Water conservancy channel editing method based on digital three-dimensional scene
By dynamically creating water conservancy channel models in digital three-dimensional scenes, the problems of low freedom and high maintenance costs in existing technologies are solved, and high-freedom and diversified water conservancy channel editing and the generation of realistic channel models are achieved.
Patent Information
- Application Number
- CN202510684089.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-26
- Publication Date
- 2025-09-12
AI Technical Summary
The existing technology has low freedom in channel model drawing, cannot dynamically draw diverse water conservancy channels, and has high modification and maintenance costs.
A water conservancy channel model is dynamically created through a water conservancy channel editing method based on a digital three-dimensional scene, including channel bottom mesh drawing, RenderTarget updating, terrain layer BOOL operation, terrain noise lerp algorithm processing, model vertex data calculation, model surface subdivision calculation and UV data calculation.
It realizes high-freedom and diversified editing of water conservancy channels, and realistically presents model materials, improves the speed and accuracy of model generation, and enhances the authenticity and coordination of water conservancy project simulation.
Smart Images
Figure CN120635378A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of water conservancy data, and in particular relates to a water conservancy channel editing method based on a digital three-dimensional scene. Background Art
[0002] With the rapid development of information technology, enabled by cutting-edge technologies such as big data, cloud computing, the Internet of Things, and artificial intelligence, the demands facing the field of smart water digital twins are becoming increasingly diverse. From precise monitoring and real-time simulation of water conservancy facilities to optimized decision-making for water resource allocation; from intelligent early warning for flood and drought prevention to precise control of water environment management, the application demand for smart water digital twin technology continues to expand and deepen. Whether it is to improve the operational efficiency of water conservancy projects, enhance the protection of aquatic ecosystems, or address complex and ever-changing water security challenges, smart water digital twins are highly anticipated, and their demand continues to evolve through practical promotion and technological iteration.
[0003] The existing technology for channel model drawing generally uses modeling software to build models according to needs and imports them into a three-dimensional scene for display. It cannot dynamically draw diverse water conservancy channels, has a relatively low degree of freedom, and has a relatively high modification and maintenance cost. Summary of the Invention
[0004] The purpose of the present invention is to provide a water conservancy channel editing method based on a digital three-dimensional scene to solve the problems mentioned in the background technology that the existing simulation software cannot dynamically draw diverse water conservancy channels, has a relatively low degree of freedom, and has a relatively high modification and maintenance cost.
[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:
[0006] The water conservancy channel editing method based on the digital three-dimensional scene includes the following steps:
[0007] Step S1: Draw the channel bottom mesh; mark points according to the required channel model, and connect the marked points into a smooth curve by adding the marked points;
[0008] Step S2, create a RenderTarget and update it to the terrain layer; capture the channel bottom model material data through the scene to the RenderTarget rendering texture;
[0009] Step S3, terrain layer BOOL operation;
[0010] Step S4, terrain noise lerp algorithm is used to perform terrain smoothing;
[0011] Step S5, model vertex data calculation: offset the Bezier points at the bottom of the channel perpendicular to the two vector directions of their respective tangents to generate Bezier surface vertex data on both sides of the channel wall, and draw the model based on these vertex data;
[0012] Step S6, model surface subdivision calculation: Model surface subdivision calculation is performed based on Bezier points and tangents. The core algorithm of surface subdivision is to refine the Bezier curve and add points in the middle to achieve subdivision calculation;
[0013] Step S7, calculating model UV data;
[0014] Step S8: draw the model according to the vertices.
[0015] According to the above technical solution, in step S1, points are marked according to the required channel model, and the marked points are connected into a smooth curve as follows:
[0016] Use keystrokes to locate the segmented points of the channel, connect these points into Bezier curves, and set the position tangent corresponding to each Bezier point. Adjust the curve shape by dragging the points to express the basic shape of the channel and use Bezier curves to draw the channel bottom model.
