Disease early warning method for permafrost roadbed

Abstract

The invention provides a permafrost roadbed disease early warning method, and relates to the technical field of permafrost roadbed disease evaluation. The method comprises the following steps: performing deformation laser scanning detection on the surface of the permafrost roadbed to obtain road surface deformation point cloud elevation data of the permafrost roadbed; and preprocessing the road surface deformation point cloud elevation data to obtain a surface elevation two-dimensional grid of the permafrost roadbed. And calculating a deformation gradient value of each grid unit in the elevation two-dimensional grid by adopting an eight-neighborhood gradient algorithm to obtain a surface deformation gradient two-dimensional grid of the permafrost roadbed. And classifying the surface deformation gradient two-dimensional grids based on a preset deformation gradient threshold value, identifying a standard-exceeding area of the permafrost roadbed, and determining a roadbed disease index based on the area of the standard-exceeding area. And determining the disease early warning level of the permafrost roadbed according to the roadbed disease index. According to the method, the development process of frozen soil roadbed diseases can be effectively predicted, and the timeliness and reliability of freeze-thaw disaster early warning are remarkably improved.

Description

Technical Field

[0001] This application relates to the technical field of frozen soil subgrade disease evaluation, and specifically relates to a warning method for diseases of permafrost subgrade. Background Art

[0002] Permafrost refers to rock and soil layers that remain in a frozen state for more than two years, mainly distributed in high-latitude and high-altitude regions. Such as the Qinghai-Tibet Plateau in China, the Greater and Lesser Hinggan Mountains in the Northeast, and the circum-Arctic region. This type of frozen soil has significant thermal sensitivity and phase change characteristics, and its mechanical properties change significantly with temperature fluctuations, manifested as the alternating action of frost heaving and thaw settlement. In the natural state, the frozen soil maintains a dynamic thermal balance with the surrounding environment, but once it is subjected to external thermal disturbances, its stability is extremely easy to be damaged, leading to irreversible engineering geological problems.

[0003] With the continuous warming of the global climate and the intensification of human engineering activities, the thermal stability of permafrost areas is further out of balance, and the degradation trend is significant. Diseases such as subgrade thermal melting settlement, longitudinal cracks, and slope collapse frequently occur due to phenomena such as the downward movement of the permafrost table and the thickening of the active layer. Taking the Qinghai-Tibet Highway as an example, the average annual temperature rise rate reaches 0.3°C to 0.4°C, and the permafrost table has decreased by an average of 0.5 m to 1.5 m in the past 30 years, directly inducing differential thaw settlement deformation of the subgrade. Such diseases not only reduce the pavement flatness, intensify vehicle bumpiness, but also cause the instability of the subgrade structure, threatening traffic safety, and even forcing the road speed limit or closure, seriously restricting the operation efficiency of the transportation network in alpine regions. Therefore, establishing an accurate warning system for frozen soil subgrade diseases can identify risk areas in advance, guide maintenance decisions, thereby reducing maintenance costs and extending the road life, which is of great significance for ensuring the safety of the transportation lifeline in cold regions.

[0004] Currently, most of the warning technologies for frozen soil subgrade are based on finite element numerical models, relying on complex thermal-mechanical coupling parameters and boundary conditions. However, factors such as the heterogeneity of frozen soil media and the uncertainty of boundary conditions limit the applicability of the models. In addition, traditional monitoring methods such as manual measurement and sensor layout have defects such as low efficiency, narrow coverage, and poor real-time performance, making it difficult to meet the dynamic assessment requirements for the health status of large-scale subgrades. Existing warning methods have significant deficiencies in parameter acquisition, calculation efficiency, and engineering practicability. There is an urgent need for an efficient, accurate, and popularizable disease warning technology to address the severe challenges faced by road engineering in permafrost regions. Summary of the Invention

[0005] The present application provides a method for early warning of permafrost subgrade diseases, including: performing deformation laser scanning detection on the surface of the permafrost subgrade to obtain the elevation data of the road surface deformation point cloud of the permafrost subgrade; preprocessing the elevation data of the road surface deformation point cloud to obtain the two-dimensional grid of the surface elevation of the permafrost subgrade; calculating the deformation gradient value of each grid unit in the two-dimensional elevation grid by using the eight-neighborhood gradient algorithm to obtain the two-dimensional grid of the surface deformation gradient of the permafrost subgrade; classifying the two-dimensional grid of the surface deformation gradient based on a preset deformation gradient threshold to identify the exceeding-standard area of the permafrost subgrade, and determining the subgrade disease index based on the area of the exceeding-standard area; and determining the disease early warning level of the permafrost subgrade according to the subgrade disease index.

