Cliff road section screening method and device

Abstract

The invention relates to the field of road safety protection, and discloses a cliff road section screening method and device, and the method comprises the steps: obtaining digital elevation model data and road vector data of a plurality of regions; performing anisotropic filtering processing for protecting topographic feature lines on the digital elevation model data; calculating a topographic slope map according to the filtered digital elevation model data, and screening the topographic slope map according to a preset slope threshold to generate a slope binary image; on the basis of the filtered digital elevation model data, generating contour line data according to a preset distance between preset contour lines, and obtaining a gradient candidate region set according to the road vector data; superposing the gradient binary image and the gradient candidate region set to obtain a cliff road section candidate region set; and obtaining a final cliff-near candidate area set according to the cliff-near road section candidate area set and the contour line data, and obtaining a complete cliff-near road section according to the road vector data and the final cliff-near candidate area set. According to the invention, the time and economic cost is obviously reduced, and the precision of cliff road section extraction is improved.

Description

A method and device for screening road sections near cliffs Technical Field

[0001] This disclosure relates to the field of road safety protection technology, specifically to the field of road safety protection data updating, and particularly to a method and device for screening road sections near cliffs. Background Technology

[0002] Rapidly obtaining information on road sections near cliffs and their dangerous sides is crucial for installing warning signs, crash barriers, and other life-saving facilities. Identifying and identifying the locations of road sections near cliffs is essential for ensuring traffic safety, preventing natural disasters, providing early warnings for protective measures, improving traffic management efficiency, and reducing economic losses. This is especially important for rural and remote areas, where quickly identifying key areas along road sections near cliffs is vital for protecting people's lives and property.

[0003] Currently, the most common technical method for identifying road sections near cliffs is through the collection of information by various sensors mounted on vehicle systems. This involves comprehensively judging the data based on video images, slope, and other factors. However, this method is labor-intensive, requires field data collection, and is unsuitable for large-scale investigation of road sections with safety hazards and for identifying road sections near cliffs, making it difficult to achieve comprehensive coverage. Furthermore, with the rapid development of spatial technology, remote sensing and GIS (Geographical Information System) technologies provide effective technical means for acquiring large-scale basic geographic information. Although remote sensing can acquire geographic information over a large area and locate road sections near cliffs by extracting ridges, valleys, and changes in road slope, the terrain of ridges, valleys, and cliffs differs greatly from that of the actual cliff, and it does not consider the vertical elevation difference between the roadside and the ridge, which can easily lead to incorrect identification.

[0004] Furthermore, digital elevation model (DEM) data, especially publicly available and free data, often suffers from quality issues due to low resolution and limitations in processing algorithms (such as interpolation methods). This can lead to the presence of localized, unreasonable elevation anomalies (such as abrupt elevation changes, local convex hulls, and depressions), which can cause misjudgments of road sections near cliffs. The most common method for handling DEM anomalies is filtering. However, common filtering methods such as mean filtering and median filtering are mostly isotropic, meaning the filtering method is the same for each pixel in all directions. This approach can easily blur certain terrain features such as ridges and valleys, affecting the identification of road sections near cliffs.

[0005] Therefore, there is an urgent need for a method that can effectively overcome the shortcomings of existing technologies, accurately and efficiently screen a large number of road sections near cliffs without requiring large-scale fieldwork, and optimize preprocessing specifically for the characteristics of DEM data to improve the accuracy of identifying road sections near cliffs. Summary of the Invention

[0006] This disclosure provides a method and apparatus for screening road sections near cliffs, which solves the technical problem of being unable to accurately extract road sections near cliffs.

[0007] According to a first aspect of this disclosure, a method for screening road sections near cliffs is provided. The method includes: acquiring digital elevation model (DEM) data and road vector data from multiple regions; performing anisotropic filtering on the DEM data oriented towards protected terrain feature lines; calculating a terrain slope map based on the filtered DEM data, filtering the terrain slope map according to a preset slope threshold, and generating a binary slope map; generating contour line data based on the filtered DEM data at preset contour line intervals and preset distances, and obtaining a set of slope candidate areas based on the road vector data; superimposing the binary slope map with the set of slope candidate areas to obtain a set of candidate areas for road sections near cliffs; obtaining a final set of candidate areas for road sections near cliffs based on the set of candidate areas for road sections near cliffs and the contour line data; and obtaining a complete road section near a cliff based on the road vector data and the final set of candidate areas for road sections near cliffs.

[0008] In addition to the above aspects and any possible implementations, a further implementation is provided, which performs anisotropic filtering processing on the digital elevation model data for protecting terrain feature lines, including: extracting terrain feature lines from the digital elevation model data, the terrain feature lines including ridge lines and valley lines; calculating the feature direction of each pixel based on the terrain feature lines; constructing an anisotropic filtering kernel for each pixel based on the feature direction; and using the anisotropic filtering kernel to filter the digital elevation model data.

[0009] In addition to the above aspects and any possible implementation methods, a further implementation method is provided, which obtains a final set of candidate areas for cliff-side road sections based on the candidate area set for cliff-side road sections and contour line data, including: overlaying contour line data on the candidate area set for cliff-side road sections; filtering each candidate area in the candidate area set for cliff-side road sections based on elevation difference conditions to obtain a set of cliff-side road section elevation difference filters; determining the relationship between elevation difference and route direction in the set of cliff-side road section elevation difference filters, and obtaining a set of cliff-side road section direction judgments based on the judgment results; and removing foreign objects from the set of cliff-side road section direction judgments to obtain the final set of candidate areas for cliff-side road sections.

