Subway room layout design method and system based on conditional generative adversarial network

Abstract

The present invention relates to the field of civil engineering technology, and discloses a subway room layout design method and system based on a conditional generative adversarial network. The method comprises: obtaining plane drawings of various projects, simplifying the plane drawings, obtaining layout drawings and heat maps corresponding to the plane drawings, and forming a drawing data set based on the corresponding layout drawings and heat maps; using the drawing data set to train a conditional generative adversarial network model to obtain a trained conditional generative adversarial network model; determining the positions of proxy points of each subway station room to be designed and the area of ​​each subway station room to be designed according to engineering requirements, obtaining a heat map of the required layout, inputting the heat map of the required layout into the trained conditional generative adversarial network model, and obtaining a layout map of the required layout as the final layout of the subway station room to be designed. The present invention can automatically and quickly generate layout plans, reduce conflicts or unreasonableness caused by design errors, and greatly shorten the design cycle.

Description

Technical Field

[0001] The present invention relates to the field of civil engineering, and in particular to a subway room layout design method and system based on a conditional generative adversarial network. Background Art

[0002] With the acceleration of China's urbanization process, rail transit construction has become one of the key factors in promoting urban development and improving the quality of life of residents. From the development of a single line to a grid layout, rail transit has not only improved the traffic conditions in the city, but also promoted the development of the regional economy and social progress. In this context, the application of artificial intelligence technology has brought revolutionary changes to the design and operation of rail transit.

[0003] In traditional rail transit design, drawing design is a time-consuming and labor-intensive process. Designers need to manually draw various layout plans according to established design specifications and project requirements, and repeatedly adjust them to ensure that they meet various requirements such as functionality and safety. This working mode is not only inefficient, but also prone to waste of resources, because many design results may not be effectively used in subsequent projects. In addition, due to the different specific circumstances of each project, designers often need to start from scratch, which further increases time and labor costs.

[0004] Therefore, there is an urgent need for a subway room layout design method and system based on conditional generative adversarial networks, which can automatically and quickly generate layout plans, reduce conflicts or unreasonableness caused by design errors, and greatly shorten the design cycle. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a subway room layout design method and system based on a conditional generative adversarial network, which can automatically and quickly generate layout plans, reduce conflicts or unreasonableness caused by design errors, and greatly shorten the design cycle.

[0006] The present invention provides a subway room layout design method based on a conditional generative adversarial network, comprising the following steps:

[0007] S1. Obtain the plan drawings of various projects, simplify the plan drawings, obtain the layout drawings and heat maps corresponding to the plan drawings, and construct a drawing data set based on the corresponding layout drawings and heat maps;

[0008] S2. Use the drawing data set to train the conditional generative adversarial network model to obtain a trained conditional generative adversarial network model;

[0009] S3. Determine the location of the proxy point of each subway station room to be designed and the area of ​​each subway station room to be designed according to the engineering requirements, obtain the heat map of the required layout, input the heat map of the required layout into the trained conditional generative adversarial network model, and obtain the layout diagram of the required layout as the final layout of the subway station room to be designed; wherein the proxy point of the room is the center point of each subway station room.

[0010] Furthermore, in S1, the plane drawing is simplified to obtain a layout drawing corresponding to the plane drawing, including:

[0011] S11. Classify the same type of plan drawings into a drawing library;

[0012] S12. Simplify the plan drawings in the drawing library, retain only the outer contours of each subway station room and proxy points, and obtain a layout drawing corresponding to the plan drawings.

