A digital asset generation and management system

Abstract

This invention discloses a digital asset generation and management system, relating to the fields of digital asset management and artificial intelligence. It comprises seven core modules, forming a closed-loop system of "generation-ownership confirmation-management-traceability-value mining-optimization-control" through a service mesh architecture. This system is adaptable to the full lifecycle management of multimodal digital assets. The system innovatively integrates human-machine collaborative ownership confirmation and blockchain-zero-knowledge proof fusion, addressing pain points such as insufficient collaboration, difficulty in ownership confirmation, and incomplete traceability in existing technologies. It achieves intelligent asset generation, secure traceability, value activation, and multi-industry adaptability, significantly improving management efficiency and reducing costs, possessing extremely high practicality and promotional value.

Description

Technical Field

[0001] This invention relates to the fields of digital asset management and artificial intelligence, and in particular to a digital asset generation and management system. Background Technology

[0002] As the digital economy enters a stage of high-quality development, digital assets have become a core production factor for enterprises. Their forms are constantly evolving, encompassing diverse types such as data assets, digital content, virtual items, NFTs, and digital twin models, and their quantity is experiencing explosive growth. According to Gartner research data, the total amount of digital assets held by global enterprises will exceed 5 ZB in 2020, a tenfold increase compared to 2020. However, enterprises are caught in the "digital asset paradox"—for every new piece of content created, three similar pieces lie dormant deep within their hard drives; 60% of marketing budgets are used to repeatedly create existing materials; and 90% of high-quality assets are never fully utilized, resulting in enormous resource waste. While current technologies and systems related to digital asset generation and management have made some progress, fundamental technical deficiencies remain, making it difficult to meet the large-scale, intelligent, secure, and compliant management needs of enterprises, especially in the field of AIGC (AI-generated content) video creation, specifically as follows:

[0003] In existing technologies, the generation of digital assets (AI generation, human creation) and subsequent storage, labeling, retrieval and management are separate modules, lacking a unified collaborative architecture. The generated assets require manual intervention for format conversion, tag addition and other operations, which is not only time-consuming and labor-intensive, but also prone to human error.

[0004] Existing digital asset management systems mostly use traditional databases to store asset information, lacking the ability to trace the entire lifecycle of digital assets without alteration. This makes it impossible to accurately trace the source of digital assets, modification records, transfer paths, and changes in ownership, which can easily lead to problems such as asset theft, tampering, and misuse. Summary of the Invention

[0005] The main objective of this invention is to address the problems in existing technologies, such as the separation of generation and management, lack of reliable traceability throughout the entire process, low asset utilization, poor industry adaptability, and weak security control. In particular, it aims to solve the problems of insufficient standardization, compliance, stability, and reusability of assets in AIGC image creation.

[0006] To address the above problems, this invention proposes a digital asset generation and management system, characterized by comprising:

[0007] The system comprises an intelligent generation module, a collaborative management module, a blockchain traceability module, an intelligent value mining module, an adaptive adaptation module, and a security control module. These modules work together through a distributed bus architecture to achieve data exchange and collaborative operation, forming a closed-loop system of "generation-management-traceability-value mining-security control".

[0008] The intelligent generation module is used to receive user generation requests, realize the intelligent generation of multiple types of digital assets based on multimodal generation models and dynamic parameter optimization algorithms, and complete the automated detection and correction of generation quality in real time. The intelligent generation module provides unified, compliant and reusable digital materials for AIGC image creation. As the core tool layer of image production, it ensures the stability of work quality from the source.

[0009] The collaborative management module is used to manage the generated digital assets throughout their entire lifecycle, including asset classification, standardization, version control, multi-dimensional retrieval, and cross-departmental collaborative operations.

[0010] The blockchain traceability module is used to record the entire process information of digital assets from generation, modification, circulation to destruction. Based on the immutability of blockchain, it achieves the accuracy and security of asset traceability and supports full-process operation auditing and accountability.

