Smart city planning simulation dynamic simulation method and system

Abstract

The invention relates to the technical field of image simulation, in particular to a dynamic simulation method and system for smart city planning simulation, and the method comprises the following steps: obtaining remote sensing image data of a city, calculating the reflectivity of a surface wave band, analyzing the height difference of surface features based on elevation data, and calculating the building coverage rate of a unit region. And classifying the differentiated material regions by using the surface material reflection features to obtain image multi-dimensional feature data. According to the invention, by analyzing the pixel gradient change rate and the building density parameter and optimizing the urban boundary determination method, the accuracy of boundary determination and the detail processing capability of urban planning are improved, and the classification of urban functional areas is accurately analyzed by combining the building height and the surface spectral reflectance analysis. The method enhances the understanding of urban spatial distribution and function division, integrates the application of multi-source data and algorithms, not only improves the efficiency of urban planning, but also increases the practicability and scientificity of a planning result.

Description

Technical Field

[0001] The present invention relates to the technical field of image simulation, and particularly to a method and system for dynamic simulation of smart city planning and simulation. Background Art

[0002] The technical field of image simulation encompasses a wide range of applications, covering multiple aspects from basic image processing to advanced visual simulation. The core content is to use computational methods to create or improve the visual representation of images and videos. Image simulation technology is not limited to the processing of static images, but also includes the real-time simulation of dynamic scenes and the development of technologies such as augmented reality. This technical field supports a wide range of applications from medical imaging, the entertainment industry to autonomous vehicles by simulating real or virtual visual effects. A systematic introduction to this technical field shows that applications range from simple image color adjustment to complex environmental simulation, all of which are achieved through algorithms and computational models to accurately reproduce real-world visual phenomena.

[0003] Among them, the method for dynamic simulation of smart city planning and simulation refers to a method that simulates various changes in the city through digital simulation technology to support urban planning and management decisions. The technical matters targeted by this patent theme include urban layout planning, traffic flow simulation, and optimization of the layout of public service facilities, etc. Specifically, it constructs a dynamic urban development simulation model by collecting urban data and applying geographic information system (GIS) and computer vision technology. By simulating the impact of different planning schemes on aspects such as the layout of urban functional areas and urban traffic, it provides scientific decision-making information for urban planners. This kind of simulation does not rely on any single module, but uses multiple data inputs and algorithms for comprehensive analysis and processing.

[0004] The existing technologies show several deficiencies in practical applications. Especially in terms of the accuracy and real-time update ability of urban planning, traditional urban planning methods rely on outdated or insufficiently comprehensive data, resulting in the planning results not being able to fully reflect the latest situation of urban development. The existing technologies lack effective dynamic simulation capabilities and are difficult to reflect urban changes in real time, making planning decisions not adapt to the rapidly changing urban environment. These deficiencies limit the flexibility and adaptability of urban planning, leading to unreasonable resource allocation or lagging updates, affecting the overall function of the city and the quality of life of residents. By analyzing the deficiencies, it can be seen that in terms of the timeliness, comprehensiveness of data, and the dynamic simulation capabilities of processing algorithms, the existing technologies need further improvement and update. Summary of the Invention

[0005] To solve the technical problems existing in the prior art in terms of the accuracy and real-time update ability of urban planning, traditional urban planning methods rely on outdated or insufficiently comprehensive data, resulting in planning results that cannot fully reflect the latest situation of urban development. The prior art lacks effective dynamic simulation capabilities and is difficult to reflect urban changes in real time, making planning decisions unsuitable for the rapidly changing urban environment. These deficiencies limit the flexibility and adaptability of urban planning, leading to unreasonable resource allocation or lagging updates, affecting the overall function of the city and the quality of life of residents. Embodiments of the present invention provide a method and system for dynamic simulation of smart city planning and simulation. The technical solutions are as follows:

[0006] On the one hand, a method for dynamic simulation of smart city planning and simulation is provided, and the method includes:

[0007] S1: Obtain remote sensing image data of the city, calculate the surface band reflectivity, analyze the height differences of ground objects based on elevation data, classify the differentiated material regions using the reflection characteristics of surface materials, and obtain multi-dimensional feature data of the image;

[0008] S2: Use the multi-dimensional feature data of the image, select a fixed pixel window to calculate the horizontal and vertical gradient components of the pixels within the window, analyze the gradient direction change rate of adjacent pixel points, and obtain green space boundary data;

[0009] S3: Based on the green space boundary data, combined with the building elevation information of the city, calculate the average building height, analyze the range of elevation gradient fluctuations between buildings, and based on the simulation data model of building elevation changes in a smart city, calculate the change in the contrast density of building spacing in the region, judge the building distribution form, and obtain the building height distribution result;

[0010] S4: Call the building height distribution result, analyze the building occupancy ratio per unit area based on building density, analyze the building coverage status of the differentiated reflectivity region, and combined with the remote sensing map of the urban functional area and the structural diagram of the simulation model nodes, analyze the functional categories of urban regions, obtain the classification record of urban functional areas, and identify and divide the regions with differentiated urban functions.

[0011] As a further solution of the present invention, the multi-dimensional feature data of the image includes band reflectivity, elevation difference rate, pixel gradient change rate, building coverage rate, and surface material category. The green space boundary data includes the distribution of boundary candidate points, edge continuity, and denoised boundary points. The building height distribution result includes the average building height, range of elevation gradient fluctuations, change in building spacing density, and building distribution form. The classification record of urban functional areas includes functional categories, regional spectral characteristics, building occupancy ratio status, and spatial distribution pattern.

