River channel three-dimensional reconstruction method and system for riverbed reconstruction

Abstract

The invention discloses a riverway three-dimensional reconstruction method and system for riverbed reconstruction, and relates to the technical field of model construction, and the method comprises the steps: dividing geographic features into slowly-changing geographic features and dramatically-changing geographic features, carrying out the clustering twice through the slowly-changing geographic features and the dramatically-changing geographic features, improving the reliability of the clustering analysis, and improving the reconstruction efficiency of the riverbed. And a reliable basis is provided for subsequent model establishment and reconstruction. A local initial region of a river topographic map is calibrated through river geographical evolution indexes, the local initial region is adjusted by analyzing the boundary effect of the initial region, and respective global and local description scales are determined according to the river geographical evolution indexes and geographical feature information in the global region and the local region. According to the method, multiple types of control points are selected in the global region and the local region, and the global model and the local model of the river channel are superposed by virtue of the multiple types of control points, so that the reasonability and the adaptability of three-dimensional modeling of the river channel are improved, and the plasticity and the description precision of river bed reconstruction are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of model construction, and in particular to a three-dimensional river reconstruction method and system for riverbed reconstruction. Background Art

[0002] As water conservancy projects increasingly demand ecological and refined design, traditional two-dimensional design methods are gradually becoming obsolete due to their inability to accurately depict complex river morphology. 3D reconstruction technology, through methods such as drone oblique photography and lidar scanning, can acquire high-precision terrain data. Combined with BIM or real-world 3D modeling software, digital models can be constructed that incorporate terrain, water bodies, structures, and other elements. For example, a river project in Shandong used a DJI M300 RTK drone to capture 0.05-meter resolution imagery. This was then combined with GodWork3D software to generate a real-world 3D model, achieving accuracy control of better than 0.15 meters in planar and 0.3 meters in elevation. This technology not only intuitively displays the spatial form of the river but also enables real-time linkage between the model and design plan through a dynamic linkage mechanism. This provides data support for terrain analysis, flow simulation, and ecological restoration design in riverbed reconstruction, significantly improving the scientific nature of engineering design and construction efficiency.

[0003] In the existing technology, the modeling of river channels uses the same description accuracy for different location areas, and cannot perform high-precision adaptive reconstruction of different local areas, resulting in poor rationality and adaptability of river channel three-dimensional reconstruction.

[0004] Therefore, how to improve the rationality and adaptability of three-dimensional reconstruction of river channels is a technical problem that needs to be solved. Summary of the Invention

[0005] The purpose of the present invention is to solve the problem of poor rationality and adaptability of river channel three-dimensional reconstruction in the prior art, and to propose a river channel three-dimensional reconstruction method for riverbed reconstruction, which includes: Obtain the geographical feature information of the river channel, establish a river channel topographic map, cluster the geographical features on the river channel topographic map, and define the river channel geographical evolution indicators; The local initial area of ​​the river topography map is demarcated by using the river geographical evolution index, and the boundary effect of the initial area is analyzed to adjust the local initial area and determine the local area; Determine the global and local description scales based on the river channel geographic evolution indicators and geographic feature information in the global and local regions, and establish the global and local models of the river channel respectively based on the description scales; Multiple types of control points are selected in the global area and the local area, and the global model and the local model of the river are superimposed with the multiple types of control points to generate an overall three-dimensional model of the river.

[0006] In some embodiments of the present application, clustering of geographical features is performed on a river topographic map, including: Extracting multiple geographical features from the geographical feature information of the river channel, and distinguishing between slowly changing geographical features and drastically changing geographical features according to the changes in the historical records of the geographical features; Determine the search space range of the first cluster, perform first-stage clustering under the spatial constraints of the slowly varying geographic features based on the search space range of the first cluster and the slowly varying geographic features, and generate multiple first cluster areas; The similarity between different first cluster regions is defined according to the drastic geographical features, and the second stage clustering is performed to generate multiple second cluster regions.

[0007] In some embodiments of the present application, determining the search space range of the first cluster includes: Establish the river network skeleton of the river channel and calculate the distance from each grid to the center line. Determine the average width of the river channel based on the distance from the grid to the center line. Set the basic radius according to the average width of the river channel and adjust the basic radius according to the terrain slope and rock anti-scour coefficient. Use the adjusted basic radius as the search space range of the first cluster.

[0008] In some embodiments of the present application, the similarity between different first cluster regions is defined based on the drastic geographical features, including: Calculate the range and change rate of the sudden change geographical feature at different locations in each first cluster area. Use this to calculate the similarity of the range and change rate of the same sudden change geographical feature in different first cluster areas. Combine the similarities of the range and change rate to define the similarity between different first cluster areas, which is recorded as the comprehensive similarity. The calculation formula for comprehensive similarity is as follows: ; in, For the The first cluster area and the The comprehensive similarity between the first cluster regions, For the The first cluster area and the The number of common drastic geographical features between the first cluster regions, For the Similar weights for the drastic geographical features, For the The first cluster area and the The first cluster of regions similarity in the extent of the drastic geographical features, For the The first cluster area and the The first cluster of regions The similarity of the rate of change of the drastic geographical features, For the The first constant of a drastically changing geographical feature.

