Terrain intrinsic structure-based dem fusion method and apparatus, device, and medium

Abstract

Embodiments of the present application relate to the technical field of hydrogeography, and provide a terrain intrinsic structure-based digital elevation model (DEM) fusion method and apparatus, a device, and a medium. The method comprises: using a hydrological network from original terrain elevation data as an identification object for iterative mainstream identification to obtain a stream set, the stream set comprising a plurality of streams and a topological relationship of the streams; setting stream orders on the basis of the topological relationship; constructing a multi-order watershed set on the basis of the stream orders; on the basis of the ideal number of watersheds, selecting a target watershed set for terrain feature point extraction; and using an extraction result to establish a DEM. The stream set obtained in the embodiments of the present application contains mainstreams at areas of different scales within the hydrological network, so that a living structure of the hydrological network can be effectively extracted, and surface morphological characteristics are accurately extracted on the basis of the living structure of the hydrological network, thereby retaining original data features as much as possible while simplifying DEM terrain.

Description

DEM synthesis method and device based on inherent structure of terrain, equipment and medium TECHNICAL FIELD

[0001] The present application relates to the technical field of hydrological geography, and particularly relates to a DEM synthesis method and device based on inherent structure of terrain, equipment and medium. BACKGROUND

[0002] Digital Elevation Model (DEM), as a digital expression of the earth's surface terrain elevation, is widely used in scientific research and production fields such as surveying and mapping, hydrology, meteorology, and engineering construction. In order to meet the diverse needs of terrain accuracy and data size in different use scenarios, various countries generally adopt the way of coexistence of multiple scales to establish a multi-scale DEM database.

[0003] The traditional multi-scale DEM database is constructed by multiple databases and multiple versions, that is, DEM data of each scale is collected independently to establish a DEM database. With the update and improvement of spatial data acquisition means such as laser radar, the amount of terrain data, the frequency of acquisition, and the complexity have increased significantly, and the traditional multi-database and multi-version database construction method gradually presents the disadvantages of data redundancy, poor consistency, difficulty in updating, and waste of manpower and material resources. To solve these drawbacks, a more ideal way is to establish only a high-precision large-scale database, and use DEM terrain automatic synthesis technology to extract and export other small-scale data from the large-scale data. This way needs to simplify the large-scale data into different small-scale data while retaining the original data characteristics. Therefore, how to simplify the DEM terrain while retaining the original data characteristics as much as possible has become a technical problem to be solved.

[0004] SUMMARY

[0005] The main purpose of the embodiments of the present application is to propose a DEM synthesis method and device based on inherent structure of terrain, equipment and medium, which aims to simplify the DEM terrain while retaining the original data characteristics as much as possible.

[0006] To achieve the above purpose, the first aspect of the embodiments of the present application proposes a DEM synthesis method based on inherent structure of terrain, which comprises:

[0007] obtaining original terrain elevation data;

[0008] identifying a hydrological network in the original terrain elevation data, the hydrological network comprising a plurality of river reaches;

[0009] The hydrological network is taken as an identification object to iteratively perform mainstream identification, each round of mainstream identification obtains at least one identification result, and the identification result includes a river and a plurality of sub-hydrological networks; each of the sub-hydrological networks is taken as an identification object of the next round of mainstream identification until an iteration stop condition is reached; after the iteration is completed, a river set is obtained, and the river set includes a plurality of rivers and a topological relationship of the rivers;

[0010] According to the topological relationship, a river level is set for each river;

[0011] For a river with a river level not less than i (i = 1, 2, …), the catchment area of each river is extracted as an i-level basin, and an i-level basin set is obtained.

[0012] An ideal basin quantity is obtained.

[0013] The n-level basin set closest to the ideal basin quantity is selected as a target basin set.

[0014] Topographic feature points are extracted from the target basin set.

[0015] A digital elevation model is established by using the extracted topographic feature points.

[0016] In some embodiments, the method further comprises: performing data preprocessing on the original topographic elevation data; and the data preprocessing on the original topographic elevation data comprises:

[0017] Depressions in the original topographic elevation data are extracted; the depression depths of the depressions are calculated to obtain a depression depth set; the head-tail segmentation method is used to segment the depression depth set into a head set and a tail set; the depressions corresponding to the tail set are selected as non-naturally formed depressions, and the non-naturally formed depressions are filled.

[0018] In some embodiments, the method further comprises:

[0019] The flow direction of each grid in the original topographic elevation data is calculated; each grid is traversed according to the flow direction, and the cumulative flow of each grid is calculated to obtain a cumulative flow set; the head-tail segmentation method is used to iteratively segment the cumulative flow set into a head set and a tail set until an iteration stop condition is met; a river section threshold is determined according to the average value of the tail set; a vectorization operation is performed on the grid with a cumulative flow value greater than the river section threshold to obtain a directed polyline set, and each directed polyline is a river section.

[0020] In some embodiments, the mainstream identification comprises:

[0021] extracting a longest river path in the recognition object as an initial main stream; removing the initial main stream in the recognition object, and dividing remaining river sections into a plurality of sub-hydrological networks according to connection relationships; extracting a longest river path in each of the sub-hydrological networks as a candidate main stream; calculating an importance degree of each of the candidate main streams relative to the initial main stream according to path lengths and path angles; if a maximum value of the importance degrees is a positive value, outputting the candidate main stream corresponding to the maximum value of the importance degrees as a river; otherwise, outputting the initial main stream as the river.

[0022] In some embodiments, the setting of the river grades according to the topological relationships comprises:

[0023] setting a river connected to other rivers only at an end water outlet as a first-grade river;

[0024] removing a jth (j=1, 2, …) grade river, setting a river connected to other rivers only at an end water outlet in remaining rivers as a (j+1)th grade river, and increasing j by 1 and repeating the step until all rivers are assigned with grades.

[0025] In some embodiments, the method further comprises: in the i-th grade basin set, if a catchment area of a low-grade river overlaps a catchment area of a high-grade river, changing the catchment area of the high-grade river into a non-overlapping area.

[0026] In some embodiments, the acquiring of the ideal number of basins comprises: acquiring a ratio of a target scale to an original scale; and determining the ideal number of basins according to the ratio and a number of first-grade basins.

[0027] In some embodiments, the method further comprises: calculating a river number adjustment amount according to a difference between the ideal number of basins and a number of basins in the target basin set; acquiring a number of target rivers equal to the river number adjustment amount; and increasing or decreasing a catchment area of the target rivers in the target basin set.

[0028] In some embodiments, the topographic feature points comprise: mountain top points, depression points, ridge lines, valley bottom lines, saddle points, ridge line nodes, and valley bottom line nodes; the ridge line node is a grid point of a ridge line of a basin that is adjacent to at least two adjacent basins; and the valley bottom line node is a grid point at an intersection of two valley bottom lines in a basin.

