An Automatic Generation Method for Urban Residential Area Form Based on Spatial Entropy

Abstract

The present invention provides an automatic generation method for urban residential area form based on spatial entropy, which relates to the field of urban planning. The automatic generation method for urban residential area form based on spatial entropy includes input and standardization of geospatial information; constructing a sample library of residential area forms to generate multiple schemes of site forms; constructing a calculation rule library of spatial entropy to calculate the spatial entropy values of multiple schemes of site forms; calculating the reasonable range and optimal value of the spatial entropy value of the residential area; screening N preferred schemes; constructing a human-computer interaction instruction library to display and compare multiple schemes of site forms; and result output. It solves the problems that the quantity of design results of automatic generation of residential area form is huge, and it is difficult for manual evaluation and screening, with problems such as low evaluation efficiency, long cycle, and low interactivity, realizes the methods of automatic generation, evaluation and screening of urban residential areas, and displays the screening results and gives real-time adjustment feedback, improving the rationality of the design results of automatic generation of residential area form.

Description

Technical Field

[0001] The present invention relates to the technical field of urban planning, and particularly to an automatic generation method for urban residential area form based on spatial entropy. Background Art

[0002] The development of digital technology provides important technical support for the generation of urban regular plot forms. The traditional methods for generating urban residential area forms mainly rely on theories such as morphology, and are digital-driven parametric designs through steps such as refining residential area prototypes, constructing rule bases, building program algorithm models, performance evaluation, and optimization. This can effectively solve problems such as plot division, road and building generation in residential area design. However, based on the existing learning of residential area forms, the number of automatically generated design results is huge, and it is difficult for manual evaluation and screening. There are problems such as low evaluation efficiency, long cycle, and low interactivity. Currently, there is a lack of a method for automatically generating and screening urban residential areas. Urban residential areas can be regarded as regular complex systems, and the entropy model provides a solution idea for evaluation and optimization in the generation of urban residential area forms. Summary of the Invention

[0003] (1) Technical Problems to be Solved

[0004] In view of the deficiencies of the prior art, the present invention provides an automatic generation method for urban residential area form based on spatial entropy. By constructing an automatic generation system for urban residential area form, it solves problems such as low evaluation efficiency, long cycle, and low interactivity in the automatic generation of urban residential areas, realizes the methods for automatic generation, evaluation, and screening of urban residential areas, and displays the screening results and provides real-time adjustment feedback, providing a scientific basis for the evaluation of urban residential area forms, and having certain social and economic benefits.

[0005] (2) Technical Solutions

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions:

[0007] On the one hand, it provides an automatic generation method for urban residential area form based on spatial entropy, including:

[0008] Collect and input the geospatial information data and planning and design condition indicators of the site, and store them in the database and the indicator library respectively;

[0009] Construct a sample library of residential area forms, input multiple cases of residential area forms, generate multiple schemes of site forms through image learning, and further generate three-dimensional building volume models corresponding to the multiple schemes and the corresponding plot ratios, densities, and maximum building heights. The cases include building layout, height information, plot ratio, density, and maximum building height;

[0010] Build a spatial entropy calculation rule library, which includes the building area complexity Ep, the building height complexity Eh, the building orientation complexity Eo, and the building interval complexity Es. Input the multi-scheme three-dimensional building block model and calculate its spatial entropy;

[0011] Calculate the spatial entropy values of the residential area form sample library under different planning and design conditions respectively, and select the middle 80% of all spatial entropy values under different planning and design conditions as the reasonable spatial entropy value interval; cluster the reasonable spatial entropy value interval, and calculate the median of each category as the optimal spatial entropy value;

[0012] Input the spatial entropy values of multiple schemes of the site form and the corresponding plot ratios, densities, and maximum building heights. Screen the schemes according to the planning and design condition indicators, and further screen the schemes according to the optimal spatial entropy value;

[0013] Input the schemes further screened according to the optimal spatial entropy value into a mixed reality head-mounted device that can accurately locate and orient, build a human-computer interaction instruction library, and determine the final scheme through the mixed reality head-mounted device in the actual scenario;

[0014] Output the standard floor plan, the three-dimensional building block model of the scheme, and the scheme indicators of the selected scheme.

