WO2023185458A1 - 建筑方案生成的方法、装置、计算机设备以及存储介质 - Google Patents

建筑方案生成的方法、装置、计算机设备以及存储介质 Download PDF

Info

Publication number
WO2023185458A1
WO2023185458A1 PCT/CN2023/081435 CN2023081435W WO2023185458A1 WO 2023185458 A1 WO2023185458 A1 WO 2023185458A1 CN 2023081435 W CN2023081435 W CN 2023081435W WO 2023185458 A1 WO2023185458 A1 WO 2023185458A1
Authority
WO
WIPO (PCT)
Prior art keywords
building
target
land
information
candidate
Prior art date
Application number
PCT/CN2023/081435
Other languages
English (en)
French (fr)
Inventor
何宛余
陈天才
杨大宇
刘镭
王子艾
杨小荻
魏启赟
Original Assignee
深圳小库科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳小库科技有限公司 filed Critical 深圳小库科技有限公司
Publication of WO2023185458A1 publication Critical patent/WO2023185458A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/08—Construction

Definitions

  • the present application relates to the field of computer technology, and specifically to a method, device, computer equipment and storage medium for generating an architectural plan.
  • Embodiments of the present application provide a method, device, computer equipment, and storage medium for generating architectural plans, which can automatically generate architectural plans based on relevant information, which is beneficial to improving the efficiency of architectural design.
  • embodiments of the present application provide a method for generating an architectural plan, including:
  • embodiments of the present application provide a device for generating architectural plans, including:
  • the acquisition module is used to obtain the generated design information of the target building land based on the surrounding supporting information and environmental information of the target building land, as well as the design demand information of the target user;
  • the rating module is used to rate the target construction land through the preset rating model based on the generated design information, and obtain the estimated rating corresponding to the target construction land;
  • a generation module configured to generate at least one candidate construction plan for the target construction site according to the preset generation rules according to the estimated rating, and conduct an intelligent review of the at least one candidate construction plan;
  • the display module is used to generate target scoring dimensions based on design demand information, score the reviewed candidate building plans based on the target scoring dimensions, and display the preset number of candidate building plans with the highest scores as recommended building plans.
  • embodiments of the present application provide a computer device, including a processor, a memory and a communication interface, which , the memory stores a computer program, the computer program is configured to be executed by the processor, the computer program includes instructions for some or all of the steps described in the first aspect of the embodiment of the present application.
  • embodiments of the present application provide a computer-readable storage medium that stores a computer program, and the computer program causes the computer to execute part or components as described in the first aspect of the embodiment of the present application. All steps.
  • Figure 1 is a schematic diagram of a system architecture provided by an embodiment of the present application.
  • Figure 2 is a schematic flowchart of a method for generating an architectural plan provided by an embodiment of the present application
  • Figure 3 is a schematic diagram of a scene generated by an architectural plan provided by the embodiment of the present application.
  • Figure 4 is a schematic diagram of a scene generated by another building plan provided by the embodiment of the present application.
  • Figure 5 is a schematic diagram of a scene generated by another building plan provided by the embodiment of the present application.
  • Figure 6 is a schematic structural diagram of a device for generating architectural plans provided by an embodiment of the present application.
  • Figure 7 is a schematic structural diagram of a computer device provided by an embodiment of the present application.
  • an embodiment means that a particular feature, structure or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application.
  • the appearances of this phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those skilled in the art understand, both explicitly and implicitly, that the embodiments described herein may be combined with other embodiments.
  • FIG. 1 is a schematic diagram of a system architecture provided by an embodiment of the present application.
  • the system architecture may include: an electronic device 101 and a server 102 .
  • the electronic device 101 and the server 102 can communicate through a network.
  • Network communication can be based on any wired and wireless network, including but not limited to the Internet, wide area network, metropolitan area network, local area network, virtual private network (virtual private network, VPN), wireless communication network, etc.
  • the method of generating a building plan provided by the embodiment of the present application can be executed by an electronic device.
  • the electronic device 101 executes the method of generating a building plan of the embodiment of the present application.
  • the electronic device 101 can send the generated recommended building plan to the server 102.
  • the server 102 can store recommended building plans, and the server 102 can send the recommended building plans to other devices.
  • the electronic device 101 can obtain the generated design information of the target building land based on the surrounding supporting information and environmental information of the target building land, as well as the design demand information of the target user; then, the target can be rated through a preset rating model based on the generated design information.
  • the construction land is rated to obtain the estimated rating corresponding to the target building land; then based on the estimated rating and in accordance with the preset generation rules, at least one candidate building plan for the target building land is generated, and at least one candidate building plan is intelligently reviewed;
  • the target scoring dimension is generated based on the design demand information, and the candidate building plans that have passed the review are scored according to the target scoring dimension, and the preset number of candidate building plans with the highest scores are displayed as recommended building plans.
  • the embodiments of this application do not limit the number of electronic devices and servers.
  • the server can provide services for multiple electronic devices at the same time.
  • the electronic device may be a personal computer (PC), a notebook computer or a smartphone, or it may be an all-in-one machine, a handheld computer, a tablet computer (pad), a smart TV playback terminal, a vehicle-mounted terminal or a convenient equipment, etc.
  • the operating systems of PC-side electronic devices, such as all-in-one computers can include but are not limited to Linux systems, Unix systems, Windows series systems (such as Windows xp, Windows 7, etc.), Mac OS X systems (operating systems of Apple computers), etc. system.
  • the operating systems of mobile electronic devices, such as smartphones may include but are not limited to Android, IOS (the operating system of Apple mobile phones), Window and other operating systems.
  • the server can be an independent server, or it can provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery network, CDN) and cloud servers for basic cloud computing services such as big data and artificial intelligence platforms.
  • the server may be implemented as a server cluster consisting of multiple servers.
  • embodiments of the present application provide a method for generating architectural plans, which can be applied to electronic devices or servers configured in the construction field.
  • architectural plans can be automatically generated based on relevant information, which is beneficial to improving the efficiency of architectural design.
  • FIG. 2 is a schematic flowchart of a method for generating a building plan provided by an embodiment of the present application. Taking the application of this method to the electronic device 101 in Figure 1 as an example, the method may include the following steps S201-S204, wherein:
  • Step S201 Obtain the generated design information of the target building land based on the surrounding supporting information and environmental information of the target building land and the design demand information of the target user.
  • the target construction land may refer to the construction land for which a construction plan is expected to be generated.
  • Construction land can refer to the land that the developer intends to use for building buildings and structures.
  • Construction land can be divided into residential land, commercial land, public facility land, industrial land, green space and special land, etc. That is to say, residences, shopping malls, office buildings, schools, hospitals, factories, etc. can be built on the construction land, which is not limited in the embodiments of the present application.
  • the geographical location information of the target building land can be represented by longitude and latitude.
  • the geographical location information of the target building land can be input by the target user and then obtained by the electronic device. It can also be obtained by the electronic device through GPS positioning. This is not the case in the embodiment of the present application.
  • Surrounding supporting information can include surrounding public facilities information, traffic information, people flow information, surrounding building type information, etc.
  • Public facility information can include the number and location of subway stations, bus stations, parks, libraries, schools and hospitals, etc.
  • Traffic information can include traffic flow, congestion level, number of traffic lights, etc.
  • People flow information can include people flow and so on.
  • the surrounding building type information may include whether the surrounding buildings are high-end residences, ordinary residences, villas, office buildings, shopping malls or factories, as well as the housing price information and floor height of the surrounding buildings, etc.
  • Environmental information can refer to the target building land and surrounding noise information, greening information, topography, soil, water quality, or whether it is near a river or the sea, etc.
  • the target users are usually designers, and more specifically may refer to users of the electronic device 101 .
  • the design demand information can be the target user's needs for the generated architectural plan, such as maximizing economic benefits, optimal floor area ratio, most reasonable layout, best quality, etc.
  • the design requirement information may also include specific building indicators, such as a floor area ratio requirement of 2.0, a green space ratio requirement of 30%, etc., or may also include some building code requirements, etc., which are not limited in the embodiments of this application.
  • the generated design information can be used as constraints for subsequent generation of candidate building proposals.
  • the generated design information can include specific building indicators, such as floor area ratio, green space rate, and building height limit, and can also include specific parameters of related building indicators, such as floor area ratio 2.0, green space rate 30%, and so on.
  • the floor area ratio refers to the ratio of the total construction area to the construction land area within a certain plot of land.
  • Floor area ratio can reflect development intensity and is an indicator often studied in architectural design.
  • the green space rate refers to the ratio of the green space area to the construction land area within a certain plot.
  • the green space rate can reflect the degree of greening and environmental comfort.
  • Building height limit refers to the limit on building height.
  • the generated design information can also include but is not limited to building density, building layout, building type, building orientation, building sun spacing, building location spacing, and values related to these indicators, etc.
  • step S201 the following steps may also be included:
  • the type of construction land includes blank construction land and construction land to be renovated; in response to the type of construction land being the construction land to be remodeled, obtaining the building information contained in the construction land,
  • the building information includes building type information, building type information and building age information; an area to be renovated in the building land is determined according to the building information, and the area to be rebuilt is used as the target building land; in response to the The type of construction land is the blank construction land, and the construction land is used as the target construction land.
  • the electronic device 101 before executing step S201, the electronic device 101 needs to select a piece of construction land as the target construction land.
  • the type of building land can be identified first, and the building Whether there are already built buildings on the land. If it is identified that there are already built buildings in the building land, then the type of the building land is the building land to be converted. If it is identified that no built building exists in the building land, the type of the building land is blank building land.
  • the specific recognition method can be through pre-training the recognition model and performing recognition through the recognition model.
  • Figure 3 is a schematic diagram of a scene generated by a building plan provided by an embodiment of the present application.
  • the electronic device 101 may respond to the preset application program that generates the construction plan.
  • Start the trigger operation to start the preset application program that generates the construction plan and the preset application program can display the corresponding construction land in the current display interface 30a, such as construction land A, construction land B, construction land C, construction land D, Building land E and building land F and so on.
  • Building land is divided by building lines.
  • the type of construction land may be determined first.
  • construction land A, construction land B, construction land C and construction land D are plots where buildings have not been built and can be used as blank construction land; construction land E and construction land F are plots where buildings have been built. The land can be used as building land to be renovated.
  • the target user can click on the construction land, for example, click on the blank construction land (for example, construction land A) to view the construction land information of the construction land.
  • the construction land information may include but is not limited to the construction land area, construction land area, and construction land area.
  • the geographical location information (for example, it can be represented by longitude and latitude), the topography of the construction land, the soil conditions of the construction land, and the surrounding supporting information and environmental information of the construction land, etc. If you click on the building land to be renovated (for example, building land E), you can also view the building information contained in the building land.
  • the building information includes building type information, building type information, building age information, etc.
  • the electronic device can pre-collect information on the buildings that have been built in the building land to be renovated, and the target user can obtain it by clicking on the corresponding building land to be renovated. After obtaining the building information of the building land to be renovated, the area to be renovated in the building land can be determined based on the building information. In a possible implementation, the electronic device can automatically divide the areas to be renovated from the building land to be rebuilt based on the pre-collected building information, and use them as target building land for subsequent generation of building plans, and use red lines or other lines to target users. Showcase the area to be remodeled.
  • the target user can also draw one or more closed polygons in the building land to be converted (for example, building land E), and the electronic device can respond to the drawing operation and draw the one or more closed polygons.
  • the area to be reconstructed is used as the area to be reconstructed, and the area to be reconstructed is used as the target building land.
  • the closed polygon can be a regular shape such as a circle or a quadrilateral, or it can be an irregular shape.
  • prompt information 30b as shown in Figure 3 can be generated (for example, it can be specifically expressed as "use building land A as the target building land"), and the prompt information 30b can include "yes" (the above target building There are two controls: land use confirmation control) and "no" (the above target building land denial control).
  • the electronic device 101 takes the construction land A as the target construction land, displays the target construction land 30c in the current display interface 30a, and provides a generated design information area 30d.
  • the target user can input the generated design information in the generated design information area 30d.
  • the target user inputs 2.0 in the floor area ratio input box in the generated design information area 30d.
  • the electronic device 101 obtains the generated design information of the target building site by reading the target user's data in the generated design information area 30d.
  • the generated design information of the target building land can also be obtained through other methods, which will be described in detail below and will not be described in detail here.
  • the electronic device 101 can close the prompt information 30b and not regard the construction land A as the target construction land.
  • prompt messages not involved in Figure 3 can also be generated, such as "Please re-select the building land as the target building land” to prompt the target user to re-select the target building land. Building land.
  • the current display interface 30a may also include controls not involved in Figure 3, such as a search box, which may be used to search for locations, construction land, etc.
  • a search box which may be used to search for locations, construction land, etc.
  • the target user can enter certain geographical location information in the search box, and then the electronic device 101 displays the geographical location on the current display interface 30a. The information corresponds to the building land. If the construction land currently displayed on the display interface 30a includes the target construction land for which a construction plan is to be generated, the target user does not need to perform a search operation in the search box.
  • the construction land displayed in the current display interface 30a can also be panned, enlarged or reduced to better observe the construction land.
  • the following steps may also be included: obtaining the urban culture of the city where the area to be renovated is located, and dividing the area to be renovated into L target areas according to the urban culture and the building information,
  • the target area includes at least one of an area to be demolished, a building reserved area, and a blank land area, and the L is a positive integer; in response to the target area being a building reserved area, obtain the area within the building reserved area.
  • the functions and layout of the included buildings generating at least one candidate building scheme for the reserved area of the building based on the layout and functions of the buildings.
  • urban culture may refer to a collection of related features such as landscape, art, customs, etc. of a city.
  • the buildings of a city can form part of a city's culture, which means that the culture of a city can also be reflected in its buildings.
  • the old buildings in the area to be renovated bear witness to the development of the city. As a mark of the times, they tell the glory of the past with their architectural form. But now most of them have lost their original functions and become divorced from the urban fabric. Therefore, old buildings in the area to be renovated can be renovated and given new life, thereby improving urban service functions and enhancing the sense of historical development of the city.
  • renovation of old buildings is inseparable from the culture of a city.
  • the urban culture of the city where the area to be renovated is located can be obtained by reading the data input by the target user, or by collecting relevant image data of building forms, natural landscapes and cultural heritage contained in the city, and extracting it using a preset model Its relevant characteristics are obtained.
  • the target areas can include areas to be demolished, building retention areas and blank land areas.
  • L is a positive integer, and the specific value can be determined by the target user. settings, or system default settings.
  • the area to be demolished refers to the area where the original old buildings need to be demolished and reconstructed in the area to be rebuilt;
  • the building retention area refers to the area to be rebuilt where the original old buildings are retained and rebuilt on the basis of the original old buildings.
  • blank land area refers to the area where no buildings are built in the area to be renovated.
  • steps S201 to S204 For the specific implementation of generating a construction plan in the area to be demolished and the blank land area, reference may be made to the relevant descriptions of steps S201 to S204, which will not be described again here.
  • the building preservation area it is necessary to consider the function and layout of the original building, as well as the original building's construction materials, shape, size, color and other related factors, and based on these related factors, combined with the urban culture and surrounding supporting information, environment Information and other factors are used to generate relevant candidate building plans to give new life to old buildings.
  • the target area is a building reserved area
  • obtain the functions and layout of the buildings included in the building reserved area and calculate the The layout and functions of the building generate at least one candidate building scheme for the reserved area of the building to realize the reconstruction of the old building, and can improve the rationality of the reconstruction.
  • step S201 may specifically include the following steps:
  • customer positioning refers to the type of customers that will be targeted after the buildings are built on the target construction land, including high-income customers, middle-income customers, low-income customers, first-time home buyers, second-time home buyers, etc.
  • Product positioning mainly refers to product quality positioning, including high-end products and mid-range products.
  • High-end products can be villas and bungalows, etc.
  • mid-range products can be high-rise buildings, etc.
  • the customer positioning can be targeted at high-income customers, and the product positioning can be high-end products such as villas or bungalows.
  • the electronic device can automatically generate relevant generative design information based on past cases of villas or bungalows. Specifically, this can be achieved through conditional generation models.
  • the electronic device obtains a large number of historical building plans and extracts relevant historical data in the historical building plans as sample training data for the conditional generation model.
  • Historical data can include building types (such as villas, bungalows, high-rises, low- and mid-rises, etc.) in historical building plans, as well as relevant generated design information about historical building plans, such as floor area ratio, green space ratio, building height limit, building density, building Layout, building type, building orientation, building sunlight spacing, building location spacing, etc.
  • a large amount of historical data is input into the conditional generation model as sample training data for training to obtain a more accurate conditional generation model.
  • the input of the conditional generation model is the building type
  • the output is the relevant generated design information corresponding to the building type.
  • RNN recurrent neural network
  • CNN convolutional neural network
  • SVM support vector machine
  • the customer positioning and product positioning of the target building land are determined based on the surrounding supporting information, environmental information and design demand information. , thereby determining the building type, and then using the model to automatically generate design information for the target building land according to the building type. Users do not need to manually enter relevant data required to generate design information, which can improve design efficiency and shorten the overall solution design time.
  • step S201 may specifically include the following steps:
  • the surrounding supporting information and environmental information of the target building land as well as the design demand information of the target user, and output the multi-modal features corresponding to the surrounding supporting information, the environmental information and the design demand information through the multi-modal model.
  • the multi-modal features are obtained by fusing the first features corresponding to the surrounding supporting information, the second features corresponding to the environmental information, and the third features corresponding to the design requirement information; the multi-modal features
  • the characteristics are input into the preset land parcel division model, and the target building land is divided into N functional zones through the preset land parcel division model.
  • the functional zones include high-rise building areas, villa areas, public facilities areas and commercial areas.
  • the N is a positive integer; obtain the building land area of each functional zone, and determine the sub-district of each functional zone according to the building land area of the functional zone and the function of the functional zone. Design conditions; determining the generated design information of the target construction land according to the sub-design conditions.
  • the geographical location information of the target construction land can be obtained through an API interface provided by a third-party map application, and the geographical location information can be represented by longitude and latitude.
  • Obtaining target construction land After obtaining the geographical location information, the surrounding supporting information and environmental information within the preset distance range of the target building land can be obtained with the target building land as the center.
  • the preset distance is a preset parameter, and the specific value can be set based on historical experience. For example, it can be set to 2 kilometers, or it can also be set based on the actual situation. This is not limited in the embodiments of the present application.
  • the public facility information in the surrounding supporting information can be obtained through an API interface provided by a third-party map application; the heat map provided by the third-party map application can be used to analyze the surrounding areas of the target building land.
  • human traffic Supporting information such as traffic information and surrounding building type information can be obtained through third-party websites such as government websites and real estate websites.
  • the surrounding supporting information of the target construction land can also be obtained through other methods, which is not limited in the embodiments of the present application.
  • the noise value in the environmental information can be obtained by installing multiple noise detection devices at preset locations on the target construction site, and communication connections can be established between the electronic device and the multiple noise detection devices.
  • the noise detection equipment is used to detect noise, and may be a sound level meter, a noise detection sensor, etc., which is not limited in the embodiments of the present application. Multiple noise detection devices can be set at different locations, and when noise detection operations are performed at the same time, multi-point noise detection can be achieved.
  • Noise detection equipment can be installed near the land contour of the target building land. The distance between each noise detection equipment can be 300 meters, 400 meters, etc. The specific value can be determined according to the actual situation.
  • the electronic device receives noise detection data sent by multiple noise detection devices within a preset time interval.
  • the preset time interval may be 10 minutes, 5 minutes, etc., which is not limited in the embodiments of the present application.
