Urban skyline image extraction method

Abstract

The invention discloses a city skyline image extraction method, relates to the technical field of digital city planning, and aims to solve the problem of low efficiency of city skyline image extraction in the prior art by constructing a simulation model and extracting a skyline in the simulation model. According to the technical scheme, the extraction efficiency of the city skyline image is greatly improved. And the efficiency is improved by at least two times. In addition, the problems that the viewpoint is single and is limited by the standing position of a photographer due to the fact that the city skyline is extracted through a picture are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of digital urban planning, and particularly to a method for extracting urban skyline images. Background Art

[0002] As the most directly recognizable urban image in visual perception, the urban skyline not only reveals the footprint of the city's historical development but also reflects the sustainability of the city's spatial development.

[0003] The skyline is an important manifestation of the urban physical outline and a major component of the urban landscape resources. Although it is only a two-dimensional elevation contour line of the city from a certain perspective, it reflects three-dimensional spatial hierarchical features. As one of the elements of the "urban image", the skyline is the most direct and specific display of the urban spatial landscape and often becomes the business card and characteristic identification line of the urban image. Therefore, urban planning and design centered on urban skyline control can effectively enhance the spatial order of regional development and improve the overall image of the city. Currently, most urban skyline images are obtained by manual photography and hand-drawing methods. This method requires high tracing ability of the drawing personnel, resulting in low efficiency in extracting urban skyline images. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for extracting urban skyline images to address the problem of low efficiency in extracting existing urban skyline images.

[0005] The technical solution adopted by the present invention to solve the above technical problems is as follows:

[0006] A method for extracting urban skyline images includes the following steps:

[0007] Step 1: Construct a three-dimensional skyline simulation model;

[0008] Step 2: Form a virtual visual plane perpendicular to the ground according to the field of view of the observer;

[0009] Step 3: In the three-dimensional skyline simulation model, determine the screening areas of urban skyline observation points and visual corridors, and obtain the urban skyline image based on the virtual visual plane.

[0010] Further, the specific steps of Step 1 are as follows:

[0011] Step 1-1: Obtain the spatial basic data of the city to be extracted, where the spatial basic data includes DEM elevation data, building data, green space data, and road network data;

[0012] Step 1-2: Use the DEM elevation data, building data, green space data, and road network data to construct a three-dimensional skyline basic model;

[0013] Step 1-3: Build a 3D skyline simulation model based on the 3D skyline basic model, and adjust the scene environment to clear weather in the 3D skyline simulation model.

[0014] Further, the specific steps of Step 3 are as follows:

[0015] Step 3-1: Based on the DEM elevation data, use GIS software to obtain the slope, aspect, plane curvature, and profile curvature of the city. Then, use the natural break interval method to classify the slope, aspect, plane curvature, and profile curvature of the city.

[0016] Step 3-2: Based on the classification results of Step 3-1, select the highest-grade slope, the highest-grade aspect, the highest-grade plane curvature, the lowest-grade slope, the lowest-grade aspect, and the lowest-grade profile curvature. Then, obtain the areas corresponding to the highest-grade slope, the highest-grade aspect, and the highest-grade plane curvature, and obtain the intersection area of the three, which is the first screening area. Then, according to the lowest-grade slope, the lowest-grade aspect, and the lowest-grade profile curvature, jointly determine the corresponding area within the city, which is the second screening area. Finally, judge whether the altitude of the first area is greater than the city height limit. If so, select the second screening area; otherwise, select the first screening area and the second screening area.

[0017] Step 3-3: Define the area selected in Step 3-2 as the third screening area.

[0018] Step 3-4: Obtain the POI data, building data, and GPS trajectory data of the third screening area, and determine the high-density public spaces in the city according to the POI data and building data of the third screening area. In the determined high-density public spaces in the city, use the GPS trajectory data of the third screening area to generate a heat map of urban land use vitality.

[0019] Step 3-5: Use the natural break interval method to classify the heat map of urban land use vitality, and select the area corresponding to the highest grade, which is the fourth screening area.

[0020] Step 3-6: Obtain the road network data of the third screening area, and perform space syntax analysis on the road network data of the third screening area. The space syntax analysis includes integration analysis, choice analysis, global depth analysis, and visibility analysis. Then, classify each analysis result by the natural break interval method, and select the area corresponding to the highest integration analysis grade, the highest choice analysis grade, the lowest global depth analysis grade, and the highest visibility analysis grade, which is the fifth screening area.

[0021] Step 3-7: Obtain the intersection area of the fourth screening area and the fifth screening area, which is the screening area for the urban skyline observation points and view corridors, and obtain the urban skyline image based on the virtual view plane.

