Hydrological monitoring multi-element real-time management system

Abstract

The invention discloses a hydrological monitoring multi-element real-time management system. A hydrological monitoring multi-element real-time management system comprises a water area vector information collection device which obtains DME data of a target area and constructs a water body vector model based on the DME data; the three-dimensional model establishing device is used for acquiring high-altitude image data of the target area and establishing a three-dimensional scene model based on the high-altitude image data; in the technical scheme provided by the invention, the water body vector model and the three-dimensional scene model are generated in advance, and then the water body vector model and the three-dimensional scene model are mutually fused, which is substantially that the water body vector model is rendered into the three-dimensional scene model, so that the water body vector model can be obtained according to the change condition of the precipitation. And the flow direction trend of the water body in the three-dimensional scene model is rendered, and the water level change conditions of different positions in the target area are accurately predicted, so that a correct decision can be made conveniently when extreme weather occurs.

Description

Technical Field

[0001] This application relates to the technical field of neural network processing, and more particularly, to a real-time management system for multi-element hydrological monitoring. Background Art

[0002] The distribution of natural disasters has strong regionality. Areas with developed water areas are long-term affected by floods and heavy rainfall. In the Poyang Lake Basin, when encountering heavy rainfall, the water area of Poyang Lake will continuously rise, and then flood the surrounding land. When extreme rainfall weather comes, not only the water level of Poyang Lake will continuously rise, but also mountainous and hilly areas around will cause flash flood problems due to precipitation. Currently, in extreme situations, people mainly judge the disaster-affected order of each region based on the water level of Poyang Lake and the terrain elevation difference around. This method is too idealized and empirical, and does not fully consider the complexity of surrounding hydrological and geographical conditions. Therefore, in extreme climates, it is impossible to accurately predict the order of flood inundation. Summary of the Invention

[0003] This part of the content of this application is used to briefly introduce concepts, which will be described in detail in the subsequent Detailed Description part. This part of the content of this application is not intended to identify the key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0004] As a first aspect of this application, to solve the technical problems mentioned in the above Background Art part, some embodiments of this application provide a real-time management system for multi-element hydrological monitoring, including: A water area vector information collection device, which acquires DME data of a target area and constructs a water body vector model based on the DME data; A three-dimensional model establishment device, which acquires aerial image data of a target area and establishes a three-dimensional scene model based on the aerial image data; A visualization rendering device, which renders the water flow to the three-dimensional scene model based on the water body vector model and generates a water flow scene diagram in the three-dimensional scene model; An information alarm device, which acquires rainfall information in real time, generates a hydrological alarm diagram based on the rainfall information and the water flow scene diagram, and issues an alarm based on the hydrological alarm diagram; Among them, the water area vector information collection device constructs a water body vector model based on the following steps: Step 1: Acquire multiple DME data of the target area, and perform coordinate transformation on the DME data by the seven-parameter transformation method to complete resampling and splicing in a unified coordinate system to generate target DME information; Step 2: Perform vector superposition on the river channel vector of the target area and the DME information, and perform cutting based on the edge of the target area to generate preliminary water area distribution information; Step 3: rasterize the preliminary water area distribution information, and assign water area vectors to each raster area to generate a water area vector map; Step 4: generate a water body vector model based on the water area vector map.

[0005] In the technical solution provided by this application, a water body vector model and a three-dimensional scene model are pre-generated, and then the water body vector model and the three-dimensional scene model are fused with each other. Substantially, it is to render the water body vector model into the three-dimensional scene model, and then be able to render the flow trend of the water body in the three-dimensional scene model according to the change of precipitation, accurately predict the water level change situation at different positions in the target area, so as to facilitate making correct decisions in case of extreme weather.

[0006] When predicting the hydrological inundation direction of the target area, usually the target area is a large administrative division range, and the accuracy of its DME data is often limited. However, if the target area is subdivided into multiple small areas with high precision, it may lead to insufficient continuity of hydrological prediction. To address this contradiction, the following technical solution is proposed in this application: Further, Step 1 includes the following steps: Step 11: obtain the DME data of each local area in the target area; Step 12: splice the DME data based on the SIFT algorithm, and crop the spliced DME data according to the scope of the target area; Step 13: reduce the dimension of the cropped DME data to generate the target DME information.