  • the noise detection data is multi-point noise detection data, which at least includes the equipment identification of each noise detection device and the detection data corresponding to each equipment identification. After the electronic device reads the noise detection results, it outputs the detected noise value of each noise detection device respectively, thereby obtaining the noise value in the environmental information of the target building land. It can be seen that this method can control multiple noise detection devices at the same time for noise detection, has good synchronization, and is conducive to improving the reliability of noise detection. Greening information, terrain, soil, water quality, etc. in environmental information can be obtained through government websites, Ecological Environment Bureau websites, or other third-party websites. In addition, the design requirement information of the target users can be obtained through the data input by the target users.
  • the multimodal model may be a model used to obtain multimodal features.
  • the electronic device can input peripheral supporting information, environmental information, and design requirement information into the multimodal model, and the multimodal model can output multimodal features corresponding to these input information.
  • the multi-modal feature is obtained by fusing the first feature corresponding to the surrounding supporting information, the second feature corresponding to the environmental information, and the third feature corresponding to the design requirement information.
  • the first feature can be features related to surrounding supporting information, such as the number and location of surrounding public facilities, etc.
  • the second feature can be features related to environmental information, such as noise value, green space rate, etc.
  • the third feature can be target users. Characteristics related to the design requirement information, such as the values of specific building indicators, etc.
  • the multi-modal features can also add other features related to building generation, which are not limited in the embodiments of the present application.
  • the multi-modal model may be a model such as a recurrent neural network (RNN), a convolutional neural network (CNN), a support vector machine (SVM), etc.
  • RNN recurrent neural network
  • CNN convolutional neural network
  • SVM support vector machine
  • the functional zoning can include at least one of high-rise building area, villa area, public facilities area and commercial area.
  • the functional zoning can also include a western-style building area, a low-rise area and a mixed area (a mixture of high-rise and low-rise buildings, a mixture of villas and western-style buildings, etc. wait wait wait.
  • N is a positive integer greater than or equal to 1. In practical applications, N can be set by the target user or preset by the system. This is not limited in the embodiments of the present application.
  • the sub-design conditions can be understood as the generated design information for each functional zone. Similar to the generated design information, the sub-design conditions can include floor area ratio, green space ratio, building height limit, building density, building layout, building type, building orientation, Building sun spacing, building location spacing, etc.
  • Figure 4 is a schematic diagram of a scene generated by another building plan provided by an embodiment of the present application.
  • the electronic device can input the multi-modal features into the preset plot division model, and use the preset plot division model to
  • the target building land is divided into N functional partitions.
  • a preset tool such as a scatter point contour algorithm (Alpha Shapes)
  • Alpha Shapes can be used to extract the land contour line of the target construction land.
  • the land outline of the target building land is usually polygonal, and multiple outline edges can be determined based on the land outline.
  • the surrounding supporting information and environmental information of each outline edge may be different.
  • the land parcel model is then used to divide the target building land into multiple functional zones. It can be understood that the area and building type of each functional zone can be the same or different.
  • the specific division rules are based on the type and number of surrounding supporting facilities and the environment corresponding to the contour lines of the target building land. information, and the design information is determined based on the needs of target users.
  • the land parcel division model is actually applied, a large amount of data can be collected for training, so that the land parcel training model can summarize the basic rules of land parcel division, so that the division of functional zoning is more accurate and reasonable, thereby improving the construction quality.
  • the rationality of business distribution is then used to divide the target building land into multiple functional zones.
  • the current display interface 40a includes a visualization area 40b and a generated design information area 40c.
  • the visualization area 40b is used to display the target building land.
  • the land parcel model can be used to divide the target building land into 4 functional zones, namely functional zone 1 (40d), Functional Partition 2 (40e), Functional Partition 3 (40f) and Functional Partition 4 (40g).
  • functional partition 1 and functional partition 4 correspond to the contour lines.
  • the surrounding supporting resources are complete, the transportation is developed, the environment is comfortable, and the noise pollution is lower.
  • functional partition 1 and functional partition 4 can be used as villa areas, and functional partition 3 can be used as High-rise building area, and use functional zone 2 as a public facilities area (can be used to build kindergartens, community offices or green areas, etc.).
  • other division methods not shown in Figure 4 may also be included.
  • embodiments of the present application may provide multiple different division methods for users to choose.
  • the target building land can also be divided according to the terrain of the target building land. For example, the terrain above the target building land is obviously higher than the terrain below.
  • the target building land can also be divided into upper and lower functional zones. Functional zones are used as low-rise building areas, functional zones below are used as high-rise building areas, and so on.
  • the target user can also draw one or more closed polygons in the visualization area 40b.
  • the electronic device can display one or more closed polygons drawn by the drawing operation in the visualization area 40b. Closed polygons, and divide the target construction land into multiple functional zones based on the one or more closed polygons.
  • the closed polygon can be a regular shape such as a circle or a quadrilateral, or it can be an irregular shape.
  • the building land area of each building land can be obtained, and then the above condition generation model is used to determine the sub-design conditions of each functional partition according to the function of the functional partition. For example, after determining that functional zone 1 will be used as a villa area, you can use the condition generation model to automatically generate sub-design conditions related to functional zone 1, such as automatically generating the floor area ratio, green space ratio, building height limit, building density, Related sub-design conditions such as building layout, building type, building orientation, building sunlight spacing, building location spacing, etc.
  • the generated design information of the target building land and the sub-design conditions of each functional partition can be displayed in the generated design information area 40c. In order to further improve the user experience, target users can also manually modify each sub-design condition.
  • the multi-modal features can be input into the preset parcel division model, and the preset parcel division
  • the model divides the target building land into multiple functional zones, then determines the sub-design conditions of each functional zone based on the functions of the functional zones, and finally determines the generated design information of the target building land based on the sub-design conditions.
  • the setting of each design condition is more reasonable, and it also helps to increase the diversity of architectural plans, thus helping to improve the rationality of the generation of architectural plans.
  • Step S202 Rating the target construction land through a preset rating model according to the generated design information, and obtain an estimated rating corresponding to the target construction land.
  • the estimated rating may refer to a preliminary assessment of the target building land. For example, based on the generated design information, it can be preliminarily determined that the target building land can be used as high-end, mid-range or low-end residential buildings, etc. In the embodiment of this application, you can pass Establish a basic model.
  • the basic model can be a convolutional neural network model, etc., obtain a large amount of historical generative design information of historical building land, and extract the historical generative design information of each historical building land and the corresponding historical building land from the historical generative design information.
  • the estimated rating data for example, the historical generated design information can be a floor area ratio of 0.5, a green space ratio of 40%, a building density of 15%, and the corresponding estimated rating of the historical building land can be a high-end community, etc. Then a large amount of historical generated design information and the corresponding estimated rating data of historical building land are used as sample rating data. And input the sample rating data into the basic model for training, and obtain the preset rating model based on the training results. Finally, the generated design information is input into the preset rating model to obtain the estimated rating of the target building land, which can further make a more refined evaluation of the target building land and help improve the accuracy of candidate building plans.
  • Step S203 Generate at least one candidate construction plan for the target construction site based on the estimated rating and preset generation rules, and conduct an intelligent review of the at least one candidate construction plan.
  • architectural generative design came into being. Compared with traditional computer-aided design, architectural generative design puts the role of computers in the design stage, so that the design process can use the powerful computing power of computers to generate a variety of possible architectural design plans. Therefore, architectural generative design can not only improve the efficiency of architectural design, but also liberate architects' thinking, break the rigidity of template design, and stimulate more design possibilities.
  • the preset generation rules may be some generation rules pre-built into the electronic device, such as some preset arrangements, preset building orientations, and preset constraint information (for example, location constraints, sunlight constraints , fire control, etc.).
  • Candidate building plans can include the plan that maximizes economic benefits, the plan with reasonable layout, the best plan, the best plan with floor area ratio, the best green space rate plan, the best quality plan, etc.
  • the economic benefit maximization plan can refer to the plan with the highest expected profit
  • the reasonable layout plan can refer to meeting the layout requirements of "high in the west and low in the east, high in the north and low in the south", “same type of group layout” and "fullness of layout”.
  • the optimal solution can refer to the solution in which the floor area ratio, green space ratio, building height limit, etc. are all close to generating design information; the best solution in floor area ratio can refer to the solution in which the floor area ratio is closest to generating design information; The best solution with green space ratio refers to the solution whose green space ratio is closest to the generated design information; the best quality solution refers to the solution with the floor area ratio close to the generated design information and the farthest distance between buildings.
  • the candidate building solutions may also include other possible solutions, and the embodiments of this application will not list them one by one.
  • the generated at least one candidate building scheme can be intelligently reviewed. Specifically, the generated at least one candidate building plan can be reviewed for sunlight, building spacing, fire protection, and other parameters according to relevant sunlight specifications, building spacing specifications, fire protection specifications, or other urban planning-related specifications. If the candidate building plan passes the review, step S204 is executed; if it does not pass the review, the plan is discarded or deleted. In this way, the rationality of the recommended building plan can be improved.
  • generating at least one candidate construction plan for the target construction site based on the estimated rating and preset generation rules in step S203 may specifically include the following steps:
  • the target building type library stores the reference building type corresponding to the estimated rating and the parameter information of the reference building type, as well as the The score value of the reference building type; the preset number of reference building types with the highest scores in the target building type library are displayed to the target user as candidate building types, and the candidate building type selected by the target user is used as the target Building type; obtain the land outline of the target building land, and generate at least one candidate building plan for the target building land according to the target building type, the preset arrangement and the land outline, so
  • the above-mentioned preset arrangements include determinant, peripheral, point group and extreme high-low arrangement.
  • the electronic device can pre-collect building types of multiple buildings, building type parameters, and corresponding score values of each building type to build a preset building type library.
  • Building types can include three bedrooms, two living rooms and two bathrooms, three bedrooms, two living rooms and one bathroom, two bedrooms, two living rooms and one bathroom, and three bedrooms, two living rooms and two bathrooms, as well as the number of balconies, terraces, bay windows, etc.
  • the parameters of the building type can include the number of households and the area of each household, the number of building floors, the height of each floor, the orientation of the building, the height of the overhead floors, etc.; they can also include the number of balconies, the number of terraces, and the number of bay windows. Number and direction of the living room, etc.
  • the apartment types with higher ratings of each developer's properties and related apartment configuration information can be included in the preset building type library in advance.
  • labels can be added to the building types of each building.
  • This tag is used to record the estimated rating corresponding to each building, and then classify the preset building type library according to this tag, for example, into a high-end residential building type library, a villa building type library or a high-rise building type library, etc., so that Find.
  • the definition of estimated rating can refer to the previous description and will not be repeated here.
  • the estimated rating of the target building land can be used as query information, and the corresponding building type library classification can be found from the preset building type library as the target building type library.
  • the target building type database can store the reference building type corresponding to the estimated rating and the parameter information of the reference building type, as well as the rating value of the reference building type. Then, a preset number of reference building types with top scores in the target building type library are displayed to the target users as candidate building types. The definition of the preset quantity can be referred to the description below and will not be repeated here.
  • the target user is prompted to select the corresponding candidate building type, and the candidate building type selected by the target user is used as the target building type.
  • the target user can select one candidate building type or multiple candidate building types, which is not limited in the embodiment of the present application.
  • the target user can modify the parameters of the target building type, such as modifying the height of the floor, the orientation of the building, the height of the elevated floor, the number of floors, etc., in order to obtain the parameters of the target building type that better meet the requirements of the target user.
  • the land outline of the target building land After obtaining the land outline of the target building land, at least one candidate building plan for the target building land can be generated based on the target building type, the preset arrangement and the land outline.
  • the preset arrangement methods include determinant, peripheral, point group and extreme high-low configuration.
  • the extreme high-low configuration can refer to the arrangement that accommodates the most buildings while meeting other indicators (such as building density, floor area ratio, sunshine, etc.).
  • the target building type library is determined based on the estimated rating, and then the candidate building type selected by the user is used as the target building type. Finally, at least one building type for the target building land is generated based on the target building type, preset arrangement and land outline. candidate building plans, which helps to improve the efficiency and accuracy of generating candidate building plans.
  • the step is to display a preset number of reference building types with top scores in the target building type library as candidate building types to the target user, and use the candidate building type selected by the target user as the target Building type may specifically include the following steps:
  • a preset number of reference building types with top scores in the target building type database are displayed to the target user as candidate building types; in response to detecting that the target user selects the joining operation, the target user is selected according to the joining operation.
  • the selected at least two candidate building types are combined to obtain at least one building type combination result, and the building type combination result is used as the target building type.
  • the combining operation refers to merging two or more candidate building types.
  • the specific assembly quantity can be determined based on the geographical location, supporting information and environmental information, and then determined based on the product positioning.
  • the candidate building types used for splitting can be exactly the same building type, or they can be different building types. It can also generate multiple building type combination results based on parameters such as different orientations and different number of combinations, and push the multiple combination results to the target user for selection, and use the building type combination result selected by the target user as the target building. type.
  • the generated combination results of various building types can also be used as target building types, and the generated building combination results can be sorted from best to worst according to product positioning or other requirements, and the plans can be arranged in sequence according to the sorting. Arrange relevant indicators that meet user needs (such as floor area ratio, coverage ratio) to generate corresponding candidate building plans.
  • the step is to obtain the land outline of the target building land, and generate at least one type of land contour for the target building land according to the target building type, the preset arrangement and the land outline.
  • Candidate building plans may include the following steps:
  • the land use information of the target building land includes the floor area and land outline of the target building land; obtain the constraint information of each building and adjacent buildings in the target building land; according to The land use information and the constraint information generate a building layout of the target building land.
  • the constraint information may include a target distance constraint value, and the target distance constraint value may be used to represent the minimum distance between each building and adjacent buildings in the target construction land.
  • the constraint information may also include a target sunshine spacing constraint value or a fire protection spacing constraint value, etc.
  • the target sunshine spacing constraint value may be used to represent the minimum sunshine spacing between each building and adjacent buildings in the target building land; fire protection spacing constraint Values may be used to represent the minimum fire separation distance between each building and adjacent buildings on the target building site. Then the building layout of the target building land is generated based on the land use information and constraint information.
  • the target building land can also be divided into different functional zones.
  • the building layout is determined according to the functional zoning, which means that the building layout of each functional zoning can be the same or different, and the specifics can be determined according to the actual situation.
  • the number of buildings in the functional zone, the building type, the number of building floors, etc. can be obtained based on the area and land contours of the functional zone, combined with other sub-design conditions, and based on each Constraint information between buildings and adjacent buildings.
  • the specific arrangement method may include one or more of determinant, peripheral, point group, and mixed methods.
  • climate and lighting conditions, as well as the specific building types within each functional zone can be taken into consideration. For example, if the lighting conditions in the city where the target building land is located are insufficient, the determinant should be used; if the city where the target building land is located is relatively cold, the peripheral type should be used.
  • the functional zoning is a single type of functional zoning such as villa area, high-rise building area or western-style building area, determinant or peripheral type should be adopted; if the functional zoning is a mixed area (mixed high-rise and low-rise buildings, mixed villas and western-style buildings, etc.), it should be adopted Point group or mixed.
  • other building layout methods are also possible, which are not limited in the embodiments of this application.
  • the building layout of the target building land is generated based on the land use information and constraint information of the target building land, and multiple building layout drawings can be generated at one time. Compared with the manual layout method, the efficiency and accuracy of the building layout can be effectively improved, and Can save a lot of time and effort.
  • the step of obtaining the constraint information of each building and adjacent buildings in the target construction land may specifically include the following steps: obtaining the user input of each building in the target construction land.
  • the first distance constraint value between the object and the adjacent building if the first distance constraint value is greater than or equal to the preset distance constraint value, then the first distance constraint value is used as the target distance constraint value; or if If the first distance constraint value is less than the preset distance constraint value, the user is prompted to adjust the first distance constraint value until the first distance constraint value is greater than or equal to the preset distance constraint value.
  • the preset distance constraint value may be used to represent the minimum prescribed distance between each building and adjacent buildings in the target construction land.
  • the specific value of the preset distance constraint value can be determined based on the architectural design specification data of the city where the target building land is located or other architectural design-related specifications. Therefore, in order to prevent the generation of If the candidate building plan does not meet the actual needs, the first distance constraint value input by the user needs to be greater than or equal to the preset distance constraint value.
  • the user needs to be prompted to adjust the first distance constraint value until the first distance constraint value is greater than or equal to the preset distance constraint value, and will be greater than Or the first distance constraint value equal to the preset distance constraint value is used as the target distance constraint value.
  • the target construction land includes the target construction area
  • at least one candidate construction plan for the target construction land is generated according to the estimated rating and preset generation rules. , which may include the following steps:
  • the setback value refers to the distance between the construction land contour line (or land red line) of the target construction land and the building construction area.
  • the target building area is an area where buildings are constructed, and the area of the target building area is smaller than the area of the target building land.
  • the setback value corresponding to the land contour line on each side can be obtained from the numerical value input by the target user, or can be automatically generated by electronic equipment according to relevant specification requirements.
  • the target construction area is determined, and at least one candidate construction plan for the target construction area is generated according to the estimated rating and preset generation rules.
  • the rationality of the generation of candidate building plans can be improved, and it can also facilitate subsequent expansion or other processing of the target building land.
  • At least one candidate construction plan for the target construction site is generated according to the preset generation rules according to the estimated rating; if the generation fails, each of the generated design information is generated according to the weight Sort from large to small; use the preset number of generated design information ranked first as the target to generate design information; generate at least one candidate for the target building land according to the target generated design information according to the preset generation rules construction plan.
  • the generated design information plays a vital role in generating the overall solution, while some of the generated design information is insignificant. When too much design information is generated, it may lead to failure in generating candidate building plans. If the generation fails, you can obtain the weight of each generated design information, and sort the generated design information according to the weight from large to small. Use the preset number of generated design information in front as the target to generate design information, and use the generated design information with the lowest weight to generate the design information. Design information is discarded, thereby reducing the constraints for generating candidate architectural plans to improve the success rate of plan generation. The determination of the weight can refer to the description below and will not be repeated here.
  • the design information is generated according to the target and according to the preset generation rules to generate at least one candidate construction plan for the target construction land.
  • the design information is generated by taking the top preset number of generated design information as the target, and then generating at least one candidate building plan for the target building site based on the target generated design information in accordance with the preset generation rules. , which can improve the success rate of plan generation, thereby improving the efficiency of architectural design.
  • the candidate building plans displayed first may not accurately match the user's design preferences, so that the user needs to spend more time screening those that meet expectations.
  • the embodiment of the present application can reduce the user's search time by executing step S204, thereby improving the design efficiency.
  • Step S204 Generate target scoring dimensions according to the design requirement information, and pass the The reviewed candidate building plans are scored, and the preset number of candidate building plans with the highest scores are displayed as recommended building plans.
  • weight parameter options can be set for the design requirement information of the target user, and the weight parameter options can be displayed for the target user to confirm or modify.
  • Weight parameter options can be identified by numerical values, for example: 1.0 represents a very important requirement, 0.8 represents an important requirement, 0.6 represents a relatively important requirement, 0.4 represents a generally important requirement and 0.2 represents an ordinary requirement.
  • option settings or other option settings which are not limited in the embodiments of this application. It is understandable that if the target users have a large amount of design demand information, some marginal demands may not be of high reference value for screening candidate building plans that have passed the review, and may not be considered. The design demand information with high weight will be regarded as the demands that the target users are more concerned about.