[0022] Further, the viewing range of the observer is obtained through the following steps:

[0023] Step 1: Obtain the latest population age data and sex ratio data of the city to be extracted from the World Pop platform, and based on the obtained data, in the age data with the largest number of people, determine the sex with the largest proportion, which is the target population;

[0024] Step 2: Obtain the height data of the target population from WorldPopulation Review, and then obtain the average height of the target population, which is used as the height of the observer;

[0025] Step 3: At the highest point of the observer's height, select a horizontal viewing angle of 120°, an upward viewing angle of 30°, and a downward viewing angle of 40° as the viewing range of the observer.

[0026] Further, the method further includes Step 38:

[0027] Step 38: According to the screening area of the urban skyline observation point and the visual corridor, obtain the skyline contour in the three-dimensional skyline simulation model, obtain the fractal dimension of the skyline contour, and finally evaluate the skyline according to the fractal dimension of the skyline contour.

[0028] Further, the fractal dimension of the skyline contour is obtained through the fraclab tool of MATLAB;

[0029] The evaluation of the skyline according to the fractal dimension of the skyline contour is expressed as:

[0030]

[0031] Further, the DEM elevation data is obtained through the TanDEM-X digital elevation model.

[0032] Further, the GPS trajectory data, building data, green space data, POI data, and road network data are obtained through the OpenStreetMap database.

[0033] Further, the three-dimensional skyline basic model is constructed by Rhino;

[0034] The three-dimensional skyline simulation model is constructed in City Engine.

[0035] Further, obtaining the urban skyline image based on the virtual vision plane is performed through Simulink.

[0036] The beneficial effects of the present invention are:

[0037] This application extracts the skyline by constructing a simulation model and performing skyline extraction in the simulation model. The technical solution of this application greatly improves the extraction efficiency of urban skyline images. The efficiency is increased by at least more than 2 times. Moreover, this application avoids the problems of single viewpoint and limitation by the photographer's standing position caused by extracting the urban skyline through photos. Description of the Drawings

[0038] Figure 1 It is a schematic diagram of the 3D skyline basic model;

[0039] Figure 2 It is a schematic diagram of the 3D skyline simulation model;

[0040] Figure 3 It is a schematic diagram of calculating the fractal dimension of the skyline using the fraclab tool of MATLAB. Detailed Embodiments

[0041] It should be specifically noted that, without conflict, the various embodiments disclosed in this application can be combined with each other.

[0042] Detailed Embodiment 1: A method for extracting urban skyline images described in this embodiment includes the following steps:

[0043] Step 1: Obtain the spatial basic data of the city to be extracted, and the spatial basic data includes DEM elevation data, building data, green space data, and road network data;

[0044] Step 2: Use the DEM elevation data, building data, green space data, and road network data to construct a 3D skyline basic model;

[0045] Step 3: Based on the 3D skyline basic model, construct a 3D skyline simulation model, and adjust the scene environment to clear weather in the 3D skyline simulation model;

[0046] Step 4: According to the field of view of the observer, form a virtual field of view plane perpendicular to the ground;

[0047] Step 5: Based on the DEM elevation data, use GIS software to obtain the slope, aspect, plane curvature, and profile curvature of the city, and then use the natural breakpoint interval method to classify the slope, aspect, plane curvature, and profile curvature of the city;

[0048] Step 6: Based on the grading results in Step 5, select the slope with the highest grade, the aspect with the highest grade, the plan curvature with the highest grade, the slope with the lowest grade, the aspect with the lowest grade, and the profile curvature with the lowest grade. Then, based on the slope with the highest grade, the aspect with the highest grade, and the plan curvature with the highest grade, jointly determine the corresponding area within the city, i.e., the first area. Then, based on the slope with the lowest grade, the aspect with the lowest grade, and the profile curvature with the lowest grade, jointly determine the corresponding area within the city, i.e., the second area. Finally, determine whether the altitude of the first area is greater than the city height limit. If so, select the second area; otherwise, select the first area and the second area;

[0049] Aspect, slope, plan curvature, and profile curvature are all in the same dimension. The high or low aspect only indicates the direction. What is really compared are the slope and plan curvature, and profile curvature. Generally, the plan curvature is high where the slope is high. Where it is not high, there may be more complex terrain, which is not suitable for urban engineering construction. Therefore, this application needs to exclude this part. Generally, the profile curvature is low where the slope is low. Where it is not low, there may be subsidence related to groundwater on the ground, which cannot be developed either. Therefore, this application also needs to exclude it. On this basis, when this application selects the areas with high or low aspect, it actually selects the areas where the aspect change is stable. A stable aspect change is conducive to urban development. So they will definitely overlap. This application needs to exclude the non-overlapping places. Then, considering the terrain, even the flat ground on the surface will have slope and aspect. Therefore, this consideration range can cover all areas of the entire city. Where there is slope does not necessarily mean there are small mounds.

