when the comparison conditions are met, the landslide area surrounded by the edge contour is used as the geological landslide recognition result, so that not only the work efficiency can be improved, but also the implicit information of landslide influence in remote sensing images and laser radar scanning data can be fully utilized by taking advantage of the advantages and avoiding the disadvantages, so as to effectively improve the landslide recognition ability, recognition accuracy and recognition effect, and benefit environmental protection and disaster reduction and prevention;

[0067] (2) After obtaining the geological landslide identification results, landslide monitoring can be carried out in real time and early warning can be triggered. In addition, the accuracy of the early warning timing can be ensured based on artificial intelligence algorithms, which is further beneficial to environmental protection and disaster reduction and prevention, and is convenient for practical application and promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0069] Figure 1 A schematic flow chart of a method for identifying a geological landslide provided in an embodiment of the present application.

[0070] Figure 2 A schematic diagram of the structure of a geological landslide identification system provided in an embodiment of the present application.

[0071] Figure 3 A schematic diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0072] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the structures of the drawings is only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can be obtained based on these embodiments without creative work. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0073] It should be understood that although the terms first and second, etc. may be used herein to describe various objects, these objects should not be limited by these terms. These terms are only used to distinguish one object from another object. For example, a first object can be referred to as a second object, and similarly, a second object can be referred to as a first object without departing from the scope of the exemplary embodiments of the present invention.

[0074] It should be understood that the term "and / or" that may appear in this document is merely a description of the association relationship between associated objects, indicating that there may be three relationships. For example, A and / or B can indicate three situations: A exists alone, B exists alone, or A and B exist at the same time. For another example, A, B and / or C can indicate the existence of any one of A, B and C or any combination of them. The term " / and" that may appear in this document describes another type of association object relationship, indicating that there may be two relationships. For example, A / and B can indicate two situations: A exists alone or A and B exist at the same time. In addition, the character " / " that may appear in this document generally indicates that the previous and next associated objects are in an "or" relationship.

[0075] Example

[0076] like Figure 1 As shown, the method for identifying geological landslides provided in the first aspect of this embodiment can be, but is not limited to, executed by a computer device with certain computing resources, such as a cloud server, an edge computer, a personal computer (PC, a multi-purpose computer with a size, price and performance suitable for personal use; desktops, laptops, small laptops, tablets and ultrabooks are all personal computers), a smart phone, a personal digital assistant (PDA) or a wearable device. Figure 1 As shown, the identification method may include but is not limited to the following steps S1 to S12.

[0077] S1. Obtain remote sensing images and lidar scanning data of the area where the target slope belongs.

[0078] In step S1, the region to which the target slope belongs is a region including the target slope, for example, a river basin region including a river and target slopes located on both sides of the river. The remote sensing image may be a satellite remote sensing image, or an aerial remote sensing image obtained based on an aircraft platform, which can be conventionally obtained. The laser radar scanning data is preferably an airborne laser radar scanning data obtained based on an aircraft platform, so as to improve the subsequent modeling accuracy, that is, since the laser radar signal can penetrate a certain vegetation cover to obtain pure surface information, and its data acquisition method can minimize the shadow effect caused by terrain cutting compared to aerial images, so a high-precision DEM model can be generated, which is conducive to improving the recognition accuracy of (potential) landslide bodies. In addition, the laser radar scanning data can be specifically but not limited to conventionally acquired by an ALS50-II airborne laser radar system (the flight altitude range of the system is 200 meters to 6000 meters, and the maximum scanning angle is 75°, the maximum laser pulse emission frequency is 150KHz, the maximum scanning frequency is 90Hz, and a maximum of four echoes can be received).

[0079] S2. For each pixel point in the remote sensing image, the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels are imported into a stratigraphic lithology classification model pre-trained based on the first artificial intelligence algorithm, and the corresponding stratigraphic lithology classification result is output, wherein the stratigraphic lithology classification result is used to indicate the corresponding pixel point and a stratigraphic lithology label value among multiple stratigraphic lithology label values, and the multiple stratigraphic lithology label values ​​correspond one-to-one to multiple formation types.

[0080] In the step S2, the surrounding pixels can be conventionally determined by covering the radius R, that is, if the second pixel is located in a circular area with the first pixel as the center and the radius R, then the second pixel is the surrounding pixel of the first pixel, otherwise the second pixel is not the surrounding pixel of the first pixel; in addition, when the aforementioned first pixel is an edge pixel, the pixel value of the first pixel can be, but is not limited to, used as the pixel value of the missing surrounding pixels. The aforementioned artificial intelligence algorithm is a core artificial intelligence algorithm that specifically studies how computers simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their own performance. It is the fundamental way to make computers intelligent; specifically, the first artificial intelligence algorithm can be, but is not limited to, specifically using machine learning algorithms based on support vector machines, stochastic gradient descent, multivariate linear regression, multilayer perceptron, decision tree, back propagation neural network or radial basis function network, etc., so as to quickly and accurately find the rules in the data. Therefore, the stratigraphic lithology classification model can be trained in advance based on a certain amount of sample data (e.g., the remote sensing pixel values ​​of pixels located in an exposed area of ​​a certain stratigraphic type and the remote sensing pixel values ​​of pixels around the pixel) and label data (e.g., the stratigraphic lithology label value of the certain stratigraphic type), through a conventional calibration verification modeling process (specifically including a calibration process and a verification process of the model, that is, first by comparing the model simulation results with the measured data, and then adjusting the model parameters according to the comparison results so that the simulation results are consistent with the actual process). Preferably, in the calibration verification modeling process of the stratigraphic lithology classification model, a Bayesian optimization algorithm based on a tree structure is used to tune the model parameters. In addition, specifically, the multiple stratigraphic types include but are not limited to Cambrian stratigraphy, Ordovician stratigraphy, Silurian stratigraphy, Devonian stratigraphy, Carboniferous stratigraphy, Permian stratigraphy and / or Triassic stratigraphy.