[0009] In some embodiments of the present application, river channel geographical evolution indicators are defined, including: The change values ​​of the drastic geographical features under each second cluster area are counted, and the change values ​​of multiple drastic geographical features are standardized. The evolution weight corresponding to each drastic geographical feature is determined by the entropy weight method; The river channel geography evolution index under each second cluster area is defined by the evolution weight and the change value of the drastic geographical characteristics.

[0010] In some embodiments of the present application, the local initial area of ​​the river topographic map is demarcated by using river geographical evolution indicators, including: Each second cluster area is divided on the river topographic map, and the river geographical evolution index under each second cluster area is marked, and the difference in river geographical evolution index between adjacent second cluster areas is calculated; If the difference in river channel geographical evolution indicators between adjacent second cluster areas is less than a preset threshold, multiple adjacent second cluster areas are merged to obtain a new second cluster area; The old second cluster region and the new second cluster region are used as local initial regions.

[0011] In some embodiments of the present application, the boundary effect of the initial region is analyzed to adjust the local initial region, including: Identify all boundary effect types involved in the river channel, match the boundary effect type corresponding to each initial area, establish a boundary effect model for each initial area, determine the degree of boundary effect of each initial area, and expand the buffer area of ​​the boundary of each initial area to adjust the local initial area.

[0012] In some embodiments of the present application, the global and local description scales are determined based on the river geographical evolution indicators and geographical feature information in the global and local regions, including: For the global description scale, multiple macro-feature change indicators are screened from the geographic feature information, and the macro-complexity is calculated by integrating the multiple macro-feature change indicators. The river geographical evolution indicators of multiple second cluster areas under the global area are integrated to calculate the evolution complexity. The description scale of the global area is determined by combining the macro-complexity and evolution complexity. For the local description scale, multiple terrain features are screened out from the geographic feature information, and the terrain complexity of each second cluster area is calculated based on the terrain features. The evolution complexity of each second cluster area is calculated based on the river geographical evolution index of the second cluster area. The description scale of each second cluster area is determined based on the terrain complexity and evolution complexity of the second cluster area.

[0013] In some embodiments of the present application, multiple types of control points are selected in the global area and the local area, including: According to their functions, control points are divided into geometric control points, dynamic control points, evolutionary control points and boundary control points. According to the respective conditions of geometric control points, dynamic control points, evolution control points and boundary control points, they are jointly screened in the global and local areas to determine each type of control points, namely geometric control points, dynamic control points, evolution control points and boundary control points, and a layered overlay strategy of geometric alignment, dynamic coupling, evolution verification and boundary optimization is established, corresponding to the geometric control points, dynamic control points, evolution control points and boundary control points respectively.

[0014] Correspondingly, the present application also provides a river channel three-dimensional reconstruction system for riverbed reconstruction, comprising: The first module is used to obtain the geographical feature information of the river channel, establish a river channel topographic map, cluster the geographical features on the river channel topographic map, and define the river channel geographical evolution indicators; The second module is used to calibrate the local initial area of ​​the river topography map through the river geographical evolution index, analyze the boundary effect of the initial area to adjust the local initial area, and determine the local area; The third module is used to determine the global and local description scales based on the river geographical evolution indicators and geographical feature information in the global and local regions, and to establish the global model and local model of the river respectively based on the description scales; The fourth module is used to select multiple types of control points in the global area and the local area, and use the multiple types of control points to superimpose the global model and the local model of the river to generate an overall three-dimensional model of the river.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. Cluster geographical features on the river topography map, dividing them into slowly changing geographical features and drastically changing geographical features. Perform two clustering operations based on these two types of features to improve the reliability of cluster analysis and provide a reliable foundation for the establishment and reconstruction of subsequent models. Demarcate the local initial area of ​​the river topography map using river geographical evolution indicators, analyze the boundary effects of the initial area to adjust the local initial area, and consider the river geographical evolution of the clustered area to initially demarcate the local area, i.e., some local areas with large changes and complex situations. Consider the boundary effects of the local initial area to appropriately adjust the regional boundaries, providing an accurate foundation for subsequent local modeling and ensuring the compatibility of the local area with the local model.

[0016] 2. Determine the global and local description scales based on the river's geographic evolution indicators and geographic characteristics within the global and local regions. Considering the model's description scale from both global and local perspectives allows for adapting to the 3D reconstruction accuracy requirements of different river regions, taking into account both the overall and partial conditions. Multiple types of control points are selected within the global and local regions. Using these control points, the global and local models of the river are superimposed, improving the rationality and adaptability of 3D river modeling and ensuring the flexibility and descriptive accuracy of riverbed reconstruction. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic flow chart of a three-dimensional river channel reconstruction method for riverbed reconstruction proposed by the present invention; Figure 2 This is a structural schematic diagram of a river channel three-dimensional reconstruction system for riverbed reconstruction proposed by the present invention. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0019] Reference Figure 1 , a three-dimensional reconstruction method of a river channel for riverbed reconstruction includes the following steps.

[0020] Step S101 , obtaining geographical feature information of a river channel, establishing a river channel topographic map, clustering geographical features on the river channel topographic map, and defining river channel geographical evolution indicators.