[0012] As a further solution of the present invention, the specific steps for obtaining the multi-dimensional feature data of the image are as follows:

[0013] S101: Obtain the remote sensing image data of the city, call the original surface multi-band reflectance data, calculate the reflectance of the band based on the received radiation flux, and obtain the surface spectral reflectance value;

[0014] S102: Based on the surface spectral reflectance value, call the elevation data, calculate the change in ground object elevation at the same geographical coordinates, screen the elevation data of adjacent areas to compare the height differences, extract the elevation change values of adjacent pixels, identify the trend of ground object height change, and combine with the regional grid distribution to calculate the height differences of continuous pixels to obtain the ground object height change amount;

[0015] S103: Call the ground object height change amount, identify the gradient direction of adjacent pixels, calculate the gradient change rate according to the gradient direction, calculate the building coverage rate per unit area in combination with the building density, classify different regions, and obtain the multi-dimensional feature data of the image.

[0016] As a further solution of the present invention, the steps for obtaining the green space boundary data are specifically as follows:

[0017] S201: Call the multi-dimensional feature data of the image, select a fixed pixel window, calculate the horizontal gradient and vertical gradient components of multiple pixels within the window, identify the pixel gradient direction based on the gradient component data, and obtain the pixel gradient distribution data;

[0018] S202: Based on the pixel gradient distribution data, analyze the gradient direction change rate of adjacent pixel points, screen the pixels with prominent gradient changes as boundary candidate points, analyze the gradient direction differences of the candidate points, eliminate the noise points with sudden direction changes, and screen the pixel points that conform to the boundary change trend to obtain the boundary candidate point distribution data;

[0019] S203: Call the boundary candidate point distribution data, evaluate the spatial distribution continuity of the boundary points, optimize the edge coherence according to the boundary point density, adjust the edge point distribution of the fracture area, and obtain the green space boundary data.

[0020] As a further solution of the present invention, the steps for obtaining the building height distribution result are specifically as follows:

[0021] S301: Call the green space boundary data, combine with the elevation information of urban buildings, calculate the height values of multiple buildings, screen the building pixel areas, calculate the average height of the buildings within the area, and obtain the building average height data;

[0022] Calculate the average height of the buildings within the area using the formula:

[0023]

[0024] where, H avgrepresents the average height of buildings within the representative area, M represents the number of building pixels, and h o represents the elevation value of the o-th building pixel, represents the average value of the elevation values of building pixels, A o represents the area value of the o-th building pixel, H represents the maximum elevation value of buildings within the green space boundary, and h min represents the lowest building elevation value within the building pixel area;

[0025] S302: Based on the building average height data, analyze the elevation gradient fluctuation range between buildings, calculate the change in the contrast density of the adjacent building spacing, analyze the building distribution pattern in combination with the building coverage rate, identify the building height distribution characteristics of the differentiated area, and obtain the building height distribution result.

[0026] As a further solution of the present invention, the obtaining steps of the urban functional area classification record are specifically as follows:

[0027] S401: Invoke the building height distribution result, calculate the building density value according to the comparison of the number of buildings in the area and the unit area, analyze the proportion of buildings per unit area under different differentiated grids, and obtain the building density ratio data;

[0028] S402: Use the building density ratio, combine with the average reflectivity of the regional surface wavebands, compare the building height and coverage rate distribution in the differentiated reflectivity area, identify the building distribution state of the density buildings, analyze the spatial proportion under the differentiated reflectivity, calibrate the building state characteristics of the differentiated area, and obtain the building coverage state data;

[0029] S403: Utilize the building coverage state data, refer to the characteristics of the building height, building density, and reflectivity intervals, analyze the combined characteristics of the differentiated areas in the spatial distribution pattern, classify the regional functional types and perform regional marking to obtain the urban functional area classification record.

[0030] As a further solution of the present invention, the formula for calculating the building density value is as follows:

[0031]

[0032] where D rp represents the building density value of the p-th grid in the r-th area, Z rp represents the number of buildings in the p-th grid in the r-th area, U rp represents the unit area of the p-th grid in the r-th area, represents the average value of the number of buildings in the grids of the r-th area, S rp represents the standard deviation of the building areas in the p-th grid in the r-th area, A rk represents the height value of the k-th building in the r-th area, B pkrepresents the floor area of the k-th building in the p-th grid, and m represents the total number of buildings in the p-th grid of the r-th area.

[0033] As a further solution of the present invention, the method further includes step S5:

[0034] S5: Using the urban functional area classification record, analyzing the fitting degrees of building heights, surface reflectivities, and building density fitting degrees in adjacent areas, calling the boundary fitting degree to adjust the classification boundary, evaluating the rationality of the comparison boundary of the category characteristics of adjacent urban areas, and optimizing the functional area segmentation according to the regional clustering relationship to obtain the dynamic boundary data of urban functional areas;

[0035] The dynamic boundary data of urban functional areas includes boundary fitting degree, regional category characteristic difference, and functional area clustering relationship.

[0036] As a further solution of the present invention, the specific steps for obtaining the dynamic boundary data of urban functional areas are as follows:

[0037] S501: Calling the urban functional area classification record, analyzing the numerical changes of building heights, surface reflectivities, and building densities in adjacent areas, calculating the fitting degree of building characteristics in adjacent areas, calculating the category characteristic difference value of the adjacent boundary of different functional areas, and analyzing the continuous change of building characteristics after boundary adjustment to obtain the optimized functional area classification boundary data;

[0038] S502: Using the optimized functional area classification boundary data, evaluating the spatial connectivity between functional areas, adjusting the distribution state at the boundary junction, optimizing the integrity of regional division, and obtaining the dynamic boundary data of urban functional areas.

