Design drawing generation method, system and device and storage medium

Abstract

The invention discloses a design drawing generation method. The method comprises the steps of obtaining a target keyword from design demand information; screening picture materials with similar keywords from a material library according to the target keyword to obtain a target picture material; extracting picture key features of the target picture material through a pre-trained feature learning model; and generating a design drawing through a generative model in combination with the key features of the picture and the design demand information. The invention further discloses a design drawing generation system, an electronic device and a computer readable storage medium. Therefore, the design efficiency can be improved, and the finally generated result can be more accurate and better meets user requirements.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and particularly to a method, system, electronic device and computer-readable storage medium for generating design drawings. Background Art

[0002] Most of the existing website and mobile application design technologies rely on the personal experience and innovation ability of designers. This manual design method is not only inefficient, but may also limit the diversity and innovation of designs.

[0003] In addition, the existing artificial intelligence-based design technologies usually need to make a compromise between design quality and generation speed. High-quality design generation often takes longer time, which may not meet the requirements of rapid iteration and real-time preview. Summary of the Invention

[0004] The main purpose of this application is to propose a method, system, electronic device and computer-readable storage medium for generating design drawings, aiming to solve the problem of how to use artificial intelligence to achieve picture design and improve design efficiency and diversity.

[0005] In a first aspect, an embodiment of this application provides a method for generating a design drawing, the method including:

[0006] Obtaining a target keyword from design requirement information;

[0007] Screening picture materials with similar keywords from a material library according to the target keyword to obtain target picture materials;

[0008] Extracting picture key features of the target picture materials through a pre-trained feature learning model;

[0009] Combining the picture key features and the design requirement information, and generating a design drawing through a generative model.

[0010] In a second aspect, an embodiment of this application provides a system for generating a design drawing, the system including:

[0011] An obtaining module, configured to obtain a target keyword from design requirement information;

[0012] A screening module, configured to screen picture materials with similar keywords from a material library according to the target keyword to obtain target picture materials;

[0013] An analysis module, configured to extract picture key features of the target picture materials through a pre-trained feature learning model;

[0014] A generating module, configured to combine the picture key features and the design requirement information, and generate a design drawing through a generative model.

[0015] In a third aspect, an embodiment of the present application provides an electronic device, which includes: a memory, a processor, and a design drawing generation program stored on the memory and executable on the processor. When the design drawing generation program is executed by the processor, it implements the design drawing generation method as described above.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a design drawing generation program is stored. When the design drawing generation program is executed by a processor, it implements the design drawing generation method as described above.

[0017] The design drawing generation method, system, electronic device, and computer-readable storage medium proposed in the embodiments of the present application do not directly generate the design drawing through AIGC based on the design requirement information input by the user. Instead, after obtaining the design requirement information, the target keyword is first obtained, and then multiple similar target picture materials are quickly screened from the material library according to the target keyword. Then, the picture key features of the target picture materials are extracted by using a pre-trained feature learning model, and the picture key features and the design requirement information input by the user are combined and used as the input of the AIGC model to generate the corresponding design drawing. This can not only improve the design efficiency, but also make the final generated result more accurate, more in line with the user's needs, and can also introduce new innovative elements through a large number of picture materials to improve the design diversity. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings herein are used to provide a further understanding of the present application and form a part of the present application. It should be understood that these drawings only depict some embodiments disclosed according to the present application and should not be regarded as limiting the scope of the present application.

[0019] Figure 1 An application environment architecture diagram for implementing various embodiments of the present application;

[0020] Figure 2 A flowchart of a design drawing generation method proposed in the first embodiment of the present application;

[0021] Figure 3 For Figure 2 A detailed flowchart diagram of step S200 in

[0022] Figure 4 A schematic diagram of a front-end interface for providing reference pictures in the present application;

[0023] Figure 5 A flowchart of a design drawing generation method proposed in the second embodiment of the present application;

[0024] Figure 6Schematic diagram of a front - end interface for presenting the design drawing and input adjustment information in this application;

[0025] Figure 7 Schematic diagram of the hardware architecture of an electronic device proposed in the third embodiment of this application;

[0026] Figure 8 Schematic diagram of the modules of a design - drawing generation system proposed in the fourth embodiment of this application;

[0027] Figure 9 Schematic diagram of the modules of a design - drawing generation system proposed in the fifth embodiment of this application. Detailed implementation manners

[0028] In order to make the objectives, technical solutions and advantages of this application more clear and understandable, the following further details this application in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of this application.