[0007] In this solution, to break through the limitations of local area hydrological information prediction, the local DME data of the target area is comprehensively collected and spliced using the SIFT algorithm, so as to ensure the construction of a water body vector model based on the complete DME data. This solution not only ensures the independence of the water body vector model, but also realizes the reasonable prediction of hydrological information in the surrounding areas while eliminating redundant data.

[0008] Step 2 includes the following steps: Step 21: obtain the river channel vector data of the target area; Step 22: perform vector overlay on the river channel vector data and the target DME information using QGIS software to generate preliminary water area distribution information.

[0009] In this solution, the water area vector map finally obtained through vector overlay not only contains DME information, but also contains vector information of the river channel water area in the target area, so it can accurately perform water analysis and flood simulation.

[0010] Further, Step 3 includes the following steps: Step 31: Add a network grid layer of a preset specification to the preliminary water area distribution information; Step 32: Set a depression threshold, fill in the data of the network grid in the preliminary water area distribution information to obtain the water flow direction information; Step 33: Describe the water flow vectors in each grid based on the tangent curvature of the grid to generate a water area vector map.

[0011] In this solution, after filling in the data of the water area vector map, the influence of depression data on water flow prediction can be significantly reduced, and the calculation efficiency can be increased.

[0012] Currently, the calculation of water flow vectors is all based on the D8 algorithm. This algorithm belongs to a typical single-flow algorithm, that is, it assumes that the water flow direction is unique, and the general water flow direction needs to be preset when constructing a water flow model. However, this simplified processing method cannot simulate the water flow diversion phenomenon in a complex geographical environment, resulting in limited accuracy of the generated water flow vector map. To solve this problem, the present application proposes the following technical solution: Step 33 includes the following steps: Step 331: Calculate the tangent curvature Cp and streamline curvature Dp of grid p; Step 332: Preset the water flow aggregation divergence value W and the water flow path swing degree threshold R in advance, where W > 0 and R > 0; Step 333: If Cp ∈ (-W, W) and Dp ∈ (-R, R), then use the D8 algorithm to determine the water flow direction of grid p and the flow distribution ratio to the surrounding grids; If Cp ∈ (-W, W) and Dp ∈ (-∞, -R] ∪ [R, +∞), use the Rho8 algorithm to determine the water flow direction of grid p and the flow distribution ratio to the surrounding grids; If Cp ∈ (-∞, -W] ∪ [W, +∞), use the Dinf algorithm to determine the water flow direction of grid p and the flow distribution ratio to the surrounding grids; Step 334: Calculate the water flow direction and flow distribution ratio of each grid in turn according to Step 333 to generate a water flow direction matrix of the target area to generate a water area vector map.

[0013] In the technical solution provided by the present application, instead of using a single D8 algorithm to calculate the water flow vectors at each position in the target area, the tangent curvature and streamline curvature are used as the judgment basis, and a suitable water flow distribution algorithm is flexibly selected. Therefore, in practical applications, the swing state of the water flow can be accurately described based on the aggregation and divergence of the water flow. Compared with the single D8 algorithm, this method has higher accuracy.

[0014] Further, ; ; ; Among them, f represents the height function of the terrain surface, which is used to describe the height of the grid; g represents the first-order partial derivative of f with respect to i, which is used to describe the slope of the terrain in the i direction; q represents the slope of f in the j direction, r represents the second-order partial derivative of f with respect to i, s represents the mixed second-order partial derivative of f with respect to i and j, t represents the second-order partial derivative of f with respect to j, and i and j respectively represent the abscissa and ordinate of the grid. Represents the derivative symbol.

[0015] In this solution, when calculating Cp and Dp, p describes the slope or inclination degree of the terrain surface in the i direction, q describes the slope or inclination degree of the terrain surface in the j direction, r describes the curvature change of the terrain surface in the i direction, s describes the interaction or distortion degree of the terrain surface in the i and j directions, t describes the curvature change of the terrain surface in the j direction. Thus, Cp can accurately describe the aggregation and divergence ability of the ground surface for water flow movement, and Dp can describe the swing degree of the water flow path or the change of the water flow direction, so as to accurately construct a hydrological model.