[0017] According to the above technical solution, the channel bottom model is drawn as follows:
[0018] Traverse the coordinates and tangent of each Bezier point, take the coordinates and tangent of each two adjacent points to generate Bezier model segments, and obtain the scale of each Bezier point to set the model width;
[0019] Set the scale of the Bezier model segment from the start to the end point: Calculate the model's vertices based on two points, and use the vertex subscripts, UV data, and the scale of each point as the model's width to draw the model;
[0020] The model shape is calculated and generated. The generated model serves as the bottom model of the channel and is used as the scene capture source for the next stage RenderTarget.
[0021] According to the above technical solution, construct the RenderTarget rendering texture:
[0022] Photograph the terrain in the entire model, capture the channel model, and render the channel; calculate the texture pixels, convert the model vertices of the terrain, and offset the model vertices according to the color of the pixels. Specifically, the size is X1*Y1, and the texture coordinates are defined as (u1,v1).
[0023] According to the above technical solution, in step S2, the material data of the channel bottom model is captured through the scene and rendered to the RenderTarget texture as follows:
[0024] Update the color data in the RenderTarget rendering texture to the terrain layer for Boolean operation to achieve dynamic terrain modification;
[0025] The algorithm triggers the terrain layer update event through delegation, triggers the RenderTarget scene capture, transfers the model data captured by the RenderTarget to the landscape layer, binds the terrain layer to the event, and after the triggering, stores the model's vertex and fragment metadata in the Debuff, then performs the model Boolean operation to dig out the terrain potholes;
[0026] After the model vertex data in RenderTareget is stored in Debuff, the Debuff data is mapped to the terrain in the terrain layer Tick logic, matrix coordinate conversion is performed, and then a Boolean model subtraction operation is performed; then the terrain height map is subjected to geometric operations to implement the digging logic.
[0027] According to the above technical solution, in step S5, the calculation formula for calculating the model vertex data is specifically:
[0028] Point(x,y,z)+tangent(x*(-1),y,z)=TargetPoint(x,y,z)ORPoint(x,y,z)+tangent(x,y*(-1),z)=TargetPoint(x,y,z)
[0029] In the formula, Point: represents vector point a; Tangent: represents the tangent of point a; TargetPoint: represents the point that is 90 degrees perpendicular to the tangent of point a.
[0030] According to the above technical solution, in step S6, the model surface subdivision calculation is specifically as follows:
[0031] Step S601, obtaining the overall length of Bezier;
[0032] Step S602, setting the subdivision length length, and calculating NUM, where NUM represents the number of segments of the Bezier curve subdivided according to the length length;
[0033] Step S603: traverse the distance between each segment boundary and the Bezier curve according to the calculated value of NUM; Distance = Index * Length is the distance between each traversed point and the Bezier curve;
[0034] Step S604: obtaining the position and tangent line at each distance according to the distance of each segment, and calculating the model vertex, model subscript and UV data according to the subdivided tangent line;
[0035] Step S605: Calling the GPU to perform model rendering according to the calculated data.
[0036] According to the above technical solution, the calculation process of model vertices, model subscripts and UV data is as follows:
[0037] The channel baseline is generated by Bezier curve control points, and the curve interpolation is performed using the De Castellio algorithm;
[0038] Dynamically adjust the tangent direction of the control point (x*(-1), y, z or x, y*(-1), z) to achieve three-dimensional deformation of the curve;
[0039] Generate segmented parameterized mesh: each segment contains vertex buffer (Position / Normal / UV) and index buffer.
[0040] Furthermore, vertex data generation is specifically as follows: Calculate the lateral offset based on the Frenet frame:
[0041] sideOffset=width*normalize(cross(tangent,binormal)).
[0042] The surface is then subdivided using an adaptive subdivision strategy:
[0043] Curvature threshold control: θ = arccos(t1·t2 / |t1||t2|)
[0044] Dynamic subdivision factor calculation: n = cei l (θ / θ_max*maxTessFactor)
[0045] Finally, map the UV data: implement non-uniform parameterization:
[0046] u=cumulativeLength / totalLength
[0047] v=sidePosition / maxWidth.