[0039] On the other hand, the smart city planning simulation dynamic simulation system is used to execute the above-mentioned smart city planning simulation dynamic simulation method, and the system includes:

[0040] The remote sensing feature analysis module obtains urban remote sensing image data, surface elevation data, and ground object reflectivity data, calculates the elevation difference to analyze the change of ground object height, and divides the surface material area by comparing the reflectivity feature differences in the material area to obtain the multi-dimensional feature data of the image;

[0041] The green space boundary recognition module calls the multi-dimensional feature data of the image, selects a fixed pixel window, calculates the horizontal and vertical gradient components of the pixels in each window, screens the boundary candidate pixels according to the gradient direction difference of adjacent pixel points, and counts the change of boundary point density to obtain the green space boundary data;

[0042] Based on the green space boundary data, the building form calculation module calls the urban area building elevation layer data and the building density value within the cell, calculates the average height value of the building unit, compares the elevation difference with the spatial spacing value to calculate the building spacing change rate within the area, determines the evenness of building distribution, and generates the building height distribution result;

[0043] The functional area category judgment module calls the building height distribution result, obtains the building coverage per unit area, the surface reflectivity, and the building distribution interval amount, evaluates the corresponding relationship between the reflectivity change in the differential area and the building density, and obtains the urban functional area classification record;

[0044] Based on the urban functional area classification record, the functional area boundary calibration module combines the building height value, the building density value, and the surface reflectivity value of the adjacent area, compares the fitting difference between the three values with the change direction of the area category boundary, and obtains the dynamic boundary data of the urban functional area.

[0045] The beneficial effects brought by the technical solutions provided in the embodiments of the present invention at least include:

[0046] By obtaining and using the remote sensing image data of the city to calculate the surface band reflectivity and the height difference of ground objects, the accuracy of the data and the richness of multi-dimensional features are effectively improved, making urban planning and management decisions more scientific and accurate. By analyzing the pixel gradient change rate and the building density parameter, the method for determining the urban boundary is optimized, and the accuracy of boundary determination and the ability to handle details in urban planning are improved. By combining the analysis of building height and surface spectral reflectivity, the classification of urban functional areas is accurately analyzed, enhancing the understanding of urban spatial distribution and functional division. The application of this comprehensive multi-source data and algorithms not only improves the efficiency of urban planning, but also increases the practicality and scientific nature of the planning results. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic diagram of the working process of the present invention;

[0048] Figure 2 It is a system flow chart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] The following describes the technical solutions in the present invention with reference to the drawings.

[0050] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to represent examples, illustrations, or explanations. Any embodiment or design solution described as "exemplary" in the present invention should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, the use of the word "example" is intended to present concepts in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one of the two can be selected.

[0051] To make the technical problems, technical solutions, and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0052] Please refer to Figure 1 , an embodiment of the present invention provides a method for dynamic simulation of smart city planning and simulation. The processing flow of this method can include the following steps:

[0053] S1: Obtain the remote sensing image data of the city, calculate the surface band reflectivity, analyze the height difference of ground objects based on elevation data, calculate the pixel gradient change rate according to the gradient direction information, calculate the building coverage rate per unit area in combination with the building density parameter, classify the differentiated material areas using the reflection characteristics of surface materials, and obtain the multi-dimensional feature data of the image;

[0054] S2: Use the multi-dimensional feature data of the image, select a fixed pixel window to calculate the horizontal and vertical gradient components of the pixels within the window, analyze the gradient direction change rate of adjacent pixel points, screen the pixels with gradient changes as boundary candidate points, eliminate the noise points with sudden direction changes, and optimize the edge continuity according to the distribution of boundary points to obtain the green space boundary data;

[0055] S3: Based on the green space boundary data, combine the building elevation information of the city, calculate the average building height, analyze the elevation gradient fluctuation range between buildings, calculate the change in the contrast density of the building spacing in the area based on the smart city building elevation change simulation data model, and judge the building distribution form in combination with the building coverage rate to obtain the building height distribution result;

[0056] S4: Call the building height distribution result, analyze the building proportion per unit area based on the building density, calculate the surface spectral reflectivity of the area, analyze the building coverage status of the differentiated reflectivity areas, combine the remote sensing map of the urban functional area and the node structure diagram of the simulation model, refer to the building height, density, and surface reflectivity, analyze the functional categories of urban areas, and divide the functional areas in combination with the spatial distribution pattern to obtain the classification record of urban functional areas;

[0057] S5: Use the classification record of urban functional areas to analyze the fitting degree of building height, surface reflectivity, and building density in adjacent areas, call the boundary fitting degree to adjust the classification boundary, evaluate the rationality of the contrast boundary of the category characteristics of adjacent urban areas, and optimize the functional area segmentation according to the regional clustering relationship to obtain the dynamic boundary data of urban functional areas;

[0058] The multi-dimensional feature data of the image includes band reflectance, elevation difference rate, pixel gradient change rate, building coverage rate, surface material category. The green space boundary data includes the distribution of boundary candidate points, edge continuity, and denoised boundary points. The building height distribution result includes average building height, elevation gradient fluctuation range, building spacing density change, and building distribution pattern. The urban functional area classification record includes functional category, regional spectral characteristics, building occupancy status, and spatial distribution pattern. The urban functional area dynamic boundary data includes boundary fitting degree, regional category feature difference, and functional area clustering relationship.

[0059] The steps for obtaining the multi-dimensional feature data of the image are specifically as follows:

[0060] S101: Obtain the remote sensing image data of the city, call the original multi-band reflectance data of the surface, and calculate the reflectance of the band based on the received radiation flux to obtain the surface spectral reflectance value;

[0061] When obtaining the remote sensing image data of the city, download the original remote sensing image containing multiple bands through platforms such as Sentinel-2 or Landsat-8, and select the analysis area (such as 116.38–116.45, 39.90–39.95) and the target time (such as June 2023). Each pixel in the image contains the digital number (DN) value of one or more bands. First, the DN value needs to be converted into the radiation flux L according to the radiation calibration coefficient λ , and the conversion formula is:

[0062] L λ = Gain × DN + Offset;

[0063] For the B4 band, if DN = 100, Gain = 0.01, Offset = 0, then:

[0064] L λ = 0.01 × 100 + 0 = 1.00;

[0065] Continue to use the solar zenith angle θ s , the solar irradiance ESUN of the band λ to calculate the surface reflectance ρ, and the formula is:

[0066]

[0067] Substitute L λ = 1.00, ESUN λ = 1895 W / m 2 ·μm, θ s = 30°, and calculate to get ρ≈0.0019. Process all pixels and multiple bands of the entire image in the same way, and output the multi-band spectral reflectance value at each pixel position.