[0016] Furthermore, step 4 includes the following steps: Step 41: Extract the water flow direction of each grid and its flow distribution ratio to the surrounding grids from the water area vector map; Step 42: Calculate the total water flow vector of each grid and assign a flow value according to the total; Step 43: Generate a water body vector model based on the flow value.

[0017] In this solution, through the total water flow vector, the complex water flow vector map can be dimensionally reduced, the data complexity can be reduced, and the water flow change areas in each region can be accurately described, so as to accurately describe the current situation of the flow direction changes in each region of the target.

[0018] Furthermore, the visualization rendering device includes: An information acquisition unit that acquires the water body vector model, the three-dimensional scene model, and the water volume prediction data; A rendering information processing unit that generates rendering information based on the water body vector model and the water volume prediction data; A rendering processing unit that defines the vertex data in the three-dimensional scene model and performs matrix transformation to convert the three-dimensional coordinates into two-dimensional screen coordinates; Calculate the final color of the pixel according to the color information and texture information included in the rendering information to generate a water flow scene map.

[0019] In the technical solution provided by this application, the three-dimensional scene model is converted into two-dimensional screen coordinates, and then each pixel in the two-dimensional screen coordinates is color-rendered, so that the change of the water level can be accurately rendered. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings constituting a part of this application are used to provide a further understanding of this application, making other features, objectives, and advantages of this application more obvious. The schematic embodiments and descriptions of the drawings of this application are used to explain this application and do not constitute an improper limitation of this application.

[0021] In addition, throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the elements and elements are not necessarily drawn to scale.

[0022] In the drawings: Figure 1 is a schematic structural diagram of a multi-element real-time management system for hydrological monitoring.

[0023] Figure 2 is the DME data of the target area.

[0024] Figure 3 is the DME data river channel vector data of the target area.

[0025] Figure 4 is the water area vector map of the target area. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] The embodiments of this application will be described in more detail below with reference to the drawings. Although some embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand this application. It should be understood that the drawings and embodiments of this application are only for exemplary purposes and are not used to limit the protection scope of this application.

[0027] In addition, it should be noted that, for the sake of convenience of description, only parts related to the relevant invention are shown in the drawings. Without conflict, the embodiments in this application and the features in the embodiments can be combined with each other.

[0028] This application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0029] Refer to Figure 1 , the multi-element real-time management system for hydrological monitoring includes: A water area vector information collection device that acquires the DME data of the target area and constructs a water body vector model based on the DME data; A three-dimensional model building device acquires aerial image data of a target area and builds a three-dimensional scene model based on the aerial image data. A visualization rendering device renders the water flow onto the three-dimensional scene model based on a water body vector model and generates a flow direction area of the water body in the three-dimensional scene model.

[0030] Specifically, the three-dimensional scene model is a three-dimensional real scene model of the target area, which needs to be jointly surveyed according to local surveying and mapping data and aerial image data in practice. The elevation of each position in the target area is marked in the three-dimensional real scene model, and thus the height difference information of each position in the target area is obtained.

[0031] Among them, the target area is a local area that needs to be monitored. The size of the target area can be set according to the actual situation, mainly to monitor the water level change near the water area to avoid the inability to allocate rescue forces in a timely and reasonable manner under heavy rainfall weather.

[0032] The core innovation of this solution is to propose a method for constructing a three-dimensional dynamic water body model. By coupling hydrological parameters with topographic features, this method can real-time simulate the inundation state of each position in the target area under heavy rainfall conditions, providing an accurate decision-making basis for the dynamic allocation of rescue resources.

[0033] Among them, the water area vector information collection device constructs a water body vector model based on the following steps: Reference Figure 2 , Step 1: Acquire multiple DME data of the target area, and perform coordinate transformation on the DME data using the seven-parameter transformation method to complete resampling and splicing in a unified coordinate system to generate target DME information.

[0034] Among them, DME data is digital elevation model data. DME data is a kind of entity ground model representing ground elevation in the form of an ordered numerical array, and DME data is a TIFF image. DME data is generally obtained through remote sensing satellite measurement, unmanned aerial vehicle measurement or ground measurement. It can be queried in a public database.

[0035] Furthermore, Step 1 includes the following steps: Step 11: Acquire DME data of each local area in the target area.