[0050] Step 7: Define the area selected in Step 6 as the first screening area;

[0051] Step 8: Obtain the POI data, building data, and GPS trajectory data of the first screening area, and determine the high-density public spaces in the city based on the POI data and building data of the first screening area. In the determined high-density public spaces in the city, generate the urban land use vitality heat map using the GPS trajectory data of the first screening area;

[0052] Step 9: Use the natural break interval method to grade the urban land use vitality heat map, and select the area corresponding to the highest grade, which is the second screening area;

[0053] Step 10: Obtain the road network data of the first screening area, and perform space syntax analysis on the road network data of the first screening area. The space syntax analysis includes integration analysis, choice analysis, global depth analysis, and visibility analysis. Then, grade each analysis result separately using the natural break interval method, and select the area corresponding to the highest integration analysis grade, the highest choice analysis grade, the lowest global depth analysis grade, and the highest visibility analysis grade, which is the third screening area;

[0054] Step 11: Obtain the intersection area of the second screening area and the third screening area, which is the screening area for the urban skyline observation points and visual corridors, and obtain the urban skyline image based on the virtual visual plane.

[0055] Obtain the basic urban spatial data required for urban skyline evaluation, including the following sub-steps:

[0056] S1. Obtain the DEM elevation data of domestic and foreign cities from the TanDEM-X global free 90-meter digital elevation model;

[0057] S2. Obtain the public GPS trajectories of the city from the OpenStreetMap database (OSM);

[0058] S3. Obtain the building, green space, POI and road network data of the city from the OpenStreetMap database (OSM);

[0059] S4. Obtain the latest population age and gender ratio data of the city from the WorldPop platform, and obtain the population height data of different genders in different age groups of the city from WorldPopulation Review.

[0060] Construct an urban simulation model, including the following sub-steps:

[0061] S2.1. Use the elevation, road network, building and green space data to construct a 3D skyline basic model in Rhino, as Figure 1 shown.

[0062] S2.2. Further construct a 3D skyline simulation model in CityEngine, as Figure 2 shown.

[0063] S2.3. Adjust the scene environment of the model to the sunny weather with the solar altitude angle on June 21st in the location where Mudanjiang City is located;

[0064] S2.4. Determine the age and gender with the largest proportion of the urban population according to the latest population age and gender ratio data of the city, and determine the observation height of the human perspective of the urban skyline;

[0065] S2.5. Take the horizontal viewing angle of the human observer as 120°, the upward viewing angle as 30°, and the downward viewing angle as 40° as the normal visual field range of the human eye, and form a virtual visual plane perpendicular to the ground.

[0066] Determine the urban viewing points and sight lines corresponding to the skyline area, including the following sub-steps:

[0067] S3.1. Analyze the slope, aspect, planar curvature, and profile curvature of the city using GIS software based on DEM elevation data. Among them, the slope, planar curvature, and aspect comprehensively reflect the steepness of the terrain; the aspect and profile curvature reflect the undulation direction and hydrological characteristics of the terrain;

[0068] S3.2. Initially determine the high-density public spaces in the city based on the city's POI data and building data;

[0069] S3.3. Generate a heat map of urban land use vitality using the city's publicly available GPS trajectory data to further determine the urban public vitality hotspots and their distribution characteristics;

[0070] S3.4. Based on the analysis results of S3.2 and S3.3, divide the urban public vitality space hotspots into 6 levels using the natural break interval method;

[0071] S3.5. Conduct a space syntax analysis on the urban road network data, mainly involving integration, choice, global depth, and visibility analysis, and divide the analysis results into 6 levels respectively using the natural break interval method;

[0072] S3.6. Based on the classification results of S3.4 and S3.5, select areas where the mountain ridge lines around the study area do not exceed the height limit as mountain observation points, select high-rise public buildings that meet the height limit requirements as urban high-rise observation points, set public spaces with open spaces such as parks and water bodies as entertainment facility observation points, and determine the urban skyline observation points and visual corridors according to the horizontal viewing angle of the virtual visual plane.

[0073] Skyline contour acquisition and fractal dimension calculation include the following sub-steps:

[0074] S4.1. Obtain the skyline contours of each observation point and visual corridor from the perspective of people based on the virtual visual plane from the 3D simulation model;

[0075] S4.2. Calculate the fractal dimension of the skyline using the fraclab tool in MATLAB. As Figure 3 shown.

[0076] Evaluate the skyline according to the following evaluation system.

[0077]

[0078] It should be noted that the specific implementation manners are only explanations and illustrations of the technical solutions of the present invention, and the scope of the rights cannot be limited thereby. Any changes that are merely partial based on the claims and specifications of the present invention should still fall within the protection scope of the present invention.

