River three-dimensional reconstruction method and system for riverbed reconstruction

Abstract

The application discloses a riverway three-dimensional reconstruction method and system for riverbed reconstruction, relates to the model construction technical field, and comprises the following steps: dividing geographical features into slowly-varying geographical features and rapidly-varying geographical features, clustering twice by virtue of the two types of features, improving the reliability of clustering analysis, and providing a reliable basis for the establishment and reconstruction of subsequent models. The local initial area of the riverway topographic map is demarcated by means of riverway geographical evolution indexes, the boundary effect of the initial area is analyzed to adjust the local initial area, and the respective description scales of the global area and the local area are determined according to the riverway geographical evolution indexes and geographical feature information in the global area and the local area. A plurality of control points are selected in the global area and the local area, the global model and the local model of the riverway are superposed by virtue of the plurality of control points, the rationality and adaptability of the riverway three-dimensional modeling are improved, and the plasticity and description precision of the riverbed reconstruction are ensured.

Description

Technical Field

[0001] This invention relates to the field of model building technology, and in particular to a method and system for three-dimensional reconstruction of river channels for riverbed restoration. Background Technology

[0002] With the increasing demands for ecological and refined water conservancy construction, traditional two-dimensional design methods are gradually being phased out due to their inability to accurately depict complex river morphologies. Three-dimensional reconstruction technology, through methods such as drone oblique photography and LiDAR scanning, can acquire high-precision topographic data. Combined with BIM or real-scene 3D modeling software, it can construct digital models containing elements such as topography, water bodies, and structures. For example, a river project in Shandong Province used a DJI M300 RTK drone to collect 0.05-meter resolution images, combined with GodWorks 3D software to generate a real-scene 3D model, achieving precision control with a planar error better than 0.15 meters and an elevation error better than 0.3 meters. This type of technology not only visually displays the spatial morphology of the river channel but also enables real-time linkage between the model and design schemes through dynamic correlation mechanisms. This provides data support for topographic analysis, flow simulation, and ecological restoration design in riverbed reconstruction, significantly improving the scientific nature of engineering design and construction efficiency.

[0003] In existing technologies, the modeling of river channels uses the same level of description accuracy for different locations, which cannot perform high-precision adaptive reconstruction of different local areas, resulting in poor rationality and adaptability of the 3D reconstruction of river channels.

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

[0005] The purpose of this invention is to address the problems of poor rationality and adaptability in existing three-dimensional river channel reconstruction techniques, and to propose a three-dimensional river channel reconstruction method for riverbed restoration, comprising:

[0006] Obtain geographical feature information of the river channel, establish a river channel topographic map, perform geographical feature clustering on the river channel topographic map, and define river channel geographical evolution indicators.

[0007] By using river geography evolution indicators, the initial local area of ​​the river topographic map is identified, and the boundary effect of the initial area is analyzed to adjust the initial local area and determine the local area.

[0008] Based on the river channel geographical evolution indicators and geographical feature information in the global and local areas, the respective description scales of the global and local areas are determined, and the global and local models of the river channel are established based on the description scales.

[0009] Multiple control points are selected in both the global and local regions. These control points are then used to overlay the global and local models of the river channel to generate a comprehensive three-dimensional model of the river channel.

[0010] In some embodiments of this application, geographic feature clustering is performed on river topographic maps, including:

[0011] Multiple geographical features are extracted from the geographical feature information of the river channel, and the gradual change geographical features and the dramatic change geographical features are distinguished based on the changes in the historical records of the geographical features.

[0012] Determine the search space range of the first cluster, and perform first-stage clustering under the spatial constraints of the gradually changing geographical features based on the search space range of the first cluster and the gradually changing geographical features to generate multiple first cluster regions;

[0013] Based on the dramatic geographical features, the similarity between different first-cluster regions is defined, and a second-stage clustering is performed to generate multiple second-cluster regions.

[0014] In some embodiments of this application, determining the search space range of the first cluster includes,

[0015] Establish the river network framework and calculate the distance from each grid to the center line. Determine the average width of the river based on the distance from the grid to the center line. Set the base radius according to the average width of the river and adjust the base radius by means of terrain slope and lithological erosion resistance coefficient. Use the adjusted base radius as the search space range of the first cluster.

[0016] In some embodiments of this application, the similarity between different first cluster regions is defined based on drastically changed geographical features, including...

[0017] Calculate the range and rate of change of the dramatic geographical features at different locations in each first cluster region. Then, calculate the similarity of the range and rate of change of the same dramatic geographical feature in different first cluster regions. Combine the similarity of the range and rate of change to define the similarity between different first cluster regions, which is denoted as the comprehensive similarity.

[0018] The formula for calculating the overall similarity is as follows:

[0019] ;

[0020] in, For the first The first cluster region and the first The overall similarity between the first cluster regions For the first The first cluster region and the first The number of identical dramatic geographical features among the first cluster regions For the first Similarity weights for drastically changed geographical features For the first The first cluster region and the first Between the first cluster regions The similarity of the range of dramatic geographical features. For the first The first cluster region and the first Between the first cluster regions The similarity of the rates of change of several dramatically changed geographical features For the first The first constant of a dramatically changing geographical feature.

[0021] In some embodiments of this application, river channel geographic evolution indicators are defined, including...

