Financial image generation method and device based on workflow arrangement, equipment and medium

Abstract

The application relates to the technical field of image generation, in particular to a financial image generation method and device based on workflow arrangement, equipment and medium, the method comprises the following steps: acquiring a work node in a workflow arrangement interface, constructing an AI financial image task flow according to the work node, and generating hierarchical parameters; based on the hierarchical parameters and the AI financial image task flow, an initial AI financial image is generated, and iterative editing is performed to generate an edited AI financial image; real-time financial data is acquired, fusion processing is performed according to a visualization rule, and an AI financial image after data fusion is obtained; the AI financial image after fusion is subjected to compliance approval, if the AI financial image meets the condition of passing the compliance approval, a plurality of distribution channels are determined, and the AI financial image that passes the compliance approval is published through the plurality of distribution channels. The application has the effect of solving the problems of low image content control precision, difficult dynamic data integration and lack of production process collaborative management in the application of financial image generation.

Description

Technical Field

[0001] This application relates to the field of image generation technology, and in particular to a method, apparatus, device and medium for generating financial images based on workflow orchestration. Background Technology

[0002] Currently, AI-powered image generation technology is widely used to create visual content based on user intent. In the financial services industry, this technology is primarily used to rapidly generate marketing materials, such as financial product posters, social media promotional images, and explanatory illustrations for online events, aiming to improve the agility of visual content production and enrich the visual elements of customer communication.

[0003] Current mainstream AI-generated image generation technologies are typically based on deep learning models such as diffusion models. These models analyze user-input text prompts and utilize massive datasets learned by the model to gradually synthesize images. While such technologies show great potential for general image generation, they lack the fine-grained control over image content when applied to financial image production, which demands high precision, timeliness, and strict compliance. Users cannot directly adjust the specific details of key visual elements such as brand logos and compliance text within the image. Furthermore, the generated image content is often separate from real-time dynamic financial data, affecting the timeliness and accuracy of financial image information. The lack of integration across multiple processes means that the entire lifecycle cannot be covered, thus leaving room for improvement. Summary of the Invention

[0004] To address the common problems in financial image generation applications, such as low precision in image content control, difficulty in dynamic data integration, and lack of collaborative management of production processes, this application provides a financial image generation method, apparatus, device, and medium based on workflow orchestration.

[0005] The above-mentioned objective of this application is achieved through the following technical solution: A workflow-based financial image generation method, comprising: Obtain the user-defined work nodes in the workflow orchestration interface, construct an AI financial image task flow based on the work nodes, and generate layered parameters, which are used to generate AI fused images; Based on the hierarchical parameters and the AI ​​financial image task flow, an initial AI financial image is generated, and the initial AI financial image is iteratively edited to generate an edited AI financial image; Real-time financial data is acquired, and according to visualization rules, the real-time financial data is fused with the edited AI financial image to obtain the fused AI financial image. The AI ​​financial image after data fusion undergoes compliance approval. If the AI ​​financial image meets the conditions for compliance approval, multiple distribution channels are determined, and the AI ​​financial image that has passed compliance approval is published through the multiple distribution channels.

[0006] By adopting the above technical solutions, and by acquiring user-defined work nodes, constructing an AI financial image task flow, and generating hierarchical parameters for AI-fused images, the financial image generation process can be flexibly customized and parameterized. This meets the diverse needs of different financial business scenarios for image content and style. By generating initial AI financial images based on hierarchical parameters and the AI ​​financial image task flow, and then iteratively editing them, it ensures that the initial AI-generated images initially meet preset requirements and provides the ability to optimize and adjust image details, thereby improving the quality of the final image and user satisfaction. This reduces the uncertainty of relying entirely on AI black-box output. By acquiring real-time financial data and performing fusion processing according to visualization rules, the latest financial data can be dynamically integrated into the image content, thereby enhancing the timeliness and accuracy of financial image information. This allows the image to truly reflect the current market situation or business data. By conducting compliance approval of the AI ​​financial images after data fusion and determining the channels for release after approval, it ensures that the final released financial images meet regulatory and brand requirements and achieve effective distribution to the target audience. This reduces the compliance risks of financial institutions, enhances the consistency of brand image, and expands the scope and influence of information dissemination.

[0007] In a preferred example, this application can be further configured as follows: obtaining user-defined work nodes in the workflow orchestration interface, constructing an AI financial image task flow based on the work nodes, and generating hierarchical parameters, specifically including: In the workflow orchestration interface, multiple user-specified work nodes are obtained by clicking or dragging, and the work nodes are connected. The AI ​​financial image task flow is constructed based on the connection relationships and corresponding functional attributes of the working nodes; The hierarchical parameters are generated in response to the AI ​​financial image task flow.

[0008] By adopting the above technical solution, multiple user-specified work nodes can be acquired and connected through click or drag-and-drop operations in the workflow orchestration interface. This allows for an intuitive and convenient definition of the initial structure of the image generation process, thereby lowering the technical threshold for constructing complex AI image generation tasks. By constructing AI financial image task flows based on the connection relationships and functional attributes of work nodes, user-defined nodes and their configurations can be transformed into executable automated task sequences. This achieves standardization and automation of the AI ​​financial image generation process, improving production efficiency and process reliability. By generating hierarchical parameters in response to the AI ​​financial image task flow, structured parameters for controlling image generation can be automatically or assisted for the user based on the specific needs of the task flow. This ensures that subsequent image generation has clear guidance and improves the efficiency and accuracy of parameter configuration.

[0009] In a preferred embodiment, this application can be further configured such that: in response to the AI ​​financial image task flow, generating the hierarchical parameters specifically includes: Obtain the set of defined parameters input by the user on the interactive interface, and identify the basic layer parameters, the business layer parameters, and the financial compliance layer parameters from the set of defined parameters; The basic layer parameters, the business layer parameters, and the financial compliance layer parameters are pre-formatted and structured to generate the layered parameters.

[0010] By adopting the above technical solution, and by acquiring the set of defined parameters input by the user on the interactive interface and identifying the parameters of the basic layer, business layer, and financial compliance layer, it is possible to accurately extract and classify control parameters of different levels from the diverse input of the user. This enables effective parsing of the user's intent, making subsequent parameter processing more targeted. By performing pre-formatted and structured processing on the parameters of the basic layer, business layer, and financial compliance layer to generate layered parameters, the original, possibly non-standardized, parameters of each layer can be transformed into a standardized and structured instruction set that the image generation model can directly understand and execute. This improves the efficiency of parameter utilization and the accuracy of the generated results of the image generation model.