[0011] The intelligent value mining module is used to intelligently analyze the usage data and relationships of digital assets, mine the value potential of the assets, and generate value assessment reports and reuse suggestions.

[0012] The adaptive module is used to automatically adjust system parameters and functional modules according to the business needs of different industries, adapt to the management needs of multimodal digital assets, and achieve functional expansion without large-scale modification.

[0013] The security management module adopts a role-based fine-grained permission management mechanism, combined with data encryption and abnormal behavior monitoring technologies, to achieve hierarchical protection, access control and security early warning of digital assets.

[0014] Preferably, the intelligent generation module includes a demand analysis unit, a multimodal generation unit, a dynamic parameter optimization unit, and a quality detection and correction unit;

[0015] The requirement parsing unit is used to receive the generation requirements input by the user, parse the requirement keywords through natural language processing algorithms, and generate standardized generation instructions. The generation instructions support structured requirement parsing for professional scenarios such as film and television production, character design, and script visualization.

[0016] The multimodal generation unit has built-in image generation model, audio generation model, text generation model, and virtual asset generation model, which supports the synchronous generation of multiple types of digital assets, and the models can communicate with each other.

[0017] The layer relies on mainstream multimodal large model APIs such as nanobanana, gptimage2, and gpt5, and adopts a modular architecture, designed according to three major modules: characters, props, and scenes, which can efficiently generate high-precision digital assets.

[0018] Character generation unit: Supports dual modes of material library access and customization, ensures compliance through authorization comparison and element blending, and solves the problems of character drift and poor stability with multi-view output and unified style rendering;

[0019] Item generation unit: Covers the generation of all types of items, relies on authoritative reference image library to ensure accurate details, supports multi-view output and style consistency, and can be reused across projects;

[0020] Scene Generation Unit: Focuses on high-precision restoration and rendering of cultural and tourism landmarks, historical buildings, and regional culture, realizing real-scene replication, ancient building restoration, and asset management;

[0021] Meanwhile, the multimodal generation unit integrates five major creative tools: intelligent image generation, multi-image fusion creation, perspective conversion, style transfer, and professional lens language, providing AIGC image creation with one-stop material generation capabilities;

[0022] The dynamic parameter optimization unit, based on reinforcement learning algorithm, collects data in real time during the generation process and dynamically adjusts the parameters of the generation model to achieve a balance between generation efficiency and generation quality.

[0023] The quality inspection and correction unit establishes a multi-dimensional quality assessment index system. Through technologies such as computer vision, audio recognition, and text review, it automatically inspects the generated digital assets. If an asset that does not meet the quality standards is detected, a correction instruction is automatically triggered to adjust the generation parameters and regenerate the asset. For video creation scenarios, the quality inspection and correction unit focuses on verifying the consistency of characters, the precision of prop details, the scene reproduction, and the uniformity of style to ensure that the generated assets meet film and television production standards.

[0024] Preferably, the collaborative management module includes an asset classification unit, a standardization processing unit, a version control unit, an intelligent retrieval unit, and a collaborative operation unit;

[0025] The asset classification unit adopts a semantic graph-based intelligent classification algorithm, combined with user-defined classification rules, to automatically classify and archive digital assets, while establishing asset association relationships. For video creation assets, it supports a multi-level classification system based on "characters / props / scenes-themes-projects" to achieve rapid classification and retrieval of materials.

[0026] The standardization processing unit automatically performs standardization processing on digital assets of different formats, such as format unification, size adjustment, and data compression, to generate asset files that conform to industry standards.

[0027] The version control unit uses incremental backup and version marking technology to record every modification operation of digital assets, supports version rollback and version comparison, and can set version permissions.

[0028] The intelligent retrieval unit supports keyword retrieval, semantic retrieval, image similarity retrieval, and cross-modal retrieval. Combined with intelligent recommendation algorithms, it pushes digital assets that users may need based on their usage habits.

[0029] The collaborative operation unit supports multi-user, cross-departmental collaborative operations, including functions such as asset sharing, permission allocation, task assignment, and progress tracking.