[0068] S102: Based on the surface spectral reflectance value, call the elevation data, calculate the elevation change of the ground object at the same geographical coordinate, screen the elevation data of adjacent regions to compare the height differences, extract the elevation change value of adjacent pixels, identify the height change trend of the ground object, and combine with the regional grid distribution to calculate the height difference of continuous pixels to obtain the elevation change of the ground object;

[0069] Find the pixel elevation value H in the corresponding elevation model according to the pixel geographical coordinate i,j , use the central difference method to compare the elevation data of adjacent pixels to extract the local elevation change. Select the pixel point (i, j) = (200, 300), and its elevation is H 200,300 = 45.2m, the adjacent pixel H 199,300 = 44.5m, H 201,300 = 46.1m, then the elevation change calculation formula is:

[0070]

[0071] Substitute the data to get:

[0072]

[0073] Perform the same calculation for all pixels in turn. If the set elevation change threshold T H = 1.0m, then judge ΔH < T H , this point is a non-mutation point, otherwise it is marked as a height difference area. Then, take a 50m×50m grid as a unit, average or median the ΔH of all pixels in the grid as the regional elevation change amount, and extract the overall height change pattern of the city through sliding window iteration. The result represents the average ground object height change value of each grid.

[0074] S103: Call the elevation change of the ground object, identify the gradient direction of adjacent pixels, calculate the gradient change rate according to the gradient direction, calculate the building coverage rate per unit area in combination with the building density, classify the differentiated areas, and obtain the multi-dimensional feature data of the image;

[0075] Extract the gradient direction and change rate of adjacent pixels, and use the Sobel algorithm to calculate the height difference gradients G x 、G y in the x and y directions, and then calculate the local gradient direction angle θ, which is defined as follows:

[0076]

[0077] If at a certain pixel G x = 1.2, G y = 1.6, then:

[0078] θ≈arctan(1.6 / 1.2)≈53.13°;

[0079] Then, calculate the building coverage rate R based on the building density D within the area and the set normalization coefficient K. The building density DQ is defined as:

[0080] DQ = NQ b / NQ;

[0081] where NQ b is the number of building pixels, and NQ is the total number of pixels;

[0082] Assume DQ = 0.4, the height change ΔH = 1.3m, and K = 1.2. Then the formula for calculating the building coverage rate is:

[0083] R = ΔH·DQ·K;

[0084] Substituting the values gives:

[0085] R = 1.3×0.4×1.2 = 0.624;

[0086] According to the threshold, divide the area types. R > 0.6 is the high-density area, 0.3 < R ≤ 0.6 is the medium-density area, and R ≤ 0.3 is the low-density area. At the same time, record the reflectivity, multi-band characteristics, gradient angle, building density, etc. of each pixel to form multi-dimensional image feature data

[0087] The specific steps for obtaining the green space boundary data are as follows:

[0088] S201: Call the multi-dimensional image feature data, select a fixed pixel window, calculate the horizontal and vertical gradient components of multiple pixels within the window, and identify the pixel gradient direction based on the gradient component data to obtain the pixel gradient distribution data;

[0089] After calling the multi-dimensional image feature data, a fixed-size sliding pixel window, such as 3×3 or 5×5, needs to be set in the entire remote sensing image, and each position of the image is scanned step by step. For each central pixel in the window, extract the multi-dimensional feature values of its surrounding pixels, especially focusing on information such as elevation value, spectral reflectivity, building density, etc. Calculate the horizontal gradient G x and the vertical gradient G y in the vertical direction, and then judge the gradient direction of the pixel in the local area. The formula for the gradient component is:

[0090] G x = f(x + 1, y) - f(x - 1, y);

[0091] G y = f(x, y + 1) - f(x, y - 1);

[0092] Among them, f(x, y) represents the grayscale value of the pixel or the constructed feature intensity value, such as the reflectance in the B4 band. If the pixel (100, 100) is the center point, the reflectance of its left pixel is 0.45, the right pixel is 0.61, the upper pixel is 0.52, and the lower pixel is 0.50, then:

[0093] G x = 0.61 - 0.45 = 0.16;

[0094] G y = 0.50 - 0.52 = -0.02;

[0095] Use these two components to calculate the gradient direction distribution and obtain the pixel gradient distribution data.

[0096] S202: Based on the pixel gradient distribution data, analyze the gradient direction change rate of adjacent pixel points, select the pixels with prominent gradient changes as boundary candidate points, analyze the gradient direction differences of the candidate points, eliminate the noise points with sudden direction changes, and select the pixel points that conform to the boundary change trend to obtain the boundary candidate point distribution data;

[0097] Calculate the gradient direction change rate between adjacent pixels, extract the positions with significant changes as boundary candidate points. For any adjacent pixels P i and P j , whose gradient directions are θ i and θ j respectively, define its change rate as:

[0098] Δθ = |θ i - θ j |;

[0099] For example, θ i = 45°, θ j = 93°;

[0100] Then Δθ = 48°. If the change threshold T θ = 40° is set, then this position is a candidate boundary point. Perform this type of determination on all pixels to initially form a set of boundary candidate points. Eliminate the pseudo-boundary points caused by local mutations or noises in this set. If the direction change of a single point is obvious but the direction difference from its neighboring points is large (for example, the change rates of the adjacent 4 points are all less than 10, and only this point exceeds 50), it is regarded as an isolated noise point and eliminated. Retain the pixels that are continuous in space and have a consistent direction change trend as the boundary candidate point distribution data.