[0013] Furthermore, in S1, the plane drawing is simplified, and the heat map corresponding to the plane drawing is obtained, including:

[0014] S13, further simplifying the layout drawing corresponding to the plan drawing, and only retaining the subway station boundary and the proxy points of each subway station room; wherein the subway station boundary is the outer contour of the area where the subway station room layout needs to be designed;

[0015] S14, dividing the space within the boundary of the subway station into a number of sub-areas with regular shapes;

[0016] S15, respectively calculate the distance between the center point of each sub-region and all proxy points. If the i-th sub-region b i The distance d between the center point and the jth proxy point ij The shortest, then sub-area b i Classified as the subway station room to which agent point j belongs;

[0017] S16. When the area of ​​the subway station room to which the proxy point j belongs is greater than or equal to the preset area, the jth proxy point is removed, and S15-S16 are repeatedly executed until all proxy points are removed, and the spatial division of each subway station room is completed;

[0018] S17. Render the drawing after completing the space division to obtain a heat map corresponding to the plane drawing.

[0019] Further, in S2, the conditional generative adversarial network model is trained using the drawing data set to obtain a trained conditional generative adversarial network model, wherein the conditional generative adversarial network model adopts a pix2pix model;

[0020] The pix2pix model includes a generator G and a discriminator D;

[0021] The generator G adopts the U-net network structure to generate the predicted layout diagram according to the heat map in the drawing dataset;

[0022] The discriminator D adopts the PatchGAN network structure to judge the similarity between the predicted layout diagram and the corresponding layout diagram in the drawing dataset.

[0023] Furthermore, in S2, the conditional generative adversarial network model is trained using the drawing dataset to obtain a trained conditional generative adversarial network model. The training process is as follows:

[0024] S21, preprocessing the heat map and layout diagram in the drawing data set, and making the preprocessed heat map and layout diagram uniformly represented by RGB three-color channel mode; wherein the preprocessing includes normalization processing and centering processing;

[0025] S22. Define the loss function of the conditional generative adversarial network model;

[0026] S23. Iteratively train the conditional generative adversarial network model through the preprocessed drawing data set, and update the parameters of the generator G and the discriminator D in real time through back propagation. When the loss function converges, the training is completed and the trained conditional generative adversarial network model is obtained.

[0027] Furthermore, in S3, the location of the proxy point of each subway station room to be designed and the area of ​​each subway station room to be designed are determined according to the engineering requirements, and the heat map of the required layout is obtained, including:

[0028] S31. Determine the boundary of the subway station to be designed according to the engineering requirements;

[0029] S32. Determine the location of the proxy point of each subway station room to be designed and the area of ​​each subway station room to be designed according to the engineering requirements;

[0030] S33, dividing the space inside the boundary of the subway station to be designed according to the boundary of the subway station to be designed, the positions of the proxy points of each room of the subway station to be designed, and the area of ​​each room of the subway station to be designed;

[0031] S34. Render the drawings that complete the space division to obtain a heat map of the required layout.

[0032] The present invention also provides a subway room layout design system based on a conditional generative adversarial network, which is used to execute any of the above-mentioned subway room layout design methods based on a conditional generative adversarial network. The system includes the following modules:

[0033] The data set construction module is used to obtain the plane drawings of various projects, pre-process the plane drawings, obtain the layout drawings and heat maps corresponding to the plane drawings, and construct the drawing data set based on the corresponding layout drawings and heat maps;

[0034] The conditional generative network training module is connected to the data set construction module and is used to train the conditional generative adversarial network model using the drawing data set to obtain a trained conditional generative adversarial network model;

[0035] The output module is connected to the conditional generative network training module, and is used to determine the location of the proxy point of each subway station room to be designed and the area of ​​each subway station room to be designed according to the engineering requirements, obtain the heat map of the required layout, input the heat map of the required layout into the trained conditional generative adversarial network model, and obtain the layout diagram of the required layout as the final layout of the subway station room to be designed; wherein the proxy point of the room is the center point of each subway station room.