[0029] In some embodiments, the method further comprises: setting grades of the topographic feature points, and screening out topographic feature points with grades lower than a preset grade threshold; and the setting of the grades of the topographic feature points comprises:

[0030] setting grades of the mountain top points, the depression points, the saddle points, the valley bottom line nodes, and the ridge line nodes;

[0031] acquiring a feature point set containing all the topographic feature points; calculating a convex hull of the feature point set; and setting a grade of a convex hull vertex corresponding topographic feature point.

[0032] constructing a triangular mesh using the ranked terrain feature points, the triangular mesh comprising a plurality of triangles;

[0033] calculating a two-dimensional plane projection of each unranked terrain feature point, calculating a two-dimensional plane projection of each triangle, and judging whether the two-dimensional plane projection of each unranked terrain feature point is within the two-dimensional plane projection of a triangle in sequence, and if so, constructing a corresponding relationship between the unranked terrain feature point and the triangle;

[0034] traversing all the triangles, calculating a distance from a current triangle to a corresponding unranked terrain feature point, taking the point with the largest distance as a candidate point, obtaining a candidate point set, screening the candidate point set according to a preset distance threshold, obtaining a set of salient points, setting a rank for each salient point, and updating the triangular mesh using the salient points to obtain the ranks of the remaining unranked terrain feature points.

[0035] To achieve the above object, a second aspect of the embodiment of the present application proposes a DEM synthesizing device based on intrinsic structure of terrain, which comprises:

[0036] a data acquisition module configured to acquire original terrain elevation data;

[0037] a hydrological network identification module configured to identify a hydrological network in the original terrain elevation data, the hydrological network comprising a plurality of river sections;

[0038] a river extraction module configured to iteratively perform main stream identification with the hydrological network as an identification object, obtain at least one identification result in each round of main stream identification, the identification result comprising a river and a plurality of sub-hydrological networks, take each of the sub-hydrological networks as an identification object of the next round of main stream identification until an iteration stop condition is reached, and obtain a river set after the iteration is completed, the river set comprising a plurality of the rivers and topological relationships of the rivers;

[0039] a watershed construction module configured to extract a catchment area of each river with a river rank no less than i (i = 1, 2, …) as an i-level watershed, and obtain an i-level watershed set;

[0040] a model reconstruction module configured to obtain an ideal watershed quantity, select an n-level watershed set with a quantity of watersheds closest to the ideal watershed quantity as a target watershed set, extract terrain feature points from the target watershed set, and establish a digital elevation model using the extracted terrain feature points.

[0041] To achieve the above object, a third aspect of the embodiments of the present application provides an electronic device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the method of the first aspect when executing the computer program.

[0042] To achieve the above object, a fourth aspect of the embodiments of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method of the first aspect.

[0043] The DEM synthesis method and device based on the inherent structure of the terrain, the equipment and the medium provided by the present application, by iteratively dividing the hydrological network into sub-hydrological networks to identify the main stream, extract the main stream of different scale regions in the hydrological network as the river, and the obtained river set reflects the vitality structure of the hydrological network; then according to the topological relationship of the river, the river is graded, and the multi-level river basin set is constructed based on the multi-level river, and the high-level river basin set is nested in the low-level river basin set, which fully considers the interaction between the river and the surface form, extracts the inherent nested substructure in the river basin based on the vitality structure of the hydrological network, and clearly defines the inherent hierarchical relationship between the river basins; then according to the ideal river basin quantity selection, the corresponding hierarchical target river basin set is extracted to extract the terrain feature points and reconstruct the digital elevation model, which can accurately extract the surface form characteristics on the basis of the vitality structure of the hydrological network, and can simplify the DEM terrain while retaining the original data characteristics as much as possible. BRIEF DESCRIPTION OF DRAWINGS

[0044] FIG. 1 is a flowchart of the DEM synthesis method based on the inherent structure of the terrain provided by the embodiments of the present application;

[0045] FIG. 2 is a schematic diagram showing the original terrain elevation data in the embodiments of the present disclosure;

[0046] FIG. 3 is a schematic diagram showing the filling of the original terrain elevation data in FIG. 2 in the embodiments of the present disclosure;

[0047] FIG. 4 is a schematic diagram showing the flow direction of water flow obtained from the terrain elevation data in FIG. 3 in the embodiments of the present disclosure;

[0048] FIG. 5 is a schematic diagram showing the cumulative flow obtained from the flow direction of water flow in FIG. 4 in the embodiments of the present disclosure;

[0049] FIG. 6 is a schematic diagram showing the hydrological network obtained from the cumulative flow in FIG. 5 in the embodiments of the present disclosure;

[0050] FIG. 7 is a schematic diagram showing the initial main stream obtained from the hydrological network in FIG. 6 in the embodiments of the present disclosure;

[0051] FIG. 8 is a schematic diagram illustrating an alternative main stream obtained from the initial main stream of FIG. 7 according to an embodiment of the present disclosure;

[0052] FIG. 9 is a schematic diagram illustrating an angle between the initial main stream of FIG. 7 and the alternative main stream of FIG. 8 according to an embodiment of the present disclosure;

[0053] FIG. 10 is a schematic diagram illustrating a river obtained from the hydrological network of FIG. 6 according to an embodiment of the present disclosure;

[0054] FIG. 11 is a schematic diagram illustrating a set of rivers obtained from the hydrological network of FIG. 6 according to an embodiment of the present disclosure;

[0055] FIG. 12 is a schematic diagram illustrating a first-level river obtained from the set of rivers of FIG. 11 according to an embodiment of the present disclosure;

[0056] FIG. 13 is a schematic diagram illustrating a second-level river obtained from the first-level river of FIG. 12 according to an embodiment of the present disclosure;

[0057] FIG. 14 is a schematic diagram illustrating a third-level river obtained from the second-level river of FIG. 13 according to an embodiment of the present disclosure;

[0058] FIG. 15 is a schematic diagram illustrating a fourth-level river obtained from the third-level river of FIG. 14 according to an embodiment of the present disclosure;

[0059] FIG. 16 is a schematic diagram illustrating a first-level catchment obtained from the river levels of FIG. 15 according to an embodiment of the present disclosure;

[0060] FIG. 17 is a schematic diagram illustrating a second-level catchment obtained from the river levels of FIG. 15 according to an embodiment of the present disclosure;

[0061] FIG. 18 is a schematic diagram illustrating a third-level catchment obtained from the river levels of FIG. 15 according to an embodiment of the present disclosure;

[0062] FIG. 19 is a schematic diagram illustrating a fourth-level catchment obtained from the river levels of FIG. 15 according to an embodiment of the present disclosure;

[0063] FIG. 20 is a schematic diagram illustrating a mountain peak according to an embodiment of the present disclosure;

[0064] FIG. 21 is a schematic diagram illustrating a depression according to an embodiment of the present disclosure;

[0065] FIG. 22 is a schematic diagram illustrating a ridge according to an embodiment of the present disclosure;