[0022] The change values ​​of dramatic geographical features under each second cluster region are statistically analyzed, and the change values ​​of multiple dramatic geographical features are standardized. The evolution weight corresponding to each dramatic geographical feature is determined by the entropy weight method.

[0023] The river geography evolution index under each second cluster region is defined by the change values ​​of evolution weights and dramatic geographical features.

[0024] In some embodiments of this application, local initial areas of river topographic maps are identified using river geography evolution indicators, including:

[0025] On the river topographic map, each second cluster region is divided, and the river geography evolution index under each second cluster region is marked. The difference in river geography evolution index between adjacent second cluster regions is calculated.

[0026] If the difference in river geography evolution index between adjacent second cluster regions is less than a preset threshold, then multiple adjacent second cluster regions will be merged to obtain a new second cluster region.

[0027] Use the old second cluster region and the new second cluster region as the local initial region.

[0028] In some embodiments of this application, the boundary effect of the initial region is analyzed to adjust the local initial region, including,

[0029] Identify all boundary effect types involved in the river channel, match the boundary effect type corresponding to each initial region, establish a boundary effect model for each initial region, determine the degree of boundary effect for each initial region, and expand the buffer zone of the boundary of each initial region to adjust the local initial region.

[0030] In some embodiments of this application, the descriptive scales for the global and local regions are determined based on river channel geographical evolution indicators and geographical feature information within the global and local regions, including:

[0031] For the global description scale, multiple macro-feature change indicators are selected from geographic feature information, and the macro-complexity is calculated by integrating multiple macro-feature change indicators. The river geography evolution indicators of multiple second cluster regions under the global region are integrated, and the evolution complexity is calculated. The description scale of the global region is determined by combining macro-complexity and evolution complexity.

[0032] For local description scales, multiple topographic features are selected from geographic feature information. The topographic complexity of each second cluster region is calculated based on the topographic features. The evolution complexity of each second cluster region is calculated based on the river geography evolution index of the second cluster region. The description scale of each second cluster region is determined based on the topographic complexity and evolution complexity of the second cluster region.

[0033] In some embodiments of this application, multiple types of control points are selected within the global and local regions, including:

[0034] According to their different functions, control points are divided into geometric control points, dynamic control points, evolutionary control points, and boundary control points.

[0035] Based on the conditions of geometric control points, dynamic control points, evolutionary control points, and boundary control points, a joint screening is conducted in both the global and local regions to determine each type of control point. A hierarchical overlay strategy of geometric alignment, dynamic coupling, evolutionary verification, and boundary optimization is established, corresponding to geometric control points, dynamic control points, evolutionary control points, and boundary control points, respectively.

[0036] Correspondingly, this application also provides a three-dimensional river channel reconstruction system for riverbed restoration, including,

[0037] The first module is used to obtain the geographical features of the river, build a river topographic map, perform geographical feature clustering on the river topographic map, and define river geographical evolution indicators.

[0038] The second module is used to identify the local initial area of ​​the river topographic map through river geography evolution indicators, analyze the boundary effect of the initial area to adjust the local initial area, and determine the local area.

[0039] The third module is used to determine the description scales of the global and local areas based on the river geographical evolution indicators and geographical feature information of the global and local areas, and to establish the global and local models of the river based on the description scales.

[0040] The fourth module is used to select multiple types of control points in the global and local regions. These control points are then used to overlay the global and local models of the river channel to generate a complete three-dimensional model of the river channel.

[0041] Compared with the prior art, the beneficial effects of this invention are as follows:

[0042] 1. Geographic feature clustering is performed on the river topographic map, categorizing geographic features into gradually changing and rapidly changing features. Two clustering operations are then performed based on these two categories to improve the reliability of the clustering analysis and provide a solid foundation for subsequent model building and reconstruction. Initial local areas on the river topographic map are identified using river geographical evolution indicators. Boundary effects of these initial areas are analyzed to adjust the local initial areas. The initial identification of local areas, i.e., some areas with significant changes and complex conditions, is considered in light of the river geographical evolution of the clustered areas. Boundary effects of the initial local areas are considered to appropriately adjust the regional boundaries, providing an accurate basis for subsequent local modeling and ensuring the fit between local areas and local models.

[0043] 2. Based on the river channel geographical evolution indicators and geographical features within the global and local areas, the respective description scales for the global and local areas are determined. The model's description scale is considered from both global and local perspectives to adapt to the 3D reconstruction accuracy requirements of different river regions, taking into account both the overall and local conditions. Multiple types of control points are selected within the global and local areas. These control points are used to overlay the global and local river channel models, improving the rationality and adaptability of the 3D river channel modeling and ensuring the plasticity and description accuracy of riverbed reconstruction. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of the process of a three-dimensional reconstruction method for riverbed restoration proposed in this invention.

[0045] Figure 2 This is a schematic diagram of a three-dimensional river reconstruction system for riverbed restoration proposed in this invention. Detailed Implementation

[0046] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0047] Reference Figure 1 A method for three-dimensional reconstruction of river channels for riverbed restoration includes the following steps.

[0048] Step S101: Obtain the geographical feature information of the river channel, establish a river channel topographic map, perform geographical feature clustering on the river channel topographic map, and define river channel geographical evolution indicators.