[0015] Preferably, the collection and input of the geospatial information data and planning and design condition indicators of the site, and the storage in the database and the indicator library respectively include:

[0016] Obtain the current geospatial information data of the site. The current geospatial information data of the site includes site boundary data, terrain data, building data, land use function data, and road traffic data. Among them, the urban terrain data and urban building data are collected by the GeoSLAM ZEB Discovery mobile laser panoramic three-dimensional scanning system, and the site boundary data, land use function data, and road traffic data are obtained from relevant departments;

[0017] Input the current geospatial information data of the site into a geospatial information processing platform. The geospatial information processing platform refers to a computer system that can input, store, query, analyze, and display geospatial data, and is a platform for analyzing and processing spatial information. Input the geospatial data through the geospatial information processing platform, convert the data into a digital form for subsequent spatial analysis. And input it into the memory and store it in the database. Among them, the shp file of the site boundary data contains boundary contour, location, and area information, the shp file of the terrain data contains elevation and location information, the shp file of the building data contains building contour, number of building floors, building height, and building location information, the shp file of the land use function data contains land use nature, land use boundary, and location information, and the line elements of the shp file of the road traffic data contain road grade, red line width, and location information;

[0018] Obtain the planning and design condition indicators of the site and store them in the indicator library. The planning and design condition indicators include land area, land use nature, land use ratio, land location, plot ratio, building density, building height, green space rate, land use compatibility requirements, building setbacks, opening requirements, sunshine requirements, and regional conditions.

[0019] Preferably, construct a residential area form sample library, input multiple residential area form cases, generate multiple site form solutions through image learning, and further generate a three-dimensional building volume model of multiple solutions and the corresponding plot ratio, density, and maximum building height. The cases include building layout, height information, plot ratio, density, and maximum building height, specifically including:

[0020] Input 10,000 residential area form cases into the residential area form sample library of a computer workstation with a computing power of 5 petaFLOPS and a memory of 320 GB. The sample library contains a plan view of the residential area design plan with a scale of 1:1000 and a resolution of 1680*1050. The residential area design plan should include building information, road information, water system information, and green space information. Among them, the road information is a planar information with an RGB value of 255-255-253, the green space information is a planar information with an RGB value of 245-250-221, the water system information is a planar information with an RGB value of 61-98-132, and the building information is a planar information containing building floor data. Among them, the RGB value of the 1st to 3rd floors is 232-232-221, the RGB value of the 4th to 6th floors of the building is 206-206-201, the RGB value of the 7th to 9th floors is 177-178-175, the RGB value of the 7th to 9th floors is 152-151-149, the RGB value of the 10th to 18th floors is 126-126-123, and the RGB value of the 19th to 26th floors is 110-110-107; where the RGB value represents height information;

[0021] Learn the sample cases of the urban design plan case library through a convolutional neural network, input the site boundary data and rasterize it, and further learn to generate the building layout and height information to obtain multiple site solutions. Vectorize the multiple site solutions to generate a three-dimensional volume model of the plan and its corresponding plot ratio, building density, and maximum building height; the sample cases include building layout, height information, plot ratio, density, and maximum building height.

[0022] Preferably, construct a spatial entropy calculation rule library, which includes the building area complexity Ep, the building height complexity Eh, the building orientation complexity Eo, and the building interval complexity Es. Input the three-dimensional building volume model of multiple solutions and calculate its spatial entropy, specifically including:

[0023] Construct a spatial entropy calculation rule library, where the spatial entropy includes the building area complexity Ep, the building height complexity Eh, the building orientation complexity Eo, and the building interval complexity Es. The specific calculation formulas are as follows. The building area complexity where P t is the ground floor building area of building numbered t, and P is the average ground floor building area; the building height complexity where H t is the building height of building numbered t, and H is the average building height; the building orientation complexity where O t is the building orientation angle of building numbered t, and O is the average building orientation angle; the building interval complexity where D t is the adjacent building distance between building numbered t and the surrounding buildings (plot boundaries), and D is the average adjacent distance;

[0024] Input the multi-scheme three-dimensional building block model, calculate the spatial entropy values of multiple schemes of the site form according to the spatial entropy calculation rule library, and input them into the geographic information database.

