Method for extracting abrupt slope, abrupt cliff and scarp based on elevation angle extreme value

Abstract

The invention discloses a method for extracting a steep slope, a steep cliff and a scarp based on an elevation angle extreme value, which comprises the following steps of: 1, acquiring a digital elevation model (DEM) of a research area, and constructing a multi-direction translation model of the DEM based on matrix convolution operation; 2, calculating a three-dimensional maximum elevation angle (3D maxEA) and a three-dimensional minimum elevation angle (3D minEA) of DEM data through a translation model; 3, determining the critical terrain gradient ST of the abrupt slope, the abrupt cliff and the scarp, comparing the relative sizes of the 3D maxEA, the-3D minEA and the ST, and extracting the raster data of the abrupt slope, the abrupt cliff and the scarp; and 4, extracting vector lines of the 3D maxEA and the-3D minEA which are equal, superposing area vectors of the abrupt slope, the abrupt cliff and the scarp, and extracting vector line data of the abrupt slope, the abrupt cliff and the scarp through a cutting method. According to the abrupt slope, abrupt cliff and scarp extraction model provided by the invention, the landform abrupt change is prevented from being smoothed, and the abrupt slope, the abrupt cliff and the scarp which are easily hidden by slope smoothing are excavated.

Description

Technical Field

[0001] The invention belongs to the field of GIS image processing, and in particular relates to a method for extracting steep slopes, steep cliffs and steep banks based on elevation angle extreme values. Background Art

[0002] In the fields of geographic information science, remote sensing applications, urban planning, land resource management, etc., terrain analysis is a crucial research topic. Terrain not only directly affects the distribution and evolution of natural elements such as hydrology, climate, soil and biodiversity, but is also closely related to the production and life of human society. Among them, steep slopes, cliffs and steep banks are important components of terrain, and their extraction and analysis have extremely high theoretical and practical value. Steep cliffs and steep banks are widely present in the natural environment, and have an important impact on the distribution and evolution of natural elements such as hydrology, climate, soil and biodiversity. Accurate extraction of steep slopes, steep cliffs and steep banks is of great significance for engineering site selection, disaster warning, environmental protection and other aspects. Therefore, studying the extraction methods and techniques of steep slopes, steep cliffs and steep banks has important theoretical and practical value.

[0003] Traditional terrain extraction methods mainly include manual interpretation of topographic maps and field surveys, which are time-consuming, labor-intensive and low-precision. With the development of remote sensing technology and geographic information systems (GIS), modern terrain extraction methods have gradually replaced traditional methods, such as digital elevation models (DEMs) and laser radar (LiDAR) technology. The digital elevation model (DEM) in raster form is the main storage format for international satellite remote sensing digital terrain products. DEM is supported by matrix mathematics theory and has high analysis efficiency, which is suitable for macro-scale topographic and geomorphological research. Previous studies have shown that DEM interpolated from 1:50,000 and 1:100,000 topographic maps has significant terrain description errors, making it difficult to depict sudden terrain features, resulting in the loss of loess cliffs with a slope of more than 70°. The upgrade of drone photogrammetry has improved the expression accuracy of surface objects. The Gaofen-7 satellite developed by my country has enabled regional and even global DEM inversion to break through to a resolution of 1m, and has created a data foundation for the study of regional landform mutations. It is urgent to develop corresponding steep slope, steep cliff, and steep ridge extraction methods to realize regional and even global scale terrain mutation analysis.

[0004] However, the slope analysis method under the GIS platform "flattens" the steep cliff terrain, which, to a certain extent, causes a sharp decrease in the slope calculation results of the steep cliff, resulting in the plane hiding of the steep slope, steep cliff and steep ridge. Therefore, constructing a steep slope, steep cliff and steep ridge extraction method based on the extreme value of elevation angle will not only help to improve the sudden terrain analysis method of the geographic information system (GIS), but also provide technical support for my country's high-resolution satellite to mine the macro pattern of sudden terrain. Summary of the invention

[0005] Purpose of the invention: The purpose of the present invention is to provide a method for extracting steep slopes, steep cliffs and steep steps based on the extreme values ​​of elevation angles, and to effectively reveal sudden changes in terrain by comparing the relative sizes of 3D maxEA, ​​-3D minEA and critical terrain ST.

