A method and system for capturing and quickly modeling urban three-dimensional space parameters
By combining aerial photography with clustering algorithms and fluid dynamics software, the problem of long simulation time and high cost in urban physical environment simulation was solved. This enabled the rapid acquisition and modeling of urban three-dimensional spatial parameters, reducing simulation costs and improving prediction accuracy.
Patent Information
- Application Number
- CN202411306837.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-19
- Publication Date
- 2026-02-27
- Estimated Expiration
- 2044-09-19
AI Technical Summary
Existing urban physical environment simulation methods are time-consuming and costly in building high-precision geometric models and mesh generation, leading to increased simulation prediction expenses and making it difficult to achieve rapid modeling and sustainable development of the urban environment.
By remotely controlling unmanned aerial vehicles equipped with high-definition infrared imaging cameras to conduct aerial photography, detailed aerial images of the city are obtained. Clustering algorithms and depth recognition technology are used to capture the three-dimensional spatial parameters of the city. Combined with fluid dynamics software, rapid modeling and simulation are performed, reducing the cost of simulating the urban physical environment.
It enables rapid construction of complex urban geometric models and efficient simulation of physical environments, significantly reducing simulation costs, ensuring that the prediction accuracy of global environmental parameters is less than 10%, and supporting fine-grained prediction at arbitrary locations and heights.
Smart Images
Figure CN119339011B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of urban physical environment simulation technology, specifically relating to a method and system for capturing and rapidly modeling urban three-dimensional spatial parameters based on aerial images. Background Technology
[0002] Rapid urbanization, accompanied by the diversification and densification of urban geometries, leads to uneven distribution of urban physical environmental parameters. This can easily result in environmental problems such as heat waves and air pollution, threatening the health of urban residents (e.g., increasing the incidence and even mortality of cardiovascular and respiratory diseases) and drastically increasing energy consumption and carbon emissions, thus impacting sustainable urban development. Therefore, exploring the relationship between urban morphology and urban physical environmental parameters is of great significance for improving urban environmental issues.
[0003] Computational fluid dynamics-based simulation prediction is a commonly used technique for exploring the relationship between urban morphology and urban physical environment parameters. This method includes steps such as geometric model construction, spatial mesh generation, defining boundary conditions, and calculating environmental parameters. Given the complex and dense nature of urban geometry, existing simulation prediction methods often face challenges such as the time-consuming and costly process of building high-precision geometric models and mesh generation. For example, a detailed simulation of the full-scale physical environment of a city requires at least 1-3 months, significantly increasing the cost of simulation prediction.
[0004] Therefore, how to efficiently capture urban three-dimensional spatial parameters and build a rapid modeling system is of great significance for reducing the cost of urban physical environment simulation and realizing the sustainable development of urban environment. Summary of the Invention
[0005] The purpose of this invention is to provide a method and system for capturing and rapidly modeling three-dimensional spatial parameters of cities, so as to realize the rapid construction of complex urban geometric models and the rapid simulation of physical environment, which helps to reduce the cost of urban physical environment simulation.
[0006] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution:
[0007] A method for capturing and rapidly modeling three-dimensional spatial parameters of a city includes the following steps:
[0008] The S100 is a remotely controlled unmanned aerial vehicle equipped with an infrared imaging high-definition camera to conduct aerial photography of the entire target city and obtain detailed aerial images of the city.
[0009] S200. The acquired detailed aerial photograph of the city is processed to obtain the three-dimensional spatial parameters of the city, including city density data and city height data.
[0010] S300, using a clustering algorithm, classifying the obtained city density data and city height data, so as to obtain a city horizontal dimension region division result and a city vertical dimension level division result;
[0011] S400, based on the obtained city horizontal dimension region division result and city vertical dimension level division result, performing depth recognition on the city fine aerial image, realizing low-dimensional processing of the complex geometric morphology of the city, so as to obtain an average geometric morphology height of the city;
[0012] S500, based on the city horizontal dimension region division result, the city vertical dimension level division result and the average geometric morphology height of the city, quickly constructing a city three-dimensional space geometric model of the target city;
[0013] S600, using the constructed city three-dimensional space geometric model, predicting a city physical environment high-resolution distribution result of the target city, and displaying the prediction result.
[0014] Further, in step S100, the city fine aerial image includes: a city digital surface model, a digital ground model and a corresponding aerial thermal image.
[0015] Further, in step S200, the processing method of the city fine aerial image is,
[0016] S201, performing image segmentation and classification processing on the aerial thermal image, so as to extract the city geometric morphology and the structure shadow of the target city;
[0017] S202, further combining the digital surface model and the digital ground model to realize efficient capture of city density data and city height data.
