Urban Design Generation Auxiliary Methods and Devices

Abstract

The application provides a kind of urban design generation auxiliary method and device, including the following steps: step S1, the existing design scheme generation original image-planning image pair of specified designer is collected;Step S2, the multiple urban design form parameters of regional planning image are calculated, and the classification result is obtained by classifying regional original image;Step S3, clustering is carried out according to urban design form parameters and classification result, and multiple design style preferences are obtained;Step S4, the training data set of each design style preference is constructed;Step S5, style design model is obtained by training according to training data set;Step S6, the optimal design style preference is selected;Step S7, the image of the region to be planned is input into the style design model corresponding to the optimal design style preference to obtain the initial design drawing;Step S8, auxiliary design drawing is obtained by multi-objective optimization on the initial design drawing.In short, the method can generate regional planning design drawing with specified designer style preference.

Description

Technical Field

[0001] This invention relates to a method for urban layout design, specifically to a method and apparatus for generating urban design. Background Technology

[0002] In traditional street design, designers create artificial scenarios based on imagined future conditions. With the development of digital modeling software, there is increasing research on using models to create urban blocks that meet designers' expectations. Existing research has thoroughly explored data analysis methods used at different design stages to solve specific problems. These include classic quantitative analysis models, big data analysis methods, and diagnostic methods that leverage AI to uncover urban development patterns in recent years. Other methods include system dynamics models, cellular automata models, agent models, game theory deduction, and predictive models—predictive methods that derive future descriptions based on objective conditions. Parametric design methods and AI-powered design generation software tools address the problem of generating design solutions by creating planning and design schemes based on knowledge.

[0003] However, the application of diverse auxiliary tools has actually increased the complexity of the design process. Designers often need to spend hours setting up, running, and testing models to complete an evaluation, which negatively impacts design efficiency. In traditional computer-aided design, designers exhibit two tendencies: one is to translate design intent into specific generation conditions and parameters, and then generate them using computer software. This strategy separates intent from the design process and is often called "pre-setting"; the other tendency is to perform multi-dimensional evaluations using software after the design is completed. This strategy separates the design process from performance optimization and is often called "post-evaluation." It has been proven that both paradigms can lead to insufficient design optimization, resulting in only minimal benefits and significant costs.

[0004] Consequently, research has begun to focus on strategies for design interaction. On the one hand, interactivity has been improved during the design process. Some researchers have established visual interfaces between generative tools and 3D modeling platforms; these software programs and plugins make it easier for generative tools to assist in the design process. On the other hand, the time-consuming nature of simulation engines hinders rapid iteration in the design process, so scholars have begun to study techniques to accelerate the generation or evaluation of design solutions. A typical approach is to simplify the underlying algorithms to accelerate the model.

[0005] However, the basic principles of the above methods are similar to those of mainstream methods, the improvement effect is very limited, and it is still difficult to obtain real-time feedback during the design process. Designers cannot directly use these tools to effectively create ideal designs. Summary of the Invention

[0006] This invention is made to solve the above-mentioned problems, and its purpose is to provide a method and apparatus for generating urban design.

[0007] This invention provides an auxiliary method for generating urban design, used to generate auxiliary design drawings with a specified designer's style for a planned area. The method includes the following steps: Step S1, collecting regional planning images of existing design schemes by a specified designer and corresponding original images of unplanned areas to generate multiple original image-planned image pairs; Step S2, calculating multiple urban design morphological parameters of the regional planning image for each original image-planned image pair, and classifying the original regional images to obtain classification results; Step S3, clustering the corresponding urban design morphological parameters and classification results based on each original image-planned image to obtain multiple classes as design style preferences; Step S4, using all corresponding original image-planned image pairs as the corresponding training dataset for each design style preference; Step S5, constructing an initial model for each design style preference and training it based on the corresponding training dataset to obtain the corresponding style design model; Step S6, selecting the corresponding design style preference as the optimal design style preference based on the planning design parameters of the planned area; Step S7, inputting the image of the planned area into the style design model corresponding to the optimal design style preference to obtain the initial design drawing; Step S8, performing multi-objective optimization on the initial design drawing to obtain the auxiliary design drawing.