[0006] According to an embodiment of the present application, the performing deformation laser scanning detection on the surface of the permafrost subgrade to obtain the elevation data of the road surface deformation point cloud of the permafrost subgrade includes: setting the flight parameters of the airborne lidar system, where the flight parameters are used to generate the flight route for flight survey and scanning; based on the flight parameters, performing spatial registration and calibration on the equipment for performing deformation laser scanning detection; arranging elevation detection points at the edge of the survey area, the beginning and end of the flight route, the flat area inside the survey area, and the overlapping area between flight strips in the flight survey area according to the flight route; performing point cloud accuracy verification based on the elevation detection points. If the accuracy verification meets the requirements, it indicates that the point cloud data collection is completed. If the accuracy verification does not meet the requirements, re-operation is performed.

[0007] According to an embodiment of the present application, the flight parameters include: the flight height is limited to 50 meters, the side overlap rate of the flight route is 50%, and the flight speed is 5 meters per second.

[0008] According to an embodiment of the present application, the preprocessing the elevation data of the road surface deformation point cloud to obtain the two-dimensional grid of the surface elevation of the permafrost subgrade includes: filtering out the high and low outliers in the point cloud elevation data based on point cloud filtering; performing point cloud segmentation on the filtered point cloud elevation data, and dividing the point cloud elevation data into multiple non-overlapping regions according to the three-dimensional coordinates, echo intensity, waveform, texture, and shape of the point cloud, and extracting the road surface point cloud data; performing point cloud thinning on the road surface point cloud data to make the road point cloud evenly distributed in the two-dimensional plane; defining the grid cell size, and matching each grid with the plane coordinates in the road surface point cloud data to generate the two-dimensional grid of the surface elevation of the permafrost subgrade.

[0009] According to an embodiment of the present application, during the process of point cloud thinning, the road surface point cloud data is resampled at equal intervals with a sampling interval of 50 centimeters; the grid cell size is 50 centimeters.

[0010] According to an embodiment of the present application, calculating the deformation gradient value of each grid unit in the elevation two-dimensional grid by using the eight-neighborhood gradient algorithm to obtain the surface deformation gradient two-dimensional grid of the permafrost subgrade includes: performing boundary expansion processing on the elevation two-dimensional grid, adding filled zero elements around the elevation two-dimensional grid to obtain a complete eight-neighborhood calculation window; in the eight-neighborhood calculation window, performing effective neighborhood determination on each central pixel. If there are at least 7 effective elevation values in the eight-neighborhood of the central pixel, calculate the deformation gradient value of the central pixel; if there are not at least 7 effective elevation values in the eight-neighborhood of the central pixel, determine that the deformation gradient value of the central pixel is a null value; traverse all the grid units to obtain the surface deformation gradient two-dimensional grid of the permafrost subgrade.

[0011] According to an embodiment of the present application, the formula for calculating the deformation gradient value of the central pixel includes:

[0012]

[0013]

[0014] Wherein, is the incremental change rate of the central pixel e in the x direction; is the incremental change rate of the central pixel e in the y direction; a, b, c, d, f, g, h, i are the eight pixels adjacent to the central pixel e in the eight-neighborhood calculation window; and are the weighted counts in the x direction of valid pixels, , are the weighted counts in the y direction of valid pixels; is the standardized pixel size; is the deformation gradient value of the central pixel.

[0015] According to an embodiment of the present application, classifying the surface deformation gradient two-dimensional grid based on a preset deformation gradient threshold, identifying the exceeding-standard area of the permafrost subgrade, and determining the subgrade disease index based on the area of the exceeding-standard area includes: performing binary classification on the surface deformation gradient two-dimensional grid based on the deformation gradient threshold, marking the pixels exceeding the deformation gradient threshold as the exceeding-standard area, and generating a binary image including a foreground area and a background area; calculating the total area of the section of the permafrost subgrade according to the total number of pixels in the binary image and the fixed pixel size; calculating the area of the differential exceeding-standard area in the binary image; determining the subgrade disease index according to the total area of the section and the area of the differential exceeding-standard area.