[0006] Technical solution: A method for extracting steep slopes, steep cliffs and steep steps based on elevation angle extreme values, comprising the following steps:

[0007] Step 1: Obtain the digital elevation model (DEM) of the study area and construct a multi-directional translation model of the DEM based on matrix convolution operation;

[0008] Step 2, calculate the elevation angle EA of the DEM data in each direction through the translation model, take the maximum and minimum values, record them as the maximum elevation angle maxEA and the minimum elevation angle minEA respectively, then compare the maxEA in each direction, take the maximum value as the three-dimensional maximum elevation angle 3D maxEA, ​​compare the minEA in each direction, take the minimum value as the three-dimensional minimum elevation angle 3DminEA;

[0009] Step 3: Set the critical terrain slope ST of the steep slope, steep cliff and steep bank according to the requirements, compare the relative size of 3D maxEA, ​​-3DminEA and ST, extract the raster data of the steep slope, steep cliff and steep bank, and convert them into district vector data;

[0010] Step 4: extract the vector lines where 3D maxEA and -3D minEA are equal, superimpose the area vectors of the steep slope, steep cliff and steep bank, and extract the vector line data of the steep slope, steep cliff and steep bank by clipping method.

[0011] Furthermore, the specific steps of step 1 are:

[0012] (11) Obtain DEM data of the study area from high-precision DEM data measured by UAVs or DEM data collected by high-resolution satellites;

[0013] (12) Construct a multi-directional translation model of DEM data based on matrix convolution.

[0014] Furthermore, the specific steps of step 2 are:

[0015] (21) Based on the DEM translation model, calculate the elevation angle EA of the neighborhood grid in each direction relative to the central grid;

[0016] (22) comparing the magnitude of EA in a specific direction to obtain the maximum elevation angle maxEA and the minimum elevation angle minEA in the specific direction;

[0017] (23) Compare the maxEA in each direction and take the maximum value to obtain the three-dimensional maximum elevation angle 3D maxEA; compare the minEA in each direction and take the minimum value to obtain the three-dimensional minimum elevation angle 3D minEA.

[0018] Furthermore, the specific steps of step 3 are:

[0019] (31) In ArcGIS, the relative sizes of 3D maxEA, ​​-3D minEA, and ST were compared. When 3D maxEA>ST or -3D minEA>ST, the raster data of steep slopes, cliffs, and steep banks were extracted;

[0020] (32) In ArcGIS, the steep slope, cliff, and bluff raster is converted into a regional vector file of the steep slope, cliff, and bluff development area.

[0021] Furthermore, the specific steps of step 4 are:

[0022] (41) Based on ArcGIS, the rasters with 3D maxEA-(-3D minEA)>0 were extracted and converted into zone vectors, and then the zone vector boundaries were converted into line vectors as the vector lines where 3D maxEA and -3D minEA were equal;

[0023] (42) Based on ArcGIS, the vector lines of step 41 and the vector areas of step 32 are superimposed, and the vector lines are clipped with the vector lines to obtain the line vector files of steep slopes, steep cliffs, and steep steps.

[0024] Furthermore, the elevation angle extreme value refers to the three-dimensional maximum elevation angle 3D maxEA and the three-dimensional minimum elevation angle 3D minEA. By comparing the relative sizes of 3D maxEA, ​​-3D minEA and the critical terrain ST, the landform development locations of steep slopes, steep cliffs and steep banks are extracted.

[0025] Furthermore, the neighborhood grid traversal adopts the DEM multi-directional translation model; specifically, the number of directions of the model is 4-360, the segmentation interval is 1-90°, and the farthest distance of the neighborhood grid considered by the model is within 1000 cells.

[0026] Furthermore, the critical terrain slopes ST of the steep slope, steep cliff and steep ridge are 45-65°, 65-75° and 65-75° respectively.

[0027] Furthermore, the relative size of 3D maxEA and -3D minEA is used as the basis for extracting the line vectors of steep slopes, steep cliffs and steep banks, providing an effective technical means for automatically drawing the characteristic lines of steep cliffs and steep banks in topographic maps or geomorphic maps.