[0008] The urban design generation assistance method provided by this invention may also have the following features: the urban design morphological parameters include building density, plot ratio, average height, building height variance, building configuration variance, green space ratio, water surface ratio, and hard plaza ratio.

[0009] The urban design generation assistance method provided by this invention may also have the following features: the classification results include unbuilt and built types; when the proportion of construction in the original image of a region is less than a preset value, the original image of the region is unbuilt; when the proportion of construction in the original image of a region is greater than a preset value, the original image of the region is built.

[0010] The urban design generation assistance method provided by this invention may also have the following features: In step S5, generating a corresponding style design model for each design style preference includes the following steps: Step T1, constructing a generator and discriminator as a generative adversarial network; Step T2, training the generative adversarial network based on the training dataset to obtain a trained generator; Step T3, inputting the original images of each region in the training dataset into the trained generator to obtain corresponding predicted planning images; Step T4, calculating the similarity between each predicted planning image and the corresponding regional planning image to obtain a similarity value; Step T5, determining whether each similarity value is greater than or equal to a preset similarity value threshold. If so, the trained generator is used as the style design model; otherwise, the original images of regions with similarity values ​​less than the similarity value threshold are removed from the training dataset, and step T2 is executed.

[0011] The urban design generation assistance method provided by the present invention may also have the following feature: in step S6, the planning and design parameters are compared with the urban design form parameters and classification results corresponding to each design style preference, and the design style preference with the highest similarity is selected as the optimal design style preference.

[0012] The urban design generation assistance method provided by the present invention may also have the following feature: in step S8, multi-objective optimization is to adjust the building combination in the initial design drawing until the parameters of the initial design drawing meet the planning design parameters, and then an auxiliary design drawing is obtained.

[0013] This invention also provides an urban design generation auxiliary device for generating auxiliary design drawings with the style of a specified designer for a planned area. The device comprises: a data extraction module for collecting regional planning images of existing design schemes by a specified designer and corresponding original images of unplanned areas, generating multiple original image-planned image pairs; a feature calculation module for calculating multiple urban design morphological parameters of the regional planning image for each original image-planned image pair, and classifying the regional original images to obtain classification results; and a classification module for clustering the corresponding urban design morphological parameters and classification results based on each original image-planned image, obtaining multiple classes as design parameters. The system comprises the following modules: a style preference calculation module; a training set construction module, which uses all original image-planning image pairs as the corresponding training dataset for each design style preference; a model construction module, which constructs the corresponding initial model for each design style preference and trains it based on the corresponding training dataset to obtain the corresponding style design model; a style selection module, which selects the corresponding design style preference as the optimal design style preference based on the planning design parameters of the area to be planned; an initial map generation module, which inputs the image of the area to be planned into the style design model corresponding to the optimal design style preference to obtain the initial design map; and an optimization module, which performs multi-objective optimization on the initial design map to obtain the auxiliary design map.

[0014] The role and effect of invention

[0015] According to the urban design generation assistance method and apparatus of the present invention, firstly, the design style preferences of existing designers' works are identified through a data extraction module, a feature calculation module, and a classification module; secondly, a corresponding style design model is trained based on each design style preference through a training set construction module and a model construction module; nextly, a suitable design style preference is selected according to the design requirements of the area to be planned through a style selection module and an initial map generation module, generating an initial design map containing that design style preference; finally, the initial design map is optimized and adjusted through an optimization module to obtain an auxiliary design map that better meets the design requirements. Therefore, the urban design generation assistance method and apparatus of the present invention can generate regional planning design maps with specified designer style preferences, providing a reference for regional planning and design. Attached Figure Description

[0016] Figure 1 This is a block diagram of the city design generation auxiliary device in an embodiment of the present invention;