[0016] According to an embodiment of the present application, a differential deformation gradient slope change rate of 0.51% is used as the deformation gradient threshold.

[0017] According to an embodiment of the present application, determining the disease warning level of the permafrost subgrade based on the subgrade disease index includes: determining the disease warning level corresponding to the permafrost subgrade according to the numerical range where the disease index is located; wherein, if the subgrade disease index is greater than 0.7 and less than or equal to 1, it is determined that the disease warning level of the permafrost subgrade is a severe disaster warning level; if the subgrade disease index is greater than 0.1 and less than or equal to 0.7, it is determined that the disease warning level of the permafrost subgrade is a moderate disaster warning level; if the subgrade disease index is greater than or equal to 0 and less than or equal to 0.1, it is determined that the disease warning level of the permafrost subgrade is a mild disaster warning level.

[0018] Compared with the prior art, the beneficial effects of the present application are: By integrating laser scanning technology and automated gradient analysis, a closed-loop monitoring system for permafrost subgrade diseases is constructed. Using three-dimensional point cloud data to generate elevation two-dimensional grids, combined with the dynamic weighted eight-neighborhood gradient algorithm to accurately identify local deformation characteristics, and quantifying the disease index based on the spatial distribution of the exceeded standard area, realizing the full-process intelligent processing from data acquisition, deformation analysis to warning classification. This method effectively solves the problems of low efficiency of traditional manual monitoring and strong model dependence, scientifically divides the disease development stage through a standardized index system, provides high-precision and quantifiable decision support for road maintenance in permafrost areas, and significantly improves the timeliness and reliability of freeze-thaw disaster warnings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the steps of the permafrost subgrade disease warning method provided by the embodiment of the present application.

[0020] Figure 2 It is a schematic diagram of the result of downsampling the point cloud data of the road surface provided by the embodiment of the present application.

[0021] Figure 3 It is a schematic diagram of the eight-neighborhood of pixels provided by the embodiment of the present application.

[0022] Figure 4 It is a schematic diagram of the deformation gradient grid of the disease section of the permafrost section A provided by the embodiment of the present application.

[0023] Figure 5 It is a schematic diagram of the binary image of the difference exceeded standard area generated by threshold classification provided by the embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] The present application will be further described in detail below in combination with test examples and specific embodiments. However, it should not be understood that the scope of the above-mentioned subject matter of the present application is limited to the following embodiments. All technologies implemented based on the content of the present application fall within the scope of protection of the present application.

[0025] Unless otherwise specified, in the description of the specific embodiments of the present application, the expression terms indicating the orientation or positional relationship such as "upper", "lower", "left", "right", "center", "inner", "outer", "side", etc. are all based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship when the product / device / device is normally used and placed. These terms of orientation or positional relationship are only for the convenience of describing the solution of the present application or simplifying the description in the specific embodiment, so as to facilitate technicians to quickly understand the solution, rather than indicating or implying that a specific device / component / element must have a specific orientation, or be constructed and operated in a specific positional relationship. Therefore, it should not be construed as a limitation to the present application.

[0026] In the description of the embodiments of the present application, the technical terms "first", "second", etc. only distinguish one entity or operation from another entity or operation, and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "a plurality" is two or more, unless otherwise specifically defined.

[0027] Referring to the embodiments herein means that the specific features, structures or characteristics described in connection with the embodiments may be included in at least one embodiment of the present application. The phrase appears in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein may be combined with other embodiments.

[0028] Please refer to Figure 1 , Figure 1 which is a schematic diagram of the steps of the warning method for permafrost subgrade diseases provided by the embodiments of the present application. The warning method for permafrost subgrade diseases may include the following steps.

[0029] S1. Perform deformation laser scanning detection on the surface of the permafrost subgrade to obtain the elevation data of the surface deformation point cloud of the permafrost subgrade.

[0030] S2. Preprocess the elevation data of the surface deformation point cloud to obtain a two-dimensional grid of the surface elevation of the permafrost subgrade.