[0010] In some possible embodiments, each candidate area in the set of candidate areas for road sections near cliffs is filtered based on elevation difference conditions to obtain a set of candidate areas for elevation difference of road sections near cliffs. This includes: overlaying contour line data onto the set of candidate areas for road sections near cliffs to obtain a set of candidate areas for road sections near cliffs after overlay; if the number of contour lines covered in each candidate area in the set of candidate areas for road sections near cliffs after overlay is greater than the preset number of contour lines, then candidate areas for road sections near cliffs with a greater than the preset number of contour lines will be extracted to generate a set of candidate areas for elevation difference of road sections near cliffs.

[0011] In some possible embodiments, the elevation difference screening set of the cliff-side road section is used to determine the relationship between the elevation difference and the direction of the route, and the direction judgment set of the cliff-side road section is obtained based on the judgment result. This includes: overlaying contour lines on each candidate area in the cliff-side road section elevation difference screening set, determining whether the contour lines in the candidate area are horizontally related to the road vector line; extracting candidate areas where the contour lines are horizontally related to the route, and obtaining the direction judgment set of the cliff-side road section.

[0012] In some possible embodiments, determining whether the contour lines and road vector lines within the candidate area are horizontally related includes: finding the intersections of the road vector lines with contour lines at different elevations within the candidate area; if the number of intersections with each contour line at a different elevation is less than a preset number, then the contour lines are determined to be horizontally related to the route; if the number of intersections with each contour line at a different elevation is greater than the preset number, then the elevation range of the road vector lines is determined to be large, and the contour lines are not horizontally related to the route.

[0013] In some possible embodiments, foreign objects are excluded from the set of direction judgments for road sections near the cliff to obtain a final set of candidate areas for the cliff. This includes: determining whether there are foreign objects in each candidate area of ​​the set of direction judgments for road sections near the cliff through remote sensing images; and extracting candidate areas without foreign objects to obtain the final set of candidate areas for the cliff.

[0014] In some possible embodiments, a complete cliff-side road segment is obtained based on road vector data and the final set of cliff-side candidate areas, including: superimposing the road vector route with the final set of cliff-side candidate areas, breaking the road segment at the boundary of the candidate area to obtain a vector road segment within the candidate area, which is denoted as a cliff-side road segment; and connecting and merging adjacent cliff-side road segments to form a complete cliff-side road segment.

[0015] In some possible embodiments, road segments are broken at the boundaries of candidate regions to obtain vector road segments within the candidate region. This includes: sequentially rasterizing each candidate region, and cropping the vectorized candidate region surfaces and road vectors respectively; retaining the original road attribute information for each cropped cliff-side road segment, and sequentially obtaining multiple vector segments; the original road attribute information includes the original road route code, route start coordinates, route end coordinates, route start station number, and route end station number.

[0016] According to a second aspect of this disclosure, a device for screening road sections near cliffs is provided. The device includes: an acquisition module for acquiring digital elevation model data and road vector data from multiple regions; a filtering module for performing anisotropic filtering on the digital elevation model data oriented towards protected terrain feature lines; a first generation module for calculating a terrain slope map based on the filtered digital elevation model data, filtering the terrain slope map according to a preset slope threshold, and generating a slope binary map; a second generation module for generating contour line data based on the filtered digital elevation model data at preset contour line intervals and preset distances, and obtaining a set of slope candidate areas based on the road vector data; an overlay module for overlaying the slope binary map with the set of slope candidate areas to obtain a set of candidate areas for road sections near cliffs; and a road section near cliff generation module for obtaining a final set of candidate areas for road sections near cliffs based on the set of candidate areas for road sections near cliffs and the contour line data, and obtaining a complete road section near cliffs based on the road vector data and the final set of candidate areas for road sections near cliffs.

[0017] According to a third aspect of this disclosure, an electronic device is provided. The electronic device includes a memory and a processor, wherein a computer program is stored in the memory, and the processor executes the program to implement the method described above.

[0018] According to a fourth aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the methods according to the first and / or second aspects of this disclosure.

[0019] The beneficial effects of this application are as follows: By using spatial data such as DEM and road vectors, there is no need for field data collection by vehicle-mounted sensors or manual on-site measurement, completely avoiding the high labor costs and long cycles of field operations. It is especially suitable for rural, remote or dangerous areas, enabling the investigation of cliff-side road sections with zero field investment, significantly reducing time and economic costs. Furthermore, it improves positioning accuracy through multi-layered condition constraints: first, high-slope basic terrain is screened using slope thresholds, then the surrounding area of ​​the road is locked using road vectors, and finally, contour line data is used to verify whether the elevation difference meets the standard. This avoids the mis-extraction caused by single conditions in traditional methods, greatly reducing the false positive and false negative rates, and providing accurate data support for hazard investigation. Combining DEM data quality and its application in cliff-side road section extraction, this invention proposes an anisotropic filtering DEM preprocessing method for protecting terrain feature lines. This method differs from the commonly used isotropic filtering method, and can retain terrain features with a greater probability while removing elevation anomalies, thus improving the accuracy of cliff-side road section extraction.

[0020] It should be understood that the description in the Summary of the Invention is not intended to limit the key or essential features of the embodiments of this disclosure, nor is it intended to restrict the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0021] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. The drawings are provided for a better understanding of the present invention and are not intended to limit the scope of this disclosure. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein: Figure 1 shows a flowchart of a cliffside road section screening method according to an embodiment of this disclosure; Figure 2 shows a schematic diagram of contour lines parallel to a route according to an embodiment of this disclosure; Figure 3 shows a schematic diagram of contour lines not parallel to a route according to an embodiment of this disclosure; Figure 4 shows an example remote sensing image a according to an embodiment of this disclosure; Figure 5 shows an example remote sensing image b according to an embodiment of this disclosure; Figure 6 shows a complete flowchart of the cliffside road section screening method according to an embodiment of this disclosure; Figure 7 shows a block diagram of a cliffside road section screening device according to an embodiment of this disclosure; Figure 8 shows a block diagram of an exemplary electronic device capable of implementing embodiments of this disclosure. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0023] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0024] Rapidly obtaining information on road sections near cliffs and their dangerous sides is crucial for installing warning signs, crash barriers, and other life-saving facilities. Identifying and identifying the locations of road sections near cliffs is essential for ensuring traffic safety, preventing natural disasters, providing early warnings for protective measures, improving traffic management efficiency, and reducing economic losses. This is especially important for rural and remote areas, where quickly identifying key areas along road sections near cliffs is vital for protecting people's lives and property.