[0025] Preferably, calculate the spatial entropy values of the residential form sample library under different planning and design conditions respectively, and select the middle 80% part of all spatial entropy values under different planning and design conditions as the reasonable spatial entropy value interval; cluster the reasonable spatial entropy value interval, and calculate the median of each category as the optimal spatial entropy value. Specifically include:

[0026] Establish a residential form sample library, classify the cases according to the climate zone where the case is located, the highest building height of the case, the plot ratio, and the building density data;

[0027] For the cases in the residential form sample library, calculate the spatial entropy values of the residential areas under different planning and design conditions respectively; the spatial entropy includes the building area complexity E p , the building height complexity E h , the building orientation complexity E o , and the building interval complexity E s ;

[0028] For the samples in the residential form sample library, eliminate the smallest and largest 10% parts of the spatial entropy values in each category from small to large, and retain the middle 80% part of the spatial entropy values of each category as the reasonable spatial entropy value interval, and store it in the storage;

[0029] Cluster the reasonable spatial entropy value intervals obtained from the samples in the residential form sample library, screen those with an Euclidean distance within 0.5 as one category, calculate the median of each category as the optimal spatial entropy value, and store it in the storage.

[0030] Preferably, for the multi - scheme spatial entropy values of the input site form and the corresponding plot ratio, density, and maximum building height, screening the schemes according to the planning and design condition indicators, and further screening the schemes according to the optimal spatial entropy value specifically includes:

[0031] Selecting a matching residential form sample library through the planning and design conditions to obtain the reasonable spatial entropy value range and the optimal spatial entropy value corresponding to the site; the planning and design conditions include the climate zone where the case is located, the maximum building height of the case, the plot ratio, and the building density data;

[0032] Comparing the generated multi - scheme spatial entropy of the site form with the optimal spatial entropy value of the corresponding residential form sample library type. If the sum of the building area complexity E p , building height complexity E h , building orientation complexity E o , and building interval complexity E s is closest to the optimal spatial entropy value of the corresponding residential form sample library type, then save the scheme as a reasonable scheme, and store the reasonable scheme as an optimal scheme, with a total of N.

[0033] Preferably, further screening the scheme according to the optimal spatial entropy value, inputting it into a mixed - reality head - mounted device that can accurately position and orient, constructing a human - machine interaction instruction library, and determining the final scheme through comparison in the actual scene by the mixed - reality head - mounted device specifically includes:

[0034] Inputting the N optimal schemes obtained by screening into a mixed - reality head - mounted device that can accurately position and orient;

[0035] Positioning and orienting according to the position of the mixed - reality head - mounted device, and outputting the corresponding image to be superimposed on the mixed - reality head - mounted device and the real scene;

[0036] Constructing a gesture instruction library, including operations such as selecting / canceling selection / stretching / switching schemes / determining the final scheme through a data glove;

[0037] Viewing the effects of each scheme in the actual scene through the mixed - reality head - mounted device and comparing, and switching different schemes through the data glove to finally determine the scheme.

[0038] Preferably, outputting the standard floor plan, the 3D building mass model, and the scheme indicators of the selected scheme specifically includes:

[0039] Outputting the standard floor plan of the selected final scheme in the database;

[0040] Exporting the 3D building mass model and the scheme indicators corresponding to the final scheme.

[0041] On the other hand, a device is provided, and the device includes:

[0042] One or more processors;

[0043] A memory for storing one or more programs,

[0044] When the one or more programs are executed by the one or more processors, the one or more processors are caused to execute the described method for automatically generating urban settlement forms based on spatial entropy.

[0045] In another aspect, there is provided a computer-readable storage medium storing a computer program, which when executed by a processor implements the described method for automatically generating urban settlement forms based on spatial entropy.