[0017] Figure 2 This is a schematic diagram of the process for generating a style design model corresponding to design style preferences in an embodiment of the present invention;

[0018] Figure 3This is a comparative schematic diagram of multi-objective optimization before and after in an embodiment of the present invention;

[0019] Figure 4 This is a flowchart illustrating the urban design generation assistance method in an embodiment of the present invention. Detailed Implementation

[0020] To make the technical means, creative features, objectives and effects of the present invention easy to understand, the following embodiments, in conjunction with the accompanying drawings, will specifically illustrate the urban design generation auxiliary method and apparatus of the present invention.

[0021] This embodiment provides an urban design generation auxiliary device for generating auxiliary design drawings with a specified designer style for a planned area.

[0022] Figure 1 This is a block diagram of a city design generation auxiliary device in an embodiment of the present invention.

[0023] like Figure 1 As shown, the urban design generation auxiliary device 100 includes a data extraction module 10, a feature calculation module 20, a classification module 30, a training set construction module 40, a model construction module 50, a style selection module 60, an initial map generation module 70, an optimization module 80, and a control module 90 that controls the operation of the above modules.

[0024] Data extraction module 10 is used to collect regional planning images of existing design schemes by a specified designer and corresponding original images of unplanned areas, generating multiple original image-planning image pairs. In this embodiment, the design scope of the existing design scheme ranges from 1km. 2 up to 200km 2 The two images are not identical. The regional planning image is a plan view of a fixed-size portion of the area in the design scheme, while the original regional image is a vector map obtained by surveying the site before planning, and then processed according to a specific format. The fixed size is 800 meters by 800 meters.

[0025] The feature calculation module 20 is used to calculate multiple urban design morphological parameters of the regional planning image for each original image-planned image pair, and to classify the original regional images to obtain classification results.

[0026] Among them, urban design morphological parameters include building density, plot ratio, average height, building height variance, building configuration variance, green space ratio, water surface ratio, and hard plaza ratio.

[0027] Building density is used to quantify the degree of building coverage on a site, and its calculation expression is as follows:

[0028]

[0029] Floor area ratio (FAR) is used to quantify the intensity of construction on a site, and its calculation formula is as follows:

[0030]

[0031] For a single building, the corresponding building area is the sum of the areas of all floors of that building.

[0032] Average height is used to represent the overall building height within a site, and its calculation formula is as follows:

[0033]

[0034] Building height variance is used to quantify the richness of variations in building heights within a site. Its calculation expression is as follows:

[0035]

[0036] Building configuration variance is used to quantify the richness of shape variations among buildings on a site. Its calculation expression is as follows:

[0037]

[0038] The green space ratio is used to quantify the proportion of green space in a site, and its calculation formula is as follows:

[0039]

[0040] The water surface ratio is used to quantify the proportion of water surface in a site, and its calculation formula is as follows:

[0041]

[0042] The hard plaza ratio is used to quantify the proportion of hard plazas in a site, and its calculation formula is as follows:

[0043]

[0044] In this embodiment, the urban design morphological parameters of the regional planning image include not only the building density, floor area ratio, average height, building height variance, building configuration variance, green space ratio, water surface ratio, and hard plaza ratio of the regional planning image, but also the average values ​​of these parameters for the surrounding plots. By using the average values ​​of the urban design morphological parameters from the surrounding regional planning images, the characteristics of the regional planning image are further supplemented and improved.

[0045] In this embodiment, urban design morphology parameters are used to reflect the roads, water bodies, building combinations and forms within the region, thereby parametrically representing the design style of the planning image.

[0046] The classification results include "undeveloped" and "developed" types, used to distinguish the initial development status. If the proportion of developed areas in the original image of a region is less than a preset value, the original image of that region is considered "undeveloped." If the proportion of developed areas in the original image of a region is greater than a preset value, the original image of that region is considered "developed."

