AI text graph wallpaper generation processing method and device based on intelligent terminal, and terminal

Abstract

The invention discloses an AI text drawing wallpaper generation processing method and device based on an intelligent terminal, and a terminal, and relates to the technical field of intelligent terminals, and the method comprises the steps: obtaining a wallpaper generation drawing request described by user voice; identifying the wallpaper generation drawing request described by the user voice to obtain wallpaper generation drawing request text information; generating drawing request text information for the filtered wallpaper, and converting the drawing request text information into cue word text information which can be more understood by a text generation graph model through a cue word enhancement model; the prompt word text information is input into a preset text graph model for automatic drawing, and a drawing graph is generated; and checking the generated drawing graph according to a preset safety compliance requirement, and displaying the drawing graph when the checking is controlled. According to the method, the operation process of user-defined wallpaper is simplified, and the updating frequency and the individuation degree of the wallpaper content are improved, so that the user satisfaction and the overall operation efficiency of the system are improved, and convenience is provided for use of the user.

Description

Technical Field

[0001] The present invention relates to the field of smart terminal technology, and in particular to a method and device for generating and processing AI-generated wallpaper based on a smart terminal, a smart terminal, and a non-temporary computer-readable storage medium. Background Art

[0002] With the development of science and technology and the continuous improvement of people's living standards, the use of various smart terminals such as smart TVs is becoming more and more popular.

[0003] In the current smart TV wallpaper market, technological innovation has led to the introduction of ultra-thin, seamlessly attached wallpaper TVs. These products function not only as display devices but also as electronic picture frames for displaying artwork or personal photos, significantly enhancing the aesthetic value of the home environment. However, existing wallpaper TV technology has significant limitations, particularly the limited nature of wallpaper content and infrequent updates, which restrict the ability to personalize the user experience.

[0004] Existing methods for obtaining wallpapers for smart TVs primarily rely on purchasing wallpapers or collaborating with content providers, resulting in slow updates and content lags. Furthermore, companies face high costs when purchasing wallpaper licenses and must go through a tedious comparison and negotiation process. Furthermore, the process for users to customize wallpapers for existing smart TVs is equally complex, typically requiring multiple steps, such as searching for images, uploading them to the device, and setting the wallpaper, which is time-consuming and laborious.

[0005] Therefore, the existing technology still needs to be improved and developed. Summary of the Invention

[0006] The technical problem to be solved by the present invention is that, in response to the above-mentioned defects of the prior art, a method, device, smart terminal and storage medium for generating and processing AI-generated wallpapers based on smart terminals are provided. The present invention simplifies the operation process of users customizing wallpapers, and improves the update frequency and personalization of wallpaper content, thereby improving user satisfaction and the overall operating efficiency of the system, and providing convenience for users.

[0007] The technical solutions adopted by the present invention to solve the problem are as follows:

[0008] This application provides an AI-based Wenshengtu wallpaper generation and processing method based on a smart terminal, and the technical solution is as follows: obtaining a wallpaper generation and painting request described by the user's voice;

[0009] Recognizing the wallpaper generation drawing request described by the user's voice to obtain text information of the wallpaper generation drawing request;

[0010] Generating painting request text information for the wallpaper and filtering out harmful element information;

[0011] Generating drawing request text information for the filtered wallpaper, and converting it into prompt word text information that can be better understood by the text-based image model through the prompt word enhancement model;

[0012] The prompt word text information converted by the prompt word enhancement model is input into the preset text-based graph model for automatic drawing to generate a drawing graphic;

[0013] The generated drawing graphics are reviewed according to the preset safety compliance requirements and displayed when they pass the review.

[0014] The method for generating and processing AI-generated wallpaper based on a smart terminal, wherein the step of obtaining a wallpaper generation request based on a user's voice description includes:

[0015] The voice module on the remote control obtains the wallpaper described by the user's voice and generates a drawing request;

[0016] And send a wallpaper generation and painting request to the smart TV end based on the user's voice description.

[0017] The method for generating and processing AI-generated wallpaper based on a smart terminal, wherein before the step of identifying the wallpaper generation drawing request described by the user's voice and obtaining the text information of the wallpaper generation drawing request, includes:

[0018] Speech recognition is performed on the wallpaper generation painting request described by the user's voice to obtain text information of the wallpaper generation painting request.

[0019] The method for generating and processing AI-generated wallpaper based on a smart terminal, wherein the steps of generating painting request text information for the wallpaper and filtering out undesirable element information include:

[0020] The wallpaper generation painting request text information is detected to determine whether it contains undesirable elements of preset keywords; the undesirable element information is filtered to obtain filtered wallpaper generation painting request text information.