[0101] S203: Call the boundary candidate point distribution data, evaluate the spatial distribution continuity of the boundary points, optimize the edge coherence based on the boundary point density, adjust the edge point distribution in the broken area, and obtain the green space boundary data;

[0102] It is necessary to further evaluate its spatial continuity and determine whether there is a break or discontinuity in the boundary. For each candidate point, the point density ρ within a certain range around it (such as a 5×5 window) is calculated, which is defined as follows:

[0103]

[0104] Among them, NC is the number of candidate points, and NW is the total number of pixels in the window;

[0105] If there are 15 candidate points in a 25-pixel window, then:

[0106] ρ = 15 / 25 = 0.6;

[0107] Set the density threshold T ρ =0.4, if ρ>T ρ If the local boundary distribution is continuous, it is considered as a broken area, and interpolation, edge fitting and other methods are needed to complete the missing edge points. For example, the missing area between two points is connected by linear interpolation, and the overall continuity after completion is evaluated again. The processed edge points are summarized as complete green space boundary data.

[0108] The specific steps for obtaining the building height distribution results are as follows:

[0109] S301: Calling green space boundary data, combining urban building elevation information, calculating the height values of multiple buildings, screening building pixel areas, calculating the average height of buildings in the area, and obtaining average building height data;

[0110] Calculate the average height of buildings in the area using the formula:

[0111]

[0112] Among them, H avg represents the average height of buildings in the area, M represents the number of building pixels, and h o represents the elevation value of the oth building pixel, Represents the average value of building pixel elevation, A o represents the area value of the oth building pixel, H represents the maximum elevation value of the building within the green space boundary, and h min Represents the lowest building elevation value in the building pixel area;

[0113] Parameter meaning and formula calculation derivation process:

[0114] The number of building pixels M represents the number of pixels identified as independent building units in the effective building area. This value is obtained by spatially superimposing high-resolution remote sensing images combined with vector data of building outline boundaries, extracting binary building mask images, and then counting the number of pixels. The processing results show that the number of effective building pixels in the area is M = 5;

[0115] The elevation value h of each building pixel o Obtained through 3D modeling of lidar point cloud data and processed by digital elevation model (DEM), the building elevation values are as follows:

[0116] h1 = 32.7 m, h2 = 34.1 m, h3 = 29.6 m, h4 = 31.2 m, h5 = 35.4 m;

[0117] The building pixel area value A o Is obtained by converting the pixel resolution in the remote sensing image. The spatial resolution of the image is 0.5 m, and the area of each pixel is 0.5 × 0.5 = 0.25 square meters, resulting in:

[0118] A1 = A2 = A3 = A4 = A5 = 0.25 square meters;

[0119] The average building elevation The calculation method is to average the elevation values of all building pixels:

[0120]

[0121] The highest building value H in the area is determined by extracting the maximum elevation value among all building pixels:

[0122] H = max(32.7, 34.1, 29.6, 31.2, 35.4) = 35.4;

[0123] The lowest building elevation value h min Is the minimum elevation value among the extracted building pixels:

[0124] h min = min(32.7, 34.1, 29.6, 31.2, 35.4) = 29.6;

[0125] Substitute the parameters into the formula and expand step by step as follows:

[0126] Calculate the expression value corresponding to each building pixel:

[0127] The 1st pixel:

[0128]

[0129] The 2nd pixel:

[0130]

[0131] The 3rd pixel:

[0132]

[0133] The 4th pixel:

[0134]

[0135] The 5th pixel:

[0136]

[0137] Find the average value of all pixel expressions:

[0138]

[0139] The result shows that the average height of building pixels in the area is 3.4711 meters. This height value is the calculation result of the average building height data and can be used as a characteristic parameter for the average three-dimensional scale of multiple building facades in the area. It can be used as a data input item in subsequent analysis scenarios such as building group distribution, skyline fitting, and urban space scale determination.

[0140] S302: Based on the average building height data, analyze the range of elevation gradient fluctuations between buildings, calculate the change in the comparison density of adjacent building spacings, combine the building coverage rate to analyze the building distribution pattern, identify the building height distribution characteristics in the differentiated area, and obtain the building height distribution result;

[0141] Further analyze the elevation fluctuation characteristics between buildings and their spatial distribution differences. Divide the urban area into grid cells of equal area, count the average building height value of each cell, and compare it with the height values of surrounding grids to extract the trend of height difference changes. For any two adjacent cells, record their average building height difference to form an urban building elevation gradient map. At the same time, combine the spatial distance information of the building center points to analyze the spacing distribution of adjacent buildings, compare it with the building quantity or density information to judge the spatial compactness of the local area. In areas with high density and small spacing, such as the old city or industrial agglomeration areas, the building height fluctuations are generally small; in newly built areas or suburbs, there are areas with large height changes but low density. By combining the building spacing with the average height and coverage rate, building form feature combinations such as "high density - medium high", "low density - high", "medium density - low" are formed. For different types of areas, further classification marks can be made to identify areas with sudden height changes, areas with rich levels, or areas with a single building pattern, and obtain the building height distribution result.

[0142] The specific steps for obtaining the classification record of urban functional areas are as follows:

[0143] S401: Call the building height distribution result, calculate the building density value according to the comparison of the number of buildings in the area and the unit area, analyze the proportion of buildings per unit area in the differentiated grid, and obtain the building density ratio data;

[0144] The formula for calculating the building density value is as follows:

[0145]

[0146] Among them, D rp represents the building density value of the p-th grid in the r-th area, Z rp represents the number of buildings in the p-th grid in the r-th area, U rp represents the unit area of the p-th grid in the r-th area, represents the average value of the number of buildings in the grids of the r-th area, S rp represents the standard deviation of the building areas in the p-th grid in the r-th area, A rk represents the height value of the k-th building in the r-th area, B pk represents the floor area of the k-th building in the p-th grid, and m represents the total number of buildings in the p-th grid in the r-th area;

[0147] Meaning of parameters and derivation process of formula calculation:

[0148] The area number is set as r = 3, and the grid number is set as p = 5. The number of buildings in the 5th grid of this area is detected as Z 35 = 24 buildings. The monitoring data of the number of buildings in all grids in area 3 is {22, 24, 26, 23, 24, 25, 21, 20, 25}. The average value can be calculated as follows:

[0149]

[0150] The grid unit area U 35 is obtained as 1200 square meters after GIS plot division. The standard deviation S of the building area 35 is statistically calculated from the floor area data of 24 buildings in this grid. The actual data set is {56, 60, 58, 62, 64, 57, 59, 61, 63, 60, 58, 56, 59, 60, 62, 64, 61, 63, 58, 59, 60, 62, 61, 60} (unit: square meters). The calculation process is as follows:

[0151] Calculate the average value of this data

[0152]

[0153] Calculate the sum of squared differences

[0154]

[0155] Standard deviation calculation

[0156]

[0157] Building height A 3kand floor area B 5k The data is collected by lidar point cloud scanning and feature segmentation mapping equipment. The combined sample data of the first 6 buildings are as follows (unit: meters):

[0158] Height data of Group A: {18.2, 20.5, 19.8, 21.0, 20.3, 19.1};

[0159] Floor area of Group B: {56, 60, 58, 62, 64, 57};

[0160] Substitute this part of the data into the radical term:

[0161]

[0162] The average value is:

[0163]

[0164] Substitute the combined part into the numerator calculation of the formula:

[0165]

[0166] Calculate the denominator part:

[0167]

[0168] The overall formula calculation is:

[0169]

[0170] This result shows that the building density value of Grid 5 in Area 3 is 1.271. This value quantitatively represents the structural strength degree of the building distribution per unit area within this grid, serves as an input item for subsequent analysis of the proportion of buildings in different grids, and further supports the output and standardization processing of building density ratio data.

[0171] S402: Use the building density ratio, combined with the average reflectance of the regional surface band, to compare the building height and coverage distribution in the differential reflectance area, identify the distribution state of density buildings, analyze the spatial proportion under differential reflectance, calibrate the building state characteristics of the differential area, and obtain building coverage state data;

[0172] Introduce the information of the average surface band reflectivity in the urban area, conduct a correlation analysis between the reflectivity data in different bands and the building density, building height, and coverage rate. In each grid area, first calculate the average band reflectivity of all pixels, and focus on selecting bands that are sensitive to differentiating ground objects, such as red light, green light, and near-infrared. Combine the reflectivity value of each grid with the corresponding average building height and density value of the grid to find the correlation pattern between the high and low reflectivity and the building characteristics. In the area with high reflectivity, if the building coverage rate is low and the height is large, it represents a high-rise sparse residential area; conversely, in the area with low reflectivity, the appearance of high-density medium and low-rise buildings may represent old low-rise residential areas or industrial areas. By constructing a regional classification table, record the combined characteristics of building density, coverage rate, and height in different reflectivity intervals, and further count the proportion of various combinations in the city according to the regional area to obtain the building coverage status data.

[0173] S403: Utilize the building coverage status data, refer to the characteristics of building height, building density, and reflectivity intervals, analyze the combined characteristics of the differentiated areas in the spatial distribution pattern, classify the functional types of the areas and conduct area marking to obtain the urban functional area classification record;

[0174] Combine the combined characteristics of building height distribution, building density ratio, and multi-band reflectivity to analyze the spatial distribution pattern of each area in the city, extract the combined characteristics of the differentiated areas and carry out functional area classification processing. First, divide the building height distribution into high-rise, mid-rise, and low-rise intervals; divide the building density into dense, moderate, and sparse categories; and divide the reflectivity into three categories according to the band characteristics: high reflectivity (for roofs and roads), medium reflectivity (bare land and low vegetation), and low reflectivity (dense vegetation and green spaces). Form a functional area recognition template with the combined characteristics. The area with the combination of high buildings, high density, and high reflectivity can be initially determined as a commercial area, the area with high buildings, low density, and low reflectivity is a new residential area or a mixed public building area, and the area with low buildings, low density, and low reflectivity is mostly a park or an open space. Match the combined characteristics for each grid in the entire urban map one by one, automatically classify them into corresponding functional types, such as residential areas, commercial areas, industrial areas, green spaces, etc., and record the results in the form of functional markings in the urban functional area layer for use in scenarios such as urban planning and resource allocation to obtain the urban functional area classification record.

[0175] The specific steps for obtaining the dynamic boundary data of urban functional areas are as follows:

[0176] S501: Call the urban functional area classification record, analyze the numerical changes in the building height, surface reflectivity, and building density of adjacent areas, calculate the fitness of the building characteristics of adjacent areas, calculate the category characteristic difference value of the adjacent boundary of the differentiated functional areas, and analyze the continuous change of the building characteristics after the boundary adjustment to obtain the optimized functional area classification boundary data;

[0177] It is necessary to compare the three indicators of building height, surface reflectivity, and building density between adjacent regions item by item, and extract the changes in building characteristics at the boundaries of each adjacent region. In specific implementation, each pair of adjacent grid cells is used as the boundary unit of the adjacent region, and the average building height, main band reflectivity value, and building density value are extracted item by item, and the differences between the three are calculated to evaluate the degree of building characteristic differences between functional areas. The average building height of Region A is 12.3 meters, the building density is 0.65, and the reflectivity is 0.38; for Region B, they are 6.5 meters, 0.35, and 0.49 respectively. Then the three differences are 5.8 meters, 0.30, and -0.11 respectively. Next, the overall fitness of the building characteristics of the two regions is calculated based on these differences. For example, the boundary difference index is obtained by using the weighted average of the normalized differences to measure whether there is an unreasonable functional area division at this boundary. If the boundary difference value is lower than the set threshold, it is considered that the characteristics of the two regions tend to be continuous, and the boundary can be merged or optimized; if the difference value is large, the original boundary is retained, and the continuity after the boundary is fine-tuned is further analyzed, that is, to observe whether the adjusted functional areas form a more coherent building characteristic sequence in space, and obtain the optimized functional area classification boundary data.