[0011] In a preferred embodiment, this application can be further configured as follows: Based on the hierarchical parameters and the AI ​​financial image task flow, an initial AI financial image is generated, and the initial AI financial image is iteratively edited to generate an edited AI financial image, specifically including: The initial AI financial image is obtained by using the text-based image model in the AI ​​financial image task flow and the layered parameters. Determine whether there are any local redraw nodes in the AI ​​financial image task flow; If the local redraw node exists in the AI ​​financial image task flow, then the region modification information of the selected area of ​​the initial AI financial image by the user is obtained, and the initial AI financial image is iteratively edited according to the region modification information to obtain the edited AI financial image.

[0012] By adopting the above technical solution, the initial AI financial image is obtained through the Wensheng image model in the AI ​​financial image task flow using layered parameters. Based on the refined layered parameters, the AI ​​model can be directly driven to generate a preliminary financial image that meets the design intent, thereby improving the quality of the initial image and its consistency with the expected goal. By judging whether there are local redrawing nodes in the task flow, and obtaining the user's regional modification information of the selected area of ​​the initial AI financial image for iterative editing when they exist, it can provide conditional editing capabilities for fine-tuning the target area and correcting the content based on the initial AI financial image. This greatly enhances the controllability and flexibility of the image content, enabling users to efficiently handle local defects in the AI-generated image or achieve specific creative fine-tuning, significantly improving the satisfaction and compliance of the final image.

[0013] In a preferred embodiment, this application can be further configured as follows: the acquisition of real-time financial data, and the fusion processing of the real-time financial data with the edited AI financial image according to visualization rules, to obtain a data-fused AI financial image, specifically includes: The real-time financial data is obtained based on multiple preset financial data source interfaces; According to the visualization rules, the mapping relationship between data and image elements is obtained. Based on the mapping relationship, the real-time financial data is mapped onto the elements in the edited AI financial image to generate the data-fused AI financial image.

[0014] By adopting the above technical solution and acquiring real-time financial data through multiple preset financial data source interfaces, the stability and timeliness of the financial data source used for image content fusion can be ensured. This provides a reliable data foundation for generating accurate financial images that reflect the latest market conditions. By obtaining the mapping relationship between data and image elements according to visualization rules, and mapping real-time financial data onto elements in the AI-edited financial image based on this mapping relationship, the automated, rule-based dynamic binding and content updating of real-time financial data and image visual elements can be achieved. This allows the financial image to dynamically display the latest data, greatly improving the information value and timeliness of the image and meeting the strict requirements of timeliness for financial information dissemination.

[0015] In a preferred embodiment, this application can be further configured as follows: The AI ​​financial image after data fusion undergoes compliance review; if the AI ​​financial image meets the conditions for compliance review, multiple distribution channels are determined, and the compliant AI financial image is published through these multiple distribution channels. Specifically, this includes: Based on financial compliance rules, a preliminary compliance screening is performed on the AI ​​financial image after data fusion to obtain the preliminary compliance screening results. Based on the preliminary compliance screening results, the AI ​​financial images that pass the preliminary compliance screening are fused and sent to designated personnel for multi-level manual review. The AI ​​financial images that pass the multi-level manual review are then used as AI financial images that have passed compliance approval. Obtain the channel distribution strategy from the AI ​​financial image task flow, and determine the multiple distribution channels based on the channel distribution strategy; Based on the multiple distribution channels, the AI ​​financial images that have passed the compliance approval are adapted to obtain multiple adapted AI financial images. The adapted AI financial images are pushed and published online to the corresponding multiple distribution channels.

[0016] By adopting the above technical solution, and performing preliminary compliance screening on the AI ​​financial images after data fusion based on financial compliance rules, automated preliminary compliance checks can be performed on the generated financial images. This improves the initial efficiency of compliance review, enabling early detection and alerting of potential compliance risks, and reducing the pressure of subsequent manual review. Based on the preliminary compliance screening results, images that pass the screening are sent to designated personnel for multi-level manual review. Images that pass the review are then used as AI financial images that have passed compliance approval. This combines the efficiency of AI preliminary review with the rigor of professional human review, ensuring the authority and accuracy of the approval results. This improves overall approval efficiency while guaranteeing the final compliance of the financial image content, meeting the stringent risk control standards of the financial industry. By obtaining channel distribution strategies from the AI ​​financial image task flow and determining multiple distribution channels accordingly, it is possible to... It enables the strategic and automated determination of distribution channels, ensuring that images are accurately placed on appropriate target channels according to preset plans, improving the efficiency and target orientation of distribution work. By adapting compliant and approved AI financial images to multiple distribution channels, multiple adapted AI financial images are obtained. The specifications and styles of the images can be automatically adjusted according to the specific display requirements of different distribution channels, thereby ensuring the best display effect and user experience of the images on each channel and reducing the tedious work of manual adaptation. By pushing and publishing multiple adapted AI financial images to the corresponding multiple distribution channels, the final image content can be automatically and one-stop published to multiple target channels, which greatly accelerates the speed of information dissemination, improves operational efficiency, and ensures that financial information can reach the target audience in a timely and effective manner.

[0017] The second objective of this invention is achieved through the following technical solution: A workflow-based financial image generation device, comprising: The workflow construction and parameter generation module is used to obtain the user-defined work nodes in the workflow orchestration interface, construct the AI ​​financial image task flow based on the work nodes, and generate hierarchical parameters, which are used to generate AI fused images. The image generation and editing module is used to generate an initial AI financial image based on the layered parameters and the AI ​​financial image task flow, and to iteratively edit the initial AI financial image to generate an edited AI financial image. The dynamic data fusion module is used to acquire real-time financial data and, according to visualization rules, fuse the real-time financial data with the edited AI financial image to obtain the fused AI financial image. The compliance approval and intelligent distribution module is used to conduct compliance approval on the AI ​​financial image after data fusion. If the AI ​​financial image meets the conditions for compliance approval, multiple distribution channels are determined and the AI ​​financial image that has passed compliance approval is published through the multiple distribution channels.

