Automatic comparison-based post-raining railway geological disaster risk identification method

Abstract

The invention discloses a post-raining railway geological disaster risk identification method based on automatic comparison, and relates to the technical field of geological disaster identification. The method comprises the following steps: carrying out pre-rain image data acquisition and post-rain image data acquisition according to forecast rainfall data and real-time rainfall data; performing three-dimensional modeling construction on the image data when the pre-rain image data and the post-rain image data are transmitted back to obtain a pre-rain railway geology three-dimensional model and a post-rain railway geology three-dimensional model; a deformation area is extracted through a geological disaster risk identification model based on an image comparison algorithm, and the risk probability and type of a disaster are calculated according to the variable quantity of the deformation area. By adopting the post-raining railway geological disaster risk identification method based on automatic comparison, the unmanned aerial vehicle is triggered to patrol according to the forecast rainfall and the real-time rainfall to obtain the image data, the post-raining three-dimensional model and the pre-raining three-dimensional model are compared and analyzed through a comparison algorithm, and the geological disaster risk hidden danger is obtained through automatic identification. And the recognition efficiency of the geological hazard hidden danger is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of geological disaster identification, and in particular to a method for identifying railway geological disaster risks after rain based on automatic comparison. Background Art

[0002] Railway flood control management is a crucial measure to ensure safe railway operations during flood seasons and rainstorms. Existing technology involves deploying sensor networks at key locations to collect data. For example, patent publication number CN118446869A discloses a railway flood control management system. This system includes a data acquisition module that uses sensors to collect real-time data from key flood control points. It also employs drone monitoring to monitor and capture images along the railway line during extreme weather conditions. While this only provides monitoring functionality, it also requires manual image review and judgment. However, due to the long distances along the railway line, this process consumes considerable time and effort, making it difficult to obtain timely information on disasters. While only collecting and processing current images, the difference between pre- and post-rain images often indicates the occurrence of geological hazards and the presence of corresponding disasters. An automated post-rain railway geological hazard risk identification method is urgently needed to identify geological hazards along the railway line. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for identifying railway geological disaster risks after rain based on automatic comparison to solve the above technical problems.

[0004] To achieve the above objectives, the present invention provides a method for identifying railway geological disaster risks after rain based on automatic comparison, the specific steps of which are as follows:

[0005] Step S1: obtaining forecast rainfall data and real-time rainfall data in real time;

[0006] Step S2: collecting pre-rain image data and post-rain image data based on the forecast rainfall data and the real-time rainfall data;

[0007] Step S3: When the pre-rain image and the post-rain image data are transmitted back, the image data are subjected to three-dimensional modeling to obtain a pre-rain railway geological three-dimensional model and a post-rain railway geological three-dimensional model;

[0008] Step S4: Input the railway geological 3D model before and after the rain into the geological disaster risk identification model based on the image contrast algorithm to extract the deformation area, and calculate the risk probability and type of the disaster according to the change in the deformation area.

[0009] Preferably, in step S2, when the forecast rainfall is greater than the set rainfall, the drones in the drone library are triggered to collect pre-rain image data of the set railway line;

[0010] Real-time rainfall is obtained in real time. When the real-time rainfall is not greater than the set rainfall and the forecast rainfall within the set time interval is less than the set rainfall, the drones in the drone library are triggered to collect post-rainfall image data for the set railway line.

[0011] Preferably, the UAV is equipped with a laser radar and photogrammetry equipment to scan the railway line to obtain three-dimensional geological point cloud data around the railway line.

[0012] Preferably, in step S3, data is preprocessed, and coordinate alignment and resolution unification operations are performed on the three-dimensional point cloud data; geological features are extracted from the preprocessed three-dimensional point cloud data through a terrain feature extraction model, the extracted features are represented as raster data, and the feature values ​​are superimposed on each unit grid; three-dimensional modeling is performed using the extracted terrain data.

[0013] Preferably, in step S4,

[0014] Step S41: extracting a three-dimensional model of railway geology before and after the rain at the same location;

[0015] Step S42: performing a point-by-point and pixel-by-pixel comparison of the railway geological 3D model before and after the rain using a direct point cloud difference method and a feature change monitoring method to calculate a feature value change;

[0016] Step S43: Determine the deformation area according to the characteristic value change, and obtain the risk probability and type of disaster according to the characteristic value and change characteristics of the identified change area.

[0017] In step S43, the characteristic value variation is calculated, and the characteristic value variation includes the time variation and the space variation.