[0081] S3. For each pixel point, the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels are imported into a vegetation cover recognition model pre-trained based on a second artificial intelligence algorithm, and the corresponding vegetation cover recognition result is output, wherein the vegetation cover recognition result is used to indicate the vegetation cover label value of the corresponding pixel point, and the vegetation cover label value uses a value of 1 to indicate that it is covered by vegetation, and uses a value of 0 to indicate that it is not covered by vegetation.

[0082] In the step S3, specifically, the second artificial intelligence algorithm may also be, but not limited to, a machine learning algorithm based on support vector machine, stochastic gradient descent, multivariate linear regression, multilayer perceptron, decision tree, back propagation neural network or radial basis function network, and may also be based on a certain amount of positive sample data (e.g., remote sensing pixel values ​​of pixels located in vegetation coverage areas and remote sensing pixel values ​​of pixels surrounding the pixel) and positive label data (e.g., vegetation coverage label values ​​with a value of 1) and negative sample data (e.g., remote sensing pixel values ​​of pixels located in non-vegetation coverage areas and remote sensing pixel values ​​of pixels surrounding the pixel) and negative label data (e.g., vegetation coverage label values ​​with a value of 0), and the vegetation coverage recognition model is obtained by training through a conventional calibration verification modeling process. In addition, preferably, in the calibration verification modeling process of the vegetation coverage recognition model, a tree-structured Bayesian optimization algorithm may also be used to tune the model parameters.

[0083] S4. For each pixel point, the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels are imported into a surface water body recognition model pre-trained based on a third artificial intelligence algorithm, and the corresponding surface water body recognition result is output, wherein the surface water body recognition result is used to indicate whether the corresponding pixel point is a water body pixel point.

[0084] In the step S4, specifically, the third artificial intelligence algorithm may also be, but not limited to, a machine learning algorithm based on support vector machine, stochastic gradient descent, multivariate linear regression, multilayer perceptron, decision tree, back propagation neural network or radial basis function network, and may also be based on a certain amount of positive sample data (e.g., remote sensing pixel values ​​of pixels located in water areas such as river or lake surface and remote sensing pixel values ​​of pixels surrounding the pixel) and positive label data (e.g., water body label values ​​with a value of 1) and negative sample data (e.g., remote sensing pixel values ​​of pixels located in non-water areas and remote sensing pixel values ​​of pixels surrounding the pixel) and negative label data (e.g., water body label values ​​with a value of 0), and the surface water body recognition model may be obtained through conventional calibration verification modeling process training. In addition, preferably, in the calibration verification modeling process of the surface water body recognition model, a tree-structured Bayesian optimization algorithm may also be used to tune the model parameters.

[0085] S5. For each target pixel point in the remote sensing image and located within the target slope, determine the water body pixel point closest to the corresponding pixel point based on the surface water body identification result of each pixel point, and calculate the distance value between the water body pixel point and the corresponding pixel point.

[0086] In step S5, the target slope is, for example, a mountain slope located on both sides of the river, which can be pre-delineated manually. In addition, the specific method of determining the nearest water pixel point can be conventionally achieved by traversing and calculating the distance between pixel points and comparing the distances.

[0087] S6. Generate a digital elevation model of the area to which the target slope belongs based on the laser radar scanning data.

[0088] In step S6, the digital elevation model is a digital simulation of the ground terrain (i.e., a digital expression of the terrain surface morphology) achieved through limited terrain elevation data. It is a physical ground model that represents the ground elevation in the form of a set of ordered numerical arrays, and can describe the spatial distribution of linear and nonlinear combinations of various geomorphic factors including elevation, such as slope, aspect, and slope change rate. When the laser radar scanning data is airborne laser radar scanning data obtained based on an aircraft platform, specifically, based on the laser radar scanning data, a digital elevation model of the area to which the target slope belongs is generated, including but not limited to the following steps S601 to S606.

[0089] S601. Using differential GPS (Global Positioning System) technology, combined with the dynamic GPS data of the airborne platform and the data of the ground GPS reference station, the three-dimensional coordinates of the flight trajectory of the airborne platform are obtained by joint solution, wherein the airborne platform refers to an aircraft platform used to carry a laser radar scanning device to obtain the laser radar scanning data.

[0090] In step S601, the differential GPS technology (Differential Global Positioning System, abbreviated as DGPS, i.e. differential global positioning system) is an existing technology that installs a GPS monitoring receiver at a precise known location (such as a base station) and calculates the distance correction number between the base station and the GPS satellite. Therefore, the three-dimensional coordinates of the flight trajectory of the airborne platform can be conventionally jointly solved.

[0091] S602. Integrate the inertial navigation data of the airborne platform into the three-dimensional coordinates of the flight trajectory, and then solve to obtain a new flight trajectory with the instantaneous position and attitude information of the laser radar scanning device.

[0092] S603. Combine the new flight trajectory and the laser radar scanning data, add system error correction parameters and coordinate projection parameters, and solve to obtain laser point cloud data with three-dimensional coordinates in the WGS-84 coordinate system.

[0093] S604. Perform ground point extraction processing on the laser point cloud data to obtain ground point cloud data.

[0094] S605. Generate a digital elevation model of the area to which the target slope belongs based on the ground point cloud data.

[0095] S7. According to the digital elevation model, obtain the elevation distribution image, slope distribution image and slope aspect distribution image of the area to which the target slope belongs, wherein the pixel value of each pixel point in the elevation distribution image is the elevation value, the pixel value of each pixel point in the slope distribution image is the slope value, and the pixel value of each pixel point in the slope aspect distribution image is the slope aspect value.

[0096] In step S7, since the digital elevation model can describe the spatial distribution of linear and nonlinear combinations of various geomorphic factors including elevation, such as slope, aspect and slope change rate, the elevation distribution image, the slope distribution image and the aspect distribution image can be conventionally obtained based on the digital elevation model. In addition, in order to ensure that their pixels have a one-to-one correspondence with the pixels in the remote sensing image, the elevation distribution image, the slope distribution image and the aspect distribution image need to have the same size as the remote sensing image.