[0018] By adopting the above technical solution, and by acquiring user-defined work nodes, constructing an AI financial image task flow, and generating hierarchical parameters for AI-fused images, flexible customization and parameterized control of the financial image generation process can be achieved. This satisfies the diverse needs of different financial business scenarios for image content and style. By generating initial AI financial images based on the hierarchical parameters and AI financial image task flow, and then iteratively editing them, it ensures that the initial AI-generated images initially meet preset requirements and provides the ability to optimize and adjust image details, thereby improving the quality of the final image and user satisfaction. This reduces the uncertainty of relying entirely on AI black-box output. By acquiring real-time financial data and performing fusion processing according to visualization rules, the latest financial data can be dynamically integrated into the image content, thereby enhancing the timeliness and accuracy of financial image information. This allows the image to truly reflect the current market situation or business data. By conducting compliance approval of the AI ​​financial images after data fusion and determining the channels for release after approval, it ensures that the final released financial images meet regulatory and brand requirements and achieve effective distribution to the target audience. This reduces the compliance risks of financial institutions, enhances the consistency of brand image, and expands the scope and influence of information dissemination.

[0019] The above-mentioned objective three of this application is achieved through the following technical solution: A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the above-described workflow-based financial image generation method.

[0020] The fourth objective of this application is achieved through the following technical solution: A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described workflow orchestration-based financial image generation method.

[0021] In summary, this application includes at least one of the following beneficial technical effects: 1. By acquiring user-defined work nodes, constructing an AI financial image task flow, and generating hierarchical parameters for AI-fused images, the financial image generation process can be flexibly customized and parameterized, thereby meeting the diverse needs of different financial business scenarios for image content and style. By generating initial AI financial images based on the hierarchical parameters and AI financial image task flow and iteratively editing them, it can ensure that the initial AI-generated images initially meet the preset requirements and provide the ability to optimize and adjust image details, thereby improving the quality of the final image and user satisfaction, reducing the uncertainty of relying entirely on AI black-box output. By acquiring real-time financial data and performing fusion processing according to visualization rules, the latest financial data can be dynamically integrated into the image content, thereby enhancing the timeliness and accuracy of financial image information, enabling the image to truly reflect the current market situation or business data. By conducting compliance approval on the AI ​​financial images after data fusion and determining the channels for release after approval, it can ensure that the final released financial images meet regulatory and brand requirements and achieve effective distribution to the target audience, thereby reducing the compliance risks of financial institutions, enhancing the consistency of brand image, and expanding the scope and influence of information dissemination. 2. By conducting preliminary compliance screening of AI-generated financial images based on financial compliance rules, automated preliminary compliance checks can be performed on the generated financial images, thereby improving the initial efficiency of compliance review. Potential compliance risks can be identified and highlighted early, reducing the pressure of subsequent manual review. Images that pass the preliminary compliance screening are sent to designated personnel for multi-level manual review, and those that pass the review are considered AI financial images approved for compliance. This combines the efficiency of AI initial review with the rigor of professional human review, ensuring the authority and accuracy of the approval results. This improves overall approval efficiency while guaranteeing the final compliance of the financial image content, meeting the stringent risk control standards of the financial industry. By obtaining channel distribution strategies from the AI ​​financial image task flow and determining multiple distribution channels accordingly, it is possible to achieve... The strategic and automated determination of distribution channels ensures that images are accurately placed on appropriate target channels according to the preset plan, improving the efficiency and target orientation of distribution. By adapting compliant and approved AI financial images to multiple distribution channels, multiple adapted AI financial images are obtained. The image specifications and styles can be automatically adjusted according to the specific display requirements of different distribution channels, thereby ensuring the best display effect and user experience of the images on each channel and reducing the tedious work of manual adaptation. By pushing and publishing multiple adapted AI financial images to the corresponding multiple distribution channels, the final image content can be automatically and one-stop published to multiple target channels, which greatly accelerates the speed of information dissemination, improves operational efficiency, and ensures that financial information can reach the target audience in a timely and effective manner. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating the implementation of a workflow orchestration-based financial image generation method in one embodiment of this application. Figure 2 This is a flowchart illustrating the implementation of step S10 in a workflow-based financial image generation method according to an embodiment of this application. Figure 3 This is a flowchart illustrating the implementation of step S13 in a workflow-based financial image generation method according to an embodiment of this application. Figure 4 This is a flowchart illustrating the implementation of step S20 in a workflow-based financial image generation method according to an embodiment of this application. Figure 5 This is a flowchart illustrating the implementation of step S30 in a workflow-based financial image generation method according to an embodiment of this application. Figure 6 This is a flowchart illustrating the implementation of step S40 in a workflow-based financial image generation method according to an embodiment of this application. Figure 7 This is a schematic diagram of a financial image generation device based on workflow orchestration in one embodiment of this application; Figure 8 This is a schematic diagram of the internal structure of a computer device according to an embodiment of this application. Detailed Implementation

[0023] The following embodiments will help those skilled in the art to further understand the function of this application, but do not limit this application in any way. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of this application. These all fall within the protection scope of this application.

[0024] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0025] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0026] The present application will be further described in detail below with reference to the accompanying drawings.

[0027] In one embodiment, such as Figure 1 As shown, this application discloses a financial image generation method based on workflow orchestration, which specifically includes the following steps: S10: Obtain the user-defined work nodes in the workflow orchestration interface, construct the AI ​​financial image task flow based on the work nodes, and generate layered parameters. The layered parameters are used to generate AI fused images.

[0028] Specifically, a work node refers to a graphical unit selected and combined by the user on the visual workflow orchestration interface, representing a specific image processing stage or functional operation. For example, a work node can be used to specify the basic artistic style of the image, such as ink painting or a technological feel, while another work node can be used to load preliminary marketing text content. These selected work nodes are then connected to form an ordered execution sequence. The AI ​​financial image task flow is a complete and automated instruction sequence built based on the connection relationship of these work nodes and the functional instructions they carry. This AI financial image task flow defines all the calculation and processing steps required from parameter input to the final AI financial image output, much like a complete production blueprint customized for creating a promotional poster for a specific financial product, such as a stable growth fund.

[0029] Furthermore, in response to the completion of this AI financial image task flow, hierarchical parameters are generated. These hierarchical parameters are a set of structured parameters assigned different control levels. They may include parameters for global image features such as overall tone and lighting atmosphere control, as well as parameters for specific content elements such as preset financial chart styles and promotional slogan text definitions. This entire set of hierarchical parameters will jointly guide the subsequent AI image generation process, ensuring that the image can initially meet the preset visual and content goals.

[0030] S20: Based on the hierarchical parameters and the AI ​​financial image task flow, generate an initial AI financial image, and iteratively edit the initial AI financial image to generate an edited AI financial image.