[0021] In this embodiment, data acquisition involves using drone oblique photography (such as the DJI Phantom 4 RTK), airborne laser radar (LiDAR), or a terrestrial 3D scanner to obtain high-precision point cloud data (resolution ≤ 5 cm) and orthophotos (GSD ≤ 2 cm) of the river. Topographic map generation involves processing the point cloud and imagery using ContextCapture or Pix4D software to generate a DEM (Digital Elevation Model) and DSM (Digital Surface Model). Geographic features such as the riverbed, shoreline, and vegetation are then overlaid to construct a 3D topographic map of the river. Geographic feature information for the river refers to all relevant geographic features of the river.

[0022] The geographical features extracted from the river topographic map must cover three types of information: morphology, hydrology, and geology (mainly these three aspects): Morphological characteristics: riverbed elevation, slope, curvature, cross-sectional shape (e.g., U-shaped / V-shaped), and beach width; Hydrological characteristics: water velocity, water depth, and sediment content (inverted through hydrological models); Geological characteristics: riverbed substrate type (sand, gravel, clay), erosion resistance (associated with geological exploration data).

[0023] In some embodiments of the present application, clustering of geographical features is performed on a river topographic map, including: Extracting multiple geographical features from the geographical feature information of the river channel, and distinguishing between slowly changing geographical features and drastically changing geographical features according to the changes in the historical records of the geographical features; Determine the search space range of the first cluster, perform first-stage clustering under the spatial constraints of the slowly varying geographic features based on the search space range of the first cluster and the slowly varying geographic features, and generate multiple first cluster areas; The similarity between different first cluster regions is defined according to the drastic geographical features, and the second stage clustering is performed to generate multiple second cluster regions.

[0024] In this example, slowly changing geographic features are those that change relatively slowly (e.g., elevation changes or slope changes), while rapidly changing geographic features are those that change frequently and rapidly (e.g., hydrodynamic conditions or sediment transport). Clustering is divided into two stages: a first clustering based on slowly changing geographic features, followed by a second clustering based on these results using rapidly changing geographic features. This two-stage clustering model achieves classification based on changes in geographic feature type.

[0025] The first clustering was performed using a dynamic radius adjustment based on the river network topology. This radius was the search space for the first cluster, adapting the search range to the scale of the river channel morphology. The second clustering was performed based on the similarity of the dramatic changes in geographical features between the first cluster areas, ultimately determining multiple second cluster areas.

[0026] In some embodiments of the present application, determining the search space range of the first cluster includes: Establish the river network skeleton of the river channel and calculate the distance from each grid to the center line. Determine the average width of the river channel based on the distance from the grid to the center line. Set the basic radius according to the average width of the river channel and adjust the basic radius according to the terrain slope and rock anti-scour coefficient. Use the adjusted basic radius as the search space range of the first cluster.

[0027] In this embodiment, HydroSHEDS data or ArcGIS Hydrology toolset is used to generate the river network skeleton, and the distance from each grid to the center line is calculated. The center line of the river network is used to generate a river buffer zone (Buffer Zone), and the average width of the river is inferred by counting the number or area of ​​grids within the buffer zone. For example, for each line segment of the center line, a buffer zone with a width of W is generated, the number of grids N(W) covered by the buffer zone is calculated, and the N(W)-W curve is drawn. When the growth of N(W) slows down (that is, the buffer zone begins to cover the area outside the river channel), the W corresponding to the inflection point is the average width of the river channel. The center line defines the mainstream direction of the river channel. In subsequent clustering, the search range will preferentially expand along both sides of the center line to avoid excessive searching perpendicular to the direction of the river channel (such as preventing the incorrect merging of the floodplain with the main channel). Base radius = data resolution Average river channel width. For each grid, the radius is modified based on the local terrain slope (Si) and the lithologic scour resistance coefficient (GRCi). This modification can be fitted from historical data. This has the effect of expanding the search range in steep slopes (large Si) or soft rock areas (small GRCi), compensating for data sparsity. Spatial constraints are achieved using the SKATER (Spatial Constrained Clustering) algorithm or the Redcap algorithm, using dynamic radius and river network connectivity as constraints to generate spatially contiguous clusters.

[0028] In some embodiments of the present application, the similarity between different first cluster regions is defined based on the drastic geographical features, including: The range and change rate of the drastic geographical feature at different locations in each first cluster area are calculated, and the similarity of the range and change rate of the same drastic geographical feature in different first cluster areas is calculated based on this. The similarity between different first cluster areas is defined by combining the similarities of the range and change rate, which is recorded as the comprehensive similarity.

[0029] In this embodiment, the calculation formula of the comprehensive similarity is as follows: ; in, For the The first cluster area and the The comprehensive similarity between the first cluster regions, For the The first cluster area and the The number of common drastic geographical features between the first cluster regions, For the Similar weights for the drastic geographical features, For the The first cluster area and the The first cluster of regions similarity in the extent of the drastic geographical features, For the The first cluster area and the The first cluster of regions The similarity of the rate of change of the drastic geographical features, For the The first constant of a drastically changing geographical feature, The similarity of the rate of change is corrected for the range similarity, and then the class average is taken It is slightly smaller than the average value to ensure the rationality of the similarity. The first constant is used to balance the size of the correction function.