[0025] Currently, the most common technical method for identifying road sections near cliffs is to collect information using various sensors mounted on vehicle systems. This involves comprehensively judging the location of road sections near cliffs using video images, slope data, and other data. However, this method is costly due to its high labor costs and the need for field data collection, making it unsuitable for large-scale investigation of road sections with safety hazards and for identifying road sections near cliffs, thus failing to achieve comprehensive coverage. With the rapid development of spatial technology, remote sensing and GIS spatial technologies have provided effective technical means for acquiring large-scale basic geographic information. DEM (Digital Elevation Model) data is used to extract information such as ridges, valleys, slopes, and slope changes along the road direction, thereby identifying key road sections near cliffs for protection. However, while ridges and valleys are typical landforms, their morphology differs significantly from that of road sections near cliffs. Directly using ridges and valleys cannot accurately locate the position of the cliff, leading to many erroneous extraction results. Meanwhile, although existing technologies consider the changes in elevation of segmented curves (elevation changes along the road direction), they focus more on the changes in elevation along the vertical road line, i.e., the roadside, when dealing with cliff sections. This can lead to many erroneous extractions in the investigation results of cliff-side road sections.

[0026] This application fully utilizes the characteristics of digital elevation models (DEMs) in spatial analysis technology, specifically targeting the features of road sections near cliffs. It leverages the distribution characteristics of DEMs and corresponding contour lines, such as slope, contour line parallelism, tangent direction, and spatial relationship with the road section, particularly the characteristics perpendicular to the roadline, to comprehensively determine the location and accuracy of road sections near cliffs. This patent enables rapid location and investigation of road sections near cliffs, achieving low-cost, field-free investigation and rapid positioning of such sections.

[0027] Figure 1 shows a flowchart of a cliffside road section screening method 100 according to an embodiment of the present disclosure, including the following steps: Step S110, acquiring digital elevation model data and road vector data for multiple regions.

[0028] In step S110, the selected digital elevation model data, i.e., DEM data, can be derived from public datasets (such as ALOS with a resolution of 12.5 meters) or DEM data acquired by UAVs (such as 0.1-0.2 meters). The higher the spatial resolution of the DEM, the more accurate the results of extracting road sections near cliffs.

[0029] Step S120: Perform anisotropic filtering on the digital elevation model data to protect the terrain feature lines.

[0030] In step 120, the anisotropic filtering for protecting terrain feature lines mainly includes the following steps: terrain feature line extraction (ridge lines and valley lines), pixel feature orientation field calculation, anisotropic filtering kernel construction, and filtering processing. Specifically: S121: Terrain feature line extraction. Terrain feature line extraction (ridge lines and valley lines) based on the DEM can be achieved using ArcMap software, and then binarized to obtain binary images of ridge lines and valley lines. For example, for ridge line extraction, ArcMap software can be used to first extract the slope aspect ratio (SOA) and positive / negative terrain (ZFDX), and then the grid calculator in the ArcToolbox can be used to calculate the ridge line using the formula shanji=(“ZFDX”>0)&(“SOA”>80).

[0031] The method for extracting the aspect variability (SOA) is as follows: First, calculate the maximum elevation value (MaxValue) of the original DEM data layer. Using the Raster Calculator tool, input the formula FDEM = MaxValue - DEM to obtain the DEM data layer "FDEM" which is opposite to the original terrain. Use the Aspect tool to extract the aspect from the "FDEM" data to obtain the aspect data "FDEM1". Then, use the Slope tool to extract the slope from the "FDEM1" data to obtain the aspect variability data "SOA" which is opposite to the terrain.

[0032] The extraction method for positive and negative terrain (ZFDX) is as follows: Use the Spatial Analysis tool, then the Neighborhood Analysis tool, and finally the Focus Statistics tool. Set the statistics type to Mean (MEAN) to obtain the layer "MEAN". Use the Raster Calculator to calculate the distribution area of ​​positive and negative terrain using the formula ZFDX = "DEM" - "MEAN".

[0033] The extraction of valley lines is similar to that of ridge lines. First, use the [Raster Calculator] to obtain inverse terrain data that is completely opposite to the DEM terrain, such as the formula: Abs(“dem”-MaxValue). Subsequent processes are the same as those for ridge line processing.

[0034] S122: Pixel Feature Orientation Field Calculation. For pixel feature orientation field calculation, the slope aspect and direction of each pixel are calculated, and combined with their relationship to terrain feature lines (ridge lines and valley lines) to form the final feature orientation. .

[0035] The slope aspect can be calculated using the Spatial Analyst tool in ArcToolbox of ArcMap software. Selecting the slope aspect will extract the slope aspect results.

[0036] The direction calculation refers to the direction parallel to the contour lines, i.e., perpendicular to the slope aspect, which is the direction that needs to be protected during the filtering process. For each pixel, its direction... It can be seen from its slope The calculation is then normalized to the range of 0-360 degrees (modulo 360). The formula is as follows: For pixel regions far from terrain feature lines, That is, the result calculated in the above steps. For determining ordinary areas far from terrain feature lines, a buffer zone of a certain radius (e.g., 20m) can be constructed based on the ridgeline and valley line extracted in the above steps. Pixels outside this buffer zone are pixel areas far from terrain feature lines.

