Regional debris flow residual slope product slope source identification model determination method, identification method and related device

Abstract

The invention discloses a regional debris flow residual slope product slope source identification model determination method, an identification method and a related device, and relates to the field of geological disaster identification and risk assessment. The method comprises the following steps: firstly, acquiring a multispectral optical satellite image, a high-precision digital elevation model, an unmanned aerial vehicle aerial orthoimage and unmanned aerial vehicle live-action three-dimensional image data of a debris flow gully in a target area; dividing a debris flow gully in a target area based on a high-precision digital elevation model to obtain a plurality of slope units; calculating a topographic factor, an average vegetation coverage and an effective migration space of each slope unit; a target region is determined through a residual slope product slope source target region judgment model and the parameters, a typical debris flow gully is selected in the target region, and then residual slope product and non-residual slope product slope sources are screened out in combination with obtained data; and finally, constructing a sample library based on the comprehensive remote sensing data and the screening result, and training a machine learning model to obtain a regional recognition model. According to the invention, automatic identification of the residual slope product slope material source can be realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of geological disaster identification and risk assessment, in particular to a regional debris flow residual slope deposition slope surface source identification model determination method, an identification method and related devices. BACKGROUND

[0002] As a highly destructive geological disaster, debris flow is often accompanied by the release of huge energy, which can instantly destroy the infrastructure along the way, such as houses, roads, bridges, etc., and even threaten people's lives and safety. With the intensification of global climate change and the increasing frequency of human activities interfering with the natural environment, the frequency and severity of debris flow disasters are showing an upward trend.

[0003] In the process of debris flow disaster prevention, accurate identification and quantification of residual slope deposition slope surface sources is one of the key links. However, the current identification method mainly relies on manual field investigation and experiential estimation, which has many limitations. First of all, manual investigation is highly subjective, and the judgment standards of different investigators may differ, making it difficult to ensure the accuracy and consistency of the investigation results. Secondly, manual investigation is inefficient and difficult to complete the investigation task of a large area in a short time, especially in complex terrain mountainous areas, the investigation is more difficult. Thirdly, the coverage of manual investigation is limited, and it is difficult to systematically investigate all potential debris flow gullies, which may miss some important residual slope deposition slope surface source areas. In addition, manual investigation cannot be dynamically updated, and cannot timely reflect the changes of the source. These limitations result in insufficient accuracy of residual slope deposition slope surface source investigation results, which is difficult to meet the needs of refined risk assessment.

[0004] More importantly, current research on residual slope deposition slope surface sources mainly focuses on the dynamic reserves that are easily observed, have participated or are extremely easy to participate in debris flow activities, while the systematic identification of static reserves is relatively weak, and the depth of research is obviously insufficient. Static reserves are an important part of potential debris flow sources, and their neglect leads to significant uncertainty in the understanding of the maximum disaster scale and long-term evolution risk of debris flow. SUMMARY

[0005] The purpose of the present application is to provide a regional debris flow residual slope deposition slope surface source identification model determination method, an identification method and related devices, which can comprehensively, accurately and efficiently realize the automatic identification of residual slope deposition slope surface sources.

[0006] To achieve the above purpose, the present application provides the following solutions:

[0007] In a first aspect, the present application provides a regional debris flow residual slope deposition slope surface source identification model determination method, comprising:

[0008] Obtaining comprehensive remote sensing data of the target area debris flow gully, the comprehensive remote sensing data comprising: multispectral optical satellite image data, high-precision digital elevation model data, unmanned aerial vehicle aerial orthographic image data, and unmanned aerial vehicle real scene three-dimensional image data;

[0009] Dividing the target area debris flow gully based on the high-precision digital elevation model data to obtain a plurality of slope units;

[0010] Based on the high-precision digital elevation model data, calculating the terrain factor and effective moving space of each slope unit, and based on the multispectral optical satellite image data, calculating the average vegetation coverage of each slope unit;

[0011] Based on the residual slope surface material source target area discrimination model and the terrain factor, average vegetation coverage and effective moving space of each slope unit, determining the residual slope surface material source target area in the target area debris flow gully;

[0012] Selecting a typical debris flow gully in the residual slope surface material source target area, and based on the typical debris flow gully, the unmanned aerial vehicle aerial orthographic image data, the high-precision digital elevation model data and the unmanned aerial vehicle real scene three-dimensional image data, screening out residual slope surface material sources and non-residual slope surface material sources from the typical debris flow gully;

[0013] Based on the comprehensive remote sensing data of the typical debris flow gully and the screened out residual slope surface material sources and non-residual slope surface material sources vector range, constructing a sample library;

[0014] Training a machine learning model according to the sample library to obtain a regional debris flow residual slope surface material source identification model; the regional debris flow residual slope surface material source identification model is used to identify residual slope surface material sources of a debris flow gully in a to-be-detected area.