[0030] The range and change rate of drastically changing geographical features. The range is the distribution of features within the region, and the change rate is the change of features at different locations within the region. Similarity is defined from the perspectives of range and change rate. Clusters are iteratively merged based on the comprehensive similarity. When the termination condition is met, the iteration stops.

[0031] In some embodiments of the present application, river channel geographical evolution indicators are defined, including: The change values ​​of the drastic geographical features under each second cluster area are counted, and the change values ​​of multiple drastic geographical features are standardized. The evolution weight corresponding to each drastic geographical feature is determined by the entropy weight method; The river channel geography evolution index under each second cluster area is defined by the evolution weight and the change value of the drastic geographical characteristics.

[0032] In this embodiment, the change value of the sudden change geographical feature is the degree of change over time. ; in, For the The river geographical evolution indicators under the second cluster area, For the The number of dramatic geographical changes under the second cluster area, For the The evolution weight of a dramatic geographical feature, For the The second cluster area The change value of a drastically changing geographic feature (the change value is different from the change rate. The change value is how the feature changes over time, and the change rate is how the feature changes with different locations). The biggest change in the drastic geographical features, For the The second constant under the second cluster region, It represents the correction of the maximum change to the average change, and the second constant is used to balance the size of the correction function.

[0033] Step S102: Demarcate a local initial area of ​​the river topographic map using the river geographical evolution index, analyze the boundary effect of the initial area to adjust the local initial area, and determine the local area.

[0034] In this embodiment, different river channel geographical evolution index intervals are set in advance, and different and adjacent second cluster areas within the same river channel geographical evolution index interval are merged to generate a merged second cluster area as a local initial area. The local area here is part of the river channel or some places that need special attention.

[0035] In some embodiments of the present application, the local initial area of ​​the river topographic map is demarcated by using river geographical evolution indicators, including: Each second cluster area is divided on the river topographic map, and the river geographical evolution index under each second cluster area is marked, and the difference in river geographical evolution index between adjacent second cluster areas is calculated; If the difference in river channel geographical evolution indicators between adjacent second cluster areas is less than a preset threshold, multiple adjacent second cluster areas are merged to obtain a new second cluster area; The old second cluster region and the new second cluster region are used as local initial regions.

[0036] In this embodiment, if the difference in the river geographical evolution index between adjacent second cluster areas is less than a preset threshold (through the river geographical evolution index interval), two or more adjacent second cluster areas that meet the requirements will be merged to obtain a new second cluster area; otherwise, the second cluster areas will not be merged, and the original second cluster areas (old second cluster areas) will be maintained.

[0037] In some embodiments of the present application, the boundary effect of the initial region is analyzed to adjust the local initial region, including: Identify all boundary effect types involved in the river channel, match the boundary effect type corresponding to each initial area, establish a boundary effect model for each initial area, determine the degree of boundary effect of each initial area, and expand the buffer area of ​​the boundary of each initial area to adjust the local initial area.

[0038] In this embodiment, by systematically analyzing the type and degree of boundary effects in the initial area of ​​the river channel, a quantitative model is established and the buffer area is expanded to achieve dynamic optimization of the local area to solve the problems of boundary information loss and analysis error accumulation caused by traditional static division. The boundary effects in the initial area are mainly caused by the following factors: Sudden changes in water flow dynamics: large differences in water velocity and direction on both sides of a boundary (e.g., the boundary between an erosion zone and a sedimentation zone); Sediment transport imbalance: Sudden changes in sediment transport rates at boundaries lead to local erosion and deposition (e.g., at the end of a revetment project); Topographic gradient changes: Gravity-driven sediment movement caused by differences in slope on both sides of a boundary (e.g., where a steep slope meets a gentle slope); Differences in vegetation cover: Different vegetation types or densities on both sides of the boundary affect erosion resistance (such as the junction of artificial forests and bare beaches).

[0039] Boundary effect types include dynamic mutation type (scenarios include meander neck cutting sections and the end of spur dikes, where the shear force of water flow suddenly increases at the boundary and the erosion range expands), transport imbalance type (tributary confluence, artificial trough edges, and differences in sediment transport rates on both sides of the boundary lead to local siltation or scouring), terrain transition type (the junction of the floodplain and the main channel, the bottom of the steep slope, where the terrain gradient at the boundary triggers gravity-driven sediment sliding), and ecological buffer type (the junction of the bank protection project and the natural river bank, the edge of the vegetation belt, and the sudden change in vegetation anti-scouring ability at the boundary affects the erosion rate).

[0040] Extended boundary: Applicable to scenarios where boundary effects spread outward (such as dynamic mutation type and ecological buffer type); Shrinking boundary: Applicable to scenarios where the boundary effect shrinks inward (such as transport imbalance and terrain transition); The degree of boundary effect is quantified through the boundary effect model, and the boundary is adjusted by the boundary adjustment direction and the degree of boundary effect (in the form of product).

[0041] Step S103 , determining the global and local description scales according to the river geographical evolution indicators and geographical feature information in the global and local regions, and establishing the global model and local model of the river respectively based on the description scales.