[0037] For pixels located at or near ridge / valley lines, their orientation should be based on the tangent direction of the feature line. The orientation of this point, determined by a weighted average of the feature line tangent, allows for the extraction of... It is more stable than local slope aspect calculation. The implementation method is as follows: 1) Calculate the distance D of each pixel to the nearest ridgeline / valleyline.

[0038] 2) Define the exponential distance decay function as shown in the following formula.

[0039] Here, L is the attenuation scale parameter, used to control the attenuation rate. L can be set according to the DEM resolution or the feature line width of a typical terrain. For example, it can be set to L = 5 × cell_size, where cell_size represents the actual size corresponding to one pixel. The larger the value of L, the wider the influence range of the feature line direction.

[0040] 3) Calculate the tangent direction from the pixel to the nearest ridgeline / valleyline. ;4) The direction of local calculation Tangent direction of the nearest feature line Weighted fusion is performed to obtain the final feature orientation. The specific calculation formula is as follows: S123: Construction of Anisotropic Filter Kernel. The purpose of constructing anisotropic filter kernels is to build a filter kernel with adaptive orientation and shape for each pixel, including two steps: construction of elliptical filter kernel and adaptive determination of filter intensity.

[0041] Construction of the elliptic filter kernel: 1) The major axis direction of the ellipse is as calculated in the above steps. 2) The direction of the minor axis of the ellipse is along... 3) Major axis length a: Determines the smoothness along the feature direction. To protect the feature line, this value should be large (e.g., 5-7 pixels), allowing more pixels to be averaged along the ridge / valley direction, making the feature line smoother. 4) Minor axis length b: Determines the smoothness perpendicular to the feature direction. To eliminate noise across the feature line, this value should be small (e.g., 1-3 pixels), limiting the smoothing range and avoiding blurring the boundaries of ridges / valleys.

[0042] Adaptive Determination of Filter Strength Setting: The filter strength is defined by a / b, which represents the anisotropic characteristic. To better protect the feature line, this patent proposes an adaptive filter strength. The adaptive method is based on the distance to the feature line. The closer the pixel is to the feature line, the larger the anisotropic ratio a / b, and the stronger the protection. This patent uses a fixed major axis length 'a', mainly adjusting the minor axis radius 'b'. That is, when D is small (close to the feature line), b should be small (strict protection); when D is large (far from the feature line), b can be larger (active noise reduction). The adaptive function of b is shown in the following formula: in, The maximum minor axis radius (e.g., 3 meters). The minimum minor axis radius (e.g., 1 meter). This is a cutoff distance; anything beyond this distance is considered too far from the feature line, and the maximum filtering intensity is applied. This can be set according to actual needs (e.g., 50 meters).

[0043] S124: Filtering Process The filtering process is implemented as follows: For each pixel (i,j) in the DEM, using it as the center, select all pixels within the elliptical filter kernel tailored to it. Apply mean filtering to the elevation values ​​of these pixels; the result is the new elevation value of the center point. Process each pixel individually to complete the filtering of the entire DEM. It should be noted that a weighted average (such as Gaussian weighted filtering) can be applied to all pixels within the elliptical filter; however, this patent uses mean filtering to improve processing efficiency.

[0044] Through step S120 above, anisotropic filtering processing oriented towards protected terrain feature lines is performed on the digital elevation model data to generate the filtered DEM result. This step S120 effectively eliminates local elevation anomalies (such as abrupt changes, convex hulls, and depressions) in the original DEM data caused by sensor errors, data processing interpolation, etc., providing a stable and reliable data foundation for subsequent slope calculation and contour line generation, avoiding misjudgment of "false cliffs," thus reducing misjudgment from the source and improving overall accuracy. At the same time, unlike ordinary filtering, the method of this invention is "anisotropic" and "oriented towards protected terrain feature lines," which can consciously protect and enhance terrain features such as ridge lines and valley lines that are crucial for identifying cliff-adjacent road sections while smoothing noise. These feature lines themselves are the areas where potential cliff-adjacent road sections are distributed. By specially protecting characteristic lines such as ridges and valleys, the filtered DEM makes the slope and contour features of these areas clearer and more continuous, enhancing the "signal strength" of truly cliff-adjacent road sections. This allows these real risk road sections to be captured more completely and accurately when generating binary slope maps and screening for elevation differences, effectively reducing the false negative rate. Furthermore, because the DEM retains clear terrain features, the generated contour lines more realistically reflect the terrain's orientation. This ensures the geometric basis for the subsequent crucial step of determining whether "contour lines and road vector lines are horizontal" is reliable, enabling a more accurate distinction between roadside cliffs and steep slopes along the road, further refining the final set of cliff-adjacent candidate areas.

[0045] Step S130: Calculate the terrain slope map based on the filtered digital elevation model data, filter the terrain slope map according to the preset slope threshold, and generate a binary slope map.

[0046] In some embodiments, a slope map is calculated based on digital elevation model (DEM) data, and regions with slopes higher than the Thangle are filtered based on the slope map to obtain a filtered binary slope map G1. In the binary slope map G1, 1 represents a region with a slope higher than the Thangle, and 0 represents a region with a slope lower than the Thangle. As an example, the Thangle can be 33 degrees, which can be fine-tuned as needed.

[0047] The selected DEM data can be sourced from publicly available datasets (such as ALOS with a resolution of 12.5 meters) or DEM data acquired by drones (such as 0.1-0.2 meters). Higher spatial resolution of the DEM results in more accurate extraction of road sections near cliffs. As a concrete example, the process of calculating slope based on the DEM can be achieved using ArcMap software. For instance, using the ArcToolbox in ArcMap, selecting "Spatial Analysis," then "Surface Analysis," and finally "Slope," will generate a slope map.