[0049] In this embodiment, data acquisition utilizes UAV oblique photography (such as DJI Phantom 4 RTK), airborne LiDAR, or a ground-based 3D scanner to obtain high-precision point cloud data (resolution ≤ 5cm) and orthophotos (GSD ≤ 2cm) of the river channel. Topographic map generation involves processing the point cloud and imagery using ContextCapture or Pix4D software to generate a DEM (Digital Elevation Model) and a DSM (Digital Surface Model). These are then overlaid with geographic features such as the riverbed, shoreline, and vegetation to construct a 3D topographic map of the river channel. The geographic feature information of the river channel includes all geographic features related to the river channel.

[0050] The geographical features extracted from river topographic maps need to cover three types of information: morphology, hydrology, and geology (mainly these three aspects).

[0051] Morphological characteristics: riverbed elevation, slope, curvature, cross-sectional shape (e.g., U-shaped / V-shaped), and width of the floodplain;

[0052] Hydrological characteristics: water flow velocity, water depth, and sediment content (derived through hydrological model inversion);

[0053] Geological characteristics: riverbed substrate type (sand, gravel, clay), erosion resistance (correlated with geological exploration data).

[0054] In some embodiments of this application, geographic feature clustering is performed on river topographic maps, including:

[0055] Multiple geographical features are extracted from the geographical feature information of the river channel, and the gradual change geographical features and the dramatic change geographical features are distinguished based on the changes in the historical records of the geographical features.

[0056] Determine the search space range of the first cluster, and perform first-stage clustering under the spatial constraints of the gradually changing geographical features based on the search space range of the first cluster and the gradually changing geographical features to generate multiple first cluster regions;

[0057] Based on the dramatic geographical features, the similarity between different first-cluster regions is defined, and a second-stage clustering is performed to generate multiple second-cluster regions.

[0058] In this embodiment, slowly changing geographical features are those that change relatively slowly (e.g., changes in elevation and slope), while rapidly changing geographical features are those that change more frequently and rapidly (e.g., hydrodynamic conditions and sediment transport). The clustering is divided into two stages: first, clustering is performed based on slowly changing geographical features; then, based on the results, a second clustering is performed using rapidly changing geographical features. This two-stage clustering model achieves classification based on changes in geographical feature types.

[0059] The first clustering is based on dynamic radius adjustment according to the river network topology. The radius is the search space range of the first cluster, thus adapting the search range to match the scale of river channel morphology. The second clustering is based on the similarity of dramatic geographical features among the first cluster regions, ultimately identifying multiple second cluster regions.

[0060] In some embodiments of this application, determining the search space range of the first cluster includes,

[0061] Establish the river network framework and calculate the distance from each grid to the center line. Determine the average width of the river based on the distance from the grid to the center line. Set the base radius according to the average width of the river and adjust the base radius by means of terrain slope and lithological erosion resistance coefficient. Use the adjusted base radius as the search space range of the first cluster.

[0062] In this embodiment, HydroSHEDS data or the ArcGIS Hydrology toolset is used to generate the river network skeleton, and the distance from each grid to the centerline is calculated. The river network centerline is used to generate a buffer zone. The average width of the river channel is inferred by statistically analyzing the number or area of ​​grids within the buffer zone. For example, for each segment of the centerline, a buffer zone of width W is generated, the number of grids N(W) covered by the buffer zone is calculated, and an N(W)-W curve is plotted. When the growth of N(W) slows down (i.e., 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 centerline defines the main direction of the river channel. In subsequent clustering, the search range will preferentially expand along both sides of the centerline to avoid over-searching perpendicular to the river channel direction (e.g., to prevent incorrect merging of floodplains and main channels). Base radius = data resolution. The average width of the river channel. For each grid, the radius is adjusted based on the local topographic slope (Si) and lithological erosion coefficient (GRCi). This adjustment can be derived from historical data. The effect is to expand the search range in steep slopes (large Si) or soft rock areas (small GRCi), compensating for data sparsity. Spatially continuous clusters are generated using the SKATER (Spatial Constrained Clustering) algorithm or the Redcap algorithm, constrained by the dynamic radius and river network connectivity, thus achieving spatial constraints.

[0063] In some embodiments of this application, the similarity between different first cluster regions is defined based on drastically changed geographical features, including...

[0064] Calculate the range and rate of change of the dramatic geographical features at different locations in each first cluster region. Then, calculate the similarity of the range and rate of change of the same dramatic geographical feature in different first cluster regions. Combine the similarity of the range and rate of change to define the similarity between different first cluster regions, which is denoted as the comprehensive similarity.

[0065] In this embodiment, the formula for calculating the overall similarity is as follows:

[0066] ;

[0067] in, For the first The first cluster region and the first The overall similarity between the first cluster regions For the first The first cluster region and the first The number of identical dramatic geographical features among the first cluster regions For the first Similarity weights for drastically changed geographical features For the first The first cluster region and the first Between the first cluster regions The similarity of the range of dramatic geographical features. For the first The first cluster region and the first Between the first cluster regions The similarity of the rates of change of several dramatically changed geographical features For the first The first constant of a dramatically changing geographical feature The similarity of the rate of change is adjusted for cases of similar range, and then the average value of the categories is taken. Slightly smaller than the average value to ensure reasonable similarity, the first constant is used to balance the size of the correction function.