[0021] The method for generating and processing AI-based Wensheng picture wallpapers based on smart terminals, wherein the step of generating drawing request text information for the filtered wallpaper and converting it into prompt word text information that is more understandable to the Wensheng picture model through the prompt word enhancement model includes:

[0022] Receive user wallpaper generation request text message after filtering out harmful elements;

[0023] Through the built-in prompt word enhancement model, semantic analysis and keyword extraction are performed on the filtered wallpaper generation drawing request text information to create prompt word text information that conforms to the understanding of the text-based image model;

[0024] The prompt word text information that conforms to the understanding of the text-graph model is structured and generated to be more in line with the input format of the text-graph model.

[0025] The method for generating AI-powered wallpaper based on a smart terminal, wherein the prompt word text information converted by the prompt word enhancement model is input into a preset Wensheng picture model for automatic drawing, and the step of generating a drawing graphic includes:

[0026] Obtaining prompt word text information converted by the prompt word enhancement model;

[0027] Input the prompt word text information obtained after transformation by the prompt word enhancement model into the text-generated image model, start the image generation process and automatically draw the image;

[0028] The text-based graph model uses its internal algorithm to perform image synthesis, interpret the intention in the prompt word text information, and generate a painting image that meets the user's description.

[0029] The method for generating and processing AI-based Wensheng picture wallpaper based on a smart terminal, wherein the step of reviewing the generated painting graphics according to preset security compliance requirements and controlling the display when the review passes includes:

[0030] Get the generated drawing graphics;

[0031] The generated drawings are reviewed according to preset security and compliance requirements, and the graphic content is analyzed using automated image recognition technology and machine learning algorithms;

[0032] When approved, the control will generate the drawing graphic ready for display.

[0033] An AI-based wallpaper generation and processing device for a smart terminal, wherein the device comprises:

[0034] A drawing request acquisition module is used to obtain the wallpaper described by the user's voice and generate a drawing request;

[0035] a drawing request recognition module, configured to recognize the wallpaper drawing request generated by the user's voice description and obtain text information of the wallpaper drawing request;

[0036] A filtering module, configured to generate drawing request text information for the wallpaper and filter out harmful elements;

[0037] A prompt word generation module is used to generate a drawing request text message for the filtered wallpaper, and convert it into prompt word text information that can be better understood by the text-based image model through a prompt word enhancement model;

[0038] An automatic drawing processing module is used to input the prompt word text information converted by the prompt word enhancement model into a preset text-based graph model for automatic drawing to generate a drawing graphic;

[0039] The review and display control module is used to review the generated drawing graphics according to the preset safety compliance requirements, and display them when the review passes the control.

[0040] An intelligent terminal includes a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors, including the method for executing any one of the methods described above.

[0041] A non-transitory computer-readable storage medium, wherein when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform any one of the methods described above.

[0042] From the above, it can be seen that the present application provides an AI Wenshengtu wallpaper generation and processing method based on a smart terminal, a device, a smart terminal and a non-temporary computer-readable storage medium. The present invention proposes an innovative engineering process, which aims to optimize the entire wallpaper content update and personalization process by introducing advanced algorithms and models. This process can significantly reduce the cost of wallpaper authorization, simplify the user's custom wallpaper operation process, and increase the update frequency and personalization of wallpaper content, thereby improving user satisfaction and the overall operating efficiency of the system. Through this process, users will be able to enjoy richer, more personalized and cost-effective wallpaper services, while enterprises will also be able to improve the efficiency and flexibility of their wallpaper content management. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 This is a flow chart of the AI Wensheng picture wallpaper generation and processing method based on the smart terminal provided by an embodiment of the present invention.

[0045] Figure 2 It is a flowchart of an AI-powered wallpaper generation and processing method based on a smart terminal provided by a specific application embodiment of the present invention.

[0046] Figure 3This is a schematic diagram of an interface in which a user of the AI Wensheng picture wallpaper generation and processing method based on a smart terminal provides a specific application embodiment of the present invention and describes a painting request to the TV through a voice button on a remote control.

[0047] Figure 4 This is a schematic diagram of an interface where a television receives a user's request and starts processing painting information according to the method for generating and processing AI Wensheng picture wallpaper based on a smart terminal provided by a specific application embodiment of the present invention.

[0048] Figure 5 It is a schematic diagram of generating a painting graphic of the AI Wensheng picture wallpaper generation processing method based on the smart terminal provided by the specific application embodiment of the present invention.

[0049] Figure 6 This is a principle block diagram of an AI-powered wallpaper generation and processing device based on a smart terminal provided by an embodiment of the present invention.

[0050] Figure 7 This is a block diagram of the internal structure of the smart terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solutions and advantages of the present invention more clear and distinct, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0052] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the various components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indications will also change accordingly.

[0053] With the widespread popularity of smart devices, users' demand for personalized, high-definition wallpapers is increasing, especially their preference for dynamic wallpapers and AI-based customized wallpapers.

