An artificial intelligence-based picture generation method and system

Abstract

The application belongs to the technical field of image processing, and particularly relates to a picture generation method and system based on artificial intelligence, which comprises S1, an uploading stage; S2, a positioning and segmentation stage; S3, an instruction analysis and region binding stage; S4, a conflict detection and negotiation stage; and S5, a backup and encryption stage. Through cooperation of the above structures, multiple people can synchronously perform picture retouching, the picture is segmented into several parts in a segmented mask manner, then a person participating in picture retouching in a group selects a corresponding mask to perform picture retouching, compared with a traditional picture retouching manner, the picture retouching manner is more convenient and fast, and the picture generation efficiency is improved based on this. When multiple people edit the same region mask, a conflict classification and priority rule and a real-time interactive negotiation manner are adopted to solve the conflict, so that the multiple people can smoothly perform collaborative editing.

Description

Technical Field

[0001] This invention belongs to the field of image processing technology, specifically an image generation method and system based on artificial intelligence. Background Technology

[0002] With the continuous progress of society and the improvement of people's quality of life, taking photos to record daily life has become a widely popular lifestyle. After taking photos with a mobile phone, many users will use software to edit the photos according to their personal aesthetics and needs to enhance the artistic effect of the photos. One-click photo editing is a common function of existing photo editing software. Although one-click photo editing is efficient, its universal design cannot meet the different editing needs of everyone. Therefore, in real society, when processing group photos, due to different personal aesthetics, users will take turns using software to edit the photos according to their own needs, resulting in a long editing process and low image generation efficiency.

[0003] Therefore, the present invention provides an image generation method and system based on artificial intelligence. Summary of the Invention

[0004] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0005] The technical solution adopted by this invention to solve its technical problem is: an image generation method based on artificial intelligence, comprising the following steps:

[0006] S1. Upload stage: After standardizing the images to be processed, upload them in a format supported by the system, while saving the original images.

[0007] S2, Localization and Segmentation Stage: The model automatically identifies people, objects, and backgrounds in the image and generates an interactive mask;

[0008] S3, Command Parsing and Region Binding Stage: Bind the parsed text commands to the specific masks segmented by the S2 positioning and segmentation stage;

[0009] S4. Conflict Detection and Negotiation Phase: Resolve conflicts arising from multiple users modifying the same mask;

[0010] S5. Backup and Encryption Stage: Backup the generated images and verify identity information when storing the images.

[0011] Furthermore, the S2 localization and segmentation stage performs recognition operations using a semantic segmentation model, the model formula of which is:

[0012] M = SAM(I)

[0013] Where M is the set of segmentation masks, which are the masks of each region obtained after segmenting the input image. Each mask corresponds to an independent region in the image, including people, objects and background; SAM is the semantic segmentation model, which is used to perform segmentation operations on the input image; I is the original input image.

[0014] Furthermore, the S2 positioning and segmentation stage allows users to fine-tune the segmented area through clicks and text descriptions.

[0015] Furthermore, the instruction parsing in the S3 instruction parsing and region binding stage is achieved through multimodal instruction parsing and context association. The multimodal instruction parsing combines natural language processing models with object detection to identify region keywords and operation keywords in the instruction, and uses neural networks to parse position keywords in the instruction.

[0016] Furthermore, the implementation of the context association includes historical instruction analysis and image content association:

[0017] Historical command analysis involves recording the user's historical commands and combining them with the current command for contextual analysis.

[0018] Image content association is reflected in combining the image segmentation mask and target detection results to analyze the relationship between instructions and image content;

[0019] Historical instruction analysis uses semantic similarity calculation to find historical instructions related to the current instruction;

[0020] Image content association maintains continuous attention to the target region through target tracking technology, infers the attributes of the target region based on context, optimizes instruction parsing, and processes sequential instructions through a long short-term memory network to capture contextual dependencies.