[0036] The local area in this solution is the source of DME data division. Generally speaking, high-precision DME requires a large amount of resources for measurement. Therefore, each data source will only have high-precision DME data for a certain area. When the precipitation range is wide, that is, the target area is large, it is necessary to obtain the DME data of each local area within the target area. The division method of the local area here depends on the storage situation of DME data in the local database. If the range of the target area is small and the DME data of the target area can be directly obtained from a certain data source, then the local area is the target area.

[0037] Step 12: Stitch the DME data based on the SIFT algorithm and crop the stitched DME data according to the range of the target area to obtain the target DME information.

[0038] Generally speaking, the DME data of multiple regions may be involved within the target area. Therefore, it is necessary to stitch these DME data. The stitching scheme is actually image stitching. The specific method is as follows: First, use the SIFT algorithm to extract the edge feature points in the image information, and then align the same feature points in the two images to be stitched, and the image stitching can be completed.

[0039] Reference Figure 3 Step 2: Perform vector superposition on the river channel vector of the target area and the DME information, and crop based on the edge of the target area to generate the preliminary water area distribution information; Step 2 includes the following steps: Step 21: Obtain the river channel vector data of the target area.

[0040] The river channel vector data "is the digital geographic information of water systems such as rivers, channels, and waterways stored in vector format" and is high-definition image information.

[0041] Step 22: Use QGIS software to perform vector superposition on the river channel vector data and the target DME information to generate the preliminary water area distribution information.

[0042] Superposing the river channel vector data and the target DME information is actually using the map of the target area as a template and superposing the DME information and the river channel vector information on this template, so that the water system situation in the target area can be clearly marked on the map, and thus the distribution situation of the water area is obtained.

[0043] Reference Figure 4 Step 3: Perform rasterization processing on the preliminary water area distribution information and assign water area vectors to each raster area to generate a water area vector map; Step 3 includes the following steps: Step 31: Add a network raster layer with a preset specification to the preliminary water area distribution information.

[0044] Step 32: Set the depression threshold and perform data filling for the network grid in the preliminary water area distribution information.

[0045] To accurately describe the water flow situation in the target area, it is necessary to determine the height at each position on the water area vector map, that is, data filling.

[0046] The specific method is as follows: Step 321: Traverse each grid in the preliminary water area distribution information to obtain the elevation value of each grid; Step 322: Compare the elevation value of each grid with the elevation values of its 8 neighboring grids. If the elevation values of all neighbors are higher than the elevation value of this grid, then mark this grid as a depression grid; Step 323: Group all the marked depression grids according to spatial connectivity to form independent depression areas;

[0047] For example: If adjacent depression grids are spatially continuous, they are grouped into the same depression network.

[0048] Step 324: Assign a unique ID to each independent depression area and record the coordinates and elevation values of all its grids; Step 325: Set the maximum number of iterations, set the elevation increment step size h0, set the initial number of iterations to 0, and execute the following loop condition: When the depression area is not an empty set and the current number of iterations is less than the maximum number of iterations, execute the loop; (1): Select the grid C with the minimum elevation value from the depression area min , the elevation value of C min is E min ; (2): Check the 8 neighboring grids of C min and filter out the set S no of grids that are not marked as depressions; (3): Calculate the minimum overflow elevation E sp ; If S no is not an empty set, then E sp = min({E k | E k ∈ S no}); If S no is an empty set, then E sp = E min + h0; (4) Adjust the elevation of the grid C min with the minimum value; If: E sp > E min , then E min = Esp +h0; If: E sp≤ E min , then C min is completed by filling depressions, removed from the depression area, and continue to fill depressions for subsequent grids.

[0049] In this way, through data filling, the DME data not only contains the corrected elevation surface, but also completely records the spatial logic and parameter details of the filling process through additional annotation information.

[0050] Step 33: Describe the water flow vectors in each grid based on the tangent curvature of the grid to generate a water area vector map.

[0051] Step 33 includes the following steps: Step 331: Calculate the tangent curvature Cp and streamline curvature Dp of grid p; Step 332: Preset the water flow aggregation divergence value W and the water flow path swing degree threshold R, W>0, R>0; the water flow aggregation divergence value W and the water flow path swing degree threshold R are preset values, generally W = 0.05, R = 0.1.