[0066] FIG. 23 is a schematic diagram illustrating a valley according to an embodiment of the present disclosure;

[0067] FIG. 24 is a schematic diagram illustrating a saddle according to an embodiment of the present disclosure;

[0068] FIG. 25 is a schematic diagram illustrating a ridge node according to an embodiment of the present disclosure;

[0069] Fig. 26 is a schematic diagram showing a valley bottom line node in an embodiment of the present disclosure;

[0070] Fig. 27 is a schematic diagram showing a 4-level watershed terrain feature in an embodiment of the present disclosure;

[0071] Fig. 28 is a schematic diagram showing three 3-level watershed terrain features in an embodiment of the present disclosure;

[0072] Fig. 29 is a schematic diagram showing a Delaunay triangular mesh in an embodiment of the present disclosure;

[0073] Fig. 30 is a schematic diagram showing significant points obtained by screening the Delaunay triangular mesh of Fig. 29 in an embodiment of the present disclosure;

[0074] Fig. 31 is a schematic diagram showing a DEM synthesis result in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0075] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application is further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0076] It should be noted that although the functional modules are divided in the device schematic diagram, and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a manner different from the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification and claims and the above drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence.

[0077] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0078] First, several terms involved in the present application are analyzed:

[0079] Watershed: a region surrounded by a drainage divide, with runoff from the region concentrated at the lowest point and flowing out. Each watershed is an independent hydrological unit.

[0080] Hydrologic Network: a network system composed of rivers, streams, lakes, reservoirs and other water bodies connected to each other in the process of surface hydrological cycle, which are connected to each other through the natural movement of water flow.

[0081] Mainstream Identification: refers to identifying the main flow path in a river system, i.e., identifying the main river in a drainage basin.

[0082] Stream Order: refers to the numerical level assigned to a river according to factors such as the number of tributaries and the mode of intersection. The stream order can represent certain characteristics of the river; for example, a first-order river is mainly land-based flow and does not have upstream concentrated flow.

[0083] Living Structure: refers to a multi-level, closely linked structural system formed by the internal parts and elements of an object. The living structure reflects the internal organization of the object. A living structure contains different levels of substructures, each of which also contains smaller levels of substructures. The number of substructures and the levels of substructures together determine the degree of vitality of the living structure.

[0084] Existing DEM terrain generalization methods can be divided into two types according to the size of the terrain range considered when simplifying the elements: local neighborhood-based terrain simplification methods and global structure-based terrain simplification methods.

[0085] Local neighborhood-based terrain simplification methods mainly include two categories: global filter-based smoothing methods and local region-based feature selection methods. The former includes downsampling, low-pass filtering, wavelet analysis, etc. This method performs the same smoothing operation on all elements, causing distortion of the terrain skeleton and failing to achieve the effect of extracting the main form and discarding the secondary form. The latter includes hierarchical method, important point method, selective filtering method, etc. This method mainly determines whether to retain the terrain elements by the degree of morphological change between local regions, but cannot identify global features with little change in local terrain.

[0086] The DEM terrain simplification method based on global structure considers the relationship of terrain features from the perspective of the whole terrain, including methods of considering geometric characteristics and methods of considering terrain structure. Among them, the method of considering geometric characteristics includes the three-dimensional Douglas-Peucher algorithm, which determines the selection of terrain points by iteratively judging the distance of three-dimensional terrain points to a specific base surface, thereby extracting important terrain feature points that meet the conditions, and can better maintain the overall contour of the terrain. However, the selection of terrain feature points does not consider the semantic characteristics of the terrain, and cannot guarantee the extraction of geographical elements such as peak points, depression points, ridge lines, and valley lines. In addition, the terrain threshold of this algorithm is affected by many factors such as terrain range, terrain fluctuation, and screening ratio, and needs to be manually set by the cartographer, with low automation degree. The method of considering terrain structure first identifies and extracts the terrain features implied in the DEM data, then evaluates the importance of them to determine the terrain features that need to be retained, and on this basis reconstructs the simplified DEM. For example, the valley filling algorithm proposed by Aitinghua and the DEM generalization method based on catchment area merging proposed by Li Jingzhong. However, the current research rarely considers the multi-level structural characteristics of the terrain structure itself, and lacks systematic consideration and quantitative differentiation of the importance of the terrain structure in the terrain simplification process. In addition, the research on terrain features and their transformation in the corresponding DEM model is also not comprehensive. Based on this, the embodiments of the present application provide a DEM generalization method and device based on the inherent structure of the terrain, equipment and medium, aiming to simplify the DEM terrain while retaining the original data characteristics as much as possible.

[0087] The DEM generalization method and device based on the inherent structure of the terrain provided by the embodiments of the present application are specifically described as follows. First, the DEM generalization method based on the inherent structure of the terrain in the embodiments of the present application is described.

[0088] The DEM generalization method based on the inherent structure of the terrain provided by the embodiments of the present application relates to the field of hydrological geography. The DEM generalization method based on the inherent structure of the terrain provided by the embodiments of the present application can be applied in a terminal, can be applied in a server end, and can also be software running in a terminal or a server end. In some embodiments, the terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc.; the server end can be configured as an independent physical server, can be configured as a server cluster or a distributed system composed of multiple physical servers, can also be configured as a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and basic cloud computing services such as big data and artificial intelligence platforms; and the software can be an application for implementing the DEM generalization method based on the inherent structure of the terrain, etc., but is not limited to the above forms.

[0089] The application is operable in a multitude of general or special computer system environments or configurations. Examples of well known computing systems, environments, and / or configurations that can be suitable for use with the application include, but are not limited to, personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like. The application can be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like, that perform particular tasks or implement particular abstract data types. The application can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules can be located in local and remote computer storage media including memory storage devices.

[0090] FIG. 1 is an optional flowchart of a DEM generalization method based on the inherent structure of terrain according to an embodiment of the application. The method in FIG. 1 can include, but is not limited to, steps S101 to S109.

[0091] Step S101, obtaining original terrain elevation data;

[0092] Step S102, identifying a hydrological network in the original terrain elevation data, the hydrological network including a plurality of river reaches;

[0093] Step S103, iteratively performing main stream identification on the hydrological network as the identification object, each round of main stream identification obtaining at least one identification result, the identification result including a river and a plurality of sub-hydrological networks; taking each of the sub-hydrological networks as the identification object of the next round of main stream identification until an iteration stopping condition is reached; after the iteration is completed, obtaining a river set, the river set including a plurality of the rivers and topological relationships of the rivers;

[0094] Step S104, setting a river level for each river according to the topological relationships;

[0095] Step S105, extracting a catchment area of each river as an i-level watershed, obtaining an i-level watershed set, for a river whose river level is not less than i (i = 1, 2, …);

[0096] Step S106, obtaining an ideal number of watersheds;

[0097] Step S107, selecting an n-level watershed set closest to the ideal number of watersheds as a target watershed set;

[0098] Step S108, extracting terrain feature points from the target watershed set;

[0099] Step S109, a digital elevation model is established by using the extracted terrain feature points.