[0018] Further, in step S300, the method for classifying the city density data and the city height data is,
[0019] S301, first, clustering the city density data (x and y directions), and performing city horizontal dimension region division according to the city layout density;
[0020] S302, then, further clustering the height data (z direction) based on the city region division, so as to realize city vertical dimension level division;
[0021] S303, finally, obtaining a clustering result of arrangement category m and height category n, wherein the arrangement category m represents the city horizontal dimension region division result, and the height category n represents the city vertical dimension level division result.
[0022] Further, in step S400, the method for low-dimensional processing of the complex geometry of the city is:
[0023] S401, according to the obtained city horizontal dimension region division result and city vertical dimension level division result, select a color and its corresponding RGB value to represent a clustering cluster, that is, there are m×n colors in the clustering result;
[0024] S402, for the division area with dense building arrangement, divide the area into small blocks based on small block size;
[0025] For the division area with sparse building arrangement, divide the area into small blocks based on medium block size;
[0026] S403, identify and count the pixel RGB value corresponding to each small block in the dense and sparse building arrangement division area, calculate the proportion of each color in each small block, and take the geometry height of the block corresponding to the maximum proportion as the average geometry height of the block.
[0027] Further, in step S500, the method for quickly building a city three-dimensional space geometry model is:
[0028] S501, collect the city horizontal dimension region division result, the city vertical dimension level division result and the city average geometry height, and use them as city three-dimensional space geometry model data for modeling;
[0029] S502, use modal recognition algorithm to extract and normalize the data features of the city three-dimensional space geometry model data, thereby constructing a homogenization data set and uploading to a cloud service platform;
[0030] S503, encode the city three-dimensional space geometry model data, implant finite volume grid algorithm, and further combine fluid mechanics (CFD) calculation software to realize city geometry adaptive and fast modeling.
[0031] Further, in step S600, the prediction method of the city physical environment is:
[0032] S601, use the built city three-dimensional space geometry model and fluid mechanics (CFD) numerical simulation software to quickly predict the city physical environment of the target city and obtain the prediction result;
[0033] S602, compare the prediction result with the city digital surface and ground thermal image capture result; when the error exceeds the set percentage threshold, cluster and divide the city three-dimensional space again and predict the physical environment until the error is lower than the set percentage threshold;
[0034] S603, finally, the urban physical environment prediction result is visualized and displayed.
[0035] A city three-dimensional space parameter capturing and rapid modeling system comprises:
[0036] The unmanned aerial vehicle is provided with an infrared imaging high-definition camera and can be remotely controlled, is responsible for remote aerial photography of a target city or region, and collects fine aerial photography of the target city or region.
[0037] The geometric model data processing integrated system is responsible for rapid modeling of the geometric morphology of the target city or region and physical environment simulation of the target city or region by using the above-mentioned city geometric morphology rapid modeling and physical environment simulation corresponding method.
[0038] A computer device comprises a processor, a memory, a communication interface and a communication bus, the processor, the memory and the communication interface complete communication with each other through the communication bus, the memory is used for storing at least one executable instruction, and the executable instruction makes the processor execute the operation corresponding to the above-mentioned city three-dimensional space parameter capturing and rapid modeling method.
[0039] A computer readable storage medium, the computer storage medium stores at least one executable instruction, and the executable instruction makes the processor execute the operation corresponding to the above-mentioned city three-dimensional space parameter capturing and rapid modeling method.
[0040] The beneficial effects of the present application are:
[0041] 1. The present application provides a city complex three-dimensional space rapid modeling system, which significantly reduces the city global scale simulation prediction cost by low-dimensional data processing of complex geometric morphology.
[0042] 2. The present application realizes high-resolution physical environment visualization and dynamic display based on the rapid modeling system, can ensure that the global environment parameter prediction accuracy is less than 10%, and can provide fine prediction data at any horizontal position and vertical height of the city in real time.
[0043] 3. The city three-dimensional space parameter capturing and rapid modeling method and system provided by the present application are suitable for rapid construction of three-dimensional space geometric models of cities of any scale and rapid simulation of physical environment, have universality, and have important significance for reducing the city physical environment simulation cost and realizing sustainable development of the city environment.
[0044] The above description is only a summary of the technical scheme of the present application, in order to more clearly understand the technical means of the application, and the content of the specification can be implemented, the following is a preferred embodiment of the present application and the detailed description of the drawings. The specific embodiments of the present application are given in detail by the following examples and their drawings. Attached Figure Description
[0045] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:
[0046] Figure 1 This is a schematic diagram of the system framework and process flow of the present invention.
