Method and system for quickly generating three-dimensional white pattern of building

Abstract

The invention relates to the technical field of three-dimensional modeling, and particularly discloses a method and system for quickly generating a building three-dimensional white model, and the method comprises the following steps: obtaining an image region range, and determining a target region in a manner of manually dragging a base map of a map and inputting a place name or latitude and longitude coordinates by a user; extracting the contour of the building and obtaining space coordinate information, analyzing the remote sensing image of the target area by using an AI image recognition algorithm, automatically extracting the contour information of the building, and reading the corresponding space coordinate data to obtain the contour data of the building; and cleaning the contour data, endowing the contour data with a height value, and cleaning the extracted building contour data. Through an AI image recognition algorithm and an automatic data processing flow, the generation time of the building three-dimensional white mold is greatly shortened, and the problem that a traditional method is long in consumed time is solved.

Description

Technical Field

[0001] This invention relates to a method for generating a 3D white model of a building, specifically a method and system for rapidly generating a 3D white model of a building, belonging to the field of 3D modeling technology. Background Technology

[0002] With the rapid development of remote sensing technology and artificial intelligence, the technology of automatically extracting building outlines from remote sensing images has been widely applied. By identifying remote sensing image maps, AI spatial image recognition algorithms can be used to extract the relevant outlines of buildings and read their spatial coordinate information, generating building outline layer data. However, in the traditional field of building white model generation, it usually relies on manually drawing building outlines, assigning height values ​​to each outline, and then using third-party software to generate white model models. This process is not only time-consuming and labor-intensive, but also difficult to meet the needs of large-scale rapid modeling. Although existing technologies can automatically extract building outline information from remote sensing images, they have significant shortcomings: they cannot directly generate white model models from the extracted outlines, still requiring conversion operations using third-party software, resulting in low efficiency and complex operation. In addition, existing technologies lack support for texture information when generating white model models, limiting the visualization effect and application scope of the models. Therefore, there is an urgent need for a method and system that can quickly generate 3D white models of buildings based on remote sensing images to solve the problems of low efficiency, cumbersome operation, and functional limitations in existing technologies. This invention aims to automatically extract and clean the outline of buildings through AI image recognition algorithms, and directly generate white model with texture information, thereby greatly improving modeling efficiency and practicality. Summary of the Invention

[0003] To address the problems in existing technologies, this invention provides a method and system for rapidly generating three-dimensional white models of buildings.

[0004] The objective of this invention can be achieved through the following technical solutions: A method for rapidly generating 3D white models of buildings includes the following steps: To obtain the image area range, users can determine the target area by manually dragging the map base, entering the place name, or latitude and longitude coordinates; The outline of the building is extracted and its spatial coordinate information is obtained. The remote sensing image of the target area is analyzed using AI image recognition algorithm to automatically extract the outline information of the building and read its corresponding spatial coordinate data to obtain the building outline data. Clean the outline data and assign height values. Clean the extracted building outline data to remove noise and error information, and assign fixed or random height values ​​to the cleaned outline data to generate building outline layer data. Generate a white model and set the texture. Based on the generated building outline layer data, read and set the texture information to generate a three-dimensional white model. By overlaying white models onto the map, the distribution and shape of buildings can be visually displayed.

[0005] Optionally, the acquired image area range includes a multi-scale zoom function, allowing users to flexibly locate area ranges of different sizes.

[0006] Optionally, the multi-scale zoom function supports selecting the target area by drawing a rectangle or polygon.

[0007] Optionally, the AI ​​image recognition algorithm employs a convolutional neural network structure and includes an edge detection module to accurately locate the building outline boundary.

[0008] Optionally, the edge detection module incorporates spatial coordinate system correction technology to ensure that the extracted building outline information has high-precision spatial coordinate attributes.

[0009] Optionally, the process of cleaning the contour data includes using morphological filtering algorithms to remove isolated points and fine noise, and using contour closure algorithms to repair incomplete contour boundaries.

[0010] Optionally, the cleaning process also includes a consistency verification mechanism for removing abnormal contour data.

[0011] Optionally, the texture settings function supports tiling, projection, and UV mapping modes, and provides a texture library for users to select from.

[0012] A system for rapidly generating 3D white models of buildings, the system being used in the method described above, and the system further includes the following modules: The region selection module is used to receive the place name, latitude and longitude coordinates, or manual map dragging operation from the user to determine the target region range; The image recognition module, based on AI image recognition algorithms, extracts building outline information from remote sensing images and obtains their spatial coordinate data; The data processing module cleans and assigns height values ​​to the extracted building outline data, generating a building outline layer. The model generation module generates a 3D white model based on the building outline layer data and supports texture setting functions; The model display module overlays the generated 3D white model onto the map, providing multiple view modes and parameter setting options for easy viewing and analysis by users.

