Intelligent city planning method and system based on multi-source remote sensing data fusion

Abstract

The invention discloses an intelligent city planning method and system based on multi-source remote sensing data fusion, and particularly relates to the technical field of city planning, and the method comprises the steps: S1, region division: dividing a planning region into M dynamic management grids based on GIS geographic coordinates and city function zoning data, carrying out the auxiliary recognition of topographic relief and building height through LiDAR point cloud data, and adjusting the boundary of the grids, establishing a bidirectional mapping relation between the grid and the multi-source sensing equipment; s2, multi-source data acquisition: constructing a city holographic data set through global coverage and a multi-modal sensing technology, acquiring target city data in real time, then executing multi-stage cleaning and space-time alignment, and performing normalization operation on each layer of data by adopting a standardized processing mode to obtain city data; according to the invention, an air-sky-ground-network global sensing network and a multi-level space-time grid framework are constructed, satellite remote sensing, unmanned aerial vehicle point cloud, ground sensors and urban operation data are deeply fused, and the limitation of single-source data of a traditional system is broken through.

Description

Technical Field

[0001] The present invention relates to the technical field of urban planning, and more specifically, to an intelligent urban planning method and system based on multi-source remote sensing data fusion. Background Art

[0002] With the acceleration of the urbanization process, the scale of cities is constantly expanding, and urban planning faces many challenges; remote sensing technology, as an efficient and rapid means of obtaining geospatial information, has been widely used in urban planning; therefore, how to make full use of the advantages of multi-source remote sensing data, realize data fusion and analysis, and provide more comprehensive and accurate information support for urban planning has become a research hotspot in the current urban planning field.

[0003] Traditional urban planning methods include data collection steps, planning analysis steps, and planning evaluation steps; among them, the data collection steps are responsible for collecting the basic data required for urban planning and preliminary sorting, and these data are important bases for planning decisions; the planning analysis steps are based on the collected data to carry out the analysis and design work of urban planning, aiming to optimize the urban spatial structure and improve urban functions; the planning evaluation steps evaluate the effect of plan implementation, collect feedback information, so as to adjust and optimize the planning scheme, which helps to improve the scientificity and adaptability of the planning.

[0004] However, in actual use, there are still some disadvantages, such as limited data sources: traditional systems mainly rely on field measurement and map surveying means to obtain data, the data source channels are relatively narrow, and it is difficult to obtain geospatial information at large areas and macro scales. Field measurement has low efficiency and limited scope; weak analysis ability, traditional systems are relatively insufficient in data intelligent analysis, mostly relying on manual analysis and simple models, and the technical means are relatively single; insufficient prediction ability, due to untimely data update and limited analysis ability, traditional systems are difficult to provide effective decision support in the face of urban emergencies.

[0005] Therefore, there is an urgent need to provide an intelligent urban planning method and system based on multi-source remote sensing data fusion to solve the problems of limited data sources, weak analysis ability, and insufficient prediction ability of existing urban planning methods. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide an intelligent urban planning method based on multi-source remote sensing data fusion, through the following solutions, to solve the problems raised in the above background art.

[0007] To achieve the above object, the present invention provides the following technical solutions: An intelligent urban planning method based on multi-source remote sensing data fusion, including:

[0008] S1. Regional division: Based on GIS geographical coordinates and urban functional zoning data, the planned area is divided into M dynamic management grids. LiDAR point cloud data is used to assist in identifying terrain undulations and building heights, adjusting grid boundaries, and establishing a two-way mapping relationship between the grids and multi-source perception devices.

[0009] S2. Multi-source data collection: A holographic urban dataset is constructed through full-domain coverage and multi-modal perception technologies to collect real-time data of the target city. Then, multi-level cleaning, spatio-temporal alignment, and a standardized processing method are performed to normalize each layer of data to obtain urban data.

[0010] S3. Calculation of data core indicators: After standardizing the data obtained in S2, a mathematical model is established to generate four types of core indicators of urban signs, namely the influence coefficient of high-altitude remote sensing layer data, the influence coefficient of low-altitude drone layer data, the influence coefficient of ground sensor layer data, and the influence coefficient of urban operation layer data.

[0011] S4. Urban planning and feedback: Based on the four types of core indicators of urban signs obtained in S3, a hierarchical decision-making method is used to guide planning strategies in different dimensions respectively to achieve multi-objective collaborative optimization.

[0012] S5. Human-computer interaction: A user interface is constructed based on Web-based interaction technology to display urban planning methods, four types of core indicators of urban signs, and target city data in an intuitive chart form. A line chart is used to show the change trend of core indicators in different regions at different times and is transmitted to the user data terminal to provide reference data for users to make adjustment measures.

[0013] Preferably, the urban data includes high-altitude remote sensing layer data, low-altitude drone layer data, ground sensor layer data, and urban operation layer data.

[0014] Preferably, the high-altitude remote sensing layer data includes the vegetation index, denoted as NDVI; the surface deformation rate, denoted as V deform ; the intensity of night light remote sensing, denoted as L; the low-altitude drone layer data includes the unevenness index of the building surface, denoted as I roughness ; the point cloud density of the drone, denoted as Dp; the heat island intensity, denoted as UHI; the ground sensor layer data includes the traffic flow density, denoted as D traffic ; the PM2.5 concentration, denoted as C; the deformation rate of the underground pipe network, denoted as ε; the urban operation layer data includes the population density, denoted as D; the carbon emission intensity, denoted as I; the regional vitality index, denoted as A.

[0015] Preferably, the influence coefficient of the high-altitude remote sensing layer represents the suppression ability of vegetation cover on the heat island effect by dividing the index by the heat island effect. Adding 0.1 avoids a zero denominator, and the product relationship reflects the synergistic effect. The logarithmic ratio of land surface deformation to night-time economy indicates that the larger the value, the more prominent the contradiction between geological risk and economic expansion, which is used to reflect the ecological-geological-economic balance relationship. Specifically:

[0016]

[0017] Where L c represents the normal night-time remote sensing light intensity, L n represents the night-time remote sensing light intensity at night, and NDVI represents the vegetation index; V deform represents the land surface deformation rate.