[0021] Furthermore, the multimodal instruction parsing formula is as follows:

[0022] E = Bind(NLP(T), M)

[0023] Where E represents the editing instruction bound to a specific region, i.e., the operation instruction for a specific region of the image parsed from the user's instruction; Bind is a function used to bind the parsed text instruction to a specific region in the image; NLP represents the Natural Language Processing model, used to parse the natural language instruction T input by the user and extract the region keywords and operation keywords; T is the natural language instruction input by the user; M is the set of segmentation masks, i.e. the image region masks obtained by the semantic segmentation model.

[0024] Furthermore, the context association formula is as follows:

[0025] E* =ContextAssociation(T, H) * M)

[0026] Among them, E * This refers to the context-optimized editing instructions, which are the result of optimizing the original editing instructions based on historical instructions and the current image content.

[0027] ContextAssociation is a function used to combine the analysis results of the current instruction T and the historical instructions H. * And perform context association and instruction optimization with the segmentation mask set M;

[0028] T represents the natural language command currently input by the user;

[0029] H * This is the result of semantic analysis of historical instructions, containing key information and contextual relationships within the historical instructions;

[0030] M is a set of segmentation masks, i.e., region masks of the image.

[0031] Furthermore, the S4 conflict detection and negotiation phase uses the NSGA-II algorithm to balance the command priorities of different users. Its principle is as follows: superior individuals are selected through non-dominated sorting and crowding distance, thereby finding a balanced solution among multiple objectives. When processing the command priorities of different users, each user's command and its priority are considered as an objective. The NSGA-II algorithm finds a solution that balances the needs of different users. Its operation process includes:

[0032] Objective function: The objective function is to consider each user's instruction and its priority. Assume there are n users, each user i issues an instruction and its corresponding priority pi, and the decision variables are the degree of execution of each user's instruction or the proportion of resource allocation.

[0033] Encoding scheme: Individuals are represented by chromosomes, each individual is a vector, and the elements correspond to the degree of execution of each user instruction;

[0034] Fitness function: Define an objective function fi(x) for each user i to measure the effect of instruction execution or satisfaction, with the goal of maximizing the satisfaction of all users;

[0035] NSGA-II algorithm flow:

[0036] Initialization: Randomly generate an initial population, with each individual representing a possible instruction execution scheme;

[0037] Non-dominated ranking: Performing a non-dominated ranking on all individuals in the population to divide them into different non-dominated strata;

[0038] Crowding distance calculation: Calculate the crowding distance for each individual in each non-dominated layer;

[0039] Selection, crossover, and mutation: Generate the next generation of individuals through tournament selection, crossover, and mutation operations;

[0040] Iteration: Repeat the above steps until the maximum number of iterations is reached.

[0041] An AI-based image generation system, applied to the aforementioned AI-based image generation method, includes: an upload module, a processing module, a storage module, and an authentication module.

[0042] The upload module is used to upload the images to be processed to the system and provide real-time feedback on the upload progress and upload status.

[0043] The processing module is used to segment the image and perform collaborative modifications on the segmented image to generate a new image.

[0044] The storage module is used to store the generated output images, and at the same time record the relevant parameters and content of the modification and generation to form preference data;

[0045] The identity verification module is used to verify the user's identity and record all user operations, including login, logout, image upload, editing, and download storage.

[0046] Furthermore, the processing module includes a parameter setting unit and a feedback unit. The setting unit is used for users to set relevant parameters for image generation, while the feedback unit is used to evaluate the quality of the generated image and make modification suggestions. The basis for the modification suggestions includes sharpness, color deviation, and contrast.

[0047] The beneficial effects of this invention are as follows:

[0048] 1. Enables simultaneous image editing by multiple users. By using a segmentation mask, the image is divided into several parts. Then, each person involved in the editing of a group photo selects a corresponding mask to edit. Compared with the traditional method of taking turns editing, this is more convenient and faster, and improves the efficiency of image generation. At the same time, when multiple people edit the same area mask, conflict classification and priority rules, as well as real-time interactive negotiation, are used to resolve conflicts, ensuring the smooth progress of multi-person collaborative editing.

[0049] 2. A feedback unit is provided, which evaluates the quality of the generated image based on sharpness, color deviation, and contrast, and provides suggestions for improvement to enhance the artistic effect of the generated image. Attached Figure Description

[0050] The invention will now be further described with reference to the accompanying drawings.