[0030] Preferably, the blockchain traceability module includes an information collection unit, a blockchain writing unit, a traceability query unit, and an auditing unit;

[0031] The information collection unit collects information on the entire process of digital assets from generation to destruction in real time, ensuring the integrity of traceability information. For image assets, it simultaneously collects compliance information such as authorization source, material call records, and modification traces, providing data support for copyright tracing.

[0032] The blockchain writing unit adopts a consortium blockchain architecture, which encrypts the collected traceability information and writes it into the blockchain node, and combines it with the Alibaba Cloud BaaS platform to achieve elastic scaling.

[0033] The traceability query unit allows users to quickly query the full-process traceability information of digital assets using keywords such as asset ID, generation time, and asset name, and generate a visual traceability report.

[0034] The audit unit automatically performs audit analysis on traceability information, identifies abnormal operations, generates audit reports, and provides data support for asset security management.

[0035] Preferably, the intelligent value mining module includes a data acquisition unit, a correlation analysis unit, a value assessment unit, and a reuse recommendation unit;

[0036] The data acquisition unit collects usage data, storage data, and associated data of digital assets in real time.

[0037] The association analysis unit uses a graph neural network algorithm to analyze the relationship between digital assets and the matching degree between assets and business scenarios, and to explore potential asset reuse scenarios. For video assets, it can analyze the reuse potential of characters / props / scenes in different themes and projects, thereby reducing the cost of repeated creation.

[0038] The value assessment unit establishes a digital asset value assessment model, combines indicators such as usage frequency, reuse potential, creation cost, and business contribution to conduct quantitative value assessment of digital assets, and generates a value assessment report.

[0039] The reuse recommendation unit pushes high-value, high-reuse-potential digital assets based on the value assessment results and the user's business needs.

[0040] Preferably, the adaptive adaptation module includes a requirement identification unit, a parameter adjustment unit, and a module expansion unit;

[0041] The demand identification unit, by collecting users' business data and operating habits, combined with industry characteristics, identifies the personalized needs of different industries. It can identify the image creation needs of different themes such as film and television, cultural tourism, intangible cultural heritage, and commerce, and automatically adapt the corresponding generation strategies and quality standards.

[0042] The parameter adjustment unit automatically adjusts various system parameters based on the identified personalized needs, without requiring manual adjustment.

[0043] The module expansion unit adopts a microservice architecture, supports the rapid addition and deletion of functional modules, supports cloud-native deployment, and achieves elastic expansion through Kubernetes + container architecture.

[0044] Preferably, the security control module includes an access control unit, a data encryption unit, an anomaly monitoring unit, and a security early warning unit;

[0045] The permission management unit adopts a role-based fine-grained permission management mechanism, which divides users into different roles and assigns different operation permissions to each role.

[0046] The data encryption unit uses a combination of symmetric and asymmetric encryption to encrypt the digital assets themselves, traceability information, and user data.

[0047] The anomaly monitoring unit monitors user actions and system operating status in real time, identifies abnormal behaviors, and records anomaly logs.

[0048] The security early warning unit automatically triggers a security early warning based on the abnormal monitoring results, and notifies the administrator to handle the situation in a timely manner.

[0049] Preferably, it also includes an edge computing module, which is used to offload some generation, detection and retrieval tasks to edge nodes, reduce the pressure on cloud servers, improve task processing speed, achieve low latency response, and at the same time synchronize data with the cloud system to ensure data consistency.

[0050] Preferably, the multimodal generation model employs a transfer learning algorithm, which can optimize the performance of the generation model based on the user's historical generation data, while also supporting users to upload custom models to adapt to the generation needs of special scenarios.

[0051] Beneficial effects:

[0052] 1. Achieve seamless integration of AI-native collaborative generation and end-to-end management, innovate the "human-machine collaboration" rights confirmation unit, quantify the contribution ratio of users and AI, solve the problem of rights confirmation for AI-generated assets, and achieve generation quality and tag accuracy that are superior to existing technologies, significantly improving collaborative efficiency. Through the standardized generation of three modules: characters, props, and scenes, it transforms scattered and random AI-generated content into standardized, storable, and reusable digital assets, providing stable and high-quality materials for AIGC video creation.