[0054] However, existing methods for obtaining wallpapers for smart TVs primarily rely on purchasing wallpapers or collaborating with content providers, resulting in slow updates and content lags. Furthermore, companies face high costs when purchasing wallpaper licenses and must go through a tedious comparison and negotiation process. Furthermore, the process for users to customize wallpapers for smart TVs is equally complex, typically requiring multiple steps, such as searching for images, uploading them to the device, and setting the wallpaper, which is time-consuming and laborious.

[0055] To address these issues, this paper proposes an innovative engineering process designed to optimize the entire wallpaper content update and customization process by introducing advanced algorithms and models. This process significantly reduces the cost of wallpaper licensing, simplifies the user-defined wallpaper process, and increases the frequency and personalization of wallpaper content updates, thereby improving user satisfaction and the overall operational efficiency of the system. Through this process, users will be able to enjoy a richer, more personalized, and more cost-effective wallpaper service, while businesses will also be able to improve the efficiency and flexibility of their wallpaper content management.

[0056] The present invention can realize 4K drawing, has a fast and easy-to-use process, and can automatically help users enhance prompt words.

[0057] Therefore, if Figure 1 As shown, this application proposes an AI-based wallpaper generation and processing method for a smart terminal, comprising the following steps:

[0058] S100, obtaining a wallpaper described by a user's voice and generating a drawing request;

[0059] The wallpaper generation drawing request described by the user's voice is a wallpaper image requirement that needs to be generated as described verbally by the user, such as "help me draw an impressionist landscape painting."

[0060] Specifically, wallpaper generation requests are generated through voice input. For example, users interact with the system through voice, describing the style, color, and theme of the wallpaper they want. This facilitates the subsequent conversion of the user's voice into text. Key information, such as color, shape, style, and theme, is extracted from the text.

[0061] This allows users to customize wallpapers based on their preferences, enhancing their sense of participation and satisfaction. Furthermore, voice input is usually faster and more convenient than text input, especially on mobile devices, where users can express their ideas more naturally.

[0062] S200: Recognize the wallpaper generation drawing request described by the user voice, and obtain text information of the wallpaper generation drawing request;

[0063] This step performs voice recognition on the wallpaper generation drawing request described by the user's voice to obtain text information of the wallpaper generation drawing request. That is, the present invention converts the description input by the user through voice into text information for subsequent processing and wallpaper generation.

[0064] Among them, for user voice input, for example, users express their creativity in natural language through a microphone, such as: "I want a wallpaper of blue ocean and white sand beach."

[0065] The present invention can convert the user's voice in real time through voice recognition technology (ASR), analyze the sound signal into readable text content, and can recognize various accents and intonations through voice recognition technology.

[0066] Then, the output text is generated. Specifically, after speech recognition processing, the present invention generates corresponding text, such as "blue ocean" and "white beach", and structures it into wallpaper generation request information. The obtained text information is then passed to the wallpaper generation module for further image generation and processing.

[0067] Compared to existing technologies, this invention allows users to express their needs in a natural way, reducing the psychological burden and discomfort of input and improving the interactive experience. Furthermore, the system can quickly analyze and convert speech into text, making the entire process more efficient than manual input and providing instant feedback. Furthermore, modern speech recognition technology can effectively reduce information input errors caused by spelling errors or typing difficulties, improving the accuracy of wallpaper generation requests.

[0068] S300, generating drawing request text information for the wallpaper and filtering out harmful element information;

[0069] In this embodiment, this step refers to reviewing and processing the text information provided by the user before generating the wallpaper to remove potential inappropriate elements.

[0070] In specific implementation, the text information of the wallpaper generation request received by the system includes the user's description, such as content, theme, color, etc. Then, the system performs filtering and identification of bad elements.

[0071] Regarding the identification and filtering of objectionable elements, the present invention can use specific algorithms and dictionaries to analyze text information and identify potentially sensitive words, inappropriate content, or elements that do not conform to ethical standards. Once objectionable elements are identified, the present invention will process them. This may include completely deleting the objectionable content or prompting the user to modify their input.

[0072] Then a compliance request is generated. That is, in the embodiment of the present invention, the filtered text information will be organized into a compliance wallpaper generation request to ensure that the content is safe and suitable for release.

[0073] For example, when a user describes their desired wallpaper, they say, "I want a dark art style painting with ferocious animals and war scenes." The present invention identifies undesirable elements: the system detects that the word "war scene" does not meet the set content standards. It then performs filtering and notifies the user through the system.

[0074] The user can then modify the request according to the prompts, and the final description may be: "I want a natural scene with ferocious animals, using a dark art style." After the filtered request is processed, the present invention can generate a compliant wallpaper pattern in the subsequent steps.