[0097] S8. For each target pixel point, the corresponding stratum lithology label value, vegetation coverage label value, distance value, elevation value, slope value and aspect value as well as the stratum lithology label value, vegetation coverage label value, distance value, elevation value, slope value and aspect value of the corresponding surrounding target pixel points are imported into the landslide body recognition model pre-trained based on the fourth artificial intelligence algorithm, and the corresponding landslide body recognition result is output, wherein the landslide body recognition result is used to indicate whether the corresponding pixel point is a landslide body pixel point.

[0098] In step S8, according to existing research, stratigraphic lithology, vegetation coverage, distance to the nearest water body, elevation, slope and slope direction are important factors affecting the occurrence of geological landslides. Therefore, the stratigraphic lithology label value, vegetation coverage label value, distance value, elevation value, slope value and slope direction value of the target pixel point and its surrounding pixels are used as model input items to determine whether the target pixel point is a landslide body pixel point through the landslide body identification model. Specifically, the fourth artificial intelligence algorithm may also be, but is not limited to, a machine learning algorithm based on support vector machine, stochastic gradient descent method, multivariate linear regression, multilayer perceptron, decision tree, back propagation neural network or radial basis function network, and may also be based on a certain amount of positive sample data (e.g., remote sensing pixel values ​​of pixels located in a known landslide area and the remote sensing pixel values ​​of pixels surrounding the pixel) and positive label data (e.g., a landslide body label value of 1) and negative sample data (e.g., remote sensing pixel values ​​of pixels located in a known non-landslide area and the remote sensing pixel values ​​of pixels surrounding the pixel) and negative label data (e.g., a landslide body label value of 0), and obtain the landslide body recognition model through conventional calibration verification modeling process training. In addition, preferably, in the calibration verification modeling process of the landslide body recognition model, a tree-structured Bayesian optimization algorithm may also be used to tune the model parameters.

[0099] S9. Generate a binary image of the landslide body distribution of the target slope according to the landslide body identification results of each target pixel point, wherein the binary image of the landslide body distribution uses a first pixel value to indicate that the corresponding pixel point is a landslide body pixel point, and uses a second pixel value to indicate that the corresponding pixel point is not a landslide body pixel point.

[0100] In the step S9, for example, the first pixel value is a value of 255, and the second pixel value is a value of 0.

[0101] S10. Performing edge contour extraction processing on the landslide body connected domain in the landslide body distribution binary image to obtain the edge contour of the landslide body connected domain.

[0102] In the step S10, in order to eliminate holes in the connected domain and improve the accuracy of edge contour extraction, preferably, edge contour extraction processing is performed on the landslide body connected domain in the landslide body distribution binary image to obtain the edge contour of the landslide body connected domain, including but not limited to the following steps: S101. The landslide body connected domain in the landslide body distribution binary image is subjected to hole removal processing by using the region growing method to obtain a new landslide body connected domain; S102. The landslide body connected new domain is subjected to edge contour extraction processing based on the Canny algorithm to obtain the edge contour of the landslide body connected domain. The aforementioned region growing method and Canny algorithm are both prior art means. In addition, the number of landslide body connected domains in the landslide body distribution binary image can be one or more, and when there are more than one, the edge contour extraction processing needs to be performed on each landslide body connected domain respectively to obtain the corresponding landslide body connected domain edge contour.

[0103] S11. Calculate the area of ​​the connected domain of the landslide body according to the edge contour of the connected domain of the landslide body.

[0104] S12. If the area of ​​the connected domain of the landslide body reaches a preset area threshold, the landslide area surrounded by the edge contour of the connected domain of the landslide body is used as a geological landslide identification result.

[0105] In the step S12, the preset area threshold may be predetermined based on historical experience. In addition, if the area of ​​the landslide body connected domain does not reach the preset area threshold, the corresponding landslide body connected domain may be regarded as noise and excluded.

[0106] Therefore, based on the recognition method described in the aforementioned steps S1 to S12, a new solution for geological landslide recognition based on the hybrid remote sensing image and LiDAR scanning data is provided, that is, after obtaining the remote sensing image and LiDAR scanning data of the area to which the target slope belongs, the stratigraphic lithology label value, vegetation coverage label value and the nearest water body distance value of each target pixel point are first obtained based on the remote sensing image preprocessing, and the elevation value, slope value and slope aspect value of each target pixel point are obtained based on the scanning data preprocessing, and then these values ​​are imported into the landslide body recognition model pre-trained based on the artificial intelligence algorithm to obtain the stratigraphic lithology label value, vegetation coverage label value and the nearest water body distance value of each target pixel point. The landslide body identification results of the primitive points are summarized and a binary image of the landslide body distribution of the target slope is generated. Finally, the edge contour extraction, connected domain area calculation and area comparison are performed on the connected domain of the landslide body in the binary image. When the comparison conditions are met, the landslide area surrounded by the edge contour is used as the geological landslide identification result. This can not only improve work efficiency, but also make full use of the implicit information of landslide impact in remote sensing images and lidar scanning data by leveraging their strengths and avoiding their weaknesses, effectively improve the landslide identification ability, recognition accuracy and recognition effect, benefit environmental protection and disaster reduction and prevention, and facilitate practical application and promotion.

[0107] Based on the technical solution of the first aspect, this embodiment also provides a possible design for how to conduct landslide monitoring and trigger early warning in real time after obtaining the geological landslide identification result. That is, after the landslide area defined by the edge contour of the connected domain of the landslide body is used as the geological landslide identification result, the method also includes but is not limited to the following steps S131 to S135.

[0108] S131. Obtain a precipitation distribution image of the area to which the target slope belongs in the current most recent K historical unit time periods and the current next future unit time period, wherein the pixel value of each pixel point in the precipitation distribution image is the precipitation, and K represents a positive integer.