  • the preset quantity can be preset by the system or determined by the target users.
  • Figure 5 is a schematic diagram of a scene generated by yet another building plan provided by an embodiment of the present application.
  • a preset number of recommended building plans with top scores may be displayed in thumbnail form in the recommended building plan list 50d in the current display interface 50a.
  • the target user selects a recommended building plan in the recommended building plan list 50d
  • the enlarged recommended building plan can be displayed in the visualization area 50b, and the parameters corresponding to the recommended building plan can be displayed.
  • the parameters can include but are not limited to buildings. Land area, plot ratio, green space ratio, building type, building type, building orientation, building arrangement, architectural style, etc. As shown in FIG.
  • the enlarged plan 1 can be displayed in the visualization area 50 b.
  • the demand design information may be specific generated design information or related building indicators, such as floor area ratio, green space ratio, building density, building height limit, etc.
  • the generated design information area 50c can display the specific value of each generated design information and the value of the corresponding weight of each generated design information.
  • the weight of each generated design information can be set by the target user or pre-set by the system. Then the target scoring dimensions are determined according to the weight in each generated design information to score the candidate building plans.
  • the level of the score can be used to characterize the correlation between the weight of the generated design information and the recommended architectural plan. The score is positively correlated with the weight of the generated design information.
  • the floor area ratio of the generated design information in the generated design information area 50c is 2.0, the green space rate is 30%, the weight corresponding to the floor area ratio is 0.98, and the weight corresponding to the green space rate is 0.75.
  • the weight value can be 0-1.0.
  • the plot ratio of Scheme 1 is 2.0 and the green area rate is 25%; the plot ratio of Scheme 2 is 1.7 and the green area rate is 30%.
  • the target user can adjust the weight of each generated design information in the generated design information area 50c at any time. After the weight is adjusted, the display order of the corresponding plans in the recommended building plan list 50d will also be adjusted accordingly. , in order to give users priority to display building plans that meet their preferences, so as to reduce the user's search time.
  • the recommended building plan to be displayed can also be verified, such as verifying sunlight, building spacing, or building height limits. Specific verification requirements can be determined by referring to the building design code data of the city where the target building site is located or other building design-related codes. If the recommended building plan to be displayed does not pass the verification, the recommended building plan will be deleted from the list to be displayed.
  • step S203 the following steps may also be included:
  • sample image data of the city where the target construction land is located includes image data associated with buildings, natural landscapes and cultural landscapes of the city where the target construction land is located;
  • Sample labels are clustered to obtain K label clusters, where K is a positive integer; determined based on the K label clusters
  • the architectural style of the city where the target building land is located calculate the similarity between the architectural style of the candidate building plan and the architectural style of the city where the target building land is located; use the preset number of candidate building plans with the highest similarity as recommended buildings Show the plan.
  • Architectural style refers to the characteristics reflected in the content and appearance of architectural design, mainly in the original and perfect artistic conception displayed in the plane layout, morphological composition, artistic processing and use of techniques of the building.
  • Common architectural styles include, but are not limited to, Gothic architectural style, Baroque architectural style, Rococo architectural style, wooden batten architectural style, garden style, conceptual style and modernism style.
  • architecture is the art of environment.
  • Architectural design is not just a simple patchwork of geometric shapes, but also considers various factors such as the context of the place to achieve a fit between form and function, so that the building takes root in the environment and becomes a part of the environment.
  • the term artistic conception is often used in artistic creation to reflect the creator's expression of the environment.
  • the clustering algorithm in machine learning can be used to extract abstract formal features such as culture, history, and nature of a city to obtain the architectural style of the city.
  • the electronic device can obtain a large number of sample image data related to the buildings, natural landscapes, and cultural landscapes of the city where the target building land is located.
  • the embodiment of this application does not limit the building type in the city where it is located.
  • the building type can be residential buildings, villa buildings, office buildings, commercial buildings, or some public buildings such as schools, museums, libraries, courts, government office buildings, etc.
  • Natural landscapes can include mountains, rivers, lakes, trees, etc.
  • cultural landscapes can include gardens, pavilions, pavilions, etc.
  • Each sample video data or sample photo data can have a known label, which can be used to represent the architectural style reflected in the sample image data. It can be understood that the label corresponding to each sample image data is determined according to the content of the sample image data, and each sample image data may have one or more labels. Similarly, the architectural style of a city can be single or diverse.
  • the electronic device can add the tags corresponding to all sample image data to the tag set, and obtain K tag clusters by clustering all the tags in the tag set, where K is a positive integer, for example, K can take the value 1, 2,...; After clustering the labels in the label set, among the above K label clusters, similar labels can be divided into the same label cluster, and different labels can be divided into different label clusters.
  • the number K of label clusters obtained after the clustering process can be a preset fixed value, or it can be determined after continuously merging labels in the label set during the clustering process.
  • label clusters The quantity is not limited.
  • the clustering algorithms used by electronic devices to cluster tags in the tag set may include but are not limited to: k-means clustering algorithm (k-means), mean shift clustering algorithm (mean shift), k-means clustering Algorithm (k-means), fuzzy c-means clustering algorithm (fuzzy c-means, FCM), density-based spatial clustering of applications with noise (DBSCAN), using Gaussian mixture model ( Gaussian mixture model (GMM) expectation-maximization (EM) clustering algorithm, hierarchical clustering algorithm, etc.
  • GBM Gaussian mixture model
  • EM expectation-maximization
  • the similarity between the architectural style of the candidate building scheme and the architectural style of the city where the target building site is located is calculated.
  • similarity can be understood as candidate
  • the architectural style of the candidate building plan and the architectural style of the city where the target building site is located can be understood as semantic tags described by natural speech. Therefore, before calculating the similarity, the electronic device can compare the architectural style of the candidate building plan and the architecture of the city where the target building site is located.
  • the style is vector converted to obtain the vector corresponding to the architectural style of the candidate building plan and the architectural style of the city where the target building site is located.
  • the vector corresponding to the architectural style of the city where the target building land is located can be called the center vector.
  • the electronic device can perform vector conversion through a word vector model.
  • the word vector model can include but is not limited to: a word vector model (word2vec) that obtains a vectorized representation of a word through the context of the word, a word vector that uses a co-occurrence matrix (glove), bidirectional encoder representations from transformers (BERT), latent semantic analysis (LSA), latent dirichlet allocation (LDA), embodiments of this application There are no restrictions on the word vector model used.
  • the distance between the vector of the architectural style of the candidate building scheme and the center vector can be calculated, measured by the distance between the two vectors.
  • the similarity between the two is then determined based on the similarity. Among them, the shorter the distance, the greater the similarity, and the greater the distance, the smaller the similarity.
  • Methods used to calculate the distance between two vectors may include but are not limited to: Euclidean distance, Manhattan distance, Minkowski distance, cosine similarity ( cosine similarity).
  • cosin is the cosine function.
  • the similarity is greater than the similarity between vector G2 and the center vector t, then the architectural style of the candidate building scheme corresponding to vector G1 is more similar to the architectural style of the city where the target building land is located than the architectural style of the candidate building scheme corresponding to vector G2. Similarity in architectural style of the city where the target building site is located. Finally, the similarity is sorted, and from high to low, a preset number of candidate building plans with the highest similarity are returned as recommended building plans for priority display to target users. In actual applications, the preset number may be set by the target user or may be preset by the system, which is not limited in the embodiments of the present application.
  • the brand culture of the developer of the target construction site can also be obtained, and by comparing the matching degree between the developer's brand culture and the candidate building plans, the candidate building plans with a high degree of matching are prioritized as the recommended building plans.
  • the candidate building plans with a high degree of matching are prioritized as the recommended building plans.
  • plans with high green space rates and low floor area ratios can be prioritized for display to target users as recommended building plans.
  • the specific implementation method of calculating the matching degree can refer to the calculation of the architectural style similarity mentioned above, which will not be described again here.
  • the architectural style of the city where the target building land is located is obtained, and then the similarity between the architectural style of the candidate building plan and the architectural style of the city where the target building land is located is calculated, and the prediction with the highest similarity is
  • a set number of candidate building plans are prioritized to be displayed to target users as recommended building plans. In this way, users are given priority to display architectural plans that match the architectural style of the city where they are located, making the architectural plans displayed to users more reasonable and more in line with the actual needs of users, thereby reducing the user's search time and improving design. efficiency purpose.
  • the generated design information is passed through the preset rating
  • the model rates the target construction land, obtains an estimated rating corresponding to the target construction land, and then generates at least one candidate construction plan for the target construction land based on the estimated rating and in accordance with preset generation rules.
  • FIG. 6 is a schematic structural diagram of a device for generating architectural plans provided by an embodiment of the present application.
  • This device can be applied to electronic equipment, such as the electronic equipment 101 in the embodiment corresponding to Figure 1, and can also be applied to servers, such as the server 102 in the embodiment corresponding to Figure 1, which will not be limited here.
  • the embodiment of the present application takes the device being applied to the electronic device 101 as an example for description.
  • the device 600 for generating architectural plans may include an acquisition module 601, a rating module 602, a generation module 603 and a display module 604. The detailed description of each module is as follows:
  • the acquisition module 601 is used to obtain the generated design information of the target building land based on the surrounding supporting information and environmental information of the target building land, as well as the design demand information of the target user;
  • the rating module 602 is used to rate the target construction land through the preset rating model according to the generated design information, and obtain the estimated rating corresponding to the target construction land;
  • the generation module 603 is configured to generate at least one candidate construction plan for the target construction land according to the estimated rating and preset generation rules, and conduct an intelligent review of the at least one candidate construction plan;
  • the display module 604 is used to generate target scoring dimensions according to the design requirement information, score the candidate building plans that have passed the review according to the target scoring dimensions, and display a preset number of candidate building plans with the highest scores as recommended building plans.
  • the acquisition module 601 may include an output unit, a dividing unit, a first determination unit and a second determination unit, where:
  • the output unit is used to obtain the surrounding supporting information and environmental information of the target building land, as well as the design demand information of the target user, and output the multi-modal features corresponding to the surrounding supporting information, environmental information and design demand information through the multi-modal model; multi-modal
  • the dynamic feature is obtained by fusing the first feature corresponding to the surrounding supporting information, the second feature corresponding to the environmental information, and the third feature corresponding to the design requirement information;
  • the division unit is used to input multi-modal features into the preset land parcel division model, and divide the target building land into N functional zones through the preset land parcel division model;
  • the functional zones include high-rise building areas, villa areas, and public facilities areas. and at least one of the commercial areas, N is a positive integer;
  • the first determination unit is used to obtain the building land area of each functional zone, and determine the sub-design conditions of each functional zone based on the building land area of the functional zone and the functions of the functional zone;
  • the second determination unit is used to determine the generated design information of the target construction land according to the sub-design conditions.
  • the generation module 603 may include a search unit, a third determination unit and a generation unit, where:
  • the search unit is used to search the target building type library corresponding to the estimated rating from the preset building type library; the target building type library stores the reference building type corresponding to the estimated rating and the parameter information of the reference building type, as well as the reference building type. Rating value;
  • the third determination unit is used to display a preset number of reference building types with top scores in the target building type library as candidate building types to the target users, and use the candidate building types selected by the target users as the target building type;
  • the generation unit is used to obtain the land contour line of the target building land, and generate at least one candidate building plan for the target building land according to the target building type, preset arrangement method and land outline line; the preset arrangement method includes determinant, Peripheral type, point group type and extreme high and low matching type.
  • the third determining unit may include a determining subunit and a splicing subunit, where:
  • the splicing subunit is used to, in response to detecting that the target user selects a splicing operation, splice at least two candidate building types selected by the target user according to the splicing operation, obtain at least one building type splicing result, and use the building type splicing result as the target Building type.
  • the device 600 may also include modules not shown in Figure 6, such as a first acquisition module, a second acquisition module, a first determination module and a second determination module, wherein:
  • the first acquisition module is used to obtain the type of construction land; the types of construction land include blank construction land and construction land to be renovated;
  • the second acquisition module is used to obtain the building information contained in the building land in response to the type of building land being a building land to be renovated; the building information includes building type information, building type information and building age information;
  • the first determination module is used to determine the area to be reconstructed in the construction land based on the building information, and use the area to be reconstructed as the target construction land;
  • the second determination module is configured to use the construction land as the target construction land in response to the type of the construction land being blank construction land.
  • the device 600 may also include modules not shown in Figure 6, such as a third acquisition module, a fourth acquisition module and a first generation module, wherein:
  • the third acquisition module is used to obtain the urban culture of the city where the area to be renovated is located, and divide the area to be renovated into L target areas based on urban culture and building information; the target areas include areas to be demolished, building retention areas, and blank land areas. At least one of, L is a positive integer;
  • a fourth acquisition module configured to obtain the functions and layout of the buildings contained in the building reservation area in response to the target area being the building reservation area;
  • a first generation module is configured to generate at least one candidate building scheme for the reserved area of the building according to the layout and functions of the building.
  • the device 600 may also include modules not shown in Figure 6, such as a fifth acquisition module, a clustering module, a third determination module, a calculation module, and a first display module, where:
  • the fifth acquisition module is used to obtain sample image data of the city where the target building land is located;
  • the sample image data includes image data related to buildings, natural landscapes and cultural landscapes in the city where the target building land is located;
  • the clustering module is used to cluster the sample labels contained in the sample image data to obtain K label clusters, where K is a positive integer;
  • the third determination module is used to determine the architectural style of the city where the target building land is located based on K label clusters;
  • the calculation module is used to calculate the similarity between the architectural style of the candidate building scheme and the architectural style of the city where the target building site is located;
  • the first display module is used to display a preset number of candidate building plans with the highest similarity as recommended building plans.
  • FIG. 7 is a schematic structural diagram of a computer device provided by an embodiment of the present application.
  • the computer device 700 can be an electronic device, for example, the electronic device 101 in the embodiment corresponding to Figure 1, or a server, for example, the server 102 in the embodiment corresponding to Figure 1. This will not be the case here. It's restricted.
  • the embodiment of this application takes the computer device as an electronic device 101 as an example.
  • the computer device 700 may include a processor 701, a memory 702, and a communication interface 703, where the memory 702 stores a computer program 704.
  • the processor 701, the memory 702, the communication interface 703 and the computer program 704 may be connected through a bus 705.
  • the above computer program 704 is used to execute instructions for the following steps:
  • the computer program 704 is specifically used to obtain the generated design information of the target building land based on the surrounding supporting information and environmental information of the target building land, as well as the target user's design demand information for the target building land. Instructions to perform the following steps:
  • the multi-modal features are for the surrounding It is obtained by merging the first feature corresponding to the supporting information, the second feature corresponding to the environmental information, and the third feature corresponding to the design requirement information;
  • the functional zones include high-rise building areas, villa areas, public facilities areas and commercial areas. At least one, N is a positive integer;
  • the computer program 704 is specifically configured to perform the following steps:
  • the target building type library stores the reference building type corresponding to the estimated rating and parameter information of the reference building type, as well as the score value of the reference building type;
  • the preset arrangement method includes determinant, perimeter, and point group. style and extreme high and low configurations.
  • Computer Program 704 Specific Instructions to perform the following steps:
  • At least two candidate building types selected by the target user are combined according to the joining operation to obtain at least one building type joining result, and the building type joining result is used as the target building type.
  • the computer program 704 before obtaining the generated design information of the target building land based on the surrounding supporting information and environmental information of the target building land, the computer program 704 is also used to execute the instructions of the following steps:
  • the building information includes building type information, building type information and building age information;
  • the building land is used as the target building land.
  • the computer program 704 is also used to perform instructions for the following steps:
  • the target area includes the area to be demolished, the building retention area and the blank land area, L is positive integer;
  • the computer program 704 is further configured to perform instructions for the following steps:
  • the sample image data includes image data related to buildings, natural landscapes and cultural landscapes in the city where the target construction land is located;
  • the preset number of candidate building plans with the highest similarity are displayed as recommended building plans.
  • the memory 702 may also be called a storage medium or a storage device, which is not limited in this embodiment of the present application.
  • the processor 701 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), or special-purpose processors.
  • Integrated circuits application specific integrated circuits, ASICs
  • off-the-shelf programmable gate arrays field-programmable gate arrays, FPGAs
  • FPGAs field-programmable gate arrays
  • the memory 702 mentioned in the embodiment of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories.
  • the non-volatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable programmable read-only memory (erasable PROM, EPROM), electrically removable memory. Erase programmable read-only memory (electrically EPROM, EEPROM) or flash memory.
  • Volatile memory may be random access memory (RAM), which is used as an external cache.
  • RAM random access memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access memory
  • synchronous DRAM synchronous DRAM, SDRAM
  • double data rate synchronous dynamic random access memory double data rate SDRAM, DDR SDRAM
  • enhanced synchronous dynamic random access memory enhanced SDRAM, ESDRAM
  • synchronous connection dynamic random access Memory synchronize link DRAM, SLDRAM
  • direct memory bus random access memory direct rambus RAM, DR RAM
  • processor 701 is a general processor, DSP, ASIC, FPGA or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component
  • the memory storage module
  • memory 702 described herein is intended to include, but is not limited to, these and any other suitable types of memory.
  • the bus 705 may also include a power bus, a control bus, a status signal bus, etc.
  • the various buses are labeled as buses in the figure.
  • each step of the above method can be completed by instructions in the form of hardware integrated logic circuits or software in the processor.
  • the steps of the methods disclosed in conjunction with the embodiments of the present application can be directly implemented by a hardware processor for execution, or can be executed by a combination of hardware and software modules in the processor.
  • the software module can be located in random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, registers and other mature storage media in this field.
  • the storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.
  • the size of the sequence numbers of the above-mentioned processes does not mean the order of execution.
  • the execution order of each process should be determined by its functions and internal logic, and should not be determined by the implementation process of the embodiments of the present application. constitute any limitation.
  • the disclosed systems, devices and methods can be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the units is only a logical function division. In actual implementation, there may be other division methods.
  • multiple units or components may be combined or can be integrated into another system, or some features can be ignored, or not implemented.
  • the coupling or direct coupling or communication connection between each other shown or discussed may be through some interfaces, and the indirect coupling or communication connection of the devices or units may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present application can be integrated into one processing unit, each unit can exist physically alone, or two or more units can be integrated into one unit.
  • the computer program product includes one or more computer instructions.
  • the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.
  • the computer instructions may be stored in or transmitted from one computer-readable storage medium to another, e.g., the computer instructions may be transferred from a website, computer, server, or data center To another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line) or wireless (such as infrared, wireless, microwave, etc.) line transmission.
  • the computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more available media integrated.
  • the available media may be magnetic media (eg, floppy disk, hard disk, tape), optical media (eg, DVD), or semiconductor media (eg, solid state drive), etc.
  • Embodiments of the present application also provide a computer storage medium.
  • the computer-readable storage medium stores a computer program.
  • the computer program is executed by a processor to implement any method of generating a building plan as described in the above method embodiments. some or all of the steps.
  • Embodiments of the present application also provide a computer program product.
  • the computer program product includes a non-transitory computer-readable storage medium storing a computer program.
  • the computer program is operable to cause the computer to execute the steps described in the above method embodiments. Some or all of the steps in any method of generating architectural plans.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Geometry (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Computer Hardware Design (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Educational Administration (AREA)
  • Primary Health Care (AREA)
  • Operations Research (AREA)
  • General Health & Medical Sciences (AREA)
  • Game Theory and Decision Science (AREA)
  • Architecture (AREA)
  • Civil Engineering (AREA)
  • Quality & Reliability (AREA)
  • Health & Medical Sciences (AREA)
  • Structural Engineering (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本申请公开了一种建筑方案生成的方法、装置、计算机设备以及存储介质。其中方法包括：根据目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，获取目标建筑用地的生成设计信息；根据生成设计信息通过预设评级模型对目标建筑用地进行评级，获得目标建筑用地对应的预估评级；根据预估评级按照预设生成规则生成针对目标建筑用地的至少一种候选建筑方案，并对至少一种候选建筑方案进行智能审核；根据设计需求信息生成目标评分维度，并根据目标评分维度对通过审核的候选建筑方案进行评分，将评分靠前的预设数量的候选建筑方案作为推荐建筑方案进行展示。实施本申请实施例，可以提高建筑设计的效率。

