Cold region high-rise residential building plane small sample training and generating method and device based on diffusion model

Abstract

The invention provides a cold region high-rise residential building plane small sample training and generating method and device based on a diffusion model. The method comprises the following steps: collecting and sorting an original data set of a standard plane graph of the high-rise residential building in the cold region, analyzing to obtain a function partition, a house type profile graph and an entrance information data set corresponding to each standard plane graph, and taking text description generated by statistical information corresponding to each house type plane as an original training data set; training a generation model from the house type contour to the house type plane graph based on a diffusion model framework; a tool platform capable of realizing intelligent design and automatic generation of a house type plane graph is developed, and a house type plane graph meeting conditions can be generated by combining graphic information such as a basic house type contour and entrances and exits with specific required text information of each functional room. The invention provides a high-efficiency and high-accuracy automatic residential building plane scheme generation method and device.

Description

Technical Field

[0001] The present invention relates to the technical field of high-rise residential building plane generation, and in particular to a method and device for training and generating small samples of high-rise residential building planes in cold regions based on a diffusion model. Background Art

[0002] Residential buildings are important places for people to live and work. With the improvement of living standards and the enhancement of individual aesthetic needs, people are paying more and more attention to the level and convenience of their own residences and living houses, and the demand for residential building unit design is becoming more and more diversified. There are also large differences in the needs of different users for residential building unit design. How to combine the needs of different families or groups to generate healthy, livable and comfortable high-rise residential building unit design solutions to meet people's diverse needs for living space is particularly important.

[0003] Due to my country's vast territory and complex geographical environment, there are significant differences in the plans of residential buildings in the north and south in a broad sense. The climate in the cold north is harsh, and higher requirements are placed on thermal performance such as building warmth and cold protection. Therefore, it is necessary to design the plans of high-rise buildings in cold regions in a more targeted manner that is more suitable.

[0004] At present, most architectural design thinking is a "top-down" way of thinking that takes the designer as the main body and makes design decisions based on the designer's subjective judgment. At present, to generate a residential building plan, it is mainly drawn manually using computer-aided design tools through human-computer interaction by designers. The main disadvantage of using traditional CAD design is that the design deepening efficiency is low, the entire process is almost manually involved, and the quality and speed of generation greatly depend on the experience and business level of the participants, and residential building plan design drawings cannot be generated in batches.

[0005] Another disadvantage is the high error rate. For some designers who do not have enough design experience, when creating residential building plans, they may encounter problems such as not meeting design specifications and standards. Moreover, architectural creation errors can easily lead to subsequent impacts such as design rework and extended construction period, and the cost of error correction is high.

[0006] In addition, since the design process is centered on the designer, there is often a problem of low design personalization. The design decisions made by the designer's subjective judgment are to some extent insufficiently personalized and cannot meet the diverse needs of residential building plans.

[0007] Therefore, how to provide a high-rise residential building plan generation design method that fits the regional climate and meets personalized multi-cloud needs is a key technical problem that needs to be solved urgently in this field. Summary of the invention

[0008] The present invention proposes a method and device for training and generating small samples of cold-region high-rise residential building plans based on a diffusion model. The method and device are used to improve the accuracy and efficiency of the plan design of cold-region high-rise residential buildings in my country, enhance its ability to meet the needs of diversified apartment types, and solve the problem that existing design technologies are difficult to efficiently and customized generate cold-region high-rise residential building plan design solutions.

[0009] The present invention is implemented by the following technical scheme. The present invention proposes a method for training and generating a small sample of a high-rise residential building plan in cold regions based on a diffusion model. The method comprises:

[0010] Obtain the floor plans of high-rise residential buildings in cold regions, analyze and obtain the residential building area, apartment area, apartment layout, apartment outline and residential structure size information corresponding to each standard floor plan as original training data;

[0011] The above-mentioned floor plan information is organized into text descriptions, which are mapped to the corresponding pictures as text input conditions; the indoor functional layout is annotated using the color block annotation method as picture input conditions; the text data and picture data are used as input conditions to train the cold-region high-rise residential building plane diffusion model through the low-order adaptive modeling LoRA of the large language model;

[0012] After the diffusion model training is completed, the text description and outline map of the target floor plan are used as input parameters, and the cold-region high-rise residential building plan generated by the trained diffusion model is used as the output.

