A method for extracting building outlines and constructing unstructured networks for industrial park simulation

Abstract

The present invention relates to the technical field of extracting three-dimensional building outlines, and discloses a method for extracting building outlines and constructing an unstructured network for simulating industrial parks, which realizes efficient extraction of building outlines and output of high-quality grids, and provides reliable data support for subsequent numerical simulations; it effectively solves the problems of insufficient building outline extraction accuracy, complex three-dimensional model construction, and low grid generation quality in the existing technology; at the same time, the application integrates multiple advanced technologies to construct an integrated process from image processing to unstructured grid generation. Compared with traditional fixed grids, it can not only accurately reflect urban terrain and building details, but also has greater flexibility, and can effectively reduce the waste of computing resources, significantly improve the accuracy and efficiency of numerical simulations, and provide solid technical support for the refined analysis of pollution conditions in complex environments of industrial parks.

Description

Technical Field

[0001] The present invention relates to the technical field of extracting three-dimensional building outlines, and more particularly to a method for extracting building outlines for simulating industrial parks and constructing a non-structural network. Background Art

[0002] With the rapid development of industrialization and urbanization, the number and scale of industrial parks are expanding. While generating economic benefits, these parks also present a range of environmental issues, including odor pollution from exhaust gases and excessive particulate matter emissions. Accurately simulating and assessing pollution in industrial parks has become a hot topic in air pollution regulation. However, the complex building layouts, irregular terrain, diverse emission source types, and variable meteorological conditions within industrial parks make traditional numerical simulation methods difficult to meet the high-precision prediction requirements.

[0003] In numerical simulations, accurately extracting building outlines and constructing unstructured grids are key to improving simulation accuracy. When the wind field flows through the boundaries of buildings and complex terrain, its flow state will show significant changes, such as vortexes, wakes, and vortices. Capturing these microscopic details requires high-resolution grid support. Traditional fixed grids are difficult to effectively adapt to complex geometric boundaries, resulting in uneven grid density distribution during the simulation process, reduced computational efficiency, and even deviations in simulation results. In contrast, unstructured grids, due to their better adaptability and flexibility, can accurately capture the detailed characteristics of complex boundaries, providing strong support for the refined simulation of industrial parks.

[0004] As the foundation for constructing unstructured grids, the precise extraction of building outlines directly determines the quality of the grid and the accuracy of numerical simulations. Currently, building outline extraction methods are primarily categorized as manual and automated. Manual methods rely on humans visually analyzing image features and manually plotting outlines. While highly accurate, these methods are extremely inefficient and rely heavily on the interpreter's experience, making them difficult to meet the rapid processing requirements of large-scale, high-resolution images in industrial parks. Automated methods utilize algorithms to analyze images to extract building outlines. Commonly used techniques include edge detection and machine learning. Edge detection methods are computationally simple and highly efficient, but are susceptible to image noise, resulting in insufficient extraction accuracy and regularity. While machine learning techniques offer powerful feature expression capabilities, their reliance on labeled data and high computing power makes them costly. Furthermore, the diversity of building morphology and the complexity of image backgrounds (such as interfering factors such as vegetation, shadows, and roads) further increase the technical difficulty of automated extraction.

[0005] In summary, existing methods have two major problems: First, the accuracy and efficiency of building outline extraction are insufficient, and they mainly rely on manual interpretation or automated algorithms. The former is inefficient and cannot cope with the needs of large-scale high-resolution image processing, while the latter is easily affected by image noise and background complexity, resulting in insufficient accuracy and high cost; second, the adaptability and optimization capabilities of traditional grids are limited, and they fail to adjust the density according to the complex environmental characteristics of industrial parks, resulting in an excessive number of grids, a waste of computing resources, and difficulty in accurately capturing boundary details, which limits the simulation accuracy. Summary of the Invention

[0006] In order to overcome the above-mentioned defects in the prior art, the present invention provides a method for simulating the extraction of building contours in industrial parks and constructing an unstructured network, which realizes high-precision building contour extraction and integrates high-resolution image processing, sample separability analysis, supervised classification, post-processing optimization and other technologies. It proposes effective solutions to the problems of insufficient building contour extraction accuracy, low data processing efficiency, high cost, and complex operation in the prior art, realizes the efficient extraction of building contours and the comprehensive integration of three-dimensional spatial information, and provides accurate and reliable data support for applications in related fields.