[0046] (III) Advantageous Effects

[0047] (1) The method for automatically generating urban settlement forms based on spatial entropy in the present invention reduces the economic and human costs required for the preliminary comparison and selection of settlement design schemes. For the screening of multiple schemes for the intelligent generation of settlement planning and design, it takes 3 days. The present invention establishes the optimal values of multiple indicators of settlement spatial entropy for automatic scheme screening, further screens and optimizes the schemes, shortens the manual screening time to 1 hour, and improves the efficiency by 72 times.

[0048] (2) The method for automatically generating urban settlement forms based on spatial entropy in the present invention realizes the interactive modification and comparison of settlement schemes. The planner can select, move and scale buildings in real time through gesture commands. By quickly and efficiently selecting and adjusting schemes on the site, it can effectively avoid the process of the planner conducting on-site research and then returning to the computer for modification and adjustment. The traditional method usually takes at least 2 days, and through on-site adjustment, it can be shortened to within 2 hours, and the efficiency is increased by 24 times. Description of the Drawings

[0049] Figure 1 It is the overall method flow and equipment framework diagram of the embodiment of the present invention;

[0050] Figure 2 It is the schematic diagram of the principle of spatial entropy calculation of the present invention;

[0051] Figure 3 It is the schematic diagram of multi-scheme spatial entropy of the embodiment of the present invention;

[0052] Figure 4 It is the three-dimensional space interaction schematic diagram of the present invention. Detailed Embodiments

[0053] Next, in combination with the accompanying drawings of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. 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.

[0054] Embodiment

[0055] As Figure 1 shown, the embodiment of the present invention discloses an automatic generation method for urban settlement form based on spatial entropy. Taking the urban design of a certain plot as an example and with the accompanying drawings, the technical solutions of the present invention will be described in detail, including the following steps:

[0056] Step S1: Collect and input the geographical spatial information data and planning and design condition indicators of the site, and store them in the database and the indicator library respectively;

[0057] Step S2: Construct a settlement form sample library, input 10,000 settlement form cases, generate multiple site form options through image learning, and further generate multiple three-dimensional building block models and the corresponding plot ratios, densities, and maximum building heights. The cases include building layout, height information (RGB values), plot ratio, density, and maximum building height;

[0058] Step S3: Construct a spatial entropy calculation rule library, including the building area complexity Ep, building height complexity Eh, building orientation complexity Eo, and building interval complexity Es. Input the multiple three-dimensional building block models and calculate their spatial entropy;

[0059] Step S4: Calculate the spatial entropy values of the settlement form sample library under different planning and design conditions respectively, and select the middle 80% part of all the spatial entropy values under different planning and design conditions as the reasonable spatial entropy value interval; cluster the reasonable spatial entropy value interval and calculate the median of each category as the optimal spatial entropy value.

[0060] Step S5: Input the spatial entropy values of the multiple site form options and the corresponding plot ratios, densities, and maximum building heights, screen the options according to the planning and design condition indicators, and further screen the options according to the optimal spatial entropy value;

[0061] Step S6: Input the above reasonable options into a mixed reality head-mounted device capable of precise positioning and orientation, construct a human-computer interaction instruction library, and the designer selects the final option through the mixed reality head-mounted device in the actual scene;

[0062] Step S7: Output the standard plan view, the three-dimensional building block model of the option, and the option indicators of the selected option.

[0063] The above-mentioned step S1 includes three steps: S1-1, S1-2, and S1-3:

[0064] Step S1-1: Obtain the current geographical spatial information data of the site. The current geographical spatial information data of the site includes site boundary data, terrain data, building data, land use function data, and road traffic data. Among them, the urban terrain data and urban building data are collected by the GeoSLAM ZEB Discovery mobile laser panoramic three-dimensional scanning system, and the site boundary data, land use function data, and road traffic data are obtained from relevant departments;

[0065] Step S1-2: Input the current geographical spatial information data of the site into the geographical information processing platform. The shp file of the site boundary data should contain boundary contour, location, and area information. The shp file of the terrain data should contain elevation and location information. The shp file of the building data should contain building contour, number of building floors, building height, and building location information. The shp file of the land use function data should contain land use nature, land use boundary, and location information. The line elements of the shp file of the road traffic data should contain road grade, red line width, and location information. Input the above standardized data into the memory and store it in the database;