[0042] In some embodiments of the present application, the global and local description scales are determined based on the river geographical evolution indicators and geographical feature information in the global and local regions, including: For the global description scale, multiple macro-feature change indicators are screened from the geographic feature information, and the macro-complexity is calculated by integrating the multiple macro-feature change indicators. The river geographical evolution indicators of multiple second cluster areas under the global area are integrated to calculate the evolution complexity. The description scale of the global area is determined by combining the macro-complexity and evolution complexity. For the local description scale, multiple terrain features are screened out from the geographic feature information, and the terrain complexity of each second cluster area is calculated based on the terrain features. The evolution complexity of each second cluster area is calculated based on the river geographical evolution index of the second cluster area. The description scale of each second cluster area is determined based on the terrain complexity and evolution complexity of the second cluster area.

[0043] In this implementation, the global model captures overall river channel evolution trends (e.g., changes in river network topology and mainstream migration); the local model fine-tunes the dynamics of key regions (e.g., erosion and deposition in bends and confluence of tributaries). This addresses the issues of traditional single-scale models, where "global coarseness leads to loss of detail" or "local granularity leads to computational overhead."

[0044] Macro indicators: give priority to indicators that can reflect overall changes, such as: River channel length change rate: reflects the trend of river channel extension or shortening; Changes in watershed area: revealing the impact of erosion or sedimentation on watershed morphology; Main channel swing amplitude: quantified by the historical channel centerline offset distance; Interannual variation in sediment transport: reflects the overall balance of erosion-transport-deposition in the basin.

[0045] The macro-complexity is calculated by integrating multiple macro-feature change indicators, and the description scale of the global area is determined by combining macro-complexity and evolutionary complexity. The description scale is comprehensively determined by considering macro-complexity and evolutionary complexity.

[0046] The local description scale is the description scale of a part of the area, and the description scale is determined by considering the complexity of the terrain and the complexity of evolution. Different description scales correspond to parameters such as spatial resolution and time step.

[0047] Scale parameterization Spatial resolution: Select a resolution based on the width of the river and the complexity of the terrain. For example, large rivers (such as the Yangtze River) can use 30-meter resolution remote sensing imagery, while small and medium-sized rivers (such as the tributaries of the Pearl River) require 10-meter resolution or higher.

[0048] Time step: Determine the time interval based on the rate of evolution. Rapidly evolving rivers (such as mountain rivers) can be analyzed on an annual basis, while slowly evolving rivers (such as alluvial plain rivers) can be analyzed on a 5-10 year basis.

[0049] The model is established according to the scale parameterization of these models.

[0050] Global model construction Input data: DEM, river network, land use and other data at the global description scale; Model type: LSTM-CA (cellular automaton) model is used to simulate river network topology and mainstream migration; Output results: river centerline migration probability map and tributary growth and decline prediction.

[0051] Local model construction Input data: high-precision DEM, water velocity, sediment concentration and other data at the local description scale; Model type: CFD (computational fluid dynamics) and DEM coupled model is used to simulate bend erosion and siltation and tributary confluence; Output results: local terrain change map, water flow dynamic field distribution.

[0052] In step S104 , multiple types of control points are selected in the global region and the local region, and the global model and the local model of the river are superimposed by means of the multiple types of control points to generate an overall three-dimensional model of the river.

[0053] In some embodiments of the present application, multiple types of control points are selected in the global area and the local area, including: According to their functions, control points are divided into geometric control points, dynamic control points, evolutionary control points and boundary control points. According to the respective conditions of geometric control points, dynamic control points, evolution control points and boundary control points, they are jointly screened in the global and local areas to determine each type of control points, namely geometric control points, dynamic control points, evolution control points and boundary control points, and a layered overlay strategy of geometric alignment, dynamic coupling, evolution verification and boundary optimization is established, corresponding to the geometric control points, dynamic control points, evolution control points and boundary control points respectively.

[0054] In this embodiment, through the layered screening and layered superposition strategy of multiple types of control points, the coordinated constraints of the geometric form, dynamic process, evolution law and boundary conditions in the river model are realized, and the simulation distortion problem caused by the decoupling of "geometry-dynamics-evolution-boundary" in the traditional model is solved.

[0055] Functionally divided into geometric control points (constraining morphology), dynamic control points (constraining flow), evolutionary control points (constraining transitions), and boundary control points (constraining ranges), this enables precise control of multiple physical fields. A progressive constraint mechanism of "geometric alignment → dynamic coupling → evolutionary verification → boundary optimization" is established to ensure the self-consistency of the model across multiple scales and factors.

[0056] 1. Geometric Control Points (GCPs) Function: Constrain the river geometry (such as centerline, river width, cross-sectional shape). Filter conditions: Global Zone: Select key nodes of river network topology (such as the intersection and bifurcation points of main and tributary rivers); Screen the center points of high-curvature river sections (curvature radius < 5 times the river width); Example: In the "Nine Bends" area of ​​the Jingjiang section of the Yangtze River, a GCP is deployed every 200m.