[0015] In a second aspect, the present application provides a regional debris flow residual slope surface material source identification method, comprising:

[0016] Obtaining comprehensive remote sensing data of a to-be-detected area debris flow gully;

[0017] Inputting the comprehensive remote sensing data of the to-be-detected area debris flow gully into a regional debris flow residual slope surface material source identification model to obtain residual slope surface material sources;

[0018] The regional debris flow residual slope surface material source identification model is a model determined based on the above-mentioned regional debris flow residual slope surface material source identification model determination method.

[0019] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to implement the steps of the method for determining the source identification model of regional debris flow colluvial slope as described above.

[0020] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for determining the source identification model of regional debris flow colluvial slope as described above.

[0021] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method for determining the source identification model of regional debris flow colluvial slope as described above.

[0022] According to the specific embodiments provided in this application, this application has the following technical effects:

[0023] This application provides a method, identification method, and related apparatus for determining the source of debris flow colluvial slopes in regional debris flow. It utilizes multiple high-precision remote sensing data, including multispectral optical satellite imagery, UAV aerial orthophotos, high-precision digital elevation model data, and UAV real-scene 3D imagery, to provide detailed topographic, vegetation, and surface information. This avoids the subjectivity, inefficiency, and limited coverage of manual field surveys. Furthermore, this multi-source data fusion provides more comprehensive and accurate surface information, effectively identifying debris sources in both vegetated and non-vegetated areas, thus improving the method's applicability. Through machine learning models, automated identification of debris sources on colluvial slopes is achieved, reducing manual intervention and improving identification efficiency and accuracy.

[0024] By constructing a target area discrimination model for residual colluvial slope sediments, this application can systematically identify static reserves. This model comprehensively considers multiple factors, including topographic factors, vegetation cover, and effective transport space, enabling a more comprehensive assessment of potential source areas and overcoming the shortcomings of existing studies in identifying static reserves on residual colluvial slopes. Furthermore, the dynamic updating capability of remote sensing data can promptly reflect changes in the source, thus providing a scientific basis for the identification of static reserves. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 An application environment diagram of a regional debris flow residual slope accumulation slope surface material source identification model determination method in an embodiment of the present application;

[0027] Figure 2 A flowchart of a regional debris flow residual slope accumulation slope surface material source identification model determination method provided in an embodiment of the present application;

[0028] Figure 3 A remote sensing interpretation sign diagram of a residual slope accumulation slope surface material source in a certain typical debris flow gully in an embodiment of the present application; Figure 3 In the middle (a), the residual slope accumulation slope surface material source is an interpretation sign in an unmanned aerial vehicle aerial orthographic image; Figure 3 In the middle (b), the residual slope accumulation slope surface material source is an interpretation sign in a mountain shadow grid image; Figure 3 In the middle (c), the residual slope accumulation slope surface material source is a slope gradient distribution range; Figure 3 In the middle (d), the residual slope accumulation slope surface material source is an interpretation sign in a mountain shadow grid image; Figure 3 In the middle (a), a profile of a line connecting two points A and B in an unmanned aerial vehicle orthographic image;

[0029] Figure 4 An unmanned aerial vehicle aerial orthographic image and a mountain shadow grid image after a residual slope accumulation slope surface material source area is circled in an embodiment of the present application; Figure 4 In the middle (a), the unmanned aerial vehicle aerial orthographic image after the residual slope accumulation slope surface material source area is circled, Figure 4 In the middle (b), the mountain shadow grid image after the residual slope accumulation slope surface material source area is circled;

[0030] Figure 5 A plurality of images after an unmanned aerial vehicle aerial orthographic image and a mountain shadow grid image are cropped and segmented in an embodiment of the present application; Figure 5 In the middle (a), the plurality of images after the unmanned aerial vehicle aerial orthographic image is cropped and segmented; Figure 5 In the middle (b), the plurality of images after the mountain shadow grid image is cropped and segmented;