[0066] Step S1-3: Obtain the planning and design condition indicators of the site. The planning and design condition indicators include land area, land use nature, land use ratio, land use location, floor area ratio, building density, building height, green space rate, land use compatibility requirements, building setbacks, opening requirements, sunshine requirements, regional conditions (the building climate zone where the site is located). The above data are obtained from the regulatory detailed planning, construction detailed planning, and relevant planning departments of the plot and stored in the indicator library;

[0067] The above-mentioned step S2 includes three steps: S2-1, S2-2, and S2-3:

[0068] Step S2-1: Input 10,000 residential form cases into the residential form sample library of a computer workstation with a computing power of 5 petaFLOPS and a memory of 320 GB. The sample library contains the plan view of the residential area design scheme with a scale of 1:1000 and a resolution of 1680*1050. The residential area design plan view should include building information, road information, water system information, and green space information. Among them, the road information is planar information with an RGB value of 255-255-253, the green space information is planar information with an RGB value of 245-250-221, the water system information is planar information with an RGB value of 61-98-132, and the building information is planar information containing building floor data. Among them, for low-rise (1 to 3 floors), the RGB value is 232-232-221, for multi-story class I (4 to 6 floors) buildings, the RGB value is 206-206-201, for multi-story class II (7 to 9 floors), the RGB value is 177-178-175, for multi-story class II (7 to 9 floors), the RGB value is 152-151-149, for high-rise class I (10 to 18 floors), the RGB value is 126-126-123, and for high-rise class I (19 to 26 floors), the RGB value is 110-110-107;

[0069] Step S2-2: Learn the sample cases in the urban design scheme case library through a convolutional neural network, input the site boundary data and rasterize it, and further learn to generate building layout and height information to obtain multiple site plans. Vectorize the above multiple site plans to generate a 3D volume model of the plan and its corresponding plot ratio, building density, and maximum building height. The sample cases include building layout, height information (RGB value), plot ratio, density, and maximum building height;

[0070] The above step S3 includes two steps: S3-1 and S3-2:

[0071] Step S3-1: Construct a spatial entropy calculation rule library. The spatial entropy includes the building area complexity Ep, the building height complexity Eh, the building orientation complexity Eo, and the building interval complexity Es. The specific calculation formulas are as follows. The building area complexity where P t is the ground floor building area of building numbered t, and P is the average ground floor building area; the building height complexity where H t is the building height of building numbered t, and H is the average building height; the building orientation complexity where O t is the building orientation angle of building numbered t, and O is the average building orientation angle; the building interval complexity where D t is the adjacent building distance between building numbered t and the surrounding buildings (plot boundary), and D is the average adjacent distance. The calculation principle is asFigure 2 as shown;

[0072] Step S3-2: Input the multi-scheme three-dimensional building block model, calculate the spatial entropy values of multiple schemes of the site form according to the spatial entropy calculation rule base, and input them into the geographic information database.

[0073] The above-mentioned step S4 includes four steps: S4-1, S4-2, S4-3, and S4-4:

[0074] Step S4-1: Establish a residential area form sample library, and classify the cases according to the climate zone where the case is located, the maximum building height of the case, the plot ratio, and the building density data (Table 1). The climate zone is based on the building climate zoning I, II, III, IV, V, VI, and VII in the "Code for Planning and Design of Urban Residential Areas" GB50180-2018. The maximum building height is divided into low-rise of 12m and below, multi-story of 12m - 30m, and high-rise of 30m and above according to the above standards. The plot ratio is divided into 1.2 and below, 1.2 - 2.7, and 2.7 and above according to the above standards. The building density is divided into 28% and above, 20% - 28%, and 20% and below according to the above standards.

[0075] Table 1 Classification Table of Residential Area Form Sample Library

[0076]

[0077]

[0078]

[0079] Step S4-2: For the cases in the residential area form sample library, calculate the spatial entropy values of the residential areas under 189 different planning and design conditions respectively by the spatial entropy calculation method in step S3. The spatial entropy includes the building area complexity E p , the building height complexity E h , the building orientation complexity E o , the building interval complexity E s .