[0048] According to the above technical solution, the procedural mesh is drawn based on the model vertex data, normals and UV data, specifically:
[0049] Model vertex data is used to draw the model surface: every three vertices form a triangle, and the triangles are stitched together to form the whole model;
[0050] Calculate the normal data of the model and correspond the calculated normal data to the vertex data one by one to indicate the normal corresponding to each vertex of the model;
[0051] The UV data array data also corresponds one-to-one to the vertex data, indicating the corresponding coordinates of each vertex unfolded on the two-dimensional UV data.
[0052] Furthermore, the normal data of the model can be obtained by the cross product of partial derivatives: \mathbf{n}=\frac{\partial\mathbf{r}}{\partial u}\times\frac{\partial\mathbf{r}}{\partial v}).
[0053] Compared with the prior art, the present invention has the following beneficial effects:
[0054] In this paper, the channel bottom mesh is drawn by dotting the desired channel model and connecting them into smooth curves. This accurately outlines the channel bottom shape. Increasing the number of dotted points improves the smoothness of the curve, thereby more closely matching the actual shape of the water conservancy channel bottom and laying a solid foundation for subsequent model construction. The scene captures the channel bottom model material data into a RenderTarget rendering texture, and then creates a RenderTarget and updates it to the terrain layer. This process ensures the accurate presentation of the model material, making the constructed water conservancy channel model more realistic in terms of material texture.
[0055] The method of this invention facilitates the simulation of irrigation, water disaster drills, and disaster relief measures in water conservancy projects by dynamically creating and editing water conservancy channels in a 3D scene without third-party modeling software. This allows for the editing of water conservancy channels with a high degree of freedom and diverse properties. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] Figure 1 Schematic diagram of the flow chart of the water conservancy channel editing method of the present invention. DETAILED DESCRIPTION
[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.
[0058] Example 1
[0059] like Figure 1 As shown, the water conservancy channel editing method based on the digital three-dimensional scene includes the following steps:
[0060] Step S1: Draw the channel bottom mesh; mark points according to the required channel model, and connect the marked points into a smooth curve by adding the marked points;
[0061] Step S2, create a RenderTarget and update it to the terrain layer; capture the channel bottom model material data through the scene to the RenderTarget rendering texture;
[0062] Step S3, terrain layer BOOL operation;
[0063] Step S4, terrain noise lerp algorithm is used to perform terrain smoothing;
[0064] Step S5, model vertex data calculation: offset the Bezier points at the bottom of the channel perpendicular to the two vector directions of their respective tangents to generate Bezier surface vertex data on both sides of the channel wall, and draw the model based on these vertex data;
[0065] Step S6, model surface subdivision calculation: Model surface subdivision calculation is performed based on Bezier points and tangents. The core algorithm of surface subdivision is to refine the Bezier curve and add points in the middle to achieve subdivision calculation;
[0066] Step S7, calculating model UV data;
[0067] Step S8: draw the model according to the vertices.
[0068] In this paper, the channel bottom mesh is drawn by dotting the desired channel model and connecting them into smooth curves. This accurately outlines the channel bottom shape. Increasing the number of dotted points improves the smoothness of the curve, thereby more closely matching the actual shape of the water conservancy channel bottom and laying a solid foundation for subsequent model construction. The scene captures the channel bottom model material data into a RenderTarget rendering texture, and then creates a RenderTarget and updates it to the terrain layer. This process ensures the accurate presentation of the model material, making the constructed water conservancy channel model more realistic in terms of material texture.
[0069] The method of this invention facilitates the simulation of irrigation, water disaster drills, and disaster relief measures in water conservancy projects by dynamically creating and editing water conservancy channels in a 3D scene without third-party modeling software. This allows for the editing of water conservancy channels with a high degree of freedom and diverse properties.
[0070] In this invention, terrain smoothing is performed by performing BOOL operations on the terrain layer and combining them with a terrain noise lerp algorithm. This logical processing of the terrain layer through BOOL operations and the use of the noise lerp algorithm for terrain smoothing allow the water conservancy channel model to blend naturally with the surrounding terrain, avoiding abrupt transitions and improving the authenticity and coordination of the entire digital 3D scene.