[0031] Figure 6 A label data diagram in an embodiment of the present application;

[0032] Figure 7 A flowchart of a regional debris flow residual slope accumulation slope surface material source identification method provided in an embodiment of the present application;

[0033] Figure 8 A structural diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0035] The above purposes, features and advantages of the present application will be more apparent and understandable. The present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] Applicants have found that in the traditional method, the identification of the residual slope surface source is mainly based on artificial field investigation and empirical estimation. This method is highly subjective to artificial investigation, which makes it difficult to guarantee the accuracy and consistency of the survey results, and is inefficient, making it difficult to complete the survey task of a large area in a short time. Secondly, the coverage of artificial investigation is limited, and it is difficult to systematically investigate all potential debris flow gullies, which may miss some important source areas. Thirdly, artificial investigation is difficult to update dynamically, and cannot reflect the changes of the source in time, resulting in insufficient accuracy of the survey results of the residual slope surface source. In addition, the current research on the residual slope surface source mainly focuses on the dynamic reserves of the easily observed, involved or easily involved debris flow activities, and the depth of systematic identification of static reserves is obviously insufficient. The neglect of this makes the cognition of the maximum disaster scale and long-term evolution risk of debris flow have great uncertainty. Therefore, there is an urgent need for a method that can comprehensively, accurately and efficiently realize the automatic identification of the residual slope surface source of debris flow.

[0037] The regional debris flow residual slope surface source identification model determination method provided by the embodiments of the present application can be applied to, for example Figure 1The application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data required by the server 104 to process. The data storage system can be set up separately, or integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the obtained comprehensive remote sensing data of the target area debris flow gully to the server 104. After the server 104 receives the comprehensive remote sensing data of the target area debris flow gully, the server 104 divides the target area debris flow gully based on the high-precision digital elevation model data to obtain a plurality of slope units; based on the high-precision digital elevation model data, calculate the terrain factor and effective moving space of each slope unit, and based on the multispectral optical satellite image data, calculate the average vegetation coverage of each slope unit; based on the residual slope surface material source target area discrimination model and the terrain factor, average vegetation coverage and effective moving space of each slope unit, determine the residual slope surface material source target area in the target area debris flow gully; select a typical debris flow gully in the residual slope surface material source target area, and based on the typical debris flow gully, the unmanned aerial orthographic image data, the high-precision digital elevation model data and the unmanned aerial real scene three-dimensional image data, screen the residual slope surface material source and the non-residual slope surface material source from the typical debris flow gully; based on the comprehensive remote sensing data of the typical debris flow gully and the screened residual slope surface material source and non-residual slope surface material source vector range, construct a sample library; train a machine learning model according to the sample library to obtain a regional debris flow residual slope surface material source identification model of the target area. The server 104 can feed back the obtained regional debris flow residual slope surface material source identification model to the terminal 102. In addition, in some embodiments, the regional debris flow residual slope surface material source identification model determination method can also be implemented by the server 104 or the terminal 102 alone, such as the terminal 102 can directly process the obtained comprehensive remote sensing data of the target area debris flow gully, or the server 104 can obtain the comprehensive remote sensing data of the target area debris flow gully from the data storage system and process it.

[0038] Among them, the terminal 102 can be but not limited to various desktop computers, notebook computers and Internet of Things devices. The server 104 can be implemented by an independent server or a server cluster composed of multiple servers, and can also be a cloud server.

[0039] In an exemplary embodiment, as Figure 2 shown, a regional debris flow residual slope surface material source identification model determination method is provided, which is executed by a computer device, specifically by a terminal or a server computer device alone, or by a terminal and a server together. In the embodiments of the present application, the method is applied to the server 104 in Figure 1 The following steps 201 to 207 are described. Among them:

[0040] Step 201, obtaining comprehensive remote sensing data of the target area debris flow gully, the comprehensive remote sensing data comprising: multispectral optical satellite image data, high-precision digital elevation model data, unmanned aerial vehicle aerial orthographic image data and unmanned aerial vehicle real scene three-dimensional image data.

[0041] Step 202, dividing the target area debris flow gully based on the high-precision digital elevation model data to obtain a plurality of slope units.