Description

建筑方案生成的方法、装置、计算机设备以及存储介质
本申请要求于2022年03月31日提交中国专利局、申请号为2022103279872、申请名称为“建筑方案生成的方法、装置、计算机设备以及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机技术领域，具体涉及一种建筑方案生成的方法、装置、计算机设备以及存储介质。
背景技术
目前的建筑方案多是设计人员根据相关的规范以及客户的要求，运用自身掌握的知识进行构思，然后利用计算机辅助设计(computer aided architectural design，CAD)来完成建筑图纸的绘制，模型的构造，模型的渲染等。这种方式极度依赖设计人员的经验和审美，需要耗费大量的时间和精力。
发明内容
本申请实施例提供了一种建筑方案生成的方法、装置、计算机设备以及存储介质，可以根据相关信息自动生成建筑方案，有利于提高建筑设计的效率。
第一方面，本申请实施例提供了一种建筑方案生成的方法，包括：
根据目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，获取目标建筑用地的生成设计信息；
根据生成设计信息通过预设评级模型对目标建筑用地进行评级，获得目标建筑用地对应的预估评级；
根据预估评级按照预设生成规则生成针对目标建筑用地的至少一种候选建筑方案，并对至少一种候选建筑方案进行智能审核；
根据设计需求信息生成目标评分维度，并根据目标评分维度对通过审核的候选建筑方案进行评分，将评分靠前的预设数量的候选建筑方案作为推荐建筑方案进行展示。
第二方面，本申请实施例提供了一种建筑方案生成的装置，包括：
获取模块，用于根据目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，获取目标建筑用地的生成设计信息；
评级模块，用于根据生成设计信息通过预设评级模型对目标建筑用地进行评级，获得目标建筑用地对应的预估评级；
生成模块，用于根据预估评级按照预设生成规则生成针对目标建筑用地的至少一种候选建筑方案，并对至少一种候选建筑方案进行智能审核；
展示模块，用于根据设计需求信息生成目标评分维度，并根据目标评分维度对通过审核的候选建筑方案进行评分，将评分靠前的预设数量的候选建筑方案作为推荐建筑方案进行展示。
第三方面，本申请实施例提供了一种计算机设备，包括处理器、存储器和通信接口，其 中，所述存储器存储有计算机程序，所述计算机程序被配置由所述处理器执行，所述计算机程序包括用于如本申请实施例第一方面中所描述的部分或全部步骤的指令。
第四方面，本申请实施例提供了一种计算机可读存储介质，所述计算机可读存储介质存储计算机程序，所述计算机程序使得计算机执行如本申请实施例第一方面中所描述的部分或全部步骤。
实施本申请实施例，将具有如下有益效果：
采用上述的建筑方案生成的方法、装置、设备以及存储介质，在根据目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，获取所述目标建筑用地的生成设计信息之后，根据生成设计信息通过预设评级模型对目标建筑用地进行评级，获得目标建筑用地对应的预估评级，然后根据预估评级按照预设生成规则生成针对目标建筑用地的至少一种候选建筑方案。通过预先对目标建筑用地进行评级，可以更为全面评估目标建筑用地，有助于提高候选建筑方案的准确性。此外，采用建筑生成设计生成建筑方案，相对于传统的计算机辅助设计而言，运用到了计算机强大的计算能力，可以生成多种可能的候选建筑方案，因此不仅可以提高建筑设计的效率，还可以解放建筑师的思维，打破模板式设计的僵硬，激发更多设计的可能性。在生成候选建筑方案之后，对至少一种候选建筑方案进行智能审核，根据设计需求信息生成目标评分维度，并根据目标评分维度对通过审核的候选建筑方案进行评分，将评分靠前的预设数量的候选建筑方案作为推荐建筑方案进行展示。如此，只对通过审核的候选建筑方案评分和展示，可以提高推荐建筑方案的准确性以及合理性。此外，可以优先展示匹配用户设计偏好的建筑方案，使得展示的建筑方案更符合用户的预期，以便降低用户查找的时间，从而提高设计的效率。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以基于这些附图获得其他的附图。其中：
图1为本申请实施例提供的一种系统架构示意图；
图2为本申请实施例提供的一种建筑方案生成的方法的流程示意图；
图3为本申请实施例提供的一种建筑方案生成的场景示意图；
图4为本申请实施例提供的另一种建筑方案生成的场景示意图；
图5为本申请实施例提供的又一种建筑方案生成的场景示意图；
图6为本申请实施例提供的一种建筑方案生成的装置的结构示意图；
图7为本申请实施例提供的一种计算机设备的结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
本申请的说明书和权利要求书及所述附图中的术语“第一”、“第二”、“第三”和“第四”等是用于区别不同对象，而不是用于描述特定顺序。此外，术语“包括”和“具有”以及它 们任何变形，意图在于覆盖不排他的包含。例如包含了一系列步骤或单元的过程、方法、系统、产品或设备没有限定于已列出的步骤或单元，而是可选地还包括没有列出的步骤或单元，或可选地还包括对于这些过程、方法、产品或设备固有的其它步骤或单元。
在本文中提及“实施例”意味着，结合实施例描述的特定特征、结构或特性可以包含在本申请的至少一个实施例中。在说明书中的各个位置出现该短语并不一定均是指相同的实施例，也不是与其它实施例互斥的独立的或备选的实施例。本领域技术人员显式地和隐式地理解的是，本文所描述的实施例可以与其它实施例相结合。
还应理解，本文中术语“和/或”，仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，本文中字符“/”，一般表示前后关联对象是一种“或”的关系。
为了更好地理解本申请实施例的技术方案，先对本申请实施例可能涉及的系统架构进行介绍。请参照图1，本申请实施例提供的一种系统架构示意图，该系统架构可以包括：电子设备101和服务器102。其中，电子设备101和服务器102之间可以通过网络通信。网络通信可以基于任何有线和无线网络，包括但不限于因特网、广域网、城域网、局域网、虚拟专用网络(virtual private network，VPN)和无线通信网络等等。
本申请实施例提供的建筑方案生成的方法可以由电子设备执行，例如，电子设备101执行本申请实施例的建筑方案生成的方法，电子设备101可以将生成的推荐建筑方案发送至服务器102，服务器102可以存储推荐建筑方案，服务器102可以将该推荐建筑方案发送至其他的设备。
具体地，电子设备101可以根据目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，获取目标建筑用地的生成设计信息；然后，可以根据生成设计信息通过预设评级模型对目标建筑用地进行评级，获得目标建筑用地对应的预估评级；再根据预估评级按照预设生成规则生成针对目标建筑用地的至少一种候选建筑方案，并对至少一种候选建筑方案进行智能审核；最后根据设计需求信息生成目标评分维度，并根据目标评分维度对通过审核的候选建筑方案进行评分，将评分靠前的预设数量的候选建筑方案作为推荐建筑方案进行展示。
本申请实施例不限定电子设备和服务器的数量，服务器可同时为多个电子设备提供服务。在本申请实施例中，电子设备可以是个人计算机(personal computer，PC)、笔记本电脑或智能手机，还可以是一体机、掌上电脑、平板电脑(pad)、智能电视播放终端、车载终端或便捷式设备等。PC端的电子设备，例如一体机等，其操作系统可以包括但不限于Linux系统、Unix系统、Windows系列系统(例如Windows xp、Windows 7等)、Mac OS X系统(苹果电脑的操作系统)等操作系统。移动端的电子设备，例如智能手机等，其操作系统可以包括但不限于安卓系统、IOS(苹果手机的操作系统)、Window系统等操作系统。
服务器可以是独立的服务器，也可以是提供云服务、云数据库、云计算、云函数、云存储、网络服务、云通信、中间件服务、域名服务、安全服务、内容分发网络(content delivery network，CDN)以及大数据和人工智能平台等基础云计算服务的云服务器。服务器或者可以通过多个服务器组成的服务器集群来实现。
目前的建筑方案多是设计人员根据相关的规范以及客户的要求，运用自身掌握的知识进行构思，然后利用计算机辅助设计(computer aided architectural design，CAD)来完成建筑图纸的绘制，模型的构造，模型的渲染等。这种方式极度依赖设计人员的经验和审美，需要耗费大量的时间和精力。
为了解决上述问题，本申请实施例提供了一种建筑方案生成的方法，该方法可以应用在建筑领域配置的电子设备或服务器上。通过实施该方法，可以根据相关信息自动生成建筑方案，有利于提高建筑设计的效率。
请参照图2，图2是本申请实施例提供的一种建筑方案生成的方法的流程示意图。以该方法应用于图1中的电子设备101为例进行举例说明，可以包括以下步骤S201-S204，其中：
步骤S201：根据目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，获取所述目标建筑用地的生成设计信息。
其中，目标建筑用地可以是指期望生成建筑方案的建筑用地。建筑用地可以是指开发商拟用于建造建筑物、构筑物的土地。建筑用地可以分为居住用地、商业用地、公共设施用地、工业用地、绿地和特殊用地等等。也就是说，可以在建筑用地上建住宅、商场、办公楼、学校、医院、工厂等等，本申请实施例对此不做限定。
目标建筑用地的地理位置信息可以采用经纬度来表示，目标建筑用地的地理位置信息可以通过目标用户自行输入，再由电子设备获取，也可以由电子设备通过GPS定位获取，本申请实施例对此不做限定。周边配套信息可以包括周边的公共设施信息、交通信息、人流信息以及周边建筑类型信息等等。公共设施信息可以包括地铁站、公交站、公园、图书馆、学校和医院的数量、位置等等。交通信息可以包括车流量、拥堵程度、红绿灯的数量等等。人流信息可以包括人流量等等。周边建筑类型信息可以包括周边建筑为高档住宅、普通住宅、别墅、写字楼、商场或者工厂，以及周边建筑的房价信息和楼层高度等等。环境信息可以是指目标建筑用地以及周边的噪声信息、绿化信息以及地形、土壤、水质，或者是否临江、临海等等。
目标用户通常为设计人员，更具体的可以是指电子设备101的使用者。设计需求信息可以是目标用户对生成的建筑方案的需求，比如，经济效益最大化、容积率最佳、布局最合理、品质最佳等要求。设计需求信息中也可以包括具体的建筑指标，例如容积率要求为2.0，绿地率要求为30％等等，或者还可以包括一些建筑规范的要求等等，本申请实施例对此不作限定。
生成设计信息可以用于作为后续候选建筑方案生成的约束。生成设计信息可以包括具体的建筑指标，例如容积率、绿地率和建筑限高等建筑指标，还可以包括相关建筑指标具体的参数，例如容积率2.0，绿地率30％等等。其中，容积率是指一定地块内，总建筑面积与建筑用地面积的比值。容积率可以反映开发强度，是建筑设计中经常研究的一项指标。绿地率是指一定地块内，绿地面积与建筑用地面积的比值，绿地率可以反映绿化程度以及环境舒适度。建筑限高是指对建筑高度进行限定。建筑限高有利于提高消防安全，倡导绿色低碳建设以及维护传统风貌，有利于实现更长远的发展。此外，生成设计信息还可以包括但不限于建筑密度、建筑布局、建筑楼型、建筑朝向、建筑日照间距、建筑位置间距以及这些指标相关的数值等等。
在一种可能的实施方式中，在步骤S201之前，还可以包括以下步骤：
获取建筑用地的类型，所述建筑用地的类型包括空白建筑用地以及待改建建筑用地；响应于所述建筑用地的类型为所述待改建建筑用地，获取所述建筑用地中所包含的建筑信息，所述建筑信息包括建筑类型信息、楼型信息以及建筑年龄信息；根据所述建筑信息确定所述建筑用地中的待改建区域，将所述待改建区域作为所述目标建筑用地；响应于所述建筑用地的类型为所述空白建筑用地，将所述建筑用地作为所述目标建筑用地。
在本申请实施例中，电子设备101在执行步骤S201之前，需要先选择一块建筑用地作为目标建筑用地。在一种可能的实施方式中，可以先对建筑用地的类型进行识别，检测该建筑 用地是否已经存在已建建筑物，如果识别出建筑用地中存在已建建筑物，那么建筑用地的类型为待改建建筑用地。如果识别出建筑用地中不存在已建建筑物，那么建筑用地的类型为空白建筑用地。具体的识别方式可以通过预先训练识别模型，通过识别模型进行识别。
请参见图3，图3是本申请实施例提供的一种建筑方案生成的场景示意图。如图3所示，当电子设备101的使用者(以下简称目标用户)对生成建筑方案的预设应用程序执行启动触发操作时，电子设备101可以响应于针对生成建筑方案的预设应用程序的启动触发操作，启动生成建筑方案的预设应用程序，并在预设应用程序可以在当前显示界面30a中显示对应的建筑用地，例如建筑用地A、建筑用地B、建筑用地C、建筑用地D、建筑用地E和建筑用地F等等。建筑用地通过建筑红线进行划分。在本申请实施例中，可以先确定建筑用地的类型，建筑用地的类型包括空白建筑用地以及待改建建筑用地。从图3可以看出，建筑用地A、建筑用地B、建筑用地C和建筑用地D为未建造建筑物的地块，可以作为空白建筑用地；建筑用地E和建筑用地F为已经建造建筑物的地块，可以作为待改建建筑用地。
在本申请实施例中，目标用户可以通过点击建筑用地，例如点击空白建筑用地(例如建筑用地A)，以查看建筑用地的建筑用地信息，建筑用地信息可以包括但不限于建筑用地面积，建筑用地的地理位置信息(例如，可以采用经纬度来表示)，建筑用地的地形，建筑用地的土壤条件，以及建筑用地的周边配套信息和环境信息等等。若点击待改建建筑用地(例如建筑用地E)，还可以查看建筑用地中所包含的建筑信息，建筑信息包括建筑类型信息、楼型信息以及建筑年龄信息等等。电子设备可以预先收集待改建建筑用地中已经建造的建筑的信息，目标用户通过点击相应的待改建建筑用地进行获取。在获得待改建建筑用地的建筑信息之后，可以根据建筑信息确定建筑用地中的待改建区域。在一种可能的实施方式中，电子设备可以根据预先收集的建筑信息，自动从待改建建筑用地中划分待改建区域，用作后续生成建筑方案的目标建筑用地，并用红线或者其他线向目标用户展示该待改建区域。在一种可能的实施方式中，目标用户还可以在待改建建筑用地(例如建筑用地E)中绘制一个或者多个闭合多边形，电子设备可以响应于该绘制操作，将该一个或者多个闭合多边形的区域作为待改建区域，并将该待改建区域作为目标建筑用地。其中，闭合多边形可以是圆形、四边形等规则的图形，也可以是不规则的图形。通过根据目标用户的绘制操作划分待改建建筑用地的待改建区域，有助于满足用户不同程度的需求。
目标用户在选择目标建筑用地时，若选取的是空白建筑用地，可以通过点击当前显示界面30a中的任一空白建筑用地，若该空白建筑用地可以作为目标建筑用地，例如，当目标用户点击到建筑用地A的区域时，可以生成如图3所示的提示信息30b(例如，可以具体表示为“将建筑用地A作为目标建筑用地”)，该提示信息30b可以包括“是”(上述目标建筑用地确认控件)和“否”(上述目标建筑用地否认控件)两个控件。当目标用户选择对“是”控件执行触发操作时，电子设备101将建筑用地A作为目标建筑用地，并在当前显示界面30a中显示目标建筑用地30c，提供生成设计信息区域30d。目标用户可以在生成设计信息区域30d中输入生成设计信息，例如目标用户在生成设计信息区域30d中的容积率输入框中输入为2.0等。电子设备101通过读取目标用户在生成设计信息区域30d的数据以获取目标建筑用地的生成设计信息。除此之外，还可以通过其他方式来获取目标建筑用地的生成设计信息，将在下文进行详细的描述，在此不做赘述。
当目标用户选择对“否”控件执行触发操作时，电子设备101可以关闭提示信息30b，不将建筑用地A作为目标建筑用地。此外，还可以生成图3中未涉及的提示消息，例如“请重新选择建筑用地作为目标建筑用地”等之类的提示信息，以提示目标用户重新选择目标建 筑用地。
在一种可能的实施方式中，当前显示界面30a还可以包括图3中未涉及的控件，例如搜索框，搜索框可以用于搜索地点、建筑用地等。例如，当目标用户想要进行生成建筑方案的目标建筑用地没有在当前显示界面30a显示时，目标用户可以在搜索框中输入某地理位置信息，然后电子设备101在当前显示界面30a显示该地理位置信息对应的建筑用地。若当前显示界面30a显示的建筑用地中包含想要进行生成建筑方案的目标建筑用地，则目标用户不需要在搜索框进行搜索操作。另外，还可以对当前显示界面30a中显示的建筑用地进行平移、放大或者缩小操作，以便更好的观察建筑用地。
可以看出，通过预先确定建筑用地的类型，然后根据建筑用地的类型的不同，从而确定不同类型的目标建筑用地，从而有助于提高后续生成的候选建筑方案的准确性。
在一种可能的实施方式中，还可以包括以下步骤：获取所述待改建区域所在城市的城市文化，根据所述城市文化以及所述建筑信息将所述待改建区域划分为L个目标区域，所述目标区域包括待拆除区域、建筑物保留区域以及空白用地区域中的至少一种，所述L为正整数；响应于所述目标区域为建筑物保留区域，获取所述建筑物保留区域内所包含的建筑物的功能和布局；根据所述建筑物的布局和功能生成针对所述建筑物保留区域的至少一种候选建筑方案。
在本申请实施例中，城市文化可以是指一个城市的景观、艺术、习俗等相关特征的集合。一个城市的建筑物可以构成一个城市文化的一部分，也就是说一个城市的文化也可以在建筑物中体现。待改建区域内的旧建筑见证了城市发展，作为时代的印记，它们用建筑形态诉说着过去的辉煌。但现在大多己经丧失了原有的使用功能，脱离了城市的肌理。因此，可以对待改建区域内的旧建筑进行改建，重新赋予其新的生命，从而能够完善城市服务功能，增强城市发展历史的厚重感。对旧建筑的改建离不开一个城市的文化，当待改建区域内的旧建筑的年龄过老，且已经明显与所属城市的文化格格不入时，则需要对其进行拆除重建；而旧建筑如果功能完善，则可以保留，并在原先的基础上，进行相关的改建。待改建区域所在城市的城市文化的获取，可以通过读取目标用户输入的数据进行获取，也可以通过收集城市所包含的建筑物形态、自然景观以及文化遗产的相关影像数据，利用预设模型提取其相关特征进行获取。