[0109] In step S131, the unit time period may be, but is not limited to, 1 hour, 1 hour, half a day or 1 day, etc. The current K most recent historical unit time periods need to include the unit time period currently being experienced; for example, if the current unit time period is 12:00 on the 5th and the value of K is 12, then it is necessary to obtain the precipitation distribution image of the area to which the target slope belongs from 1:00 on the 5th to 13:00 on the 5th, where 1:00 on the 5th to 12:00 on the 5th is the current most recent K historical unit time periods, 13:00 on the 5th is the current next future unit time period, and the precipitation distribution image at 13:00 on the 5th is conventional weather forecast data. In addition, the precipitation distribution image can be obtained specifically from a conventional query on a weather forecast server.

[0110] S132. According to the precipitation distribution image, the average precipitation of all pixel points located in the landslide area in the current recent multiple historical unit time periods and the current next future unit time period is obtained by statistics.

[0111] In step S132, it is necessary to obtain the average precipitation of all pixels in the landslide area in the corresponding period for each unit period in the current recent multiple historical unit periods and the current next future unit period, so as to form an average precipitation time series data. In addition, the specific statistical method is an existing conventional technical means.

[0112] S133. According to the formation lithology classification results of the respective pixel points, the proportions of the various pixel points in all the pixel points are statistically obtained, wherein the various pixel points correspond one-to-one to the various formation types.

[0113] In the step S133, specifically, according to the stratigraphic lithology classification results of the respective pixel points, the proportions of the various pixel points among all the pixel points are statistically obtained, including but not limited to the following steps: for various stratigraphic types among the multiple stratigraphic types, according to the stratigraphic lithology classification results of the respective pixel points, the number of pixel points among all the pixel points and having corresponding stratigraphic lithology label values ​​is statistically obtained, and the ratio of the number to the total number of pixel points of all the pixel points is used as the proportion of the corresponding pixel points among all the pixel points. For example, if the multiple stratum types include Cambrian strata, Ordovician strata, Silurian strata, Devonian strata, Carboniferous strata, Permian strata and Triassic strata, and the total number of pixel points of all pixel points is 216, the number of pixel points with stratum lithology label values ​​corresponding to the Cambrian strata is 10, the number of pixel points with stratum lithology label values ​​corresponding to the Ordovician strata is 33, the number of pixel points with stratum lithology label values ​​corresponding to the Silurian strata is 4, and the number of pixel points with stratum lithology label values ​​corresponding to the Devonian strata is 216. The number of pixel points with stratigraphic lithology label values ​​corresponding to the basin-system strata is 0, the number of pixel points with stratigraphic lithology label values ​​corresponding to the Carboniferous strata is 27, the number of pixel points with stratigraphic lithology label values ​​corresponding to the Permian strata is 54, and the number of pixel points with stratigraphic lithology label values ​​corresponding to the Triassic strata is 88. It can be statistically obtained that the proportions of various pixel points in all the pixel points are 0.046, 0.153, 0.019, 0, 0.125, 0.25 and 0.407 respectively.

[0114] S134. The connected domain area of ​​the landslide body, the average precipitation, the proportion of the multiple pixel points, and the average vegetation coverage label value, average distance value, average elevation value, average slope value and average slope aspect value of all pixel points are imported into the landslide event prediction model pre-trained based on the fifth artificial intelligence algorithm, and the probability value of the occurrence of a landslide event in the current next future unit time period is output.

[0115] In the step S134, the average vegetation coverage label value, average distance value, average elevation value, average slope value and average slope direction value of all the pixels can be obtained based on the vegetation coverage recognition results of each pixel, the distance value corresponding to each target pixel, the elevation distribution image, the slope distribution image and the slope direction distribution image. Since the landslide area, precipitation, stratum lithology, whether vegetation is covered, the distance to the nearest water body, elevation, slope and slope direction are important factors that trigger geological landslide events, the area of ​​the landslide body connected domain, the average precipitation, the proportion of the multiple pixel points and the average vegetation coverage label value, average distance value, average elevation value, average slope value and average slope direction value of all the pixels are used as model input items, so as to predict the probability of a landslide event occurring in the current next future unit period through the landslide event occurrence prediction model.

[0116] In the step S134, specifically, the fifth artificial intelligence algorithm may also be but not limited to a machine learning algorithm based on support vector machine, stochastic gradient descent, multivariate linear regression, multilayer perceptron, decision tree, back propagation neural network or radial basis function network, and in detail, the landslide event prediction model is pre-trained according to the following steps S1341 to S1342.

[0117] S1341. Acquire multiple negative sample data and multiple positive sample data corresponding to multiple historical landslide events, wherein the negative sample data contains a first model input item and a first model output item with a value of 0, the first model input item includes the landslide body connected domain area of ​​the first known landslide area, the average precipitation in the continuous K+1 historical unit time period when no landslide event occurs, the proportion of multiple pixel points in all pixel points located in the first known landslide area, and the average vegetation coverage label value, average distance value, and average elevation value of all pixel points located in the first known landslide area. , an average slope value and an average slope aspect value, the positive sample data contains a second model input item and a second model output item with a value of 1, the second model input item includes the landslide body connected domain area of ​​the second known landslide area, the average precipitation in the most recent K historical unit time periods before the corresponding landslide event and the historical unit time period when the corresponding landslide event occurred, the proportion of multiple pixel points in all pixel points located in the second known landslide area, and the average vegetation coverage label value, average distance value, average elevation value, average slope value and average slope aspect value of all pixel points located in the second known landslide area.

[0118] In step S1341, the specific method for obtaining the above-mentioned average precipitation, the proportion of multiple pixel points, the average vegetation coverage label value, the average distance value, the average elevation value, the average slope value and the average slope aspect value can be obtained by referring to the conventional derivation of the previous steps, and will not be repeated here.

[0119] S1342. Apply the multiple negative sample data and the multiple positive sample data to calibrate and verify the machine learning model based on the fifth artificial intelligence algorithm to obtain the landslide event prediction model, wherein the landslide event prediction model and the confidence level of the output value of 1 are used as the probability value of the landslide event occurring in the next unit period.