[0075] Through the above process, not only the appropriateness of the generated content is ensured, but also friendly feedback is provided to users to help them better express their creativity.

[0076] Compared to existing solutions, which lack the ability to filter out harmful elements, this invention effectively prevents the creation of harmful or inappropriate wallpapers by filtering out harmful elements, protecting users from negative content and ensuring content security. Furthermore, this filtering mechanism ensures that user creation requests are more aligned with social and ethical standards, enhancing user trust in the system and improving the overall user experience.

[0077] S400: Generate painting request text information for the filtered wallpaper, and convert it into prompt word text information that can be better understood by the text-based image model through the prompt word enhancement model;

[0078] In the embodiment of this step, the filtered wallpaper generation request information is first obtained: that is, this step first receives the user wallpaper generation request text information after filtering out the bad elements, such as the user's modified description: "I want a natural scene with ferocious animals, using a dark art style."

[0079] This is then transformed using an enhanced prompt word model. The present invention uses a built-in prompt word enhancement module to perform semantic analysis and keyword extraction on the user's text description to create prompt word text information that more accurately conforms to the understanding of the text graph model. This module can use natural language processing (NLP) technology, and in specific implementations, word vector models (such as Word2Vec or GloVe) can be used to extract keywords and phrases.

[0080] This is then converted into prompt text: During the enhancement process, the system structures the user's description to generate text that better matches the input format of the text-based image model. For example, combining and reorganizing keywords like "ferocious animals," "natural scenes," and "dark art style" might generate a prompt text like "dark art style natural scenes, including ferocious animals."

[0081] The final request is then generated: The enhanced prompt text information is integrated into the final wallpaper generation request, which is then passed to the text-based image model in the subsequent steps. This process ensures that the generated information is more specific and easier for the AI model to understand, making it more accurate when generating images.

[0082] Of course, in the embodiment of the present invention, verification and feedback can be further performed. That is, before the newly generated prompt word text information is submitted to the text graph model, preliminary verification can be performed to ensure that no information is lost, and the final prompt information is displayed to the user, and modification suggestions are provided if necessary.

[0083] For example, when the user's final wallpaper generation request is: "I want a wallpaper that includes a freely flying eagle and open natural scenery." The embodiment of the present invention records the user's description: "Includes a freely flying eagle and open natural scenery" and filters out harmful elements (in this example, there is no harmful content).

[0084] The enhanced prompt word model then extracts the keywords "free-flying eagle" and "open natural scenery." A word embedding model analyzes the contextual relationship between "freedom" and "nature," enhancing their description. This is then converted into prompt word text information, generating the enhanced prompt word text information: "free-flying eagle in the vast natural scenery."

[0085] And generate the final request, the enhanced prompt word text is integrated and submitted to the text-based graph model for image generation, so that the generated wallpaper can accurately reflect the user's expectations.

[0086] Through the above steps, not only the understanding of user needs by the cultural graph model is optimized, but also the interaction and creative realization between users and technology are enhanced.

[0087] Compared with the existing technology, traditional solutions have basically not achieved the conversion of prompt words through the prompt word enhancement model, and cannot better meet user needs. However, through the prompt word enhancement of the present invention, the system can generate more accurate images that meet user expectations, greatly improving the quality of wallpapers and user satisfaction, and improving the generation quality. In addition, the present invention can better understand user intentions through the model, promote the diversity and creativity of AI-generated results, and meet the needs of different users. Moreover, through automated prompt word enhancement, manual intervention is reduced, making the entire wallpaper generation process more efficient. In addition, the present invention can also improve user participation, because the user's adjustment of the description and real-time feedback in the process can continuously improve the final result, allowing users to feel more involved in the creation process.

[0088] S500, inputting the prompt word text information converted by the prompt word enhancement model into a preset text-based graph model for automatic drawing to generate a drawing graphic;

[0089] The core of this step is to pass the user's text information, which has been enhanced with prompt words, to the text-generated graph model, so that it automatically generates a graphic work that matches the text description. This process automates the creation, reduces manual intervention, and improves generation efficiency.

[0090] In specific implementation, the step S500 specifically includes:

[0091] S501. Obtaining prompt word text information after conversion by the prompt word enhancement model;

[0092] Regarding the prompt text: Before this step begins, the text information obtained by the prompt enhancement model includes the user's creative description, which has been subjected to keyword extraction and semantic optimization. For example, after enhancement, the user's original input description becomes a more structured prompt, such as "beach at sunset, and seagulls flying in the sky."

[0093] S502. The prompt word text information obtained by the prompt word enhancement model after conversion is input into the Wensheng graph model to start the image generation process for automatic drawing;

[0094] The text-based image model in this embodiment of the present invention is a generative model based on deep learning. After training, it can generate corresponding images based on input text information. Specifically, the system passes the prompt word text information generated in the previous step to this model, starting the image generation process and automatically drawing the image.