[0031] Specifically, the initial AI financial image is a draft or first design synthesized by the underlying image generation model based on the aforementioned layered parameters and AI generation instructions contained in the AI ​​financial image task flow. This image generation model could be a text-based image model based on a diffusion mechanism. The initial draft or first design might be a preliminary design drawing showcasing the core features and application scenarios of a newly launched credit card from a bank, already possessing basic composition and elements. Iterative editing refers to the subsequent optimization and adjustment process of this initial AI financial image. This process allows users or automated scripts to modify the image multiple times, either locally or entirely. For example, a user might want to adjust the angle of a close-up shot of a product in the initial image, or change the color scheme of a background element to better align with marketing expectations or aesthetic preferences. Each such modification is an editing iteration. After one or more such iterative edits, the resulting AI financial image is a more refined version, closer to the final requirements in terms of detail, content, or style compared to the initial version.

[0032] S30: Acquire real-time financial data, and according to visualization rules, fuse the real-time financial data with the edited AI financial image to obtain the fused AI financial image.

[0033] Specifically, real-time financial data refers to the latest data actively queried or passively received from one or more designated sources that can provide immediate financial market information or user-specific financial information. The sources of real-time financial data can be obtained through stock exchange market interfaces, fund company net asset value release systems, or user personal account databases. For example, obtaining the current latest price of a stock, the past 24-hour return fluctuation data of a fund product, or the real-time profit and loss status of a specific client's investment portfolio. Visualization rules, on the other hand, are a set of pre-configured or dynamically specified transformation and display logics in the task flow. They stipulate how the acquired real-time financial data should be interpreted and transformed into visual elements reflected on an image. For example, one rule may define that when a fund's return is positive, its value is marked with a red upward arrow, and when it is negative, it is marked with a green downward arrow. Another rule may specify that a series of historical net asset value data should be rendered into a smooth curve.

[0034] More specifically, fusion processing involves applying these visualization rules to visually embed or update the information contained in the acquired real-time financial data into the corresponding areas of the AI ​​financial image. For example, in the reserved data display area of ​​the original financial product promotional image, the latest annualized rate of return of the product is dynamically filled in, and the display color is adjusted according to its positive or negative value. The resulting AI financial image after data fusion is a dynamically updated financial image that contains both the basic design elements generated by AI and reflects the latest financial data information in real time.

[0035] S40: Conduct compliance approval for the AI ​​financial image after data fusion. If the AI ​​financial image meets the conditions for compliance approval, determine multiple distribution channels and release the compliant AI financial image through multiple distribution channels.

[0036] Specifically, compliance approval is a review process to ensure that image content complies with relevant laws and regulations in the financial industry, advertising guidelines, and the company's internal brand and compliance policies. For example, approvers or automated approval scripts will check whether the image contains any misleading statements, whether necessary risk warnings are missing, and whether the financial data used is accurate and from compliant sources. If the AI ​​financial image successfully passes all review stages and meets the compliance approval conditions, it will enter the channel selection and release stage. Multiple distribution channels refer to selecting one or more media or platforms to place the image based on the purpose of the image generation or the marketing campaign plan. These channels may include the company's official website announcement area, emails sent to specific customer groups, mainstream social media platforms such as third-party public accounts, promotional posts on third-party enterprise accounts, or specific display positions within financial service apps. Finally, through actual placement, the image is formatted or resized according to the technical specifications of each channel, then uploaded to the backend system of the corresponding platform, and the release time is set or it goes live immediately so that it can be viewed or received by the target audience, thus achieving release through multiple distribution channels.

[0037] In one embodiment, such as Figure 2 As shown, in step S10, the user-defined work nodes in the workflow orchestration interface are obtained, an AI financial image task flow is constructed based on the work nodes, and hierarchical parameters are generated, specifically including: S11: In the workflow orchestration interface, multiple user-specified work nodes can be obtained by clicking or dragging, and the work nodes can be connected.

[0038] Specifically, the workflow orchestration interface is a visual operating environment, such as a graphical user interface with a toolbar and a main editing area. Users can see a list or icons of available work nodes representing different functions on this interface. Clicking or dragging is how users interact with the interface to select these work nodes. For example, a user can left-click on a text input node from the toolbar and drag it to the canvas in the main editing area, or select an image stylization node. These work nodes are pre-defined modular units that encapsulate specific image processing logic or data input / output functions. Subsequently, users connect work nodes by drawing directed lines between specific connection points of different nodes, thereby initially defining the execution sequence and data transfer direction between nodes. For example, the output of a product information acquisition node can be connected to the input of a poster template rendering node, laying the foundation for building a task flow.

[0039] S12: Construct the AI ​​financial image task flow based on the connection relationship and corresponding functional attributes of the working nodes.

[0040] Specifically, the connection relationships between work nodes refer to the topology and execution dependencies established by the user through connections in the aforementioned steps. For example, node A must execute after node B, or the output of node C is the input of node D. The corresponding functional attributes refer to the specific operation definitions and configurable items carried within each work node. For instance, the functional attributes of a data chart generation node might include selecting the chart type (e.g., line chart, bar chart), setting the data source fields, and configuring chart colors and labels. Therefore, based on these connection relationships and the functional attributes of each node, an AI financial image task flow is constructed. This involves combining these interconnected work nodes, each with initial attributes, into a complete, ordered, and automatically executable sequence of instructions. This sequence starts with initial data input, goes through a series of processing steps defined by the work nodes, and finally generates the target AI financial image. For example, this could form a dedicated task flow for automatically generating charts for weekly fund performance reports, which includes a series of sequential node-based operations such as data acquisition, chart drawing, and risk warning embedding.

[0041] S13: Responds to the AI ​​financial image task flow and generates hierarchical parameters.

[0042] Specifically, responding to the AI ​​financial image task flow refers to the process of generating parameters after the structure of the AI ​​financial image task flow and its constituent work nodes and their initial configurations, as described in the previous step, are determined, or when the task flow reaches a stage requiring specific parameter input. This triggers the parameter generation process, thereby generating hierarchical parameters to precisely control the final AI image output based on the image generation requirements embodied in the constructed task flow and the functional settings of each node. These hierarchical parameters are divided into different control levels or categories. For example, some parameters may be used to control the overall visual performance of the image, such as basic color tone and composition, while other parameters may be used to define specific business information or content elements that need to be embedded in the image. This forms a structured and hierarchical parameter system to ensure that the subsequent AI model can generate images that meet expectations based on these comprehensive instructions. For example, if the task flow contains a template node for a holiday greeting poster, it may generate initial hierarchical parameters that include holiday theme colors, greeting text placeholders, and recommended product information placeholders.