[0029] It should be noted that the descriptions involving "first", "second", etc. in the embodiments of this application are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first", "second" may explicitly or implicitly include at least one of such features. In addition, the technical solutions between various embodiments can be combined with each other, but it must be based on the fact that those of ordinary skill in the art can implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0030] The following provides explanations of the terms involved in this application:

[0031] Artificial Intelligence (AI): It is a new technical science that studies, develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. Artificial intelligence includes a very wide range of sciences and is composed of different fields, such as machine learning, computer vision, etc. Generally speaking, one of the main goals of artificial - intelligence research is to enable machines to be competent for some complex tasks that usually require human intelligence to complete.

[0032] Artificial Intelligence Generated Content (AIGC): It refers to the technical methods of artificial intelligence such as generative adversarial networks and large pre-trained models. Through the learning and recognition of existing data, it is a technology that can generate relevant content with appropriate generalization ability. The core idea of AIGC technology is to use artificial intelligence algorithms to generate content with certain creativity and quality. Through training the model and learning a large amount of data, AIGC can generate relevant content according to the input conditions or instructions, including articles, images, audio, etc. For example, AIGC can generate pictures that match the input keywords, descriptions, or samples through text-to-image or image-to-image methods. Among them, the text-to-image technology is a mode of generating images based on text prompts. Different from the text-to-image technology that only uses text prompts, the image-to-image technology takes the reference image and text prompts as common inputs and performs secondary creation on the original reference image.

[0033] Stable Diffusion (SD) model: A deep learning text-to-image generation model, which is a type of AIGC model and is mainly used to draw imaginary pictures according to text descriptions. It adopts a more stable and controllable diffusion process, so that high-quality images can be generated. By inputting text prompts, the model will output an image that matches the prompts.

[0034] Prompt: It is the input of the AIGC model and is used to draw images. Generally, Prompts are divided into positive prompts and negative prompts.

[0035] Based on the problems existing in the existing picture design solutions, the embodiments of this application provide a new design drawing generation solution, which can utilize a large number of picture materials and keywords, and through training the feature learning model, realize autonomous learning and generate approximate website designs, mobile application (APP) design prototypes and design drafts. The design drawing generation solution can not only improve the design efficiency, but also introduce new innovative elements and improve the design diversity.

[0036] Guided by the above inventive concepts, the technical solutions of this application will be specifically described below in combination with each embodiment, taking the page designs of websites and mobile applications (APPs) as examples.

[0037] Please refer to Figure 1 , Figure 1 For an application environment architecture diagram for implementing each embodiment of this application. This application can be applied to an application environment including, but not limited to, client 2, server 4, and network 6.

[0038] Among them, the client 2 is used to receive the design requirement information input by the user, including inputting prompt words, design requirement descriptions, or selecting a reference picture, and displaying the finally generated design drawing to the user, etc. The client 2 can be a terminal device such as a PC (Personal Computer), mobile phone, tablet computer, portable computer, wearable device, etc.

[0039] The server 4 is used to provide data and technical support for the client 2. For example, it obtains the target keyword from the design requirement information, screens the picture materials with similar keywords from the material library according to the target keyword to obtain the target picture materials. Then it extracts the key picture features of the target picture materials through a pre-trained feature learning model. Finally, it combines the key picture features and the design requirement information to generate a design drawing through a generative model. The server 4 can be a computing device such as a rack server, blade server, tower server, or cabinet server, and can be an independent server or a server cluster composed of multiple servers.

[0040] The network 6 can be an enterprise internal network (Intranet), Internet, Global System of Mobile communication (GSM), Wideband Code Division Multiple Access (WCDMA), 4G network, 5G network, Bluetooth, Wi-Fi and other wireless or wired networks. The server 4 and one or more clients 2 are communicatively connected through the network 6 for data transmission and interaction.