然后根据城市文化以及建筑信息将所述待改建区域划分为L个目标区域，目标区域可以包括待拆除区域、建筑物保留区域以及空白用地区域，L为正整数，具体取值可以根据目标用户自行设定，或者系统默认设置。划分的具体实现方式可以参考下文功能分区的划分，在此不再赘述。待拆除区域是指待改建区域中需要对原有旧建筑进行拆除重建的区域；建筑物保留区域是指待改建区域中保留原有旧建筑物，在原有旧建筑物的基础上对其进行改建的区域；空白用地区域是指待改建区域中没有建造建筑物的区域。对于在待拆除区域以及空白用地区域生成建筑方案的具体实施方式，可以参考步骤S201-步骤S204的相关描述，在此不再赘述。对于建筑物保留区域而言，则需要考虑原来建筑物的功能和布局，以及原来建筑物的建造材料、外形、尺寸、颜色等相关因素，并且根据这些相关因素结合城市文化以及周边配套信息、环境信息等因素，生成相关的候选建筑方案，以赋予旧建筑物新的生命。
可以看出，通过获取待改建区域所在城市的城市文化，然后根据城市文化以及建筑信息将待改建区域划分为待拆除区域、建筑物保留区域以及空白用地区域等在内的目标区域，可以根据不同的目标区域生成不同的候选建筑方案，有助于提高建筑设计的效率。若目标区域为建筑物保留区域，则获取建筑物保留区域内所包含的建筑物的功能和布局，并根据建筑物 的布局和功能生成针对建筑物保留区域的至少一种候选建筑方案，以实现对旧建筑物的改建，并且可以提高改建的合理性。
在一种可能的实施方式中，步骤S201具体可以包括以下步骤：
获取目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息；根据所述周边配套信息、所述环境信息以及所述设计需要信息确定所述目标建筑用地的客户定位和产品定位；根据所述客户定位和产品定位确定所述目标建筑用地的生成设计信息。
在本申请实施例中，客户定位是指目标建筑用地建成建筑物之后所面向的客户的类型，包括高收入客户、中等收入客户、低收入客户或者首次置业客户、二次置业客户等等。产品定位主要是指产品品质定位，包括高端产品和中端产品，高端产品可以是别墅和洋房等，中端产品可以是高层等。例如，目标建筑用地所在的周边配套资源完善，交通发达，环境舒适，噪声污染低，水质良好，那么客户定位可以定位与高收入客户，产品定位可以为别墅或者洋房等高端产品。此时，电子设备可以根据以往别墅或者洋房的案例自动生成相关的生成设计信息。具体可以通过条件生成模型来实现。电子设备通过获取大量的历史建筑方案，并提取历史建筑方案中相关的历史数据作为条件生成模型的样本训练数据。历史数据可以包括关于历史建筑方案中的建筑类型(例如别墅、洋房、高层、中低层等)，以及历史建筑方案的相关生成设计信息，例如容积率、绿地率、建筑限高、建筑密度、建筑布局、建筑楼型、建筑朝向、建筑日照间距、建筑位置间距等等。将大量的历史数据作为样本训练数据输入至条件生成模型中进行训练，得到更准确的条件生成模型。其中，条件生成模型的输入为建筑类型，输出为建筑类型对应的相关生成设计信息。可以理解的是，由于条件生成模型对人为输入的大量的历史数据进行了充分的学习，在条件生成模型在训练完成之后，可以总结出相关条件生成的基本规律。因此，在应用条件生成模型时，在输入建筑类型之后，可以生成较为准确且合理的生成设计信息。在本申请实施例中，条件生成模型可以是循环神经网络(recurrent neural network，RNN)、卷积神经网络(convolutional neural network，CNN)、支持向量机(support vector machine，SVM)等模型中的一种或多种。
可以看出，在根据地理位置信息确定目标建筑用地的周边配套信息、环境信息以及目标用户的设计需求信息之后，根据周边配套信息、环境信息和设计需求信息确定目标建筑用地的客户定位和产品定位，从而确定建筑类型，然后根据建筑类型利用模型自动生成目标建筑用地的生成设计信息。用户无需手动输入生成设计信息所需的相关数据，可以提高设计的效率，缩短整体方案设计的时间。
在一种可能的实施方式中，步骤S201具体可以包括以下步骤：
获取目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，通过多模态模型输出所述周边配套信息、所述环境信息以及所述设计需求信息对应的多模态特征，所述多模态特征是对所述周边配套信息对应的第一特征、所述环境信息对应的第二特征，以及所述设计需求信息对应的第三特征进行融合得到的；将所述多模态特征输入至预设地块划分模型，通过所述预设地块划分模型将所述目标建筑用地划分为N个功能分区，所述功能分区包括高层建筑区、别墅区、公共设施区和商业区中的至少一种，所述N为正整数；获取每一所述功能分区的建筑用地面积，根据所述功能分区的建筑用地面积以及所述功能分区的功能确定每一所述功能分区的子设计条件；根据所述子设计条件确定所述目标建筑用地的生成设计信息。
在本申请实施例中，可以通过第三方地图应用程序所提供的API接口来获取以获取到目标建筑用地的地理位置信息，该地理位置信息可以采用经纬度来表示。在获取目标建筑用地 的地理位置信息之后，可以以目标建筑用地为中心，获取目标建筑用地预设距离范围内的周边配套信息和环境信息。预设距离为预先设置的参数，具体的取值可以根据历史经验来设定，例如可以设置为2千米，或者也可以根据实际情况来进行设定，本申请实施例对此不做限定。在一种可能的实施方式中，可以通过第三方地图应用程序所提供的API接口来获取周边配套信息中的公共设施信息；可以通过第三方地图应用程序提供的热力图来分析目标建筑用地周边的人流量。交通信息以及周边建筑类型信息等配套信息可以通过政府网站和房产网站等第三方网站获取。或者，还可以通过其他的方式获得目标建筑用地的周边配套信息，本申请实施例对此不做限定。
环境信息中的噪声值可以通过在目标建筑用地的预设位置安装多个噪声检测设备，电子设备可以与多个噪声检测设备之间可以建立通信连接。其中，噪声检测设备用于检测噪声，可以为声级计、噪声检测传感器等，本申请实施例对此不作限定。多个噪声检测设备所设置的位置可以不同，当同时执行噪声检测操作时，能够实现多点位噪声检测。噪声检测设备可以安装在目标建筑用地的用地轮廓线附近，每个噪声检测设备间隔的距离可以是300米、400米等，具体取值可以根据实际情况确定。电子设备在预设时间间隔内，接收多个噪声检测设备发送的噪声检测数据。该预设时间间隔可以为10分钟、5分钟等，本申请实施例对此不作限定。噪声检测数据即为多点位噪声检测数据，至少包括每个噪声检测装置的设备标识以及与每个设备标识对应的检测数据。电子设备读取噪声检测结果之后，分别输出每个噪声检测设备的检测到的噪声值，从而获得目标建筑用地环境信息中的噪声值。可以看出，该方法能够同时控制多个噪声检测设备进行噪声检测，同步性好，有利于提升噪声检测的可靠性。环境信息中的绿化信息以及地形、土壤、水质等等可以通过政府网站、生态环境局网站或者其他第三方网站中获取。此外，目标用户的设计需求信息可以通过目标用户输入的数据进行获取。
多模态模型可以是用于获取多模态特征的模型。具体地，电子设备可以将周边配套信息、环境信息以及设计需求信息输入至多模态模型中，通过多模态模型可以输出这些输入信息对应的多模态特征。其中，多模态特征是对周边配套信息对应的第一特征、环境信息对应的第二特征，以及设计需求信息对应的第三特征进行融合得到的。第一特征可以是周边配套信息的相关的特征，如周边的公共设施数量及位置等等；第二特征可以是环境信息相关的特征，如噪声值、绿地率等等；第三特征可以目标用户的设计需求信息相关的特征，如具体的建筑指标的取值等等。在实际应用中，多模态特征还可以加入其他与建筑生成相关的特征，本申请实施例对此不作限定。在本申请实施例中，多模态模型可以是由循环神经网络(recurrent neural network，RNN)、卷积神经网络(convolutional neural network，CNN)、支持向量机(support vector machine，SVM)等模型中的一种或多种经过训练得到的模型。
其中，功能分区可以包括高层建筑区、别墅区、公共设施区和商业区中的至少一种，功能分区还可以包括洋楼区、矮层区以及混合区(高矮层混合、别墅和洋楼混合等等)等等。N为大于或等于1的正整数，在实际应用中，N可以由目标用户自行设定，也可以由系统预先设定，本申请实施例对此不做出限定。子设计条件可以理解为每一功能分区的生成设计信息，与生成设计信息类似地，子设计条件可以包括容积率、绿地率、建筑限高、建筑密度、建筑布局、建筑楼型、建筑朝向、建筑日照间距、建筑位置间距等等。
请参见图4，图4是本申请实施例提供的另一种建筑方案生成的场景示意图。电子设备在获取目标建筑用地周边配套信息和环境信息，以及需求设计信息等有关的多模态特征之后，可以将多模态特征输入至预设地块划分模型，通过预设地块划分模型将目标建筑用地划分为 N个功能分区。具体地，可以利用预设工具(例如散点轮廓算法(Alpha Shapes))提取目标建筑用地的用地轮廓线。一般而言，目标建筑用地的用地轮廓通常为多边形，可以根据用地轮廓线确定多条轮廓边线，每条轮廓边线的周边配套信息和环境信息可能不同。然后利用地块划分模型将目标建筑用地划分为多个功能分区。可以理解的是，每个功能分区的面积和建筑类型可以相同，也可以不相同，具体的划分规则根据目标建筑用地的用地轮廓线的各边轮廓边线对应的周边配套设施的种类和数量以及环境信息，以及目标用户的需求设计信息确定。此外，地块划分模型在实际应用之前，可以采集大量的数据进行训练，从而使得地块训练模型，可以总结出地块划分的基本规律，以使功能分区的划分较为准确且合理，从而提高建筑业态分布的合理性。
如图4所示，当前显示界面40a中包括可视化区域40b和生成设计信息区域40c。可视化区域40b用于展示目标建筑用地，在确定目标建筑用地周边配套设施信息以及环境信息之后，可以利用地块划分模型将目标建筑用地划分为4个功能分区，分别为功能分区1(40d)、功能分区2(40e)、功能分区3(40f)和功能分区4(40g)。其中，功能分区1和功能分区4为对应轮廓边线的周边配套资源完善，交通发达，环境舒适，噪声污染更低，因此可以将功能分区1和功能分区4作为别墅区，将功能分区3用作高层建筑区，并将功能分区2用作公共设施区(可以用于建造幼儿园、社区办事处或者绿化区等等)。当然，还可以包括图4中未示出的其他的划分方式，为了提高灵活性，本申请实施例可以提供多种不同的划分方式供用户选择。在一种可能的实施方式中，还可以根据目标建筑用地的地形进行划分，如目标建筑用地上方的地势明显比下方的地势高，也可以将目标建筑用地划分为上下两个功能分区，上方的功能分区用作矮层建筑区，下方的功能分区用作高层建筑区等等。在一种可能的实施方式中，目标用户还可以在可视化区域40b中绘制一个或者多个闭合多边形，电子设备可以响应于该绘制操作，在可视化区域40b中显示绘制操作所绘制的一个或者多个闭合多边形，并根据该一个或者多个闭合多边形将目标建筑用地划分为多个功能分区。其中，闭合多边形可以是圆形、四边形等规则的图形，也可以是不规则的图形。通过根据目标用户的绘制操作将目标建筑用地划分为多个功能分区，有助于满足用户不同程度的需求。
如图4所示，在得到多个功能分区之后，可以获取每个建筑用地的建筑用地面积，然后根据功能分区的功能利用上述的条件生成模型确定每一功能分区的子设计条件。例如，在确定功能分区1用作别墅区之后，可以利用条件生成模型，自动生成功能分区1相关的子设计条件，如自动生成功能分区1的容积率、绿地率、建筑限高、建筑密度、建筑布局、建筑楼型、建筑朝向、建筑日照间距、建筑位置间距等等相关的子设计条件。目标建筑用地的生成设计信息以及各个功能分区的子设计条件可以在生成设计信息区域40c中显示。为了进一步提高用户体验，目标用户还可以对各个子设计条件进行手动修改。
可以看出，在获取目标建筑用地周边配套信息和环境信息，以及需求设计信息等有关的多模态特征之后，可以将多模态特征输入至预设地块划分模型，通过预设地块划分模型将目标建筑用地划分为多个功能分区，然后再根据功能分区的功能确定各个功能分区的子设计条件，最后根据子设计条件确定目标建筑用地的生成设计信息。如此，使得各个设计条件的设定更加合理，同时也有助于增加建筑方案的多样性，从而有助于提高建筑方案生成的合理性。
步骤S202：根据所述生成设计信息通过预设评级模型对所述目标建筑用地进行评级，获得所述目标建筑用地对应的预估评级。
预估评级可以是指对目标建筑用地作一个初步的评定，例如，根据生成设计信息可以初步判定目标建筑用地可以用作高档、中档或者低档住宅等等。在本申请实施例中，可以通过 建立基础模型，基础模型可以是卷积神经网络模型等，获取大量的历史建筑用地的历史生成设计信息，从历史生成设计信息中提取出各历史建筑用地的历史生成设计信息以及对应的历史建筑用地的预估评级数据，例如历史生成设计信息可以是容积率0.5，绿地率40％，建筑密度15％，其对应的历史建筑用地的预估评级可以是高档小区等等。然后将大量的历史生成设计信息以及对应的历史建筑用地的预估评级数据作为样品评级数据。并将样本评级数据输入至基础模型中进行训练，根据训练结果获得预设评级模型。最后将生成设计信息输入至预设评级模型，从而得到目标建筑用地的预估评级，从而可以进一步对目标建筑用地作更精细化的评估，有助于提高候选建筑方案的准确性。
步骤S203：根据所述预估评级按照预设生成规则生成针对所述目标建筑用地的至少一种候选建筑方案，并对所述至少一种候选建筑方案进行智能审核。
随着数字技术的不断发展，计算机学科和建筑学科的不断融合和渗透，计算机作为建筑设计的实现平台，建筑学学科开始走向计算机辅助建筑设计的时代。在传统的计算机辅助设计中，通常首先是建筑师运用自身掌握的知识进行构思，然后使用计算机辅助设计来来完成建筑图纸的绘制，模型的构造，模型的渲染等。然而，传统的计算机辅助设计需要耗费建筑师大量的时间和精力，导致建筑设计的效率低下。
为了提高建筑设计效率，建筑生成设计应运而生。相对传统的计算机辅助设计来说，建筑生成设计是将计算机的作用在设计阶段前置，使得设计过程能够运用到计算机强大的计算能力，生成多种可能的建筑设计方案。因此，建筑生成设计不仅可以提高建筑设计的效率，还可以解放建筑师的思维，打破模板式设计的僵硬，激发更多设计的可能性。
在本申请实施例中，预设生成规则可以是预先在电子设备中内置的一些生成规则，如一些预设排布方式、预设建筑朝向，以及预设约束信息(例如，位置约束、日照约束、消防约束等等)。候选建筑方案可以包括经济效益最大化方案、布局合理方案、最优方案、容积率最佳方案、绿地率最佳方案以及品质最佳方案等等。其中，经济效益最大化方案可以是指根据预计利润最高的方案；布局合理方案可以是指满足“西高东低、北高南低”“同类型组团布局”“布局饱满性”的布局要求等要求的方案；最优方案可以是指方案中的容积率、绿地率、建筑限高等均接近生成设计信息的方案；容积率最佳方案可以是指方案中的容积率最接近生成设计信息的方案；绿地率最佳方案是指方案中的绿地率最接近生成设计信息的方案；品质最佳方案是指方案中的容积率接近生成设计信息，且建筑之间位置间距最远的方案。除上述所提及的方案之外，候选建筑方案还可以包括其他可能的方案，本申请实施例不进行一一列举。
在生成至少一种候选建筑方案之后，可以对生成的至少一种候选建筑方案进行智能审核。具体地，可以根据相关的日照规范、楼间距规范、消防规范或者其他城市规划相关的规范，对生成的至少一种候选建筑方案进行日照、楼间距或者消防以及其他参数的审核。若候选建筑方案通过审核，则执行步骤S204；若没有通过审核，则将该方案进行舍弃或者从删除。如此，可以提高推荐建筑方案的合理性。
在一种可能的实施方式中，步骤S203中的根据所述预估评级按照预设生成规则生成针对所述目标建筑用地的至少一种候选建筑方案，具体可以包括以下步骤：
从预设楼型库查找所述预估评级对应的目标楼型库，所述目标楼型库中存储有所述预估评级对应的参考楼型和所述参考楼型的参数信息，以及所述参考楼型的评分值；将所述目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向所述目标用户展示，将所述目标用户选中的候选楼型作为目标楼型；获取所述目标建筑用地的用地轮廓线，根据所述目标楼型、预设排布方式以及所述用地轮廓线生成针对所述目标建筑用地的至少一种候选建筑方案，所 述预设排布方式包括行列式、周边式、点群式以及极限高低配式。
在本申请实施例中，电子设备可以预先收集多个楼盘的楼型、楼型的参数以及各楼型相应的评分值构建预设楼型库。楼型可以包括三室两厅两卫、三室两厅一卫、两室两厅一卫和三室两厅两卫，以及阳台、露台、飘窗的数量等等。楼型的参数可以包括户数和每户的面积，也可以包括建筑物层数、每层高度、建筑物朝向、架空层高度等等；还可以包括阳台个数、露台个数、飘窗个数以及客厅朝向等等。具体地，可以预先将各个开发商楼盘评分较高的户型以及相关的户型配置信息收录在预设楼型库中，在构建预设楼型库时，可以对每一楼盘的楼型添加标签，该标签用于记录每一楼盘对应的预估评级，然后根据该标签对预设楼型库进行分类，例如分为高档住宅楼型库，别墅楼型库或者高层建筑楼型库等等，以便查找。预估评级的定义可以参考前文的描述，在此不再赘述。在本申请实施例中，可以将目标建筑用地的预估评级作为查询信息，从预设楼型库中查找到对应的楼型库分类，以作为目标楼型库。目标楼型库可以存储有估评级对应的参考楼型和参考楼型的参数信息，以及参考楼型的评分值。然后，将目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向目标用户展示。预设数量可以参考的定义可以参考下文的描述，在此不再赘述。在确定目标建筑用地对应的目标楼型库之后，提示目标用户选择相应的候选楼型，并将目标用户选中的候选楼型作为目标楼型。其中，目标用户可以选中一种候选楼型，也可以选中多种候选楼型，本申请实施例对此不作限定。