[0018] Preferably, the influence coefficient of the low-altitude UAV layer intensifies the effect of the heat island through complex building methods, and its risk can be mitigated by using high-precision point clouds. The land surface temperature gradient is inhibited by vegetation, and the dimensional difference is eliminated through the square root and the vegetation index, which is used to reflect the health of the built environment and the microclimate. Specifically:

[0019]

[0020] Where I roughness represents the unevenness index of the building surface; Dp represents the UAV point cloud density; UHI represents the heat island intensity; NDVI represents the vegetation index.

[0021] Preferably, the influence coefficient of the ground sensor layer reflects the traffic pollution pressure by the product of vehicle flow and PM2.5 being reduced by vegetation. The negative correlation between pipeline network deformation and vegetation cover reflects the underground deformation risk. Vegetation, as a public adjustment factor, achieves dimensional normalization, which is used to reflect the dynamic pressure of traffic-pollution-infrastructure. Specifically:

[0022]

[0023] Where D traffic represents the vehicle flow density; C represents the PM2.5 concentration; ε represents the underground pipeline network deformation rate; NDVI represents the vegetation index.

[0024] Preferably, the influence coefficient of the urban operation layer reflects the population carbon footprint pressure by the product of population and carbon emissions being suppressed by both regional vitality and vegetation, which is used to reflect the population-energy-society collaborative resilience. Specifically:

[0025]

[0026] Where D represents the urban operation layer data including population density; I represents the carbon emission intensity; A represents the regional vitality index.

[0027] Preferably, the hierarchical decision-making method is specifically as follows:

[0028] Align the dynamic management grid divided in step S1 with four layers of data based on the UTM coordinate system, and then define the health thresholds of each coefficient according to historical data and urban regulations; then perform priority judgment based on the high-altitude remote sensing layer > low-altitude UAV layer > ground sensor layer > urban operation layer, and then specifically implement urban planning according to the size relationship between each coefficient and the health threshold of each coefficient; then generate new virtual urban data according to the urban planning strategy and feedback it to step S2 to obtain new virtual core indicators of four types of urban signs.

[0029] The health threshold is specifically the mean value of the core indicators of the four types of urban signs under the historical core indicators of the four types of urban signs and the target urban regulations.

[0030] Preferably, the urban planning method is specifically as follows:

[0031] When α < the health threshold of the high-altitude remote sensing layer, it indicates a high-risk grid, and restricted development is required, and green space compensation is increased. For high-risk grids, 30% of the construction land is compulsorily reserved and converted into green space, native drought-tolerant vegetation is planted, and an intelligent irrigation system is configured; if Vdeform > 5 mm / yr, high-risk buildings are demolished and backfilled with soil, and InSAR monitoring piles are synchronously arranged to achieve real-time deformation warning; if Lc - Ln > 10% and Vdeform > 3 mm / yr, the floor area ratio of newly built high-rise buildings is restricted to ≤ 2.5, and tax incentives are used to guide enterprises to invest in underground reinforcement projects, and new high-altitude remote sensing layer data is predicted according to urban planning measures and transmitted to S1; when α > the health threshold of the high-altitude remote sensing layer, it indicates a low-risk grid, and development is allowed, but the deformation rate needs to be monitored synchronously; the health threshold of the high-altitude remote sensing layer is the mean value of the historical influence coefficient of the high-altitude remote sensing layer data and the influence coefficient of the high-altitude remote sensing layer data under urban regulations;

[0032] When β > the health threshold of the low-altitude UAV layer, it indicates a high thermal risk, and mandatory building facade renovation and additional ventilation corridors are required. Reflective coatings are used on the roof, and three-dimensional greening is carried out to make the solar reflectance ≥ 0.8, the coverage rate of climbing plants ≥ 60%, and the utilization rate of the roof area ≥ 30%; if UHI / d > 0.2 °C / m, buildings occupying the air duct are demolished, and ventilation corridors with a width ≥ 50 m are set; if Dp < 500 points / m 2 , the UAV performs supplementary measurement every quarter, generates a building safety rating in combination with the BIM model, and predicts new low-altitude UAV layer data according to urban planning measures and transmits it to S1; when β < the health threshold of the low-altitude UAV layer, it indicates a low-risk area, and the original construction is supported, but the UAV point cloud data needs to be updated regularly; the health threshold of the low-altitude UAV layer is the mean value of the historical influence coefficient of the low-altitude UAV layer data and the influence coefficient of the low-altitude UAV layer data under urban regulations;

[0033] When δ > the health threshold of the ground sensor layer, it indicates high pollution pressure. It is necessary to implement traffic restriction policies, upgrade the drainage network, construct a green isolation belt with a width ≥ 10m, configure plants such as oleander to adsorb PM2.5, adjust traffic command according to the traffic flow density, and reduce the congestion duration by 25%; if ε > 500 με, use CIPP lining to repair and extend the service life, use distributed optical fiber for monitoring, and predict new ground sensor layer data according to urban planning measures and transmit it to S1; when δ < the health threshold of the ground sensor layer, it indicates a low-risk area, and the original traffic signal timing needs to be maintained; where the health threshold of the ground sensor layer is the mean value of the historical ground sensor layer data influence coefficient and the ground sensor layer data influence coefficient under urban norms.

[0034] When γ > the health threshold of the urban operation layer, it indicates high carbon. It is necessary to promote distributed photovoltaic coverage, restrict the access of high-energy-consuming industries, shut down enterprises with a steel tonnage energy consumption exceeding 600 kgce, and introduce zero-carbon data centers; if A < 0.5, layout a 15-minute community commercial complex, make the walkability rate reach 100%, provide a 50% rent subsidy, and predict new urban operation layer data according to urban planning measures and transmit it to S1; when γ < the health threshold of the urban operation layer, it indicates a high-vitality area, and it is necessary to maintain the original commercial complex layout and increase the public service density by 20%; where the health threshold of the urban operation layer is specifically the mean value of the historical urban operation layer data influence coefficient and the urban operation layer data influence coefficient under urban norms.