[0007] The above technical objectives of the present invention are achieved through the following technical solutions: a method for simulating and extracting building outlines to construct an unstructured network for an industrial park, comprising the following steps:

[0008] S1. Data Preparation: Based on the basemap Tiantutu-image data of ArcGIS Pro software, high-resolution tif images of the required area of the industrial park were exported in blocks and data preprocessing was performed. Data preprocessing included radiometric calibration and atmospheric correction to eliminate radiometric errors and interference caused by sensors, lighting conditions, and atmospheric composition.

[0009] S2. Supervised classification:

[0010] (1) Sample selection and evaluation: Samples are selected according to different categories of objects, such as buildings, roads, vegetation, and shadows. Regions of interest are set for each sample, and the separability of the training samples is calculated. The values of any two parameters are between 0 and 2.0. If the value is greater than 1.9, it means that the samples are well separable and are qualified samples. If the value is less than 1.8, the samples need to be edited or reselected. If the value is less than 1, the two types of samples can be combined into one type of sample.

[0011] (2) Support vector machine classification: The support vector machine classification method in ENVI 5.3 software is used for supervised classification, and the pixels in the satellite image are divided into different categories based on the training samples.

[0012] (3) Data quality assessment: After the classification is completed, the confusion matrix is used to evaluate the classification accuracy and reliability. By calculating indicators such as overall accuracy and Kappa coefficient, it is ensured that the classification results meet the requirements, and the final classification results are converted into vector data format;

[0013] S3. Classification result processing:

[0014] (1) Data analysis: Perform maximum / minimum value analysis on the results and remove or reclassify some patches generated by supervised classification;

[0015] (2) Data simplification: Compare the original image with the classification results, simplify the classification results using the editing function of ArcMap 10.8 software, and delete the misclassified objects and redundant attributes;

[0016] (3) Graphics normalization: ArcMap 10.8 was used to normalize the geometric shapes of features and correct irregularities in boundaries or misclassified areas;

[0017] S4. Data integration and improvement: Merge the vector data shpfiles after block processing to form the overall building outline layer of the study area; based on the building parameters and high-resolution satellite image information of the study area, add a field representing the building height to the merged layer attributes, and input the building height data according to the corresponding latitude and longitude coordinates to ensure the integrity of the building spatial information and provide input data for subsequent geometric modeling;

[0018] S5. Geometric modeling: This program calls a pre-written Python program for geometric modeling, using the Fiona, Collections, and Math libraries for data processing. Fiona is used to read geographic information, while Collections and Math are used for mathematical calculations. The program is responsible for reading the shpfile to extract building outlines and height information, creating building objects, generating complete geometric graphics files, and setting the initial building grid to 10 meters for subsequent unstructured grid construction.

[0019] S6. Construct an unstructured grid: Use GMSH software to read the geometry file and construct the corresponding unstructured grid based on the position information, feature distribution, and feature resolution information in the file.

[0020] Furthermore, in step S1, the radiometric calibration uses an absolute calibration method based on an image metadata model to convert image pixel values into ground object reflectance.

[0021] Furthermore, in step S1, atmospheric correction is implemented through the FLAASH module in ENVI 5.3 to correct the absorption and scattering effects between bands and generate normalized data that is physically consistent and easy to classify.

[0022] Furthermore, in step S2, the support vector machine classifier classifies the pixels in the satellite image by selecting representative training samples and setting appropriate classification parameters.

[0023] Furthermore, in step S3, the data analysis uses a method similar to convolution filtering to classify the false pixels in the larger category into the category, define a kernel window, and replace the category of the central pixel with the category of the pixel that occupies a dominant position in the kernel window, or replace the category of the central pixel with the category of the pixel that occupies a secondary position.

[0024] In summary, the present invention has the following beneficial effects: The present invention provides a method for extracting building contours and constructing unstructured grids for refined simulation of industrial parks, which realizes efficient extraction of building contours and output of high-quality grids, and provides reliable data support for subsequent numerical simulations. The integration of high-resolution image processing, supervised classification, geometric modeling and unstructured grid generation technologies effectively solves the problems of insufficient building contour extraction accuracy, complex three-dimensional model construction, and low grid generation quality in the existing technology. Compared with the automated method that relies on machine learning, the present invention avoids the dependence on a large amount of labeled data and complex model training, and can directly realize high-resolution building contour extraction. At the same time, by integrating a variety of advanced technologies, an integrated process from image processing to unstructured grid generation is constructed. Compared with traditional fixed grids, the unstructured grid constructed by the present invention can not only accurately reflect urban terrain and building details, but also has greater flexibility, and can effectively reduce the waste of computing resources, significantly improve the accuracy and efficiency of numerical simulation, and provide solid technical support for the refined analysis of pollution conditions in complex environments of industrial parks. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a flow chart of a method for extracting building outlines and constructing unstructured grids for refined simulation of industrial parks according to an embodiment of the present invention;