[0120] S135. If the probability value reaches a preset probability threshold, triggering a landslide event warning action for the landslide area.

[0121] In step S135, the preset probability threshold can also be predetermined based on historical experience. In addition, if the probability value does not reach the preset probability threshold, the landslide event warning action is not triggered, and the landslide event warning action is specifically but not limited to an audible and visual alarm action.

[0122] Therefore, based on the aforementioned possible design one, landslide monitoring can be carried out in real time and early warning can be triggered after obtaining the geological landslide identification results, and the accuracy of the early warning timing can also be ensured based on artificial intelligence algorithms, which is further beneficial to environmental protection and disaster reduction and prevention.

[0123] like Figure 2 As shown, the second aspect of this embodiment provides a virtual system for implementing the identification method described in the first aspect or any possible design of the first aspect, including an image data acquisition unit, a stratum lithology classification unit, a vegetation cover identification unit, a surface water body identification unit, a near-water distance calculation unit, a digital model generation unit, a distribution image acquisition unit, a landslide body identification unit, a binary image generation unit, an edge contour extraction unit, a connected domain area calculation unit and a landslide area determination unit;

[0124] The image data acquisition unit is used to acquire remote sensing images and laser radar scanning data of the area to which the target slope belongs;

[0125] The stratigraphic lithology classification unit is communicatively connected to the image data acquisition unit, and is used for, for each pixel point in the remote sensing image, importing the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels into a stratigraphic lithology classification model pre-trained based on the first artificial intelligence algorithm, and outputting a corresponding stratigraphic lithology classification result, wherein the stratigraphic lithology classification result is used to indicate a stratigraphic lithology label value of the corresponding pixel point and one of the multiple stratigraphic lithology label values, and the multiple stratigraphic lithology label values ​​correspond one to one to multiple stratigraphic types;

[0126] The vegetation coverage identification unit is communicatively connected to the image data acquisition unit, and is used for importing the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels into the vegetation coverage identification model pre-trained based on the second artificial intelligence algorithm for each pixel point, and outputting the corresponding vegetation coverage identification result, wherein the vegetation coverage identification result is used to indicate the vegetation coverage label value of the corresponding pixel point, and the vegetation coverage label value uses a value of 1 to indicate that it is covered by vegetation, and uses a value of 0 to indicate that it is not covered by vegetation;

[0127] The surface water body recognition unit is communicatively connected to the image data acquisition unit, and is used for importing the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels into the surface water body recognition model pre-trained based on the third artificial intelligence algorithm for each pixel point, and outputting the corresponding surface water body recognition result, wherein the surface water body recognition result is used to indicate whether the corresponding pixel point is a water body pixel point;

[0128] The near-water distance calculation unit is communicatively connected to the surface water body identification unit, and is used to determine the water body pixel point closest to the corresponding pixel point according to the surface water body identification result of each pixel point in the remote sensing image and located in the target slope, and calculate the distance value between the water body pixel point and the corresponding pixel point;

[0129] The digital model generating unit is communicatively connected to the image data acquiring unit and is used to generate a digital elevation model of the area to which the target slope belongs according to the laser radar scanning data;

[0130] The distribution image acquisition unit is communicatively connected to the digital model generation unit, and is used to acquire the elevation distribution image, the slope distribution image and the slope aspect distribution image of the area to which the target slope belongs according to the digital elevation model, wherein the pixel value of each pixel point in the elevation distribution image is the elevation value, the pixel value of each pixel point in the slope distribution image is the slope value, and the pixel value of each pixel point in the slope aspect distribution image is the slope aspect value;

[0131] The landslide body identification unit is respectively communicatively connected to the stratum lithology classification unit, the vegetation cover identification unit, the near-water distance calculation unit and the distribution image acquisition unit, and is used for importing the corresponding stratum lithology label value, vegetation cover label value, distance value, elevation value, slope value and slope aspect value of each target pixel point and the stratum lithology label value, vegetation cover label value, distance value, elevation value, slope value and slope aspect value of the corresponding surrounding target pixel points into the landslide body identification model pre-trained based on the fourth artificial intelligence algorithm, and outputting the corresponding landslide body identification result, wherein the landslide body identification result is used to indicate whether the corresponding pixel point is a landslide body pixel point;

[0132] The binary image generating unit is communicatively connected to the landslide body identifying unit, and is used to generate a binary image of the landslide body distribution of the target slope according to the landslide body identification results of each target pixel point, wherein the binary image of the landslide body distribution uses a first pixel value to indicate that the corresponding pixel point is a landslide body pixel point, and uses a second pixel value to indicate that the corresponding pixel point is not a landslide body pixel point;

[0133] The edge contour extraction unit is communicatively connected to the binary image generation unit and is used to perform edge contour extraction processing on the landslide body connected domain in the landslide body distribution binary image to obtain the edge contour of the landslide body connected domain;

[0134] The connected domain area calculation unit is communicatively connected to the edge contour extraction unit and is used to calculate the connected domain area of ​​the landslide body according to the edge contour of the connected domain of the landslide body;

[0135] The landslide area determination unit is respectively communicatively connected to the edge contour extraction unit and the connected domain area calculation unit, and is used to use the landslide area surrounded by the edge contour of the connected domain of the landslide body as a geological landslide identification result if the area of ​​the connected domain of the landslide body reaches a preset area threshold.