[0035] Preferably, the intelligent urban planning system based on multi-source remote sensing data fusion specifically includes:

[0036] Regional division module: Based on GIS geographical coordinates and urban functional zoning data, divide the planning area into M dynamic management grids. Use LiDAR point cloud data to assist in identifying terrain undulations and building heights, adjust the grid boundaries, and establish a two-way mapping relationship between the grids and multi-source sensing devices;

[0037] Multi-source data acquisition module: Construct a city holographic data set through full-domain coverage and multi-modal sensing technology, collect target city data in real time, and then perform multi-level cleaning, spatio-temporal alignment, and normalization operations on each layer of data using a standardized processing method to obtain city data;

[0038] Data core index calculation module: Based on the data obtained from the multi-source data acquisition module, after standardization processing, establish a mathematical model to generate four types of core indicators of urban signs, namely the high-altitude remote sensing layer data influence coefficient, the low-altitude UAV layer data influence coefficient, the ground sensor layer data influence coefficient, and the urban operation layer data influence coefficient;

[0039] Urban Planning and Feedback Module: Based on the four types of core urban feature indicators obtained by the Data Core Indicator Calculation Module, the hierarchical decision-making method is used to guide the planning strategies in different dimensions respectively, so as to achieve multi-objective collaborative optimization.

[0040] Human-Computer Interaction Module: Based on Web-based interaction technology, a user interface is constructed to display urban planning methods, four types of core urban feature indicators, and target city data in an intuitive chart form. The line chart is used to display the change trend of core indicators in different regions at different times, and it is transmitted to the user data terminal to provide reference data for users to make adjustment measures.

[0041] Technical Effects and Advantages of the Present Invention:

[0042] 1. By constructing an all-region perception network of space-air-ground-network and a multi-level spatio-temporal grid framework, the present invention deeply integrates satellite remote sensing, UAV point cloud, ground sensors and urban operation data, breaking through the limitations of single-source data of traditional systems.

[0043] 2. Based on the hierarchical dynamic influence coefficient model and the adaptive feedback mechanism, the present invention realizes the non-linear collaborative optimization of multiple objectives, and solves the linear simplification defect of traditional static analysis models for complex urban problems.

[0044] 3. The multi-source remote sensing technology and the dynamic iteration driven by real-time data of the present invention significantly improve the prediction accuracy of urban development trends and the robustness of solutions, making up for the deficiency of traditional planning relying on empirical deduction. Description of the Drawings

[0045] Figure 1 It is a schematic diagram of the overall structure of the present invention.

[0046] Figure 2 It is a schematic diagram of the system structure of the present invention. Detailed Embodiments

[0047] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0048] As shown in the attached Figure 1 The intelligent urban planning method based on multi-source remote sensing data fusion includes:

[0049] S1. Region division: Based on GIS geographic coordinates and urban functional zoning data, the planning region is divided into M dynamic management grids. LiDAR point cloud data is used to assist in identifying terrain undulations and building heights, adjusting the grid boundaries, and establishing a two-way mapping relationship between the grids and multi-source perception devices.

[0050] S2. Multi-source data collection. A city holographic dataset is constructed through full-domain coverage and multi-modal perception technologies to collect data of the target city in real time. Then, multi-level cleaning, spatio-temporal alignment, and a standardized processing method are performed to normalize each layer of data, obtaining city data.

[0051] In this embodiment, it should be specifically noted that the standardized processing method is specifically as follows:

[0052]

[0053] where X norm is the normalized data, X represents the original data, μ represents the mean of the original data, and σ represents the standard deviation of the original data, making data with different dimensions in the same scale range for subsequent calculation and comparison.

[0054] In this embodiment, it should be specifically noted that the city data includes high-altitude remote sensing layer data, low-altitude drone layer data, ground sensor layer data, and city operation layer data.

[0055] In this embodiment, it should be specifically noted that the high-altitude remote sensing layer data includes the vegetation index, denoted as NDVI; the surface deformation rate, denoted as V deform ; the night light remote sensing light intensity, denoted as L; the low-altitude drone layer data includes the building surface unevenness index, denoted as I roughness ; the drone point cloud density, denoted as Dp; the heat island intensity, denoted as UHI; the ground sensor layer data includes the traffic flow density, denoted as D traffic ; the PM2.5 concentration, denoted as C; the underground pipe network deformation rate, denoted as ε; the city operation layer data includes the population density, denoted as D; the carbon emission intensity, denoted as I; the regional vitality index, denoted as A.

[0056] In this embodiment, it should be specifically noted that the vegetation index uses the near-infrared and red light bands of an optical satellite to collect surface reflectance data, and then corrects the original radiation value based on the solar altitude angle and atmospheric parameters to obtain the vegetation index, specifically as follows:

[0057]

[0058] where ρ nir represents the surface reflectance in the near-infrared band; ρ red represents the surface reflectance in the red light band; the specific formula for surface reflectance radiation correction is as follows:

[0059]

[0060] where L λ represents the sensor radiation value, d represents the Earth-Sun distance; ESUNλ Represents the spectral solar irradiance. θ solar Represents the solar elevation angle.

[0061] In this embodiment, it should be specifically noted that the surface deformation rate is obtained by calling the satellite C-band SAR sensor to collect two scenes of SAR images and calculating the surface deformation rate through interferometric measurement technology. Specifically:

[0062]

[0063] where λ is the radar wavelength, Δφ represents the phase difference between the two scenes of SAR images, and T represents the time interval between the two images.

[0064] In this embodiment, it should be specifically noted that the night light remote sensing light intensity is extracted by collecting the night light remote sensing day and night band data.

[0065] In this embodiment, it should be specifically noted that the UAV point cloud density is obtained by carrying a five-lens oblique camera to collect RGB + near-infrared images at a ground resolution of 2 cm, then generating a three-dimensional point cloud, and recording the attitude angle of the aircraft to correct the image distortion.