[0051] Figure 1 This is a flowchart of an image generation method and system based on artificial intelligence according to the present invention;

[0052] Figure 2 This is a system structure diagram of an image generation method and system based on artificial intelligence according to the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] Please see Figure 1-2 This invention provides: an image generation method and system based on artificial intelligence; Example 1

[0055] An AI-based image generation method includes the following steps:

[0056] S1. Upload stage: After standardizing the images to be processed, upload them in a format supported by the system, while saving the original images.

[0057] S2, Localization and Segmentation Stage: The model automatically identifies people, objects, and backgrounds in the image and generates an interactive mask;

[0058] The S2 localization and segmentation stage uses a semantic segmentation model for identification. The model formula is as follows:

[0059] M = SAM(I)

[0060] Where M is the segmentation mask set, which is the mask of each region obtained after segmenting the input image. Each mask corresponds to an independent region in the image, which includes people, objects and background; SAM is the semantic segmentation model, which is a model used to perform segmentation operations on the input image; I is the original input image. Suppose the input image I is a group photo containing three people A, B, C and the background. After being processed by the semantic segmentation model SAM, the segmentation mask set M is obtained, which includes the mask M1 for person A, the mask M2 for person B, the mask M3 for person C and the background mask M4. These masks M1, M2, M3 and M4 together constitute the segmentation mask set M.

[0061] The S2 positioning and segmentation stage allows users to fine-tune the segmented area by clicking and using text descriptions. For example, if user 1 enters "Character A is wearing a Santa hat" and user 2 enters "Character A is wearing a birthday hat", the system will display a pop-up message: "Character A's hat style is conflicting. Please select to keep the Santa hat or the birthday hat."

[0062] S3, Command Parsing and Region Binding Stage: Bind the parsed text commands to the specific masks segmented by the S2 positioning and segmentation stage;

[0063] The instruction parsing in the S3 instruction parsing and region binding stage is achieved through multimodal instruction parsing and context association. Multimodal instruction parsing combines natural language processing models with object detection to identify region keywords and operation keywords in the instruction. It also uses neural networks to parse positional keywords in the instruction. For example, if a user inputs instruction T: "Remove the scar on person A's face".

[0064] First, the instruction is parsed using a Natural Language Processing (NLP) model to extract the region keyword "person A" and the operation keyword "scar removal". Then, the Bind function finds the corresponding mask MA-face in the segmentation mask set M based on the region keyword "person A" and binds the operation instruction "scar removal" to the mask to generate the editing instruction E, which is to perform scar removal operation on the person A region.

[0065] Context association can be implemented through historical instruction analysis and image content association.

[0066] Historical command analysis involves recording the user's historical commands and combining them with the current command for contextual analysis.

[0067] Image content association is reflected in combining the image segmentation mask and target detection results to analyze the relationship between instructions and image content;

[0068] Historical instruction analysis uses semantic similarity calculation to find historical instructions related to the current instruction;

[0069] Image content association maintains continuous attention to the target region through target tracking technology, infers the attributes of the target region based on context, optimizes instruction parsing, and processes sequential instructions through a long short-term memory network to capture contextual dependencies.

[0070] The multimodal instruction parsing formula is:

[0071] E = Bind(NLP(T), M)

[0072] Where E represents the editing instruction bound to a specific region, i.e., the operation instruction for a specific region of the image parsed from the user's instruction; Bind is a function used to bind the parsed text instruction to a specific region in the image; NLP represents the Natural Language Processing model, used to parse the natural language instruction T input by the user and extract the region keywords and operation keywords; T is the natural language instruction input by the user; M is the set of segmentation masks, i.e. the image region masks obtained by the semantic segmentation model.

[0073] The formula for contextual association is:

[0074] E * =ContextAssociation(T, H) * M)

[0075] Among them, E * This refers to the context-optimized editing instructions, which are the result of optimizing the original editing instructions based on historical instructions and the current image content.