[0071] By calculating the offset of the Bezier points at the bottom of the channel perpendicular to the two vector directions of their respective tangents, the Bezier surface vertex data on both sides of the ditch wall is quickly generated and used to draw the model. This method can efficiently determine the key vertex positions of the model, providing the necessary data support for building a complete water conservancy channel model and greatly improving the speed of model generation.
[0072] The model's surface is subdivided based on Bezier points and tangents. The core algorithm involves refining Bezier curves and adding points in between. This refines the model's surface, enriching the waterway model's details and more accurately simulating the complex shapes of actual waterways, improving model accuracy.
[0073] UV data calculation provides the correct coordinate information for mapping the model surface texture. Combined with vertex-based model drawing, it can ensure the correct presentation of the model texture and ultimately construct a complete, accurate and realistic three-dimensional model of the water conservancy channel.
[0074] Example 2
[0075] This embodiment provides a specific implementation method.
[0076] Step 1: On the terrain of the model, mark points according to the required channel model. By pressing the key, try to increase the number of points to connect them into a smooth curve. The overall shape of the channel can be determined according to the shape of the curve. The speed of drawing the model can be optimized by adjusting the distance of the point subdivision. The specific steps include the following:
[0077] Step 101: Use keystrokes to locate the channel segment points, connect these points into Bezier curves, and set the position tangent corresponding to each Bezier point. Adjust the curve shape by dragging the points to express the general shape of the channel and use Bezier curves to draw the channel bottom model.
[0078] Traverse the coordinates and tangent of each Bezier point, take the coordinates and tangent of each two adjacent points to generate Bezier model segments, and obtain the scale at each Bezier point (Bezier curve is divided into segments according to the number of segments, and the model face is calculated according to the curve of each segment. The scale here is used to set the width and height of each model face); used to set the model width, set the scale from the start point to the end point of the Bezier model segment, calculate the model's vertices, vertex subscripts, UV data and the scale of each point as the model width to draw the model; calculate the model shape to generate the model. The generated model is used as the bottom model of the channel and is used as the scene capture source of the next stage RenderTarget.
[0079] Step 102: Construct a RenderTarget rendering texture.
[0080] Photograph the terrain in the entire model, capture the channel model, and render the channel; calculate the texture pixels, convert the model vertices of the terrain, and offset the model vertices according to the color of the pixel points. The size is X1*Y1, and the texture coordinates are defined as (u1,v1).
[0081] Step 2: Capture the channel bottom model material data through the scene to the RenderTarget rendering texture.
[0082] The color data in the RenderTarget rendering texture is updated to the terrain layer for Boolean operations to achieve dynamic terrain modification. The algorithm triggers the terrain layer update event through delegation and triggers the RenderTarget scene capture, and transfers the model data captured by the RenderTarget to the landscape terrain layer. The terrain layer binds the event and triggers it. After the event is triggered, the vertex and fragment metadata of the model (fragment metadata refers to the data contained in each point on the two-dimensional image generated by the primitive through the rasterization process in the computer graphics processing process. This data includes information such as color, depth and texture, and is called a fragment).
[0083] The process of obtaining fragment data: Vertices are combined into primitives: First, geometric vertices are combined into primitives, such as points, line segments or polygons. These primitives are determined by the user and the model he creates.
[0084] Primitives are converted to fragments: After appropriate cropping and adjustments to color and texture data, primitives are converted to fragments through the rasterization process. Rasterization involves projecting the primitives onto the screen and determining the pixels covered by their projected area. Each image point contains data such as color, depth, and texture.
[0085] Fragments are converted to pixels: Finally, fragments need to go through processes such as depth testing to compete to become the final pixels. In this process, fragments contain more information than pixels, such as depth values, normals, texture coordinates, etc. After passing the test, unnecessary information is removed and finally becomes the pixels of the image we see.
[0086] Store it in Debuff, then perform model Boolean operations to dig out holes in the terrain.