[0047] The classification module 30 is used to cluster the corresponding urban design morphological parameters and classification results based on each original image-planning image to obtain multiple classes as design style preferences.

[0048] In this embodiment, four categories are generated based on a designer's existing works: a greening category (planning undeveloped areas primarily as green facilities), an industrial category (planning undeveloped areas as industrial parks), a residential category (planning developed areas as community clusters), and a financial category (planning undeveloped areas as financial and commercial districts). Based on different designer works, the feature calculation module 20 and classification module 30 can generate various different categories. Then, by analyzing the specific urban design morphology parameters and classification results of these categories, the actual meaning of the design style preference corresponding to each category, i.e., the design focus, is obtained.

[0049] The training set construction module 40 is used to construct training datasets for each design style preference, using all the corresponding original image-planning image pairs as the training dataset.

[0050] The model building module 50 is used to build the corresponding initial model for each design style preference and train it according to the corresponding training dataset to obtain the corresponding style design model.

[0051] Figure 2 This is a schematic diagram of the process for generating a style design model corresponding to design style preferences in an embodiment of the present invention.

[0052] like Figure 2 As shown, the process of generating a style design model corresponding to a design style preference in the model building module 50 includes the following steps:

[0053] Step T1: Construct the generator and discriminator as a generative adversarial network.

[0054] Step T2: Train the generative adversarial network based on the training dataset to obtain the trained generator.

[0055] In step T2 of this embodiment, the original image from the original image-planned image pair in the training dataset is selected and input into the generator to obtain the corresponding generated image. The generated image is then input into the discriminator to obtain the discrimination result. Finally, the parameters of the generator and the discriminator are iteratively optimized based on the discrimination result to obtain the trained generator.

[0056] Step T3: Input the original images of each region in the training dataset into the trained generator to obtain the corresponding predicted planning images.

[0057] Step T4: For each predicted planning image, calculate the similarity between the predicted planning image and the corresponding regional planning image to obtain the similarity value.

[0058] The expression for calculating similarity values ​​in this embodiment is:

[0059]

[0060] In the formula, the predicted image index is the morphological index corresponding to the predicted planning image, and the target image index is the morphological index corresponding to the regional planning image. That is, the similarity between the predicted planning image and the corresponding regional planning image is quantified by the morphological index in the field of architecture.

[0061] Step T5: Determine whether each similarity value is greater than or equal to the preset similarity value threshold. If yes, use the trained generator as the style design model. If no, remove the original images of regions with similarity values ​​less than the similarity value threshold from the training dataset and execute step T2.

[0062] In this embodiment, a last-place elimination method is used to optimize the training effect. Specifically, for the trained generator, the similarity values ​​of the original images of each region in the training dataset are sorted from largest to smallest. If the similarity value of the last element in the sorted queue is less than the similarity value threshold, it indicates that the original image-regional planning image of the corresponding region in the training dataset is not significant for that feature, meaning that the trained generator cannot accurately imitate the regional planning image of that region. Therefore, the original image-planning image pair is removed from the training dataset, and the generator is trained again using the updated training dataset until the last element's similarity value is greater than the similarity threshold, thereby improving the generator's ability to generate corresponding style design preferences based on the features in the original images of the regions.

[0063] The style selection module 60 is used to select the corresponding design style preference as the optimal design style preference based on the planning and design parameters of the area to be planned.

[0064] The process involves comparing the similarity between the planning and design parameters and the corresponding urban design morphological parameters and classification results for each design style preference. The design style preference with the highest similarity is selected as the optimal design style preference. This ensures consistency between the design style and the training samples, as well as consistency in the generated results across eight morphological parameters: building density, plot ratio, average height, building height variance, building configuration variance, green space ratio, water surface ratio, and hard plaza ratio, thus guaranteeing the reliability of the generated results.

[0065] The initial map generation module 70 is used to input the image of the area to be planned into the style design model corresponding to the optimal design style preference to obtain the initial design map.

[0066] The optimization module 80 is used to perform multi-objective optimization on the initial design drawing to obtain an auxiliary design drawing.