[0076] ContextAssociation is a function used to combine the analysis results of the current instruction T and the historical instructions H. * And perform context association and instruction optimization with the segmentation mask set M;

[0077] T represents the natural language command currently input by the user;

[0078] H * This is the result of semantic analysis of historical instructions, containing key information and contextual relationships within the historical instructions;

[0079] M is a set of segmentation masks, i.e., region masks of the image;

[0080] For example, suppose the user previously entered the command "give character B a red short-sleeved shirt", and the current user enters the command T: "give character A a hat";

[0081] First, analyze the historical instructions to obtain H. * This includes information about clothing modifications made to character B; then, the function ContextAssociation combines the current instruction T and the analysis results H from the history instructions. * And the segmentation mask set M, generate optimized editing instructions E * This means that after considering historical instructions, when putting a hat on character A, it may automatically select a more matching hat style based on the previous clothing modification style of character B, or ensure that the editing of character A will not conflict with the editing of character B.

[0082] S4. Conflict Detection and Negotiation Phase: Resolve conflicts arising from multiple users modifying the same mask;

[0083]

[0084] The S4 conflict detection and negotiation phase uses the NSGA-II algorithm to balance the command priorities of different users. Its principle is as follows: superior individuals are selected through non-dominated ordering and crowding distance, thereby finding a balanced solution among multiple objectives. When processing the command priorities of different users, each user's command and its priority are considered as an objective. The NSGA-II algorithm finds a solution that balances the needs of different users. Its operation process includes:

[0085] Objective function: The objective function is to consider each user's instruction and its priority. Assume there are n users, each user i issues an instruction and its corresponding priority pi, and the decision variables are the degree of execution of each user's instruction or the proportion of resource allocation.

[0086] Encoding scheme: Individuals are represented by chromosomes, each individual is a vector, and the elements correspond to the degree of execution of each user instruction;

[0087] Fitness function: Define an objective function fi(x) for each user i to measure the effect of instruction execution or satisfaction, with the goal of maximizing the satisfaction of all users;

[0088] NSGA-II algorithm flow:

[0089] Initialization: Randomly generate an initial population, with each individual representing a possible instruction execution scheme;

[0090] Non-dominated ranking: Performing a non-dominated ranking on all individuals in the population to divide them into different non-dominated strata;

[0091] Crowding distance calculation: Calculate the crowding distance for each individual in each non-dominated layer;

[0092] Selection, crossover, and mutation: Generate the next generation of individuals through tournament selection, crossover, and mutation operations;

[0093] Iteration: Repeat the above steps until the maximum number of iterations is reached.

[0094] For example, suppose there are three users A, B, and C who each issue an instruction to modify a group photo. The priorities of each instruction are pA=3, pB=2, and pC=1, respectively. Our goal is to maximize overall satisfaction by adjusting the execution level of each instruction XA, XB, and XC.

[0095] Objective function:

[0096] fA(x)=3·XA

[0097] fB(x)=2·XB

[0098] fC(x)=1·XC

[0099] Constraints: Assume total resources are finite, such as XA+XB+XC≤1, and xi≥0;

[0100] NSGA-II algorithm:

[0101] Initialization: Randomly generate the initial population:

[0102]

[0103] Non-dominated sorting:

[0104] Calculate the objective function value for each individual:

[0105]

[0106] Crowding distance calculation: For individuals in F1, calculate the distance to each target and sum them;

[0107] Environmental selection: Individuals are selected to enter the next generation according to the non-dominant layer order and crowding distance, such as through crossover [0.45, 0.35, 0.2].

[0108] Genetic manipulation: crossover and mutation are performed on selected individuals to generate new individuals;

[0109] Iteration: A set of equilibrium solutions is obtained: x = [0.5, 0.3, 0.2], which means that under resource constraints, the execution degree of instructions for users A, B, and C is 0.5, 0.3, and 0.2 respectively, satisfying the priority of each user as much as possible.

[0110] S5. Backup and Encryption Stage: Backup the generated images and verify identity information when storing the images.