[0042] Step 203, calculating the terrain factor and effective migration space of each slope unit based on the high-precision digital elevation model data, and calculating the average vegetation coverage of each slope unit based on the multispectral optical satellite image data.

[0043] Step 204, determining the residual slope surface material source target area in the target area debris flow gully based on the residual slope surface material source target area discrimination model and the terrain factor, average vegetation coverage and effective migration space of each slope unit.

[0044] Step 205, selecting a typical debris flow gully in the residual slope surface material source target area, and screening the residual slope surface material source and the non-residual slope surface material source from the typical debris flow gully based on the typical debris flow gully, the unmanned aerial vehicle aerial orthographic image data, the high-precision digital elevation model data and the unmanned aerial vehicle real scene three-dimensional image data.

[0045] Step 206, constructing a sample library based on the comprehensive remote sensing data of the typical debris flow gully and the screened residual slope surface material source and non-residual slope surface material source.

[0046] Step 207, training a machine learning model according to the sample library to obtain a regional debris flow residual slope surface material source identification model.

[0047] Implementing the above steps 201 to 207 can obtain a regional debris flow residual slope surface material source identification model, which can accurately capture the spatial distribution characteristics and static reserve differences of the residual slope surface material source when identifying the residual slope surface material source, and can comprehensively and efficiently complete the residual slope surface material source investigation and dynamic monitoring of a large area, improve the accuracy, consistency and timeliness of the investigation results, and provide strong support for regional debris flow residual slope surface material source identification.

[0048] In another exemplary embodiment of the present application, the above step 202 is replaced by steps 301-302.

[0049] Step 301: using a terrain feature line extraction algorithm to process the high-precision digital elevation model data to obtain ridge lines, valley lines and profile curvature zero points.

[0050] Step 302: dividing the target area debris flow gully based on the ridge line, the valley line and the profile curvature zero point, to obtain a plurality of slope units.

[0051] In the embodiment, taking a certain regional debris flow gully in the western region of Sichuan Province as an example, the comprehensive remote sensing data of the target area debris flow gully includes: Landsat 8 multispectral optical satellite image data obtained through the official website of geographic space data cloud, unmanned aerial vehicle aerial orthographic image data with a spatial resolution of 0.2 m, high-precision digital elevation model data with a spatial resolution of 0.5 m, and unmanned aerial vehicle real scene three-dimensional image data with a spatial resolution of 0.2 m obtained by using a flying horse unmanned aerial vehicle and a matching sensor.

[0052] Based on the high-precision digital elevation model data with a spatial resolution of 0.5 m, the ridge line and the valley line of the target area debris flow gully are extracted by using ArcGIS10.8 software, and the target area debris flow gully is divided based on the ridge line, the valley line and the profile curvature zero point of the target area debris flow gully, to obtain a plurality of slope units.

[0053] In another exemplary embodiment of the application, one slope unit contains a plurality of pixels; and the above step 203 is replaced by steps 401-405.

[0054] Step 401: based on the high-precision digital elevation model data, the slope and the profile curvature of each pixel in each slope unit are calculated.

[0055] Step 402: the slope and the profile curvature of all pixels in each slope unit are calculated respectively by using an arithmetic mean algorithm, to obtain the first-order terrain factor average slope and the second-order terrain factor average profile curvature of each slope unit; wherein the terrain factor includes the first-order terrain factor average slope and the second-order terrain factor average profile curvature.

[0056] Step 403: based on the multispectral optical satellite image data, the vegetation coverage of each pixel in each slope unit is calculated.

[0057] Step 404: the vegetation coverage of all pixels in each slope unit is calculated by using an arithmetic mean algorithm, to obtain the average vegetation coverage of each slope unit.

[0058] Step 405: based on the high-precision digital elevation model data, the effective migration space of each slope unit is calculated.

[0059] In the embodiment, the high-precision digital elevation model data obtained in the previous exemplary embodiment is used to calculate the first-order terrain factor slope A1 and the second-order terrain factor profile curvature B1 of each pixel in each slope unit by using ArcGIS 10.8 software; the slope and the profile curvature of all pixels in a slope unit are calculated respectively to obtain the first-order terrain factor average slope and the second-order terrain factor average profile curvature of each slope unit. The calculation formula of the first-order terrain factor average slope of a slope unit is A=(A11+A12+...+A1n) / n, wherein A11, A12,..., A1n represent the first-order terrain factor slope of each pixel within the range of the slope unit, and n represents the number of pixels in the slope unit; the calculation formula of the second-order terrain factor average profile curvature of a slope unit is B=(B11+B12+...+B1n) / n, wherein B11, B12,..., B1n represent the second-order terrain factor profile curvature of each pixel within the range of the slope unit, and n represents the number of pixels in the slope unit.