[0013] The beneficial effects of this invention are: By employing AI image recognition algorithms and automated data processing workflows, the generation time for 3D white models of buildings is significantly reduced, solving the problem of time-consuming traditional methods. Data cleaning and height value assignment steps effectively improve model accuracy, while texture setting functions further enhance the model's realism and visualization. The various region selection methods and model display options provided by this invention allow users to flexibly adjust parameters according to their needs to obtain optimal modeling and display effects. In the field of geographic information systems, users can quickly generate 3D white model images of a city using this invention, and combine them with texture setting functions to generate realistic urban landscapes, providing important references for urban planning and management. Attached Figure Description

[0014] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0015] Figure 1 This is a flowchart of the method for rapidly generating a 3D white model of a building according to the present invention; Figure 2 This is a schematic diagram of the interface for determining the region outline range and reading the outline information of remote sensing images in this invention. Figure 3 This is a schematic diagram of the display interface for overlaying a white model onto a map according to the present invention. Detailed Implementation

[0016] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Please see Figure 1-3 As shown, a method and system for rapidly generating 3D white models of buildings are presented. The overall process of this method includes acquiring the image area range, extracting building outline information, cleaning the data and assigning height values, generating the white model, and overlaying the model onto a map. In practical applications, users determine the target area range by manually dragging the map base, entering a place name, or using latitude and longitude coordinates. The manual dragging of the map base is implemented based on the map interaction interface, allowing users to adjust the view position using a mouse or touchscreen to select the target area. When a user enters a place name, the system calls a geocoding service to convert the place name into the corresponding geographic coordinate range; when latitude and longitude coordinates are entered, the system directly parses the coordinate information and locates the target area. The target area selection method supports multi-scale zoom functionality, ensuring that users can flexibly locate areas of different sizes. This function is particularly important in practical scenarios; for example, urban planners can use this function to quickly select a specific block as the target area.

[0018] After identifying the target area, the system proceeds to the building outline extraction stage. (See also...) Figure 2 The AI ​​image recognition algorithm analyzes remote sensing images of selected areas, automatically extracting building outlines and their corresponding spatial coordinates. This algorithm employs a convolutional neural network structure, classifying pixel features in the remote sensing image to distinguish buildings from other land cover types. It includes an edge detection module for accurately locating building outline boundaries, and combines spatial coordinate system correction technology to ensure the extracted outlines have high-precision spatial coordinate attributes. In practice, users can view the extraction results through the interface and fine-tune the outline information as needed. For example, in complex scenes, such as dense building clusters or areas with obstructions, users can use the bounding box tool to edit and clean the identified outlines, correcting incorrect areas and further improving data accuracy.

[0019] After contour extraction, the system enters the data cleaning and height value assignment stage. The cleaning process includes several sub-steps: first, morphological filtering algorithms are used to remove isolated points and minor noise; then, contour closure algorithms are used to repair incomplete contour boundaries; and finally, a consistency check mechanism is employed to remove abnormal contour data. The cleaned contour data is assigned either a fixed height value or a random height value. The fixed height value is specified by user-preset parameters, while the random height value is dynamically generated based on a user-defined height range. The height value assignment process incorporates a building type classification algorithm, allocating reasonable height ranges according to different building types to enhance the model's realism. For example, in urban centers, the height value of high-rise buildings is typically set to over 30 meters, while in suburban or rural areas, the height value of low-rise buildings is set to under 10 meters. This height value assignment strategy significantly improves the model's visual appeal and credibility in practical applications.

[0020] The system reads and sets texture information based on the generated building outline layer data to generate a 3D white model. The texture setting function supports multiple texture mapping modes, including tiling, projection mapping, and UV mapping. The system provides a texture library for users to choose from, containing surface textures of common building materials such as brick walls, glass curtain walls, and concrete. Users can also customize texture parameters, such as color, reflectivity, and roughness, to meet personalized needs. In practice, users can intuitively adjust texture parameters through the interface and view the changes in the model in real time. For example, in the field of architectural design, users can set different textures for different types of buildings according to the design scheme, thereby generating more realistic 3D white models.