[0036] The embodiments of the present invention have the following technical effects:

[0037] The present invention processes based on plane drawings to obtain a drawing data set consisting of paired data of corresponding layout drawings and heat maps, and constructs a conditional generative adversarial network model. The conditional generative adversarial network model is trained through the drawing data set to obtain a trained conditional generative adversarial network model. The positions of proxy points of subway station rooms and the areas of subway station rooms are set according to engineering requirements to obtain corresponding heat maps. The layout drawing of the subway station room layout design is obtained according to the heat maps and the trained conditional generative adversarial network model. The conditional generative adversarial network model can quickly generate layout plans that meet specific requirements by learning a large number of historical cases, thereby reducing a large number of repetitive manual operations, and can automatically and quickly generate layout plans, thereby reducing conflicts or unreasonableness caused by design errors, and greatly shortening the design cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 It is a flow chart of a subway room layout design method based on a conditional generative adversarial network provided by an embodiment of the present invention;

[0040] Figure 2 It is a schematic diagram of a layout drawing corresponding to the plane drawing obtained by simplifying the plane drawing provided by an embodiment of the present invention;

[0041] Figure 3 It is a schematic diagram of a heat map corresponding to a plane drawing obtained by simplifying the plane drawing provided by an embodiment of the present invention;

[0042] Figure 4 It is a schematic diagram of forming a drawing data set based on a layout diagram and a heat map provided by an embodiment of the present invention;

[0043] Figure 5 It is a schematic diagram of a conditional generative adversarial network model training principle provided by an embodiment of the present invention;

[0044] Figure 6 It is a schematic diagram of a layout diagram of training a conditional generative adversarial network model and generating a required layout based on the trained conditional generative adversarial network model provided by an embodiment of the present invention;

[0045] Figure 7 It is a structural schematic diagram of a subway room layout design system based on a conditional generative adversarial network provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0047] This paper proposes a subway room layout design method based on conditional generative adversarial network. Figure 1 is a flowchart of a subway room layout design method based on a conditional generative adversarial network provided by an embodiment of the present invention, see Figure 1 , including:

[0048] S1. Obtain the plan drawings of various projects, simplify the plan drawings, obtain the layout drawings and heat maps corresponding to the plan drawings, and construct a drawing dataset based on the corresponding layout drawings and heat maps.

[0049] Specifically, Figure 2 is a schematic diagram of a layout diagram corresponding to a plane drawing obtained by simplifying the plane drawing provided by an embodiment of the present invention, Figure 3 is a schematic diagram of a heat map corresponding to a plane drawing obtained by simplifying the plane drawing provided by an embodiment of the present invention, see Figure 2 and Figure 3 , simplify the plane drawing, and obtain the layout diagram and heat map corresponding to the plane drawing, including:

[0050] S11. Classify the same type of plan drawings into a drawing library.

[0051] S12. Simplify the plan drawings in the drawing library, retain only the outer contours of each subway station room and proxy points, and obtain a layout drawing corresponding to the plan drawings.

[0052] S13. Based on the layout diagram corresponding to the plan drawing, further simplification is performed to retain only the boundaries of the subway station and the proxy points of each subway station room.

[0053] The subway station boundary is the outer contour of the area where the subway station room layout needs to be designed.

[0054] S14. Divide the space within the boundary of the subway station into a number of sub-areas with regular shapes.

[0055] Specifically, the sum of the areas of all sub-areas is the total area within the boundary of the subway station, i.e., b i satisfy , where S represents the total area within the boundary of the subway station.

[0056] S15, respectively calculate the distance between the center point of each sub-region and all proxy points. If the i-th sub-region b i The distance d between the center point and the jth proxy point ij The shortest, then sub-area b i Classified as the subway station room to which agent point j belongs.

[0057] Specifically, different color markings are used to distinguish the subway station rooms represented by different proxy points. i is classified as the subway station room to which the agent point j belongs, then the i-th sub-area b i The color of the room marked as the subway station to which the agent point j belongs.

[0058] S16. When the area of ​​the subway station room to which the proxy point j belongs is greater than or equal to the preset area, the jth proxy point is removed, and S15-S16 are repeated until all proxy points are removed, completing the spatial division of each subway station room.