[0013] Furthermore, the method also includes: evaluating whether the generated cold-region high-rise residential building plan meets the specification requirements through an evaluation model, and if not, regenerating it until it meets the requirements.

[0014] Furthermore, the evaluation model's criteria during the evaluation process include:

[0015] a. Whether the adjacent functional areas are reasonable; b. Whether the area of ​​a functional area is reasonable; c. Whether it is generated completely according to the input contour; d. Whether all required functions are generated;

[0016] If it does not meet the evaluation criteria, it will be regenerated until it meets the requirements.

[0017] Furthermore, the floor plan information is organized into text descriptions to form a corresponding relationship with the corresponding pictures, specifically:

[0018] The floor plan information is organized into a standard format for machine recognition, such as "x room x hall x bathroom x kitchen x balcony - x square meters", and a corresponding relationship is formed with the corresponding pictures as a text training data set;

[0019] Use the color block annotation method to annotate the indoor functional layout as the image input condition, specifically:

[0020] The functional areas, outlines and entrances of the collected floor plans are labeled with different colors to form an image training dataset;

[0021] The specific marking method is: main entrance is red, balcony is orange, bedroom is yellow, kitchen is green, dining room is blue, and bathroom is purple.

[0022] Furthermore, the text description and outline of the target floor plan are used as input parameters, specifically:

[0023] Enter the requirement of "x room x hall x bathroom x kitchen x balcony - x square meters" in the format of the floor plan text description as the text input condition;

[0024] Input the target floor plan outline map, use the control network in the diffusion model to perform edge detection on the outline map and convert it into an edge line map as the image input condition.

[0025] Furthermore, the output images that meet the evaluation requirements are vectorized, specifically:

[0026] Import the images that meet the evaluation requirements into the vectorization module, eliminate irrelevant information in the image through image preprocessing, and restore useful real information;

[0027] In the vectorization module, content recognition is performed through edge detection, contour tracking and contour expression, and the data of walls, windows and doors are layered to complete the vectorization of layered graphics.

[0028] Furthermore, the method further comprises:

[0029] The area of ​​specific functional areas in the standard floor plan of high-rise residential buildings in cold regions is obtained, and the area is trained with the corresponding apartment layout of high-rise residential buildings in cold regions to obtain an AI evaluation model for the area of ​​specific functional areas;

[0030] According to the AI ​​evaluation model for the area of ​​specific functional areas, it is evaluated that the area of ​​specific functional areas in the generated high-rise residential building in cold regions is within a reasonable range, the outer contour of the generated high-rise residential building in cold regions is consistent with the target floor plan outline at the input end, and the plane of the high-rise residential building in cold regions is confirmed and output.

[0031] The present invention proposes a small sample training and generation device for cold-region high-rise residential building planes based on a diffusion model, the device comprising:

[0032] Image data processing module: used to process the plane data of high-rise residential buildings in cold regions during the training process. The image correction unit corrects and scales the image to a uniform resolution of 512×512 pixels. After that, the data annotation module annotates the functional areas, outlines and entrances of the corrected plane map with different colors; the specific annotation method is: the main entrance is red, the balcony is orange, the bedroom is yellow, the kitchen is green, the dining room is blue, and the bathroom is purple;

[0033] Text information annotation module: used to annotate the text information describing the characteristics of high-rise residential buildings in cold regions. On the basis of the standard floor area, apartment area, apartment layout and apartment outline data corresponding to the floor plan of the high-rise residential building obtained by the information extraction unit, the information annotation unit uses the standard format of "x room x hall x bathroom x kitchen x balcony - x square meters" as the picture description for machine recognition, as the text training data set;

[0034] Small sample training module: used to train the plane feature model of high-rise residential buildings in cold regions based on a small sample size. It uses a small amount of text and image training data sets as input for training to build a plane diffusion model of high-rise residential buildings in cold regions.