[0043] In one embodiment, such as Figure 3 As shown, in step S13, in response to the AI ​​financial image task flow, hierarchical parameters are generated, specifically including: S131: Obtain the set of defined parameters input by the user on the interactive interface, and identify the basic layer parameters, business layer parameters and financial compliance layer parameters from the set of defined parameters.

[0044] Specifically, on a structured interactive interface dedicated to parameter configuration, users input and select various elements for generating AI financial images, forming a data set—the defined parameter set. For example, on this interface, a user might specify a modern, technological feel as the visual style for a newly launched fund product's promotional image, input the product name (e.g., Steady Growth Treasure), select the required risk warning level (R3), and upload the institution's logo file. These diverse inputs collectively constitute the initial defined parameter set. This raw, potentially loosely structured set is then analyzed and categorized, assigning specific instructions or data to three predefined control layers: basic layer parameters, business layer parameters, and financial compliance layer parameters. For instance, the user-selected modern, technological style is categorized as a basic layer parameter controlling the overall aesthetics and layout of the image; the input product name and logo information are categorized as business layer parameters reflecting specific business content and brand elements; and the selected R3 risk warning level and its corresponding standard warning requirements are categorized as financial compliance layer parameters ensuring the image complies with regulations and internal guidelines.

[0045] S132: Perform preset formatting and structuring processing on the basic layer parameters, business layer parameters, and financial compliance layer parameters to generate layered parameters.

[0046] Specifically, the parameters identified in the previous step—namely, the basic layer parameters, business layer parameters, and financial compliance layer parameters—will undergo further pre-defined formatting and structuring. This processing aims to transform the preliminary parameter information from user interaction or system parsing into standardized instructions that the subsequent AI image generation model can directly understand and execute. For example, formatting may involve converting the user's chosen style preference for a modern, technological feel into a set of specific AI model control parameters, such as specific seed value ranges, color guidance weights, or texture descriptors; unifying the user's input text information into a specific encoding format; or converting the uploaded logo image data into a format suitable for embedding. Structuring refers to organizing these formatted basic, business, and financial compliance layer parameters according to predefined hierarchical relationships and data structures, such as a nested JSON object or XML document, to form a logically clear, hierarchical, complete, and unambiguous set of parameters. After such formatting and structuring, hierarchical parameters are generated. This is a highly organized set of parameters that can be directly input into the AI ​​image generation core engine to comprehensively guide and constrain the creation of subsequent image content.

[0047] In one embodiment, such as Figure 4 As shown, in step S20, based on the hierarchical parameters and the AI ​​financial image task flow, an initial AI financial image is generated, and the initial AI financial image is iteratively edited to generate an edited AI financial image, specifically including: S21: Using the text-based image model in the AI ​​financial image task flow, the initial AI financial image is obtained by utilizing the hierarchical parameters.

[0048] Specifically, when executing the previously constructed AI financial image task flow, one or more nodes or modules of the text-based image model specifically designed for image content synthesis are activated. This model, for example, is a deep learning-based diffusion model that can understand and process complex input instructions. It uses the set of structured hierarchical parameters, which are ultimately generated above and contain specific settings for the basic layer, business layer, and compliance layer, as the core input to the text-based image model. The text-based image model will create an image from scratch according to the instructions defined in these parameters, such as the overall style, specific business elements, text content, composition, and compliance requirements. This results in the first version of a financial-themed image generated by AI based on complete parameter instructions, i.e., the initial AI financial image, such as a preliminary draft of a promotional poster for a bank's credit card promotion that meets the requirements.

[0049] S22: Determine whether there are local redraw nodes in the AI ​​financial image task flow.

[0050] Specifically, during the execution of the AI ​​financial image task flow, after the initial AI financial image has been generated, a process condition check is performed. This check is a specific functional node that the user may selectively add to the task flow based on the expectation that a fine-tuning of a specific part of the image may be required. For example, if the user anticipates that the product appearance or background may need to be fine-tuned according to different promotion periods when designing an advertising image that includes a dynamic product display area, a local redraw node may be added after the main generation node. The purpose of this judgment step is to check whether such a node is included in the currently executed task flow definition. If it is included, it means that a subsequent editing step specifically targeting a part of the image needs to be triggered.

[0051] S23: If there is a local redraw node in the AI ​​financial image task flow, obtain the region modification information of the user's selected area of ​​the initial AI financial image, and iteratively edit the initial AI financial image according to the region modification information to obtain the edited AI financial image.

[0052] Specifically, when the judgment result indicates the existence of local redraw nodes, the system may pause the automated process and present an interactive interface, allowing the user to precisely specify one or more areas to be modified on the generated initial AI financial image using tools such as mouse selection and lasso tools, and input corresponding area modification information for these areas. This information can be a new descriptive text, such as inputting instructions to replace the background of the selected area with a sunny beach, specific parameter adjustment instructions such as increasing the brightness of this area by 30%, or selecting replacement materials, etc., and then calling the corresponding image processing algorithm, such as local redraw or image restoration algorithm, to regenerate or adjust the content only for the user-specified selected area based on the obtained area modification information, while keeping the remaining unselected parts of the image unchanged or as minimally affected as possible. This editing process may involve one or more adjustments and previews until the user is satisfied, and the output is the edited AI financial image. This is the AI ​​financial image version that has been obtained after this targeted and iterative local modification, and is more in line with the user's specific requirements in terms of details.

[0053] More specifically, every significant operation performed by the user or automated process—such as every adjustment and setting of layered parameters (e.g., selecting visual style keywords in the base layer, inputting financial product name text in the business layer, or selecting risk warning statements in the compliance layer), every time a new intermediate image version is generated by the driving model, and every local area modification or overall editing adjustment of the image—is captured and registered. The type, time, specific parameter values, and resulting image version snapshots or their differential data are systematically saved in a structured editing history sequence. This editing history is designed so that users can clearly view the entire editing process through the interactive interface and have the ability to perform version rollback operations (restoring to any previous image version in the history sequence) and operation undo (canceling the most recent one or more editing actions). This provides users with a non-destructive, freely explorable editing experience while preserving the complete creative trajectory.

[0054] In one embodiment, such as Figure 5 As shown, in step S30, real-time financial data is acquired, and according to visualization rules, the real-time financial data is fused with the edited AI financial image to obtain the fused AI financial image. Specifically, this includes: S31: Obtain real-time financial data based on multiple preset financial data source interfaces.