[0059] Among them, the preset area can be determined according to actual conditions and design specifications, and the preset areas of different types of subway station rooms are different.

[0060] S17. Render the drawing after completing the space division to obtain a heat map corresponding to the plane drawing.

[0061] S2. Use the drawing dataset to train the conditional generative adversarial network model to obtain a trained conditional generative adversarial network model.

[0062] Among them, the conditional generative adversarial network model adopts the pix2pix model;

[0063] The pix2pix model includes a generator G and a discriminator D;

[0064] The generator G adopts the U-net network structure to generate the predicted layout diagram according to the heat map in the drawing dataset;

[0065] The discriminator D adopts the PatchGAN network structure to judge the similarity between the predicted layout diagram and the corresponding layout diagram in the drawing dataset.

[0066] Specifically, Figure 4 is a schematic diagram of forming a drawing data set based on a layout diagram and a heat map provided by an embodiment of the present invention, Figure 5 is a schematic diagram of a conditional generative adversarial network model training principle provided by an embodiment of the present invention, see Figure 4 and Figure 5 , the process of training the conditional generative adversarial network model using the drawing dataset is as follows:

[0067] S21. Preprocess the heat map and layout diagram in the drawing data set, and make the preprocessed heat map and layout diagram uniformly represented by RGB three-color channel mode.

[0068] Among them, preprocessing includes normalization processing and centralization processing.

[0069] S22. Define the loss function of the conditional generative adversarial network model.

[0070] Specifically, the objective function of the conditional generative adversarial network is defined as:

[0071] ;

[0072] Among them, L cGAN represents the loss function (i.e., objective function) of high-frequency features, E represents expectation, x represents the input heat map, y represents the corresponding layout diagram, and obeys the distribution of x, y~P(y|x); z represents the input noise, G represents the generator, which is responsible for mapping the distribution of the input noise z to x, z~P(G(x,z)|x) and gradually approaching the distribution of the real data (i.e., the corresponding layout diagram y); D represents the discriminator, which is responsible for outputting probability values ​​between 0 and 1.

[0073] The generator G attempts to minimize the above objective function, and the discriminator D attempts to maximize the above objective function. The two optimize their respective performance through a zero-sum game, which can effectively learn the high-frequency features between corresponding images.

[0074] Then, introduce L 1The loss function is used to ensure the similarity between the input heat map and the corresponding layout map in terms of low-frequency features:

[0075] ;

[0076] in, Indicates L 1 The value of the loss function, L 1 The loss function is the loss function of low-frequency features.

[0077] According to the loss function of high-frequency features and the loss function of low-frequency features, the final loss function, that is, the loss function of the conditional generative adversarial network model, can be obtained:

[0078] ;

[0079] in, represents the final loss function, μ represents the hyperparameter used to adjust L 1 The weight of the loss function in the overall loss function.

[0080] S23. Iteratively train the conditional generative adversarial network model through the preprocessed drawing data set, and update the parameters of the generator G and the discriminator D in real time through back propagation. When the loss function converges, the training is completed and the trained conditional generative adversarial network model is obtained.

[0081] Among them, the mapping relationship between the heat map and the layout map can be obtained according to the trained conditional generative adversarial network model.

[0082] S3. Determine the location of the proxy points of each subway station room to be designed and the area of ​​each subway station room to be designed according to the engineering requirements, obtain the heat map of the required layout, input the heat map of the required layout into the trained conditional generative adversarial network model, and obtain the layout diagram of the required layout as the final layout of the subway station room to be designed.

[0083] The proxy point of the room is the center point of each subway station room, and the engineering requirements can include actual engineering needs and design specifications. Figure 6 is a schematic diagram of a layout diagram for training a conditional generative adversarial network model and generating a required layout based on the trained conditional generative adversarial network model provided by an embodiment of the present invention, see Figure 6 :

[0084] S31. Determine the boundaries of the subway station to be designed based on project requirements.