[0095] S503: The text-based graph model uses its internal algorithm (such as GANs or variational autoencoders) to perform image synthesis, interpret the intention in the prompt word text information, and generate a painting image that meets the user's description.

[0096] In this embodiment of the present invention, after receiving the prompt text, the text-based graph model uses its internal algorithms (such as GANs or variational autoencoders) to synthesize an image. The model interprets the intent of the text and generates a drawing that matches the user's description, such as "a golden beach under the setting sun, with a few seagulls flying in the sky."

[0097] And output the result, that is, in the embodiment of the present invention, after automatic drawing processing, the system will generate an image as the final result and return it to the user. This image will directly reflect the user's previous description and embody the artistic style and theme expressed.

[0098] For example, when the enhanced prompt word text information submitted by the user is: "On a summer morning, the sun shines through the leaves on the path, and someone is jogging." After being processed by the prompt word enhancement model of the present invention, the text information is organized into: "Summer morning", "sunlight shines on the path", and "runner".

[0099] Then enter the text graph model: input this final prompt word text information into the text graph model.

[0100] Automatically generated drawing: The Wensheng graph model automatically generates a drawing based on the description, showing a sunny morning scene with someone running on a green path.

[0101] Output: The generated image will be displayed and the user can review it and choose whether to make further adjustments or keep it intact.

[0102] By following these steps, users can not only quickly create drawings that meet their needs, but also experience the joy of efficiency and convenience during the creation process. Compared to traditional art creation methods, this technology clearly provides a wider range of creative possibilities and convenience.

[0103] As can be seen, compared to traditional manual drawing or using simple graphics generation tools, this invention integrates advanced deep learning technology to make the image generation process more efficient and faster, and has higher creative expression. Traditional methods usually require users to have their own drawing skills, but this solution effectively lowers the threshold for creation, allowing any user to easily obtain high-quality visual works. In addition, this invention has the following advantages:

[0104] 1) Improved drawing efficiency: The automated drawing process significantly reduces creation time, eliminating the need for users to manually draw each element.

[0105] 2) It is conducive to the realization of user creativity: the user's creativity is more fully demonstrated through the model, and the system can generate vivid images without being limited by personal drawing ability.

[0106] 3) Seamless interaction is possible: the process from text description to graphic generation is smoothly connected, and users can quickly get feedback and make adjustments.

[0107] S600: Review the generated drawing graphics according to the preset safety compliance requirements, and display them when they pass the review.

[0108] In this embodiment, the generated drawing is reviewed to ensure that its content does not contain inappropriate, sensitive, or illegal elements. Once approved, the image is displayed to the user. This mechanism protects users from inappropriate content and ensures compliance with laws and regulations.

[0109] Specifically, the step S600 includes:

[0110] S601. Get the generated drawing graphics;

[0111] The generated drawing in this step is the one generated in the previous step. The text information is processed by the text-based graph model to generate the corresponding image, for example, the graphic work "Runner on a path with sunlight filtering through leaves" requested by the user.

[0112] S602. Review the generated drawings according to pre-set security and compliance requirements, using automated image recognition technology and machine learning algorithms to analyze the graphic content.

[0113] Regarding the audit mechanism in this invention, the generated image is submitted to the security compliance audit module, which will check the image according to the preset standards. The audit standards include:

[0114] Ensure that the image does not infringe copyright or portrait rights of others;

[0115] Check for inappropriate symbols or signs;

[0116] Implementation: Audits can use automated image recognition technology and machine learning algorithms to analyze image content. Combined with image classification and annotation models, the system can quickly assess whether an image meets set standards.

[0117] S603. When the review is passed, the control will generate a drawing ready for display;

[0118] In the embodiment of the present invention, once the examination is passed, the system will be ready to display the generated drawing graphics. At this moment, the system will automatically start the display module so that the user can view the final image effect.

[0119] If the image fails the review, the present invention will return the review result to the user, indicating the content that may need to be adjusted, so that the user can regenerate or modify the image description.

[0120] In this way, compared with the traditional content generation method, which usually requires users to review and screen suitable images themselves, this solution significantly improves the efficiency and objectivity of content review by introducing an automatic review mechanism, thereby reducing the burden on users.

[0121] For example, if a user generates a text-generated image of a woman quietly reading a book in a park, the system generates the image and prepares it for review. Review for security compliance requirements: The review module automatically analyzes the image to determine if there are any potential compliance issues, such as whether it contains inappropriate content or copyright violations.

[0122] Audit Result: When the system passes the audit, it confirms that the image meets all compliance requirements.

[0123] Display: After the review is passed, the system will control the image display module to pop up and display the work, allowing users to view and save it.