[0031] S3. Use the eight-neighborhood gradient algorithm to calculate the deformation gradient value of each grid unit in the two-dimensional elevation grid to obtain a two-dimensional grid of the surface deformation gradient of the permafrost subgrade.

[0032] S4. Classify the two-dimensional grid of surface deformation gradients based on a preset deformation gradient threshold, identify the exceeded-standard areas of the permafrost subgrade, and determine the subgrade disease index based on the area of the exceeded-standard areas.

[0033] S5. Determine the disease warning level of the permafrost subgrade according to the subgrade disease index.

[0034] In the following content of this application, the method provided in this application is described by taking the disease warning of permafrost section A as an example. This section is located in the permafrost area, with an altitude of 4300 m, an average temperature of -3.0°C to -4.1°C, an average annual ground temperature of permafrost of -0.3°C to -0.5°C, a volume ice content of permafrost of about 15% to 40%, and the ice content type of permafrost being rich-ice and saturated frozen soil, belonging to high-temperature and high-ice-content permafrost. The natural permafrost upper limit is about 1.8 m, and the risk of permafrost thawing is extremely high.

[0035] Optionally, step S1 may include: Set the flight parameters of the airborne lidar system, where the flight parameters are used to generate the flight route for flight surveying and scanning. Among them, the airborne lidar system can be used to fly and survey above the permafrost disease section. The flight route is set according to the regional area. The flight altitude can be set to 50 m, the side overlap rate of the route is set to 50%, and the flight speed is 5 m / s.

[0036] Based on the flight parameters, perform spatial registration and calibration on the equipment for performing deformation laser scanning detection. Among them, the equipment for performing deformation laser scanning detection may include each observation unit, such as GPS, IMU, laser scanner, etc.

[0037] Layout elevation detection points at the survey area edge, the beginning and end of the flight route, the flat area inside the survey area, and the overlapping area between flight strips in the flight surveying and scanning area according to the flight route. Perform point cloud accuracy verification based on the elevation detection points. If the accuracy verification meets the requirements, it indicates that the point cloud data acquisition is completed. If the accuracy verification does not meet the requirements, rework is required.

[0038] Optionally, step S2 may include: Filter out the high and low outliers in the point cloud elevation data based on point cloud filtering. Among them, point cloud filtering filters out the high and low outliers in the point cloud. The filtering method searches for the specified number n of neighboring points for each point and calculates the average distance from this point to its neighboring points. Among these n points, if the average distance of a certain point differs by more than 5 times the standard deviation, it is regarded as a noise point and excluded.

[0039] Perform point cloud segmentation on the filtered point cloud elevation data, and divide the point cloud elevation data into multiple non-overlapping regions according to the three-dimensional coordinates, echo intensity, waveform, texture, and shape of the point cloud, and extract the point cloud data of the road surface.

[0040] Perform point cloud thinning on the point cloud data of the road surface to make the road point clouds evenly distributed in the two-dimensional plane. Among them, resample the point cloud data of the road surface at an equal interval of 50 cm, so that the road point clouds are evenly distributed in the two-dimensional plane. Please refer to Figure 2 , Figure 2 which is a schematic diagram of the result of point cloud thinning for the point cloud data of the road surface provided by the embodiment of the present application.

[0041] Define the grid cell size, match each grid with the planar coordinates in the point cloud data of the road surface, and generate a two-dimensional grid of the surface elevation of the permafrost subgrade. Among them, the grid cell size can be 50 cm, each grid is matched with the planar coordinates of the point elements in the point cloud data, and each cell has and only has one point to generate a two-dimensional grid of the surface elevation.

[0042] Optionally, in step S3, when performing the calculation of the road surface deformation gradient, a sliding window calculation method of 3×3 pixels (8-neighborhood) can be used to perform the calculation. The gradient is calculated and measured according to the maximum rate of change of the value in the direction from one pixel to its eight neighboring pixels, that is, the ratio of the maximum change in the elevation of the two-dimensional grid of the subgrade surface elevation as the distance between the pixel and its eight neighboring pixels changes, to identify the maximum slope drop or maximum slope rise of the subgrade surface elevation, and a two-dimensional grid of the surface deformation gradient is calculated from the two-dimensional grid of the surface elevation.