[0060] The near-infrared band reflectance value and the infrared band reflectance value of the multispectral optical satellite image data obtained in the previous exemplary embodiment are used to calculate the normalized difference vegetation index NDVI of each pixel, and the calculation formula is wherein NIR is the near-infrared band reflectance value, and RED is the infrared band reflectance value; the normalized difference vegetation index NDVI cumulative percentage of 5% is selected as the completely non-vegetation coverage normalized difference vegetation index value NDVI1, and the normalized difference vegetation index NDVI cumulative percentage of 95% is selected as the completely vegetation coverage normalized difference vegetation index value NDVI2; the vegetation coverage C1 of a single pixel is calculated based on the normalized difference vegetation index, the completely non-vegetation coverage normalized difference vegetation index value, and the completely vegetation coverage normalized difference vegetation index value of each pixel, and the calculation formula is Further, the average vegetation coverage of a slope unit is calculated based on the vegetation coverage of each pixel, and the calculation formula is C=(C11+C12+...+C1n) / n, wherein C11, C12,..., C1n represent the vegetation coverage of each pixel within the range of the slope unit, and n represents the number of pixels in the slope unit.

[0061] The effective migration space D of each slope unit is calculated based on the high-precision digital elevation model data obtained in the previous exemplary embodiment.

[0062] As an example, step 204 is specifically as follows:

[0063] For each slope unit, the following operations are performed:

[0064] The debris-flow source target area discrimination model is established based on a terrain factor, an average vegetation coverage, and an effective migration space of the slope unit.

[0065] The debris-flow source target area discrimination model is:

[0066] F(A,B,C,D) = H(A-15) H(45-A) H(B) H(0.6-C) H(200-D);

[0067] wherein H(·) is a unit step function: x is (A-15) or (45-A) or B or (0.6-C) or (200-D).

[0068] F(A,B,C,D) is a discrimination value, if the value of F(A,B,C,D) is 0, it is determined that the slope unit is not a debris-flow source target area, if the value of F(A,B,C,D) is 1, it is determined that the slope unit is a debris-flow source target area; wherein A is a first-order terrain factor average slope of a slope unit, unit is °, B is an average profile curvature of a slope unit, dimensionless, C is an average vegetation coverage of a slope unit, dimensionless, D is an effective migration space of a slope unit, which is a slope distance between the top of the slope unit and the nearest gully, unit is m.

[0069] In another exemplary embodiment of the present application, the above step 205 is replaced by steps 501-505.

[0070] Step 501: the debris-flow source target area is screened by using a first screening condition to obtain a typical debris-flow gully; the first screening condition is that the average longitudinal slope of the main gully is greater than 150 ‰, the front and rear edge height difference is greater than 1000 m, and there is an activity history in the past 20 years.

[0071] Step 502: determining the UAV aerial orthographic image data of the typical debris-flow gully according to the UAV aerial orthographic image data; determining the UAV real scene three-dimensional image data of the typical debris-flow gully according to the UAV real scene three-dimensional image data.

[0072] Step 503: determining the mountain shadow raster image data of the typical debris-flow gully based on the high-precision digital elevation model data.

[0073] Step 504: based on the UAV aerial orthographic image data of the typical debris-flow gully, such as Figure 3 part (a) in the above, the mountain shadow raster image data, such as Figure 3The UAV real scene three-dimensional image data of the middle part (b) and the typical debris flow gully, the direct interpretation signs and the indirect interpretation signs of the residual slope surface material source are established.

[0074] Step 505: Based on the direct interpretation signs and the indirect interpretation signs, the residual slope surface material source area is explained and circled, such as Figure 3 The residual slope surface material source area is explained and circled, and the area in the target area debris flow gully except the circled residual slope surface material source area is determined as the non-residual slope surface material source area.

[0075] In the embodiment, the direct interpretation signs of the residual slope surface material source include the color, the shape and the texture; the indirect interpretation signs of the residual slope surface material source include the land use type, the vegetation coverage and the slope gradient (such as Figure 3 The middle part (c), the slope shape). The specific illustration of the direct interpretation signs and the indirect interpretation signs of the residual slope surface material source is shown in Table 1.