[0035] Building plan generation module: used to generate the plan of high-rise residential buildings in cold regions based on text feature description, with the conditional image input unit of the target apartment outline and the location of the entrance and exit of the target apartment, and the demand text input unit of the high-rise residential building apartment text description as input, the diffusion model obtained by training as the black box processing module, and the plan of high-rise residential buildings in cold regions as output;

[0036] Post-evaluation module: used to evaluate whether the floor plan generated by the floor plan generation module meets the design requirements of residential buildings. If not, it will be regenerated until it meets the requirements.

[0037] The present invention also proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a method for training and generating a small sample of a cold-region high-rise residential building plan based on a diffusion model are implemented.

[0038] The present invention also proposes a computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the steps of a method for training and generating a small sample of a cold-region high-rise residential building plan based on a diffusion model.

[0039] Beneficial effects of the present invention:

[0040] Compared with the existing methods, the innovative points of the method proposed in the present invention are: 1) breaking through the limitation that the existing high-rise building apartment design is limited to the personal experience accumulation of the designer, and realizing the generation of high-rise residential building apartment plans that conform to the regional characteristics of cold regions and meet diversified needs by integrating deep learning modeling technology; 2) Compared with the existing methods integrating deep learning modeling, the method proposed in the present invention realizes small sample training through low-order adaptive modeling of a large language model, which solves the problem that the existing training methods require large-scale training data sets and the training data is difficult to obtain, and avoids the bottleneck problem of large-scale data set annotation process with large workload and low efficiency in the existing methods; 3) The proposed method significantly improves the interaction efficiency between the designer and the model by introducing multi-modal information of "text-image" collaboration, reduces the promotion threshold of the proposed method in the engineering practice design process, and has strong scalability. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0042] Figure 1 It is a schematic diagram of a deep learning model training process for automatically generating a plan of a high-rise residential building in a cold region in an embodiment of the present invention;

[0043] Figure 2 This is a schematic diagram of a design flow for automatically generating a plan of a high-rise residential building in a cold region in this embodiment;

[0044] Figure 3 It is a schematic diagram of the operation of a diffusion model used in this embodiment to automatically generate a plane model of a high-rise residential building in a cold region;

[0045] Figure 4 This example includes the whole process of training apartment description words and generating floor plans;

[0046] Figure 5 It is an original floor plan example of a high-rise residential building in a cold region as training data;

[0047] Figure 6 It is a marked plan of a high-rise residential building in a cold region;

[0048] Figure 7 It is a floor plan of a high-rise residential building in a cold region automatically generated by a diffusion model shown in this embodiment;

[0049] Figure 8This is a schematic diagram of the functional modules of a small sample training and generation device for the plan of high-rise residential buildings in cold regions based on a diffusion model.

[0050] Fig. 9 This is a schematic diagram of the functional units within the module of a small sample training and generation device for the plan of high-rise residential buildings in cold regions based on a diffusion model. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. 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 are within the scope of protection of the present invention.

[0052] The present invention proposes a method and device for training and generating small samples of high-rise residential building plans in cold regions based on a diffusion model, which integrates deep learning modeling technology to automatically generate high-rise residential building plans. The input is the outer contour of the apartment, the location of the main entrance of the apartment, and a text description of the apartment, and the output is a high-rise residential building plan that meets user needs. Users can perform vector processing on the generated results to achieve in-depth design of the high-rise residential building plan. The method for automatically generating residential building plans in this embodiment is obtained by selecting a small number of high-quality standard floor plans from more than a thousand high-rise residential plans in cold regions as a data set and training a diffusion model. The results output by the model meet standard specifications while ensuring diversity. Figures 1 to 6 As shown, Figure 1 It is a schematic diagram of a deep learning model training process for automatically generating a plan of a high-rise residential building in a cold region in an embodiment of the present invention; Figure 2 This is a schematic diagram of a design flow for automatically generating a plan of a high-rise residential building in a cold region in this embodiment; Figure 3 It is an operating principle diagram of a diffusion model used in this embodiment to automatically generate a plane generation model for high-rise residential buildings in cold regions; Figure 4 This example includes the whole process of training apartment description words and generating floor plans; Figure 5 It is an original floor plan example of a high-rise residential building in a cold region as training data; Figure 6 This is a marked floor plan of a high-rise residential building in cold regions.