[0085] S32. Determine the location of the proxy point of each subway station room to be designed and the area of ​​each subway station room to be designed according to the engineering requirements.

[0086] S33. Divide the space inside the boundary of the subway station to be designed according to the boundary of the subway station to be designed, the positions of the proxy points of the rooms of the subway station to be designed, and the areas of the rooms of the subway station to be designed.

[0087] Among them, according to the boundary of the subway station to be designed, the location of the proxy point of each subway station room to be designed and the area of ​​each subway station room to be designed, the method of dividing the space inside the boundary of the subway station to be designed is the same as that in S14-S16, and will not be repeated here.

[0088] S34. Render the drawings that complete the space division to obtain a heat map of the required layout.

[0089] The present invention processes based on plane drawings to obtain a drawing data set consisting of paired data of corresponding layout drawings and heat maps, and constructs a conditional generative adversarial network model. The conditional generative adversarial network model is trained through the drawing data set to obtain a trained conditional generative adversarial network model. The positions of proxy points of subway station rooms and the areas of subway station rooms are set according to engineering requirements to obtain corresponding heat maps. The layout drawing of the subway station room layout design is obtained according to the heat maps and the trained conditional generative adversarial network model. The conditional generative adversarial network model can quickly generate layout plans that meet specific requirements by learning a large number of historical cases, thereby reducing a large number of repetitive manual operations, and can automatically and quickly generate layout plans, thereby reducing conflicts or unreasonableness caused by design errors, and greatly shortening the design cycle.

[0090] Figure 7 is a schematic diagram of the structure of a subway room layout design system based on a conditional generative adversarial network provided by an embodiment of the present invention, and the system is used to execute the subway room layout design method based on a conditional generative adversarial network described in the above embodiment, such as Figure 7 As shown, the system includes the following modules:

[0091] The data set construction module is used to obtain the plane drawings of various projects, pre-process the plane drawings, obtain the layout drawings and heat maps corresponding to the plane drawings, and construct the drawing data set based on the corresponding layout drawings and heat maps;

[0092] The conditional generative network training module is connected to the data set construction module and is used to train the conditional generative adversarial network model using the drawing data set to obtain a trained conditional generative adversarial network model;

[0093] The output module is connected to the conditional generative network training module, and is used to determine the location of the proxy point of each subway station room to be designed and the area of ​​each subway station room to be designed according to the engineering requirements, obtain the heat map of the required layout, input the heat map of the required layout into the trained conditional generative adversarial network model, and obtain the layout diagram of the required layout as the final layout of the subway station room to be designed; wherein the proxy point of the room is the center point of each subway station room.

[0094] It should be noted that the terms used in the present invention are only for describing specific embodiments, rather than limiting the scope of the present application. As shown in the present specification, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of more restrictions, the elements defined by the sentence "include one..." do not exclude the presence of other identical elements in the process, method or device including the elements.

[0095] It should also be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", etc. should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be a connection between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A subway room layout design method based on conditional generative adversarial network, characterized in that: The steps include: S1. Obtaining plan drawings of various types of projects, simplifying the plan drawings, obtaining layout drawings and heat maps corresponding to the plan drawings, and constructing a drawing dataset based on the corresponding layout drawings and heat maps; S11. Classify the same type of plan drawings into a drawing library; S12, simplifying the plan drawings in the drawing library, retaining only the outer contours of each subway station room and proxy points, and obtaining a layout drawing corresponding to the plan drawings; S13, further simplifying the layout drawing corresponding to the plan drawing, and only retaining the subway station boundary and the proxy points of each subway station room; wherein the subway station boundary is the outer contour of the area where the subway station room layout needs to be designed; S14, dividing the space within the boundary of the subway station into a number of sub-areas with regular shapes; S15, respectively calculate the distance between the center point of each sub-region and all proxy points. If the i-th sub-region b i The distance d between the center point and the jth proxy point ij shortest, then the sub-region b i Classified as the subway station room to which the agent point j belongs; S16. When the area of ​​the subway station room to which the proxy point j belongs is greater than or equal to the preset area, the jth proxy point is removed, and S15-S16 are repeatedly executed until all proxy points are removed, and the spatial division of each subway station room is completed; S17, rendering the drawings after completing the space division to obtain a heat map corresponding to the plane drawings; S2. Use the drawing data set to train the conditional generative adversarial network model to obtain a trained conditional generative adversarial network model; S3. Determine the location of the proxy point of each subway station room to be designed and the area of ​​each subway station room to be designed according to the engineering requirements, obtain a heat map of the required layout, input the heat map of the required layout into the trained conditional generative adversarial network model, and obtain a layout diagram of the required layout as the final layout of the subway station room to be designed; wherein the proxy point of the room is the center point of each subway station room.