[0076] Based on the direct interpretation signs and the indirect interpretation signs of the residual slope surface material source, the residual slope surface material source area 112 is explained and circled in the target area typical debris flow gully by using the ArcGIS software platform in the man-machine interactive mode. The UAV aerial orthographic image data and the vector range of the residual slope surface material source are superimposed, the mountain shadow raster image data and the vector range of the residual slope surface material source are superimposed, the residual slope surface material source in the vector range of the residual slope surface material source is determined as the residual slope surface material source, the residual slope surface material source catalog database is formed, the area in the target area debris flow gully except the circled residual slope surface material source area is determined as the non-residual slope surface material source area, the UAV aerial orthographic image after the residual slope surface material source area is circled is shown in Figure 4 (a), the mountain shadow raster image after the residual slope surface material source area is circled is shown in Figure 4 (b).

[0077] Table 1: Remote sensing interpretation signs of residual slope surface material source

[0078]

[0079]

[0080] In another exemplary embodiment of the present application, the step 206 is specifically as follows:

[0081] With a spatial resolution of 0.2m, the UAV aerial orthographic image data, mountain shadow raster image data and vector range of 112 residual slope surface sources and non-residual slope surface sources are sequentially cropped, segmented, and formed into a plurality of images to produce sample images and sample labels, and the cropped and segmented images are used as input data; the cropped and segmented images of the UAV aerial orthographic image are as shown in Figure 5 (a), and the cropped and segmented images of the mountain shadow raster image are as shown in Figure 5 (b). The "residual slope surface source" and "non-residual slope surface source" are used as label data as shown in Figure 6 The input data and label data are corresponded to obtain a plurality of sample data, and the sample data is subjected to data enhancement processing to construct a sample library.

[0082] Further, the data enhancement processing includes flipping, rotating, brightness adjustment, and contrast adjustment.

[0083] In a specific embodiment, the residual slope surface source sample and the non-residual slope surface source sample are subjected to horizontal flipping, vertical flipping, diagonal flipping, 90° clockwise rotation, 180° clockwise rotation, and 270° clockwise rotation to expand the number of samples.

[0084] In another exemplary embodiment of the present application, the above step 207 is replaced by steps 601-603.

[0085] Step 601: randomly dividing the training set and the test set according to the sample library.

[0086] Step 602: training different machine learning models.

[0087] Step 603: evaluating the recognition accuracy of different models; the recognition accuracy evaluation index is F1 score≥0.85, if satisfied, selecting the model with the highest recognition accuracy evaluation index as the optimal model, and using the optimal model to carry out automatic recognition of the residual slope surface source of the debris flow gully in the to-be-detected area.

[0088] Further, the different machine learning models include random forest, support vector machine, convolutional neural network, etc.

[0089] As an example, the F1 score is used to evaluate the recognition accuracy of different models, and the F1 score values of the random forest model, the support vector machine model, and the convolutional neural network model are 0.82, 0.83, and 0.89, respectively. The recognition accuracy of the convolutional neural network model meets the evaluation index F1 score≥0.85, and it is selected as the optimal model to carry out automatic recognition of the residual slope surface source of the debris flow gully in the to-be-detected area.

[0090] Based on the same inventive concept, this application also provides a method for identifying the source of debris flow residual slope deposits, such as... Figure 7 As shown, this method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 701 to 702. Wherein:

[0091] Step 701: Obtain comprehensive remote sensing data of debris flow gullies in the area to be detected.

[0092] Step 702: Input the comprehensive remote sensing data of the debris flow gully in the area to be detected into the regional debris flow residual slope source identification model to obtain the identification result of the residual slope source; wherein, the regional debris flow residual slope source identification model is determined based on the regional debris flow residual slope source identification model determination method described in any of the above embodiments.

[0093] This application also provides an application scenario in which the above-mentioned method for identifying the source of residual colluvial debris from debris flows is applied. Specifically, the method for identifying the source of residual colluvial debris from debris flows provided in this embodiment can be applied in a scenario for identifying the source of residual colluvial debris from debris flows. The scenario for identifying the source of residual colluvial debris from debris flows includes an identification stage and an identification result display stage. The identification stage is used to automatically identify the source of residual colluvial debris from debris flow gullies in the area to be detected, and the identification result display stage is used to display the identification results. The method for identifying the source of residual colluvial debris from debris flows provided in this embodiment belongs to the identification stage.