[0057] Local area: In typical geomorphic units such as bends, mid-shoals, and deep troughs, they are laid out at equal intervals (e.g., 10m) or at characteristic points (e.g., bend apex); Example: For the bend of the Yellow River in Mengjin, three GCPs are placed in the convex bank scour zone and the concave bank siltation zone. Coordinates (X, Y, Z), river width (B), and cross-sectional morphological parameters (such as sump elevation and slope coefficient) are provided.

[0058] 2. Dynamic Control Points (DCPs) Function: Constrain the dynamic characteristics of water flow (such as flow rate, water level, and sediment content). Filter conditions: Global Zone: Select long-term observation points such as hydrological stations and flow monitoring sections; Screen key points along the flood peak propagation path (such as river channel narrowing sections and bridge piers); Example: In the Pearl River Delta network river area, DCPs are deployed at the entrance of each branch channel.

[0059] Local area: In dynamically complex areas such as bend circulation areas and tributary confluences, layout should be based on velocity gradient (e.g., denser layout should be implemented in areas with velocity variation coefficient > 0.3); Example: At the confluence of the tributaries in the Three Gorges Reservoir area of ​​the Yangtze River, five DCPs are deployed at the junction of the mainstream and tributaries.

[0060] Output data: Flow velocity (U, V, W), water level (H), sediment content (C), dynamic gradient ( ).

[0061] 3. Evolution Control Points (ECPs) Function: Constrain the evolution of river channels (such as river bank migration, main channel swing, erosion and sedimentation). Filter conditions: Global Zone: Select areas with significant historical river course changes (such as the remains of the Yellow River diversion and the Yangtze River straightening section); Screen river sections with annual erosion / sedimentation > 100,000 m³; Example: In the Gaocun to Aishan section of the lower Yellow River, an ECP is deployed every 5 km.

[0062] Local area: In areas with high evolution activity, such as scour pits on curved banks and the bottom of deep troughs, the layout should be based on the scouring and silting rate (e.g., more dense layout should be applied in areas with annual scouring and silting depth > 0.5m); Example: In the Lugou Bridge section of the Yongding River bend, three ECPs were deployed in the center of the scour pit.

[0063] Output data: Bankline migration rate (V_bank), main channel swing amplitude (A_channel), and erosion and deposition volume (ΔV).

[0064] 4. Boundary Control Points (BCPs) Function: Constrain model calculation boundaries (such as watershed range, inflow / outflow sections, terrain cutoff lines). Filter conditions: Global Zone: Select natural / artificial boundaries such as watersheds, large reservoir dam sites, and estuaries; Screening for sudden changes in terrain elevation (such as the junction of mountains and plains); Example: The Haihe River Basin is located with the Yanshan-Taihang Mountain watershed as its northern boundary and a BCP is laid out.

[0065] Local area: At the edge of the model calculation domain (e.g. outside 500m on both sides of the river), layout is based on terrain continuity; Example: A local model of the Taihu Lake basin uses the lake embankment as the boundary and deploys a BCP every 100m.

[0066] Output data: Boundary coordinates (X, Y, Z), boundary type (inflow / outflow / solid wall), boundary condition parameters (such as flow Q, water level H).

[0067] 1. Geometric alignment to ensure that the model geometry is consistent with the measured data.

[0068] Global alignment: The GCPs coordinates were matched with the river network centerlines extracted from remote sensing images, and the model grid node positions were adjusted by thin plate spline interpolation (TPS); Example: In the Jingjiang section of the Yangtze River model, TPS was used to reduce the GCPs error from 15m to 2m.

[0069] Local alignment: In areas such as bends and deep troughs, local radial basis function (RBF) interpolation is used to refine the grid with GCPs as constraints; Example: In the Yellow River Mengjin section bend model, the local grid resolution was increased from 50m to 10m.

[0070] Verification indicators: Geometric error: river width error <5%, centerline deviation <1 times the grid size.

[0071] 2. Dynamic coupling to ensure that the water flow dynamic process is consistent with the measured data.

[0072] Global coupling: The flow velocity and water level data of DCPs are used as strong constraints to optimize the model roughness parameters through the adjoint equation method; Example: In the Pearl River Delta model, after roughness optimization, the peak flow simulation error was reduced from 25% to 8%.

[0073] Local coupling: In areas such as bend circulation areas and tributary confluences, the velocity field is dynamically adjusted using data assimilation (EnKF) with DCPs as observation points; Example: In the Yangtze River Three Gorges Reservoir model, EnKF was used to reduce the velocity error in the tributary confluence area from 40% to 15%.

[0074] Verification indicators: Dynamic error: velocity correlation coefficient R²>0.85, water level root mean square error RMSE<0.2m.

[0075] Evolution verification ensures that the evolution patterns of river channels are consistent with historical data.

[0076] Global Validation: The bank migration rate and main channel swing amplitude data of ECPs were compared with the model prediction results, and the consistency of the evolution trend was verified by the Mann-Kendall trend test; Example: The lower Yellow River model has a 90% accuracy rate in predicting the main channel swing direction.

[0077] Local verification: In areas such as bend scour pits and deep troughs, the ECPs were used as observation points and the erosion and deposition balance method was used to verify the reliability of the model; Example: Yongding River Lugou Bridge section model, annual scouring and silting simulation error <10%.