[0066] In this embodiment, it should be specifically noted that the building contour concavity and convexity index is obtained by emitting 1064 nm laser pulses, combining GNSS positioning data to generate a digital elevation model, and extracting the building contour concavity and convexity index. Specifically:

[0067]

[0068] where σ(Z) represents the standard deviation of the building surface elevation, and μ(Z) represents the average value of the building surface elevation.

[0069] In this embodiment, it should be specifically noted that the urban heat island intensity is obtained by using a UAV carrying an infrared thermal imager to collect the surface temperatures of the urban center and the suburbs and directly subtracting them.

[0070] In this embodiment, it should be specifically noted that the calculation formula of the urban heat island intensity is specifically:

[0071] UHI = T urban -T rural ,

[0072] where T urban represents the average surface temperature of the urban center area, and T rural represents the surface temperature of the suburban reference point.

[0073] In this embodiment, it should be specifically noted that the traffic flow density is detected by deploying millimeter-wave radars and 4K cameras at the main road intersections in real time to detect the vehicle types and speeds, and then the traffic flow density is statistically calculated. Specifically:

[0074]

[0075] Where N vehicle represents the number of detected vehicles, and L road represents the monitored road length; T sample represents the sampling time.

[0076] In this embodiment, it should be specifically noted that the PM2.5 concentration is directly collected by deploying micro-meteorological stations.

[0077] In this embodiment, it should be specifically noted that the deformation rate of the monitored pipeline is directly monitored by deploying distributed fiber optic strain sensors along the drainage pipeline.

[0078] In this embodiment, it should be specifically noted that the population density is obtained by generating a population heat map density from anonymous mobile phone signaling data.

[0079] In this embodiment, it should be specifically noted that the carbon emission intensity is calculated by accessing the time series data of the building's electricity consumption power collected by smart electric meters (with a sampling interval of 15 minutes) and combining it with the gas meter data. Specifically:

[0080] I = 0.92×Pe + 2.15×Vg,

[0081] where Pe is the electricity consumption power and Vg represents the gas consumption.

[0082] In this embodiment, it should be specifically noted that the regional vitality index is calculated by crawling the POI check-in data on Weibo, counting the number of check-ins, and analyzing the text sentiment polarity. Specifically:

[0083] A = ln[Nc×(Ss + 1)],

[0084] where Nc represents the number of POI check-ins and Ss represents the sentiment polarity score.

[0085] S3. Calculation of data core indicators: Based on the data obtained in S2, after standardization processing, a mathematical model is established to generate four types of core indicators of urban signs, namely, the influence coefficient of high-altitude remote sensing layer data, the influence coefficient of low-altitude UAV layer data, the influence coefficient of ground sensor layer data, and the influence coefficient of urban operation layer data.

[0086] In this embodiment, it should be specifically noted that the influence coefficient of the high-altitude remote sensing layer represents the inhibitory ability of vegetation cover on the heat island effect by dividing the index by the heat island effect. Adding 0.1 avoids a denominator of 0, and the product relationship reflects the synergistic effect. The logarithmic ratio of land surface deformation to night-time economy indicates that the greater the value, the more prominent the contradiction between geological risk and economic expansion, which is used to reflect the balance relationship among ecology, geology, and economy. Specifically:

[0087]

[0088] Where L c represents the normal night-time remote sensing light intensity, L n represents the night-time remote sensing light intensity at night, and NDVI represents the vegetation index; V deform represents the land surface deformation rate.

[0089] In this embodiment, it should be specifically noted that the influence coefficient of the low-altitude UAV layer intensifies the effect of the heat island through complex building methods, and its risk can be mitigated by using high-precision point clouds. The land surface temperature gradient is inhibited by vegetation, and the dimensional difference is eliminated through the square root and the vegetation index, which is used to reflect the health of the built environment and the microclimate. Specifically:

[0090]

[0091] Where I roughness represents the unevenness index of the building surface; Dp represents the UAV point cloud density; UHI represents the heat island intensity; NDVI represents the vegetation index.

[0092] In this embodiment, it should be specifically noted that the influence coefficient of the ground sensor layer reflects the traffic pollution pressure through the product of traffic flow and PM2.5 being reduced by vegetation. The negative correlation between pipeline network deformation and vegetation cover reflects the risk of underground deformation. Vegetation, as a public adjustment factor, achieves dimensional normalization and is used to reflect the dynamic pressure of traffic, pollution, and infrastructure. Specifically:

[0093]

[0094] Where D traffic represents the traffic flow density; C represents the PM2.5 concentration; ε represents the underground pipeline network deformation rate; NDVI represents the vegetation index.

[0095] In this embodiment, it should be specifically noted that the influence coefficient of the urban operation layer reflects the population carbon footprint pressure through the product of population and carbon emissions being inhibited by both regional vitality and vegetation, and is used to reflect the synergistic resilience of population, energy, and society. Specifically:

[0096]

[0097] Among them, D represents the data of the urban operation layer including population density; I represents the carbon emission intensity; A represents the regional vitality index.

[0098] S4. Urban planning and feedback. Based on the four core indicators of urban signs obtained in S3, the hierarchical decision-making method is used to guide the planning strategies in different dimensions respectively, so as to achieve multi-objective collaborative optimization.

[0099] In this embodiment, it should be specifically noted that the hierarchical decision-making method is specifically as follows:

[0100] Align the dynamic management grids divided in step S1 with the four-layer data based on the UTM coordinate system, and then define the health thresholds of each coefficient according to historical data and urban specifications; then, based on the high-altitude remote sensing layer > low-altitude UAV layer > ground sensor layer > urban operation layer, perform priority judgment, and then specifically implement urban planning according to the size relationship between each coefficient and the health threshold of each coefficient; then generate new virtual urban data according to the urban planning strategy and feedback it to step S2 to obtain new virtual four-core indicators of urban signs;

[0101] In this embodiment, it should be specifically noted that the health threshold is specifically the mean value of the historical four-core indicators of urban signs and the four-core indicators of urban signs under the target urban specifications.