[0018] The time variation calculation formula is as follows:

[0019] ΔV t =V(t)-V(t-1)

[0020] Where, ΔV t Indicates the time variation of the eigenvalue, V(t) is the eigenvalue before the rain, and V(t-1) is the eigenvalue after the rain;

[0021] The calculation formula of spatial variation is as follows:

[0022] ΔV s =V(x)-V(x-1)

[0023] Among them, V(x) is the eigenvalue of a certain spatial position, V(x-1) is the eigenvalue of the adjacent position, ΔV s Represents the spatial variation of eigenvalues;

[0024] The conditions for determining the deformation area are as follows:

[0025] When |ΔV t |>T t or |ΔV S |>T s , it is determined to be a deformation area, where T t is the threshold of time variation, T s is the threshold of spatial variation.

[0026] Preferably, in step S43,

[0027] The formula for calculating risk probability is as follows:

[0028]

[0029] Among them, P is the risk probability of disaster occurrence, α and β are the weight coefficients of temporal variation and spatial variation, respectively.

[0030] Preferably, the disaster types include flooding, landslide, debris flow, landslide, roadbed settlement and bridge and culvert damage.

[0031] Therefore, the present invention adopts the above-mentioned method for identifying railway geological disaster risks after rain based on automatic comparison, which has the following beneficial effects: triggering the drone patrol conditions according to the forecast rainfall and real-time rainfall, flying according to the existing planned route under the influence of the meteorological environment, and taking three-dimensional model aerial photos of the surrounding environment along the line. After the aerial photos, the remote sensing image data is automatically transmitted to the data processing server, and the three-dimensional model after rain is compared and analyzed with the three-dimensional model before rain through the three-dimensional model comparison algorithm, and the geological disaster risk hazards are automatically identified, thereby improving the identification efficiency of geological disaster hazards and improving the technical prevention capabilities of railway disaster prevention and flood control.

[0032] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a flow chart of a method for identifying railway geological disaster risks after rain based on automatic comparison according to the present invention. DETAILED DESCRIPTION

[0034] Example 1

[0035] like Figure 1 As shown in FIG, a method for identifying railway geological disaster risks after rain based on automatic comparison is shown, and the specific steps are as follows:

[0036] Step S1: Acquire forecast rainfall data and real-time rainfall data in real time. Rainfall data is acquired by establishing a connection with a meteorological platform to realize meteorological data acquisition.

[0037] Step S2: Pre-rainfall and post-rainfall image data are collected based on the forecast and real-time rainfall data. When the forecast rainfall exceeds the set rainfall amount, drones in the drone library are triggered to collect pre-rainfall image data for the designated railway line. Real-time rainfall is obtained in real time. When the real-time rainfall is not greater than the set rainfall amount and the forecast rainfall within a set time interval is less than the set rainfall amount, drones in the drone library are triggered to collect post-rainfall image data for the designated railway line. Drones equipped with lidar and photogrammetry equipment scan the railway line to obtain 3D point cloud data of the surrounding geology.

[0038] Step S3: When the pre-rain and post-rain image data are transmitted back, the image data is subjected to 3D modeling to construct a 3D railway geological model before and after the rain. Data preprocessing involves aligning the coordinates of the 3D point cloud data and unifying its resolution. A terrain feature extraction model is used to extract geological features from the preprocessed 3D point cloud data. The extracted features are represented as raster data, and the feature values ​​are superimposed on each unit grid. 3D modeling is then performed using the extracted terrain data.

[0039] Step S4: Input the railway geological 3D model before and after the rain into the geological disaster risk identification model based on the image contrast algorithm to extract the deformation area, and calculate the risk probability and type of the disaster according to the change in the deformation area.

[0040] Step S41: extracting a three-dimensional model of railway geology before and after the rain at the same location;

[0041] Step S42: performing a point-by-point and pixel-by-pixel comparison of the railway geological 3D model before and after the rain using a direct point cloud difference method and a feature change monitoring method to calculate a feature value change;

[0042] Step S43: Determine the deformation area based on the change in the characteristic value, and determine the risk probability and type of disaster based on the characteristic value and change characteristics of the identified change area. Disaster types include flooding, landslide, debris flow, landslide, roadbed settlement, and bridge and culvert damage.

[0043] In step S43, the characteristic value variation is calculated, and the characteristic value variation includes the time variation and the space variation.