[0068] The range and rate of change of drastically changing geographical features are used. The range is the distribution of features within the region, and the rate of change is the variation of features at different locations within the region. Similarity is defined from both the range and the rate of change perspectives. Iterative merging of clusters is performed based on the comprehensive similarity. The iteration stops when the termination condition is met.

[0069] In some embodiments of this application, river channel geographic evolution indicators are defined, including...

[0070] The change values ​​of dramatic geographical features under each second cluster region are statistically analyzed, and the change values ​​of multiple dramatic geographical features are standardized. The evolution weight corresponding to each dramatic geographical feature is determined by the entropy weight method.

[0071] The river geography evolution index under each second cluster region is defined by the change values ​​of evolution weights and dramatic geographical features.

[0072] In this embodiment, the change value of the dramatic geographical features is the degree of change over time.

[0073] ;

[0074] in, For the first River geography evolution indicators under the second cluster region For the first The number of dramatic geographical features within the second cluster region For the first The evolutionary weight of a dramatically changing geographical feature For the first The second cluster region under the first The change value of a dramatically changing geographical feature (the change value is different from the rate of change; the change value is how the feature changes over time, while the rate of change is how the feature changes depending on its location). This represents the most significant change among the dramatic geographical features. For the first The second constant under the second cluster region, The first constant represents the correction of the average change by the maximum change, and the second constant is used to balance the magnitude of the correction function.

[0075] Step S102: The initial local area of ​​the river topographic map is marked by the river geographical evolution index, the boundary effect of the initial area is analyzed to adjust the initial local area, and the local area is determined.

[0076] In this embodiment, different river geography evolution index intervals are pre-defined, and different and adjacent second cluster regions within the same river geography evolution index interval are merged to generate a merged second cluster region as a local initial region. Here, the local region is a part of the river channel or certain places that need to be focused on.

[0077] In some embodiments of this application, local initial areas of river topographic maps are identified using river geography evolution indicators, including:

[0078] On the river topographic map, each second cluster region is divided, and the river geography evolution index under each second cluster region is marked. The difference in river geography evolution index between adjacent second cluster regions is calculated.

[0079] If the difference in river geography evolution index between adjacent second cluster regions is less than a preset threshold, then multiple adjacent second cluster regions will be merged to obtain a new second cluster region.

[0080] Use the old second cluster region and the new second cluster region as the local initial region.

[0081] In this embodiment, if the difference in river geography evolution index between adjacent second cluster regions is less than a preset threshold (through the river geography evolution index range), two or more adjacent and qualified second cluster regions are merged to obtain a new second cluster region; otherwise, the second cluster regions are not merged, and the original second cluster regions (old second cluster regions) are maintained.

[0082] In some embodiments of this application, the boundary effect of the initial region is analyzed to adjust the local initial region, including,

[0083] Identify all boundary effect types involved in the river channel, match the boundary effect type corresponding to each initial region, establish a boundary effect model for each initial region, determine the degree of boundary effect for each initial region, and expand the buffer zone of the boundary of each initial region to adjust the local initial region.

[0084] In this embodiment, the type and degree of boundary effects in the initial region of the river channel are analyzed systematically, a quantitative model is established, and the buffer zone is expanded to achieve dynamic optimization of the local area, thereby solving problems such as loss of boundary information and accumulation of analysis errors caused by traditional static partitioning. The boundary effects in the initial region are mainly caused by the following factors:

[0085] Abrupt hydrodynamic changes: Large differences in flow velocity and direction on both sides of the boundary (such as the boundary between erosion and deposition zones).

[0086] Imbalance in sediment transport: Abrupt changes in sediment transport rate at the boundary lead to local scouring and silting (such as at the end of bank protection projects).

[0087] Topographic gradient changes: Slope differences on both sides of the boundary cause gravity-driven sediment slippage (such as the boundary between steep and gentle slopes).

[0088] Differences in vegetation cover: Different vegetation types or densities on both sides of the boundary affect erosion resistance (such as the boundary between plantations and bare beaches).

[0089] Boundary effects can be categorized into several types: dynamic mutation type (scenarios include the neck of a meandering river and the end of a groyne, where the shear force of the water flow increases sharply at the boundary, expanding the erosion range); transport imbalance type (tributary confluence and the edge of an artificial trench, where differences in sediment transport rates on both sides of the boundary lead to local siltation or scouring); topographic transition type (the boundary between a floodplain and the main channel, and the bottom of a steep embankment, where topographic gradients at the boundary induce gravity-driven sediment slippage); and ecological buffer type (the boundary between a revetment project and the natural riverbank, and the edge of a vegetation zone, where abrupt changes in vegetation erosion resistance at the boundary affect the erosion rate).

[0090] Extended Boundary: Applicable to scenarios where boundary effects spread outward (such as dynamic mutation type and ecological buffer type).

[0091] Contraction boundary: Applicable to scenarios where the boundary effect contracts inward (such as transport imbalance type and terrain transition type);

[0092] The degree of boundary effect is quantified by using a boundary effect model, and the boundary is adjusted by adjusting the direction of boundary adjustment and the degree of boundary effect (in the form of a product).