Claims

1. A method for extracting urban skyline images, characterized in that It includes the following steps: Step 1: Construct a 3D skyline simulation model; Step 2: According to the field of view of the observer, form a virtual view plane perpendicular to the ground; Step 3: In the 3D skyline simulation model, determine the screening areas of the urban skyline observation points and visual corridors, and based on the virtual view plane, obtain the urban skyline image.

2. The method for extracting an urban skyline image according to claim 1, wherein The specific steps of Step 1 are as follows: Step 1-1: Obtain the spatial basic data of the city to be extracted, and the spatial basic data includes DEM elevation data, building data, green space data, and road network data; Step 1-2: Use the DEM elevation data, building data, green space data, and road network data to construct a 3D skyline basic model; Step 1-3: Based on the 3D skyline basic model, construct a 3D skyline simulation model, and adjust the scene environment to clear weather in the 3D skyline simulation model.

3. The method for extracting an urban skyline image according to claim 2, wherein The specific steps of Step 3 are as follows: Step 3-1: Based on the DEM elevation data, use GIS software to obtain the slope, aspect, plane curvature, and profile curvature of the city, and then use the natural break interval method to classify the slope, aspect, plane curvature, and profile curvature of the city; Step 3-2: Based on the classification results of Step 3-1, select the highest-grade slope, the highest-grade aspect, the highest-grade plane curvature, the lowest-grade slope, the lowest-grade aspect, and the lowest-grade profile curvature, and then obtain the areas corresponding to the highest-grade slope, the highest-grade aspect, and the highest-grade plane curvature, and obtain the intersection area of the three, that is, the first screening area. Then, according to the lowest-grade slope, the lowest-grade aspect, and the lowest-grade profile curvature, jointly determine the corresponding area within the city, that is, the second screening area. Finally, judge whether the altitude of the first area is greater than the city height limit. If so, select the second screening area; otherwise, select the first screening area and the second screening area; Step 3-3: Define the area selected in Step 3-2 as the third screening area; Step 3-4: Obtain the POI data, building data, and GPS trajectory data of the third screening area, and determine the high-density public spaces in the city according to the POI data and building data of the third screening area. In the determined high-density public spaces in the city, use the GPS trajectory data of the third screening area to generate an urban land use vitality heat map; Step 3-5: Use the natural break interval method to classify the urban land use vitality heat map, and select the area corresponding to the highest grade, that is, the fourth screening area; Step 3-6: Obtain the road network data of the third screening area, and perform spatial syntax analysis on the road network data of the third screening area. The spatial syntax analysis includes integration analysis, choice analysis, global depth analysis, and visibility analysis, and classify each analysis result through the natural break interval method, and select the area corresponding to the highest integration analysis grade, the highest choice analysis grade, the lowest global depth analysis grade, and the highest visibility analysis grade, that is, the fifth screening area; Step 37: Obtain the intersection area of the fourth screening area and the fifth screening area, which is the screening area for the urban skyline observation points and visual corridors, and obtain the urban skyline image based on the virtual visual plane.

4. The method for extracting an urban skyline image according to claim 3, wherein The field of view of the observer is obtained through the following steps: Step 1: Obtain the latest population age data and gender ratio data of the city to be extracted from the World Pop platform, and based on the obtained data, obtain the gender with the largest proportion in the age data with the largest number of people, that is, the target population; Step 2: Obtain the height data of the target population from the World Population Review, and then obtain the average height of the target population and use it as the observer's height; Step 3: At the highest point of the observer's height, select a horizontal viewing angle of 120°, an upward viewing angle of 30°, and a downward viewing angle of 40° as the observer's field of view.

5. A method for extracting urban skyline images according to claim 4, characterized in that The method further includes Step 38: Step 38: Obtain the skyline contour in the three-dimensional skyline simulation model according to the screening area of the urban skyline observation points and visual corridors, obtain the fractal dimension of the skyline contour, and finally evaluate the skyline according to the fractal dimension of the skyline contour.

6. The method for extracting an urban skyline image according to claim 5, wherein The fractal dimension of the skyline contour is obtained through the fraclab tool of MATLAB; The evaluation of the skyline according to the fractal dimension of the skyline contour is expressed as:

7. The method for extracting an urban skyline image according to claim 1, characterized in that The DEM elevation data is obtained through the TanDEM-X digital elevation model.

8. The method for extracting an urban skyline image according to claim 1, wherein The GPS trajectory data, building data, green space data, POI data, and road network data are obtained through the OpenStreetMap database.

9. A method for extracting urban skyline images according to claim 1, characterized in that The three-dimensional skyline basic model is constructed by Rhino; The three-dimensional skyline simulation model is constructed in City Engine.

10. The method for extracting an urban skyline image according to claim 1, wherein Obtaining the urban skyline image based on the virtual visual plane is performed through Simulink.

Images