目标用户可以对目标楼型的参数进行修改，如修改楼层的高度，建筑物的朝向，架空层的高度，楼层的层数等等，以便得到更符合目标用户要求的目标楼型参数。在获取目标建筑用地的用地轮廓线之后，可以根据目标楼型、预设排布方式以及用地轮廓线生成针对目标建筑用地的至少一种候选建筑方案。其中，预设排布方式包括行列式、周边式、点群式以及极限高低配式。极限高低配式可以是指在满足其他指标(例如建筑密度、容积率以及日照等)的情况下，容纳最多建筑物的排布方式。
可以看出，根据预估评级确定目标楼型库，再将用户选中的候选楼型作为目标楼型，最后根据目标楼型、预设排布方式以及用地轮廓线生成针对目标建筑用地的至少一种候选建筑方案，有助于提高候选建筑方案生成的效率以及准确性。
在一种可能的实施方式中，步骤将所述目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向目标用户展示，将所述目标用户选中的候选楼型作为目标楼型，可以具体包括以下步骤：
将所述目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向目标用户展示；响应于检测到所述目标用户选中拼合操作，根据所述拼合操作将所述目标用户选中的至少两个候选楼型进行拼合，得到至少一种楼型拼合结果，并将所述楼型拼合结果作为目标楼型。
在本申请实施例中，拼合操作是指将两个以及两个以上的候选楼型进行合并。具体的拼合数量可以根据地理位置、配套信息以及环境信息，确定产品定位，然后根据产品定位确定。用作拼合的候选楼型可以是完全相同的楼型，也可以不同的楼型。还可以根据不用的朝向以及不同的拼合数量等等参数生成多种楼型拼合结果，并将多种拼合结果推送给目标用户，供目标用户选择，将目标用户选中的楼型拼合结果作为目标楼型。此外，也可以将生成的多种楼型拼合结果均作为目标楼型，根据产品定位或者其他的要求对生成的楼型拼合结果按照从优到劣依次排序，并按照排序依次进行方案排布，直到排布满足用户所需的相关指标(例如容积率，覆盖率)，从而生成相应的候选建筑方案。
可以看出，将多个候选楼型进行拼合，从而得到多种楼型拼合结果的拼合操作，有助于 解放建筑师的思维，打破模板式设计的僵硬，激发更多设计的可能性，从而提高候选建筑方案的多样性。
建筑布局作为建筑设计的重要步骤，决定着该建筑功能布局和流线设计的合理性，在设计过程中通常需要花费了建筑师的大量时间和精力。在一种可能的实施方式中，步骤获取所述目标建筑用地的用地轮廓线，根据所述目标楼型、预设排布方式以及所述用地轮廓线生成针对所述目标建筑用地的至少一种候选建筑方案，具体可以包括以下步骤：
获取所述目标建筑用地的用地信息，所述用地信息包括所述目标建筑用地的占地面积和用地轮廓线；获取所述目标建筑用地中每一建筑物与相邻建筑物的约束信息；根据所述用地信息和所述约束信息生成所述目标建筑用地的建筑布局。
在本申请实施例中，约束信息可以包括目标距离约束值，目标距离约束值可以用于表示目标建筑用地中每一建筑物与相邻建筑物之间的最小距离。约束信息还可以包括目标日照间距约束值或者消防间距约束值等等，目标日照间距约束值可以用于表示目标建筑用地中每一建筑物与相邻建筑物之间的最小日照间距；消防间距约束值可以用于表示目标建筑用地中每一建筑物与相邻建筑物之间的最小消防间距。然后根据用地信息和约束信息生成目标建筑用地的建筑布局。
此外，还可以将目标建筑用地划分为不同的功能分区，具体的划分方式可以参考前文的描述。根据功能分区确定建筑布局，也就是说每一功能分区的建筑布局可以相同，也可以不同，具体可以根据实际情况确定。在本申请实施例中，可以根据功能分区的占地面积和用地轮廓线，结合其他子设计条件，获取得到功能分区中建筑物的数量、建筑楼型以及建筑层数等等，并根据每一建筑物与相邻建筑物的约束信息，以约束信息为目标距离约束值为例，每一建筑物与相邻建筑物的物理距离需要大于或等于目标距离约束值，从而生成一种或者多种满足约束信息的建筑布局。具体的排布方式可以包括行列式、周边式、点群式以及混合式中一种或者多种。在确定排布方式时，可以考虑气候以及光照条件以及每一功能分区中具体的建筑类型。示例地，若目标建筑用地所在城市的光照条件不充足时，宜采用行列式；若目标建筑用地所在城市较为寒冷时，宜采用周边式。若功能分区为别墅区、高层建筑区或者洋楼区等单一类型的功能分区，宜采用行列式或者周边式；如功能分区为混合区(高矮层混合、别墅和洋楼混合等等)，宜采用点群式或者混合式。此外，还可以是其他的建筑布局方式，本申请实施例对此不做限定。
可以看出，根据目标建筑用地的用地信息和约束信息生成目标建筑用地的建筑布局，可一次性生成多种建筑布局图，相比于人工布局方式，可以有效提升建筑布局效率和准确度，且可以节省大量的时间和精力。
在一种可能的实施方式中，获取所述目标建筑用地中每一建筑物与相邻建筑物的约束信息的步骤，可以具体包括以下步骤：获取用户输入的所述目标建筑用地中每一建筑物与相邻建筑物的第一距离约束值；若所述第一距离约束值大于或等于预设距离约束值，则将所述第一距离约束值作为所述目标距离约束值；或者若所述第一距离约束值小于所述预设距离约束值，则提示所述用户对所述第一距离约束值进行调整，直至所述第一距离约束值大于或等于所述预设距离约束值。
在本申请实施例中，预设距离约束值可以用于表示目标建筑用地中每一建筑物与相邻建筑物之间的最小规定距离。当每一建筑物与相邻建筑物之间的物理距离小于预设距离约束值时，对应的建筑方案不满足规范，将无法使用。预设距离约束值的具体取值可以根据目标建筑用地所在城市的建筑设计规范数据或者其他建筑设计相关规范确定。因此，为了防止生成 的候选建筑方案不满足实际需要，用户输入的第一距离约束值需要大于或等于预设距离约束值。若第一距离约束值小于预设距离约束值，则需要提示用户该第一距离约束值对第一距离约束值进行调整，直至第一距离约束值大于或等于预设距离约束值，并将大于或等于预设距离约束值的第一距离约束值作为目标距离约束值。
可以看出，将大于或等于预设距离约束值的第一距离约束值作为目标距离约束值，可以使得生成的候选建筑方案的建筑布局较为合理，进一步可以提高候选建筑方案的合理性和准确性。
在一种可能的实施方式中，所述目标建筑用地包括所述目标建筑区域，步骤S203中的根据所述预估评级按照预设生成规则生成针对所述目标建筑用地的至少一种候选建筑方案，具体可以包括以下步骤：
获取所述目标建筑用地的建筑用地轮廓线；获取各边所述建筑用地轮廓线对应的退线值；根据所述退线值确定目标建筑区域；根据所述预估评级按照预设生成规则生成针对所述目标建筑区域的至少一种候选建筑方案。
在本申请实施例中，退线值是指目标建筑用地的建筑用地轮廓线(或者用地红线)与建筑物建造区域的距离。目标建筑区域为建造建筑物的区域，目标建筑区域的面积小于目标建筑用地的面积。建筑用地轮廓线的获取可以参考前文的具体实现方式。各边用地轮廓线对应的退线值可以由目标用户输入的数值获取，也可以根据相关的规范要求由电子设备自动生成。在确定各边用地轮廓线对应的退线值之后，确定目标建筑区域，并根据预估评级按照预设生成规则生成针对目标建筑区域的至少一种候选建筑方案。具体的实现方式可以参考前文的描述，在此不做赘述。如此，可以提高候选建筑方案生成的合理性，也便于后续对目标建筑用地作扩建或者其他处理。
在一种可能的实施方式中，根据所述预估评级按照预设生成规则生成针对所述目标建筑用地的至少一种候选建筑方案；若生成失败，则将每一所述生成设计信息按照权重从大到小进行排序；并将排在前面的预设数量的生成设计信息作为目标生成设计信息；根据所述目标生成设计信息按照预设生成规则生成针对所述目标建筑用地的至少一种候选建筑方案。
在本申请实施例中，生成设计信息可能有多个，有的生成设计信息对整体方案的生成起着至关重要的作用，而有的生成设计信息则无足轻重。当生成设计信息过多时，可以会导致生成候选建筑方案失败。如果生成失败，可以获取每一生成设计信息的权重，并按照权重从大到小将生成设计信息进行排序，将排在前面的预设数量的生成设计信息作为目标生成设计信息，将权重低的生成设计信息舍弃，从而减少生成候选建筑方案的约束条件，以提高方案生成的成功率。权重的确定可以参考下文的描述，在此不再赘述。最后根据目标生成设计信息按照预设生成规则生成针对所述目标建筑用地的至少一种候选建筑方案。具体生成过程的实现方式可以参考前文的描述吗，在此不再赘述。在生成失败的情况下，通过将排在前面的预设数量的生成设计信息作为目标生成设计信息，然后再根据目标生成设计信息按照预设生成规则生成针对目标建筑用地的至少一种候选建筑方案，可以提高方案生成的成功率，从而提高建筑设计的效率。
在本申请实施例中，由于可以生成多种候选建筑方案，这就导致了靠前展示的候选建筑方案可能无法准确匹配用户的设计偏好，从而使得用户需要耗费较多的时间来筛选符合预期的候选建筑方案。对此，本申请实施例可以通过执行步骤S204，以降低用户查找的时间，从而提高设计的效率。
步骤S204：根据所述设计需求信息生成目标评分维度，并根据所述目标评分维度对通过 审核的候选建筑方案进行评分，将评分靠前的预设数量的候选建筑方案作为推荐建筑方案进行展示。
在本申请实施例中，可以通过对目标用户的设计需求信息设置多个权重参数选项，将权重参数选项进行展示，供目标用户确认或者修改。权重参数选项可以通过数值标识，比如：1.0表示非常重要的需求，0.8表示重要的需求，0.6表示比较重要的需求，0.4表示一般重要的需求以及0.2表示普通需求。还可以是更多的选项设置或者其他选项设置，本申请实施例对此不做限定。可理解的是，若目标用户的设计需求信息的数量繁多，有些边缘需求对通过审核的候选建筑方案的筛选参考价值不高，可以不作考虑，权重高的设计需求信息作为目标用户比较关注的需求，将其作为评分点，更能体现目标用户的实际生成设计需求，并将权重值排在前面的预设数量的设计需求信息作为目标评分维度，以使展示的推荐建筑方案能够贴合目标用户最关注的需求。在实际应用中，预设数量可以有系统预先设定，也可以由目标用户自行确定。
请参见图5，图5是本申请实施例提供的又一种建筑方案生成的场景示意图。如图5所示，在生成候选建筑方案之后，可以在当前显示界面50a中的推荐建筑方案列表50d中可以缩略展示评分靠前的预设数量的推荐建筑方案。当目标用户选择推荐建筑方案列表50d中的某一个推荐建筑方案时，可以在可视化区域50b中显示放大后的推荐建筑方案，并且可以显示该推荐建筑方案对应的参数，参数可以包括但不限于建筑用地面积、容积率、绿地率、建筑楼型、建筑类型、建筑朝向、建筑排布、建筑风格等等。如图5所示，以选中方案1为例，在可视化区域50b中可以显示放大后的方案1。在本申请实施例中，需求设计信息可以是具体的生成设计信息或者相关的建筑指标，例如容积率、绿地率、建筑密度以及建筑限高等等。生成设计信息区域50c可以显示每一生成设计信息的具体数值以及每一生成设计信息对应权重的数值。每一生成设计信息的权重可以由目标用户自行设定，也可以系统预先设定。然后根据每一生成设计信息中权重的高低来确定目标评分维度，以对候选建筑方案进行评分。评分的高低可以用于表征生成设计信息的权重与推荐建筑方案的相关程度，评分与生成设计信息的权重呈正相关。评分越高，推荐建筑方案列表50d中方案排序越靠前。例如，生成设计信息区域50c中生成设计信息的容积率为2.0，绿地率为30％，容积率对应的权重为0.98，绿地率对应的权重为0.75。其中，权重取值可以为0-1.0。方案1的容积率为2.0，绿地率为25％；方案2的容积率为1.7，绿地率为30％。由于容积率的权重大于绿地率的权重，而方案1的容积率更接近目标建筑用地生成设计信息说对应的容积率，因此，方案1的评分高于建筑方案2的评分，那么方案1的展示顺序比方案2靠前。在本申请实施例中，目标用户可以随时对生成设计信息区域50c中每一生成设计信息的权重进行调整，在对权重进行调整之后，推荐建筑方案列表50d对应方案的展示顺序也会随之调整，以便优先给用户展示满足其偏好的建筑方案，以降低用户查找的时间。
在一种可能的实施方式中，还可以对待展示的推荐建筑方案进行校验，如对日照、建筑楼栋间距或者建筑限高进行校验。具体的校验要求可以参考目标建筑用地所在城市的建筑设计规范数据或者其他建筑设计相关规范来确定。若待展示的推荐建筑方案没有通过校验，则将该推荐建筑方案从待展示列表中删除。
在一种可能的实施方式中，在步骤S203之后，还可以包括以下步骤：
获取所述目标建筑用地所在城市的样本影像数据，所述样本影像数据包括所述目标建筑用地所在城市的建筑物、自然景观以及人文景观关联的影像数据；对所述样本影像数据中所包含的样本标签进行聚类，得到K个标签簇，所述K为正整数；根据所述K个标签簇确定 所述目标建筑用地所在城市的建筑风格；计算所述候选建筑方案的建筑风格与所述目标建筑用地所在城市的建筑风格的相似度；将相似度最高的预设数量的候选建筑方案作为推荐建筑方案进行展示。
建筑风格指建筑设计中在内容和外貌方面所反映的特征，主要在于建筑的平面布局、形态构成、艺术处理和手法运用等方面所显示的独创和完美的意境。常见的建筑风格包括但不限于哥特式建筑风格、巴洛克建筑风格、洛可可建筑风格、木条式建筑风格、园林风格、概念式风格以及现代主义风格。众所周知，建筑是环境的艺术。建筑设计并不只是简单的几何体拼凑，还要考虑场所文脉等多种因素，达到形式与功能的契合，使得建筑在环境中落根，成为环境的一部分。意境一词常用在艺术创作中以体现创作者对于环境的表达。例如，中国传统山水画表现了大自然的意境，其背后的涵义是画家将自然的波澜壮阔的景色在脑海中抽象概括，用绘画语言创作再现。而对于建筑而言，考虑的环境更多的是建筑所坐落于的城市。因此，需要提取一座城市的文化、历史、自然等抽象形式化特征，再将其转译为建筑语言，以得到这座城市的建筑风格。例如，A城市经济较为发达，生活节奏快。那么，A城市所对应的建筑风格现代主义风格，所应答的意境应该是“简约时尚”。换言之，A城市由于较为现代化，更新换代快，对应的建筑风格需要简约不容易过时，经得起时代的考验，同时要具有一定的时尚感和艺术感。综上所述，在考虑容积率等功能性的因素之外，从城市的文化、历史、自然等中提取出一个城市的建筑风格，再将城市的建筑风格融入到建筑生成设计中，可以使得生成的建筑方案更加的合理，也更贴合用户的实际需求。
在本申请实施例中，可以利用机器学习中的聚类算法提取一座城市的文化、历史、自然等抽象形式化特征，以得到这座城市的建筑风格。具体地，电子设备可以获取目标建筑用地所在城市的建筑、自然景观以及人文景观等有关的大量样本影像数据。本申请实施例对于所在城市的建筑类型不做限定，建筑类型可以是住宅建筑、别墅建筑、写字楼建筑、商业建筑、或者一些如学校、博物馆、图书馆、法院、政府办公大楼等公共建筑等。自然景观可以包括山川、河流、湖泊、树木等等。人文景观可以包括园林、亭台楼阁等等。每个样本视频数据或样本照片数据都可以具有已知的标签，该标签可以用于表示样本影像数据中所体现出的建筑风格。可以理解的是，每个样本影像数据所对应的标签根据样本影像数据的内容所确定，每个样本影像数据可以有一个或者多个标签。类似地，一个城市的建筑风格可以是单一的，也可以是多样化的。电子设备可以将所有样本影像数据对应的标签都添加到标签集合中，通过对标签集合中的所有标签进行聚类，得到K个标签簇，其中K为正整数，如K可以取值为1，2，……；对标签集合中的标签进行聚类处理后，在上述K个标签簇中，相似的标签可以被划分到同一个标签簇中，不同的标签可以被划分到不同的标签簇中，然后根据最终得到的标签簇确定该城市的建筑风格。其中，聚类处理后所得到的标签簇的数量K可以是预先设置的固定数值，也可以是在聚类处理过程中不断合并标签集合中的标签后所确定的，本申请实施例对标签簇的数量不做限定。电子设备对标签集合中的标签进行聚类处理所采用的聚类算法可以包括但不限于：k均值聚类算法(k-means)、均值偏移聚类算法(mean shift)、k均值聚类算法(k-means)、模糊C均值聚类算法(fuzzy c-means，FCM)、具有噪声的基于密度的聚类方法(density-based spatial clustering of applications with noise，DBSCAN)、使用高斯混合模型(Gaussian mixture model，GMM)的期望最大化(expectation-maximization，EM)聚类算法、层次聚类算法等等，本申请实施例对所采用的聚类算法的类型不做限定。
在根据最终生成的标签簇确定目标建筑用地所在城市的建筑风格之后，计算候选建筑方案的建筑风格与目标建筑用地所在城市的建筑风格的相似度。其中，相似度可以理解为候选 建筑方案的建筑风格与目标建筑用地所在城市的建筑风格的关联程度，关联程度越高，则相似度越高。候选建筑方案的建筑风格与目标建筑用地所在城市的建筑风格可以理解为自然语音描述的语义标签，因此电子设备在计算相似度之前，可以对候选建筑方案的建筑风格以及目标建筑用地所在城市的建筑风格进行向量转换，分别得到候选建筑方案的建筑风格以及目标建筑用地所在城市的建筑风格所对应的向量。为了方便描述，可以将目标建筑用地所在城市的建筑风格所对应的向量称为中心向量。其中，电子设备可以通过词向量模型进行向量转换，该词向量模型可以包括但不限于：通过词的上下文得到词的向量化表示的词向量模型(word2vec)、一种利用共现矩阵的词向量(glove)、双向注意力神经网络模型(bidirectional encoder representations from transformers，BERT)、潜在语义分析模型(latent semantic analysis，LSA)、潜在狄利克雷分配模型(latent dirichlet allocation，LDA)，本申请实施例对所采用的词向量模型不做限定。