[0124] Through the above steps, this technical solution effectively ensures the legality and appropriateness of the generated graphics, allowing users to use the generated content with peace of mind. In addition, the automatic review mechanism improves the security and efficiency of content generation.

[0125] As can be seen, the embodiments of the present invention ensure that generated images comply with social and legal standards through automated review, protecting the legitimate rights and interests of users and platforms and ensuring content compliance. Furthermore, they can guarantee the security and compliance of generated content, enhance user trust in the system, and improve the user experience. Furthermore, through early review, legal liability arising from inappropriate content is reduced, avoiding unnecessary disputes.

[0126] The present application further proposes that a wallpaper generation and painting request based on the user's voice description can be obtained through a voice module on a remote control; and the wallpaper generation and painting request based on the user's voice description can be sent to a smart TV.

[0127] For example, users can describe their drawing requests to the TV using the voice button on the remote control. This ensures a quick and easy-to-use process.

[0128] The present application further proposes to detect the wallpaper generation painting request text information to determine whether it contains undesirable elements of preset keywords; filter the undesirable element information to obtain filtered wallpaper generation painting request text information.

[0129] In this embodiment of the present invention, after receiving a user request, the television determines whether the user's text contains any objectionable elements. If so, the television filters out the objectionable elements and obtains the filtered text information for wallpaper generation. This ensures that the request information used to generate the image is free of objectionable elements.

[0130] like Figure 2 As shown, a specific application embodiment of the present invention provides an AI Wensheng picture wallpaper generation and processing method based on a smart terminal, comprising the following steps:

[0131] S11: Platform as a Service starts, obtains the drawing request described by the user's voice, and enters S12;

[0132] S12: Determine whether the service is busy. If not, proceed to S13. If yes, proceed to S19 and then to S13.

[0133] S13: Perform text review on the drawing request described by the user's voice. If the review passes, proceed to S14; if not, proceed to S19;

[0134] S14: prompt word enhancement and enter S15;

[0135] That is, in the embodiment of the present invention, after receiving the user's request, the TV will determine whether the text described by the user conforms to undesirable elements such as ethics and morality, and will enhance the prompt words if the description is civilized;

[0136] S15: Perform a cultural image. If image review is required, proceed to S17; otherwise, proceed to step S16.

[0137] S16: upload the picture and proceed to S18;

[0138] S17: Determine whether the image has passed the review. If it has passed, proceed to S18. If it has not passed, proceed to S19.

[0139] S18: Perform data synchronization; then enter S19;

[0140] S19: result returned;

[0141] In a specific embodiment of the present invention, the user can describe the drawing request to the TV through the voice button on the remote control, such as Figure 3 As shown in FIG, the user says "help me draw an impressionist landscape painting"; then Figure 4 As shown, the TV receives the user's request, starts processing the drawing information, and displays "Drawing quickly, you can leave the current page, and the drawing will be pushed to you after it is completed." Then as Figure 5 As shown, 3) generate a drawing graphic.

[0142] After receiving a user request, the television in this invention determines whether the user's description contains ethical or other negative elements. If the description is civilized, it will enhance the prompt word. This enhancement helps the text-based graph model quickly draw the image. After the image is generated, the image is immediately uploaded and reviewed in parallel. After the image passes the review, it is presented to the user.

[0143] It can be seen that the present invention can improve the execution efficiency and generation quality of AI Wensheng pictures, and solve the problems of low wallpaper update efficiency and unstable AI Wensheng picture process in the wallpaper industry of major TV manufacturers.

[0144] Exemplary devices

[0145] like Figure 6 As shown in , an embodiment of the present invention provides an AI Wensheng picture wallpaper generation and processing device based on a smart terminal, the device comprising:

[0146] A drawing request acquisition module 310 is used to acquire the wallpaper described by the user's voice and generate a drawing request;

[0147] A drawing request recognition module 320 is configured to recognize the wallpaper drawing request generated by the user's voice description and obtain text information of the wallpaper drawing request;

[0148] A filtering module 330 is used to generate drawing request text information for the wallpaper and filter out harmful elements;

[0149] A prompt word generation module 340 is used to generate a drawing request text message for the filtered wallpaper, and convert it into prompt word text information that can be better understood by the text-based image model through a prompt word enhancement model;

[0150] The automatic drawing processing module 350 is used to input the prompt word text information converted by the prompt word enhancement model into a preset text-based graph model for automatic drawing to generate a drawing graphic;

[0151] The review and display control module 360 is used to review the generated drawing graphics according to the preset safety compliance requirements, and display them when the review passes, as described above.

[0152] Based on the above embodiment, the present invention also provides an intelligent terminal, whose principle block diagram can be shown as follows: Figure 7 As shown. The smart terminal can be a smart TV, including a processor, memory, network interface, display screen, and database connected through a system bus. Among them, the processor of the smart terminal is used to provide computing and control capabilities. The memory of the smart terminal includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the smart terminal is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it realizes a method for generating and processing AI Wensheng picture wallpaper based on the smart terminal. The database of the smart terminal is used to store an AI Wensheng picture wallpaper generation processing program based on the smart terminal.