[0100] The steps S101 to S109 shown in the embodiments of the present application, the main stream is identified by iteratively dividing the hydrological network into sub-hydrological networks, the main stream in different scale regions of the hydrological network is extracted as a river, and the obtained river set reflects the vitality structure of the hydrological network; then, the rivers are classified according to the topological relationship of the rivers, the multi-level river set is constructed based on the multi-level rivers, and the high-level river set is nested in the low-level river set, the interaction between the rivers and the surface morphology is fully considered, the inherent nested substructure in the river basin is extracted based on the vitality structure of the hydrological network, and the inherent hierarchical relationship between the river basins is determined; then, the terrain feature points are extracted from the corresponding hierarchical target river set according to the ideal river basin quantity selection to reconstruct the digital elevation model, the surface morphology characteristics are accurately extracted based on the vitality structure of the hydrological network, and the original data characteristics can be retained as much as possible while the DEM terrain is simplified.

[0101] After step S101 of some embodiments, the DEM synthesis method based on the inherent structure of the terrain further includes: performing data preprocessing on the original terrain elevation data. The data preprocessing can be data cropping or depression filling, and is not limited thereto.

[0102] In some embodiments, the data preprocessing is depression filling, and the data preprocessing on the original terrain elevation data can include but is not limited to steps S201 to S204.

[0103] Step S201, extracting a depression in the original terrain elevation data;

[0104] Step S202, calculating the depression depth of each depression to obtain a depression depth set;

[0105] Step S203, dividing the depression depth set into a head set and a tail set by using a head-tail segmentation method;

[0106] Step S204, selecting the depression corresponding to the tail set as a non-naturally formed depression, and filling the non-naturally formed depression.

[0107] In a large-scale terrain, due to the data precision and the error caused by collection, there can be a large number of depressions with very small depths, i.e., non-natural depressions. The steps S201 to S204 shown in the embodiments of the present application identify the non-natural depressions by using the head-tail segmentation method, and the non-natural depressions are automatically identified based on the inherent structural characteristics of the data instead of relying on the manually set depression depth threshold, thereby avoiding the subjective decision and threshold setting of the human being and improving the automation level of the DEM synthesis method, which can meet the demand of high-frequency update of the map.

[0108] In some embodiments, step S201 can include, but is not limited to, the following steps: calculating the water flow direction of each grid in the DEM by using the D8 algorithm; determining the sink according to the water flow direction, the sink being a grid point whose water flow direction cannot point to other grid points; and taking the sink as the water outlet, and inversely tracing the water flow direction to obtain a set of grid points flowing to the sink, which is the depression corresponding to the sink. The principle of calculating the flow direction by using the D8 algorithm is that the water flow in a grid will flow to the grid with the largest slope in the eight adjacent grids and not higher than the current grid. The slope is determined by the ratio of the height difference and the distance between the grids. In this example, the distance of the four grids in the horizontal direction and the vertical direction is set to 1, and the distance of the diagonal direction is set to

[0109] In step S202 of some embodiments, for each depression, the difference between the minimum elevation of the adjacent grid of the depression and the minimum elevation of the grid in the depression is the depth of the current depression.

[0110] Taking the original terrain elevation data shown in FIG. 2 as an example, the original terrain elevation data is filled with depressions: extracting the sink, such as the two grid points in the 5th row and the 9th column and the 11th row and the 5th column in FIG. 2, whose elevations are lower than those of the eight surrounding grid points, and recording the two grid points as the sink. Obtaining the depressions corresponding to the two sinks, the minimum elevations in the two depressions are 18 and 11 respectively, and the minimum elevations in the adjacent grids are 19 and 12 respectively, so the depths of the two depressions are both 1. The two depressions are non-naturally formed depressions, and the minimum values in the depression grids are replaced by 19 and 12 respectively after filling the depressions, as shown in FIG. 3. After filling the depressions, when the water flows through these depressions, it can continue to flow to the adjacent grid with the minimum elevation, and will not cause the river network to be disconnected unnaturally.

[0111] In some embodiments, step S102 can include, but is not limited to, steps S301 to S305:

[0112] Step S301, calculating the water flow direction of each grid in the original terrain elevation data;

[0113] Step S302, traversing each grid according to the water flow direction, and calculating the cumulative flow of each grid to obtain a cumulative flow set;

[0114] Step S303, using the head-tail segmentation method to iteratively segment the cumulative flow set into a head set and a tail set until the iteration stopping condition is met;

[0115] Step S304, determining the river segment threshold according to the average value of the tail set;

[0116] Step S305, performing a vectorization operation on the grid with a cumulative flow value greater than the river segment threshold to obtain a directed polyline set, each directed polyline being a river segment.

[0117] The steps S301 to S305 shown in the embodiments of the present application determine the river section threshold value by the head-tail segmentation method, automatically identify the river section based on the inherent structural features of the data without relying on the manually set depression depth threshold value, avoid the subjective decision and threshold setting of the human, improve the automation level of the DEM comprehensive method, and meet the demand of high frequency update of the map.

[0118] In some embodiments, the step S302 can include but is not limited to including: traversing all the grids, finding the grid without water flow inflow as the river source point, and accumulating the flow as 0; then tracking the next grid according to the water flow direction, calculating the total number of grids with water flow inflow into the grid as n (n>0), and then calculating the cumulative flow of the grid as the sum of the cumulative flow of each inflow direction plus n; traversing all the grids according to the flow direction until the cumulative flow of all the grids is calculated.

[0119] Taking the topographic elevation data of the completed depression filling shown in FIG. 3 as an example, the hydrological network in the original topographic elevation data is identified: the water flow direction of each grid is calculated by the D8 algorithm, as shown in FIG. 4. The cumulative flow of each grid is calculated, as shown in FIG. 5. According to the average value 5.55 of the cumulative flow, the cumulative flow set is divided into a head set and a tail set, and the average value 0.42 of the tail set is taken as the river section threshold value. The grid with the cumulative flow value greater than the river section threshold value is subjected to a vectorization operation to obtain the hydrological network, as shown in FIG. 6.

[0120] The main stream identification in the step S103 of some embodiments can include but is not limited to including steps S401 to S405:

[0121] Step S401, extracting the longest river path in the identification object as an initial main stream;

[0122] Step S402, removing the initial main stream in the identification object, and dividing the remaining river section into a plurality of sub-hydrological networks according to the connection relationship;

[0123] Step S403, extracting the longest river path in each sub-hydrological network as a candidate main stream;

[0124] Step S404, calculating the importance of each candidate main stream relative to the initial main stream according to the path length and the path angle;

[0125] Step S405, if the maximum value of the importance is positive, outputting the candidate main stream corresponding to the maximum value of the importance as the river; otherwise, outputting the initial main stream as the river.