将候选建筑方案的建筑风格以及目标建筑用地所在城市的建筑风格进行向量转换之后，可以计算候选建筑方案的建筑风格的向量与中心向量之间的距离，通过用两个向量之间的距离来衡量两者之间的相似度，然后根据相似度确定匹配值。其中，距离越短，相似度越大，距离越大，相似度越小。计算两个向量之间的距离所采用的方法可以包括但不限于：欧几里得距离(Eucledian distance)、曼哈顿距离(Manhattan distance)、明可夫斯基距离(Minkowski distance)、余弦相似度(cosine similarity)。以余弦相似度为例，对于计算候选建筑方案的建筑风格的向量Gj与中心向量t，其两者之间的距离可以表示为：dis<Gj，t>＝cosin(Gj，t)，其中，dis<Gj，t>用于表示向量Gj与中心向量t之间的距离，cosin为余弦函数，若向量G1和向量G2与中心向量t的距离分别为1和5，说明向量G1与中心向量t的相似度大于向量G2与中心向量t的相似度，那么向量G1对应的候选建筑方案的建筑风格与目标建筑用地所在城市的建筑风格的相似度高于向量G2对应的候选建筑方案的建筑风格与目标建筑用地所在城市的建筑风格的相似度。最后，对相似度进行排序，并按照相似度从高到低，返回相似度最高的预设数量的候选建筑方案作为推荐建筑方案优先向目标用户进行展示。在实际应用中，预设数量可以由目标用户自行设定，也可以由系统预先设定，本申请实施例对此不做出限定。
在一种可能的实施方式中，还可以获取目标建筑用地的开发商的品牌文化，通过对比开发商的品牌文化与候选建筑方案的匹配程度，将匹配程度高的候选建筑方案作为推荐建筑方案优先向目标用户进行展示。例如，若开发商的品牌文化是以舒适度为主打，则可以将绿地率高、容积率低的方案作为推荐建筑方案优先向目标用户进行展示。其匹配程度的计算的具体实现方式可以参考上述建筑风格相似度的计算，在此不再赘述。
可以看出，在生成候选建筑方案之后，通过获取目标建筑用地所在城市的建筑风格，然后计算候选建筑方案的建筑风格与目标建筑用地所在城市的建筑风格的相似度，并将相似度最高的预设数量的候选建筑方案作为推荐建筑方案优先向目标用户进行展示。如此，优先向用户展示建筑风格与所在城市建筑风格匹配的建筑方案，使得向用户展示的建筑方案更加的合理，也更贴合用户的实际需求，从而可以降低用户查找的时间，以达到提高设计效率的目的。
在图2所示的方法中，在根据目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，获取所述目标建筑用地的生成设计信息之后，根据生成设计信息通过预设评级模型对目标建筑用地进行评级，获得目标建筑用地对应的预估评级，然后根据预估评级按照预设生成规则生成针对目标建筑用地的至少一种候选建筑方案。通过预先对目标建筑用 地进行评级，可以更为全面评估目标建筑用地，有助于提高候选建筑方案的准确性。此外，采用建筑生成设计生成建筑方案，相对于传统的计算机辅助设计而言，运用到了计算机强大的计算能力，可以生成多种可能的候选建筑方案，因此不仅可以提高建筑设计的效率，还可以解放建筑师的思维，打破模板式设计的僵硬，激发更多设计的可能性。在生成候选建筑方案之后，对至少一种候选建筑方案进行智能审核，根据设计需求信息生成目标评分维度，并根据目标评分维度对通过审核的候选建筑方案进行评分，将评分靠前的预设数量的候选建筑方案作为推荐建筑方案进行展示。如此，只对通过审核的候选建筑方案评分和展示，可以提高推荐建筑方案的准确性以及合理性。此外，可以优先展示匹配用户设计偏好的建筑方案，使得展示的建筑方案更符合用户的预期，以便降低用户查找的时间，从而提高设计的效率。
上述详细阐述了本申请实施例的方法，下面提供了本申请实施例的装置。
请参照图6，图6是本申请实施例提供的一种建筑方案生成的装置的结构示意图。该装置可以应用于电子设备，例如，图1所对应实施例中的电子设备101，还可以应用于服务器，例如，图1所对应实施例中的服务器102，这里将不对其进行限制。为便于理解，本申请实施例以该装置应用于电子设备101为例进行说明。如图6所示，该建筑方案生成的装置600可以包括获取模块601、评级模块602、生成模块603和展示模块604，各个模块的详细描述如下：
获取模块601，用于根据目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，获取目标建筑用地的生成设计信息；
评级模块602，用于根据生成设计信息通过预设评级模型对目标建筑用地进行评级，获得目标建筑用地对应的预估评级；
生成模块603，用于根据预估评级按照预设生成规则生成针对目标建筑用地的至少一种候选建筑方案，并对至少一种候选建筑方案进行智能审核；
展示模块604，用于根据设计需求信息生成目标评分维度，并根据目标评分维度对通过审核的候选建筑方案进行评分，将评分靠前的预设数量的候选建筑方案作为推荐建筑方案进行展示。
其中，获取模块601、评级模块602、生成模块603和展示模块604的具体功能实现方式可以参见图2所对应实施例中的步骤S201-S204，这里不再进行赘述。
在一种可能的实施方式中，获取模块601可以包括输出单元，划分单元，第一确定单元和第二确定单元，其中：
输出单元，用于获取目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，通过多模态模型输出周边配套信息、环境信息以及设计需求信息对应的多模态特征；多模态特征是对周边配套信息对应的第一特征、环境信息对应的第二特征，以及设计需求信息对应的第三特征进行融合得到的；
划分单元，用于将多模态特征输入至预设地块划分模型，通过预设地块划分模型将目标建筑用地划分为N个功能分区；功能分区包括高层建筑区、别墅区、公共设施区和商业区中的至少一种，N为正整数；
第一确定单元，用于获取每一功能分区的建筑用地面积，根据功能分区的建筑用地面积以及功能分区的功能确定每一功能分区的子设计条件；
第二确定单元，用于根据子设计条件确定目标建筑用地的生成设计信息。
在一种可能的实施方式中，生成模块603可以包括查找单元，第三确定单元和生成单元，其中：
查找单元，用于从预设楼型库查找预估评级对应的目标楼型库；目标楼型库中存储有预估评级对应的参考楼型和参考楼型的参数信息，以及参考楼型的评分值；
第三确定单元，用于将目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向目标用户展示，将目标用户选中的候选楼型作为目标楼型；
生成单元，用于获取目标建筑用地的用地轮廓线，根据目标楼型、预设排布方式以及用地轮廓线生成针对目标建筑用地的至少一种候选建筑方案；预设排布方式包括行列式、周边式、点群式以及极限高低配式。
在一种可能的实施方式中，第三确定单元可以包括确定子单元和拼接子单元，其中：
确定子单元，用于将目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向目标用户展示；
拼接子单元，用于响应于检测到目标用户选中拼合操作，根据拼合操作将目标用户选中的至少两个候选楼型进行拼合，得到至少一种楼型拼合结果，并将楼型拼合结果作为目标楼型。
在一种可能的实施方式中，装置600还可以包括图6中未示出的模块，例如第一获取模块、第二获取模块、第一确定模块以及第二确定模块，其中：
第一获取模块，用于获取建筑用地的类型；建筑用地的类型包括空白建筑用地以及待改建建筑用地；
第二获取模块，用于响应于建筑用地的类型为待改建建筑用地，获取建筑用地中所包含的建筑信息；建筑信息包括建筑类型信息、楼型信息以及建筑年龄信息；
第一确定模块，用于根据建筑信息确定建筑用地中的待改建区域，将待改建区域作为目标建筑用地；
第二确定模块，用于响应于建筑用地的类型为空白建筑用地，将建筑用地作为目标建筑用地。
在一种可能的实施方式中，装置600还可以包括图6中未示出的模块，例如第三获取模块、第四获取模块以及第一生成模块，其中：
第三获取模块，用于获取待改建区域所在城市的城市文化，根据城市文化以及建筑信息将待改建区域划分为L个目标区域；目标区域包括待拆除区域、建筑物保留区域以及空白用地区域中的至少一种，L为正整数；
第四获取模块，用于响应于目标区域为建筑物保留区域，获取建筑物保留区域内所包含的建筑物的功能和布局；
第一生成模块，用于根据建筑物的布局和功能生成针对建筑物保留区域的至少一种候选建筑方案。
在一种可能的实施方式中，装置600还可以包括图6中未示出的模块，例如第五获取模块、聚类模块、第三确定模块、计算模块以及第一展示模块，其中：
第五获取模块，用于获取目标建筑用地所在城市的样本影像数据；样本影像数据包括目标建筑用地所在城市的建筑物、自然景观以及人文景观关联的影像数据；
聚类模块，用于对样本影像数据中所包含的样本标签进行聚类，得到K个标签簇，K为正整数；
第三确定模块，用于根据K个标签簇确定目标建筑用地所在城市的建筑风格；
计算模块，用于计算候选建筑方案的建筑风格与目标建筑用地所在城市的建筑风格的相似度；
第一展示模块，用于将相似度最高的预设数量的候选建筑方案作为推荐建筑方案进行展示。
请参照图7，图7是本申请实施例提供的一种计算机设备的结构示意图。如图7所示，该计算机设备700可以为电子设备，例如，图1所对应实施例中的电子设备101，还可以为服务器，例如，图1所对应实施例中的服务器102，这里将不对其进行限制。为便于理解，本申请实施例以计算机设备为电子设备101为例，该计算机设备700可以包括处理器701、存储器702和通信接口703，其中存储器702存储有计算机程序704。处理器701、存储器702、通信接口703以及计算机程序704之间可以通过总线705连接。
当计算机设备为电子设备时，上述计算机程序704用于执行以下步骤的指令：
根据目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，获取目标建筑用地的生成设计信息；
根据生成设计信息通过预设评级模型对目标建筑用地进行评级，获得目标建筑用地对应的预估评级；
根据预估评级按照预设生成规则生成针对目标建筑用地的至少一种候选建筑方案，并对至少一种候选建筑方案进行智能审核；
根据设计需求信息生成目标评分维度，并根据目标评分维度对通过审核的候选建筑方案进行评分，将评分靠前的预设数量的候选建筑方案作为推荐建筑方案进行展示。
在一种可能的实施方式中，在根据目标建筑用地的周边配套信息和环境信息，以及目标用户针对目标建筑用地的设计需求信息，获取目标建筑用地的生成设计信息方面，计算机程序704具体用于执行以下步骤的指令：
获取目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，通过多模态模型输出周边配套信息、环境信息以及设计需求信息对应的多模态特征，多模态特征是对周边配套信息对应的第一特征、环境信息对应的第二特征，以及设计需求信息对应的第三特征进行融合得到的；
将多模态特征输入至预设地块划分模型，通过预设地块划分模型将目标建筑用地划分为N个功能分区，功能分区包括高层建筑区、别墅区、公共设施区和商业区中的至少一种，N为正整数；
获取每一功能分区的建筑用地面积，根据功能分区的建筑用地面积以及功能分区的功能确定每一功能分区的子设计条件；
根据子设计条件确定目标建筑用地的生成设计信息。
在一种可能的实施方式中，在根据预估评级按照预设生成规则生成针对目标建筑用地的至少一种候选建筑方案方面，计算机程序704具体于执行以下步骤的指令：
从预设楼型库查找预估评级对应的目标楼型库，目标楼型库中存储有预估评级对应的参考楼型和参考楼型的参数信息，以及参考楼型的评分值；
将目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向目标用户展示，将目标用户选中的候选楼型作为目标楼型；
获取目标建筑用地的用地轮廓线，根据目标楼型、预设排布方式以及用地轮廓线生成针对目标建筑用地的至少一种候选建筑方案，预设排布方式包括行列式、周边式、点群式以及极限高低配式。
在一种可能的实施方式中，在将目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向目标用户展示，将目标用户选中的候选楼型作为目标楼型方面，计算机程序704具体 用于执行以下步骤的指令：
将目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向目标用户展示；
响应于检测到目标用户选中拼合操作，根据拼合操作将目标用户选中的至少两个候选楼型进行拼合，得到至少一种楼型拼合结果，并将楼型拼合结果作为目标楼型。
在一种可能的实施方式中，在根据目标建筑用地的周边配套信息和环境信息，获取目标建筑用地的生成设计信息之前，计算机程序704还用于执行以下步骤的指令：
获取建筑用地的类型，建筑用地的类型包括空白建筑用地以及待改建建筑用地；
响应于建筑用地的类型为待改建建筑用地，获取建筑用地中所包含的建筑信息，建筑信息包括建筑类型信息、楼型信息以及建筑年龄信息；
根据建筑信息确定建筑用地中的待改建区域，将待改建区域作为目标建筑用地；
响应于建筑用地的类型为空白建筑用地，将建筑用地作为目标建筑用地。
在一种可能的实施方式中，计算机程序704还用于执行以下步骤的指令：
获取待改建区域所在城市的城市文化，根据城市文化以及建筑信息将待改建区域划分为L个目标区域中的至少一种，目标区域包括待拆除区域、建筑物保留区域以及空白用地区域，L为正整数；
响应于目标区域为建筑物保留区域，获取建筑物保留区域内所包含的建筑物的功能和布局；
根据建筑物的布局和功能生成针对建筑物保留区域的至少一种候选建筑方案。
在一种可能的实施方式中，在根据预估评级按照预设生成规则生成针对目标建筑用地的至少一种候选建筑方案之后，计算机程序704还用于执行以下步骤的指令：
获取目标建筑用地所在城市的样本影像数据，样本影像数据包括目标建筑用地所在城市的建筑物、自然景观以及人文景观关联的影像数据；
对样本影像数据中所包含的样本标签进行聚类，得到K个标签簇，K为正整数；
根据K个标签簇确定目标建筑用地所在城市的建筑风格；
计算候选建筑方案的建筑风格与目标建筑用地所在城市的建筑风格的相似度；
将相似度最高的预设数量的候选建筑方案作为推荐建筑方案进行展示。
本领域技术人员可以理解，为了便于说明，图7中仅示出了一个存储器和处理器。在实际的终端或服务器中，可以存在多个处理器和存储器。存储器702也可以称为存储介质或者存储设备等，本申请实施例对此不做限定。
应理解，在本申请实施例中，处理器701可以是中央处理单元(central processing unit，CPU)，该处理器还可以是其他通用处理器、数字信号处理器(digital signal processing，DSP)、专用集成电路(application specific integrated circuit，ASIC)、现成可编程门阵列(field-programmable gate array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。
还应理解，本申请实施例中提及的存储器702可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。通过示例性但不是限制性说明，许多形式的RAM可用，例如静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(dynamic RAM，DRAM)、同步动态随机存 取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器synchronize link DRAM，SLDRAM)和直接内存总线随机存取存储器(direct rambus RAM，DR RAM)。
需要说明的是，当处理器701为通用处理器、DSP、ASIC、FPGA或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件时，存储器(存储模块)集成在处理器中。
应注意，本文描述的存储器702旨在包括但不限于这些和任意其它适合类型的存储器。
该总线705除包括数据总线之外，还可以包括电源总线、控制总线和状态信号总线等。但是为了清楚说明起见，在图中将各种总线都标为总线。
在实现过程中，上述方法的各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。结合本申请实施例所公开的方法的步骤可以直接体现为硬件处理器执行完成，或者用处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器，处理器读取存储器中的信息，结合其硬件完成上述方法的步骤。为避免重复，这里不再详细描述。
在本申请的各种实施例中，上述各过程的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各种说明性逻辑块(illustrative logical block，ILB)和步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线)或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进 行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质，(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质(例如固态硬盘)等。
本申请实施例还提供一种计算机存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行以实现如上述方法实施例中记载的任何一种建筑方案生成的方法的部分或全部步骤。
本申请实施例还提供一种计算机程序产品，所述计算机程序产品包括存储了计算机程序的非瞬时性计算机可读存储介质，所述计算机程序可操作来使计算机执行如上述方法实施例中记载的任何一种建筑方案生成的方法的部分或全部步骤。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以权利要求的保护范围为准。