[0053] Combination Figure 1-Figure 9 The present invention proposes a method for training and generating a small sample of a high-rise residential building plan in cold regions based on a diffusion model, the method comprising:

[0054] Obtain the floor plans of high-rise residential buildings in cold regions, analyze and obtain the residential building area, apartment area, apartment layout, apartment outline and residential structure size information corresponding to each standard floor plan as original training data;

[0055] The above-mentioned floor plan information is organized into text descriptions, and a mapping relationship is formed with the corresponding pictures as text input conditions; the indoor functional layout is annotated using the color block annotation method as picture input conditions; the text data and picture data are used as input conditions, and the cold-region high-rise residential building plane diffusion model is trained through the low-rank adaptation of large language models (LoRA);

[0056] After the diffusion model training is completed, the target floor plan description and outline map are used as input parameters, and the cold-region high-rise residential building plane generated by the trained diffusion model is used as the output end; specifically: the cold-region high-rise residential building plane diffusion model constructed above is applied, with the target floor plan outline of the cold-region high-rise residential building, the location of the target floor plan entrance and exit, and the high-rise residential building floor plan text description as input, and the cold-region high-rise residential building plane generated by the trained diffusion model is used as the output.

[0057] Furthermore, the method also includes: evaluating whether the generated cold-region high-rise residential building plan meets the specification requirements through an evaluation model, and if not, regenerating it until it meets the requirements.

[0058] Furthermore, the evaluation model's criteria during the evaluation process include:

[0059] a. Whether the adjacent functional areas are reasonable; b. Whether the area of ​​a functional area is reasonable; c. Whether it is generated completely according to the input contour; d. Whether all required functions are generated;

[0060] If it does not meet the evaluation criteria, it will be regenerated until it meets the requirements.

[0061] Furthermore, the floor plan information is organized into text descriptions to form a corresponding relationship with the corresponding pictures, specifically:

[0062] The floor plan information is organized into a standard format for machine recognition, such as "x room x hall x bathroom x kitchen x balcony - x square meters", and a corresponding relationship is formed with the corresponding pictures as a text training data set;

[0063] Use the color block annotation method to annotate the indoor functional layout as the image input condition, specifically:

[0064] The functional areas, outlines and entrances of the collected floor plans are labeled with different colors to form an image training dataset;

[0065] The specific marking method is: main entrance is red, balcony is orange, bedroom is yellow, kitchen is green, dining room is blue, and bathroom is purple.

[0066] Furthermore, the text description and outline of the target floor plan are used as input parameters, specifically:

[0067] Enter the requirement of "x room x hall x bathroom x kitchen x balcony - x square meters" in the format of the floor plan text description as the text input condition;

[0068] Input the target floor plan outline map, use the control network in the diffusion model to perform edge detection on the outline map and convert it into an edge line map as the image input condition.

[0069] Furthermore, the output images that meet the evaluation requirements are vectorized, specifically:

[0070] Import the images that meet the evaluation requirements into the vectorization module, eliminate irrelevant information in the image through image preprocessing, and restore useful real information;

[0071] In the vectorization module, content recognition is performed through edge detection, contour tracking and contour expression, and the data of walls, windows and doors are layered to complete the vectorization of layered graphics.

[0072] Furthermore, the method further comprises:

[0073] The area of ​​specific functional areas in the standard floor plan of high-rise residential buildings in cold regions is obtained, and the area is trained with the corresponding apartment layout of high-rise residential buildings in cold regions to obtain an AI evaluation model for the area of ​​specific functional areas;

[0074] According to the AI ​​evaluation model for the area of ​​specific functional areas, it is evaluated that the area of ​​specific functional areas in the generated high-rise residential building in cold regions is within a reasonable range, the outer contour of the generated high-rise residential building in cold regions is consistent with the target floor plan outline at the input end, and the plane of the high-rise residential building in cold regions is confirmed and output.