[0078] Verification indicators: Evolution error: riverbank migration rate deviation <15%, erosion and deposition deviation <20%.

[0079] Boundary optimization ensures that the model boundary conditions are reasonable and the calculation is stable.

[0080] Global optimization: With BCPs as constraints, the inflow flow hydrograph is optimized using a genetic algorithm to make the model output water level consistent with the measured data; Example: In the Haihe River Basin model, the water level simulation error was reduced from 30% to 12% after inflow flow optimization.

[0081] Local optimization: At the edge of the model calculation domain, BCPs are used as observation points, and a buffer layering method is used to reduce boundary reflection effects; Example: For the Taihu Lake Basin model, after the buffer width is set to 500m, the boundary reflection error is <5%.

[0082] Verification indicators: Boundary stability: water level fluctuation amplitude <0.1m, flow velocity gradient <0.01s-¹.

[0083] Correspondingly, the present application also provides a river channel three-dimensional reconstruction system for riverbed reconstruction, such as Figure 2 Shown, including, The first module is used to obtain the geographical feature information of the river channel, establish a river channel topographic map, cluster the geographical features on the river channel topographic map, and define the river channel geographical evolution indicators; The second module is used to calibrate the local initial area of ​​the river topography map through the river geographical evolution index, analyze the boundary effect of the initial area to adjust the local initial area, and determine the local area; The third module is used to determine the global and local description scales based on the river geographical evolution indicators and geographical feature information in the global and local regions, and to establish the global model and local model of the river respectively based on the description scales; The fourth module is used to select multiple types of control points in the global area and the local area, and use the multiple types of control points to superimpose the global model and the local model of the river to generate an overall three-dimensional model of the river.

[0084] Compared with the prior art, the present invention has the following beneficial effects: 1. Cluster geographical features on the river topography map, dividing them into slowly changing geographical features and drastically changing geographical features. Perform two clustering operations based on these two types of features to improve the reliability of cluster analysis and provide a reliable foundation for the establishment and reconstruction of subsequent models. Demarcate the local initial area of ​​the river topography map using river geographical evolution indicators, analyze the boundary effects of the initial area to adjust the local initial area, and consider the river geographical evolution of the clustered area to initially demarcate the local area, i.e., some local areas with large changes and complex situations. Consider the boundary effects of the local initial area to appropriately adjust the regional boundaries, providing an accurate foundation for subsequent local modeling and ensuring the compatibility of the local area with the local model.

[0085] 2. Determine the global and local description scales based on the river's geographic evolution indicators and geographic characteristics within the global and local regions. Considering the model's description scale from both global and local perspectives allows for adapting to the 3D reconstruction accuracy requirements of different river regions, taking into account both the overall and partial conditions. Multiple types of control points are selected within the global and local regions. Using these control points, the global and local models of the river are superimposed, improving the rationality and adaptability of 3D river modeling and ensuring the flexibility and descriptive accuracy of riverbed reconstruction.

[0086] Through the above description of the embodiments, those skilled in the art will clearly understand that the present invention can be implemented via hardware or via software combined with a necessary general-purpose hardware platform. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product. This software product can be stored on a non-volatile storage medium (such as a CD-ROM, USB flash drive, or external hard drive) and includes instructions for enabling a computer device (such as a personal computer, server, or network device) to execute the methods described in various implementation scenarios of the present invention.

[0087] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0088] Those skilled in the art will appreciate that the modules in the system of the implementation scenario can be distributed in the system of the implementation scenario according to the implementation scenario description, or can be modified accordingly and located in one or more systems different from the implementation scenario. The modules of the above implementation scenario can be combined into one module or further divided into multiple submodules.

[0089] The above serial numbers of the present invention are for description only and do not represent the advantages or disadvantages of the implementation scenarios.

[0090] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A three-dimensional river reconstruction method for riverbed reconstruction, characterized in that: include, Obtain the geographical feature information of the river channel, establish a river channel topographic map, cluster the geographical features on the river channel topographic map, and define the river channel geographical evolution indicators; The local initial area of ​​the river topography map is demarcated by using the river geographical evolution index, and the boundary effect of the initial area is analyzed to adjust the local initial area and determine the local area; Determine the global and local description scales based on the river channel geographic evolution indicators and geographic feature information in the global and local regions, and establish the global and local models of the river channel respectively based on the description scales; Multiple types of control points are selected in the global area and the local area, and the global model and the local model of the river are superimposed with the multiple types of control points to generate an overall three-dimensional model of the river.

2. The three-dimensional river reconstruction method for riverbed reconstruction according to claim 1, characterized in that: Clustering of geographical features on river topography maps, include, Extracting multiple geographical features from the geographical feature information of the river channel, and distinguishing between slowly changing geographical features and drastically changing geographical features according to the changes in the historical records of the geographical features; Determine the search space range of the first cluster, perform first-stage clustering under the spatial constraints of the slowly varying geographic features based on the search space range of the first cluster and the slowly varying geographic features, and generate multiple first cluster areas; The similarity between different first cluster regions is defined according to the drastic geographical features, and the second stage clustering is performed to generate multiple second cluster regions.