[0087] The algorithm works by transferring the color vertex data in the RenderTarget to the terrain layer for height Boolean operation.
[0088] After the model vertex data in RenderTareget is stored in the Debuff, the terrain layer's Tick logic maps the Debuff data to the terrain, performs matrix coordinate conversion, and then performs a Boolean model subtraction operation. The terrain height map is then scaled to implement the pit digging logic.
[0089] Step 3: Offset the Bezier points at the bottom of the channel perpendicular to the two vector directions of their respective tangents to generate Bezier surface vertex data on both sides of the ditch wall. Procedural model drawing is performed based on these vertex data.
[0090] Bezier vertical tangent vector direction calculation formula:
[0091] Point(x,y,z)+tangent(x*(-1),y,z)=TargetPoint(x,y,z)ORPoint(x,y,z)+tangent(x,y*(-1),z)=TargetPoint(x,y,z)
[0092] Step 301: Draw the model using the model vertex data. This algorithm calculates the vertex data of a rectangle using line segments, and then draws the model by calculating the model vertices, vertex subscripts, and UV data coordinates.
[0093] Step 302: Model surface subdivision calculation; Model surface subdivision calculation is performed based on Bezier points and tangents. The core algorithm of surface subdivision is to refine the Bezier curve and add points in the middle to achieve subdivision calculation.
[0094] Specific subdivision steps:
[0095] 1. Get the overall length of Bessel;
[0096] 2. Set the subdivision length, usually 1024 or 2048;
[0097] 3. NUM = Bezier overall length / length;
[0098] 4. According to the value of NUM, traverse from 0;
[0099] 5. Distance = Index * lenght is the distance of each traversed point at Bessel;
[0100] 6. Obtain the position and tangent of each distance according to the distance of each segment;
[0101] 7. Calculate the model vertices, model subscripts and UV data based on the subdivided tangents;
[0102] 8. Call the GPU to render the model based on the calculated data.
[0103] The algorithm draws a procedural mesh based on the model's vertex data, normals, and UV data:
[0104] Model vertex data is used to draw the model surface: every three vertices form a triangle, and these triangles are spliced together to form the entire model;
[0105] Normal data corresponds to vertex data one by one, indicating the normal corresponding to each model vertex;
[0106] The UV data array data also corresponds one-to-one to the vertex data, indicating the corresponding coordinates of each vertex unfolded on the two-dimensional UV data.
[0107] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.
[0108] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.
Claims
1. A water conservancy channel editing method based on a digital three-dimensional scene, characterized by: The following steps are involved: Step S1: Draw the channel bottom mesh; mark points according to the required channel model, and connect the marked points into a smooth curve by adding the marked points; Step S2, create a RenderTarget and update it to the terrain layer; capture the channel bottom model material data through the scene to the RenderTarget rendering texture; Step S3, terrain layer BOOL operation; Step S4, terrain noise lerp algorithm is used to perform terrain smoothing; Step S5, model vertex data calculation: offset the Bezier points at the bottom of the channel perpendicular to the two vector directions of their respective tangents to generate Bezier surface vertex data on both sides of the channel wall, and draw the model based on these vertex data; Step S6, model surface subdivision calculation: Model surface subdivision calculation is performed based on Bezier points and tangents. The core algorithm of surface subdivision is to refine the Bezier curve and add points in the middle to achieve subdivision calculation; Step S7, calculating model UV data; Step S8: draw the model according to the vertices.
2. The water conservancy channel editing method based on digital three-dimensional scene according to claim 1 is characterized by: In step S1, points are marked according to the required channel model, and the marked points are connected into a smooth curve. Specifically: Use keystrokes to locate the segmented points of the channel, connect these points into Bezier curves, and set the position tangent corresponding to each Bezier point. Adjust the curve shape by dragging the points to express the basic shape of the channel and use Bezier curves to draw the channel bottom model.