[0093] Step S103: Determine the description scales for the global and local areas based on the river geography evolution indicators and geography features of the global and local areas, and establish the global and local models of the river based on the description scales.

[0094] In some embodiments of this application, the descriptive scales for the global and local regions are determined based on river channel geographical evolution indicators and geographical feature information within the global and local regions, including:

[0095] For the global description scale, multiple macro-feature change indicators are selected from geographic feature information, and the macro-complexity is calculated by integrating multiple macro-feature change indicators. The river geography evolution indicators of multiple second cluster regions under the global region are integrated, and the evolution complexity is calculated. The description scale of the global region is determined by combining macro-complexity and evolution complexity.

[0096] For local description scales, multiple topographic features are selected from geographic feature information. The topographic complexity of each second cluster region is calculated based on the topographic features. The evolution complexity of each second cluster region is calculated based on the river geography evolution index of the second cluster region. The description scale of each second cluster region is determined based on the topographic complexity and evolution complexity of the second cluster region.

[0097] In this embodiment, the global model captures the overall evolution trend of the river channel (such as changes in river network topology and main stream migration); the local model provides a refined simulation of the dynamics of key areas (such as scouring and sedimentation in bends and tributary confluence). This addresses the problems of "overly coarse global model leading to loss of detail" or "overly fine local model leading to computational explosion" in traditional single-scale models.

[0098] Macroeconomic indicators: Prioritize indicators that reflect overall changes, such as:

[0099] River length change rate: reflects the trend of river extension or shortening;

[0100] Changes in watershed area: revealing the impact of erosion or deposition on watershed morphology;

[0101] Main channel swing amplitude: quantified by the historical offset distance of the channel centerline;

[0102] Interannual variation in sediment transport: reflects the overall balance of erosion-transportation-deposition in the watershed.

[0103] The macroscopic complexity is calculated by combining multiple macroscopic feature change indicators. The description scale of the global region is determined by combining macroscopic complexity and evolutionary complexity. The description scale is determined by comprehensively considering macroscopic complexity and evolutionary complexity.

[0104] Local descriptive scales refer to the descriptive scale of a specific region, determined by considering the complexity of terrain and evolution. Different descriptive scales correspond to parameters such as spatial resolution and time step.

[0105] Scale parameterization

[0106] Spatial resolution: The resolution is selected based on the width of the river channel and the complexity of the terrain. For example, large rivers (such as the Yangtze River) can use remote sensing images with a resolution of 30 meters, while small and medium-sized rivers (such as tributaries of the Pearl River) require a resolution of 10 meters or higher.

[0107] Time step: The time interval is determined based on the rate of evolution. Rapidly evolving river channels (such as mountain rivers) can be analyzed on an annual basis, while slowly evolving river channels (such as alluvial plain rivers) can be analyzed on a 5-10 year basis.

[0108] The model is built by using the scale parameterization of these models.

[0109] Global model building

[0110] Input data: DEM, river network, land use, and other data at a global descriptive scale;

[0111] Model type: LSTM-CA (cellular automata) model is used to simulate river network topology and main stream migration;

[0112] Output results: Probability map of river centerline migration, and prediction of tributary growth and decline.

[0113] Local model construction

[0114] Input data: High-precision DEM, water flow velocity, sediment concentration, and other data at a local descriptive scale;

[0115] Model type: A coupled CFD (Computational Fluid Dynamics) and DEM model is used to simulate scouring and silting in bends and the confluence of tributaries;

[0116] Output results: Local topographic change map, hydrodynamic field distribution.

[0117] Step S104: Select multiple control points in the global and local regions, and use these multiple control points to overlay the global and local models of the river channel to generate an overall three-dimensional model of the river channel.

[0118] In some embodiments of this application, multiple types of control points are selected within the global and local regions, including:

[0119] According to their different functions, control points are divided into geometric control points, dynamic control points, evolutionary control points, and boundary control points.

[0120] Based on the conditions of geometric control points, dynamic control points, evolutionary control points, and boundary control points, a joint screening is conducted in both the global and local regions to determine each type of control point. A hierarchical overlay strategy of geometric alignment, dynamic coupling, evolutionary verification, and boundary optimization is established, corresponding to geometric control points, dynamic control points, evolutionary control points, and boundary control points, respectively.

[0121] In this embodiment, a multi-type control point hierarchical screening and hierarchical superposition strategy is used to achieve coordinated constraints on geometric morphology, dynamic process, evolution law and boundary conditions in the river model, thereby solving the simulation distortion problem caused by the decoupling of "geometry-dynamic-evolution-boundary" in traditional models.

[0122] The system is functionally divided into geometric control points (constraint morphology), dynamic control points (constraint flow), evolution control points (constraint transition), and boundary control points (constraint range) to achieve precise control of multiphysics fields. A progressive constraint mechanism of "geometric alignment → dynamic coupling → evolution verification → boundary optimization" is established to ensure the self-consistency of the model under multiple scales and factors.

[0123] 1. Geometric Control Points (GCPs)

[0124] Function: Constrains river channel geometry (e.g., centerline, width, cross-sectional shape). Filtering criteria:

[0125] Global region:

[0126] Select key nodes in the river network topology (such as the confluence of main streams and tributaries, and the points where streams branch off).