Claims (16)

  1. 一种建筑方案生成的方法，其特征在于，包括：
    根据目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，获取所述目标建筑用地的生成设计信息；
    根据所述生成设计信息通过预设评级模型对所述目标建筑用地进行评级，获得所述目标建筑用地对应的预估评级；
    根据所述预估评级按照预设生成规则生成针对所述目标建筑用地的至少一种候选建筑方案，并对所述至少一种候选建筑方案进行智能审核；
    根据所述设计需求信息生成目标评分维度，并根据所述目标评分维度对通过审核的候选建筑方案进行评分，将评分靠前的预设数量的候选建筑方案作为推荐建筑方案进行展示。
  2. 根据权利要求1所述的方法，其特征在于，所述根据目标建筑用地的周边配套信息和环境信息，以及目标用户针对所述目标建筑用地的设计需求信息，获取所述目标建筑用地的生成设计信息，包括：
    获取目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，通过多模态模型输出所述周边配套信息、所述环境信息以及所述设计需求信息对应的多模态特征；所述多模态特征是对所述周边配套信息对应的第一特征、所述环境信息对应的第二特征，以及所述设计需求信息对应的第三特征进行融合得到的；
    将所述多模态特征输入至预设地块划分模型，通过所述预设地块划分模型将所述目标建筑用地划分为N个功能分区；所述功能分区包括高层建筑区、别墅区、公共设施区和商业区中的至少一种，所述N为正整数；
    获取每一所述功能分区的建筑用地面积，根据所述功能分区的建筑用地面积以及所述功能分区的功能确定每一所述功能分区的子设计条件；
    根据所述子设计条件确定所述目标建筑用地的生成设计信息。
  3. 根据权利要求1或2所述的方法，其特征在于，所述根据所述预估评级按照预设生成规则生成针对所述目标建筑用地的至少一种候选建筑方案，包括：
    从预设楼型库查找所述预估评级对应的目标楼型库；所述目标楼型库中存储有所述预估评级对应的参考楼型和所述参考楼型的参数信息，以及所述参考楼型的评分值；
    将所述目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向所述目标用户展示，将所述目标用户选中的候选楼型作为目标楼型；
    获取所述目标建筑用地的用地轮廓线，根据所述目标楼型、预设排布方式以及所述用地轮廓线生成针对所述目标建筑用地的至少一种候选建筑方案；所述预设排布方式包括行列式、周边式、点群式以及极限高低配式。
  4. 根据权利要求3所述的方法，其特征在于，所述将所述目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向目标用户展示，将所述目标用户选中的候选楼型作为目标楼型，包括：
    将所述目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向目标用户展示；
    响应于检测到所述目标用户选中拼合操作，根据所述拼合操作将所述目标用户选中的至 少两个候选楼型进行拼合，得到至少一种楼型拼合结果，并将所述楼型拼合结果作为目标楼型。
  5. 根据权利要求1所述的方法，其特征在于，在所述根据目标建筑用地的周边配套信息和环境信息，获取所述目标建筑用地的生成设计信息之前，所述方法还包括：
    获取建筑用地的类型；所述建筑用地的类型包括空白建筑用地以及待改建建筑用地；
    响应于所述建筑用地的类型为所述待改建建筑用地，获取所述建筑用地中所包含的建筑信息；所述建筑信息包括建筑类型信息、楼型信息以及建筑年龄信息；
    根据所述建筑信息确定所述建筑用地中的待改建区域，将所述待改建区域作为所述目标建筑用地；
    响应于所述建筑用地的类型为所述空白建筑用地，将所述建筑用地作为所述目标建筑用地。
  6. 根据权利要求5所述的方法，其特征在于，所述方法还包括：
    获取所述待改建区域所在城市的城市文化，根据所述城市文化以及所述建筑信息将所述待改建区域划分为L个目标区域；所述目标区域包括待拆除区域、建筑物保留区域以及空白用地区域中的至少一种，所述L为正整数；
    响应于所述目标区域为建筑物保留区域，获取所述建筑物保留区域内所包含的建筑物的功能和布局；
    根据所述建筑物的布局和功能生成针对所述建筑物保留区域的至少一种候选建筑方案。
  7. 根据权利要求1所述的方法，其特征在于，在所述根据所述预估评级按照预设生成规则生成针对所述目标建筑用地的至少一种候选建筑方案之后，所述方法还包括：
    获取所述目标建筑用地所在城市的样本影像数据；所述样本影像数据包括所述目标建筑用地所在城市的建筑物、自然景观以及人文景观关联的影像数据；
    对所述样本影像数据中所包含的样本标签进行聚类，得到K个标签簇；所述K为正整数；
    根据所述K个标签簇确定所述目标建筑用地所在城市的建筑风格；
    计算所述候选建筑方案的建筑风格与所述目标建筑用地所在城市的建筑风格的相似度；
    将相似度最高的预设数量的候选建筑方案作为推荐建筑方案进行展示。
  8. 一种建筑方案生成的装置，其特征在于，包括：
    获取模块，用于根据目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，获取所述目标建筑用地的生成设计信息；
    评级模块，用于根据所述生成设计信息通过预设评级模型对所述目标建筑用地进行评级，获得所述目标建筑用地对应的预估评级；
    生成模块，用于根据所述预估评级按照预设生成规则生成针对所述目标建筑用地的至少一种候选建筑方案，并对所述至少一种候选建筑方案进行智能审核；
    展示模块，用于根据所述设计需求信息生成目标评分维度，并根据所述目标评分维度对通过审核的候选建筑方案进行评分，将评分靠前的预设数量的候选建筑方案作为推荐建筑方案进行展示。
  9. 根据权利要求8所述的装置，其特征在于，所述获取模块包括：
    输出单元，用于获取目标建筑用地的周边配套信息和环境信息，以及目标用户的设计需求信息，通过多模态模型输出所述周边配套信息、所述环境信息以及所述设计需求信息对应的多模态特征；所述多模态特征是对所述周边配套信息对应的第一特征、所述环境信息对应的第二特征，以及所述设计需求信息对应的第三特征进行融合得到的；
    划分单元，用于将所述多模态特征输入至预设地块划分模型，通过所述预设地块划分模型将所述目标建筑用地划分为N个功能分区；所述功能分区包括高层建筑区、别墅区、公共设施区和商业区中的至少一种，所述N为正整数；
    第一确定单元，用于获取每一所述功能分区的建筑用地面积，根据所述功能分区的建筑用地面积以及所述功能分区的功能确定每一所述功能分区的子设计条件；
    第二确定单元，用于根据所述子设计条件确定所述目标建筑用地的生成设计信息。
  10. 根据权利要求8或9所述的装置，其特征在于，所述生成模块包括：
    查找单元，用于从预设楼型库查找所述预估评级对应的目标楼型库；所述目标楼型库中存储有所述预估评级对应的参考楼型和所述参考楼型的参数信息，以及所述参考楼型的评分值；
    第三确定单元，用于将所述目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向所述目标用户展示，将所述目标用户选中的候选楼型作为目标楼型；
    生成单元，用于获取所述目标建筑用地的用地轮廓线，根据所述目标楼型、预设排布方式以及所述用地轮廓线生成针对所述目标建筑用地的至少一种候选建筑方案；所述预设排布方式包括行列式、周边式、点群式以及极限高低配式。
  11. 根据权利要求9所述的装置，其特征在于，所述第三确定单元包括：
    确定子单元，用于将所述目标楼型库中评分靠前的预设数量的参考楼型作为候选楼型向目标用户展示；
    拼接子单元，用于响应于检测到所述目标用户选中拼合操作，根据所述拼合操作将所述目标用户选中的至少两个候选楼型进行拼合，得到至少一种楼型拼合结果，并将所述楼型拼合结果作为目标楼型。
  12. 根据权利要求8所述的装置，其特征在于，所述装置还包括：
    第一获取模块，用于获取建筑用地的类型；所述建筑用地的类型包括空白建筑用地以及待改建建筑用地；
    第二获取模块，用于响应于所述建筑用地的类型为所述待改建建筑用地，获取所述建筑用地中所包含的建筑信息；所述建筑信息包括建筑类型信息、楼型信息以及建筑年龄信息；
    第一确定模块，用于根据所述建筑信息确定所述建筑用地中的待改建区域，将所述待改建区域作为所述目标建筑用地；
    第二确定模块，用于响应于所述建筑用地的类型为所述空白建筑用地，将所述建筑用地作为所述目标建筑用地。
  13. 根据权利要求12所述的装置，其特征在于，所述装置还包括：
    第三获取模块，用于获取所述待改建区域所在城市的城市文化，根据所述城市文化以及 所述建筑信息将所述待改建区域划分为L个目标区域；所述目标区域包括待拆除区域、建筑物保留区域以及空白用地区域中的至少一种，所述L为正整数；
    第四获取模块，用于响应于所述目标区域为建筑物保留区域，获取所述建筑物保留区域内所包含的建筑物的功能和布局；
    第一生成模块，用于根据所述建筑物的布局和功能生成针对所述建筑物保留区域的至少一种候选建筑方案。
  14. 根据权利要求8所述的装置，其特征在于，所述装置还包括：
    第五获取模块，用于获取所述目标建筑用地所在城市的样本影像数据；所述样本影像数据包括所述目标建筑用地所在城市的建筑物、自然景观以及人文景观关联的影像数据；
    聚类模块，用于对所述样本影像数据中所包含的样本标签进行聚类，得到K个标签簇，所述K为正整数；
    第三确定模块，用于根据所述K个标签簇确定所述目标建筑用地所在城市的建筑风格；
    计算模块，用于计算所述候选建筑方案的建筑风格与所述目标建筑用地所在城市的建筑风格的相似度；
    第一展示模块，用于将相似度最高的预设数量的候选建筑方案作为推荐建筑方案进行展示。
  15. 一种计算机设备，其特征在于，包括处理器、存储器和通信接口，其中，所述存储器存储有计算机程序，所述计算机程序被配置由所述处理器执行，所述计算机程序包括用于执行权利要求1-7中任一项所述的方法中的步骤的指令。
  16. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储计算机程序，所述计算机程序使得计算机执行以实现权利要求1-7中任一项所述的方法。
PCT/CN2023/081435 2022-03-31 2023-03-14 建筑方案生成的方法、装置、计算机设备以及存储介质 WO2023185458A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210327987.2A CN114491773B (zh) 2022-03-31 2022-03-31 建筑方案生成的方法、装置、计算机设备以及存储介质
CN202210327987.2 2022-03-31

Publications (1)

Publication Number Publication Date
WO2023185458A1 true WO2023185458A1 (zh) 2023-10-05

Family

ID=81487759

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/081435 WO2023185458A1 (zh) 2022-03-31 2023-03-14 建筑方案生成的方法、装置、计算机设备以及存储介质

Country Status (2)

Country Link
CN (1) CN114491773B (zh)
WO (1) WO2023185458A1 (zh)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117113515A (zh) * 2023-10-23 2023-11-24 湖南大学 一种路面设计方法、装置、设备及存储介质
CN117314198A (zh) * 2023-10-25 2023-12-29 北京华清安地建筑设计有限公司 历史文化街区功能更新综合分析方法及系统
CN117391343A (zh) * 2023-10-08 2024-01-12 江苏科能电力工程咨询有限公司 一种基于城市规划的用电预测方法、装置和存储介质
CN117725661A (zh) * 2024-02-07 2024-03-19 中建科工集团绿色科技有限公司 模块化建筑的模块单元标准化方法、装置、设备及介质
CN117744233A (zh) * 2024-02-21 2024-03-22 沈阳航空工业发展有限公司 一种基于数据分析的建筑设计专家智能分配系统及方法
CN117828705A (zh) * 2024-01-11 2024-04-05 北京建筑大学 基于遗传算法搜索的住宅布局生成设计方法和系统

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114491773B (zh) * 2022-03-31 2022-07-05 深圳小库科技有限公司 建筑方案生成的方法、装置、计算机设备以及存储介质
CN114943112B (zh) * 2022-07-20 2022-11-11 深圳小库科技有限公司 建筑沿线排布方案的自动生成方法、装置、设备及介质
CN115880479A (zh) * 2023-02-01 2023-03-31 北京有竹居网络技术有限公司 改造方案的生成方法、生成装置、电子设备和存储介质
CN116051059B (zh) * 2023-03-31 2023-07-28 佛山市龙生光启科技有限公司 一种基于大数据分析的城市规划方案智能管理系统
CN116502770B (zh) * 2023-06-13 2024-04-02 佛山市银锠精密五金有限公司 生成太空舱装配方案方法、装置、设备及可读存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111401726A (zh) * 2020-03-12 2020-07-10 深圳小库科技有限公司 一种建筑设计方法、装置、系统及存储介质
CN113158312A (zh) * 2021-04-26 2021-07-23 深圳小库科技有限公司 建设用地的建筑方案生成方法、装置、存储介质及处理器
CN113515708A (zh) * 2021-04-26 2021-10-19 深圳小库科技有限公司 针对建设用地的项目推荐方法、装置、存储介质及处理器
US20220035962A1 (en) * 2021-10-14 2022-02-03 Shenzhen Xkool Technology Co., Ltd. Building design method, device, system and storage medium
CN114491773A (zh) * 2022-03-31 2022-05-13 深圳小库科技有限公司 建筑方案生成的方法、装置、计算机设备以及存储介质

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109635511B (zh) * 2019-01-16 2020-08-07 哈尔滨工业大学 一种基于条件生成对抗网络的高层居住区强排方案生成设计方法
CN110110353A (zh) * 2019-03-20 2019-08-09 广西大学 面向绿色住区设计方案初期的参数化生成优化设计方法
WO2021009389A1 (es) * 2019-07-17 2021-01-21 Smartscapes Studio, S.L. Método y sistema de generación de soluciones de diseño asistido por inteligencia artificial y método y sistema de entrenamiento del mismo
CN111415035A (zh) * 2020-03-12 2020-07-14 深圳小库科技有限公司 一种预估建筑功能户型比的方法及装置
CN112651059B (zh) * 2020-12-18 2022-11-08 东南大学 一种基于人工智能的控规地块城市设计多方案生成方法
CN113032890B (zh) * 2021-05-31 2021-08-24 北京盈建科软件股份有限公司 建筑模型的生成方法及装置
CN113761622B (zh) * 2021-08-27 2022-08-19 广州市城市规划勘测设计研究院 一种参数化城市建模方法、装置、存储介质及终端设备
CN114004006B (zh) * 2022-01-04 2022-04-19 深圳小库科技有限公司 建筑群排布方法、装置、设备及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111401726A (zh) * 2020-03-12 2020-07-10 深圳小库科技有限公司 一种建筑设计方法、装置、系统及存储介质
CN113158312A (zh) * 2021-04-26 2021-07-23 深圳小库科技有限公司 建设用地的建筑方案生成方法、装置、存储介质及处理器
CN113515708A (zh) * 2021-04-26 2021-10-19 深圳小库科技有限公司 针对建设用地的项目推荐方法、装置、存储介质及处理器
US20220035962A1 (en) * 2021-10-14 2022-02-03 Shenzhen Xkool Technology Co., Ltd. Building design method, device, system and storage medium
CN114491773A (zh) * 2022-03-31 2022-05-13 深圳小库科技有限公司 建筑方案生成的方法、装置、计算机设备以及存储介质

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117391343A (zh) * 2023-10-08 2024-01-12 江苏科能电力工程咨询有限公司 一种基于城市规划的用电预测方法、装置和存储介质
CN117391343B (zh) * 2023-10-08 2024-03-12 江苏科能电力工程咨询有限公司 一种基于城市规划的用电预测方法、装置和存储介质
CN117113515A (zh) * 2023-10-23 2023-11-24 湖南大学 一种路面设计方法、装置、设备及存储介质
CN117113515B (zh) * 2023-10-23 2024-01-05 湖南大学 一种路面设计方法、装置、设备及存储介质
CN117314198A (zh) * 2023-10-25 2023-12-29 北京华清安地建筑设计有限公司 历史文化街区功能更新综合分析方法及系统
CN117314198B (zh) * 2023-10-25 2024-03-05 北京华清安地建筑设计有限公司 历史文化街区功能更新综合分析方法及系统
CN117828705A (zh) * 2024-01-11 2024-04-05 北京建筑大学 基于遗传算法搜索的住宅布局生成设计方法和系统
CN117725661A (zh) * 2024-02-07 2024-03-19 中建科工集团绿色科技有限公司 模块化建筑的模块单元标准化方法、装置、设备及介质
CN117725661B (zh) * 2024-02-07 2024-05-07 中建科工集团绿色科技有限公司 模块化建筑的模块单元标准化方法、装置、设备及介质
CN117744233A (zh) * 2024-02-21 2024-03-22 沈阳航空工业发展有限公司 一种基于数据分析的建筑设计专家智能分配系统及方法

Also Published As

Publication number Publication date
CN114491773A (zh) 2022-05-13
CN114491773B (zh) 2022-07-05

Similar Documents

Publication Publication Date Title
WO2023185458A1 (zh) 建筑方案生成的方法、装置、计算机设备以及存储介质
US11468206B2 (en) Machine learning system for building renderings and building information modeling data
US20140278280A1 (en) System and method for designing buildings
JP6602774B2 (ja) ジオフェンスのプロビジョニングのための方法および装置
US20110173572A1 (en) Method and interface for displaying locations associated with annotations
CN106462624A (zh) 基于图块的地理编码器
WO2021203728A1 (zh) 一种展业区域的选址方法、装置、计算机设备和介质
WO2020098252A1 (zh) 一种路径规划方法、服务器及计算机存储介质
KR102164538B1 (ko) 부동산 개발에 관한 정보를 제공하는 방법, 시스템 및 비일시성의 컴퓨터 판독 가능 기록 매체
US20220130004A1 (en) Interface for uncompleted homes planning
US20200043110A1 (en) Land acquisition and property development analysis platform
CN117015771A (zh) 通过开发和呈现摄影测量现实网格来处理和显示与房产相关的信息的系统和方法
US11893073B2 (en) Method and apparatus for displaying map points of interest, and electronic device
CN111881377A (zh) 位置兴趣点的处理方法及装置
CN110309433A (zh) 一种数据处理方法、装置及服务器
CN114332439A (zh) 一种三维数据编辑生成系统
KR102204458B1 (ko) 인테리어 설계 및 비용 시뮬레이션 방법 및 장치
JP2001338037A (ja) 三次元コンピュータグラフィック画像生成用サーバ
Sun et al. Automatic building age prediction from street view images
WO2021081767A1 (zh) 位置点确定方法、装置、电子设备及计算机可读介质
Haghbayan et al. Visual description of the indoor space of Real Estate in crowd-sourcing environments
JP7304925B2 (ja) 情報処理装置、情報処理方法及び情報処理プログラム
EP4328868A1 (en) Automated generation and use of building information from analysis of floor plans and acquired building images
CN114996586B (zh) 一种基于时空大数据的个性化出行路径生成方法及系统
US20240111792A1 (en) Computer Systems and Methods for Identifying Location Entities and Generating a Location Entity Data Taxonomy

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23777833

Country of ref document: EP

Kind code of ref document: A1