[0075] The present invention proposes a small sample training and generation device for cold-region high-rise residential building planes based on a diffusion model, the device comprising:

[0076] Image data processing module: used to process the plane data of high-rise residential buildings in cold regions during the training process. The image correction unit corrects and scales the image to a uniform resolution of 512×512 pixels. After that, the data annotation module annotates the functional areas, outlines and entrances of the corrected plane map with different colors; the specific annotation method is: the main entrance is red, the balcony is orange, the bedroom is yellow, the kitchen is green, the dining room is blue, and the bathroom is purple;

[0077] Text information annotation module: used to annotate the text information describing the characteristics of high-rise residential buildings in cold regions. On the basis of the standard floor area, apartment area, apartment layout and apartment outline data corresponding to the floor plan of the high-rise residential building obtained by the information extraction unit, the information annotation unit uses the standard format of "x room x hall x bathroom x kitchen x balcony - x square meters" as the picture description for machine recognition, as the text training data set;

[0078] Small sample training module: used to train the plane feature model of high-rise residential buildings in cold regions based on a small sample size. It uses a small amount of text and image training data sets as input for training to build a plane diffusion model of high-rise residential buildings in cold regions.

[0079] Building plan generation module: used to generate the plan of high-rise residential buildings in cold regions based on text feature description, with the conditional image input unit of the target apartment outline and the location of the entrance and exit of the target apartment, and the demand text input unit of the high-rise residential building apartment text description as input, the diffusion model obtained by training as the black box processing module, and the plan of high-rise residential buildings in cold regions as output;

[0080] Post-evaluation module: used to evaluate whether the floor plan generated by the floor plan generation module meets the design requirements of residential buildings. If not, it will be regenerated until it meets the requirements.

[0081] The present invention proposes a method and device for training and generating small samples of cold-region high-rise residential building plans based on a diffusion model. The method includes: collecting and collating the original data set of standard floor plans of cold-region high-rise residential buildings, analyzing and obtaining the functional zoning, apartment outline and entrance information data sets corresponding to each standard floor plan, and the text description generated by the statistical information corresponding to each apartment plan as the original training data set; training the generation model from apartment outline to apartment plan based on the diffusion model framework; developing a tool platform that can realize intelligent design and automatic generation of apartment plan, and generating apartment plan that meets the conditions by combining the graphic information such as basic apartment outline and entrances and exits with the text information of specific requirements for each functional room. The device includes: an image data processing module, a text information annotation module, a small sample training module, a building plan generation module, and a post-evaluation module. The present invention provides a method and device for automatically generating residential building plan plans with high efficiency and accuracy.

[0082] The present invention also proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a method for training and generating a small sample of a cold-region high-rise residential building plan based on a diffusion model are implemented.

[0083] The present invention also proposes a computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the steps of a method for training and generating a small sample of a cold-region high-rise residential building plan based on a diffusion model.

[0084] The memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus RAM (DRRAM). It should be noted that the memory of the method described in the present invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0085] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. 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 a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a high-density digital video disc (DVD)), or a semiconductor medium (eg, a solid state disc (SSD)).

[0086] In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in a processor or an instruction in the form of software. The steps of the method disclosed in conjunction with the embodiment of the present application can be directly embodied as a hardware processor for execution, or a combination of hardware and software modules in a processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in conjunction with its hardware. To avoid repetition, it is not described in detail here.

[0087] It should be noted that the processor in the embodiment of the present application can be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiment can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to perform, or the hardware and software modules in the decoding processor can be combined and performed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0088] The above is a detailed introduction to the method and device for training and generating small samples of high-rise residential building plans in cold regions based on a diffusion model proposed in the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for training and generating small samples of high-rise residential building plans in cold regions based on a diffusion model, characterized in that: The method comprises: Obtain the floor plans of high-rise residential buildings in cold regions, analyze and obtain the residential building area, apartment area, apartment layout, apartment outline and residential structure size information corresponding to each standard floor plan as original training data; The above-mentioned floor plan information is organized into text descriptions, which are mapped to the corresponding pictures as text input conditions; the indoor functional layout is annotated using the color block annotation method as picture input conditions; the text data and picture data are used as input conditions to train the cold-region high-rise residential building plane diffusion model through the low-order adaptive modeling LoRA of the large language model; After the diffusion model training is completed, the text description and outline map of the target floor plan are used as input parameters, and the cold-region high-rise residential building plan generated by the trained diffusion model is used as the output.

2. The method according to claim 1, characterized in that The method further comprises: evaluating whether the generated cold-region high-rise residential building plan meets the specification requirements through an evaluation model, and if not, regenerating it until it meets the requirements.