[0111] Example 2

[0112] This embodiment provides:

[0113] The S4 conflict detection and negotiation phase also includes version management and collaborative rollback, supporting version control and history rollback during multi-user collaboration. This is achieved through a Git-style version tree: each commit generates a version node, supporting branching and merging. For example:

[0114] Version V1.0 (Initial Group Photo)

[0115] ├─Branch A: User 1 "Change Character B's Hairstyle" → V1.1 │ └─User 2 "Add Snowflakes to the Background" → V1.2

[0116] └─Branch B: User 3 "Character C with glasses" → V1.3 └─Merge Branch A → V2.0 (Final Group Photo);

[0117] In summary, the processing procedures for S2, S3, and S4 are as follows:

[0118]

[0119]

[0120] Example 3

[0121] This embodiment provides:

[0122] The S2 localization and segmentation stage also includes voice commands. For example, suppose we have a voice message "Hello, how are you?", the following is the speech-to-text process:

[0123] Signal preprocessing: Remove background noise from the speech and determine the start and end points of the speech.

[0124] Feature extraction: Calculate the MFCC features of the speech signal.

[0125] Acoustic modeling: Using a DNN to map MFCC features to phoneme sequences, such as "HELLO, HO WARE YO U?".

[0126] Language modeling: Using an LSTM language model to predict word sequences corresponding to phoneme sequences.

[0127] Decoding: Find the most likely word sequence "Hel lo, how are you?" through cluster search.

[0128] Through the above steps, the voice signal is successfully converted into text. Then, text commands can be bound to specific masks segmented in the S2 positioning and segmentation stage to meet the image editing needs of special users.

[0129] Example 4

[0130] This embodiment provides:

[0131] An AI-based image generation system, applied to an AI-based image generation method, includes: an upload module, a processing module, a storage module, and an authentication module.

[0132] The upload module is used to upload images to be processed to the system and provide real-time feedback on the upload progress and status.

[0133] The processing module is used to segment images and perform collaborative modifications on the segmented images to generate new images.

[0134] The storage module is used to store the output images and record the relevant parameters and content of the modifications and generation to form preference data.

[0135] The authentication module is used to verify user identity and record all user operations, including login, logout, image upload, editing, and download storage.

[0136] The processing module includes a parameter setting unit and a feedback unit. The setting unit is used by the user to set relevant parameters for image generation, while the feedback unit is used to evaluate the quality of the generated image and provide modification suggestions. The modification suggestions are based on factors such as sharpness, color deviation, and contrast.

[0137] It is important to note that insufficient image sharpness is a concern. If the image sharpness is low, it is recommended to adjust the parameters in the image generation process, such as increasing the number of iterations or adjusting the learning rate of the optimization algorithm.

[0138] Color deviation: If the image has color deviation, it is recommended to adjust the color balance parameters or use a color correction algorithm to correct it;

[0139] Poor contrast: If the image has insufficient contrast, it is recommended to adjust the contrast enhancement parameters or use histogram equalization to enhance the contrast.

[0140] For example, assuming the generated image has a blur problem, the feedback unit calculates the image gradient and finds that the sharpness is low. The system will then propose the following modification suggestions:

[0141] Adjust generation parameters: Increase the number of iterations in the image generation process from 100 to 200 to improve image clarity.

[0142] The terms “center,” “longitudinal,” “lateral,” “front,” “rear,” “left,” “right,” “vertical,” “horizontal,” “top,” “bottom,” “inner,” and “outer,” etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are merely simplified descriptions for the convenience of describing the present invention and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting the scope of protection of the present invention.