[0080] Step S4-3: For the 189 types of samples in the residential area form sample library, eliminate the 10% part with the smallest and largest spatial entropy values in each type of sample from small to large, and retain the middle 80% part of the spatial entropy values of each type of sample as the reasonable spatial entropy value interval, and store it in the memory.

[0081] Step S4-4: Cluster the reasonable spatial entropy value intervals obtained from the 189 types of samples in the residential area form sample library, screen those with an Euclidean distance within 0.5 as one category, calculate the median of each category as the optimal spatial entropy value, and store it in the memory. The results are shown in Table 2.

[0082] Table 2 Optimal Spatial Entropy Values of Residential Area Morphology Sample Database

[0083]

[0084] The above-mentioned step S5 includes two steps: S5-1 and S5-2

[0085] Step S5-1: Select the matching residential area morphology sample database in S4 through the planning and design conditions to obtain the reasonable spatial entropy value range and the optimal spatial entropy value corresponding to the site. The planning and design conditions include the climate zone where the case is located, the maximum building height of the case, the plot ratio, and the building density data

[0086] Step S5-2: Compare the spatial entropy of the multiple-site morphology schemes generated in S3 with the optimal spatial entropy value of the corresponding residential area morphology sample database type. If the sum of the floor area complexity E p 、building height complexity E h 、building orientation complexity E o 、building spacing complexity E s of the scheme is closest to the optimal spatial entropy value of the corresponding residential area morphology sample database type, then save the scheme as a reasonable scheme and store the above scheme as a preferred scheme, a total of N, as Figure 3 shown

[0087] The above-mentioned step S6 includes four steps: S6-1, S6-2, S6-3, and S6-4

[0088] Step S6-1: Input the N preferred schemes screened in S5-2 into a mixed reality head-mounted device capable of precise positioning and orientation

[0089] Step S6-2: Perform positioning and orientation according to the position of the mixed reality head-mounted device, and output the corresponding image to the mixed reality head-mounted device to overlay with the real scene

[0090] Step S6-3: Construct a gesture instruction library, including operations such as selecting / canceling selection / stretching / switching schemes / determining the final scheme through a data glove. The detailed operation methods are shown in the following table

[0091] Table 3 Operation Method Table of Data Glove

[0092]

[0093]

[0094] Step S6-4: The designer views the effects of each scheme in the actual scene through the mixed reality head-mounted device and makes comparisons and selections, and switches different schemes through the data glove to finally determine the scheme, as Figure 4 shown

[0095] The above-mentioned step S7 includes two steps: S7-1 and S7-2:

[0096] Step S7-1: Output the standard floor plan of the final solution selected in S6-4 in the database;

[0097] Step S7-2: Export the 3D building block model and the solution indicators corresponding to the above-mentioned final solution.

[0098] As another embodiment of the present invention, there is provided a device, the device includes:

[0099] One or more processors;

[0100] A memory for storing one or more programs,

[0101] When the one or more programs are executed by the one or more processors, the one or more processors are caused to execute a method for automatically generating an urban settlement form based on spatial entropy in the above-mentioned embodiment.

[0102] As another embodiment of the present invention, there is provided a computer-readable storage medium storing a computer program, and when the program is executed by a processor, it implements a method for automatically generating an urban settlement form based on spatial entropy in the above-mentioned embodiment.

[0103] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method,

[0104] article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of another identical element in the process, method, article or device including the element.