[0136] In a possible design, it also includes a precipitation acquisition unit, a precipitation averaging unit, a lithology proportion statistics unit, a landslide event prediction unit and a warning action execution unit;

[0137] The precipitation acquisition unit is communicatively connected to the landslide area determination unit, and is used to acquire the precipitation distribution image of the area to which the target slope belongs in the current most recent K historical unit time periods and the current next future unit time period after taking the landslide area defined by the edge contour of the connected domain of the landslide body as the geological landslide identification result, wherein the pixel value of each pixel point in the precipitation distribution image is the precipitation, and K represents a positive integer;

[0138] The precipitation averaging unit is communicatively connected to the precipitation acquisition unit, and is used to obtain, based on the precipitation distribution image, the average precipitation of all pixels located in the landslide area in the current recent multiple historical unit time periods and the current next future unit time period;

[0139] The lithology proportion statistics unit is communicatively connected to the stratum lithology classification unit, and is used to obtain the proportion of multiple pixel points among all the pixel points according to the stratum lithology classification results of each pixel point, wherein the multiple pixel points correspond to multiple stratum types one by one;

[0140] The landslide event prediction unit is respectively communicatively connected to the connected domain area calculation unit, the precipitation averaging unit, the lithology proportion statistics unit, the vegetation coverage identification unit, the near-water distance calculation unit and the distribution image acquisition unit, and is used to import the connected domain area of ​​the landslide body, the average precipitation, the proportion of the multiple pixel points, and the average vegetation coverage label value, average distance value, average elevation value, average slope value and average slope aspect value of all pixel points into the landslide event occurrence prediction model pre-trained based on the fifth artificial intelligence algorithm, and output a probability value of a landslide event occurring in the current next future unit time period;

[0141] The early warning action execution unit is communicatively connected to the landslide event prediction unit, and is used to trigger the execution of a landslide event early warning action for the landslide area if the probability value reaches a preset probability threshold.

[0142] The working process, working details and technical effects of the aforementioned system provided in the second aspect of this embodiment can be referred to the first aspect or any possible design of the identification method described in the first aspect, and will not be repeated here.

[0143] like Figure 3As shown, the third aspect of the present embodiment provides a computer device for executing the identification method described in any possible design as in the first aspect or the first aspect, including a memory, a processor and a transceiver connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program, and execute the identification method described in any possible design as in the first aspect or the first aspect. Specifically, the memory may include, but is not limited to, a random access memory (Random-Access Memory, RAM), a read-only memory (Read-Only Memory, ROM), a flash memory (Flash Memory), a first-in-first-out memory (First Input First Output, FIFO) and / or a first-in-last-out memory (First Input Last Output, FILO), etc.; the processor may include, but is not limited to, a microprocessor of the STM32F105 series. In addition, the computer device may also include, but is not limited to, a power module, a display screen and other necessary components.

[0144] The working process, working details and technical effects of the aforementioned computer device provided in the third aspect of this embodiment can be referred to the identification method described in the first aspect or any possible design of the first aspect, and will not be repeated here.

[0145] The fourth aspect of this embodiment provides a computer-readable storage medium storing instructions including the identification method described in the first aspect or any possible design of the first aspect, that is, the computer-readable storage medium stores instructions, and when the instructions are run on a computer, the identification method described in the first aspect or any possible design of the first aspect is executed. The computer-readable storage medium refers to a carrier for storing data, which may include but is not limited to computer-readable storage media such as floppy disks, optical disks, hard disks, flash memories, USB flash drives, and / or memory sticks, and the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0146] The working process, working details and technical effects of the aforementioned computer-readable storage medium provided in the fourth aspect of this embodiment can be referred to the identification method described in the first aspect or any possible design of the first aspect, and will not be repeated here.

[0147] A fifth aspect of this embodiment provides a computer program product, including a computer program or instructions, wherein the computer program or instructions, when executed by a computer, implements the identification method as described in the first aspect or any possible design of the first aspect. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.

[0148] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for identifying a geological landslide, characterized in that: include: Obtain remote sensing images and lidar scanning data of the area where the target slope belongs; For each pixel point in the remote sensing image, the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels are imported together into a stratigraphic lithology classification model pre-trained based on the first artificial intelligence algorithm, and the corresponding stratigraphic lithology classification result is output, wherein the stratigraphic lithology classification result is used to indicate a stratigraphic lithology label value of the corresponding pixel point and one of multiple stratigraphic lithology label values, and the multiple stratigraphic lithology label values ​​correspond one to one to multiple stratigraphic types; For each pixel point, the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels are imported together into a vegetation cover recognition model pre-trained based on a second artificial intelligence algorithm, and a corresponding vegetation cover recognition result is output, wherein the vegetation cover recognition result is used to indicate a vegetation cover label value of the corresponding pixel point, and the vegetation cover label value uses a value of 1 to indicate that it is covered by vegetation, and uses a value of 0 to indicate that it is not covered by vegetation; For each pixel point, the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels are imported into a surface water body recognition model pre-trained based on a third artificial intelligence algorithm, and the corresponding surface water body recognition result is output, wherein the surface water body recognition result is used to indicate whether the corresponding pixel point is a water body pixel point; For each target pixel point in the remote sensing image and located within the target slope, determine the water body pixel point closest to the corresponding pixel point according to the surface water body recognition result of each pixel point, and calculate the distance value between the water body pixel point and the corresponding pixel point; Generating a digital elevation model of the area to which the target slope belongs based on the laser radar scanning data; According to the digital elevation model, an elevation distribution image, a slope distribution image and a slope aspect distribution image of the area to which the target slope belongs are obtained, wherein the pixel value of each pixel point in the elevation distribution image is an elevation value, the pixel value of each pixel point in the slope distribution image is a slope value, and the pixel value of each pixel point in the slope aspect distribution image is a slope aspect value; For each target pixel point, the corresponding stratum lithology label value, vegetation coverage label value, distance value, elevation value, slope value and aspect value as well as the stratum lithology label value, vegetation coverage label value, distance value, elevation value, slope value and aspect value of the corresponding surrounding target pixel points are imported into the landslide body recognition model pre-trained based on the fourth artificial intelligence algorithm, and the corresponding landslide body recognition result is output, wherein the landslide body recognition result is used to indicate whether the corresponding pixel point is a landslide body pixel point; Generate a landslide body distribution binary image of the target slope according to the landslide body identification results of each target pixel point, wherein the landslide body distribution binary image uses a first pixel value to indicate that the corresponding pixel point is a landslide body pixel point, and uses a second pixel value to indicate that the corresponding pixel point is not a landslide body pixel point; Performing edge contour extraction processing on the landslide body connected domain in the landslide body distribution binary image to obtain the edge contour of the landslide body connected domain; According to the edge contour of the landslide body connected domain, the area of ​​the landslide body connected domain is calculated; If the area of ​​the connected domain of the landslide body reaches a preset area threshold, the landslide area surrounded by the edge contour of the connected domain of the landslide body is taken as a geological landslide identification result.