[0043] Step S3 may include: Perform boundary extension processing on the two-dimensional elevation grid, add zero-padding elements around the two-dimensional elevation grid to obtain a complete eight-neighborhood calculation window. Among them, add zero-padding elements at the outermost periphery of the two-dimensional grid of the subgrade surface elevation to ensure that the boundary pixel information is fully utilized.

[0044] In the eight-neighborhood calculation window, effective neighborhood determination is performed for each central pixel. If there are at least 7 effective elevation values in the eight-neighborhood of the central pixel, the deformation gradient value of the central pixel is calculated; if there are not at least 7 effective elevation values in the eight-neighborhood of the central pixel, the deformation gradient value of the central pixel is determined to be a null value. Among them, when calculating the deformation gradient, the pixel deformation gradient depends on the incremental change rates in the horizontal and vertical directions determined by the elevation values of the surface starting from the central pixel and its adjacent eight pixels; these adjacent pixels are identified by letters a to i, where e represents the pixel for which the gradient is currently being calculated; the calculation requires that at least 7 of the pixels adjacent to the processed pixel in the 8-neighborhood have valid values. If there are fewer than 7 adjacent valid pixels, the calculation is not performed, and the processed pixel output is a null value.

[0045] Please refer to Figure 3 , Figure 3 which is a schematic diagram of the eight-neighborhood of pixels provided by the embodiments of this application. Specifically, the formula for calculating the deformation gradient value of the central pixel includes:

[0046]

[0047]

[0048] Among them, is the incremental change rate of the central pixel e in the x direction; is the incremental change rate of the central pixel e in the y direction; a, b, c, d, f, g, h, i are the eight pixels adjacent to the central pixel e in the eight-neighborhood calculation window; and are the weighted counts in the x direction of valid pixels, , are the weighted counts in the y direction of valid pixels; is the standardized pixel size; is the deformation gradient value of the central pixel. For example, if c, f, and i all have one valid value, then =(1 + 2 * 1 + 1)=4; if i is a null value, then =(1 + 2 * 1 + 0)=3; if f is a null value, then =(1 + 2 * 0 + 1)=2. Based on the above deformation gradient calculation, the deformation gradient map of this section of the road is obtained. Please refer to Figure 4 , Figure 4It is a schematic diagram of the deformation gradient grid of the disease section of the permafrost road section A provided by the embodiment of the present application. In the schematic diagram of the deformation gradient grid of the disease section, different colors represent the point cloud deformation gradient increments in different numerical ranges. Among them, different shades of green areas represent the lowest interval of deformation gradient increment, corresponding to slight deformation or no significant differential settlement. The yellow area indicates that the deformation gradient enters the low-risk interval, representing local deformation in the initial stage. Different shades of orange areas reflect medium-strength deformation gradients, corresponding to the transition stage of the decline in the stability of the subgrade structure. The red area marks the highest interval of deformation gradient, representing severe differential settlement or high-risk areas of potential diseases.

[0049] Traverse all the grid cells to obtain the two-dimensional grid of the surface deformation gradient of the permafrost subgrade; among them, traverse the two-dimensional elevation grid using the calculation method in the above steps to obtain the two-dimensional grid of the surface deformation gradient.

[0050] Optionally, step S4 may include: Perform binary classification on the two-dimensional grid of the surface deformation gradient based on the deformation gradient threshold, mark the pixels exceeding the deformation gradient threshold as the exceeded standard area, and generate a binary image including the foreground area and the background area; among them, the differential deformation gradient slope rate of 0.51% can be used as the deformation gradient threshold, and use this threshold to classify the pixels of the two-dimensional grid of the surface deformation gradient. The exceeded standard area is used as the foreground area, and the non-exceeded standard area is used as the background area, and different colors or values are used to mark the pixels of different categories respectively, so as to generate a binary image of the differential deformation exceeded standard area. Please refer to Figure 5 , Figure 5 It is a schematic diagram of the binary image of the differential exceeded standard area generated by threshold classification provided by the embodiment of the present application. Among them, the green part is the non-exceeded standard area where the point cloud deformation gradient increment is less than 0.51%, and the red part is the exceeded standard area where the point cloud deformation gradient increment is greater than 0.51%.