[0067] In step S8, multi-objective optimization involves adjusting the building combination in the initial design drawing until the parameters of the initial design drawing meet the planning and design parameters, thus obtaining the auxiliary design drawing.

[0068] This embodiment employs a genetic algorithm for multi-objective optimization. Specifically, the parameters and basic location of each building are treated as locally adjustable parameters. The activity level is determined by using benchmark parameter values ​​(multiple target values) on the urban morphology parameters. Subsequently, urban indicators are calculated based on these benchmark values ​​and compared with those of the benchmark sample. This process, under constraints of limited optimization adjustments and various local building fine-tuning, identifies a combination of morphological parameters that closely approximates the benchmark sample.

[0069] Figure 3 This is a comparative diagram showing the multi-objective optimization before and after in an embodiment of the present invention.

[0070] like Figure 3 As shown, (a) is the initial design drawing, and (b) is the auxiliary design drawing generated after multi-objective optimization. The objective values ​​of the parameters are: height diversity +10%, configuration diversity -10%, average building height <35, plot ratio >1.5, and building density +30%. It can be seen that multi-objective optimization can generate a design drawing that better meets the requirements.

[0071] The control module 90 stores the control program that controls the operation of each module.

[0072] The process of using the urban design generation assistance device 100 to assist in urban design generation will be described below with reference to the accompanying drawings.

[0073] Figure 4 This is a flowchart illustrating the urban design generation assistance method in an embodiment of the present invention.

[0074] like Figure 4 As shown, the urban design generation assistance method includes the following steps:

[0075] Step S1: The data extraction module 10 is used to collect the regional planning images of the existing design schemes of the designated designer and the corresponding original images of the unplanned areas, generating multiple pairs of original images and planning images.

[0076] Step S2: The feature calculation module 20 calculates multiple urban design morphological parameters of the regional planning image for each original image-planned image pair, and classifies the original regional images to obtain classification results.

[0077] Step S3: The classification module 30 clusters the corresponding urban design morphological parameters and classification results based on each original image-planned image to obtain multiple classes as design style preferences.

[0078] Step S4: Use the training set construction module 40 to construct each design style preference and use all the corresponding original image-planning image pairs as the corresponding training dataset.

[0079] Step S5: Using the model building module 50, construct the corresponding initial model for each design style preference and train it according to the corresponding training dataset to obtain the corresponding style design model.

[0080] Step S6: The style selection module 60 selects the corresponding design style preference as the optimal design style preference based on the planning and design parameters of the area to be planned.

[0081] Step S7: The initial design drawing is obtained by inputting the image of the area to be planned into the style design model corresponding to the optimal design style preference using the initial drawing generation module 70.

[0082] Step S8: Use optimization module 80 to perform multi-objective optimization on the initial design drawing to obtain auxiliary design drawing.

[0083] The role and effect of the embodiments

[0084] According to the urban design generation assistance method and apparatus involved in this embodiment, firstly, the design style preferences of existing designers' works are identified through a data extraction module, a feature calculation module, and a classification module; secondly, a training set construction module and a model construction module train corresponding style design models based on each design style preference; nextly, a style selection module and an initial map generation module select appropriate design style preferences according to the design requirements of the area to be planned, generating an initial design map containing the design style preference; finally, an optimization module optimizes and adjusts the initial design map to obtain an auxiliary design map that better meets the design requirements. In summary, this method can generate regional planning design maps with specified designer style preferences, providing a reference for regional planning and design.