[0102] In this embodiment, it should be specifically noted that the urban planning method is specifically as follows:

[0103] When α < the health threshold of the high-altitude remote sensing layer, it represents a high-risk grid, and restricted development is required, and green space compensation is increased. For high-risk grids, 30% of the construction land is compulsorily reserved and converted into green space, native drought-tolerant vegetation is planted, and an intelligent irrigation system is equipped; if V deform > 5 mm / yr, demolish high-risk buildings and backfill the soil, and synchronously install InSAR monitoring piles to achieve real-time deformation warning; if L c -L n > 10% and V deform > 3 mm / yr, restrict the floor area ratio of newly built high-rise buildings ≤ 2.5, and use tax incentives to guide enterprises to invest in underground reinforcement projects, and predict new high-altitude remote sensing layer data according to urban planning measures and transmit it to S1; when α > the health threshold of the high-altitude remote sensing layer, it represents a low-risk grid, and development is allowed, but the deformation rate needs to be monitored synchronously; among them, the health threshold of the high-altitude remote sensing layer is the mean value of the historical high-altitude remote sensing layer data influence coefficient and the high-altitude remote sensing layer data influence coefficient under urban specifications.

[0104] When β > the health threshold of the low-altitude UAV layer, it indicates a high thermal risk, and mandatory building facade renovation, additional ventilation corridors need to be carried out, reflective coatings are used on the roof, and vertical greening is implemented to make the solar reflectance ≥ 0.8, the coverage rate of climbing plants ≥ 60%, and the utilization rate of roof area ≥ 30%; if UHI / d > 0.2 °C / m, buildings occupying the air duct are demolished, and ventilation corridors with a width ≥ 50 m are set up; if Dp < 500 points / m 2 , the UAV is supplemented and measured quarterly, the building safety rating is generated in combination with the BIM model, and new low-altitude UAV layer data is predicted according to urban planning measures and transmitted to S1; when β < the health threshold of the low-altitude UAV layer, it indicates a low-risk area, and the original construction is supported, but the UAV point cloud data needs to be updated regularly; among them, the health threshold of the low-altitude UAV layer is the average value of the influence coefficient of historical low-altitude UAV layer data and the influence coefficient of low-altitude UAV layer data under urban regulations.

[0105] When δ > the health threshold of the ground sensor layer, it indicates a high pollution pressure, and traffic restriction policies need to be implemented, the drainage pipe network needs to be upgraded, a green isolation belt with a construction width ≥ 10 m needs to be implemented, plants that can adsorb PM2.5 such as oleander are configured, and traffic command is adjusted according to the traffic flow density to reduce the congestion duration by 25%; if ε > 500 με, CIPP lining is used for repair to extend the service life, distributed optical fibers are used for monitoring, and new ground sensor layer data is predicted according to urban planning measures and transmitted to S1; when δ < the health threshold of the ground sensor layer, it indicates a low-risk area, and the original traffic signal timing needs to be maintained; among them, the health threshold of the ground sensor layer is the average value of the influence coefficient of historical ground sensor layer data and the influence coefficient of ground sensor layer data under urban regulations.

[0106] When γ > the health threshold of the urban operation layer, it indicates high carbon, and distributed photovoltaic coverage needs to be promoted, access to high-energy-consuming industries needs to be restricted, enterprises with a steel tonnage energy consumption exceeding 600 kgce need to be shut down, and zero-carbon data centers need to be introduced; if A < 0.5, a 15-minute community commercial complex is laid out to make the walkability rate reach 100%, with a 50% rent subsidy, and new urban operation layer data is predicted according to urban planning measures and transmitted to S1; when γ < the health threshold of the urban operation layer, it indicates a high-vitality area, and the original commercial complex layout needs to be maintained and the public service density needs to be increased, and the public service investment needs to be increased by 20%; among them, the health threshold of the urban operation layer is specifically the average value of the influence coefficient of historical urban operation layer data and the influence coefficient of urban operation layer data under urban regulations.

[0107] S5, Human-computer interaction, a user interface is constructed based on Web-based interaction technology, and urban planning methods, the core indicators of the four types of urban physical signs, and the target city data are displayed in an intuitive chart form. The change trend of the core indicators in different regions at different times is displayed with a line chart and transmitted to the user data terminal to provide reference data for users to make adjustment measures.

[0108] As attachedFigure 2 As shown in Figure 2 , the present embodiment provides an intelligent city planning system based on multi-source remote sensing data fusion, including:

[0109] Regional division module: Based on GIS geographical coordinates and urban functional zoning data, divide the planning area into M dynamic management grids. Use LiDAR point cloud data to assist in identifying terrain undulations and building heights, adjust the grid boundaries, and establish a two-way mapping relationship between the grids and multi-source perception devices.

[0110] Multi-source data acquisition module: Construct a city holographic data set through global coverage and multi-modal perception technology, collect target city data in real time, and then perform multi-level cleaning, spatio-temporal alignment, and normalization operations on each layer of data using a standardized processing method to obtain city data.

[0111] Data core index calculation module: Establish a mathematical model based on the data obtained from the multi-source data acquisition module after standardization processing, and generate four types of core indicators of urban signs, namely the influence coefficient of high-altitude remote sensing layer data, the influence coefficient of low-altitude drone layer data, the influence coefficient of ground sensor layer data, and the influence coefficient of urban operation layer data.

[0112] Urban planning and feedback module: Based on the four types of core indicators of urban signs obtained by the data core index calculation module, use the hierarchical decision-making method to guide planning strategies in different dimensions respectively, and achieve multi-objective collaborative optimization.

[0113] Human-computer interaction module: Construct a user interface based on Web-based interaction technology, display urban planning methods, four types of core indicators of urban signs, and target city data in an intuitive chart form, display the change trend of core indicators in each region at different times with a line chart, and transmit it to the user data terminal to provide reference data for users to make adjustment measures.