2. The identification method according to claim 1, characterized in that: The first artificial intelligence algorithm, the second artificial intelligence algorithm, the third artificial intelligence algorithm or the fourth artificial intelligence algorithm adopts a machine learning algorithm based on support vector machine, stochastic gradient descent, multivariate linear regression, multilayer perceptron, decision tree, back propagation neural network or radial basis function network.

3. The identification method according to claim 1, characterized in that: The multiple strata types include Cambrian strata, Ordovician strata, Silurian strata, Devonian strata, Carboniferous strata, Permian strata and / or Triassic strata.

4. The identification method according to claim 1, characterized in that: Generating a digital elevation model of the area to which the target slope belongs according to the laser radar scanning data, including: Using differential GPS technology, combined with dynamic GPS data of the airborne platform and data from a ground GPS reference station, the three-dimensional coordinates of the flight trajectory of the airborne platform are obtained by joint solution, wherein the airborne platform refers to an aircraft platform used to carry a laser radar scanning device to obtain the laser radar scanning data; Integrate the inertial navigation data of the airborne platform into the three-dimensional coordinates of the flight trajectory, and then solve to obtain a new flight trajectory with the instantaneous position and attitude information of the laser radar scanning device; Combining the new flight trajectory and the laser radar scanning data, and adding system error correction parameters and coordinate projection parameters, to obtain laser point cloud data with three-dimensional coordinates in the WGS-84 coordinate system; Performing ground point extraction processing on the laser point cloud data to obtain ground point cloud data; A digital elevation model of the area to which the target slope belongs is generated according to the ground point cloud data.

5. The identification method according to claim 1, characterized in that: The edge contour extraction process is performed on the landslide body connected domain in the landslide body distribution binary image to obtain the edge contour of the landslide body connected domain, including: The region growing method is used to remove holes from the landslide body connected domain in the landslide body distribution binary image to obtain a new landslide body connected domain; The edge contour extraction process of the connected new domain of the landslide body is performed based on the Canny algorithm to obtain the edge contour of the connected domain of the landslide body.

6. The identification method according to claim 1, characterized in that: After taking the landslide area defined by the edge contour of the landslide body connected domain as the geological landslide identification result, the method further includes: Obtain a precipitation distribution image of the area to which the target slope belongs in the current most recent K historical unit time periods and the current next future unit time period, wherein the pixel value of each pixel point in the precipitation distribution image is the precipitation, and K represents a positive integer; According to the precipitation distribution image, the average precipitation of all pixel points located in the landslide area in the current recent multiple historical unit time periods and the current next future unit time period is obtained by statistics; According to the formation lithology classification results of the respective pixel points, the proportions of the various pixel points in the total number of pixel points are statistically obtained, wherein the various pixel points correspond one to one to the various formation types; The connected domain area of ​​the landslide body, the average precipitation, the proportion of the multiple pixel points, and the average vegetation coverage label value, average distance value, average elevation value, average slope value, and average slope aspect value of all pixel points are imported into a landslide event prediction model pre-trained based on the fifth artificial intelligence algorithm, and the probability value of the landslide event occurring in the next future unit period is output; If the probability value reaches a preset probability threshold, a landslide event warning action is triggered for the landslide area.

7. The identification method according to claim 6, characterized in that: According to the formation lithology classification results of each pixel point, the proportions of various pixel points in all the pixel points are statistically obtained, including: For each formation type among the multiple formation types, according to the formation lithology classification results of each pixel point, the number of pixel points with corresponding formation lithology label values ​​among all the pixel points is statistically obtained, and the ratio of this number to the total number of pixel points of all the pixel points is taken as the proportion of the corresponding pixel points among all the pixel points.

8. The identification method according to claim 6, characterized in that: The landslide event prediction model is pre-trained in the following way: Acquire multiple negative sample data and multiple positive sample data corresponding to multiple historical landslide events, wherein the negative sample data contains a first model input item and a first model output item with a value of 0, the first model input item includes the landslide body connected domain area of ​​the first known landslide area, the average precipitation in the continuous K+1 historical unit time period when no landslide event occurs, the proportion of multiple pixel points in all pixel points located in the first known landslide area, and the average vegetation coverage label value, average distance value, average elevation value, average vegetation coverage label ... The positive sample data contains a second model input item and a second model output item with a value of 1, wherein the second model input item includes the landslide body connected domain area of ​​the second known landslide area, the average precipitation in the most recent K historical unit time periods before the corresponding landslide event and the historical unit time period when the corresponding landslide event occurred, the proportion of multiple pixel points in all pixel points located in the second known landslide area, and the average vegetation coverage label value, average distance value, average elevation value, average slope value and average aspect value of all pixel points located in the second known landslide area; The multiple negative sample data and the multiple positive sample data are applied to calibrate and verify the machine learning model based on the fifth artificial intelligence algorithm to obtain the landslide event occurrence prediction model, wherein the landslide event occurrence prediction model and the confidence level with an output value of 1 are used as the probability value of a landslide event occurring in the next unit period.