[0094] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 8As shown in the figure. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store processing data. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with the terminal outside through the network connection. The computer program is executed by the processor to realize a regional debris flow residual slope surface material source identification model determination method.

[0095] Those skilled in the art can understand that, Figure 8 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement. In one exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to realize the steps in each of the above method embodiments.

[0096] In one exemplary embodiment, a computer readable storage medium is provided, storing a computer program, which is executed by a processor to realize the steps in each of the above method embodiments.

[0097] In one exemplary embodiment, a computer program product is provided, including a computer program, which is executed by a processor to realize the steps in each of the above method embodiments.

[0098] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0099] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, databases or other media used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0100] The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a blockchain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0101] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described, but as long as the combinations of the technical features do not exist, they should be considered as the scope of the present application.

[0102] The principles and implementation modes of the present application are described by applying specific examples herein. The above description of the embodiments is only used to help understand the method and its core idea of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range can be changed. In conclusion, the content of the present application should not be understood as a limitation.

Claims

1. A method for determining a regional debris flow residual slope surface material source identification model, characterized in that, The regional debris flow residual slope surface material source identification model determination method comprises: Obtaining comprehensive remote sensing data of a target area debris flow gully, wherein the comprehensive remote sensing data comprises multispectral optical satellite image data, high-precision digital elevation model data, unmanned aerial vehicle aerial orthographic image data and unmanned aerial vehicle real scene three-dimensional image data; Dividing the target area debris flow gully based on the high-precision digital elevation model data to obtain a plurality of slope units; Based on the high-precision digital elevation model data, calculating the terrain factor and effective migration space of each slope unit, and based on the multispectral optical satellite image data, calculating the average vegetation coverage of each slope unit; Based on the residual slope surface material source target area discrimination model and the terrain factor, average vegetation coverage and effective migration space of each slope unit, determining the residual slope surface material source target area in the target area debris flow gully; Selecting a typical debris flow gully in the residual slope surface material source target area, and based on the typical debris flow gully, the unmanned aerial vehicle aerial orthographic image data, the high-precision digital elevation model data and the unmanned aerial vehicle real scene three-dimensional image data, screening out residual slope surface material sources and non-residual slope surface material sources from the typical debris flow gully; Based on the comprehensive remote sensing data of the typical debris flow gully and the screened out residual slope surface material source and non-residual slope surface material source vector range, constructing a sample library; Training a machine learning model according to the sample library to obtain a regional debris flow residual slope surface material source identification model; the regional debris flow residual slope surface material source identification model is used to identify the residual slope surface material source of a to-be-detected area debris flow gully.

2. The method according to claim 1, wherein, Based on the high-precision digital elevation model data, the target area debris flow gully is divided to obtain a plurality of slope units, specifically comprising: Using a terrain feature line extraction algorithm to process the high-precision digital elevation model data to obtain ridge lines, valley lines and profile curvature zero points, and based on the ridge lines, the valley lines and the profile curvature zero points, the target area debris flow gully is divided to obtain a plurality of slope units.

3. The method of claim 1, wherein the method further comprises: One slope unit contains a plurality of pixels; based on the high-precision digital elevation model data, the terrain factor and effective migration space of each slope unit are calculated, and based on the multispectral optical satellite image data, the average vegetation coverage of each slope unit is calculated, specifically comprising: The terrain factor comprises a first-order terrain factor average slope and a second-order terrain factor average profile curvature; Based on the high-precision digital elevation model data, the slope and profile curvature of each pixel in each slope unit are calculated; Using an arithmetic mean algorithm, the slope and profile curvature of all pixels in each slope unit are calculated respectively to obtain the first-order terrain factor average slope and the second-order terrain factor average profile curvature of each slope unit; Based on the high-precision digital elevation model data, the effective migration space of each slope unit is calculated; Based on the multispectral optical satellite image data, the vegetation coverage of each pixel in each slope unit is calculated; The vegetation coverage of all pixels in each of the slope units is calculated by using an arithmetic mean algorithm to obtain the average vegetation coverage of each of the slope units.