[0127] Select the center point of a high-curvature river section (curvature radius < 5 times the river width);

[0128] Example: In the winding section of the Yangtze River in Jingjiang, a GCP is set up every 200m.

[0129] Local area:

[0130] Within 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).

[0131] Example: In the bend of the Yellow River at Mengjin, three GCPs are set up in both 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 thalweg elevation and slope coefficient).

[0132] 2. Dynamic Control Points (DCPs)

[0133] Function: To constrain the dynamic characteristics of water flow (such as flow velocity, water level, and sediment concentration). Screening criteria:

[0134] Global region:

[0135] Select long-term observation points such as hydrological stations and flow monitoring sections;

[0136] Select key points along the flood peak propagation path (such as narrowing sections of the river channel and bridge piers).

[0137] Example: In the Pearl River Delta river network area, DCPs are deployed at the entrance of each channel.

[0138] Local area:

[0139] In dynamically complex areas such as bend circulation zones and tributary confluences, the flow rate gradient should be used for the deployment of flow rate gradients (e.g., denser deployment in areas where the flow rate variation coefficient is >0.3).

[0140] Example: At the confluence of tributaries in the Three Gorges Reservoir area of ​​the Yangtze River, five DCPs are deployed at the boundary between the main stream and the tributaries.

[0141] Output data:

[0142] Flow velocity (U, V, W), water level (H), sediment concentration (C), dynamic gradient ( ).

[0143] 3. Evolution Control Points (ECPs)

[0144] Function: To constrain river channel evolution patterns (such as shoreline migration, main channel sway, and scour and deposition). Screening criteria:

[0145] Global region:

[0146] Select areas with significant historical river course changes (such as the Yellow River diversion relics and the straightening sections of the Yangtze River).

[0147] Select river sections with an annual erosion / siltation volume > 100,000 m³;

[0148] Example: In the section from Gaocun to Aishan in the lower reaches of the Yellow River, one ECP is deployed every 5km.

[0149] Local area:

[0150] In areas with high evolutionary activity, such as scour pits on convex banks of bends and the bottom of deep channels, the scour and sedimentation rates should be determined by the scour and sedimentation rates (e.g., denser deployment in areas with an annual scour and sedimentation depth > 0.5m).

[0151] Example: Three ECPs are installed in the center of the scour pit at the bend of the Yongding River near Lugou Bridge.

[0152] Output data:

[0153] Riverbank migration rate (V_bank), main channel oscillation amplitude (A_channel), and scouring and sedimentation volume (ΔV).

[0154] 4. Boundary Control Points (BCPs)

[0155] Function: Constrains the calculation boundaries of the model (e.g., watershed extent, inflow / outflow cross-sections, topographic cutoff lines). Filtering criteria:

[0156] Global region:

[0157] Select natural / artificial boundaries such as watersheds, large reservoir dam sites, and estuaries;

[0158] Filter out points where terrain elevation changes abruptly (such as the boundary between mountains and plains);

[0159] Example: The BCP is deployed in the Haihe River Basin with the Yanshan-Taihang Mountains watershed as its northern boundary.

[0160] Local area:

[0161] At the edge of the model's computational domain (such as within 500m on both sides of a river), the model is laid out according to the continuity of the terrain.

[0162] Example: Local model of Taihu Lake Basin, with the lakeside dike as the boundary, one BCP is set up every 100m.

[0163] Output data:

[0164] Boundary coordinates (X,Y,Z), boundary type (inflow / outflow / solid wall), boundary condition parameters (such as flow rate Q, water level H).

[0165] 1. Geometric alignment to ensure that the model's geometry matches the measured data.

[0166] Global alignment:

[0167] The coordinates of GCPs are matched with the river network centerline extracted from remote sensing images, and the position of the model grid nodes is adjusted by thin plate spline interpolation (TPS).

[0168] Example: In the model of the Jingjiang section of the Yangtze River, the GCP error was reduced from 15m to 2m using TPS.

[0169] Local alignment:

[0170] In areas such as curves and deep trenches, the mesh is refined by interpolation using local radial basis functions (RBF) with GCPs as constraints.

[0171] Example: The local mesh resolution of the Yellow River Mengjin section bend model is increased from 50m to 10m.

[0172] Verification metrics:

[0173] Geometric errors: River width error <5%, centerline deviation <1 times the grid size.

[0174] 2. Dynamic coupling ensures that the dynamic process of water flow is consistent with the measured data.

[0175] Global coupling:

[0176] Using the flow velocity and water level data of DCPs as strong constraints, the roughness parameters of the model are optimized by the adjoint equation method.

[0177] Example: In the Pearl River Delta model, after roughness optimization, the simulation error of peak flow was reduced from 25% to 8%.

[0178] Local coupling:

[0179] In areas such as bend circulation zones and tributary confluences, the velocity field is dynamically adjusted using data assimilation (EnKF) with DCPs as observation points.

[0180] Example: In the Three Gorges Reservoir area model, the velocity error in the tributary confluence area was reduced from 40% to 15% using EnKF.

[0181] Verification metrics:

[0182] Dynamic error: velocity correlation coefficient R²>0.85, water level root mean square error RMSE<0.2m.