[0153] Those skilled in the art will understand that Figure 7 The principle block diagram shown in the figure is only a block diagram of a partial structure related to the solution of the present invention and does not constitute a limitation on the smart terminal to which the solution of the present invention is applied. The specific smart terminal may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0154] In one embodiment, a smart terminal is provided, comprising a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors to implement a control function of AI Wenshengtu wallpaper generation processing based on the smart terminal.

[0155] Among them, the memory refers to a hardware component used to store executable program code, which can be implemented by a flash memory chip or a solid-state hard drive. Its function is to save program files containing AI Wenshengtu wallpaper generation processing.

[0156] Among them, the processor refers to a computing unit that executes program instructions, which can be implemented by a multi-core mobile processor. Its function is to dynamically monitor the AI ​​wenshengtu wallpaper generation processing function according to the program code.

[0157] Among them, the program refers to a set of codes containing computer instructions, which can be implemented by combining Android system services with application layer API calls. Its function is to monitor the execution of the AI Wenshengtu wallpaper generation processing program.

[0158] Specifically, after the program is loaded by the processor, the steps of the AI Wensheng picture wallpaper generation processing method based on the smart terminal are executed.

[0159] The present application further proposes a non-transitory computer-readable storage medium. When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform a method for generating AI-powered wallpaper based on a smart terminal. The method includes:

[0160] Get the wallpaper described by the user's voice and generate a drawing request;

[0161] Recognizing the wallpaper generation drawing request described by the user's voice to obtain text information of the wallpaper generation drawing request;

[0162] Generating painting request text information for the wallpaper and filtering out harmful element information;

[0163] Generating drawing request text information for the filtered wallpaper, and converting it into prompt word text information that can be better understood by the text-based image model through the prompt word enhancement model;

[0164] The prompt word text information converted by the prompt word enhancement model is input into the preset text-based graph model for automatic drawing to generate a drawing graphic;

[0165] The generated drawing graphics are reviewed according to the preset safety compliance requirements and displayed when they pass the review.

[0166] The method for generating and processing AI-generated wallpaper based on a smart terminal, wherein the step of obtaining a wallpaper generation request based on a user's voice description includes:

[0167] The voice module on the remote control obtains the wallpaper described by the user's voice and generates a drawing request;

[0168] And send a wallpaper generation and painting request to the smart TV end based on the user's voice description.

[0169] The method for generating and processing AI-generated wallpaper based on a smart terminal, wherein before the step of identifying the wallpaper generation drawing request described by the user's voice and obtaining the text information of the wallpaper generation drawing request, includes:

[0170] Speech recognition is performed on the wallpaper generation painting request described by the user's voice to obtain text information of the wallpaper generation painting request.

[0171] The method for generating and processing AI-generated wallpaper based on a smart terminal, wherein the steps of generating painting request text information for the wallpaper and filtering out undesirable element information include:

[0172] The wallpaper generation painting request text information is detected to determine whether it contains undesirable elements of preset keywords; the undesirable element information is filtered to obtain filtered wallpaper generation painting request text information.

[0173] The method for generating and processing AI-based Wensheng picture wallpapers based on smart terminals, wherein the step of generating drawing request text information for the filtered wallpaper and converting it into prompt word text information that is more understandable to the Wensheng picture model through the prompt word enhancement model includes:

[0174] Receive user wallpaper generation request text message after filtering out harmful elements;

[0175] Through the built-in prompt word enhancement model, semantic analysis and keyword extraction are performed on the filtered wallpaper generation drawing request text information to create prompt word text information that conforms to the understanding of the text-based image model;

[0176] The prompt word text information that conforms to the understanding of the text-graph model is structured and generated to be more in line with the input format of the text-graph model.

[0177] The method for generating AI-powered wallpaper based on a smart terminal, wherein the prompt word text information converted by the prompt word enhancement model is input into a preset Wensheng picture model for automatic drawing, and the step of generating a drawing graphic includes:

[0178] Obtaining prompt word text information converted by the prompt word enhancement model;

[0179] Input the prompt word text information obtained after transformation by the prompt word enhancement model into the text-generated image model, start the image generation process and automatically draw the image;

[0180] The text-based graph model uses its internal algorithm to perform image synthesis, interpret the intention in the prompt word text information, and generate a painting image that meets the user's description.