3. The method according to claim 2, characterized in that The evaluation model's criteria in the evaluation process include: a. Whether the adjacent functional areas are reasonable; b. Whether the area of ​​a functional area is reasonable; c. Whether it is generated completely according to the input contour; d. Whether all required functions are generated; If it does not meet the evaluation criteria, it will be regenerated until it meets the requirements.

4. The method according to claim 1, characterized in that: The floor plan information is organized into text descriptions and corresponding to the corresponding pictures, specifically: The floor plan information is organized into a standard format for machine recognition, such as "x room x hall x bathroom x kitchen x balcony - x square meters", and a corresponding relationship is formed with the corresponding pictures as a text training data set; Use the color block annotation method to annotate the indoor functional layout as the image input condition, specifically: The functional areas, outlines and entrances of the collected floor plans are labeled with different colors to form an image training dataset; The specific marking method is: main entrance is red, balcony is orange, bedroom is yellow, kitchen is green, dining room is blue, and bathroom is purple.

5. The method according to claim 1, characterized in that The target floor plan text description and outline map are used as input parameters, specifically: Enter the requirement "x room x hall x bathroom x kitchen x balcony - x square meters" in the format of the floor plan text description as the text input condition; Input the target floor plan outline map, use the control network in the diffusion model to perform edge detection on the outline map and convert it into an edge line map as the image input condition.

6. The method according to claim 3, characterized in that Vectorize the output images that meet the evaluation requirements, specifically: Import the images that meet the evaluation requirements into the vectorization module, eliminate irrelevant information in the image through image preprocessing, and restore useful real information; In the vectorization module, content recognition is performed through edge detection, contour tracking and contour expression, and the data of walls, windows and doors are layered to complete the vectorization of layered graphics.

7. The method according to claim 1, characterized in that The method further comprises: The area of ​​specific functional areas in the standard floor plan of high-rise residential buildings in cold regions is obtained, and the area is trained with the corresponding apartment layout of high-rise residential buildings in cold regions to obtain an AI evaluation model for the area of ​​specific functional areas; According to the AI ​​evaluation model for the area of ​​specific functional areas, it is evaluated that the area of ​​specific functional areas in the generated high-rise residential building in cold regions is within a reasonable range, the outer contour of the generated high-rise residential building in cold regions is consistent with the target floor plan outline at the input end, and the plane of the high-rise residential building in cold regions is confirmed and output.

8. A small sample training and generation device for high-rise residential building plans in cold regions based on a diffusion model, characterized in that: The device comprises: Image data processing module: used to process the plane data of high-rise residential buildings in cold regions during the training process. The image correction unit corrects and scales the image to a uniform resolution of 512×512 pixels. After that, the data annotation module annotates the functional areas, outlines and entrances of the corrected plane map with different colors; the specific annotation method is: the main entrance is red, the balcony is orange, the bedroom is yellow, the kitchen is green, the dining room is blue, and the bathroom is purple; Text information annotation module: used to annotate the text information describing the characteristics of high-rise residential buildings in cold regions. On the basis of the standard floor area, apartment area, apartment layout and apartment outline data corresponding to the floor plan of the high-rise residential building obtained by the information extraction unit, the information annotation unit uses the standard format of "x room x hall x bathroom x kitchen x balcony - x square meters" as the picture description for machine recognition and as the text training data set; Small sample training module: used to train the plane feature model of high-rise residential buildings in cold regions based on a small sample size. It uses a small amount of text and image training data sets as input for training to build a plane diffusion model of high-rise residential buildings in cold regions. Building plan generation module: used to generate the plan of high-rise residential buildings in cold regions based on text feature description, with the conditional image input unit of the target apartment outline and the location of the entrance and exit of the target apartment, and the demand text input unit of the high-rise residential building apartment text description as input, the diffusion model obtained by training as the black box processing module, and the plan of high-rise residential buildings in cold regions as output; Post-evaluation module: used to evaluate whether the floor plan generated by the floor plan generation module meets the design requirements of residential buildings. If not, it will be regenerated until it meets the requirements.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium for storing computer instructions, characterized in that: When the computer instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images