[0026] Figure 2 1 is a schematic diagram of the process of extracting building outlines in an embodiment of the present invention, in which: (a) a satellite image of an industrial park; (b) a satellite image of a portion of an area; (c) a supervised classification result map of a portion of an area; (d) a spatial distribution map of buildings in a portion of an area;

[0027] Figure 3 Schematic diagram of constructing an unstructured grid in an embodiment of the present invention, including: (a) a satellite image of an industrial park; and (b) a distribution map of unstructured grids on the surface of a building. DETAILED DESCRIPTION

[0028] The following is combined with Figure 1-3 The present invention is described in further detail.

[0029] Example 1: A method for simulating and extracting building outlines to construct an unstructured network for an industrial park, such as Figure 1 Said method comprises the following steps:

[0030] 1. Obtain high-resolution satellite imagery: Based on the basemap Sky Map of ArcGIS Pro software, image data is divided into blocks to export high-resolution tif images of the area required for the industrial park.

[0031] 2. Data Preprocessing: Data preprocessing includes radiometric calibration and atmospheric correction to eliminate radiometric errors and interference caused by the sensor, lighting conditions, and atmospheric composition. Radiometric calibration uses an absolute calibration method based on an image metadata model to convert image pixel values into ground reflectance. Atmospheric correction is implemented using the FLAASH module in ENVI 5.3 to correct for inter-band absorption and scattering effects, generating physically consistent, normalized data that is easy to classify.

[0032] 3. Supervised Classification:

[0033] 3.1 Set regions of interest based on the types of features, such as buildings, roads, vegetation, and shadows, and calculate the separability of the training samples. The values of any two parameters are between 0 and 2.0. If the value is greater than 1.9, it means that the samples are well separable and are qualified samples; if it is less than 1.8, the samples need to be edited or reselected; if the value is less than 1, the two types of samples can be combined into one type of sample;

[0034] 3.2 Use the support vector machine classification method in ENVI 5.3 software to perform supervised classification. By selecting representative training samples and setting appropriate classification parameters, the pixels in the satellite image are classified.

[0035] 3.3 After the classification is completed, the confusion matrix is used to evaluate the classification accuracy and reliability. By calculating indicators such as overall accuracy and Kappa coefficient, it is ensured that the classification results meet the requirements;

[0036] 3.4 Convert the final classification results into vector data format.

[0037] 4. Standardize graphics:

[0038] 4.1 Perform a maximum / minimum analysis on the results to remove or reclassify some of the patches generated by supervised classification. Specifically, a method similar to convolution filtering is used to assign false pixels in a larger category to that category. Then, a kernel window is defined and the category of the central pixel is replaced by the category of the dominant pixel in the kernel window, or the category of the secondary pixel is replaced by the category of the central pixel.

[0039] 4.2 Compare the original image of the industrial park with the classification results, and use ArcMap10.8 software to refine the classification results and delete incorrectly classified objects and redundant attributes;

[0040] 4.3 Further optimize the building graphics, use ArcMap10.8 software to normalize the geometric shapes of the features, and correct irregular parts of the boundaries or misclassified areas.

[0041] 5. Data integration and improvement:

[0042] 5.1 Merge the vector data shpfile after block processing to form the building outline layer in the industrial park;

[0043] 5.2 Based on the building parameters and high-resolution satellite image information, add the field height representing the building height in the shpfile layer properties, and enter the building height data according to the corresponding latitude and longitude coordinates.

[0044] 6. Geometric modeling:

[0045] 6.1 Call pre-written Python programs to process data using the fiona, collections, and math libraries. Fiona is used to read geographic information, while collections and math are used for mathematical calculations.

[0046] 6.2 Extract the building outline and height information from the industrial park shpfile, create building objects, and generate complete geometric graphics files; and set the initial building grid to 10m for the next step of constructing the unstructured grid.

[0047] 7. Construct an unstructured mesh: Use GMSH software to read the industrial park geometry file generated in the previous step and construct the corresponding unstructured mesh based on the location information, feature distribution, and feature resolution information in the file. During the mesh generation process, the mesh size is set hierarchically based on the feature characteristics and resolution requirements to improve simulation efficiency and ensure the accuracy of the simulation results.