[0041] Embodiment 1

[0042] As Figure 2 shown, it is a flowchart of a design drawing generation method proposed in the first embodiment of this application. It can be understood that the flowchart in the embodiment of this method is not used to limit the order of execution steps. According to needs, some steps in this flowchart can also be added or deleted. The following takes the server as the execution subject to illustrate this method.

[0043] This method includes the following steps:

[0044] S200, obtain the target keyword from the design requirement information.

[0045] The purpose of this embodiment is to automatically generate a website page design drawing, a mobile APP page design drawing, or other types of design drawings that meet the design requirements for the user. The design requirements include a specified color scheme, layout style, or specific functional requirements, etc.

[0046] In this embodiment, various ways can be provided for users to input design requirement information, including but not limited to directly inputting prompt words or text descriptions of design requirements, or providing a variety of reference design drawings to the users for them to select a reference picture close to their own requirements.

[0047] In an alternative embodiment, the user can customize and describe the page design they want, such as inputting the text description of design requirements "Generate a personal page containing common personal information and with an orange theme".

[0048] In an alternative embodiment, in order to lower the operation threshold, drop-down options for various page features can also be provided on the front-end page, with multiple preset alternative prompt words provided for each feature to simplify the user operation. The user only needs to select the corresponding target prompt word in the corresponding drop-down box according to the desired page design. For example, drop-down options for features such as page type and background color are provided to the user, and the user can select "personal page" in the page type option and "orange" in the background color option, etc.

[0049] For the case of inputting prompt words, the server directly obtains the prompt words to get the target keywords for subsequent processing.

[0050] For the case of inputting text descriptions of design requirements, after the server obtains the text description of design requirements input by the user, it needs to extract the target keywords from the text description of design requirements and then perform subsequent processing. For example, for the text description of design requirements input by the user "Generate a personal page containing common personal information and with an orange theme", the keywords "personal page", "personal information", and "orange theme" can be extracted.

[0051] In an alternative embodiment, for the case of selecting a reference picture, further refer to Figure 3 , which is a detailed flowchart of step S200 above. It can be understood that this flowchart is not used to limit the order of execution steps. As needed, some steps in this flowchart can also be added or deleted. In this embodiment, step S200 specifically includes:

[0052] S2000, provide multiple reference pictures on the front-end interface for the user to select according to the design requirements.

[0053] In order to allow the user to select the design style they need, this embodiment provides a variety of website page or mobile APP page reference pictures for the user in the front-end interface. Optionally, the reference pictures can be a part of the picture materials in the following material library.

[0054] Such as Figure 4As shown, it is a schematic diagram of the front-end interface that provides reference pictures in an embodiment of the present application. In Figure 4 multiple types of classifications are made for website pages or mobile APP pages, such as login pages, activity pages, detail pages, personal centers, navigation pages, system setting pages, etc. Multiple reference pictures are provided for each type. Users can first select the desired page type according to their own design needs, and then select a reference picture in this page type as input.

[0055] S2002, in response to the user's selection operation, obtain the selected reference picture as the reference picture.

[0056] After the user selects a reference picture from the front-end page, in response to the user's selection operation, the server obtains this picture as the reference picture to extract the specific design needs of the user according to this reference picture.

[0057] S2004, extract the key features in the reference picture through the feature learning model to obtain the target keywords corresponding to the key features.

[0058] The key features include the size, type, content layout, color, etc. of the reference picture. The specific process of the feature learning model extracting the key features in the reference picture is described in detail in the following steps and will not be elaborated here. Users only need to select a picture from the front-end interface to very simply determine their own needs and transmit them to the server.

[0059] Of course, in other embodiments, it is also possible to combine the above multiple forms of input, such as inputting both the design requirement description text and selecting the reference picture to obtain the target keywords.

[0060] Return to Figure 2 , S202, screen the picture materials with similar keywords from the material library according to the target keywords to obtain the target picture materials.