[0048] Step S140: Based on the digital elevation model data, generate contour line data at preset intervals and distances, and obtain a set of slope candidate areas based on the road vector data.

[0049] As an example, contour data can be generated based on digital elevation model (DEM) data at preset intervals. The contour interval is ΔH. Specifically, generating contour data from DEM data can be achieved using ArcMap software. For example, using the ArcToolbox in ArcMap, select "Grid Surface" in the 3D Analyst tool and perform "Contour" analysis. Since a road section near a cliff is defined as a section with a steep cliff or deep ditch greater than H (6-8m) on the roadside, the contour interval ΔH can be set to 5m, and fine-tuned as needed.

[0050] As an example, a set of slope candidate areas is obtained based on road vector data. For instance, based on road vector data, a buffer area with a buffer radius of Buffer is generated for the road vector line, forming a set of slope candidate areas B. The value within the buffer area is 1, and the value in other areas is 0.

[0051] Step S150: Overlay the slope binary map with the slope candidate area set to obtain the candidate area set for cliff-side road sections.

[0052] The intersection of the slope binary map G1 and the slope candidate area set B is calculated to obtain the cliff-side road segment candidate area set G2. The cliff-side road segment candidate area set G2 represents the area near the road where there is a slope higher than Thangle, and each area in the cliff-side road segment candidate area set G2 is a candidate area for a cliff-side road segment.

[0053] Step S160: Based on the candidate area set of the cliffside road segment and the contour data, obtain the final candidate area set of the cliffside road segment. Based on the road vector data and the final candidate area set of the cliffside road segment, obtain the complete cliffside road segment.

[0054] In some embodiments, the final set of candidate areas for road sections near cliffs is obtained based on the candidate area set for road sections near cliffs and contour line data, including: overlaying contour line data onto the candidate area set for road sections near cliffs; filtering each candidate area in the candidate area set for road sections near cliffs based on elevation difference conditions to obtain a set of elevation difference filtering for road sections near cliffs; determining the relationship between elevation difference and route direction in the set of elevation difference filtering for road sections near cliffs, and obtaining a set of direction judgments for road sections near cliffs based on the judgment results; and removing foreign objects from the set of direction judgments for road sections near cliffs to obtain the final set of candidate areas for road sections near cliffs.

[0055] In some embodiments, each candidate area in the set of candidate areas for road sections near cliffs is filtered based on elevation difference conditions to obtain a set of candidate areas for elevation difference of road sections near cliffs. This includes: overlaying contour line data onto the set of candidate areas for road sections near cliffs to obtain a set of candidate areas for road sections near cliffs after overlay; if the number of contour lines covered in each candidate area in the set of candidate areas for road sections near cliffs after overlay is greater than the number of preset contour lines, then candidate areas for road sections near cliffs with a number greater than the preset number of contour lines will be extracted to generate a set of candidate areas for elevation difference of road sections near cliffs.

[0056] As an example, contour data (Contour) is overlaid on the candidate area set G2 for road sections near cliffs. The number of contour lines covered in each candidate area of ​​G2 is then determined. If the number of contour lines is greater than the preset number of contour lines (NContour), it proves that the elevation difference within the candidate area meets the condition of being near a cliff; otherwise, it does not. The set of areas within G2 that meet the elevation difference condition is then designated as the cliff elevation difference screening set G3, and the process proceeds to the next step.

[0057] Specifically, simply calculating the absolute value of the difference between the maximum and minimum values ​​within a region to judge the situation within that region can easily lead to misjudgments of the overall elevation trend due to isolated local anomalies. The elevation difference characteristics of road sections near cliffs should be manifested as a gradual transition in elevation along a slope, with the elevation difference exceeding a certain threshold; contour lines can reflect this characteristic. To avoid judging elevation differences based on local anomalies in the DEM image, this method uses contour lines to determine elevation differences. The number of contour lines within a region represents the trend of elevation differences within that region. The preset number of contour lines, NContour, can be calculated using NContour = H / ΔH, and can be fine-tuned as needed.

[0058] In some embodiments, the relationship between the elevation difference screening set of road sections near cliffs and the direction of the route is determined, and the direction judgment set of road sections near cliffs is obtained based on the judgment result. This includes: overlaying contour lines on each candidate area in the elevation difference screening set of road sections near cliffs, determining whether the contour lines in the candidate area are horizontally related to the road vector line; extracting candidate areas where the contour lines are horizontally related to the route, and obtaining the direction judgment set of road sections near cliffs.

[0059] In some embodiments, determining whether the contour lines and road vector lines within the candidate area are horizontally related includes: finding the intersections of the road vector lines with contour lines at different elevations within the candidate area; if the number of intersections with each contour line at a different elevation is less than a preset number of intersections, then the contour lines are determined to be horizontally related to the route; if the number of intersections with each contour line at a different elevation is greater than the preset number of intersections, then the elevation range of the road vector lines is determined to be large, and the contour lines are not horizontally related to the route.

[0060] As an example, determining the relationship between elevation difference and route direction specifically includes: for each candidate region in the elevation difference screening set G3 for cliff-side road sections, superimposing contour lines to determine the spatial relationship between the contour lines and the road direction. Figure 2 shows a schematic diagram of contour lines parallel to the route, and Figure 3 shows a schematic diagram of contour lines not being parallel to the route. If the contour lines in a candidate region are basically horizontal to the route, they are added to the candidate region set, forming the cliff-side road section direction determination set G4, and proceeding to the next step of the judgment.

[0061] Specifically, the method for determining whether a contour line is basically horizontal to the route is as follows: the road vector line intersects with contour lines at different elevations within the candidate area. If the number of intersections with each contour line at a different elevation is less than the preset number of intersections N, then the contour line is considered to be basically horizontal to the route. If the number of intersections is greater than the preset number of intersections N, then the elevation range of the road vector line is considered to be large, and the contour line is not basically horizontal to the route. The preset number of intersections N can be set to 1-2 and fine-tuned as needed.