[0126] The steps S401 to S405 shown in the embodiments of the present application select an initial mainstream in the identification object, select a candidate path in the sub-hydrological network of the identification object, and select the optimal one from the initial mainstream and the candidate mainstream according to the path length and the path angle to output as the mainstream, which considers the influences of the river length and the river path form, and the obtained mainstream runs through the identification object and reflects the relative low point of the surface form of the region where the identification object is located, which is beneficial to reveal the surface form structure characteristics of the region.

[0127] Taking the hydrological network shown in FIG. 6 as an example, the step S103 can include but is not limited to steps S501 to S505.

[0128] In step S501, all river sections without inflow of other water flows are found, and the starting points of these river sections are included in a set S0, such as the black points in FIG. 7; river sections without inflow of other rivers are found, and the ending points thereof are taken as water outlets E, such as the gray points in FIG. 7; the distances of all points in S0 to the water outlet E along the river path are calculated, and the longest river path is taken as the initial mainstream p0, such as the black multi-segment line in FIG. 7.

[0129] In step S502, a copy H0 of the hydrological network H is constructed, and the initial mainstream p0 is removed from H0. The remaining river sections in the copy H0 are divided into a plurality of independent sub-hydrological networks according to whether they are connected, such as the gray points in FIG. 8, which are the water outlets of the sub-hydrological networks, and the black points, which are the starting points of the sub-hydrological networks. The longest river path in each sub-hydrological network is solved as a candidate mainstream p according to the description in step S501, such as the dashed multi-segment line in FIG. 8. The candidate mainstream p is added to a set P1.

[0130] In step S503, for each candidate mainstream p in P1, the river length and the river angle are comprehensively considered, and the Stream Significance Index (SSI) is calculated according to the following formula:

[0131] Wherein, d is the total length of the path of the candidate mainstream p from the starting point to the water outlet; d0 is the total length of the path of the initial mainstream p0 from the starting point to the water outlet; θ is the included angle formed by the candidate mainstream p and the initial mainstream p0 at the intersection point, as shown in FIG. 9. In this example, the SSI value of the mainstream path p0 is taken as 0.

[0132] In step S504, the maximum SSI value SSI max is obtained. max If SSI max > 0, the starting point of the candidate mainstream p corresponding to SSI max is taken as the river starting point, and the river from the river starting point to the water outlet E is output; if SSI max ≤ 0, the initial mainstream p0 is output as the river. In this example, SSImax 0, output the black polyline shown in FIG. 10 as a river.

[0133] Step S505, remove the river from the hydrological network H, and divide the remaining rivers into a plurality of sub-hydrological networks according to whether they are connected. Take each sub-hydrological network as an identification object, and iteratively execute steps S501 to S505 until all river segments are included in different rivers, to obtain a river set as shown in FIG. 11.

[0134] In some embodiments, step S104 can include but is not limited to steps S601 to S602:

[0135] Step S601, set the river connected to other rivers only at the terminal outfall as a first-level river;

[0136] Step S602, remove the j-level river, and set the river connected to other rivers only at the terminal outfall in the remaining river as a j+1-level river (j = 1, 2,...), j is incremented by 1, and the step is repeated until all rivers are assigned a level.

[0137] The steps S601 to S602 shown in the embodiments of the present application divide the river basin based on the catchment area corresponding to the river outfall, and extract the vitality structure of the river basin; the division of the river basin is determined by the method of extracting the inherent hierarchical structure, avoiding artificial subjective decision and threshold setting, improving the automation level, and meeting the demand of high frequency update of the map.

[0138] Taking the river set shown in FIG. 11 as an example, step S104 can include but is not limited to steps S701 to S704:

[0139] Step S701, for the set R0 composed of all rivers, find all rivers connected to other rivers only at the terminal outfall, i.e. rivers with a connection degree of 1, as shown by the black polyline in FIG. 12, add them to the set R1, and set the level of the rivers in R1 to 1;

[0140] Step S702, remove the rivers belonging to R1 from R0 to obtain the set R L_1 , as shown by the gray polyline in FIG. 12, find the rivers R2 connected to other rivers only at the terminal in R L_1 , as shown by the black polyline in FIG. 13, set the level of the rivers in R2 to 2;

[0141] Step S703, remove the rivers belonging to R2 from R L_1 , to obtain the set R L_2 , as shown by the thick gray polyline in FIG. 13. In R L_2R3, which is the thick black polyline in Fig. 14, and set the grade of the rivers in R3 as 3;

[0142] Step S704, remove the rivers belonging to R3 from R L_2 , and obtain the set of the remaining rivers R L_3 . At this time, only one river remains, which is the thick gray polyline in Fig. 14. Add this river to R4, and assign it as a river of grade 4, which is the thickest black polyline in Fig. 15. At this time, all the rivers have been assigned grades, and the calculation is stopped.

[0143] In some embodiments, after step S105, the DEM synthesis method based on the inherent structure of the terrain can further include: in the set of i-level river basins, if the catchment area of a low-grade river overlaps with the catchment area of a high-grade river, change the catchment area of the high-grade river to a non-overlapping area.

[0144] Taking the river grades shown in Fig. 15 as an example, in some embodiments, step S105 can include but is not limited to steps S801 to S804:

[0145] Step S801, obtain all rivers with a grade not less than 1, as shown by the gray polylines in Fig. 16, where the width of the polylines represents different river grades. Add the outlets of all the rivers to set P snap1 , as shown by the dots in Fig. 16. Calculate the catchment areas of all the outlets in set P snap1 . When the catchment area of an outlet of a high-grade river contains the catchment area of an outlet of a low-grade river, only the non-overlapping part of the two catchment areas is taken as the catchment area of the outlet of the high-grade river. The set of these catchment areas is the set of river basins included in grade 1. In Fig. 16, each independent area represents a river basin.

[0146] Step S802, obtain all rivers with a grade not less than 2, as shown by the gray polylines in Fig. 17, where the width of the polylines represents different river grades. Add the outlets of all the rivers to set P snap2 , as shown by the dots in Fig. 17. Calculate the catchment areas of all the outlets in set P snap2 . When the catchment area of an outlet of a high-grade river contains the catchment area of an outlet of a low-grade river, only the non-overlapping part of the two catchment areas is taken as the catchment area of the outlet of the high-grade river. The set of these catchment areas is the set of river basins included in grade 2. In Fig. 17, each independent area represents a river basin.

[0147] Step S803, obtain all rivers with a grade not less than 3, as shown by the gray polylines in Fig. 18, where the width of the polylines represents different river grades. Add the outlets of all the rivers to set P snap3 , as shown by the dots in Fig. 18. Calculate the catchment areas of all the outlets in set Psnap3 When the catchment area of the outfall of the high-level river includes the catchment area of the outfall of the low-level river, only the non-overlapping part of the two catchment areas is taken as the catchment area of the outfall of the high-level river. The set of these catchment areas is the set of the basins included in level 3. In FIG. 18, each independent area represents a basin.