[0061] In this embodiment, the material library includes a large number of pre-collected picture materials, and also records the keyword information corresponding to each picture material. The picture materials are screenshots of website pages or mobile APP pages. And the keywords corresponding to each picture material are obtained after the feature learning model extracts the key features in the picture material. The specific process of the feature learning model extracting the key features in the picture material is described in detail in the following steps and will not be elaborated here either. In addition, in the process of the feature learning model learning the key features of pictures through the large number of picture materials, a visual and semantic database is also formed. According to the keywords corresponding to the key features, a label can be formed, and then each label is used as a field and stored in the visual and semantic database.

[0062] According to the target keyword in the design requirement information, find the keywords similar to the target keyword from the visual and semantic database, and then obtain the picture materials corresponding to the similar keywords from the material library according to the mapping relationship between the picture materials and keywords recorded in the material library, which are the target picture materials corresponding to the design requirement information. Generally, there are multiple target picture materials.

[0063] In an alternative embodiment, after screening out the picture materials corresponding to the similar keywords, the screened picture materials can also be displayed to the user for selection. In response to the selection operation on the screened picture materials, the selected picture materials are used as the target picture materials.

[0064] S204, extract the picture key features of the target picture materials through a pre-trained feature learning model.

[0065] In this embodiment, adding multiple target picture materials as the input when generating the design drawing can make the output design drawing more accurate. For the target picture materials, it is also necessary to first analyze and extract the picture key features therein, that is, the keywords. The picture key features include picture size (width and height), picture type, picture content rules, color information, etc. Input the target picture materials into the pre-trained feature learning model, and analyze the picture size, picture type, picture content rules, and color information of the target picture materials to obtain the corresponding picture key features.

[0066] Among them, analyzing the picture type of the target picture materials includes: obtaining the page path corresponding to the target picture materials, or identifying the type identifier in the target picture materials; determining the picture type of the target picture materials according to the page path or the type identifier. In this embodiment, the target picture materials are screenshots of website pages or mobile APP pages. When intercepting the target picture materials, the page path of the page to which the target picture materials belong is also recorded. According to the page path, it can be known which layer of the website page or mobile APP page the target picture materials are in, belonging to the first-level page, second-level page, third-level page, etc. For example, the personal settings page belongs to the next-level page of the personal page. In addition, the page path may also include information such as the page name of the page, so that the page type can be obtained, such as the login page, personal page, detail page, system settings page, etc.

[0067] The rules for analyzing the picture content of the target picture material include: parsing the picture content of the target picture material to obtain the title information of the target picture material, the picture information included in the picture of the target picture material, and the layout information of each block in the target picture material. Additionally, for the case where the target picture material is a screenshot of a website page or an APP page, it may also include brand information, etc.

[0068] The rules for analyzing the color information of the target picture material include: identifying the main visual area of the target picture material, obtaining the color information that appears most frequently in the main visual area; obtaining the color information of the largest block in the main visual area; obtaining the color information of buttons or other small blocks (blocks other than the largest block and the buttons) in the target picture material.

[0069] Additionally, for the special application scenario where the target picture material is a screenshot of a website page or a mobile APP page, it is also possible to obtain the device information, version information, etc. when taking the corresponding page screenshot of the target picture material, because the displayed picture styles are different on different devices or on different versions of web pages or APP pages.

[0070] Although the target picture material has undergone a picture feature extraction through the feature learning model when it is collected into the material library, and the corresponding keyword information is recorded. However, since the time of the last feature extraction of the target picture material may be relatively long, or the result of a single extraction may not be completely accurate, therefore, in this embodiment, after obtaining the target picture material, the picture key features of the target picture material are extracted again through the feature learning model, which can ensure that the extraction result is more accurate.

[0071] It should be noted that the feature learning model in this embodiment is not a deep learning model, but a non-deep learning self-learning model. Compared with traditional deep learning models, the feature learning model is more efficient and can achieve self-learning through multiple picture materials and key features, improving the flexibility of generating designs.

[0072] The training process of the feature learning model includes: collecting multiple picture materials to obtain a training set for training the feature learning model. Training the AI model according to the training set to learn and identify the visual and semantic information related to design in the multiple picture materials, that is, the picture key features, including picture size, picture type, picture content rules, color information, etc., to obtain the feature learning model and generate the visual and semantic database.