[0028] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: by calculating and comparing the elevation angle extreme values ​​of DEM, the present invention extracts the area vector and line vector data of steep slopes, steep cliffs and steep steps, avoiding the terrain slope smoothing effect in the traditional slope threshold method, and exploring small-scale steep slopes, steep cliffs and steep steps. With the growth of DEM data from high-resolution satellites and drone remote sensing, the present invention has a very broad application prospect. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a flow chart of the method for extracting steep slopes, steep cliffs and steep banks based on the extreme value of elevation angle;

[0030] Figure 2 This is a schematic diagram of the digital elevation model data (a) and remote sensing image (b) of the implementation sample area;

[0031] Figure 3 The DEM multi-directional translation model (a), matrix convolution operation (b), and elevation angle calculation schematic diagram (c)-(d);

[0032] Figure 4 -3D minEA and 3D maxEA are calculated from DEM data;

[0033] Figure 5 It is the steep cliff grid effect extracted when the critical terrain ST = 70°;

[0034] Figure 6 It is the vector effect of the cliff line extracted when the critical terrain ST = 70°;

[0035] Figure 7 It is a statistical comparison between the extraction results of the present invention and the slope extraction results;

[0036] Figure 8 It is a schematic diagram of the principle that the present invention is superior to the slope extraction method. DETAILED DESCRIPTION

[0037] The following takes the extraction of steep cliffs in a typical small watershed in the Loess Plateau as an example, and further illustrates the technical solution of the present invention in conjunction with the accompanying drawings.

[0038] The present invention proposes a method for extracting steep slopes, steep cliffs and steep steps based on the extreme value of elevation angle. Figure 1 As shown, the specific steps include:

[0039] Step 1: Obtain the digital elevation model (DEM) of the study area and construct a multi-directional translation model of the DEM based on matrix convolution operation;

[0040] DEM (Digital Elevation Model) is the abbreviation of digital elevation model. It is a digital expression of the terrain surface and contains rich terrain and geomorphic information. The DEM data used in this example is a sample area of ​​the Loess Plateau, located at E110°20′08″-E110°21′31″, N37°34′56″-N37°35′46″. The data is a high-resolution DEM data measured by a drone in the field with a resolution of 1m. The DEM and remote sensing images of the experimental sample area are as follows: Figure 2 Selecting 1m resolution can simulate the DEM data resolution obtained by the Gaofen-7 satellite.

[0041] The steps of DEM multi-directional translation model are as follows:

[0042] 1) Decompose 360° into N terrain profiles at equal intervals. Translate one grid at a time along each terrain profile direction, and traverse the neighborhood grids in that specific direction in turn, such as Figure 3 (a)

[0043] 2) Integrate N terrain profiles, that is, realize multi-directional selective traversal of the neighborhood grid, such as Figure 3 (a)

[0044] 3) Each DEM translation is implemented using matrix convolution operations, such as Figure 3 (b) As shown; and for multi-directional translation, other terrain profiles in directions other than E, S, W, and N can be decomposed into convolution operations in the E, S, W, and N directions.

[0045] Step 2: Calculate the three-dimensional maximum elevation angle 3D maxEA and the three-dimensional minimum elevation angle 3D minEA of the DEM data through the translation model;

[0046] 3D maxEA and 3D minEA are regarded as the extreme values ​​of elevation angle, and the calculation steps are as follows:

[0047] 1) Calculate the elevation angle EA of the neighborhood grid relative to the central grid in a specific direction, such as Figure 3 As shown in (c), the neighborhood grid is higher than the central grid, and the elevation angle calculation result is a positive value; Figure 3 As shown in (d), the neighborhood grid is lower than the central grid, and the elevation angle calculation result is a negative value;

[0048] 2) Compare the EA sizes of the neighboring grids relative to the central grid in all specific directions, take the maximum value to get maxEA, ​​and take the minimum value to get minEA.

[0049] 3) Compare the minEAs obtained in multiple directions, take the minimum value to get the three-dimensional minimum elevation angle 3D minEA, and take the opposite number to get the numerical value (-3D minEA), such as Figure 4 (a); compare the maxEA obtained in multiple directions and take the maximum value to obtain the three-dimensional maximum elevation angle 3D maxEA, ​​as shown in Figure 4 (b)

[0050] Step 3: Determine the critical terrain slope ST of the steep slope, cliff and steep bank, compare the relative size of 3D maxEA, ​​-3D minEA and ST, and extract the raster data of the steep slope, cliff and steep bank, such as Figure 5 As shown;

[0051] The specific steps are:

[0052] 1) In ArcGIS, compare the relative sizes of 3D maxEA, ​​-3D minEA and ST. When 3D maxEA>ST or -3DminEA>ST, extract the steep slope, steep cliff and steep ridge grids;

[0053] 2) In ArcGIS, the steep slope, steep cliff and steep ridge raster is converted into the area vector file ScarpArea of ​​the steep slope, steep cliff and steep ridge development location.