[0148] In step S804, the rivers of level 4 are obtained, as shown by the gray multi-segment lines in FIG. 19. The catchment area of the outfall of the river of level 4 is obtained, i.e., the basin of level 4.

[0149] In some embodiments, step S106 can include, but is not limited to, obtaining the ratio of the target scale to the original scale; and determining the ideal number of basins according to the ratio and the number of basins of level 1.

[0150] In step S106 of some embodiments, the ideal number of basins is determined according to the ratio and the number of basins of level 1, which is expressed by the following formula:

[0151] In the formula, M d is the target scale; M o is the original scale; N o is the number of basins of level 1; and N d is the ideal number of basins.

[0152] The following is an example: If the scale of the original DEM data is M o = 1:1000, and the target scales M d to be obtained are 1:1500, 1:2000, 1:4000, and 1:8000, respectively, and the numbers N c of basins of levels 1, 2, 3, and 4 are 29, 9, 3, and 1, respectively. According to the above formula, when the target scales M d are 1:1500, 1:2000, 1:4000, and 1:8000, respectively, N d are 13, 7, 2, and 1, respectively.

[0153] The ideal number of basins determination method shown in the embodiments of the present application determines the ideal number of basins according to the target scale and the number of basins of level 1, establishes the corresponding relationship between the nesting relationship of the basin structure and the multi-scale expression of the map, and the map of an arbitrary scale can be obtained from high-precision original terrain data.

[0154] After step S107 of some embodiments, the DEM generalization method based on the inherent structure of terrain can further comprise: calculating a river quantity adjustment amount according to a difference between the ideal basin quantity and a basin quantity of the target basin set; obtaining a number of target rivers equal to the river adjustment quantity; and adding or reducing the catchment area of the target rivers in the target basin set.

[0155] In some embodiments, considering that adding or reducing one river leads to adding or reducing two basins, a river quantity adjustment amount calculation formula is constructed as follows:

[0156] ΔN s =(N d -N c ) / 2

[0157] In the formula, ΔN s is the river quantity adjustment amount; N c is the basin quantity of the target basin set; and N d is the ideal basin quantity. The length of a river corresponding to a basin is taken as an index for measuring the importance of the basin. When ΔN s > 0, the catchment area of the first ΔN d longest rivers in the level l-1 is added to the target basin set; and when ΔN s < 0, the catchment area of the first |ΔN| shortest rivers in the target basin set is subtracted.

[0158] An example is shown as follows: when the target scale M d = 1:1500, the ideal basin quantity N d = 13, which is between level 1 (N d = 29) and level 2 (N d = 9) and closer to level 2. Therefore, the target scale M d = 1:1500, the level 2 basin set is taken as the target basin set. Similarly, when the target scale M d is 1:2000, 1:4000 and 1:8000, the corresponding target basin set levels are 2, 3 and 4 respectively. On the basis of the target basin set, the river quantity in the target basin set is adjusted so that the basin quantity N c of the target basin set is consistent with the ideal basin quantity N d , as shown in the following table:

[0159] Taking the target scale M d = 1:1500 as an example, ΔN s = (13-9) / 2 = 2 is obtained, so the longest two rivers in level 1 are taken to obtain their corresponding catchment areas and add them to the target basin set, thereby increasing four basins.

[0160] In step S107 of some embodiments, the terrain feature points include: a peak point, a depression point, a ridge line, a valley line, a saddle point, a ridge line node, and a valley line node.

[0161] The peak point is a grid point with the largest elevation in a drainage basin. As shown in FIG. 20, the dark gray grid point with an elevation value of 3 is the grid point with the largest elevation in the drainage basin, and is the peak point.

[0162] The depression point is a grid point with the smallest elevation in a drainage basin. As shown in FIG. 21, the dark gray grid point with an elevation value of 1 is the grid point with the smallest elevation in the drainage basin, and is the depression point.

[0163] The ridge line is a collection of boundary grids of a drainage basin. As shown in FIG. 22, the left and right columns of grids with an elevation value of 1 belong to two different drainage basins, and the collection of grids with an elevation value of 2 in the middle column is the ridge line.

[0164] The valley line is a collection of grids where a river is located in a drainage basin. As shown in FIG. 23, the collection of grids with an elevation value of 1 is the valley line.

[0165] The saddle point is a grid point where a first-order river corresponds to an intersection point of a valley line and a ridge line. As shown in FIG. 24, the gray grid in the center represents two valley lines starting from the center grid, and the dark grid represents a ridge line passing through the center grid, so the center grid is the intersection point of the valley line and the ridge line, and is the saddle point.

[0166] The ridge line node is a grid point of a ridge line of a target drainage basin, which is adjacent to at least two adjacent drainage basins. As shown in FIG. 25, the collection of dark grids is the ridge line, and the gray grids on both sides and above the ridge line belong to three different drainage basins, so the center grid is the ridge line node.

[0167] The valley line node is a grid point of an intersection point of two valley lines in a target drainage basin. As shown in FIG. 26, the collection of gray grids is the valley line, and the two valley lines intersect at the center grid, so the center grid is the valley line node.

[0168] The ridge line node and the valley line node shown in the embodiments of the present application reflect the terrain features at the intersection of the ridge line and the intersection of the valley line, and are beneficial to extracting more accurate terrain structures. The valley line node is a point where two rivers converge, and the flow of the river will increase sharply at this position. Accurate extraction of this feature point is of great significance for hydrological analysis and disaster prevention. The ridge line node is an intersection point of two mountain ridge lines, which is a characteristic point where the mountain terrain changes abruptly, and extraction of this feature is beneficial to fixing the shape of the mountain.

[0169] It needs to be particularly pointed out that, unlike the mountain top point and the depression point in the traditional DEM synthesis method which are extreme points in the digital elevation model, the mountain top point and the depression point extracted by the present application are local features. In addition, for the basins which are not the source of the river (i.e. do not contain the river of level 1), the starting point of the valley bottom line in the basin is the feature point which is classified as the valley bottom line node by the embodiments of the present application.

[0170] The terrain feature points are illustrated by taking FIG. 27 and FIG. 28 as examples: FIG. 27 shows the terrain features in the basin of level 4, including the mountain top point, the depression point, the saddle point, the ridge line, and the valley bottom line 5 types of terrain features. FIG. 28 shows the terrain features in three basins of level 3, including the mountain top point, the depression point, the ridge line, the valley bottom line, and the ridge line intersection point. In addition, there is a saddle point at the source of the river in FIG. 28, and there is a valley bottom line intersection point in the downstream basin when two rivers intersect into one river.

[0171] After step S108 of some embodiments, the DEM synthesis method based on the internal structure of the terrain can further include: setting the levels of the terrain feature points, and screening out the terrain feature points with a level lower than a preset level threshold; the setting of the levels of the terrain feature points can include but is not limited to steps S901 to S905.