[0143] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0144] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An image generation method based on artificial intelligence, characterized in that, Includes the following steps: S1. Upload stage: After standardizing the images to be processed, upload them in a format supported by the system, while saving the original images. S2. Localization and Segmentation Stage: The model automatically identifies people, objects, and background in the image and generates an interactive mask. The model formula is as follows: ； Where M is the set of segmentation masks, which are the masks for each region obtained after segmenting the input image. Each mask corresponds to an independent region in the image, including people, objects and background; SAM is the semantic segmentation model, which is used to perform segmentation operations on the input image; I is the original input image. S3, Command Parsing and Region Binding Stage: The user inputs a natural language command, and command parsing is achieved through multimodal command parsing and context association. Multimodal command parsing combines a natural language processing model with object detection to identify region keywords and operation keywords in the command, and uses a neural network to parse position keywords in the command. The multimodal instruction parsing formula is: ； in, Editing instructions are bound to a specific area, that is, operation instructions for a specific area of ​​the image parsed from user instructions; It is a function used to bind parsed text instructions to a specific region in an image; This represents a natural language processing model used to parse natural language instructions input by the user. Extract the regional keywords and operational keywords; The natural language instructions input by the user; M is the set of segmentation masks, i.e., the image region masks obtained previously through the semantic segmentation model; Context association can be implemented through historical instruction analysis and image content association. Historical command analysis involves recording the user's historical commands and combining them with the current command for contextual analysis. Image content association is reflected in combining the image segmentation mask and target detection results to analyze the relationship between instructions and image content; Historical instruction analysis uses semantic similarity calculation to find historical instructions related to the current instruction; Image content association maintains continuous attention to the target region through target tracking technology, infers the attributes of the target region based on context, optimizes instruction parsing, and processes sequential instructions through a long short-term memory network to capture contextual dependencies. The formula for contextual association is: ； in, This refers to the context-optimized editing instructions, which are the result of optimizing the original editing instructions based on historical instructions and the current image content. It is a function used to combine the current instruction. Historical command analysis results and segmentation mask set Perform context association and instruction optimization; Natural language commands input by the current user; This is the result of semantic analysis of historical instructions, containing key information and contextual relationships within the historical instructions; This is a set of segmentation masks, i.e., region masks of the image; S4. Conflict Detection and Negotiation Phase: Based on the parsed and optimized editing instructions, multiple users modify the corresponding mask areas in the segmented image. When conflicts arise due to multiple users modifying the same mask, a conflict resolution mechanism is used. The algorithm balances the command priorities of different users. Its principle is to select superior individuals through non-dominated sorting and crowding distance, thereby finding a balanced solution among multiple objectives. When processing the command priorities of different users, each user's command and its priority are treated as an objective. The algorithm finds a set of solutions that balance the needs of different users, and its operation process includes: Objective function: Consider each user's instructions and their priorities as the objective function. Assume there are... Each user Instructions and their corresponding priorities The decision variables are the degree of execution of each user instruction or the proportion of resources allocated; Encoding scheme: Individuals are represented by chromosomes, each individual is a vector, and the elements correspond to the degree of execution of each user instruction; Fitness function: Define the objective function for each user i. To measure the effectiveness or satisfaction of command execution, with the goal of maximizing the satisfaction of all users; Algorithm flow: Initialization: Randomly generate an initial population, with each individual representing a possible instruction execution scheme; Non-dominated ranking: Performing a non-dominated ranking on all individuals in the population to divide them into different non-dominated strata; Crowding distance calculation: Calculate the crowding distance for each individual in each non-dominated layer; Selection, crossover, and mutation: Generate the next generation of individuals through tournament selection, crossover, and mutation operations; Iteration: Repeat the above steps until the maximum number of iterations is reached; S5. Backup and Encryption Stage: Backup the generated images and verify identity information when storing the images.

2. The image generation method based on artificial intelligence according to claim 1, characterized in that: The S2 positioning and segmentation stage allows users to fine-tune the segmented area through clicks and text descriptions.

3. An artificial intelligence-based image generation system, applied to the artificial intelligence-based image generation method as described in any one of claims 1-2, characterized in that, include: Upload module, processing module, storage module, and authentication module. The upload module is used to upload the images to be processed to the system and provide real-time feedback on the upload progress and upload status. The processing module is used to segment the image and perform collaborative modifications on the segmented image to generate a new image. The storage module is used to store the generated output images, and at the same time record the relevant parameters and content of the modification and generation to form preference data; The identity verification module is used to verify the user's identity and record all user operations, including login, logout, image upload, editing, and download storage.

4. The image generation system based on artificial intelligence according to claim 3, characterized in that: The processing module includes a parameter setting unit and a feedback unit. The setting unit is used for users to set relevant parameters for image generation, while the feedback unit is used to evaluate the quality of the generated image and provide modification suggestions. The modification suggestions are based on factors such as sharpness, color deviation, and contrast.

Images