[0054] Step 4: Extract the vector line where 3D maxEA and -3D minEA are equal, superimpose the area vectors of steep slopes, steep cliffs, and steep steps, and extract the vector line data of steep slopes, steep cliffs, and steep steps by clipping method, such as Figure 6 As shown;

[0055] The specific steps are:

[0056] 1) Based on ArcGIS, extract the grids with 3D maxEA-(-3D minEA)>0 and convert them into district vectors. Then convert the district vector boundaries into line vectors as the vector line ScarpCentre where 3D maxEA and -3D minEA are equal;

[0057] 2) Based on ArcGIS, superimpose the vector line ScarpCentre and the vector area ScarpArea, and use the latter to clip the former to obtain the vector line ScarpLine of the steep slope, steep cliff, and steep ridge, such as Figure 6 (a)

[0058] 3) The line vector ScarpLine of the steep slope, cliff and steep ridge can be smoothed to obtain the smoothed line vector, such as Figure 6 (b) as shown.

[0059] Figure 5(b) presents the steep cliff grid extracted by the present invention. The steep cliff grid further compares the relative sizes of 3D maxEA, ​​-3D minEA and ST, and is divided into three parts: the steep cliff face, the steep cliff top and the steep cliff bottom. When 3D maxEA>70° and -3D minEA>70°, the steep cliff face grid is extracted; when -3D minEA>70° and 3D maxEA<70°, the steep cliff top grid is extracted; when -3D minEA<70° and 3D maxEA>70°, the steep cliff bottom grid is extracted. Figure 5 (c) shows the steep cliff raster extracted from the same area with a slope > 70°.

[0060] It can be observed that the scale of the cliffs extracted by the slope is smaller than that of the cliffs extracted by the present invention. Some cliffs cannot be extracted by the slope but can be extracted by the present invention. For this reason, the cliffs extracted by the present invention are compared with the cliffs extracted by the slope>70°. Figure 7 shown.

[0061] Figure 7 (a) and (b) show the grid area statistics of the steep cliffs extracted by the present invention and the slope. It can be seen that the steep cliff areas extracted by the present invention are mostly small, while the steep cliff areas extracted by the slope are relatively large. Figure 7 (c) The area of ​​the steep cliff grid that can be extracted by the present invention but cannot be extracted by the slope is statistically analyzed, and it is found that most of the areas are less than 10m 2 , indicating that small steep cliffs are easily smoothed by slope.

[0062] Figure 7 (d) and (e) present the steep cliff grid height statistics of the present invention and the slope extraction. It can be seen that the steep cliff height extracted by the present invention is relatively small, while the steep cliff height extracted by the slope is relatively large. Figure 7 (f) The heights of the steep cliff grids that can be extracted by the present invention but cannot be extracted by the slope are statistically analyzed, and it is found that most of the heights are less than 2-6m 2 , indicating that low steep cliffs are easily smoothed by slope.

[0063] Figure 7 (g) and (h) present the maximum slope statistics of the steep cliff grids of the present invention and the slope extraction. It can be seen that the steep cliff slopes extracted by the present invention are mostly between 50-80°, while the maximum slopes of the steep cliffs extracted by slope are mostly between 70-80°. Figure 7 (i) The maximum slope of the steep cliff grid that can be extracted by the present invention but not by the slope is counted, and it is found that most of the slopes are between 50-70°, indicating that the steep cliffs extracted by the present invention are mostly smoothed by the slope and are not easy to be extracted by a slope > 70°, which also illustrates the effectiveness of the present method.

[0064] Figure 8The difference between the present invention and the slope extraction of steep slopes, cliffs and steep steps is explained. When the steep slope, cliff and steep step are small in scale, the slope grid is often missing between the A and B grids, and the existence of the steep slope, cliff and steep step can only be determined by the inclination angle between the A and B points. Point A is the top grid of the slope, and the calculated slope is smoothed by the neighborhood and is less than the actual inclination angle between A and B, while 3DminEA can. Similarly, point B is the bottom grid of the slope, and the calculated slope is less than the actual inclination angle between B and A, while 3D maxEA can. Therefore, it is difficult to extract steep slopes, cliffs and steep steps of smaller scale by slope, while 3D minEA and 3D maxEA can extract steep slopes, cliffs and steep steps of smaller scale by measuring the inclination angles of the top and bottom elements of the slope. With the growth of DEM data of high-resolution satellites and UAV remote sensing, the present invention has a very broad application prospect.