[0062] The purpose of determining the relationship between the elevation difference direction and the route direction is to identify sections that are steep slopes rather than cliff-side sections if the elevation difference is along the route direction. This would lead to the incorrect identification of sections that are actually cliff-side, thus eliminating incorrect identification caused by elevation differences along the road direction.

[0063] In some embodiments, foreign objects are excluded from the set of direction judgments for road sections near the cliff to obtain a final set of candidate areas for the cliff, including: determining whether there are foreign objects in each candidate area of ​​the set of direction judgments for road sections near the cliff through remote sensing images; and extracting candidate areas without foreign objects to obtain the final set of candidate areas for the cliff.

[0064] As an example, Figure 4 shows example image a of remote sensing image, and Figure 5 shows example image b of remote sensing image. Information on terraced fields and trees in remote sensing images can be obtained through manual visual interpretation or automatic extraction aided by artificial intelligence. Other influencing features in the remote sensing images are excluded because terraced fields and other features can easily have similar characteristics to cliffs in the DEM. For each candidate area in the cliff-side road segment direction judgment set G4, the presence of terraced fields and other features in each candidate area is determined through remote sensing imagery. The final cliff-side candidate area set G5 is then used as the final range for cliff-side road segment extraction.

[0065] In some embodiments, obtaining a complete cliff-side road segment based on road vector data and the final set of cliff-side candidate areas includes: superimposing the road vector route with the final set of cliff-side candidate areas, breaking the road segment at the boundary of the candidate area to obtain a vector road segment within the candidate area, denoted as a cliff-side road segment; and connecting and merging adjacent cliff-side road segments to form a complete cliff-side road segment.

[0066] Due to the potential for anomalies in some DEM elevation data, directly extracted cliffside road sections may exhibit gaps between adjacent road sections. To ensure the completeness of the extraction results, adjacent cliffside road sections are connected and merged to form a complete cliffside road section.

[0067] As an example, the specific method for processing the expanded fragmented road segments is as follows: For each cliff-side road segment with the same route code, calculate the route station numbers corresponding to the start and end points of each cliff-side road segment, and sort them from smallest to largest based on the start station numbers of each cliff-side road segment to obtain the arrangement order of each cliff-side road segment along the route. The start and end station numbers of each cliff-side road segment are recorded as: (Q0,Z0),(Q1,Z1),...,(Qi,Zi),...,(Qn,Zn), where Qi is the start station number of the i-th cliff-side road segment, and Zi is the end station number of the i-th cliff-side road segment. If the distance |Qi+1-Zi| between the end station number Zi of the previous cliff-side road segment and the start station number Qi+1 of the next cliff-side road segment is less than the threshold ΔD, then the two cliff-side road segments need to be connected.

[0068] Specifically, the calculation method for the station number corresponding to the start or end point of the road section near the cliff can be achieved by using the QueryPointAndDistance tool in ArcEngine / Arcpy to realize the position of a point in space to the nearest point on the curve and the distance from the start point, that is, the station number from the start point of the route.

[0069] Specifically, the process of connecting two cliff-side road sections is as follows: In order to maintain the original route alignment after the expansion of the fragmented road sections, the original route vector is broken based on the position of the end point of the previous cliff-side road section and the starting point of the next cliff-side road section, and the broken road sections are added to the gap between the two cliff-side road sections.

[0070] In some embodiments, the road segment is broken at the boundary of the candidate area to obtain vector road segments within the candidate area, including: sequentially performing raster vectorization on each candidate area, and cropping the vectorized candidate area surface and road vector respectively; retaining the original road attribute information for each cropped cliff-side road segment, and sequentially obtaining multiple vector segments; the original road attribute information includes the original road route code, route start coordinates, route end coordinates, route start station number, and route end station number.

[0071] As an example, Figure 6 shows the complete flowchart of the cliff-adjacent road section screening method, which includes: acquiring DEM data, road vector data containing route codes / start and end station numbers / locations, and high-resolution remote sensing imagery. A slope map is generated based on the DEM, and a high-slope candidate area G1 (identifying potential cliff-adjacent slope terrain) is obtained through the "slope screening range"; contour lines are generated based on the DEM (for subsequent elevation difference analysis); and a road buffer zone B is generated based on the road vectors (delineating the area around the road to be analyzed). Intersecting G1 with space B, we generate region G2 around the road with a slope that meets the requirements. Using contour lines, we filter G2 by elevation difference to obtain region G3 (selecting regions with acceptable elevation differences). We determine the directional relationship between the elevation difference in G3 and the route to obtain region G4 (ensuring the elevation difference is perpendicular to the road, meeting the "roadside elevation difference" characteristic of cliffside). Using high-resolution remote sensing imagery, we eliminate interference from features such as terraces in G4 to obtain region G5. Based on G5, we extract the cliffside road segment. After processing the extracted road segment by expanding it into smaller segments, we finally obtain the complete cliffside road segment.

[0072] According to the embodiments of this disclosure, the following technical effects are achieved: By using spatial data such as DEM and road vectors, there is no need for field data collection by vehicle-mounted sensors or manual on-site measurement, completely avoiding the problems of high labor costs and long cycles in field operations. It is especially suitable for rural, remote or dangerous areas, enabling the investigation of cliff-side road sections with zero field investment, significantly reducing time and economic costs. Furthermore, the positioning accuracy is improved through multi-layered condition constraints: first, high-slope basic terrain is screened using slope thresholds, then the surrounding area of ​​the road is locked using road vectors, and finally, contour line data is used to verify whether the elevation difference meets the standard. This avoids the mis-extraction caused by a single condition in traditional methods, greatly reducing the false positive rate and false negative rate, and providing accurate data support for hazard investigation.