[0055] Specifically, the preset financial data source interface refers to the data input prepared according to the previously configured or specified connection channels for accessing various financial information sources during the execution of the AI ​​financial image task flow. These financial data source interfaces can be application programming interfaces (APIs) that directly connect to the internal product database of financial institutions, standardized data service interfaces for obtaining public market information such as stock prices and index points, or interfaces for extracting personalized financial data from customer relationship management systems (CRM) for specific user groups. Then, through these preset interfaces, data requests are actively initiated or the latest information pushed is received to obtain real-time financial data, such as the latest net asset value of a fund on that day, the real-time rise and fall of a specific stock, or the current market hot spot analysis data. This is the most up-to-date dynamic information that can reflect the current situation as far as possible up to the time of image processing.

[0056] S32: Based on the visualization rules, obtain the mapping relationship between data and image elements. Based on the mapping relationship, map real-time financial data onto the elements in the AI ​​financial image to generate the AI ​​financial image after data fusion.

[0057] Specifically, the visualization rules specify in detail how different real-time financial data should be visually presented and which parts of the image they should be associated with. By parsing these rules, the mapping relationship between data and image elements is clearly identified. For example, the latest net asset value of a fund should correspond to and update the text box element with the ID "NAV_Display_Area" in the AI ​​financial image. Meanwhile, the real-time stock price fluctuation data should affect the color and shape of the chart element with the ID "Stock_Trend_Chart". The actual real-time financial data value or its derived information, such as the "rising" or "falling" state determined by the value, is then applied to these image elements with established mapping relationships. For example, the obtained fund net asset value "1.532" is filled into the corresponding text box, or a certain indicator arrow in the chart area is changed to red or green according to the stock price fluctuation. Through such mapping and content updates, the AI ​​financial image after data fusion is generated. This is an AI financial image that is not only beautifully designed, but also reflects the latest financial dynamics in real time and has vivid content.

[0058] In one embodiment, such as Figure 6 As shown, in step S40, the AI ​​financial image after data fusion undergoes compliance approval. If the AI ​​financial image meets the conditions for compliance approval, multiple distribution channels are determined, and the compliant AI financial image is released through these channels. Specifically, this includes: S41: Based on financial compliance rules, conduct preliminary compliance screening on the AI ​​financial images after data fusion to obtain preliminary compliance screening results.

[0059] Specifically, financial compliance rules refer to a set of pre-established and dynamically updated rules that include regulations on financial advertising, industry self-regulatory guidelines, and internal branding and information disclosure standards for enterprises. Therefore, the AI ​​financial images after data fusion are subjected to preliminary compliance screening in accordance with these financial compliance rules. This screening is usually an automated detection process. For example, it uses image content analysis and text recognition technology to check whether prohibited words are used in the image, whether necessary risk warnings are missing, whether the presented numerical data conforms to common sense or prescribed formats, and whether the use of brand logos is standardized. After the screening is completed, a preliminary compliance screening result is obtained. This result may be a report or status mark that includes specific inspection items, a judgment of whether it passes or fails, and a warning of potential risks. For example, the screening result may indicate that the font size of the risk warning does not meet the requirements or that the image contains an unauthorized third-party logo.

[0060] S42: Based on the preliminary compliance screening results, the AI ​​financial images that have passed the preliminary compliance screening will be sent to designated personnel for multi-level manual review. The AI ​​financial images that have passed the multi-level manual review will be used as the AI ​​financial images that have passed the compliance approval.

[0061] Specifically, if the initial screening result is "pass" or suggests manual review, the manual review process is initiated. Alternatively, for cases where all images must undergo manual review, this result can also serve as reference information for the manual review. The image and its initial screening report are then forwarded to designated personnel with different review responsibilities, forming a multi-level manual review process. For example, first, the compliance specialist in the legal department reviews legal compliance; then, the brand manager in the marketing department may oversee brand tone and marketing messaging; and finally, the business department head confirms the final accuracy and applicability for business purposes. During this process, only after the image successfully passes all designated manual review levels and receives approval from each reviewer is the AI ​​financial image ultimately marked as compliant and approved, qualifying for public release.

[0062] More specifically, once the AI ​​financial image enters the compliance approval stage, all activities and status changes throughout the entire approval process are comprehensively and meticulously recorded and archived. This includes the compliance rule version upon which the AI's automatic initial compliance screening is based, the specific results obtained from the screening, and the marked risk points. It also includes the identity of the reviewers, the submitted review comments (such as approval, rejection, or modification suggestions), any possible modifications to the image (if the approval process allows for modifications during review), and every change in the image's approval status (from pending initial review to pending secondary review, then to approval or rejection) and its corresponding accurate timestamp. All of this is systematically recorded and linked to the corresponding AI financial image version, forming a complete and tamper-proof approval log. Through this meticulous recording and archiving of the entire process of generation, editing, and approval, a high degree of transparency and comprehensive traceability of the entire lifecycle of the AI ​​financial image is achieved. This provides a solid data foundation and evidentiary support for subsequent quality management, compliance auditing, problem identification, and process optimization.

[0063] S43: Obtain the channel distribution strategy from the AI ​​financial image task flow, and determine multiple distribution channels based on the channel distribution strategy.

[0064] Specifically, after the image has been confirmed to be compliant and approved, the channel distribution strategy set or associated in the AI ​​financial image task flow is searched and invoked. This strategy may include preset rules such as the target release channel type, priority, release time window, or specific audience profile for this type of AI financial image. For example, the task flow for promoting a new fund product may have a preset channel distribution strategy that prioritizes push notifications on the official APP, followed by third-party public accounts and partner financial media websites. Based on these preset strategy rules, combined with the specific attributes of the current image, such as the product line, target customer group, or current marketing plan, a list of one or more multi-distribution channels to actually release the image is selected. For example, the final release channels are determined to be the official APP homepage banner, the image in the Tuesday post of a third-party service account, and the illustration in a column article on a certain financial website.

[0065] S44: Based on multiple distribution channels, adapt the AI ​​financial images that have passed compliance approval to obtain multiple adapted AI financial images.