[0052] Step 333: If Cp∈(-W, W) and Dp∈(-R, R), then use the D8 algorithm to determine the water flow direction of grid p and its flow distribution ratio to the surrounding grids; If Cp∈(-W, W) and Dp∈(-∞, -R]∪[R, +∞), use the Rho8 algorithm to determine the water flow direction of grid p and its flow distribution ratio to the surrounding grids; If Cp∈(-∞, -W]∪[W, +∞), use the Dinf algorithm to determine the water flow direction of grid p and its flow distribution ratio to the surrounding grids; Step 334: Calculate the water flow direction and flow distribution ratio of each grid in turn according to Step 333 to generate a water flow direction matrix of the target area to generate a water area vector map.

[0053] Furthermore, the calculation formulas for the tangent curvature Cp and streamline curvature Dp are: ; ; ; where f represents the height function of the terrain surface, used to describe the height of the grid; g represents the first-order partial derivative of f with respect to i, used to describe the slope of the terrain in the i direction; q represents the slope of f in the j direction, r represents the second-order partial derivative of f with respect to i, s represents the mixed second-order partial derivative of f with respect to i and j, t represents the second-order partial derivative of f with respect to j, and i and j respectively represent the abscissa and ordinate of the grid, Represents the derivative symbol.

[0054] Step 4: Generate a water body vector model based on the water area vector map.

[0055] Furthermore, Step 4 includes the following steps: Step 41: Extract the water flow direction of each grid and its flow distribution ratio to the surrounding grids from the water area vector map. In Step 334, the water flow direction and flow distribution ratio of each grid are generated, and here they are extracted.

[0056] Step 42: Calculate the total water flow vector of each grid and assign a flow value according to the total.

[0057] The total water flow vector is the vector addition of the flow in each direction in the grid, that is, the addition with direction. The water flow into the grid is positive, and the water flow out of the grid is negative; Step 43: Generate a water body vector model based on the flow value.

[0058] The water body vector model is the water flow in each network grid under each time period. Dynamically updating the water body vector model can obtain the water flow change situation in each area.

[0059] The above is the water level change situation at each position in the target area. After obtaining the water level change situation at each position in the target area, in order to increase intuitiveness, it is necessary to render the water level change situation into the 3D scene model. Thus, it is necessary to first establish a 3D scene model.

[0060] The 3D model establishment device constructs the 3D scene model of the target area based on the following method, S1: Obtain the high-altitude top view and altitude distribution map of the target area, and integrate the altitude information in the altitude distribution map into the high-altitude top view to generate the fusion information of the target area; In the fusion information, each pixel point contains information in 3 dimensions. The information in the first dimension is the color information of the pixel point, that is, the color value of the three-channel RGB behind each pixel point. The information in the second dimension is the position information of the pixel point, and the information in the third dimension is the altitude information at the corresponding position.

[0061] Among them, the information in the first dimension is actually 3D information, and the information in the second dimension is 2D information, but both need to be reduced to 1D information. To avoid information loss, the following solution is adopted in this scheme.

[0062] For the first - dimension information, set it as an encoding that is a multiple of three. For example, set it as a 3 - bit encoding. The first bit of the encoding is used to represent the information of the R channel, the second bit is used to represent the information of the G channel, and the third bit is used to represent the information of the B channel. If the first - dimension information of a certain pixel point is 123, it means that its R channel is 1, its B channel is 2, and its G channel is 3. The value of each channel of the RGB channels here needs to be normalized and then rounded. Of course, if the encoding multiple is increased, more corresponding digits can be retained when rounding. Generally speaking, the more encoding digits, the more complete the information.

[0063] The second - dimension information is sequence information, which is actually the position of the corresponding pixel point. In order to represent the position information of each pixel point, in this solution, the linear indexing method or the Hilbert curve is used to describe the second - dimension information.

[0064] S2: Obtain the three - dimensional vector P(x, y, z) of each pixel point in the fusion information, and establish a three - dimensional vector matrix T; Among them, i represents the first - dimension information of the three - dimensional vector x, y represents the second - dimension information of the three - dimensional vector P, and z represents the third - dimension information of the three - dimensional vector P.

[0065] S3: Based on principal component analysis, divide the three - dimensional vector matrix T into several feature regions.