3. The three-dimensional river reconstruction method for riverbed reconstruction according to claim 2, characterized in that: Determine the search space scope of the first cluster, including, Establish the river network skeleton of the river channel and calculate the distance from each grid to the center line. Determine the average width of the river channel based on the distance from the grid to the center line. Set the basic radius according to the average width of the river channel and adjust the basic radius according to the terrain slope and rock anti-scour coefficient. Use the adjusted basic radius as the search space range of the first cluster.

4. The three-dimensional river reconstruction method for riverbed reconstruction according to claim 2, characterized in that: The similarity between different first cluster regions is defined based on the drastic geographical features. include, Calculate the range and change rate of the sudden change geographical feature at different locations in each first cluster area. Use this to calculate the similarity of the range and change rate of the same sudden change geographical feature in different first cluster areas. Combine the similarities of the range and change rate to define the similarity between different first cluster areas, which is recorded as the comprehensive similarity. The calculation formula for comprehensive similarity is as follows: ； in, For the The first cluster area and the The comprehensive similarity between the first cluster regions, For the The first cluster area and the The number of common drastic geographical features between the first cluster regions, For the Similar weights for the drastic geographical features, For the The first cluster area and the The first cluster of regions similarity in the extent of the drastic geographical features, For the The first cluster area and the The first cluster of regions The similarity of the rate of change of the drastic geographical features, For the The first constant of a drastically changing geographical feature.

5. The three-dimensional river reconstruction method for riverbed reconstruction according to claim 2, characterized in that: Define indicators of river channel geographical evolution, include, The change values ​​of the drastic geographical features under each second cluster area are counted, and the change values ​​of multiple drastic geographical features are standardized. The evolution weight corresponding to each drastic geographical feature is determined by the entropy weight method; The river channel geography evolution index under each second cluster area is defined by the evolution weight and the change value of the drastic geographical characteristics.

6. The three-dimensional river reconstruction method for riverbed reconstruction according to claim 2, characterized in that: The local initial area of ​​the river topography map is demarcated through the river geographical evolution index. include, Each second cluster area is divided on the river topographic map, and the river geographical evolution index under each second cluster area is marked, and the difference in river geographical evolution index between adjacent second cluster areas is calculated; If the difference in river channel geographical evolution indicators between adjacent second cluster areas is less than a preset threshold, multiple adjacent second cluster areas are merged to obtain a new second cluster area; The old second cluster region and the new second cluster region are used as local initial regions.

7. The three-dimensional river reconstruction method for riverbed reconstruction according to claim 6, characterized in that: Analyze the boundary effect of the initial area to adjust the local initial area, include, Identify all boundary effect types involved in the river channel, match the boundary effect type corresponding to each initial area, establish a boundary effect model for each initial area, determine the degree of boundary effect of each initial area, and expand the buffer area of ​​the boundary of each initial area to adjust the local initial area.

8. The three-dimensional river reconstruction method for riverbed reconstruction according to claim 5, characterized in that: Determine the global and local description scales based on the river geographical evolution indicators and geographical feature information in the global and local regions, including: For the global description scale, multiple macro-feature change indicators are screened from the geographic feature information, and the macro-complexity is calculated by integrating the multiple macro-feature change indicators. The river geographical evolution indicators of multiple second cluster areas under the global area are integrated to calculate the evolution complexity. The description scale of the global area is determined by combining the macro-complexity and evolution complexity. For the local description scale, multiple terrain features are screened out from the geographic feature information, and the terrain complexity of each second cluster area is calculated based on the terrain features. The evolution complexity of each second cluster area is calculated based on the river geographical evolution index of the second cluster area. The description scale of each second cluster area is determined based on the terrain complexity and evolution complexity of the second cluster area.

9. The three-dimensional river reconstruction method for riverbed reconstruction according to claim 2, characterized in that: Multiple types of control points are selected in the global and local regions, including: According to their functions, control points are divided into geometric control points, dynamic control points, evolutionary control points and boundary control points. According to the respective conditions of geometric control points, dynamic control points, evolution control points and boundary control points, they are jointly screened in the global and local areas to determine each type of control points, namely geometric control points, dynamic control points, evolution control points and boundary control points, and a layered overlay strategy of geometric alignment, dynamic coupling, evolution verification and boundary optimization is established, corresponding to the geometric control points, dynamic control points, evolution control points and boundary control points respectively.

10. A three-dimensional river reconstruction system for riverbed reconstruction, characterized in that: include, The first module is used to obtain the geographical feature information of the river channel, establish a river channel topographic map, cluster the geographical features on the river channel topographic map, and define the river channel geographical evolution indicators; The second module is used to calibrate the local initial area of ​​the river topography map through the river geographical evolution index, analyze the boundary effect of the initial area to adjust the local initial area, and determine the local area; The third module is used to determine the global and local description scales based on the river geographical evolution indicators and geographical feature information in the global and local regions, and to establish the global model and local model of the river respectively based on the description scales; The fourth module is used to select multiple types of control points in the global area and the local area, and use the multiple types of control points to superimpose the global model and the local model of the river to generate an overall three-dimensional model of the river.

Images