[0073] S206, combine the picture key features and the design requirement information, and generate the final design drawing through a generative model.

[0074] After obtaining the key features of the picture and the design requirement information (corresponding keywords), an existing AIGC-related model can be used to automatically generate the design drawing of the website page or the mobile APP page required by the user. For example, the generative model can adopt the SD model. The design drawing may include the color matching, page layout, etc. required by the user to meet the user's expectations. The more keyword information there is and the more accurate the description is, the more accurate the generated content of the picture will be.

[0075] In this embodiment, not only the design requirement information is used as the model input, but the design requirement information is combined with the key features of the picture analyzed from multiple target picture materials, so that the model has more learning objects in the process of generating the design drawing and can generate a design drawing that more accurately meets the user's needs.

[0076] The design drawing generation method proposed in this embodiment does not directly generate the design drawing through AIGC based on the design requirement information (prompt words, design requirement description text, or reference picture) input by the user. Instead, after obtaining the design requirement information, the target keyword is first obtained, and then multiple similar target picture materials are quickly screened from the material library according to the target keyword. Then, the key features of the target picture materials are extracted by using a pre-trained feature learning model, and the key features of the picture are combined with the design requirement information input by the user and used as the input of the AIGC model to generate the corresponding design drawing. This can not only improve the design efficiency, but also make the final generated result more accurate and more in line with the user's needs. It can also introduce new innovative elements through a large number of picture materials and improve the diversity of the design.

[0077] Embodiment 2

[0078] As Figure 5 shown, it is a flowchart of a design drawing generation method proposed in the second embodiment of the present application. In the second embodiment, on the basis of the above first embodiment, the design drawing generation method further includes steps S308-S312. It can be understood that the flowchart in the embodiment of the present method is not used to limit the execution order of the steps. According to needs, some steps in this flowchart can also be added or deleted.

[0079] The method includes the following steps:

[0080] S300, obtain the target keyword from the design requirement information.

[0081] In this embodiment, various ways can be provided for the user to input design requirement information, including but not limited to directly inputting a prompt word or descriptive text of the design requirement, or providing the user with a variety of reference design drawings for the user to select a reference picture close to their own requirements.

[0082] In the case of inputting a prompt word, the server directly obtains the prompt word to get the target keyword and then proceeds with subsequent processing. In the case of inputting descriptive text of the design requirement, after the server obtains the descriptive text of the design requirement input by the user, it needs to extract the target keyword from the descriptive text of the design requirement and then proceed with subsequent processing. In the case of selecting a reference picture, it is necessary to extract the key features from the reference picture to obtain the target keyword corresponding to the key features.

[0083] S302, Screen out picture materials with similar keywords from the material library according to the target keyword to obtain target picture materials.

[0084] In this embodiment, the material library includes a large number of pre-collected picture materials, and also records the keyword information corresponding to each picture material. The picture materials are screenshots of website pages or mobile APP pages. In addition, during the process of the feature learning model learning the key features of pictures through the large number of picture materials, a visual and semantic database is also formed.

[0085] According to the target keyword in the design requirement information, search for keywords similar to the target keyword in the visual and semantic database, and then according to the mapping relationship between the picture materials and keywords recorded in the material library, obtain the picture materials corresponding to the similar keywords from the material library, which are the target picture materials corresponding to the design requirement information. The target picture materials are generally multiple pictures.

[0086] S304, Extract the key features of the target picture materials through a pre-trained feature learning model.

[0087] The key features of the pictures include picture size (width and height), picture type, picture content rules, color information, etc. Input the target picture materials into the pre-trained feature learning model, analyze the picture size, picture type, picture content rules, and color information of the target picture materials, and the corresponding key features of the pictures can be obtained.

[0088] S306, Combine the key features of the pictures and the design requirement information to generate a design drawing through a generative model.

[0089] After obtaining the key features of the picture and the design requirement information (corresponding keywords), an existing AIGC-related model can be used to automatically generate the design drawings of the website page or the mobile APP page required by the user. For example, the generative model can adopt the SD model. In this embodiment, not only the design requirement information is used as the model input, but the design requirement information is combined with the key features of the picture analyzed from multiple target picture materials, so that the model has more learning objects in the process of generating the design drawings and can more accurately generate the design drawings that meet the user's needs.