Claims

1. An automatic generation method for urban residential area form based on spatial entropy, characterized in that Including: Collect and input the geospatial information data and planning and design condition indicators of the site, and store them in the database and the indicator library respectively; Construct a residential area form sample library, input multiple residential area form cases, generate multiple site form options through image learning, and further generate three-dimensional building block models corresponding to the multiple options and the corresponding plot ratios, densities, and maximum building heights. The cases include building layouts, height information, plot ratios, densities, and maximum building heights; Build a spatial entropy calculation rule library, where the spatial entropy includes the complexity of building floor area E p , the complexity of building height E h , the complexity of building orientation E o , the complexity of building spacing E s , input a multi-scheme three-dimensional building block model and calculate its spatial entropy. The specific calculation formula is as follows: Floor area complexity , where P t is the ground floor area of building numbered t, and P is the average ground floor area; Building height complexity , where H t is the building height of building numbered t, and H is the average building height; Building orientation complexity , where O t is the building orientation angle of building numbered t, and O is the average building orientation angle; Building spacing complexity , where D t is the adjacent building distance between building numbered t and surrounding buildings, and D is the average adjacent distance; Input the three-dimensional building block models of multiple options, calculate the spatial entropy values of the multiple site form options according to the spatial entropy calculation rule library, and input them into the geospatial information database; Calculate the spatial entropy values of the residential area form sample library under different planning and design conditions respectively, and select the middle 80% of all spatial entropy values under different planning and design conditions as the reasonable spatial entropy value interval; cluster the reasonable spatial entropy value interval, and calculate the median of each category as the optimal spatial entropy value; Input the spatial entropy values of the multiple site form options and the corresponding plot ratios, densities, and maximum building heights, screen the options according to the planning and design condition indicators, and further screen the options according to the optimal spatial entropy value; Input the options further screened according to the optimal spatial entropy value into a mixed reality head-mounted device capable of precise positioning and orientation, construct a human-computer interaction instruction library, and determine the final option through comparison in the actual scene by the mixed reality head-mounted device; Output the standard floor plan, the three-dimensional building block model of the option, and the option indicators of the selected option.

2. The automatic generation method of urban settlement form based on spatial entropy according to claim 1, characterized in that: The specific steps of collecting and inputting the geospatial information data and planning and design condition indicators of the site and storing them in the database and the indicator library respectively include: Obtain the current geospatial information data of the site. The current geospatial information data of the site includes site boundary data, terrain data, building data, land use function data, and road traffic data. Among them, urban terrain data and urban building data are collected by the GeoSLAM ZEB Discovery mobile laser panoramic three-dimensional scanning system, and site boundary data, land use function data, and road traffic data are obtained from relevant departments; Store the current geospatial information data of the site in the geospatial information processing platform and input it into the memory for storage in the database. Among them, the site boundary data includes boundary contour, location, and area information, the terrain data includes elevation and location information, the building data includes building contour, number of building floors, building height, and building location information, the land use function data includes land use nature, land use boundary, and location information, and the line elements of the road traffic data include road grade, red line width, and location information; Obtain the planning and design condition indicators of the site and store them in the indicator library. The planning and design condition indicators include land area, land use nature, land use ratio, land use location, plot ratio, building density, building height, green space rate, land use compatibility requirements, building setback, opening requirements, sunshine requirements, and regional conditions.

3. The automatic generation method of urban residential area form based on spatial entropy according to claim 2, characterized in that: Construct a residential form sample library, input multiple residential form cases, generate multiple site form solutions through image learning, and further generate three-dimensional building block models corresponding to the solutions, as well as the corresponding plot ratios, densities, and maximum building heights. The cases include building layout, height information, plot ratio, density, and maximum building height, specifically including: Input 10,000 residential form cases into the residential form sample library of a computer workstation with a computing power of 5 petaFLOPS and a memory of 320 GB. The sample library contains a plan view of the residential area design plan with a scale of 1:1000 and a resolution of 1680 1050. The residential area design plan should include building information, road information, water system information, and green space information. Among them, the road information is a planar information with an RGB value of 255-255-253, the green space information is a planar information with an RGB value of 245-250-221, the water system information is a planar information with an RGB value of 61-98-132, and the building information is a planar information containing building floor data. Among them, the RGB value of the 1st to 3rd floors is 232-232-221, the RGB value of the 4th to 6th floors is 206-206-201, the RGB value of the 7th to 9th floors is 177-178-175, the RGB value of the 7th to 9th floors is 152-151-149, the RGB value of the 10th to 18th floors is 126-126-123, and the RGB value of the 19th to 26th floors is 110-110-107; where the RGB value represents height information; Learn sample cases in the urban design solution case library through a convolutional neural network, input site boundary data and rasterize it, and further learn to generate building layout and height information to obtain multiple site solutions. Vectorize the multiple site solutions to generate a three-dimensional block model of the solution and its corresponding plot ratio, building density, and maximum building height. The sample cases include building layout, height information, plot ratio, density, and maximum building height.