[0051] Calculate the total area of the road section of the permafrost subgrade according to the total number of pixels of the binary image and the fixed pixel size; among them, the formula for calculating the total area of the road section of the permafrost subgrade is:

[0052] Among them, n is the number of pixels counted in the road section area, and 0.5 is the size of the grid pixel, with the unit of meter.

[0053] The total area of the road section calculated in the embodiment of the present application is:

[0054] Calculate the area of the differential exceeded standard area in the binary image; among them, since the differential settlement areas in the permafrost road section are distributed in patches, it is necessary to calculate the areas of each differential settlement exceeded standard area respectively:

[0055] Wherein: is the total area of the area with excessive differential deformation, to are the areas of m different excessive differential settlement ranges within the road section area, in square meters. Among them, the area of the area with excessive deformation gradient in this disease road section is calculated to be 12123.75 square meters.

[0056] Determine the subgrade disease index according to the total area of the road section and the area of the excessive difference area. Among them, the formula for calculating the subgrade disease index is:

[0057] In the embodiment of the present application, the subgrade disease index calculated is:

[0058] Optionally, step S5 may include: Determine the disease warning level corresponding to the permafrost subgrade according to the numerical range where the disease index is located; Wherein, if the subgrade disease index is greater than 0.7 and less than or equal to 1, determine that the disease warning level of the permafrost subgrade is a severe disaster warning level; If the subgrade disease index is greater than 0.1 and less than or equal to 0.7, determine that the disease warning level of the permafrost subgrade is a moderate disaster warning level; If the subgrade disease index is greater than or equal to 0 and less than or equal to 0.1, determine that the disease warning level of the permafrost subgrade is a mild disaster warning level.

[0059] Among them, according to the warning classification standard of the permafrost subgrade disease index, the warning classification may include red level I warning (severe 0.7 < EDI ≤ 1), yellow level II warning (moderate 0.1 < EDI ≤ 0.7), and blue level III warning (mild 0 ≤ EDI ≤ 0.1). In the above embodiment, the permafrost subgrade disease index EDI of the road section reaches 0.953, belonging to the red level I warning of the disease.

[0060] In summary, the method provided by the embodiments of the present application can construct a closed-loop monitoring system for frozen soil subgrade diseases by integrating laser scanning technology and automated gradient analysis. The elevation two-dimensional grid is generated using three-dimensional point cloud data, and the local deformation features are accurately identified by combining the dynamic weighted eight-neighborhood gradient algorithm. The disease index is quantified based on the spatial distribution of the exceeded standard areas, realizing the full-process intelligent processing from data acquisition, deformation analysis to early warning classification. This method effectively solves the problems of low efficiency and strong model dependence in traditional manual monitoring. By means of a standardized index system, the development stages of diseases are scientifically divided, providing high-precision and quantifiable decision-making support for road maintenance in frozen soil areas, and significantly improving the timeliness and reliability of freeze-thaw disaster early warning.

[0061] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A warning method for diseases of permafrost subgrade, characterized in that, Including: Performing deformation laser scanning detection on the surface of the permafrost subgrade to obtain the elevation data of the surface deformation point cloud of the permafrost subgrade; Preprocessing the elevation data of the surface deformation point cloud to obtain the two-dimensional grid of the surface elevation of the permafrost subgrade; Calculating the deformation gradient value of each grid unit in the two-dimensional elevation grid by using the eight-neighborhood gradient algorithm to obtain the two-dimensional grid of the surface deformation gradient of the permafrost subgrade; Classifying the two-dimensional grid of the surface deformation gradient based on a preset deformation gradient threshold, identifying the exceeded-standard area of the permafrost subgrade, and determining the subgrade disease index based on the area of the exceeded-standard area; Determining the disease warning level of the permafrost subgrade according to the subgrade disease index.

2. The method according to claim 1, wherein The performing deformation laser scanning detection on the surface of the permafrost subgrade to obtain the elevation data of the surface deformation point cloud of the permafrost subgrade includes: Setting the flight parameters of the airborne lidar system, where the flight parameters are used to generate the flight route for flight survey and scan; Performing spatial registration and calibration on the equipment for performing deformation laser scanning detection based on the flight parameters; Arranging elevation detection points at the edge of the survey area, the beginning and end of the flight route, the flat area inside the survey area, and the overlapping area between flight strips in the flight survey area according to the flight route; Performing point cloud accuracy verification based on the elevation detection points. If the accuracy verification meets the requirements, it indicates that the point cloud data collection is completed. If the accuracy verification does not meet the requirements, re-operation is required.