[0114] In a specific example, assuming a renovation project in the old urban area of a medium-sized city, this embodiment uses satellite remote sensing technology to obtain the vegetation index, surface deformation rate, and night light remote sensing light intensity data of this area, with a frequency of once a month, and obtains NDVI = 0.35, V deform = -3mm / y, Lc = 300nW / cm2 / sr, Ln = 700nW / cm 2 / sr; Use drones to regularly collect building roughness, urban heat island effect, and impervious surface ratio data, and conduct data collection once a month to obtain I roughness = 0.65, Dp = 180 points / m 2 , UHI = 5.2°C; Arrange various ground sensors in the urban area to monitor traffic flow density in real time; PM2.5 concentration; underground pipeline deformation rate data, and obtain D traffic = 1200 vehicles / h, C = 0.045g / m 3, ε = 0.2%; Obtain population density, carbon emission intensity, and regional vitality index data from operator data, and get D = 8000 people / km 2 , I = 1.5 tons CO2 / 10,000 yuan, A = 65;

[0115] Then calculate the core indicators. The influence coefficient of the high-altitude remote sensing layer data is specifically:

[0116]

[0117] Substitute the parameters NDVI = 0.35, UHI = 5.2, Lc = 300, Ln = 700, Vdeform = -3,

[0118] Calculate

[0119] The influence coefficient of the low-altitude UAV layer data is specifically:

[0120]

[0121] Substitute the parameters UHI = 5.2, I roughness = 0.65, Dp = 180, NDVI = 0.35,

[0122] Calculate

[0123] The influence coefficient of the ground sensor layer data is specifically:

[0124]

[0125] Substitute the parameter D traffic = 1200, C = 0.045, NDVI = 0.35, ε = 0.2%,

[0126] Calculate

[0127]

[0128] The influence coefficient of the urban operation layer data is specifically:

[0129]

[0130] Substitute the parameters D = 8000, I = 1.5, A = 65, NDVI = 0.35

[0131] Calculate

[0132] Based on the mean values of the core indicators of the four types of urban signs in history and the core indicators of the four types of urban signs under the norms of the target city, the health threshold of the high-altitude remote sensing layer is 2.05, which indicates a high-risk grid. Restricted development is required, and green space compensation should be increased. For high-risk grids, 30% of the construction land is compulsorily reserved and converted into green space, native drought-tolerant vegetation is planted, and an intelligent irrigation system is equipped; and (Lc - Ln) / Ln > 10% and V deform > 3 mm / yr, restrict the new high-rise building plot ratio ≤ 2.5, use tax incentives to guide enterprises to invest in underground reinforcement projects, and predict new high-altitude remote sensing layer data according to urban planning measures and transmit it to S1; when the health threshold of the low-altitude drone layer is 1.37, it indicates a high thermal risk. Compulsory building facade renovation and additional ventilation corridors are required. Use reflective coatings on the roof, conduct three-dimensional greening, make the solar reflectivity ≥ 0.8, the coverage rate of climbing plants ≥ 60%, and the utilization rate of the roof area ≥ 30%; Dp is 180 points / m 2 , the drone is resurveyed quarterly, combined with the BIM model to generate a building safety rating, and predict new low-altitude drone layer data according to urban planning measures and transmit it to S1; the health threshold of the ground sensor layer is -100.47, indicating high pollution pressure. Implement traffic restriction policies, upgrade the drainage network, implement a green isolation belt with a construction width ≥ 10 m, configure plants such as oleander to adsorb PM2.5, adjust traffic command according to the traffic flow density, and reduce the congestion duration by 25%; and predict new ground sensor layer data according to urban planning measures and transmit it to S1; the health threshold of the urban operation layer is 164.2, indicating high carbon. Promote the coverage of distributed photovoltaics, restrict the access of high-energy-consuming industries, shut down enterprises with a steel tonnage energy consumption exceeding 600 kgce, introduce zero-carbon data centers, and predict new urban operation layer data according to urban planning measures and transmit it to S1.

[0133] Finally, a user interface is constructed based on web-based interactive technology, and the urban planning method, the core indicators of the four types of urban signs, and the target city data are displayed in an intuitive chart form. A line chart is used to show the change trend of the core indicators in different regions at different times, and it is transmitted to the user data terminal.

[0134] Secondly: In the accompanying drawings of the disclosed embodiments of the present invention, only the structures related to the disclosed embodiments are involved. Other structures can refer to the general design. Without conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0135] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent urban planning method based on multi-source remote sensing data fusion, characterized in that, Including: S1. Regional division: Based on GIS geographical coordinates and urban functional zoning data, the planning area is divided into M dynamic management grids. LiDAR point cloud data is used to assist in identifying terrain undulations and building heights, adjusting grid boundaries, and establishing a two-way mapping relationship between the grids and multi-source perception devices; S2. Multi-source data collection: A city holographic dataset is constructed through full-domain coverage and multi-modal perception technology to collect real-time data of the target city. Then, multi-level cleaning, spatio-temporal alignment, and standardization processing methods are performed to normalize each layer of data to obtain city data; S3. Calculation of data core indicators: After standardizing the data obtained in S2, a mathematical model is established to generate four types of core indicators of urban signs, namely the influence coefficient of high-altitude remote sensing layer data, the influence coefficient of low-altitude drone layer data, the influence coefficient of ground sensor layer data, and the influence coefficient of urban operation layer data; S4. Urban planning and feedback: Based on the four types of core indicators of urban signs obtained in S3, the hierarchical decision-making method is used to guide planning strategies in different dimensions respectively to achieve multi-objective collaborative optimization; S5. Human-computer interaction: A user interface is constructed based on Web-based interaction technology to display urban planning methods, four types of core indicators of urban signs, and target city data in an intuitive chart form. The line chart is used to display the change trend of core indicators in different regions at different times and is transmitted to the user data terminal to provide reference data for users to make adjustment measures.