[0183] Evolution verification ensures that the river's evolution patterns are consistent with historical data.

[0184] Global verification:

[0185] The data on the migration rate of the riverbank and the swing amplitude of the main channel of ECPs were compared with the model prediction results, and the consistency of the evolution trend was verified by the Mann-Kendall trend test.

[0186] Example: The model for the lower reaches of the Yellow River achieves a 90% accuracy rate in predicting the direction of the main channel's oscillation.

[0187] Local verification:

[0188] In areas such as scour pits and deep trenches in bends, the reliability of the model is verified by using ECPs as observation points and the scour-deposition balance method.

[0189] Example: The simulation error of the annual scouring and silting volume of the Lugou Bridge section of the Yongding River is less than 10%.

[0190] Verification metrics:

[0191] Evolution error: deviation of riverbank migration rate <15%, deviation of scouring and deposition <20%.

[0192] Boundary optimization ensures that the model's boundary conditions are reasonable and the computation is stable.

[0193] Global optimization:

[0194] Using BCPs as constraints, the inflow process curve is optimized through a genetic algorithm to make the model output water level match the measured data.

[0195] Example: In the Haihe River Basin model, the water level simulation error decreased from 30% to 12% after inflow optimization.

[0196] Local optimization:

[0197] At the edge of the model computational domain, using BCPs as observation points, a buffer layering method is employed to reduce boundary reflection effects;

[0198] Example: In the Taihu Lake basin model, after setting the buffer width to 500m, the boundary reflection error is <5%.

[0199] Verification metrics:

[0200] Boundary stability: water level fluctuation amplitude <0.1m, flow velocity gradient <0.01s-¹.

[0201] Correspondingly, this application also provides a three-dimensional river channel reconstruction system for riverbed restoration, such as... Figure 2 As shown, including,

[0202] The first module is used to obtain the geographical features of the river, build a river topographic map, perform geographical feature clustering on the river topographic map, and define river geographical evolution indicators.

[0203] The second module is used to identify the local initial area of ​​the river topographic map through river geography evolution indicators, analyze the boundary effect of the initial area to adjust the local initial area, and determine the local area.

[0204] The third module is used to determine the description scales of the global and local areas based on the river geographical evolution indicators and geographical feature information of the global and local areas, and to establish the global and local models of the river based on the description scales.

[0205] The fourth module is used to select multiple types of control points in the global and local regions. These control points are then used to overlay the global and local models of the river channel to generate a complete three-dimensional model of the river channel.

[0206] Compared with the prior art, the beneficial effects of this invention are as follows:

[0207] 1. Geographic feature clustering is performed on the river topographic map, categorizing geographic features into gradually changing and rapidly changing features. Two clustering operations are then performed based on these two categories to improve the reliability of the clustering analysis and provide a solid foundation for subsequent model building and reconstruction. Initial local areas on the river topographic map are identified using river geographical evolution indicators. Boundary effects of these initial areas are analyzed to adjust the local initial areas. The initial identification of local areas, i.e., some areas with significant changes and complex conditions, is considered in light of the river geographical evolution of the clustered areas. Boundary effects of the initial local areas are considered to appropriately adjust the regional boundaries, providing an accurate basis for subsequent local modeling and ensuring the fit between local areas and local models.

[0208] 2. Based on the river channel geographical evolution indicators and geographical features within the global and local areas, the respective description scales for the global and local areas are determined. The model's description scale is considered from both global and local perspectives to adapt to the 3D reconstruction accuracy requirements of different river regions, taking into account both the overall and local conditions. Multiple types of control points are selected within the global and local areas. These control points are used to overlay the global and local river channel models, improving the rationality and adaptability of the 3D river channel modeling and ensuring the plasticity and description accuracy of riverbed reconstruction.

[0209] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented in hardware or by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0210] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.

[0211] Those skilled in the art will understand that the modules in the system of the implementation scenario can be distributed throughout the system of the implementation scenario as described, or they can be modified to reside in one or more systems different from this implementation scenario. The modules of the above-mentioned implementation scenario can be merged into one module, or they can be further divided into multiple sub-modules.

[0212] The serial numbers of the present invention mentioned above are for descriptive purposes only and do not represent the superiority or inferiority of the implementation scenarios.

[0213] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A river channel three-dimensional reconstruction method for riverbed reconstruction, characterized by, include, Obtain geographical feature information of the river channel, establish a river channel topographic map, perform geographical feature clustering on the river channel topographic map, and define river channel geographical evolution indicators. By using river geography evolution indicators, the initial local area of ​​the river topographic map is identified, and the boundary effect of the initial area is analyzed to adjust the initial local area and determine the local area. Based on the river channel geographical evolution indicators and geographical feature information in the global and local areas, the respective description scales of the global and local areas are determined, and the global and local models of the river channel are established based on the description scales. Multiple control points are selected in the global and local regions. The global and local models of the river channel are overlaid using these control points to generate a 3D model of the river channel. in, Geographic feature clustering is performed on river topographic maps, including: Multiple geographical features are extracted from the geographical feature information of the river channel, and the gradual change geographical features and the dramatic change geographical features are distinguished based on the changes in the historical records of the geographical features. Determine the search space range of the first cluster, and perform first-stage clustering under the spatial constraints of the gradually changing geographical features based on the search space range of the first cluster and the gradually changing geographical features to generate multiple first cluster regions; Based on the dramatic geographical features, the similarity between different first-cluster regions is defined, and a second-stage clustering is performed to generate multiple second-cluster regions. The descriptive scales for the global and local regions are determined based on river channel evolution indicators and geographical features within the global and local areas, including: For the global description scale, multiple macro-feature change indicators are selected from geographic feature information, and the macro-complexity is calculated by integrating multiple macro-feature change indicators. The river geography evolution indicators of multiple second cluster regions under the global region are integrated, and the evolution complexity is calculated. The description scale of the global region is determined by combining macro-complexity and evolution complexity. For local description scales, multiple topographic features are selected from geographic feature information. The topographic complexity of each second cluster region is calculated based on the topographic features. The evolution complexity of each second cluster region is calculated based on the river geography evolution index of the second cluster region. The description scale of each second cluster region is determined based on the topographic complexity and evolution complexity of the second cluster region.