Claims

1. A method for extracting steep slopes, steep cliffs and steep steps based on elevation angle extreme values, characterized in that: The steps include: Step 1: Obtain the digital elevation model (DEM) of the study area and construct a multi-directional translation model of the DEM based on matrix convolution operation; Step 2, calculate the elevation angle EA of the DEM data in each direction through the translation model, take the maximum and minimum values, record them as the maximum elevation angle maxEA and the minimum elevation angle minEA respectively, then compare the maxEA in each direction, take the maximum value as the three-dimensional maximum elevation angle 3D maxEA, ​​compare the minEA in each direction, take the minimum value as the three-dimensional minimum elevation angle 3DminEA; Step 3: Set the critical terrain slope ST of the steep slope, steep cliff and steep bank according to the requirements, compare the relative size of 3D maxEA, ​​-3D minEA and ST, extract the raster data of the steep slope, steep cliff and steep bank, and convert them into district vector data; Step 4: extract the vector lines where 3D maxEA and -3D minEA are equal, superimpose the area vectors of the steep slope, steep cliff and steep bank, and extract the vector line data of the steep slope, steep cliff and steep bank by clipping method.

2. The method for extracting steep slopes, steep cliffs and steep steps based on elevation angle extreme values ​​according to claim 1, characterized in that: The specific steps of step 1 are: (11) Obtain DEM data of the study area from high-precision DEM data measured by UAVs or DEM data collected by high-resolution satellites; (12) Construct a multi-directional translation model of DEM data based on matrix convolution.

3. The method for extracting steep slopes, steep cliffs and steep steps based on elevation angle extreme values ​​according to claim 1, characterized in that: The specific steps of step 2 are: (21) Based on the DEM translation model, calculate the elevation angle EA of the neighborhood grid in each direction relative to the central grid; (22) comparing the magnitude of EA in a specific direction to obtain the maximum elevation angle maxEA and the minimum elevation angle minEA in the specific direction; (23) Compare the maxEA in each direction and take the maximum value to obtain the three-dimensional maximum elevation angle 3D maxEA; compare the minEA in each direction and take the minimum value to obtain the three-dimensional minimum elevation angle 3D minEA.

4. The method for extracting steep slopes, steep cliffs and steep steps based on elevation angle extreme values ​​according to claim 1, characterized in that: The specific steps of step 3 are: (31) In ArcGIS, the relative sizes of 3D maxEA, ​​-3D minEA, and ST were compared. When 3D maxEA>ST or -3DminEA>ST, the raster data of steep slopes, cliffs, and steep banks were extracted; (32) In ArcGIS, the steep slope, cliff, and bluff raster is converted into a regional vector file of the steep slope, cliff, and bluff development area.

5. The method for extracting steep slopes, steep cliffs and steep steps based on elevation angle extreme values ​​according to claim 1, characterized in that: The specific steps of step 4 are: (41) Based on ArcGIS, the rasters with 3D maxEA-(-3D minEA)>0 were extracted and converted into district vectors, and then the district vector boundaries were converted into line vectors as the vector lines where 3D maxEA and -3D minEA were equal; (42) Based on ArcGIS, the vector lines of step 41 and the vector areas of step 32 are superimposed, and the vector lines are clipped with the vector lines to obtain the line vector files of steep slopes, steep cliffs, and steep steps.

6. The method for extracting steep slopes, steep cliffs and steep steps based on elevation angle extreme values ​​according to claim 4, characterized in that: The elevation angle extreme values ​​refer to the three-dimensional maximum elevation angle 3D maxEA and the three-dimensional minimum elevation angle 3D minEA. By comparing the relative sizes of 3D maxEA, ​​-3D minEA and the critical terrain ST, the landform development locations of steep slopes, steep cliffs and steep banks are extracted.

7. The method for extracting steep slopes, steep cliffs and steep steps based on elevation angle extreme values ​​according to claim 3, characterized in that: The neighborhood grid traversal adopts the DEM multi-directional translation model; specifically, the number of directions of the model is 4-360, the segmentation interval is 1-90°, and the farthest distance of the neighborhood grid taken into account by the model is within 1000 cells.

8. The method for extracting steep slopes, steep cliffs and steep steps based on elevation angle extreme values ​​according to claim 4, characterized in that: The critical terrain slopes ST of steep slopes, steep cliffs and steep banks are 45-65°, 65-75° and 65-75° respectively.

9. The method for extracting steep slopes, steep cliffs and steep steps based on elevation angle extreme values ​​according to claim 5, characterized in that: The relative size of 3D maxEA and -3D minEA is used as the basis for extracting the line vectors of steep slopes, steep cliffs and steep banks, providing an effective technical means for automatically drawing the characteristic lines of steep cliffs and steep banks in topographic maps or geomorphic maps.

Images