[0047] Figure 2 This is the clustering result of urban building morphology obtained by the method of the present invention.
[0048] Figure 3 This refers to the average geometric height result obtained in the method of this invention.
[0049] Figure 4 This refers to the urban physical environment prediction results obtained using the method of this invention. Detailed Implementation
[0050] The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings to provide a clearer understanding of the invention's purpose, features, and advantages. It should be understood that the embodiments shown in the drawings are not intended to limit the scope of the invention, but are merely illustrative of the essential spirit of the invention's technical solution.
[0051] In the following description, certain specific details are set forth for the purpose of illustrating various disclosed embodiments in order to provide a thorough understanding of the various disclosed embodiments. However, those skilled in the art will recognize that embodiments may be practiced without one or more of these specific details. In other instances, well-known apparatuses, structures, and techniques associated with this application may not have been shown or described in detail to avoid unnecessarily obscuring the description of the embodiments.
[0052] Unless the context requires otherwise, throughout the specification and claims, the word “comprising” and its variations, such as “including” and “having”, shall be understood to have an open, inclusive meaning, that is, to be interpreted as “including, but not limited to”.
[0053] Throughout this specification, references to "an embodiment" or "an embodiment" indicate that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. Therefore, the appearance of "in an embodiment" or "an embodiment" in various places throughout the specification does not necessarily refer to the same embodiment. Furthermore, a particular feature, structure, or characteristic may be combined in any manner in one or more embodiments.
[0054] As used in this specification and the appended claims, the singular forms "a," "an" and "the" include plural referents unless the content clearly dictates otherwise. It should be noted that the term "or" is generally employed in its sense including "and / or" unless the content clearly dictates otherwise.
[0055] Moreover, the technical features involved in the different embodiments of the application described below can be combined with each other as long as there is no conflict.
[0056] Referring to Figure 1 As shown in the drawings, a method for capturing parameters of urban three-dimensional space and rapid modeling comprises the following steps:
[0057] S100, remotely control the unmanned aerial vehicle 1 carrying an infrared imaging high-definition camera 2 to take aerial photographs of the target city overall space, and obtain fine aerial photographs of the city.
[0058] The fine aerial photographs of the city include a digital surface model of the city, a digital ground model, and corresponding aerial thermal images.
[0059] S200, process the obtained fine aerial photographs of the city to obtain urban three-dimensional space parameters, which include urban density data and urban height data. The specific method is:
[0060] The aerial thermal image is subjected to image segmentation and classification processing, so as to extract the geometric shape and structural shadow of the target city, and then the digital surface model and the digital ground model are combined to efficiently capture the urban density data and the urban height data.
[0061] S300, use a clustering algorithm to classify the obtained urban density data and urban height data to obtain urban horizontal dimension regional division results and urban vertical dimension level division results. The specific method is:
[0062] First, the urban density data (x and y directions) is clustered to divide the urban horizontal dimension regions according to the urban layout density; then, on the basis of the urban regional division, the height data (z direction) is further clustered to realize the urban vertical dimension level division; finally, the clustering results of the arrangement category m and the height category n are obtained, wherein the arrangement category m represents the urban horizontal dimension regional division result, and the height category n represents the urban vertical dimension level division result.
[0063] The specific calculation formula is as follows:
[0064] Given data samples:
[0065] ;
[0066] (2) Select the number of clustering clusters based on the sum of squared errors SSE:
[0067] ;
[0068] (3) Select the initial clustering center based on the shortest distance between the sample data and the centroid;
[0069] (4) Calculate the distance between the sample data and each clustering center, and assign it to the nearest clustering cluster:
[0070] ;
[0071] (5) Calculate the average value of each clustering sample data to get the new clustering center;
[0072] (6) Repeat steps (4) and (5) until the sample data assignment remains unchanged.
[0073] S400, based on the obtained city horizontal dimension region division result and city vertical dimension level division result, the city fine aerial image is recognized in depth, the city complex geometric shape low dimension processing is realized, and the city average geometric shape height is obtained. The specific method is,
[0074] According to the obtained city horizontal dimension region division result and city vertical dimension level division result (arrangement category m, height category n), select a color and its corresponding RGB value to represent a clustering cluster, that is, there are m×n colors in the clustering result;
[0075] For the division area with dense building arrangement, small block division is performed on the area with small block size (such as 180m) as the reference; for the division area with sparse building arrangement, small block division is performed on the area with medium block size (such as 400m) as the reference;
[0076] Identify the pixel RGB value corresponding to each small block in the dense and sparse building arrangement division area and count it to calculate the proportion of each color in each small block, and take the geometric shape height of the block corresponding to the maximum proportion as the average geometric shape height of the block.