[0044] The time variation calculation formula is as follows:

[0045] ΔV t =V(t)-V(t-1)

[0046] Where, ΔV t Indicates the time variation of the eigenvalue, V(t) is the eigenvalue before the rain, and V(t-1) is the eigenvalue after the rain;

[0047] The calculation formula of spatial variation is as follows:

[0048] ΔV s =V(x)-V(x-1)

[0049] Among them, V(x) is the eigenvalue of a certain spatial position, V(x-1) is the eigenvalue of the adjacent position, ΔV s Represents the spatial variation of eigenvalues;

[0050] The conditions for determining the deformation area are as follows:

[0051] When |ΔV t |>T t or |ΔV S |>T s , it is determined to be a deformation area, where T t is the threshold of time variation, T s is the threshold of spatial variation.

[0052] Preferably, in step S43,

[0053] The risk probability calculation formula is as follows:

[0054]

[0055] Among them, P is the risk probability of disaster occurrence, α and β are the weight coefficients of temporal variation and spatial variation, respectively.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solutions of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for identifying railway geological disaster risks after rain based on automatic comparison, characterized in that: The specific steps are as follows: Step S1: obtaining forecast rainfall data and real-time rainfall data in real time; Step S2: collecting pre-rain image data and post-rain image data based on the forecast rainfall data and the real-time rainfall data; Step S3: When the pre-rain image and the post-rain image data are transmitted back, the image data are subjected to three-dimensional modeling to obtain a pre-rain railway geological three-dimensional model and a post-rain railway geological three-dimensional model; Step S4: Input the railway geological 3D model before and after the rain into the geological disaster risk identification model based on the image contrast algorithm to extract the deformation area, and calculate the risk probability and type of the disaster according to the change in the deformation area.

2. The method for identifying railway geological disaster risks after rain based on automatic comparison according to claim 1, characterized in that: In step S2, when the predicted rainfall is greater than the set rainfall, the drones in the drone library are triggered to collect pre-rain image data of the set railway line; Real-time rainfall is obtained in real time. When the real-time rainfall is not greater than the set rainfall and the forecast rainfall within the set time interval is less than the set rainfall, the drones in the drone library are triggered to collect post-rainfall image data for the set railway line.

3. The method for identifying railway geological disaster risks after rain based on automatic comparison according to claim 2, characterized in that: The drone is equipped with lidar and photogrammetry equipment to scan the railway line to obtain three-dimensional geological point cloud data around the railway line.

4. The method for identifying railway geological disaster risks after rain based on automatic comparison according to claim 3 is characterized by: In step S3, data is preprocessed to perform coordinate alignment and resolution unification operations on the 3D point cloud data; geological features are extracted from the preprocessed 3D point cloud data using a terrain feature extraction model, the extracted features are represented as raster data, and the feature values ​​are superimposed on each unit grid; 3D modeling is performed using extracted terrain data.

5. The method for identifying railway geological disaster risks after rain based on automatic comparison according to claim 4 is characterized in that: In step S4, Step S41: extracting a three-dimensional model of railway geology before and after the rain at the same location; Step S42: performing a point-by-point and pixel-by-pixel comparison of the railway geological 3D model before and after the rain using a direct point cloud difference method and a feature change monitoring method to calculate a feature value change; Step S43: Determine the deformation area according to the characteristic value change, and obtain the risk probability and disaster type of the disaster according to the characteristic value and change characteristics of the identified change area.

6. The method for identifying railway geological disaster risks after rain based on automatic comparison according to claim 4 is characterized in that: In step S43, Calculate the eigenvalue variation, which includes time variation and space variation. The time variation calculation formula is as follows: ΔV t =V(t)-V(t-1) Where, ΔV t Indicates the time variation of the eigenvalue, V(t) is the eigenvalue before the rain, and V(t-1) is the eigenvalue after the rain; The calculation formula of spatial variation is as follows: ΔV s =V(x)-V(x-1) Among them, V(x) is the eigenvalue of a certain spatial position, V(x-1) is the eigenvalue of the adjacent position, ΔV s Represents the spatial variation of eigenvalues; The conditions for determining the deformation area are as follows: When |ΔV t |>T t or |ΔV S |>T s , it is determined to be a deformation area, where T t is the threshold of time variation, T s is the threshold of spatial variation.

7. The method for identifying railway geological disaster risks after rain based on automatic comparison according to claim 6, characterized in that: In step S43, The risk probability calculation formula is as follows: Among them, P is the risk probability of disaster occurrence, α and β are the weight coefficients of temporal variation and spatial variation, respectively.

8. The method for identifying railway geological disaster risks after rain based on automatic comparison according to claim 7, characterized in that: Disaster types include flooding, landslides, debris flows, mudslides, roadbed subsidence, and bridge and culvert damage.

Images