[0090] The implementation of the above steps S300 - S306 is the same as the implementation principle of steps S200 - S206 in the foregoing first embodiment. The specific implementation process can refer to the description in the first embodiment, and will not be elaborated herein in this embodiment of the present application.

[0091] S308, display the design drawing to the user for viewing.

[0092] Through the front-end interface of the client, the design drawing output by the model can be displayed to the user for viewing. If the design drawing meets the user's expectations, the process ends. If the user is not very satisfied with the design drawing and some adjustments or modifications are required, it can be iterated repeatedly according to the user's feedback until the user is satisfied.

[0093] S310, obtain the adjustment information input by the user for the design drawing.

[0094] In this embodiment, when the user is not very satisfied with the design drawing, the adjustment information for the design drawing can be input in the front-end interface. The adjustment information can be input through descriptive text or can be a reselected drop-down option, etc.

[0095] Taking the adjustment information as descriptive text as an example, as Figure 6 shown, it is a schematic diagram of a front-end interface for displaying the design drawing and inputting the adjustment information in an embodiment of the present application. In Figure 6 it, if the user is very satisfied with the design drawing displayed in the front-end interface, the "Download" button can be clicked to download the design drawing. If the user is not very satisfied with the design drawing displayed in the front-end interface, the text description of the content that needs to be adjusted can be input in the input area above the interface, and the "Update" button can be clicked. The server extracts keywords according to the text description to obtain the adjustment information for the design drawing. If the design drawing displayed in the front-end interface completely does not meet the user's needs, the "Reset" button can also be clicked to repeat all the above steps to regenerate the desired design drawing.

[0096] S312. According to the design drawing and the adjustment information, regenerate a new design drawing through the generative model.

[0097] After obtaining the adjustment information on the server side, it is possible to fine-tune based on the adjustment information on the basis of the previously generated design drawing, so as to generate a new design drawing that better meets the user's needs.

[0098] Specifically, in an alternative embodiment, it is possible to directly modify the corresponding keywords on the previously generated design drawing through the generative model according to the adjustment information to obtain the new design drawing.

[0099] In another alternative embodiment, it is also possible to first filter similar picture materials through the feature learning model according to the keywords in the previously generated design drawing and the adjustment information, and then combine the key picture features extracted from the picture materials with the adjustment information to generate the new design drawing through the generative model.

[0100] The design drawing generation method proposed in this embodiment can, after obtaining the design requirement information, first obtain the target keywords, then quickly filter multiple similar target picture materials from the material library according to the target keywords, and then use the pre-trained feature learning model to extract the key picture features of the target picture materials, and combine the key picture features and the design requirement information input by the user as the input of the AIGC model to generate the corresponding design drawing. This can not only improve the design efficiency, but also make the final generated result more accurate, better meet the user's needs, and can also introduce new innovative elements through a large number of picture materials to improve the design diversity. Moreover, this method allows the user to provide feedback and adjustment on the design drawing generated by the model. If the design drawing does not fully meet the user's needs, it is possible to fine-tune the design drawing by inputting adjustment information, and repeat the iteration until the user is satisfied, further improving the accuracy of the obtained design drawing and enhancing the user experience.

[0101] Embodiment III

[0102] As Figure 7 shown, the following is a schematic diagram of the hardware architecture of an electronic device 20 proposed in the third embodiment of the present application. In this embodiment, the electronic device 20 may include, but is not limited to, a memory 21, a processor 22, and a network interface 23 that are communicatively connected to each other through a system bus. It should be noted that Figure 7 Only the electronic device 20 with components 21-23 is shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented. In this embodiment, the electronic device 20 may be a server.