2. The intelligent city planning method based on multi-source remote sensing data fusion according to claim 1, characterized in that: The city data includes high-altitude remote sensing layer data, low-altitude drone layer data, ground sensor layer data, and urban operation layer data; The high-altitude remote sensing layer data includes the vegetation index, denoted as NDVI; The surface deformation rate, denoted as Vdeform; the night light remote sensing light intensity, denoted as L; the low-altitude drone layer data includes the building surface roughness index, denoted as Iroughness; the drone point cloud density, denoted as Dp; the heat island intensity, denoted as UHI; the ground sensor layer data includes the traffic flow density, denoted as Dtraffic; the PM2.5 concentration, denoted as C; the underground pipe network deformation rate, denoted as ε; the urban operation layer data includes the population density, denoted as D; the carbon emission intensity, denoted as I; the regional vitality index, denoted as A.

3. The intelligent city planning method based on multi-source remote sensing data fusion according to claim 2, characterized in that: The vegetation index uses the near-infrared and red light bands of optical satellites to collect surface reflectance data, and then corrects the original radiation value based on the solar altitude angle and atmospheric parameters to obtain the vegetation index. Specifically: where ρ nir represents the surface reflectance in the near-infrared band; ρ red represents the surface reflectance in the red band; the specific formula for the radiometric correction of surface reflectance is as follows: where L λ represents the sensor radiation value, d represents the Earth-Sun distance; ESUN λ represents the spectral solar irradiance; θ solar represents the solar altitude angle; The surface deformation rate is calculated by calling the satellite C-band SAR sensor to collect two scenes of SAR images and calculating the surface deformation rate through interferometric measurement technology. Specifically: Where λ is the radar wavelength, Δφ represents the phase difference between the two scenes of SAR images, and T represents the time interval between the two scenes of images; The night light remote sensing light intensity is extracted by collecting the night and day band data of night light remote sensing; The drone point cloud density is obtained by carrying a five-lens oblique camera to collect RGB + near-infrared images with a ground resolution of 2 cm, then generating a three-dimensional point cloud, and recording the attitude angle of the aircraft to correct image distortion. The building contour concavity-convex index generates a digital elevation model by emitting 1064nm laser pulses and combining GNSS positioning data, and extracts the building contour concavity-convex index, specifically: where σ(Z) represents the standard deviation of the building surface elevation, and μ(Z) represents the average value of the building surface elevation; The urban heat island intensity is obtained by using an unmanned aerial vehicle (UAV) equipped with a thermal imager to collect the surface temperatures of the urban center and the suburbs and directly subtracting them; The specific formula for calculating the urban heat island intensity is: UHI = T urban -T rural , Among which T urban represents the average surface temperature of the urban central area, and T rural represents the surface temperature of the suburban reference point; The traffic flow density is detected by deploying millimeter-wave radars and 4K cameras at the intersections of arterial roads to detect vehicle types and speeds in real time and count the traffic flow density, specifically: where N vehicle represents the number of detected vehicles, L road represents the monitored road length; T sample represents the sampling time. The PM2.5 concentration is directly collected by deploying micro-meteorological stations; The monitoring pipeline deformation rate is directly monitored by deploying distributed fiber optic strain sensors along the drainage pipeline; The population density is obtained by generating a population heat map density from anonymous mobile phone signaling data; The carbon emission intensity is calculated by accessing the time series data of the building's electricity consumption power collected by a smart electricity meter (with a sampling interval of 15 minutes) and combining it with the gas meter data. Specifically: I = 0.92×Pe + 2.15×Vg, where Pe is the electricity consumption power and Vg represents the gas consumption; The regional vitality index is calculated by crawling the POI check-in data on Weibo, counting the number of check-ins, and analyzing the text sentiment polarity. Specifically: A = ln[Nc×(Ss + 1)], where Nc represents the number of POI check-ins and Ss represents the sentiment polarity score.

4. The intelligent city planning method based on multi-source remote sensing data fusion according to claim 1, characterized in that: The influence coefficient of the high-altitude remote sensing layer data reflects the inhibitory ability of vegetation cover on the urban heat island effect by dividing the index by the urban heat island effect, adding 0.1 to avoid a zero denominator, and the product relationship reflects the synergistic effect. The logarithmic ratio of surface deformation to the night light economy, the larger the value, the more prominent the contradiction between geological risk and economic expansion, and is used to reflect the ecological-geological-economic balance relationship. Specifically: where L c represents the normal night-time light intensity of remote sensing, and L n represents the night-time light intensity of remote sensing, and NDVI represents the vegetation index; V deform represents the surface deformation rate.

5. The intelligent urban planning method based on multi-source remote sensing data fusion according to claim 1, characterized in that: The influence coefficient of the low-altitude UAV layer data exacerbates the effect of the urban heat island through complex building methods, and its risk can be alleviated by using high-precision point clouds. The surface temperature gradient is inhibited by vegetation, and the dimensional difference is eliminated through the square root and the vegetation index, and is used to reflect the health of the building environment and the microclimate. Specifically: Among which I roughness represents the unevenness index of the building surface; Dp represents the drone point cloud density; UHI represents the urban heat island intensity; NDVI represents the vegetation index.

6. The intelligent city planning method based on multi-source remote sensing data fusion according to claim 1, characterized in that: The influence coefficient of the ground sensor layer reflects the traffic pollution pressure by the product of traffic flow and PM2.5 being reduced by vegetation, and the negative correlation between pipeline deformation and vegetation cover reflects the underground deformation risk. Vegetation is used as a public adjustment factor to achieve dimensional normalization, and is used to reflect the dynamic pressure of traffic-pollution-infrastructure. Specifically: Among them, D traffic represents the traffic flow density; C represents the PM2.5 concentration; ε represents the deformation rate of the underground pipe network; NDVI represents the vegetation index.

7. The intelligent city planning method based on multi-source remote sensing data fusion according to claim 1, characterized in that: The influence coefficient of the urban operation layer data reflects the population carbon footprint pressure by the product of population and carbon emission being doubly inhibited by regional vitality and vegetation, and is used to reflect the population-energy-society collaborative resilience. Specifically: where D represents the urban operation layer data including population density; I represents the carbon emission intensity; A represents the regional vitality index.