3. The water conservancy channel editing method based on digital three-dimensional scene according to claim 2 is characterized by: Drawing the channel bottom model is as follows: Traverse the coordinates and tangent of each Bezier point, take the coordinates and tangent of each two adjacent points to generate Bezier model segments, and obtain the scale of each Bezier point to set the model width; Set the scale of the Bezier model segment from the start to the end point: Calculate the model's vertices based on two points, and use the vertex subscripts, UV data, and the scale of each point as the model's width to draw the model; The model shape is calculated and generated. The generated model serves as the bottom model of the channel and is used as the scene capture source for the next stage RenderTarget.
4. The water conservancy channel editing method based on digital three-dimensional scene according to claim 3 is characterized by: Construct a RenderTarget render texture: Photograph the terrain in the entire model, capture the channel model, and render the channel; calculate the texture pixels, convert the model vertices of the terrain, and offset the model vertices according to the color of the pixels. Specifically, the size is X1*Y1, and the texture coordinates are defined as (u1,v1).
5. The water conservancy channel editing method based on digital three-dimensional scene according to claim 4 is characterized by: In step S2, the material data of the channel bottom model is captured through the scene and rendered to the RenderTarget texture as follows: Update the color data in the RenderTarget rendering texture to the terrain layer for Boolean operations to achieve dynamic terrain modification; The algorithm triggers the terrain layer update event through delegation, triggers the RenderTarget scene capture, transfers the model data captured by the RenderTarget to the landscape layer, binds the terrain layer to the event, and after the triggering, stores the model's vertex and fragment metadata in the Debuff, then performs the model Boolean operation to dig out the terrain potholes; After the model vertex data in RenderTareget is stored in Debuff, the Debuff data is mapped to the terrain in the terrain layer Tick logic, matrix coordinate conversion is performed, and then a Boolean model subtraction operation is performed; then the terrain height map is subjected to geometric operations to implement the digging logic.
6. The water conservancy channel editing method based on digital three-dimensional scene according to claim 5 is characterized by: In step S5, the calculation formula for the model vertex data is specifically: Point(x,y,z)+tangent(x*(-1),y,z)=TargetPoint(x,y,z)ORPoint(x,y,z)+tangent(x,y*(-1),z)=TargetPoint(x,y,z) In the formula, Point: represents vector point a; Tangent: represents the tangent of point a; TargetPoint: represents the point 90 degrees perpendicular to the tangent of point a.
7. The water conservancy channel editing method based on digital three-dimensional scene according to claim 6 is characterized by: In step S6, the model surface subdivision calculation is specifically as follows: Step S601, obtaining the overall length of Bezier; Step S602, setting the subdivision length length, and calculating NUM, where NUM represents the number of segments of the Bezier curve subdivided according to the length length; Step S603: traverse the distance between each segment boundary and the Bezier curve according to the calculated value of NUM; Distance = Index * Length is the distance between each traversed point and the Bezier curve; Step S604: obtaining the position and tangent line at each distance according to the distance of each segment, and calculating the model vertex, model subscript and UV data according to the subdivided tangent line; Step S605: Calling the GPU to perform model rendering according to the calculated data.
8. The water conservancy channel editing method based on digital three-dimensional scene according to claim 7 is characterized by: The calculation process of model vertices, model subscripts and UV data is as follows: The channel baseline is generated by Bezier curve control points, and the curve interpolation is performed using the De Castellio algorithm; Dynamically adjust the tangent direction of the control point (x*(-1), y, z or x, y*(-1), z) to achieve three-dimensional deformation of the curve; Generate segmented parameterized mesh: each segment contains vertex buffer (Position / Normal / UV) and index buffer.
9. The water conservancy channel editing method based on digital three-dimensional scene according to claim 7, characterized in that: Draw a procedural mesh based on the model's vertex data, normals, and UV data. Specifically: Model vertex data is used to draw the model surface: every three vertices form a triangle, and the triangles are stitched together to form the whole model; Calculate the normal data of the model and correspond the calculated normal data to the vertex data one by one to indicate the normal corresponding to each vertex of the model; The UV data array data also corresponds one-to-one to the vertex data, indicating the corresponding coordinates of each vertex unfolded on the two-dimensional UV data.