[0066] S3: includes: S31: Take each point in the three - dimensional vector matrix as a query point; for each query point a, traverse the points around it to find v nearest points to form a point set A of the query point. a . Apply PCA to A a for component extraction, to obtain the first - component vector b1, the second - component vector b2, and the third - component vector b3. Among them, b1 < b2 < b3, and take b1 as the normal vector.

[0067] PCA is short for Principal Components Analysis, which is a commonly used data - dimensionality reduction technique. This technique can divide the data into multiple component vectors. In this solution, only the first 3 - proportion component vectors are selected.

[0068] S32: Traverse the vector a formed by the query point a and each point k in the point set A a and project the vector a k along the direction of the first - component vector b1 onto the two - dimensional plane formed by b2 and b3 to obtain the projection vector a k `; Calculate each projection vector a k `; kThe projection angle with b2; preset an angle threshold. If the projection angle generated by the query point a and the point k is greater than the angle threshold, then the point k is used as a boundary point; obtain all the boundary points, use the boundary points as the boundary area d, traverse all the query points to obtain all the boundary areas, and generate a boundary area set D, where d is the index of the boundary area; S33: For the boundary area d, divide it into m0 reference areas, where m0 > 1; diffuse the boundary area d according to the following conditions; (1) The boundary area d must be within the search radius of the reference area m, where m represents the index of the reference area, and d u represents the u-th reference area; (2) The distance from the center point of the boundary area d to the plane where the reference area m is located is less than the set distance threshold σ, and f j and f u are classified into the same type of area; (3) If the reference area m is connected to the rest of the boundary areas, it is not used as a reference area, otherwise it continues to be used as a reference area.

[0069] S34: Obtain the number of three-dimensional vectors in the boundary area after diffusion is completed, and divide it into n categories according to the number of three-dimensional vectors, where n > 3.

[0070] S35: For the boundary areas d of the same category, set the same search radius, and search for the rest of the boundary areas of the same category; if an adjacent boundary area d1 is found, then merge the boundary area d and the boundary area d1; the boundary area d1 represents the l-th boundary area of the same category within the search radius of the boundary area d; If no adjacent boundary area is found, the boundary area d remains independent, and the finally remaining boundary areas are used as the feature areas of the three-dimensional vectors. S36: Preset a point cloud extraction method. According to the average gray value in each feature area, convert the feature area into the average gray value of the high-altitude image data feature area. The larger the average gray value of the feature area, the more the number of pixel points extracted from the corresponding feature area.

[0071] In this way, the three-dimensional high-altitude image data of the target area is converted into point cloud data with three-dimensional coordinates. The key factor of this application is to use the principal component analysis method and the boundary expansion method for the collected three-dimensional information to automatically mark the three-dimensional space model in the three-dimensional high-altitude image.

[0072] The visualization rendering device includes: An information acquisition unit that acquires a water body vector model, a three-dimensional scene model, and water volume prediction data; A rendering information processing unit that generates rendering information based on the water body vector model and the water volume prediction data; A rendering processing unit that defines the vertex data in the three-dimensional scene model and performs matrix transformation to convert the three-dimensional coordinates into two-dimensional screen coordinates; Calculate the final color of the pixels based on the color information and texture information contained in the rendering information to generate a water flow scene map. The rendering process is actually to render the corresponding colors on the areas in the three-dimensional scene model according to the dynamic changes of the water level. Generally, red is used to represent the water body, so the areas submerged by water will be rendered in red.

[0073] The information alarm device obtains rainfall information in real time, generates a hydrological alarm map based on the rainfall information and the water flow scene map, and issues an alarm based on the hydrological alarm map.

[0074] The alarm process here is that based on the rainfall information, the water volume change situation in the target area can be obtained, and based on the water volume change situation, a hydrological monitoring map of the target area is generated, and then alarm information is generated.

[0075] The above description is only some preferred embodiments of the present application and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present application is not limited to the technical solutions formed by the specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) having similar functions disclosed in the embodiments of the present application.