[0073] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this disclosure is not limited to the described order of actions, because according to this disclosure, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this disclosure.

[0074] The above is an introduction to the method embodiments. The following describes the present disclosure further through device embodiments.

[0075] Figure 7 shows a block diagram of a cliffside road section screening device 700 according to an embodiment of the present disclosure. As shown in Figure 7, the device 700 includes: an acquisition module 710, a filtering module 720, a first generation module 730, a second generation module 740, an overlay module 750, and a cliffside road section generation module 760. The system includes: an acquisition module for acquiring digital elevation model data and road vector data from multiple regions; a filtering module for performing anisotropic filtering on the digital elevation model data to protect terrain feature lines; a first generation module for calculating a terrain slope map based on the filtered digital elevation model data, filtering the terrain slope map according to a preset slope threshold, and generating a slope binary map; a second generation module for generating contour line data based on the filtered digital elevation model data at preset contour line intervals and distances, and obtaining a set of slope candidate areas based on the road vector data; an overlay module for overlaying the slope binary map with the set of slope candidate areas to obtain a set of cliff-side road segment candidate areas; and a cliff-side road segment generation module for obtaining a final set of cliff-side candidate areas based on the cliff-side road segment candidate area set and contour line data, and obtaining a complete cliff-side road segment based on the road vector data and the final set of cliff-side candidate areas. By utilizing spatial data such as DEM and road vectors, this method eliminates the need for vehicle-mounted sensors for field data collection or manual on-site measurements, completely avoiding the high labor costs and long lead times associated with fieldwork. It is particularly suitable for rural, remote, or dangerous areas, enabling the investigation of cliff-side road sections with zero fieldwork investment, significantly reducing time and economic costs. Furthermore, it improves positioning accuracy through multi-layered constraints: first, it filters high-slope base terrain using slope thresholds; then, it locks the surrounding area of ​​the road using road vectors; and finally, it verifies whether the elevation difference meets the standards using contour line data. This avoids the mis-extraction caused by single conditions in traditional methods, greatly reducing the false positive and false negative rates, and providing accurate data support for hazard investigation. By fully integrating the elevation characteristics of cliff-side areas, an anisotropic filtering DEM preprocessing method oriented towards protecting terrain feature lines is proposed. This not only improves the accuracy of cliff-side road section extraction but also preserves terrain features as much as possible, outperforming general filtering methods.

[0076] In some possible embodiments, the cliffside road segment generation module is used to overlay contour data onto the cliffside road segment candidate area set, filter each candidate area in the cliffside road segment candidate area set based on elevation difference conditions to obtain the cliffside road segment elevation difference filter set; determine the relationship between elevation difference and route direction in the cliffside road segment elevation difference filter set, and obtain the cliffside road segment direction judgment set based on the judgment result; remove foreign objects from the cliffside road segment direction judgment set to obtain the final cliffside candidate area set.

[0077] In some possible embodiments, the cliffside road segment generation module is used to overlay contour data onto the cliffside road segment candidate area set to obtain the overlaid cliffside road segment candidate area set; if the number of contour lines covered in each candidate area in the overlaid cliffside road segment candidate area set is greater than the preset number of contour lines, then the cliffside road segment candidate areas with a greater than the preset number of contour lines will be extracted to generate the cliffside road segment elevation difference screening set.

[0078] In some possible embodiments, the cliffside road segment generation module is used to overlay contour lines on each candidate region in the cliffside road segment elevation difference screening set, determine whether the contour lines in the candidate region are horizontally related to the road vector line, and extract the candidate regions where the contour lines are horizontally related to the route to obtain the cliffside road segment direction judgment set.

[0079] In some possible embodiments, the cliffside road section generation module is used to find the intersection of the road vector line with contour lines at different heights within the candidate area. If the number of intersections with the contour lines at different heights is less than the preset number of intersections, it is determined that the contour lines are horizontal to the route. If the number of intersections with the contour lines at different heights is greater than the preset number of intersections, it is determined that the elevation range of the road vector line is large and the contour lines are not horizontal to the route.

[0080] In some possible embodiments, the cliffside road section generation module is used to determine whether there are foreign objects in each candidate area of ​​the cliffside road section direction judgment set by remote sensing imagery; and to extract the candidate areas where there are no foreign objects to obtain the final cliffside candidate area set.

[0081] In some possible embodiments, the cliffside road segment generation module is used to sequentially perform raster vectorization on each candidate area, and then clip the vectorized candidate area surfaces and road vectors respectively; for each clipped cliffside road segment, the original road attribute information is retained, and multiple vector segments are obtained sequentially; the original road attribute information includes the original road route code, route start coordinates, route end coordinates, route start station number, and route end station number.

[0082] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the described module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0083] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0084] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0085] Figure 8 illustrates a schematic block diagram of an electronic device 800 that can be used to implement embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0086] Electronic device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in ROM 802 or a computer program loaded into RAM 803 from storage unit 808. RAM 803 can also store various programs and data required for the operation of electronic device 800. The computing unit 801, ROM 802, and RAM 803 are interconnected via bus 804. I / O interface 805 is also connected to bus 804.

[0087] Multiple components in electronic device 800 are connected to I / O interface 805, including: input unit 806, such as keyboard, mouse, etc.; output unit 807, such as various types of displays, speakers, etc.; storage unit 808, such as disk, optical disk, etc.; and communication unit 809, such as network card, modem, wireless transceiver, etc. Communication unit 809 allows electronic device 800 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0088] The computing unit 801 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above, such as the cliffside road segment screening method 100. For example, in some embodiments, the cliffside road segment screening method 100 can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of the cliffside road segment screening method 100 described above can be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to perform the cliffside road section screening method 100 by any other suitable means (e.g., by means of firmware).