4. The method of claim 1, wherein the method further comprises: The debris flow gully target area is determined based on the debris deposit slope surface material source target area discrimination model and the terrain factor, the average vegetation coverage and the effective migration space of each of the slope units, and specifically includes the following steps. For each of the slope units, the following operations are performed: Based on the debris deposit slope surface material source target area discrimination model and the terrain factor, the average vegetation coverage and the effective migration space of a slope unit, it is determined whether the slope unit is a debris deposit slope surface material source target area. If yes, the slope unit is determined as a debris deposit slope surface material source target area. The debris deposit slope surface material source target area in the target area debris flow gully is obtained by collecting the slope units determined as the debris deposit slope surface material source target area. The debris deposit slope surface material source target area discrimination model is as follows: F(A, B, C, D) = H(A-15)·H(45-A)·H(B)·H(0.6-C)·H(200-D); where H( ) is the unit step function: x is (A-15), (45-A), B, (0.6-C), or (200-D); F(A, B, C, D) is a determination value, if the value of F(A, B, C, D) is 0, it is determined that the slope unit is not a debris deposit slope surface material source target area, if the value of F(A, B, C, D) is 1, it is determined that the slope unit is a debris deposit slope surface material source target area, A is a first-order terrain factor average slope of a slope unit, unit is °, B is a second-order terrain factor average profile curvature of a slope unit, dimensionless, C is an average vegetation coverage of a slope unit, dimensionless, and D is an effective migration space of a slope unit, unit is m.

5. The method of claim 1, wherein the method further comprises: Typical debris flow gullies are selected in the debris deposit slope surface material source target area, and specifically include the following steps. The first screening condition is that the average longitudinal slope of the main channel is greater than 150‰, the height difference between the front and rear edges is greater than 1000m, and there is an activity history in the past 20 years. Based on the typical debris flow gully, the unmanned aerial orthographic image data, the high-precision digital elevation model data and the unmanned aerial real scene three-dimensional image data, the debris deposit slope surface material source and the non-debris deposit slope surface material source are screened out from the typical debris flow gully, and specifically include the following steps.

6. The method of claim 1, wherein the method further comprises: The unmanned aerial orthographic image data of the typical debris flow gully is determined according to the unmanned aerial orthographic image data, and the unmanned aerial real scene three-dimensional image data of the typical debris flow gully is determined according to the unmanned aerial real scene three-dimensional image data; Based on the high-precision digital elevation model data, the mountain shadow raster image data of the typical debris flow gully is determined; Based on the unmanned aerial orthographic image data, the mountain shadow raster image data and the unmanned aerial real scene three-dimensional image data of the typical debris flow gully, the direct interpretation signs and the indirect interpretation signs of the debris deposit slope surface material source are established; Based on the direct interpretation signs and the indirect interpretation signs, the debris deposit slope surface material source area is explained and circled, a debris deposit slope surface material source catalog database is formed, and the area other than the circled debris deposit slope surface material source area in the typical debris flow gully is determined as a non-debris deposit slope surface material source area. ​ 7. A method for identifying a regional debris flow residual deposit slope surface material source, characterized in that, The regional debris flow residual slope deposit surface source identification method comprises: comprehensive remote sensing data of a debris flow gully in the to-be-detected region is acquired; the comprehensive remote sensing data of the debris flow gully in the to-be-detected region is input into a regional debris flow residual slope deposit surface source identification model to obtain residual slope deposit surface sources; wherein the regional debris flow residual slope deposit surface source identification model is determined based on the regional debris flow residual slope deposit surface source identification model determination method of any one of claims 1-6.

8. A computer device comprising: The memory, the processor, and the computer program stored on the memory and capable of running on the processor are characterized in that the processor executes the computer program to implement the regional debris flow residual slope deposit surface source identification model determination method of any one of claims 1-6 or the regional debris flow residual slope deposit surface source identification method of claim 7.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the regional debris flow residual slope deposit surface source identification model determination method of any one of claims 1-6 or the regional debris flow residual slope deposit surface source identification method of claim 7.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the regional debris flow residual slope deposit surface source identification model determination method of any one of claims 1-6 or the regional debris flow residual slope deposit surface source identification method of claim 7. The computer program is executed by the processor to implement the regional debris flow residual slope deposit surface source identification model determination method of any one of claims 1-6 or the regional debris flow residual slope deposit surface source identification method of claim 7.

Images