[0066] Specifically, based on the established multiple distribution channels—that is, the defined list of specific channels—the AI ​​financial images that have passed compliance approval need to be adapted. This is because different distribution channels often have different technical specifications and content display requirements for image size, aspect ratio, file size, resolution, and even subtle style aspects such as whether dynamic effects are allowed and whether channel-specific badges need to be added. Adaptation processing may include automated image cropping, scaling, format conversion such as converting from high-definition PNG to compressed JPG or WEBP suitable for the web, resolution adjustment, or adding specific watermarks or borders according to channel characteristics. Since there are multiple target channels, this adaptation process may generate an optimized and adjusted image version for each selected distribution channel, resulting in multiple adapted AI financial images. For example, the same promotional poster may be adapted into a square version suitable for WeChat Moments, a long strip version suitable for website banners, and a full-screen vertical version suitable for APP launch pages.

[0067] S45: Push and publish multiple adapted AI financial images to corresponding distribution channels online.

[0068] Specifically, multiple AI financial images adapted for each channel will be accurately transmitted to the corresponding multi-distribution channels. The adapted images will be automatically uploaded to the material libraries of various social media platforms via API interfaces and published at preset times, or embedded into email marketing templates being edited, or deployed to the company's official website's specific page content management system (CMS) and activated, or sent to target user groups as part of rich media messages through mobile application push services. This ensures that each adapted image can be displayed with the best effect on its corresponding channel, thereby realizing content push and online publishing.

[0069] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0070] In one embodiment, a workflow-based financial image generation apparatus is provided, which corresponds one-to-one with the workflow-based financial image generation method described in the above embodiments. For example... Figure 7 As shown, this workflow-based financial image generation device includes a workflow construction and parameter generation module, an image generation and editing module, a dynamic data fusion module, and a compliance approval and intelligent distribution module. Detailed descriptions of each functional module are as follows: The workflow construction and parameter generation module is used to obtain the work nodes determined by the user in the workflow orchestration interface, construct the AI ​​financial image task flow based on the work nodes, and generate hierarchical parameters. The hierarchical parameters are used to generate AI fused images. The image generation and editing module is used to generate an initial AI financial image based on hierarchical parameters and the AI ​​financial image task flow, and to iteratively edit the initial AI financial image to generate an edited AI financial image. The dynamic data fusion module is used to acquire real-time financial data and, according to visualization rules, fuse the real-time financial data with the edited AI financial image to obtain the fused AI financial image. The compliance approval and intelligent distribution module is used to conduct compliance approval on the AI ​​financial images after data fusion. If the AI ​​financial image meets the conditions for compliance approval, multiple distribution channels are determined and the AI ​​financial image that has passed compliance approval is released through multiple distribution channels.

[0071] Optional, the workflow construction and parameter generation module specifically includes: The Visual Node Selection and Connection submodule is used to select and connect multiple user-specified work nodes in the workflow orchestration interface by clicking or dragging. The task flow logic construction submodule is used to construct the AI ​​financial image task flow based on the connection relationship and corresponding functional attributes of the work nodes; The layered parameter generation trigger submodule is used to generate layered parameters in response to the AI ​​financial image task flow.

[0072] Optional, the hierarchical parameter generation trigger submodule includes: The user-defined parameter acquisition and identification unit is used to acquire the set of defined parameters input by the user on the interactive interface, and to identify the basic layer parameters, business layer parameters and financial compliance layer parameters from the set of defined parameters. The parameter formatting and structuring unit is used to perform preset formatting and structuring processing on basic layer parameters, business layer parameters, and financial compliance layer parameters to generate layered parameters.

[0073] Optional image generation and editing module, specifically including: The AI ​​image initial generation submodule is used to obtain the initial AI financial image by using the Wensheng image model in the AI ​​financial image task flow and utilizing the hierarchical parameters. The path decision submodule is used to determine whether there are local redraw nodes in the AI ​​financial image task flow; The image region iterative editing submodule is used to obtain the region modification information of the selected area of ​​the initial AI financial image if there is a local redraw node in the AI ​​financial image task flow. Based on the region modification information, the initial AI financial image is iteratively edited to obtain the edited AI financial image.

[0074] Optional, dynamic data fusion module, specifically including: The real-time data acquisition submodule is used to acquire real-time financial data based on multiple preset financial data source interfaces; The data image mapping and fusion submodule is used to obtain the mapping relationship between data and image elements according to visualization rules. Based on the mapping relationship, real-time financial data is mapped to elements in the AI ​​financial image being edited, generating the AI ​​financial image after data fusion.

[0075] Optional compliance approval and intelligent distribution module, specifically including: The automated compliance screening submodule is used to perform preliminary compliance screening on the AI ​​financial images after data fusion based on financial compliance rules, and obtain preliminary compliance screening results; The manual multi-level review management submodule is used to send the AI ​​financial images that have passed the initial compliance screening to designated personnel for manual multi-level review based on the initial compliance screening results. The AI ​​financial images that have passed the manual multi-level review are used as AI financial images that have passed compliance approval. The channel strategy analysis and determination submodule is used to obtain the channel distribution strategy from the AI ​​financial image task flow and determine multiple distribution channels based on the channel distribution strategy. The image channel adaptation submodule is used to adapt AI financial images that have passed compliance approval based on multiple distribution channels, so as to obtain multiple adapted AI financial images. The multi-channel content publishing submodule is used to push and publish multiple adapted AI financial images to corresponding multi-distribution channels online.

[0076] Specific limitations regarding the workflow-based financial image generation device can be found in the limitations of the workflow-based financial image generation method described above, and will not be repeated here. Each module in the aforementioned workflow-based financial image generation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0077] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data information such as AI financial image task flows, visualization rules, and AI financial images. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer program implements a workflow-based financial image generation method.

[0078] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps: Obtain the user-defined work nodes from the workflow orchestration interface, construct the AI ​​financial image task flow based on the work nodes, and generate layered parameters. The layered parameters are used to generate AI fused images. Based on the hierarchical parameters and the AI ​​financial image task flow, an initial AI financial image is generated, and the initial AI financial image is iteratively edited to generate an edited AI financial image. Acquire real-time financial data, and according to visualization rules, fuse the real-time financial data with the edited AI financial image to obtain the fused AI financial image; The AI ​​financial images after data fusion undergo compliance approval. If the AI ​​financial images meet the conditions for compliance approval, multiple distribution channels are determined, and the AI ​​financial images that have passed compliance approval are released through multiple distribution channels.