4. A method for automatically generating urban residential area forms based on spatial entropy according to claim 1, characterized in that: Calculate the spatial entropy values of the residential form sample library under different planning and design conditions respectively, and select the middle 80% of all spatial entropy values under different planning and design conditions as the reasonable spatial entropy value range. Cluster the reasonable spatial entropy value range and calculate the median of each cluster as the optimal spatial entropy value, specifically including: Establish a residential form sample library, classify the cases according to the climate zone where the cases are located, the maximum building height of the cases, plot ratio, and building density data. For the cases in the residential area form sample library, calculate the spatial entropy values of the residential areas under different planning and design conditions respectively; the spatial entropy includes the building area complexity E p , the building height complexity E h , the building orientation complexity E o , the building spacing complexity E s ; For the samples in the residential form sample library, remove the smallest and largest 10% of the spatial entropy values in each category from smallest to largest, and retain the middle 80% of the spatial entropy values in each category as the reasonable spatial entropy value range, and store it in the storage. Cluster the reasonable spatial entropy value ranges obtained from the samples in the residential form sample library, select those with an Euclidean distance within 0.5 as one category, calculate the median of each category as the optimal spatial entropy value, and store it in the storage.

5. The automatic generation method of urban settlement form based on spatial entropy according to claim 4, characterized in that: Input the spatial entropy values of the multiple site form solutions and the corresponding plot ratios, densities, and maximum building heights, screen the solutions according to the planning and design condition indicators, and further screen the solutions according to the optimal spatial entropy value, specifically including: Select a matching residential form sample library through the planning and design conditions to obtain the corresponding reasonable spatial entropy value range and the optimal spatial entropy value for the site. The planning and design conditions include the climate zone where the case is located, the maximum building height of the case, plot ratio, and building density data. Compare the spatial entropy of the generated multiple site form solutions with the optimal spatial entropy value of the corresponding residential form sample library type. If the sum of the building area complexity E p , building height complexity E h , building orientation complexity E o , and building spacing complexity E s of the solution is closest to the optimal spatial entropy value of the corresponding residential form sample library type, then save the solution as a reasonable solution, with a total of N solutions.

6. The automatic generation method of urban settlement form based on spatial entropy according to claim 5, characterized in that: Input the solutions further screened according to the optimal spatial entropy value into a mixed reality head-mounted device that can accurately locate and orient, construct a human-computer interaction instruction library, and determine the final solution through the mixed reality head-mounted device in the actual scene, specifically including: Input the N reasonable solutions obtained by screening into a mixed reality head-mounted device that can accurately locate and orient. Locate and orient according to the position of the mixed reality head-mounted device, and output the corresponding image to the mixed reality head-mounted device to be superimposed on the real scene. Construct a gesture instruction library, including operations such as selecting / canceling selection / stretching / switching solutions / determining the final solution through a data glove. View the effects of each solution in the actual scene through the mixed reality head-mounted device and compare them, and switch different solutions through the data glove to finally determine the solution.

7. A method for automatically generating urban residential area form based on spatial entropy according to claim 6, characterized in that: Output the standard floor plan, three-dimensional building block model of the solution, and solution indicators of the selected solution, specifically including: Output the standard floor plan of the selected final solution in the database. Export the 3D building block model corresponding to the final solution and the solution indicators.

8. An automatic generation device for urban residential area form based on spatial entropy, characterized in that, The device includes: One or more processors; A memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors are caused to execute a method for automatically generating an urban settlement form based on spatial entropy as described in any one of claims 1-7.

9. A computer-readable storage medium storing a computer program, characterized in that, When the program is executed by the processor, it implements a method for automatically generating an urban settlement form based on spatial entropy as described in any one of claims 1-7.

Images