[0181] The method for generating and processing AI-based Wensheng picture wallpaper based on a smart terminal, wherein the step of reviewing the generated painting graphics according to preset security compliance requirements and controlling the display when the review passes includes:

[0182] Get the generated drawing graphics;

[0183] The generated drawings are reviewed according to preset security and compliance requirements, and the graphic content is analyzed using automated image recognition technology and machine learning algorithms;

[0184] When approved, the control will generate the drawing graphic ready for display.

[0185] Those skilled in the art will appreciate that all or part of the processes in the above-described embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-described embodiments. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAM bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0186] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, various modifications and variations of the present application are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for generating and processing AI-generated wallpaper based on an intelligent terminal, characterized in that: include: Get the wallpaper described by the user's voice and generate a drawing request; Recognizing the wallpaper generation drawing request described by the user's voice to obtain text information of the wallpaper generation drawing request; Generating painting request text information for the wallpaper and filtering out harmful element information; Generating drawing request text information for the filtered wallpaper, and converting it into prompt word text information that can be better understood by the text-based image model through the prompt word enhancement model; The prompt word text information converted by the prompt word enhancement model is input into the preset text-based graph model for automatic drawing to generate a drawing graphic; The generated drawing graphics are reviewed according to the preset safety compliance requirements and displayed when they pass the review.

2. The AI-based wallpaper generation and processing method for intelligent terminals according to claim 1 is characterized in that: The step of obtaining the wallpaper described by the user's voice and generating a drawing request includes: The voice module on the remote control obtains the wallpaper described by the user's voice and generates a drawing request; And send a wallpaper generation and painting request to the smart TV end based on the user's voice description.

3. The AI-based wallpaper generation and processing method based on a smart terminal according to claim 1 is characterized in that: Before the step of recognizing the wallpaper generation drawing request described by the user voice and obtaining the wallpaper generation drawing request text information, the following steps are included: Speech recognition is performed on the wallpaper generation painting request described by the user's voice to obtain text information of the wallpaper generation painting request.

4. The AI-based wallpaper generation and processing method for intelligent terminals according to claim 1 is characterized in that: The steps of generating drawing request text information for the wallpaper and filtering out harmful element information include: The wallpaper generation painting request text information is detected to determine whether it contains undesirable elements of preset keywords; the undesirable element information is filtered to obtain filtered wallpaper generation painting request text information.

5. The AI-based wallpaper generation and processing method for intelligent terminals according to claim 1 is characterized in that: The step of generating a painting request text message for the filtered wallpaper and converting the message into prompt word text information that can be better understood by the text-based image model through the prompt word enhancement model includes: Receive user wallpaper generation request text message after filtering out harmful elements; Through the built-in prompt word enhancement model, semantic analysis and keyword extraction are performed on the filtered wallpaper generation drawing request text information to create prompt word text information that conforms to the understanding of the text-based image model; The prompt word text information that conforms to the understanding of the text-graph model is structured and generated to be more in line with the input format of the text-graph model.

6. The AI-based wallpaper generation and processing method for intelligent terminals according to claim 1 is characterized in that: The step of inputting the prompt word text information converted by the prompt word enhancement model into a preset text-based graph model for automatic drawing to generate a drawing graph includes: Obtaining prompt word text information converted by the prompt word enhancement model; Input the prompt word text information obtained after transformation by the prompt word enhancement model into the text-generated image model, start the image generation process and automatically draw the image; The text-based graph model uses its internal algorithm to perform image synthesis, interpret the intention in the prompt word text information, and generate a painting image that meets the user's description.

7. The AI-based wallpaper generation and processing method for intelligent terminals according to claim 1 is characterized in that: The step of reviewing the generated drawing graphics according to the preset safety compliance requirements and displaying the graphics after passing the review includes: Get the generated drawing graphics; The generated drawings are reviewed according to preset security and compliance requirements, and the graphic content is analyzed using automated image recognition technology and machine learning algorithms; When approved, the control will generate the drawing graphic ready for display.

8. An AI-based wallpaper generation and processing device based on a smart terminal, characterized in that: The device comprises: A drawing request acquisition module is used to obtain the wallpaper described by the user's voice and generate a drawing request; a drawing request recognition module, configured to recognize the wallpaper drawing request generated by the user's voice description and obtain text information of the wallpaper drawing request; A filtering module, configured to generate drawing request text information for the wallpaper and filter out harmful elements; A prompt word generation module is used to generate a drawing request text message for the filtered wallpaper, and convert it into prompt word text information that can be better understood by the text-based image model through a prompt word enhancement model; An automatic drawing processing module is used to input the prompt word text information converted by the prompt word enhancement model into a preset text-based graph model for automatic drawing to generate a drawing graphic; The review and display control module is used to review the generated drawing graphics according to the preset safety compliance requirements, and display them when the review passes the control.

9. An intelligent terminal, characterized in that: The device comprises a memory and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by one or more processors, and the one or more programs include being used to execute the method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method according to any one of claims 1 to 7.

Images