[0089] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0090] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0091] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0092] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including voice input, speech input, or tactile input).

[0093] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0094] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0095] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0096] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for selecting road sections near cliffs, characterized in that, The method includes: acquiring digital elevation model data and road vector data for multiple regions; performing anisotropic filtering on the digital elevation model data oriented towards protected terrain feature lines; calculating a terrain slope map based on the filtered digital elevation model data, filtering the terrain slope map according to a preset slope threshold, and generating a slope binary map; generating contour line data based on the filtered digital elevation model data at preset intervals and distances, and obtaining a slope candidate area set based on the road vector data; superimposing the slope binary map with the slope candidate area set to obtain a cliff-side road segment candidate area set; obtaining a final cliff-side candidate area set based on the cliff-side road segment candidate area set and the contour line data; and obtaining a complete cliff-side road segment based on the road vector data and the final cliff-side candidate area set.

2. The method according to claim 1, characterized in that, Anisotropic filtering processing oriented towards protecting terrain feature lines is performed on the digital elevation model data, including: extracting terrain feature lines from the digital elevation model data, the terrain feature lines including ridge lines and valley lines; calculating the feature direction of each pixel based on the terrain feature lines; constructing an anisotropic filtering kernel for each pixel based on the feature direction; and using the anisotropic filtering kernel to filter the digital elevation model data.

3. The method according to claim 1, characterized in that, Based on the candidate area set for road sections near the cliff and the contour data, a final candidate area set for road sections near the cliff is obtained, including: overlaying the contour data onto the candidate area set for road sections near the cliff; filtering each candidate area in the candidate area set for road sections near the cliff based on elevation difference conditions to obtain a set of elevation difference filtering for road sections near the cliff; determining the relationship between elevation difference and route direction in the set of elevation difference filtering for road sections near the cliff, and obtaining a set of direction judgments for road sections near the cliff based on the judgment results; and removing foreign objects from the set of direction judgments for road sections near the cliff to obtain the final candidate area set for road sections near the cliff.

4. The method according to claim 3, characterized in that, Each candidate area in the set of candidate areas for cliffside road sections is filtered based on elevation difference conditions to obtain a set of cliffside road section elevation difference filters. This includes: overlaying the contour line data onto the set of candidate areas for cliffside road sections to obtain an overlaid set of candidate areas for cliffside road sections; if the number of contour lines covered in each candidate area in the overlaid set of candidate areas for cliffside road sections is greater than a preset number of contour lines, then candidate areas for cliffside road sections with a greater than preset number of contour lines will be extracted to generate a set of cliffside road section elevation difference filters.

5. The method according to claim 3, characterized in that, The relationship between the elevation difference and the direction of the route is determined in the set of elevation differences of the cliff-side road sections, and the direction judgment set of the cliff-side road sections is obtained based on the judgment results. This includes: superimposing the contour lines on each candidate region in the set of elevation differences of the cliff-side road sections, determining whether the contour lines in the candidate regions are horizontal to the road vector lines; extracting the candidate regions where the contour lines are horizontal to the route, and obtaining the direction judgment set of the cliff-side road sections.

6. The method according to claim 5, characterized in that, Determining whether the contour lines and road vector lines within the candidate area are horizontally related includes: finding the intersections of the road vector lines with contour lines at different elevations within the candidate area; if the number of intersections with each contour line at a different elevation is less than a preset number, then the contour lines are determined to be horizontally related to the road; if the number of intersections with each contour line at a different elevation is greater than the preset number, then the elevation range of the road vector line is determined to be large, and the contour lines are not horizontally related to the road.

7. The method according to claim 3, characterized in that, Foreign object exclusion is performed on the set of direction judgments for road sections near the cliff to obtain a final set of candidate areas for the cliff, including: determining whether there are foreign objects in each candidate area of ​​the set of direction judgments for road sections near the cliff through remote sensing images; extracting candidate areas without foreign objects to obtain the final set of candidate areas for the cliff.

8. The method according to claim 1, characterized in that, Based on the road vector data and the final set of cliff candidate areas, a complete cliff road segment is obtained, including: superimposing the road vector route with the final set of cliff candidate areas, breaking the road segment at the boundary of the candidate area to obtain the vector road segment within the candidate area, which is denoted as the cliff road segment; connecting and merging adjacent cliff road segments to form a complete cliff road segment.

9. The method according to claim 1, characterized in that, The road segment is broken at the boundary of the candidate area to obtain vector road segments within the candidate area, including: sequentially rasterizing each candidate area, and cropping the vectorized candidate area surface and road vector respectively; retaining the original road attribute information for each cropped cliff-side road segment, and sequentially obtaining multiple vector segments; the attribute information of the original road includes the original road route code, route start coordinates, route end coordinates, route start station number, and route end station number.

10. A screening device for road sections near cliffs, characterized in that, include: The acquisition module is used to acquire digital elevation model data and road vector data for multiple regions; The filtering module is used to perform anisotropic filtering on the digital elevation model data for protecting terrain feature lines. The first generation module is used to calculate the terrain slope map based on the filtered digital elevation model data, filter the terrain slope map according to a preset slope threshold, and generate a slope binary map. The second generation module is used to generate contour data based on the filtered digital elevation model data, according to a preset contour line interval and a preset distance, and to obtain a set of slope candidate areas based on the road vector data. The overlay module is used to overlay the slope binary map with the slope candidate area set to obtain the cliff-side road segment candidate area set; the cliff-side road segment generation module is used to obtain the final cliff-side candidate area set based on the cliff-side road segment candidate area set and the contour line data, and to obtain the complete cliff-side road segment based on the road vector data and the final cliff-side candidate area set.

Images