[0178] S502: Use the optimized functional area classification boundary data to evaluate the spatial connectivity between functional areas, adjust the distribution state at the boundary intersection, optimize the integrity of the area division, and obtain the dynamic boundary data of urban functional areas;

[0179] Enter the comprehensive evaluation and optimization stage of regional spatial connectivity and boundary integrity. For the overall urban pattern, check the boundary contact situation of each functional area one by one, especially whether there are isolated small areas or fragmented areas at the boundary intersections. Superimpose the optimized boundary lines with the original functional area distribution to identify the continuity and unity of the spatial structure of the boundary area. If it is found that there are many scattered small units in a certain area, it is necessary to judge whether they can be incorporated into the adjacent main area. If they have similar building characteristics and functional categories, the merger process is executed. At the same time, in the boundary intersection area, according to the degree of spatial gradient transition of the building characteristics, appropriately adjust the division range of the intersection point to break through the fractured structure between functional areas and improve the integrity of the area division. At the intersection of the residential area and the commercial area, if there is a small area with relatively low reflectivity and moderate building height, it is classified into the residential functional area according to its attributes. Throughout the optimization process, the basic principles of maintaining the continuity of building characteristics and the rationality of spatial relationships are always adhered to, and the dynamic boundary data of urban functional areas are obtained.

[0180] Please refer to Figure 2 , a smart city planning simulation dynamic simulation system, the system includes:

[0181] The remote sensing feature analysis module obtains urban remote sensing image data, surface elevation data, and ground object reflectivity data, calculates the elevation difference to analyze the change in the height of ground objects, compares the difference in reflectivity characteristics of the material area to divide the surface material area, and obtains the multi-dimensional feature data of the image;

[0182] The green space boundary recognition module calls the multi-dimensional feature data of the image, selects a fixed pixel window, calculates the horizontal and vertical gradient components of the pixels in each window, screens the boundary candidate pixels according to the gradient direction difference of adjacent pixel points, and counts the change in the boundary point density to obtain the green space boundary data;

[0183] The building form calculation module, based on the green space boundary data, calls the building elevation layer data of the urban area and the building density value within the cell, calculates the average height value of the building unit, compares the elevation difference with the spatial spacing value to calculate the building spacing change rate in the area, judges the uniformity of building distribution, and generates the building height distribution result;

[0184] The functional area category judgment module calls the building height distribution result, obtains the building coverage per unit area, surface reflectivity, and building distribution interval, evaluates the corresponding relationship between the change in reflectivity of the differentiated area and the building density, and obtains the classification record of urban functional areas;

[0185] The functional area boundary calibration module, based on the classification record of urban functional areas, combines the building height value, building density value, and surface reflectivity value of adjacent areas, compares the fitting difference between the three values and the change direction of the area category boundary, and obtains the dynamic boundary data of urban functional areas.

[0186] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for dynamic simulation of smart city planning and simulation, characterized in that, It includes the following steps: S1: Obtain the remote sensing image data of the city, calculate the surface band reflectance, analyze the height differences of ground objects based on elevation data, classify the differentiated material areas using the reflection characteristics of surface materials, and obtain the multi-dimensional feature data of the image; S2: Adopt the multi-dimensional feature data of the image, select a fixed pixel window to calculate the horizontal and vertical gradient components of the pixels within the window, analyze the gradient direction change rate of adjacent pixel points, and obtain the green space boundary data; S3: Based on the green space boundary data, combine the building elevation information of the city, calculate the average building height, analyze the elevation gradient fluctuation range between buildings, and based on the smart city building elevation change simulation data model, calculate the change in the building spacing comparison density of the area, judge the building distribution form, and obtain the building height distribution result; S4: Call the building height distribution result, analyze the building proportion per unit area based on building density, analyze the building coverage status of the differentiated reflectance area, combine the remote sensing atlas of the urban functional area and the structural diagram of the simulation model nodes, analyze the urban area function category, obtain the classification record of the urban functional area, and identify and divide the areas with differentiated urban functions.

2. The dynamic simulation method for smart city planning and simulation according to claim 1, characterized in that The multi-dimensional feature data of the image includes band reflectance, elevation difference rate, pixel gradient change rate, building coverage rate, and surface material category. The green space boundary data includes the distribution of boundary candidate points, edge continuity, and denoised boundary points. The building height distribution result includes the average building height, elevation gradient fluctuation range, building spacing density change, and building distribution form. The classification record of the urban functional area includes function category, regional spectral characteristics, building proportion status, and spatial distribution pattern.

3. The dynamic simulation method for smart city planning simulation according to claim 1, characterized in that The specific steps for obtaining the multi-dimensional feature data of the image are as follows: S101: Obtain the remote sensing image data of the city, call the original surface multi-band reflectance data, calculate the reflectance of the band based on the received radiation flux, and obtain the surface spectral reflectance value; S102: Based on the surface spectral reflectance value, call the elevation data, calculate the elevation change amount of ground objects at the same geographical coordinates, screen the elevation data of adjacent areas to compare the height differences, extract the elevation change values of adjacent pixels, identify the height change trend of ground objects, and combine the regional grid distribution to calculate the continuous pixel height difference to obtain the elevation change amount of ground objects; S103: Call the elevation change amount of ground objects, identify the gradient direction of adjacent pixels, calculate the gradient change rate according to the gradient direction, calculate the building coverage rate per unit area in combination with the building density, classify the differentiated areas, and obtain the multi-dimensional feature data of the image.