[0085] Those skilled in the art should understand that this invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to this invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. An urban design generation assistance method for generating an auxiliary design drawing with a specified designer style corresponding to a region to be planned, characterized by, The method comprises the following steps: Step S1, collecting the regional planning image and the corresponding un-planned area original image of the existing design scheme of the designated designer, and generating a plurality of original image-planning image pairs; Step S2, for each original image-planning image pair, calculating a plurality of urban design form parameters of the regional planning image, and classifying the regional original image to obtain a classification result; Step S3, clustering the urban design form parameters and the classification result corresponding to each original image-planning image pair to obtain a plurality of classes as design style preferences; Step S4, for each design style preference, all original image-planning image pairs corresponding thereto are taken as corresponding training data sets; Step S5, for each design style preference, an initial model is constructed and trained according to the corresponding training data set to obtain a corresponding style design model; Step S6, according to the planning design parameters of the to-be-planned area, selecting the corresponding design style preference as the optimal design style preference; Step S7, inputting the image of the to-be-planned area into the style design model corresponding to the optimal design style preference to obtain an initial design drawing; Step S8, performing multi-objective optimization on the initial design drawing to obtain the auxiliary design drawing, In step S1, the un-planned area original image has a fixed size of 800m by 800m, In step S3, the original image-planning image pair is further supplemented and improved by the mean value of the urban design form parameters of the surrounding area of the regional planning image, In step S6, the planning design parameters, the urban design form parameters and the classification result corresponding to each design style preference are compared in similarity, and the design style preference with the highest similarity is selected as the optimal design style preference, In step S8, the multi-objective optimization is to adjust the building combination in the initial design drawing until the parameters of the initial design drawing meet the planning design parameters, and then the auxiliary design drawing is obtained.

2. The urban design generation auxiliary method according to claim 1, wherein: wherein The urban design form parameters include building density, volume rate, average height, building height variance, building configuration variance, green space rate, water surface rate, and hard square rate.

3. The urban design generation auxiliary method according to claim 1, wherein: wherein The classification result includes un-built type and built type, When the construction proportion in the regional original image is less than a preset value, the regional original image is of un-built type, When the construction proportion in the regional original image is greater than a preset value, the regional original image is of built type.

4. The urban design generation assistance method according to Claim 1, In step S5, for each design style preference, generating a corresponding style design model comprises the following steps: Step T1, constructing a generator and a discriminator as a generative adversarial network; Step T2, training the generative adversarial network according to the training data set to obtain a trained generator; ​ Step T3, input each of the region original images in the training data set into the trained generator respectively, and obtain the corresponding predicted planning image respectively; Step T4, for each of the predicted planning images, calculate the similarity between the predicted planning image and the corresponding region planning image to obtain a similarity value; Step T5, determine whether each of the similarity values is greater than or equal to a preset similarity threshold value, if yes, the trained generator is used as the style design model, if no, the region original image with a similarity value less than the similarity threshold value is removed from the training data set, and the step T2 is executed.

5. An urban design generation assistance device for generating an auxiliary design map having a specified designer style corresponding to a region to be planned, characterized by comprising: a design map generation unit configured to generate the auxiliary design map having the specified designer style corresponding to the region to be planned; and a design map output unit configured to output the auxiliary design map generated by the design map generation unit. Comprise: A data extraction module is configured to collect region planning images and corresponding un-planned region original images of existing design schemes of the specified designer, and generate a plurality of original image-planning image pairs; A feature calculation module is configured to calculate a plurality of urban design form parameters of the region planning image for each of the original image-planning image pairs, and classify the region original image to obtain a classification result; A classification module is configured to cluster the urban design form parameters and the classification result corresponding to each of the original image-planning image pairs to obtain a plurality of classes as design style preferences; A training set construction module is configured to, for each of the design style preferences, use all of the original image-planning image pairs corresponding thereto as a corresponding training data set; A model construction module is configured to, for each of the design style preferences, construct a corresponding initial model and train the initial model according to the corresponding training data set to obtain a corresponding style design model; A style selection module is configured to select a corresponding design style preference as an optimal design style preference according to a planning design parameter of the region to be planned; An initial image generation module is configured to input an image of the region to be planned into the style design model corresponding to the optimal design style preference to obtain an initial design image; An optimization module is configured to perform multi-objective optimization on the initial design image to obtain the auxiliary design image.

Images