9. A geological landslide identification system, characterized in that: It includes an image data acquisition unit, a stratum lithology classification unit, a vegetation cover identification unit, a surface water body identification unit, a near-water distance calculation unit, a digital model generation unit, a distribution image acquisition unit, a landslide body identification unit, a binary image generation unit, an edge contour extraction unit, a connected domain area calculation unit and a landslide area determination unit; The image data acquisition unit is used to acquire remote sensing images and laser radar scanning data of the area to which the target slope belongs; The stratigraphic lithology classification unit is communicatively connected to the image data acquisition unit, and is used for, for each pixel point in the remote sensing image, importing the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels into a stratigraphic lithology classification model pre-trained based on the first artificial intelligence algorithm, and outputting a corresponding stratigraphic lithology classification result, wherein the stratigraphic lithology classification result is used to indicate a stratigraphic lithology label value of the corresponding pixel point and one of the multiple stratigraphic lithology label values, and the multiple stratigraphic lithology label values ​​correspond one to one to multiple stratigraphic types; The vegetation coverage identification unit is communicatively connected to the image data acquisition unit, and is used for importing the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels into the vegetation coverage identification model pre-trained based on the second artificial intelligence algorithm for each pixel point, and outputting the corresponding vegetation coverage identification result, wherein the vegetation coverage identification result is used to indicate the vegetation coverage label value of the corresponding pixel point, and the vegetation coverage label value uses a value of 1 to indicate that it is covered by vegetation, and uses a value of 0 to indicate that it is not covered by vegetation; The surface water body recognition unit is communicatively connected to the image data acquisition unit, and is used for importing the corresponding pixel value and the pixel values ​​of the corresponding surrounding pixels into the surface water body recognition model pre-trained based on the third artificial intelligence algorithm for each pixel point, and outputting the corresponding surface water body recognition result, wherein the surface water body recognition result is used to indicate whether the corresponding pixel point is a water body pixel point; The near-water distance calculation unit is communicatively connected to the surface water body identification unit, and is used to determine the water body pixel point closest to the corresponding pixel point according to the surface water body identification result of each pixel point in the remote sensing image and located in the target slope, and calculate the distance value between the water body pixel point and the corresponding pixel point; The digital model generating unit is communicatively connected to the image data acquiring unit and is used to generate a digital elevation model of the area to which the target slope belongs according to the laser radar scanning data; The distribution image acquisition unit is communicatively connected to the digital model generation unit, and is used to acquire the elevation distribution image, the slope distribution image and the slope aspect distribution image of the area to which the target slope belongs according to the digital elevation model, wherein the pixel value of each pixel point in the elevation distribution image is the elevation value, the pixel value of each pixel point in the slope distribution image is the slope value, and the pixel value of each pixel point in the slope aspect distribution image is the slope aspect value; The landslide body identification unit is respectively communicatively connected to the stratum lithology classification unit, the vegetation cover identification unit, the near-water distance calculation unit and the distribution image acquisition unit, and is used for importing the corresponding stratum lithology label value, vegetation cover label value, distance value, elevation value, slope value and slope aspect value of each target pixel point and the stratum lithology label value, vegetation cover label value, distance value, elevation value, slope value and slope aspect value of the corresponding surrounding target pixel points into the landslide body identification model pre-trained based on the fourth artificial intelligence algorithm, and outputting the corresponding landslide body identification result, wherein the landslide body identification result is used to indicate whether the corresponding pixel point is a landslide body pixel point; The binary image generating unit is communicatively connected to the landslide body identifying unit, and is used to generate a binary image of the landslide body distribution of the target slope according to the landslide body identification results of each target pixel point, wherein the binary image of the landslide body distribution uses a first pixel value to indicate that the corresponding pixel point is a landslide body pixel point, and uses a second pixel value to indicate that the corresponding pixel point is not a landslide body pixel point; The edge contour extraction unit is communicatively connected to the binary image generation unit and is used to perform edge contour extraction processing on the landslide body connected domain in the landslide body distribution binary image to obtain the edge contour of the landslide body connected domain; The connected domain area calculation unit is communicatively connected to the edge contour extraction unit and is used to calculate the connected domain area of ​​the landslide body according to the edge contour of the connected domain of the landslide body; The landslide area determination unit is respectively communicatively connected to the edge contour extraction unit and the connected domain area calculation unit, and is used to use the landslide area surrounded by the edge contour of the connected domain of the landslide body as a geological landslide identification result if the area of ​​the connected domain of the landslide body reaches a preset area threshold.

10. The identification system according to claim 9, characterized in that It also includes a precipitation acquisition unit, a precipitation averaging unit, a lithology proportion statistics unit, a landslide event prediction unit and a warning action execution unit; The precipitation acquisition unit is communicatively connected to the landslide area determination unit, and is used to acquire the precipitation distribution image of the area to which the target slope belongs in the current most recent K historical unit time periods and the current next future unit time period after taking the landslide area defined by the edge contour of the connected domain of the landslide body as the geological landslide identification result, wherein the pixel value of each pixel point in the precipitation distribution image is the precipitation, and K represents a positive integer; The precipitation averaging unit is communicatively connected to the precipitation acquisition unit, and is used to obtain, based on the precipitation distribution image, the average precipitation of all pixels located in the landslide area in the current recent multiple historical unit time periods and the current next future unit time period; The lithology proportion statistics unit is communicatively connected to the stratum lithology classification unit, and is used to obtain the proportion of multiple pixel points among all the pixel points according to the stratum lithology classification results of each pixel point, wherein the multiple pixel points correspond to multiple stratum types one by one; The landslide event prediction unit is respectively communicatively connected to the connected domain area calculation unit, the precipitation averaging unit, the lithology proportion statistics unit, the vegetation coverage identification unit, the near-water distance calculation unit and the distribution image acquisition unit, and is used to import the connected domain area of ​​the landslide body, the average precipitation, the proportion of the multiple pixel points, and the average vegetation coverage label value, average distance value, average elevation value, average slope value and average slope aspect value of all pixel points into the landslide event occurrence prediction model pre-trained based on the fifth artificial intelligence algorithm, and output a probability value of a landslide event occurring in the current next future unit time period; The early warning action execution unit is communicatively connected to the landslide event prediction unit, and is used to trigger the execution of a landslide event early warning action for the landslide area if the probability value reaches a preset probability threshold.

Images