[0172] Step S901: setting the levels of the mountain top point, the depression point, the saddle point, the valley bottom line node, and the ridge line node;

[0173] Step S902: obtaining a feature point set containing all the terrain feature points; calculating the convex hull of the feature point set, and setting the level of the terrain feature point corresponding to the vertex of the convex hull;

[0174] Step S903: constructing a triangular mesh using the terrain feature points with set levels, and the triangular mesh includes a plurality of triangles;

[0175] Step S904: calculating the two-dimensional plane projection of each terrain feature point without set level; calculating the two-dimensional plane projection of each triangle; sequentially judging whether the two-dimensional plane projection of each terrain feature point without set level is within the two-dimensional plane projection of the triangle, and if so, constructing the corresponding relationship between the terrain feature point without set level and the triangle;

[0176] Step S905: traversing all the triangles, calculating the distance from the current triangle to the corresponding terrain feature point without set level, taking the point with the maximum distance as a candidate point, and obtaining a candidate point set; screening the candidate point set according to a preset distance threshold to obtain a significant point set; setting the level of each significant point; and updating the triangular mesh using the significant points to obtain the levels of the remaining terrain feature points without set level.

[0177] Taking the terrain feature points shown in Figure 27 as an example, in some embodiments, the setting of the levels of each terrain feature point may include, but is not limited to, steps S1001 to S1006.

[0178] Step S1001: Classify the terrain feature points in Figure 27 by assigning levels 1 to 5 to the mountain peaks, depressions, saddles, valley line nodes, and ridge line nodes respectively. Establish set U, and add the grid points contained in the valley lines and ridge lines as unclassified feature points to set U.

[0179] Step S1002: Find the convex hull of the point set composed of all terrain feature points, and assign level 6 to the vertices of the convex hull.

[0180] Step S1003: Set the current level L = 7. Construct a Delaunay triangular mesh T using the terrain feature points of the set level (black dots as shown in Figure 29). L (As shown by the dashed line segment in Figure 29). Based on whether the two-dimensional planar projection of the feature points in set U lies within the Delaunay triangular mesh T... L Within the two-dimensional plane projection of the triangle, establish the correspondence between terrain feature points and the triangle.

[0181] Step S1004, traverse the Delaunay triangular mesh T L For all triangles in the array, perform the following operations: For the terrain feature points corresponding to triangle i (the dots shown in Figure 29), calculate the distances from these points to the two-dimensional plane containing triangle i, and denote the terrain feature point with the largest distance as... Maximum distance is The terrain feature points of triangle i are sequentially... and maximum distance value The points are placed into candidate point sets P and D respectively. In this example, the distances from the terrain feature points in set P to the faces of the corresponding triangles are d and d, respectively. A =0.352; d B =0.640; d C =13.9; d D =0.2; d E =13.5; d F =12.8; d G =0.2; d H =12.7; d I =11.0; d J =4.8; d K =1.4.

[0182] Step S1005: Find the maximum value d in set D. max In this example, d max =dC = 13.9. To ensure that the saliencies of the same grade terrain feature points are generally similar, take half of d max If the following condition is satisfied , then the distance corresponding to the terrain feature point is added to the salient point set P F , and the point is deleted from the set U, and a new vertex is taken as the terrain feature point, and three new triangles are formed with the original three edges of the triangle, and the Delaunay triangulation is updated. If the following condition is not satisfied , then the point is still left in the set U. After all values of are traversed, jump to the next step. In this example, the points in the set P F are C, E, F, H, and I shown in FIG. 30, which are the terrain feature points of the current grade L.

[0183] Step S1006, grade L is added by 1, and the calculation steps of steps S1004-S1005 are repeatedly executed until all terrain feature points are determined grades, i.e., the set U is empty, and the calculation is ended.

[0184] In some embodiments, step S109 can include but is not limited to the following: the feature extraction result (such as the sparse matrix feature point) obtained in step S108 is interpolated by using the Kriging method to obtain a digital elevation model.

[0185] Taking a certain basin in Yunnan as an example, the DEM synthesis method based on the inherent structure of the terrain is executed, and a plurality of digital elevation models of different scales obtained are shown in FIG. 31. It can be seen that in the synthesis results of different levels, different levels of terrain are retained, and the synthesis results of each level reflect the inherent hierarchical structure of the terrain.

[0186] The embodiments of the present application also provide a DEM synthesis device based on the inherent structure of the terrain, which can implement the above-mentioned DEM synthesis method based on the inherent structure of the terrain. The device comprises:

[0187] a data acquisition module configured to acquire original terrain elevation data;

[0188] a hydrological network identification module configured to identify a hydrological network in the original terrain elevation data, the hydrological network comprising a plurality of river sections;

[0189] ​The river extraction module is configured to iteratively perform main stream identification on the hydrological network as an identification object, and each round of main stream identification obtains at least one identification result, which includes a river and a plurality of sub-hydrological networks; each of the sub-hydrological networks is taken as an identification object of the next round of main stream identification until an iteration stop condition is reached; and after the iteration is completed, a river set is obtained, which includes a plurality of the rivers and topological relationships of the rivers.

[0190] The basin construction module is configured to extract a catchment area of each river as an i-level basin, and obtain an i-level basin set, for a river with a river level no less than i (i = 1, 2, …).

[0191] The model reconstruction module is configured to obtain an ideal number of basins; select an n-level basin set closest to the ideal number of basins as a target basin set; perform terrain feature point extraction on the target basin set; and establish a digital elevation model by using the extracted terrain feature points.

[0192] The specific implementation of the device is basically the same as that of the above-mentioned specific embodiment of the DEM synthesis method based on the internal structure of the terrain, and will not be described here.

[0193] The embodiments of the present application also provide an electronic device, which includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned DEM synthesis method based on the internal structure of the terrain when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a vehicle-mounted computer, etc.

[0194] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the above-mentioned DEM synthesis method based on the internal structure of the terrain.

[0195] The memory is a non-transitory computer readable storage medium, which can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0196] The DEM synthesis method and device based on the internal structure of the terrain, the electronic device and the storage medium provided by the embodiments of the present application can identify the main stream by iteratively dividing the hydrological network into sub-hydrological networks, extract the main stream of different scale regions in the hydrological network as a river, and reflect the vitality structure of the hydrological network by the obtained river set. Then, the rivers are classified according to the topological relationship of the rivers, a multi-level river basin set is constructed based on the multi-level rivers, and the high-level river basin set is nested in the low-level river basin set. The interaction between the river and the surface morphology is fully considered, the inherent nested substructure in the river basin is extracted based on the vitality structure of the hydrological network, the inherent hierarchical relationship between the river basins is clarified, the terrain feature points are extracted from the corresponding hierarchical target river basin set according to the ideal river basin quantity selection, and the digital elevation model is reconstructed. The surface morphology characteristics are accurately extracted based on the vitality structure of the hydrological network, the DEM terrain can be simplified, and the original data characteristics can be retained as much as possible.