2. The subway room layout design method based on conditional generative adversarial network according to claim 1 is characterized in that: In S2, the conditional generative adversarial network model is trained using the drawing data set to obtain a trained conditional generative adversarial network model, wherein the conditional generative adversarial network model adopts a pix2pix model; The pix2pix model includes a generator G and a discriminator D; The generator G adopts a U-net network structure and is used to generate a predicted layout diagram according to the heat map in the drawing data set; The discriminator D adopts a PatchGAN network structure to determine the similarity between the predicted layout diagram and the corresponding layout diagram in the drawing dataset.

3. The subway room layout design method based on conditional generative adversarial network according to claim 2 is characterized in that: In S2, the conditional generative adversarial network model is trained using the drawing data set to obtain a trained conditional generative adversarial network model. The training process is as follows: S21, preprocessing the heat map and layout diagram in the drawing data set, and making the preprocessed heat map and layout diagram uniformly represented by RGB three-color channel mode; wherein the preprocessing includes normalization processing and centering processing; S22. Define the loss function of the conditional generative adversarial network model; S23. Iteratively train the conditional generative adversarial network model through the preprocessed drawing data set, and update the parameters of the generator G and the discriminator D in real time through back propagation. When the loss function converges, the training is completed and the trained conditional generative adversarial network model is obtained.

4. The subway room layout design method based on conditional generative adversarial network according to claim 1 is characterized in that: In S3, the location of the proxy point of each subway station room to be designed and the area of ​​each subway station room to be designed are determined according to the engineering requirements, and the heat map of the required layout is obtained, including: S31. Determine the boundary of the subway station to be designed according to the engineering requirements; S32. Determine the location of the proxy point of each subway station room to be designed and the area of ​​each subway station room to be designed according to the engineering requirements; S33, dividing the space inside the boundary of the subway station to be designed according to the boundary of the subway station to be designed, the positions of the proxy points of the rooms of the subway station to be designed, and the areas of the rooms of the subway station to be designed; S34. Render the drawings that complete the space division to obtain a heat map of the required layout.

5. A subway room layout design system based on conditional generative adversarial networks, used to implement the subway room layout design method based on conditional generative adversarial networks as described in any one of claims 1 to 4, characterized in that: The system includes the following modules: A data set construction module is used to obtain the plane drawings of various projects, pre-process the plane drawings, obtain the layout drawings and heat maps corresponding to the plane drawings, and construct a drawing data set based on the corresponding layout drawings and heat maps; A conditional generative network training module, connected to the data set construction module, is used to train the conditional generative adversarial network model using the drawing data set to obtain a trained conditional generative adversarial network model; The output module is connected to the conditional generative network training module, and is used to determine the location of the proxy point of each subway station room to be designed and the area of ​​each subway station room to be designed according to the engineering requirements, obtain the heat map of the required layout, input the heat map of the required layout into the trained conditional generative adversarial network model, and obtain the layout diagram of the required layout as the final layout of the subway station room to be designed; wherein the proxy point of the room is the center point of each subway station room.

Images