3. The method according to claim 2, wherein The flight parameters include: the flight height is limited to 50 meters, the side overlap rate of the flight route is 50%, and the flight speed is 5 meters per second.

4. The method according to claim 1, characterized in that, The preprocessing the elevation data of the surface deformation point cloud to obtain the two-dimensional grid of the surface elevation of the permafrost subgrade includes: Filtering out the high and low outliers in the elevation data of the point cloud based on point cloud filtering; Performing point cloud segmentation on the filtered elevation data of the point cloud, dividing the elevation data of the point cloud into multiple non-overlapping regions according to the three-dimensional coordinates, echo intensity, waveform, texture, and shape of the point cloud, and extracting the point cloud data of the road surface; Performing point cloud thinning on the point cloud data of the road surface to make the road point cloud evenly distributed in the two-dimensional plane; Defining the grid cell size, matching each grid with the plane coordinates in the point cloud data of the road surface, and generating the two-dimensional grid of the surface elevation of the permafrost subgrade.

5. The method according to claim 4, characterized in that, During the process of point cloud thinning, resampling the point cloud data of the road surface at an equal interval with a sampling interval of 50 cm; the grid cell size is 50 cm.

6. The method according to claim 1, wherein The calculating the deformation gradient value of each grid unit in the two-dimensional elevation grid by using the eight-neighborhood gradient algorithm to obtain the two-dimensional grid of the surface deformation gradient of the permafrost subgrade includes: Performing boundary extension processing on the two-dimensional elevation grid, adding filled zero elements outside the two-dimensional elevation grid to obtain a complete eight-neighborhood calculation window; In the eight-neighborhood calculation window, perform an effective neighborhood determination for each central pixel. If there are at least 7 effective elevation values in the eight-neighborhood of the central pixel, calculate the deformation gradient value of the central pixel; if there are not at least 7 effective elevation values in the eight-neighborhood of the central pixel, determine that the deformation gradient value of the central pixel is a null value. Traverse all the grid cells to obtain a two-dimensional grid of the surface deformation gradient of the permafrost subgrade.

7. The method according to claim 6, wherein The formula for calculating the deformation gradient value of the central pixel includes: Among them, is the incremental change rate of the central pixel e in the x direction; is the incremental change rate of the central pixel e in the y direction; a, b, c, d, f, g, h, i are the eight pixels adjacent to the central pixel e in the eight-neighborhood calculation window; and are the weighted counts of valid pixels in the x direction, , are the weighted counts of valid pixels in the y direction; is the standardized pixel size; is the deformation gradient value of the central pixel.

8. The method according to claim 1, wherein Classify the two-dimensional grid of the surface deformation gradient based on a preset deformation gradient threshold, identify the over-standard area of the permafrost subgrade, and determine the subgrade disease index based on the area of the over-standard area, including: Perform binary classification on the two-dimensional grid of the surface deformation gradient based on the deformation gradient threshold, mark the pixels exceeding the deformation gradient threshold as the over-standard area, and generate a binary image including a foreground area and a background area. Calculate the total area of the section of the permafrost subgrade according to the total number of pixels in the binary image and the fixed pixel size. Calculate the area of the differential over-standard area in the binary image. Determine the subgrade disease index based on the total area of the section and the area of the differential over-standard area.

9. The method according to claim 8, wherein Use a differential deformation gradient slope rate of 0.51% as the deformation gradient threshold.

10. The method according to claim 1, characterized in that, Determine the disease warning level of the permafrost subgrade according to the subgrade disease index, including: Determine the corresponding disease warning level of the permafrost subgrade according to the numerical interval where the disease index is located. Among them, if the subgrade disease index is greater than 0.7 and less than or equal to 1, determine that the disease warning level of the permafrost subgrade is a severe disaster warning level. If the subgrade disease index is greater than 0.1 and less than or equal to 0.7, determine that the disease warning level of the permafrost subgrade is a moderate disaster warning level. If the subgrade disease index is greater than or equal to 0 and less than or equal to 0.1, determine that the disease warning level of the permafrost subgrade is a mild disaster warning level.

Images