Claims

1. A real-time management system for multiple elements of hydrological monitoring, characterized in that: include: A water body vector information collection device is used to obtain DME data of the target area and construct a water body vector model based on the DME data; A three-dimensional model building device is used to obtain high-altitude image data of the target area and build a three-dimensional scene model based on the high-altitude image data; A visualization rendering device, which renders the water flow into a three-dimensional scene model based on a water body vector model, and generates a flow area of ​​the water body in the three-dimensional scene model; The water body vector information collection device constructs a water body vector model based on the following steps: Step 1: Obtain multiple DME data of the target area, and use the seven-parameter transformation method to transform the coordinate system of the DME data, complete re-cutting and splicing in a unified coordinate system to generate the target DME information; Step 2: Superimpose the river vector of the target area with the DME information, and cut it based on the edge of the target area to generate preliminary water distribution information; Step 3: Rasterize the preliminary water area distribution information and assign water area vectors to each grid area to generate a water area vector map; Step 4: Generate a water body vector model based on the water body vector map.

2. The hydrological monitoring multi-factor real-time management system according to claim 1 is characterized by: Step 1 includes the following steps: Step 11: Obtain DME data of each local area in the target area; Step 12: stitch the DME data based on the SIFT algorithm, and crop the stitched DME data according to the range of the target area; Step 13: Perform dimensionality reduction on the cropped DME data to generate target DME information.

3. The hydrological monitoring multi-factor real-time management system according to claim 2 is characterized by: Step 2 includes the following steps: Step 21: Obtain river vector data of the target area; Step 22: Use QGIS software to perform vector overlay of river vector data and target DME information to generate preliminary water distribution information.

4. The hydrological monitoring multi-factor real-time management system according to claim 3 is characterized by: Step 3 includes the following steps: Step 31: Add a network grid layer of preset specifications to the preliminary water distribution information; Step 32: Set the depression threshold, fill the network grid in the preliminary water area distribution information with data, and obtain the water flow direction information; Step 33: Describe the water flow vector in each grid based on the tangent curvature of the grid to generate a water area vector map.

5. The hydrological monitoring multi-factor real-time management system according to claim 3 is characterized by: Step 33 includes the following steps: Step 331: Calculate the tangent curvature Cp and streamline curvature Dp of the grid p; Step 332: pre-set the water flow convergence divergence value W and the water flow path swing degree threshold R, W>0, R>0; Step 333: If Cp∈(-W, W) and Dp∈(-R, R), then the D8 algorithm is used to determine the flow direction of grid p and its flow distribution ratio to surrounding grids; If Cp∈(-W, W) and Dp∈(-∞,-R]∪[R,+∞), the Rho8 algorithm is used to determine the flow direction of grid p and its flow distribution ratio to the surrounding grids; If Cp∈(-∞, -W]∪[W, +∞), the Dinf algorithm is used to determine the flow direction of grid p and its flow distribution ratio to the surrounding grids; Step 334: According to step 333, the water flow direction and flow distribution ratio of each grid are calculated in turn to generate a water flow direction matrix of the target area to generate a water area vector map.

6. The hydrological monitoring multi-factor real-time management system according to claim 5 is characterized by: The calculation formulas for tangent curvature Cp and streamline curvature Dp are: ； ； Among them, f represents the height function of the terrain surface, which is used to describe the height of the grid; g represents the first-order partial derivative of f with respect to i, which is used to describe the slope of the terrain in the i direction; q represents the slope of f with respect to the j direction, r represents the second-order partial derivative of f with respect to i, s represents the mixed second-order partial derivative of f with respect to i and j, t represents the second-order partial derivative of f with respect to j, i and j represent the horizontal and vertical coordinates of the grid respectively, Represents the derivative symbol.

7. The hydrological monitoring multi-factor real-time management system according to claim 6 is characterized by: Step 4 includes the following steps: Step 41: extracting the flow direction of each grid and its flow distribution ratio to surrounding grids from the water area vector diagram; Step 42: Calculate the sum of the water flow vectors of each grid and assign a flow value according to the sum; Step 43: Generate a water body vector model based on the flow value.

8. The hydrological monitoring multi-factor real-time management system according to claim 1 is characterized by: The visualization rendering device includes: An information acquisition unit, which acquires a water body vector model, a three-dimensional scene model, and water volume prediction data; A rendering information processing unit, which generates rendering information based on a water body vector model and water volume prediction data; A rendering processing unit defines vertex data in a three-dimensional scene model and performs matrix transformation to convert three-dimensional coordinates into two-dimensional screen coordinates; The final color of the pixel is calculated according to the color information and texture information contained in the rendering information to generate a water flow scene graph.

Images