[0103] The memory 21 includes at least one type of readable storage medium, which includes flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disc, etc. In some embodiments, the memory 21 may be an internal storage unit of the electronic device 20, such as the hard disk or memory of the electronic device 20. In other embodiments, the memory 21 may also be an external storage device of the electronic device 20, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc. equipped on the electronic device 20. Of course, the memory 21 may also include both the internal storage unit and the external storage device of the electronic device 20. In this embodiment, the memory 21 is generally used to store the operating system and various application software installed in the electronic device 20, such as the program code of the design drawing generation system 60. In addition, the memory 21 may also be used to temporarily store various types of data that have been output or will be output.

[0104] In some embodiments, the processor 22 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips. The processor 22 is generally used to control the overall operation of the electronic device 20. In this embodiment, the processor 22 is used to run the program code stored in the memory 21 or process data, such as running the design drawing generation system 60, etc.

[0105] The network interface 23 may include a wireless network interface or a wired network interface, and the network interface 23 is generally used to establish a communication connection between the electronic device 20 and other electronic devices.

[0106] Embodiment Four

[0107] As Figure 8 shown, a schematic diagram of modules of a design drawing generation system 60 is proposed in the fourth embodiment of this application. The design drawing generation system 60 can be divided into one or more program modules, and one or more program modules are stored in a storage medium and executed by one or more processors to complete the embodiments of this application. The program modules referred to in the embodiments of this application refer to a series of computer program instruction segments that can complete specific functions. The following description will specifically introduce the functions of each program module in this embodiment.

[0108] In this embodiment, the design drawing generation system 60 includes:

[0109] An acquisition module 600, configured to acquire target keywords from design requirement information.

[0110] A screening module 602, configured to screen picture materials with similar keywords from a material library according to the target keywords to obtain target picture materials.

[0111] An analysis module 604, configured to extract picture key features of the target picture materials through a pre-trained feature learning model.

[0112] A generation module 606, configured to generate a design drawing through a generative model by combining the picture key features and the design requirement information.

[0113] For the specific implementation processes of the functions of the above modules, reference can be made to the descriptions in the first embodiment above, and details are not described herein again.

[0114] Embodiment Five

[0115] As Figure 9 shown, this is a schematic diagram of the modules of a design drawing generation system 60 proposed in the fifth embodiment of this application. In this embodiment, in addition to including the acquisition module 600, screening module 602, analysis module 604, and generation module 606 in the fourth embodiment, the design drawing generation system 60 further includes a display module 608 and an adjustment module 610.

[0116] The display module 608 is configured to display the design drawing to the user for viewing through a client.

[0117] The adjustment module 610 is configured to acquire adjustment information input by the user for the design drawing, and regenerate a new design drawing through a generative model according to the design drawing and the adjustment information.

[0118] For the specific implementation processes of the functions of the above modules, reference can be made to the descriptions in the second embodiment above, and details are not described herein again.

[0119] Embodiment Six

[0120] This application also provides another implementation manner, that is, to provide a computer-readable storage medium storing a design drawing generation program, and the design drawing generation program can be executed by at least one processor, so that the at least one processor executes the steps of the design drawing generation method as described above.

[0121] In this embodiment, the computer-readable storage medium includes flash memory, hard disks, multimedia cards, card-type memories (e.g., SD or DX memories, etc.), random access memories (RAM), static random access memories (SRAM), read-only memories (ROM), electrically erasable programmable read-only memories (EEPROM), programmable read-only memories (PROM), magnetic memories, magnetic disks, optical disks, etc. In some embodiments, the computer-readable storage medium may be an internal storage unit of a computer device, such as the hard disk or memory of the computer device. In other embodiments, the computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc., equipped on the computer device. Of course, the computer-readable storage medium may also include both the internal storage unit and the external storage device of the computer device. In this embodiment, the computer-readable storage medium is generally used to store the operating system installed on the computer device and various application software, such as the program code of the design drawing generation method in the embodiment. In addition, the computer-readable storage medium may also be used to temporarily store various data that have been output or are to be output.

[0122] It should be noted that in this article, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including that element.

[0123] The serial numbers of the embodiments of the present application above are only for description and do not represent the superiority or inferiority of the embodiments.

[0124] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the embodiments of the present application can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. Optionally, they can be implemented by program codes executable by the computing device, so that they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a different order from here, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module to be implemented. In this way, the embodiments of the present application are not limited to any specific combination of hardware and software.