4. The method for dynamic simulation of smart city planning and simulation according to claim 1, wherein The specific steps for obtaining the green space boundary data are as follows: S201: Call the multi-dimensional feature data of the image, select a fixed pixel window, calculate the horizontal gradient and vertical gradient components of multiple pixels within the window, identify the pixel gradient direction based on the gradient component data, and obtain the pixel gradient distribution data; S202: Based on the pixel gradient distribution data, analyze the gradient direction change rate of adjacent pixel points, screen the pixels with prominent gradient changes as boundary candidate points, analyze the gradient direction differences of the candidate points, remove the noise points with sudden direction changes, and screen the pixel points that conform to the boundary change trend to obtain the distribution data of the boundary candidate points; S203: Invoke the boundary candidate point distribution data, evaluate the spatial distribution continuity of boundary points, optimize the edge coherence based on the boundary point density, adjust the distribution of edge points in the fracture area, and obtain the green space boundary data.

5. The method for dynamic simulation of smart city planning and simulation according to claim 1, wherein The specific steps for obtaining the building height distribution result are as follows: S301: Invoke the green space boundary data, combine with the elevation information of urban buildings, calculate the height values of multiple buildings, screen the building pixel areas, calculate the average height of the buildings within the area, and obtain the building average height data; To calculate the average height of the buildings within the area, use the formula: Among them, H avg represents the average height of buildings in the area, M represents the number of building pixels, h o represents the elevation value of the o-th building pixel, represents the average value of the elevation values of building pixels, A o represents the area value of the o-th building pixel, H represents the maximum elevation value of the buildings within the green space boundary, h min represents the lowest building elevation value in the building pixel area; S302: Based on the building average height data, analyze the elevation gradient fluctuation range between buildings, calculate the change in the contrast density of the adjacent building spacing, combine with the building coverage rate to analyze the building distribution pattern, identify the building height distribution characteristics in the differentiated area, and obtain the building height distribution result.

6. The smart city planning simulation dynamic simulation method according to claim 1, wherein The specific steps for obtaining the urban functional area classification record are as follows: S401: Invoke the building height distribution result, calculate the building density value according to the comparison of the number of buildings in the area and the unit area, analyze the proportion of buildings per unit area in the differentiated grid, and obtain the building density ratio data; S402: Use the building density ratio, combine with the average reflectance of the regional surface band, compare the building height and coverage rate distribution in the differentiated reflectance area, identify the building distribution state of the density buildings, analyze the spatial occupancy ratio under the differentiated reflectance, calibrate the building state characteristics of the difference area, and obtain the building coverage state data; S403: Use the building coverage state data, refer to the characteristics of the building height, building density, and reflectance interval, analyze the combined characteristics of the differentiated areas in the spatial distribution pattern, classify the regional function types and perform regional marking to obtain the urban functional area classification record.

7. The method for dynamic simulation of smart city planning and simulation according to claim 6, characterized in that The formula for calculating the building density value is as follows: Among them, D rp represents the building density value of the p-th grid in the r-th area, Z rp represents the number of buildings in the p-th grid within the r-th area, U rp represents the unit area of the p-th grid in the r-th area, represents the average value of the number of buildings in the grids of the r-th area, S rp represents the standard deviation of the building areas in the p-th grid within the r-th area, A rk represents the height value of the k-th building in the r-th area, B pk represents the floor area of the k-th building in the p-th grid, and m represents the total number of buildings in the p-th grid within the r-th area.

8. The dynamic simulation method for smart city planning and simulation according to claim 1, wherein The method further includes step S5: S5: Use the urban functional area classification record to analyze the fitting degree of the building height, surface reflectance, and building density in adjacent areas, invoke the boundary fitting degree to adjust the classification boundary, evaluate the rationality of the comparison boundary of the category characteristics of adjacent urban areas, and optimize the functional area segmentation according to the regional clustering relationship to obtain the urban functional area dynamic boundary data; The urban functional area dynamic boundary data includes the boundary fitting degree, the difference in regional category characteristics, and the functional area clustering relationship.

9. The dynamic simulation method for smart city planning and simulation according to claim 1, characterized in that The specific steps for obtaining the urban functional area dynamic boundary data are as follows: S501: Invoke the urban functional area classification record, analyze the numerical changes in the building height, surface reflectance, and building density in adjacent areas, calculate the fitting degree of the building characteristics in adjacent areas, calculate the difference value of the category characteristics of the adjacent boundaries of the differentiated functional areas, and analyze the continuous change of the building characteristics after the boundary adjustment to obtain the optimized functional area classification boundary data; S502: Use the optimized functional area classification boundary data to evaluate the spatial connectivity between functional areas, adjust the distribution state at the boundary junction, and optimize the integrity of the regional division to obtain the urban functional area dynamic boundary data.

10. A smart city planning simulation dynamic simulation system, characterized in that, According to the smart city planning simulation dynamic simulation method described in any one of claims 1-9, the system includes: The remote sensing feature analysis module obtains urban remote sensing image data, surface elevation data, and ground object reflectivity data, calculates the elevation difference to analyze the change in ground object height, compares the reflectivity feature differences in the material area to divide the surface material area, and obtains the multi-dimensional feature data of the image; The green space boundary recognition module calls the multi-dimensional feature data of the image, selects a fixed pixel window, calculates the horizontal and vertical gradient components of the pixels in each window, screens the boundary candidate pixels based on the gradient direction difference of adjacent pixel points, and statistically analyzes the change in boundary point density to obtain the green space boundary data; The building form calculation module, based on the green space boundary data, calls the building elevation layer data and the building density value in the cell of the urban area, calculates the average height value of the building unit, compares the elevation difference and the spatial spacing value to calculate the building spacing change rate in the area, determines the uniformity of building distribution, and generates the building height distribution result; The functional area category judgment module calls the building height distribution result, obtains the building coverage per unit area, surface reflectivity, and building distribution interval, evaluates the corresponding relationship between the reflectivity change in the differentiated area and the building density, and obtains the urban functional area classification record; The functional area boundary calibration module, based on the urban functional area classification record, combines the building height value, building density value, and surface reflectivity value of the adjacent area, compares the fitting difference between the three values and the change direction of the area category boundary, and obtains the dynamic boundary data of the urban functional area.

Images