[0077] S500, based on the city horizontal dimension region division result, the city vertical dimension level division result and the city average geometric shape height, quickly construct the city three-dimensional space geometric model of the target city. The specific method is,
[0078] Collect the city horizontal dimension region division result, the city vertical dimension level division result and the city average geometric shape height, and use them as city three-dimensional space geometric model data for modeling;
[0079] The data characteristics of the urban three-dimensional space geometric model data are extracted and normalized by using a modal recognition algorithm, so as to construct a homogenized data set and upload a cloud service platform.
[0080] The urban three-dimensional space geometric model data is coded and converted, and is implanted into a finite volume grid algorithm, and further combined with an OpenFOAM open source software platform, so as to realize adaptive and rapid modeling of urban geometric morphology.
[0081] S600, using the constructed urban three-dimensional space geometric model, the high-resolution distribution result of the urban physical environment of the target city is predicted, and the prediction result is displayed. The specific method is,
[0082] The urban three-dimensional space geometric model and the fluid mechanics (CFD) numerical simulation software are used to quickly predict the urban physical environment of the target city, and the prediction result is obtained; the prediction result is compared with the error of the urban digital surface and the ground thermal image capture result; when the error exceeds the set percentage threshold (such as 10%), the urban three-dimensional space is clustered and divided again and the physical environment is predicted, until the error is less than the set percentage threshold (such as 10%); finally, the urban physical environment prediction result is visualized and displayed.
[0083] The following is an example of Nanjing Xijiekou and surrounding areas, referring to the method of the application, the following steps are implemented:
[0084] 1) Remote control of the unmanned aerial vehicle carrying an infrared imaging high-definition camera to take aerial photographs of Nanjing Xijiekou and surrounding areas, obtain fine aerial images, perform image segmentation and classification on the fine aerial images, extract geometric morphology and its shadow, and obtain building height and density data of Nanjing Xijiekou and surrounding areas.
[0085] 2) Cluster the density data, divide the area in the horizontal dimension, and further cluster the height data on this basis, divide the level in the vertical dimension, refer to Figure 2 , and finally obtain the clustering result of arrangement category m and height category n.
[0086] 3) Identify the RGB values of the three clustering colors in the obtained clustering result, calculate the color proportion of each area, refer to Figure 3 , and take the geometric morphology height in the block corresponding to the maximum proportion value as the average geometric morphology height of the block.
[0087] 4) Based on the urban area division result in step 2) and the average geometric morphology height in step 3), quickly construct a three-dimensional space geometric morphology model of the city, and further predict the high-resolution distribution result of the urban physical environment, and refer to Figure 4As shown, the prediction result is displayed.
[0088] The application also provides a city three-dimensional space parameter capturing and rapid modeling system, comprising: an unmanned aerial vehicle 1 carrying an infrared imaging high-definition camera 2 and being remotely controllable, and a geometric model data processing integrated system.
[0089] The unmanned aerial vehicle carrying an infrared imaging high-definition camera and being remotely controllable is responsible for remote aerial photography of a target city or region and collection of fine aerial photography of the target city or region.
[0090] The geometric model data processing integrated system is responsible for rapid modeling of the city geometric shape and physical environment simulation of the target city or region by using the above-mentioned corresponding method.
[0091] The application also provides a computer device, comprising: a processor, a memory, a communication interface and a communication bus, the processor, the memory and the communication interface complete mutual communication through the communication bus, the memory is used for depositing at least one executable instruction, the executable instruction makes the processor execute the operation corresponding to the above-mentioned city three-dimensional space parameter capturing and rapid modeling method.
[0092] The application also provides a computer readable storage medium, the computer storage medium has at least one executable instruction, the executable instruction makes the processor execute the operation corresponding to the above-mentioned city three-dimensional space parameter capturing and rapid modeling method.
[0093] The above only is the preferred embodiment of the application and is not used for limiting the application, for the person skilled in the art, the application can have various changes and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application should be included in the protection scope of the application.