[0048] like Figure 2 、 Figure 3 As shown, taking an industrial park as an example, Figure 2 (a) is a satellite image of the industrial park. Figure 2 (b) is a satellite image of part of the industrial park. Figure 2 (c) is a supervised classification result map obtained for the region using the method provided in this application; Figure 2 (d) is a spatial distribution map of buildings in the area obtained by the method provided by this application. Figure 2 (b) to Figure 2 (c) is to conduct supervised classification of satellite images of part of the industrial park; Figure 2 (c) to Figure 2 (d) is the spatial distribution map of buildings in part of the industrial park extracted based on the supervised classification results.

[0049] Figure 3 (a) is a satellite image of an area in the industrial park; Figure 3 (b) The unstructured grid distribution map of the building surface in a certain area of the industrial park obtained by the method provided in this application.

[0050] This specific embodiment is merely an explanation of the present invention and is not intended to limit the present invention. After reading this specification, those skilled in the art may make non-creative modifications to this embodiment as needed. However, as long as such modifications are within the scope of the claims of the present invention, they are protected by patent law.

Claims

1. A method for simulating and extracting building outlines to construct an unstructured network for industrial parks, characterized by: The following steps are involved: S1. Data Preparation: Based on the basemap Tiantutu-image data of ArcGIS Pro software, high-resolution tif images of the required area of the industrial park were exported in blocks and data preprocessing was performed. Data preprocessing included radiometric calibration and atmospheric correction to eliminate radiometric errors and interference caused by sensors, lighting conditions, and atmospheric composition. S2. Supervised classification: (1) Sample selection and evaluation: According to the different categories of ground objects, the regions of interest are set and the separability of the training samples is calculated. The values of any two parameters are between 0 and 2.

0. If the value is greater than 1.9, it means that the samples are well separable and are qualified samples; if the value is less than 1.8, the samples need to be edited or reselected; if the value is less than 1, the two types of samples are combined into one type of sample; (2) Support vector machine classification: supervised classification is performed using the support vector machine classification method in ENVI 5.3 software to classify the pixels in the satellite image into different categories based on the training samples; (3) Data quality assessment: After the classification is completed, the confusion matrix is used to evaluate the classification accuracy and reliability. By calculating the overall accuracy and Kappa coefficient indicators, it is ensured that the classification results meet the requirements, and the final classification results are converted into vector data format; S3. Classification result processing: (1) Data analysis: Perform maximum / minimum value analysis on the results and remove or reclassify some patches generated by supervised classification; (2) Data simplification: Compare the original image with the classification results, simplify the classification results using the editing function of ArcMap 10.8 software, and delete the misclassified objects and redundant attributes; (3) Graphics normalization: ArcMap 10.8 was used to normalize the geometric shapes of features and correct irregularities in boundaries or misclassified areas; S4. Data integration and improvement: Merge the vector data shpfiles after block processing to form the overall building outline layer of the study area; based on the building parameters and high-resolution satellite image information of the study area, add a field representing the building height to the merged layer attributes, and input the building height data according to the corresponding latitude and longitude coordinates to ensure the integrity of the building spatial information and provide input data for subsequent geometric modeling; S5. Geometric modeling: This program calls a pre-written Python program for geometric modeling, using the Fiona, Collections, and Math libraries for data processing. Fiona is used to read geographic information, while Collections and Math are used for mathematical calculations. The program is responsible for reading the shpfile to extract building outlines and height information, creating building objects, generating complete geometric graphics files, and setting the initial building grid to 10 meters for subsequent unstructured grid construction. S6. Construct an unstructured grid: Use GMSH software to read the geometry file and construct the corresponding unstructured grid based on the position information, feature distribution, and feature resolution information in the file.

2. The method for simulating and extracting building outlines to construct an unstructured network for an industrial park according to claim 1, characterized in that: In step S1, the radiometric calibration uses an absolute calibration method based on an image metadata model to convert image pixel values into ground object reflectance.

3. The method for simulating and extracting building outlines to construct an unstructured network for an industrial park according to claim 1, characterized in that: Atmospheric correction is implemented through the FLAASH module in ENVI 5.3 to correct the absorption and scattering effects between bands and generate normalized data that is physically consistent and easy to classify.

4. The method for simulating and extracting building outlines to construct an unstructured network for an industrial park according to claim 1, characterized in that: In step S2, the support vector machine classifier selects representative training samples and sets appropriate classification parameters to classify the pixels in the satellite image.

5. The method for simulating and extracting building outlines to construct an unstructured network for an industrial park according to claim 1, characterized in that: In step S3, the data analysis uses the convolution filtering method to classify the false pixels in the larger category into the category, define a kernel window, and replace the category of the central pixel with the category of the pixel that occupies the main position in the kernel window, or replace the category of the central pixel with the category of the pixel that occupies the secondary position.

Images