2. The river three-dimensional reconstruction method for riverbed reconstruction according to claim 1, characterized in that, Determine the search space range for the first cluster, including: Establish the river network framework and calculate the distance from each grid to the center line. Determine the average width of the river based on the distance from the grid to the center line. Set the base radius according to the average width of the river and adjust the base radius by means of terrain slope and lithological erosion resistance coefficient. Use the adjusted base radius as the search space range of the first cluster.

3. The river three-dimensional reconstruction method for riverbed reconstruction according to claim 1, characterized in that, The similarity between different first-cluster regions is defined based on the characteristics of dramatic geographical changes. include, Calculate the range and rate of change of the dramatic geographical features at different locations in each first cluster region. Then, calculate the similarity of the range and rate of change of the same dramatic geographical feature in different first cluster regions. Combine the similarity of the range and rate of change to define the similarity between different first cluster regions, which is denoted as the comprehensive similarity. The formula for calculating the overall similarity is as follows: ； in, For the first The first cluster region and the first The overall similarity between the first cluster regions For the first The first cluster region and the first The number of identical dramatic geographical features among the first cluster regions For the first Similarity weights for drastically changed geographical features For the first The first cluster region and the first Between the first cluster regions The similarity of the range of dramatic geographical features. For the first The first cluster region and the first Between the first cluster regions The similarity of the rates of change of several dramatically changed geographical features For the first The first constant of a dramatically changing geographical feature.

4. The method for three-dimensional reconstruction of river channels for riverbed restoration according to claim 1, characterized in that, Define indicators of river channel geographical evolution. include, The change values ​​of dramatic geographical features under each second cluster region are statistically analyzed, and the change values ​​of multiple dramatic geographical features are standardized. The evolution weight corresponding to each dramatic geographical feature is determined by the entropy weight method. The river geography evolution index under each second cluster region is defined by the change values ​​of evolution weights and dramatic geographical features.

5. The method for three-dimensional reconstruction of river channels for riverbed restoration according to claim 1, characterized in that, Local initial areas of river topographic maps are identified using river geography evolution indicators. include, On the river topographic map, each second cluster region is divided, and the river geography evolution index under each second cluster region is marked. The difference in river geography evolution index between adjacent second cluster regions is calculated. If the difference in river geography evolution index between adjacent second cluster regions is less than a preset threshold, then multiple adjacent second cluster regions will be merged to obtain a new second cluster region. Use the old second cluster region and the new second cluster region as the local initial region.

6. The method for three-dimensional reconstruction of river channels for riverbed restoration according to claim 5, characterized in that, Analyze the boundary effects of the initial region to adjust the local initial region. include, Identify all boundary effect types involved in the river channel, match the boundary effect type corresponding to each initial region, establish a boundary effect model for each initial region, determine the degree of boundary effect for each initial region, and expand the buffer zone of the boundary of each initial region to adjust the local initial region.

7. The method for three-dimensional reconstruction of river channels for riverbed restoration according to claim 1, characterized in that, Multiple types of control points are selected within the global and local regions, including: According to their different functions, control points are divided into geometric control points, dynamic control points, evolutionary control points, and boundary control points. Based on the conditions of geometric control points, dynamic control points, evolutionary control points, and boundary control points, a joint screening is conducted in both the global and local regions to determine each type of control point. A hierarchical overlay strategy of geometric alignment, dynamic coupling, evolutionary verification, and boundary optimization is established, corresponding to geometric control points, dynamic control points, evolutionary control points, and boundary control points, respectively.

8. A three-dimensional river channel reconstruction system for riverbed restoration, characterized in that, The system is used to implement the three-dimensional river channel reconstruction method for riverbed restoration as described in any one of claims 1-7, the system comprising: The first module is used to obtain the geographical features of the river, build a river topographic map, perform geographical feature clustering on the river topographic map, and define river geographical evolution indicators. The second module is used to identify the local initial area of ​​the river topographic map through river geography evolution indicators, 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 description scales of the global and local areas based on the river geographical evolution indicators and geographical feature information of the global and local areas, and to establish the global and local models of the river based on the description scales. The fourth module is used to select multiple types of control points in the global and local regions. These control points are then used to overlay the global and local models of the river channel to generate a complete three-dimensional model of the river channel.

Images