[0079] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor: Obtain the user-defined work nodes from the workflow orchestration interface, construct the AI ​​financial image task flow based on the work nodes, and generate layered parameters. The layered parameters are used to generate AI fused images. Based on the hierarchical parameters and the AI ​​financial image task flow, an initial AI financial image is generated, and the initial AI financial image is iteratively edited to generate an edited AI financial image. Acquire real-time financial data, and according to visualization rules, fuse the real-time financial data with the edited AI financial image to obtain the fused AI financial image; The AI ​​financial images after data fusion undergo compliance approval. If the AI ​​financial images meet the conditions for compliance approval, multiple distribution channels are determined, and the AI ​​financial images that have passed compliance approval are released through multiple distribution channels.

[0080] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0081] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0082] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A financial image generation method based on workflow orchestration, characterized in that, The workflow orchestration-based financial image generation method includes: Obtain the user-defined work nodes from the workflow orchestration interface, construct an AI financial image task flow based on the work nodes, and generate hierarchical parameters. These hierarchical parameters are used to generate AI fused images, specifically including: In the workflow orchestration interface, multiple user-specified work nodes are obtained by clicking or dragging, and the work nodes are connected. The AI ​​financial image task flow is constructed based on the connection relationships and corresponding functional attributes of the working nodes; In response to the AI ​​financial image task flow, the hierarchical parameters are generated, specifically including: Obtain the set of defined parameters input by the user on the interactive interface, and identify the basic layer parameters, business layer parameters and financial compliance layer parameters from the set of defined parameters; The basic layer parameters, the business layer parameters, and the financial compliance layer parameters are formatted and structured in a preset manner to generate the layered parameters; Based on the hierarchical parameters and the AI ​​financial image task flow, an initial AI financial image is generated, and the initial AI financial image is iteratively edited to generate an edited AI financial image, specifically including: The initial AI financial image is obtained by using the text-based image model in the AI ​​financial image task flow and the layered parameters. Determine whether there are any local redraw nodes in the AI ​​financial image task flow; If the local redraw node exists in the AI ​​financial image task flow, then obtain the region modification information of the selected area of ​​the initial AI financial image by the user, and iteratively edit the initial AI financial image according to the region modification information to obtain the edited AI financial image; Acquire real-time financial data, and according to visualization rules, fuse the real-time financial data with the edited AI financial image to obtain a data-fused AI financial image, specifically including: The real-time financial data is obtained based on multiple preset financial data source interfaces; According to the visualization rules, the mapping relationship between data and image elements is obtained. Based on the mapping relationship, the real-time financial data is mapped onto the elements in the edited AI financial image to generate the data-fused AI financial image. The AI ​​financial image after data fusion undergoes compliance approval. If the AI ​​financial image after data fusion meets the conditions for compliance approval, multiple distribution channels are determined, and the AI ​​financial image that has passed compliance approval is published through the multiple distribution channels.

2. The financial image generation method based on workflow orchestration according to claim 1, characterized in that, The process involves conducting compliance approval on the AI ​​financial image after data fusion. If the AI ​​financial image meets the conditions for compliance approval, multiple distribution channels are determined, and the compliant AI financial image is published through these channels. Specifically, this includes: Based on financial compliance rules, a preliminary compliance screening is performed on the AI ​​financial image after data fusion to obtain the preliminary compliance screening results. Based on the preliminary compliance screening results, the AI ​​financial image fused from the preliminary compliance screening data will be sent to designated personnel for multi-level manual review. The AI ​​financial image that passes the multi-level manual review will be used as the AI ​​financial image that has passed the compliance approval. Obtain the channel distribution strategy from the AI ​​financial image task flow, and determine the multiple distribution channels based on the channel distribution strategy; Based on the multiple distribution channels, the AI ​​financial images that have passed the compliance approval are adapted to obtain multiple adapted AI financial images. The adapted AI financial images are pushed and published online to the corresponding multiple distribution channels.

3. A financial image generation device based on workflow orchestration, characterized in that, The workflow orchestration-based financial image generation device includes: The workflow construction and parameter generation module is used to obtain the user-defined work nodes in the workflow orchestration interface, construct the AI ​​financial image task flow based on the work nodes, and generate hierarchical parameters. These hierarchical parameters are used to generate AI fused images, specifically including: The visualization node selection and connection submodule is used to obtain multiple user-specified work nodes in the workflow orchestration interface through click or drag operations, and connect the work nodes. The task flow logic construction submodule is used to construct the AI ​​financial image task flow based on the connection relationship and corresponding functional attributes of the work nodes; The layered parameter generation trigger submodule is used to generate the layered parameters in response to the AI ​​financial image task flow, specifically including: The user-defined parameter acquisition and identification unit is used to acquire the set of defined parameters input by the user on the interactive interface, and to identify the basic layer parameters, business layer parameters and financial compliance layer parameters from the set of defined parameters. The parameter formatting and structuring unit is used to perform preset formatting and structuring processing on basic layer parameters, business layer parameters, and financial compliance layer parameters to generate layered parameters. The image generation and editing module is used to generate an initial AI financial image based on the layered parameters and the AI ​​financial image task flow, and to iteratively edit the initial AI financial image to generate an edited AI financial image, specifically including: The AI ​​image initial generation submodule is used to obtain the initial AI financial image by using the Wensheng image model in the AI ​​financial image task flow and utilizing the hierarchical parameters. The path decision submodule is used to determine whether there are local redraw nodes in the AI ​​financial image task flow; The image region iterative editing submodule is used to obtain the region modification information of the selected region of the initial AI financial image by the user if there is a local redraw node in the AI ​​financial image task flow, and to iteratively edit the initial AI financial image according to the region modification information to obtain the edited AI financial image. A dynamic data fusion module is used to acquire real-time financial data and, according to visualization rules, fuse the real-time financial data with the edited AI financial image to obtain a fused AI financial image, specifically including: The real-time data acquisition submodule is used to acquire real-time financial data based on multiple preset financial data source interfaces; The data image mapping and fusion submodule is used to obtain the mapping relationship between data and image elements according to visualization rules. Based on the mapping relationship, real-time financial data is mapped to the elements in the AI ​​financial image to generate the AI ​​financial image after data fusion. The compliance approval and intelligent distribution module is used to conduct compliance approval on the AI ​​financial image after data fusion. If the AI ​​financial image after data fusion meets the conditions for compliance approval, multiple distribution channels are determined, and the AI ​​financial image that has passed compliance approval is published through the multiple distribution channels.

4. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the workflow-based financial image generation method as described in any one of claims 1 to 2.

5. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the workflow orchestration-based financial image generation method as described in any one of claims 1 to 2.

Images