[0197] The embodiments described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0198] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and can include more or fewer steps than those shown in the figures, or combine certain steps or different steps.

Claims

1. A DEM generalization method based on the intrinsic structure of terrain, characterized in that, The method comprises: obtaining original terrain elevation data; identifying a hydrological network in the original terrain elevation data, the hydrological network comprising a plurality of river sections; iteratively performing main stream identification on the hydrological network as an identification object, each round of main stream identification obtaining at least one identification result, the identification result comprising a river and a plurality of sub-hydrological networks; taking each of the sub-hydrological networks as an identification object of the next round of main stream identification until an iteration stop condition is reached; after the iteration is completed, obtaining a river set, the river set comprising a plurality of the rivers and topological relationships of the rivers; setting a river level for each of the rivers according to the topological relationships; extracting a catchment area of each of the rivers with a river level no less than i (i = 1, 2, …) as an i-level watershed to obtain an i-level watershed set; obtaining an ideal number of watersheds; selecting an n-level watershed set closest to the ideal number of watersheds as a target watershed set; extracting terrain feature points from the target watershed set; establishing a digital elevation model by using the extracted terrain feature points.

2. The method of claim 1, wherein, Further comprising data preprocessing on the original terrain elevation data; the data preprocessing on the original terrain elevation data comprises: extracting depressions in the original terrain elevation data; calculating depression depths of the depressions to obtain a depression depth set; dividing the depression depth set into a head set and a tail set by using a head-tail segmentation method; selecting depressions corresponding to the tail set as non-naturally formed depressions, and filling the non-naturally formed depressions.

3. The method of claim 1, wherein, The identification of the hydrological network in the original terrain elevation data comprises: calculating water flow directions of each grid in the original terrain elevation data; traversing each grid according to the water flow directions, calculating cumulative flow values of each grid to obtain a cumulative flow set; iteratively dividing the cumulative flow set into a head set and a tail set by using a head-tail segmentation method until an iteration stop condition is met; determining a river section threshold according to an average value of the tail set; performing a vectorization operation on grids with cumulative flow values greater than the river section threshold to obtain a directed polyline set, each directed polyline being a river section.

4. The method of claim 1, wherein, The main stream identification comprises: extracting a longest river path in the identification object as an initial main stream; removing the initial main stream from the identification object, dividing remaining river sections into a plurality of sub-hydrological networks according to connection relationships; extracting a longest river path in each of the sub-hydrological networks as a candidate main stream; calculating an importance of each of the candidate main streams relative to the initial main stream according to path lengths and path angles; if a maximum value of the importance is positive, outputting the candidate main stream corresponding to the maximum value of the importance as the river; otherwise, outputting the initial main stream as the river.

5. The method of claim 1, wherein, The setting of the river level for each of the rivers according to the topological relationships comprises: setting a river with only an end water outlet connected to other rivers as a 1-level river; removing j (j = 1, 2, …) level rivers, setting a river with only an end water outlet connected to other rivers among remaining rivers as a j+1 level river, and increasing j by 1, and repeating the step until all rivers are assigned a level.

6. The method of claim 1, wherein, Further comprising: in the i-level watershed set, if a catchment area of a low-level river overlaps with a catchment area of a high-level river, changing the catchment area of the high-level river to a non-overlapping area.

7. The method of claim 1, wherein, The acquiring the ideal number of watersheds comprises: acquiring a ratio of a target scale to an original scale; and determining the ideal number of watersheds according to the ratio and a first-level number of watersheds.

8. The method of claim 1, wherein, Further comprising: calculating a river number adjustment amount according to a difference between the ideal number of watersheds and a number of watersheds in a target watershed set; acquiring a number of target rivers equal to the river number adjustment amount; and adding or reducing catchment areas of the target rivers in the target watershed set.

9. The method of claim 1, wherein, The terrain feature points comprise: mountain top points, depression points, ridge lines, valley bottom lines, saddle points, ridge line nodes and valley bottom line nodes; the ridge line node is a grid point of a ridge line of a watershed that is adjacent to at least two adjacent watersheds; and the valley bottom line node is a grid point at an intersection of two valley bottom lines in the watershed.

10. The method of claim 9, wherein, Further comprising: setting levels of the terrain feature points, and screening out terrain feature points with levels lower than a preset level threshold; the setting of the levels of the terrain feature points comprises: setting levels of the mountain top points, the depression points, the saddle points, the valley bottom line nodes and the ridge line nodes; acquiring a feature point set containing all the terrain feature points; calculating a convex hull of the feature point set; and setting levels of terrain feature points corresponding to the convex hull vertices; constructing a triangular mesh using the terrain feature points with the set levels, the triangular mesh comprising a plurality of triangles; calculating two-dimensional plane projections of the terrain feature points without the set levels; calculating two-dimensional plane projections of the triangles; and sequentially judging whether the two-dimensional plane projections of the terrain feature points without the set levels are within the two-dimensional plane projections of the triangles, and if so, constructing a corresponding relationship between the terrain feature points without the set levels and the triangles; traversing all the triangles, calculating distances from a current triangle to corresponding terrain feature points without the set levels, taking a terrain feature point with the largest distance as a candidate point, and obtaining a candidate point set; and according to a preset distance threshold, screening the candidate point set to obtain a set of salient points; setting levels of the salient points; and using the set of salient points to update the triangular mesh to obtain levels of remaining terrain feature points without the set levels.

11. A DEM synthesizing apparatus based on inherent structure of terrain, characterized by, The device comprises: a data acquisition module configured to acquire original terrain elevation data; a hydrological network identification module configured to identify a hydrological network in the original terrain elevation data, the hydrological network comprising a plurality of river segments; a river extraction module configured to iteratively perform main stream identification using the hydrological network as an identification object, each round of main stream identification obtaining at least one identification result, the identification result comprising a river and a plurality of sub-hydrological networks; using each of the sub-hydrological networks as an identification object for the next round of main stream identification until an iteration stop condition is reached; and after the iteration is completed, obtaining a river set comprising a plurality of the rivers and topological relationships of the rivers; a watershed construction module configured to extract a catchment area of each river with a river level not less than i (i = 1, 2, …) as an i-level watershed, and obtain an i-level watershed set; a model reconstruction module configured to acquire an ideal number of watersheds; select an n-level watershed set with a number of watersheds closest to the ideal number of watersheds as a target watershed set; perform terrain feature point extraction on the target watershed set; and use the extracted terrain feature points to establish a digital elevation model.

12. An electronic device, comprising: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method in any one of claims 1 to 10 when executing the computer program.

13. A computer-readable storage medium, the computer-readable storage medium storing a computer program, characterized in that, The computer program is executed by the processor to implement the method in any one of claims 1 to 10.

Images