Claims
1. A method for capturing and rapidly modeling three-dimensional spatial parameters of a city, characterized in that, Includes the following steps: The S100 is a remotely controlled unmanned aerial vehicle equipped with an infrared imaging high-definition camera to conduct aerial photography of the entire target city and obtain detailed aerial images of the city. The detailed urban aerial images include: urban digital surface models, digital ground models, and their corresponding aerial thermal images; S200. The acquired detailed aerial photograph of the city is processed to obtain the three-dimensional spatial parameters of the city, including city density data and city height data. The processing method for the refined aerial photographs of the city is as follows: S201. Perform image segmentation and classification processing on the aerial thermal image to extract the urban geometry and structural shadows of the target city. S202. Then, by combining the digital surface model and the digital ground model, efficient acquisition of urban density data and urban height data is achieved; S300. Using clustering algorithms, the obtained urban density data and urban height data are classified to obtain the results of urban horizontal dimension regional division and urban vertical dimension level division. The method for classifying the city density data and the city height data is as follows: S301. First, cluster the urban density data and divide the urban areas into horizontal dimensions according to the density of the urban layout. S302. Then, based on the urban area division, further cluster the height data to achieve the vertical dimension level division of the city; S303. Finally, the clustering results of layout category m and height category n are obtained, where layout category m represents the city's horizontal dimension region division result and height category n represents the city's vertical dimension level division result. S400. Based on the obtained urban horizontal dimension region division results and urban vertical dimension level division results, perform depth recognition on the refined aerial image of the city to achieve low-dimensional processing of the complex geometric shape of the city, thereby obtaining the average geometric shape height of the city. The method for reducing the dimensionality of complex urban geometries is as follows: S401. Based on the obtained urban horizontal dimension region division results and urban vertical dimension level division results, select a color and its corresponding RGB value to represent a cluster, that is, there are a total of m×n colors in the clustering results; S402. For densely populated areas, the area shall be divided into smaller blocks based on the size of a small street block. For sparsely populated areas, the area is divided into smaller blocks based on the size of a medium-sized street block. S403. Identify and count the pixel RGB values corresponding to each small block in the area divided by dense and sparse buildings, calculate the proportion of each color in each small block, and take the geometric height of the block corresponding to the maximum proportion as the average geometric height of the block. S500: Based on the city's horizontal dimension region division results, the city's vertical dimension level division results, and the city's average geometric shape height, quickly construct the city's three-dimensional spatial geometric model. S600: Using the constructed three-dimensional spatial geometric model of the city, predict the high-resolution distribution of the urban physical environment of the target city and display the prediction results.
2. The method for capturing and rapidly modeling urban three-dimensional spatial parameters according to claim 1, characterized in that, In step S500, the method for rapidly constructing the three-dimensional spatial geometric model of the city is as follows: S501. Collect the results of the horizontal dimension region division of the city, the results of the vertical dimension level division of the city, and the average geometric shape height of the city, and use them as the three-dimensional spatial geometric model data of the city for modeling. S502. Using a modal recognition algorithm, extract and normalize the data features of the urban three-dimensional spatial geometric model data to construct a homogenized dataset and upload it to the cloud service platform. S503. The city's three-dimensional spatial geometric model data is encoded and converted, and a finite volume mesh algorithm is embedded. This is further combined with fluid dynamics calculation software to achieve adaptive and rapid modeling of the city's geometric shape.
3. The method for capturing and rapidly modeling urban three-dimensional spatial parameters according to claim 1, characterized in that, In step S600, the method for predicting the urban physical environment is as follows: S601. Using the established three-dimensional spatial geometric model of the city and fluid dynamics numerical simulation software, the urban physical environment of the target city is quickly predicted, and the prediction results are obtained. S602. Compare the prediction results with the results of urban digital surface and ground thermal image capture; when the error exceeds the set percentage threshold, cluster the urban three-dimensional space again and predict the physical environment until the error is lower than the set percentage threshold. S603. Finally, the prediction results of the urban physical environment will be visualized.
4. A system for capturing and rapidly modeling three-dimensional spatial parameters in cities, characterized in that, include: Equipped with a high-definition infrared imaging camera and remotely controllable, the unmanned aerial vehicle is responsible for remote aerial photography of target cities or regions, and for collecting detailed aerial images of the target cities or regions. The geometric model data processing integration system is responsible for rapidly modeling the urban geometry and simulating the physical environment of a target city or region using the urban three-dimensional spatial parameter capture and rapid modeling method as described in any one of claims 1-3.
5. A computer device, characterized in that, include: The system includes a processor, a memory, a communication interface, and a communication bus. The processor, the memory, and the communication interface communicate with each other via the communication bus. The memory stores at least one executable instruction, which causes the processor to perform the operation corresponding to the urban three-dimensional spatial parameter capture and rapid modeling method as described in any one of claims 1-3.
6. A computer-readable storage medium, characterized in that, The computer storage medium stores at least one executable instruction that causes the processor to perform the operation corresponding to the urban three-dimensional spatial parameter capture and rapid modeling method as described in any one of claims 1-3.