[0125] The above are only the preferred embodiments of the embodiments of the present application, and do not limit the patent scope of the embodiments of the present application. Any equivalent structure or equivalent process transformation made by using the description and drawings of the embodiments of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the embodiments of the present application.

Claims

1. A method for generating a design drawing, characterized in that, The method includes: Obtaining target keywords from the design requirement information; Filtering image materials with similar keywords from the material library according to the target keywords to obtain target image materials; Extracting the key image features of the target image materials through a pre-trained feature learning model; Combining the key image features and the design requirement information, and generating a design drawing through a generative model.

2. The method for generating a design drawing according to claim 1, wherein The filtering of image materials with similar keywords from the material library according to the target keywords includes: Searching for keywords similar to the target keywords in the visual and semantic databases; According to the mapping relationship between the image materials recorded in the material library and the keywords, obtaining the image materials corresponding to the similar keywords from the material library.

3. The method for generating a design drawing according to claim 1, wherein The extracting of the key image features of the target image materials through a pre-trained feature learning model includes: Inputting the target image materials into the feature learning model, and analyzing at least one of the image size, image type, image content rules, and color information of the target image materials to obtain corresponding key image features.

4. The design drawing generation method according to claim 3, wherein: The analyzing of the image type of the target image materials includes: Obtaining the page path corresponding to the target image materials, or identifying the type identifier in the target image materials; Determining the image type of the target image materials according to the page path or the type identifier; The analyzing of the image content rules of the target image materials includes: Performing image content parsing on the target image materials to obtain the title information of the target image materials, the image information included in the picture in the target image materials, and the layout information of each block in the target image materials; The analyzing of the color information of the target image materials includes: Identifying the main visual area of the target image materials, and obtaining the color information that appears most frequently in the main visual area; Obtaining the color information of the largest block in the main visual area; Obtaining the color information of the buttons or other small blocks in the target image materials.

5. The method for generating a design drawing according to claim 1 or 3, characterized in that The feature learning model is a non-deep learning autonomous learning model, and the training process of the feature learning model includes: Collecting a plurality of image materials to obtain a training set for training the feature learning model; Training an artificial intelligence model according to the training set, learning and identifying the visual and semantic information related to design in the plurality of image materials to obtain the feature learning model, and generating a visual and semantic database; the visual and semantic information related to design includes at least one of the following: image size, image type, image content rules, color information.

6. The method for generating a design drawing according to claim 1, wherein, The obtaining of target keywords from the design requirement information includes: Obtaining the prompt words input by the user to obtain the target keywords; or Obtaining the design requirement description text input by the user, and extracting the target keywords from the design requirement description text.

7. The method for generating a design drawing according to claim 1 or 6, characterized in that, The obtaining of target keywords from the design requirement information includes: Providing a variety of reference pictures on the front-end interface for the user to select according to the design requirements; Responding to the user's selection operation, obtaining the selected reference picture as a reference picture; Extract the key features in the reference picture through the feature learning model to obtain the target keyword corresponding to the key features.

8. The design drawing generation method according to claim 1, wherein The method further includes: Display the design drawing to the user for viewing; Obtain the adjustment information input by the user for the design drawing; According to the design drawing and the adjustment information, regenerate a new design drawing through the generative model.

9. A design drawing generation system, characterized in that, The system includes: An acquisition module, configured to acquire a target keyword from design requirement information; A screening module, configured to screen picture materials with similar keywords from a material library according to the target keyword to obtain target picture materials; An analysis module, configured to extract the picture key features of the target picture materials through a pre-trained feature learning model; A generation module, configured to generate a design drawing through a generative model by combining the picture key features and the design requirement information.

10. An electronic device, characterized in that, The electronic device includes: a memory, a processor, and a design drawing generation program stored on the memory and executable on the processor. When the design drawing generation program is executed by the processor, it implements the design drawing generation method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, A design drawing generation program is stored on the computer-readable storage medium. When the design drawing generation program is executed by a processor, it implements the design drawing generation method according to any one of claims 1 to 8.

Images