8. The intelligent city planning method based on multi-source remote sensing data fusion according to claim 1, characterized in that: The specific hierarchical decision-making method is: Align the dynamically managed grids divided in step S1 with the four-layer data based on the UTM coordinate system, and then define the health thresholds of each coefficient according to historical data and urban specifications; then perform priority judgment based on the high-altitude remote sensing layer > low-altitude drone layer > ground sensor layer > urban operation layer, and then specifically implement urban planning according to the size relationship between each coefficient and the health threshold of each coefficient; then generate new virtual city data according to the urban planning strategy and feedback it to step S2 to obtain new virtual core indicators of the four types of urban signs. The health threshold is specifically the mean value of the core indicators of the four types of urban signs in historical data and the core indicators of the four types of urban signs under the target urban specifications.

9. The intelligent city planning method based on multi-source remote sensing data fusion according to claim 8, wherein: The specific urban planning method is as follows: When α < the health threshold of the high-altitude remote sensing layer, it indicates a high-risk grid, and restricted development is required, and green space compensation is increased. For high-risk grids, 30% of the construction land is compulsorily reserved and converted into green space, native drought-tolerant vegetation is planted, and an intelligent irrigation system is configured; if Vdeform > 5 mm / yr, high-risk buildings are demolished and the earthwork is backfilled, and InSAR monitoring piles are synchronously arranged to achieve real-time deformation warning; if Lc - Ln > 10% and Vdeform > 3 mm / yr, the floor area ratio of newly built high-rise buildings is restricted to ≤ 2.5, and tax incentives are used to guide enterprises to invest in underground reinforcement projects, and new high-altitude remote sensing layer data is predicted according to urban planning measures and transmitted to S1; when α > the health threshold of the high-altitude remote sensing layer, it indicates a low-risk grid, and development is allowed, but the deformation rate needs to be monitored synchronously; the health threshold of the high-altitude remote sensing layer is the mean value of the influence coefficient of historical high-altitude remote sensing layer data and the influence coefficient of high-altitude remote sensing layer data under urban specifications. When β > the health threshold of the low-altitude UAV layer, it indicates a high thermal risk. Compulsory building facade renovation, additional ventilation corridors need to be carried out. Use reflective coatings on the roof, conduct three-dimensional greening, so that the solar reflectance ≥ 0.8, the coverage rate of climbing plants ≥ 60%, and the utilization rate of the roof area ≥ 30%; If UHI / d > 0.2 °C / m, demolish the buildings occupying the air ducts and set up ventilation corridors with a width ≥ 50m; If Dp < 500 points / m 2 , conduct supplementary UAV surveys every quarter, generate building safety ratings in combination with the BIM model, and predict new low-altitude UAV layer data according to urban planning measures and transmit it to S1; When β < the health threshold of the low-altitude drone layer, it indicates a low-risk area, and the original construction is supported, but the drone point cloud data needs to be updated regularly; the health threshold of the low-altitude drone layer is the mean value of the influence coefficient of historical low-altitude drone layer data and the influence coefficient of low-altitude drone layer data under urban specifications. When δ > the health threshold of the ground sensor layer, it indicates high pollution pressure, and a traffic restriction policy needs to be implemented, the drainage network needs to be upgraded, a green isolation belt with a width of ≥ 10 m needs to be built, plants that can adsorb PM2.5 such as oleander are configured, and traffic command is adjusted according to the traffic flow density to reduce the congestion duration by 25%; if ε > 500 με, CIPP lining is used for repair to extend the service life, distributed optical fiber is used for monitoring, and new ground sensor layer data is predicted according to urban planning measures and transmitted to S1; when δ < the health threshold of the ground sensor layer, it indicates a low-risk area, and the original traffic signal timing needs to be maintained. The health threshold of the ground sensor layer is the mean value of the influence coefficient of historical ground sensor layer data and the influence coefficient of ground sensor layer data under urban specifications. When γ > the health threshold of the urban operation layer, it indicates high carbon. It is necessary to promote the coverage of distributed photovoltaics, restrict the access of high-energy-consuming industries, shut down enterprises with a steel tonnage energy consumption exceeding 600 kgce, and introduce zero-carbon data centers. If A < 0.5, layout 15-minute community commercial complexes to achieve a 100% walkability rate, provide a 50% rent subsidy, and predict new urban operation layer data according to urban planning measures and transmit it to S1. When γ < the health threshold of the urban operation layer, it indicates a high-vitality area. It is necessary to maintain the original layout of commercial complexes and increase the density of public services, and increase public service investment by 20%. Among them, the health threshold of the urban operation layer is specifically the mean value of the historical urban operation layer data influence coefficient and the urban operation layer data influence coefficient under urban regulations.

10. An intelligent urban planning system based on multi-source remote sensing data fusion, for implementing the intelligent urban planning method based on multi-source remote sensing data fusion according to any one of claims 1-9 above, characterized in that, Including: Regional division module: Based on GIS geographical coordinates and urban functional zoning data, divide the planned area into M dynamic management grids. Use LiDAR point cloud data to assist in identifying terrain undulations and building heights, adjust grid boundaries, and establish a two-way mapping relationship between grids and multi-source sensing devices. Multi-source data collection module: Construct a holographic urban dataset through full-domain coverage and multi-modal sensing technology, collect target urban data in real time, and then perform multi-level cleaning, spatio-temporal alignment, and normalization operations on each layer of data using a standardized processing method to obtain urban data. Data core index calculation module: Based on the data obtained from the multi-source data collection module, establish a mathematical model after standardization processing to generate four types of core indicators of urban signs, namely the high-altitude remote sensing layer data influence coefficient, the low-altitude drone layer data influence coefficient, the ground sensor layer data influence coefficient, and the urban operation layer data influence coefficient. Urban planning and feedback module: Based on the four types of core indicators of urban signs obtained from the data core index calculation module, use the hierarchical decision-making method to guide planning strategies in different dimensions respectively to achieve multi-objective collaborative optimization. Human-computer interaction module: Build a user interface based on Web-based interaction technology, display urban planning methods, four types of core indicators of urban signs, and target urban data in an intuitive chart form, show the change trend of core indicators in each region at different times with a line chart, and transmit it to the user data terminal to provide reference data for users to make adjustment measures.

Images