[0021] After generating the white model, the system overlays it onto the map, visually displaying the distribution and form of buildings. The model display module supports multiple view modes, including top view, side view, and perspective view, which users can switch between via the interface. The system provides parameter setting options, allowing users to adjust parameters such as lighting effects, sky background, and shadow intensity to optimize the model's display effect. The model overlay process uses real-time rendering technology to ensure seamless integration between the model and the map base map. In practical applications, users can rotate, zoom, and move the model's viewpoint using interactive functions to observe the distribution and form of buildings from different angles. For example, in urban planning projects, users can use this function to quickly assess the building density and layout rationality of a specific area.

[0022] This invention also provides a system for rapidly generating 3D white models of buildings, including a region selection module, an image recognition module, a data processing module, a model generation module, and a model display module. The region selection module receives user input of place names, latitude and longitude coordinates, or manual map dragging to determine the target area. This module integrates a geographic information system interface, supporting multi-source map data access to ensure accurate target area positioning. The image recognition module, based on AI image recognition algorithms, extracts building outline information from remote sensing imagery and obtains its spatial coordinate data. This module adopts a distributed computing architecture, supporting parallel processing of large-scale remote sensing imagery and significantly improving outline extraction efficiency. The data processing module cleans and assigns height values ​​to the extracted building outline data, generating a building outline layer. The model generation module generates a 3D white model based on the building outline layer data and supports texture setting functions. The model display module overlays the generated 3D white model onto a map, providing multiple view modes and parameter setting options for easy user viewing and analysis.

[0023] This invention presents an AI-based algorithm for automatically extracting ground feature outlines. By employing deep learning to analyze pixel features in remote sensing imagery, this algorithm achieves high-precision extraction of building outlines in any region, avoiding the inefficiency of traditional methods that rely on manual drawing. In practical applications, this algorithm can handle various complex scenes, such as high-density building clusters, vegetated areas, and regions with significant terrain undulations. This invention optimizes the data cleaning and height value assignment processes. Through multi-level cleaning and intelligent height value assignment of the extracted building outline data, it improves data accuracy and usability, providing high-quality foundational data for subsequent model generation. Furthermore, this invention introduces multi-modal texture mapping technology during texture setting and model generation, making the generated 3D white model more realistic and visually appealing.

[0024] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method of rapidly producing a three-dimensional white model of a building, characterized by, The method comprises the following steps: Obtaining the image area range, the user determines the target area by manually dragging the map base map, inputting the place name or latitude and longitude coordinates; Extracting the building contour and obtaining the spatial coordinate information, using AI image recognition algorithm to analyze the remote sensing image of the target area, automatically extracting the contour information of the building, and reading the corresponding spatial coordinate data to obtain the building contour data; Cleaning the contour data and assigning a height value, cleaning the extracted building contour data to remove noise and error information, and assigning a fixed height value or a random height value to the cleaned contour data to generate building contour layer data; Generating a white model and setting the texture, based on the generated building contour layer data, reading and setting the texture information to generate a three-dimensional white model; Superimposing the white model on the map to intuitively show the distribution and morphology of the building.

2. The method of claim 1, wherein, The image area range acquisition includes a multi-scale zoom function for flexible positioning of different sizes of the area range.

3. The method of claim 2, wherein, The multi-scale zoom function supports drawing a rectangular or polygonal frame to select the target area.

4. The method of claim 1, wherein, The AI image recognition algorithm uses a convolutional neural network structure and includes an edge detection module to accurately locate the building contour boundary.

5. The method of claim 4, wherein, The edge detection module combines a spatial coordinate system correction technique to ensure that the extracted building contour information has high-precision spatial coordinate properties.

6. The method of claim 1, wherein, The contour data cleaning process includes using a morphological filtering algorithm to remove isolated points and noise, and using a contour closure algorithm to repair incomplete contour boundaries.

7. The method of claim 6, wherein, The cleaning process also includes a consistency verification mechanism to eliminate abnormal contour data.

8. The method of claim 1, wherein, The texture setting function supports tiling, projection mapping and UV mapping modes, and provides a texture library for user selection.

9. A system for rapid production of a three-dimensional white model of a building, characterized by The system is used for the method of the above claims 1-8, and the system further comprises the following modules: A region selection module for receiving user input of place names, latitude and longitude coordinates or manual map dragging operations to determine the target area range; An image recognition module based on AI image recognition algorithm to extract building contour information from remote sensing images and obtain its spatial coordinate data; A data processing module for cleaning and assigning height values to the extracted building contour data to generate building contour layers; A model generation module for generating a three-dimensional white model based on the building contour layer data and supporting texture setting function; A model display module for superimposing the generated three-dimensional white model on the map, providing multiple view modes and parameter setting options for user viewing and analysis.

Images