[0053] 2. An innovative blockchain-zero-knowledge proof fusion solution enables full-process traceability of digital assets, resolving the contradiction between privacy protection and transparent traceability. At the same time, it combines intellectual property rights confirmation and rights division to activate the value of ownership, break through the limitations of existing traceability technologies, and provide a reliable basis for copyright tracing and compliance auditing by on-chaining authorized material access and compliance information for image assets.

[0054] 3. Construct an intelligent value mining and intellectual property evaluation system, combined with graph neural network algorithms, to realize asset value quantification and reuse recommendation, distinguish between asset legal value and accounting value, activate idle assets, significantly reduce the cost of repeated creation for enterprises, recommend historical materials with high reuse potential for video creation projects, and improve the efficiency of creative workflow by more than 300%.

[0055] 4. Adopting a "microservices + service mesh + cloud-native" architecture, it enables automatic adaptation of system parameters and elastic scaling, solving the problems of rigid and poor scalability of existing system architectures, reducing maintenance and expansion costs. The overall technical route is mature and stable, and the underlying layer relies on mainstream multimodal large model APIs, eliminating the need to develop basic models from scratch, and significantly reducing development risks and costs.

[0056] 5. Establish a closed-loop intelligent optimization process, dynamically optimizing system parameters and strategies through reinforcement learning algorithms to overcome the limitations of static management and ensure that system efficiency steadily improves with asset scale. For video creation scenarios, this involves multi-view... Figure 1 Consistent rendering and style unification control solve industry pain points such as character drift and style fragmentation.

[0057] 6. Innovate a fine-grained security and compliance management mechanism to achieve hierarchical access control for individual assets, integrate encryption, anomaly monitoring and compliance auditing, comprehensively prevent security and compliance risks, and meet the compliance needs of multiple industries. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is an overall architecture diagram of the digital asset generation and management system in an embodiment of the present invention;

[0060] Figure 2 This is a schematic diagram of the internal structure of the intelligent generation module in an embodiment of the present invention;

[0061] Figure 3 This is a schematic diagram of the internal structure of the collaborative management module in an embodiment of the present invention;

[0062] Figure 4 This is a schematic diagram of the internal structure of the blockchain traceability module in an embodiment of the present invention;

[0063] Figure 5 This is a schematic diagram of the internal structure of the intelligent value mining module in an embodiment of the present invention;

[0064] Figure 6 This is a schematic diagram of the internal structure of the adaptive adaptation module in an embodiment of the present invention;

[0065] Figure 7 This is a schematic diagram of the internal structure of the security control module in an embodiment of the present invention;

[0066] Figure 8 This is a schematic diagram of the structure of the three generation modules for characters, props, and scenes for AIGC image creation in an embodiment of the present invention. Detailed Implementation

[0067] 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 a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0068] To achieve the aforementioned objectives, this invention provides a digital asset generation and management system, comprising: an intelligent generation module, a collaborative management module, a blockchain traceability module, an intelligent value mining module, an adaptive adaptation module, and a security control module. Each module achieves data interoperability and collaborative work through a distributed bus architecture, forming a closed-loop system of "generation-management-traceability-value mining-security control." Simultaneously, the system also includes an edge computing module, used to offload some generation, detection, and retrieval tasks to edge nodes, reducing the pressure on cloud servers, improving task processing speed, and achieving low-latency response.

[0069] like Figure 2 As shown, the intelligent generation module includes a requirement analysis unit, a multimodal generation unit, a dynamic parameter optimization unit, and a quality detection and correction unit.

[0070] The requirement parsing unit receives the generation requirements input by the user. For example, if the user inputs "generate 10 e-commerce women's clothing product images, requiring high definition, minimalist style, and a size of 800*800px", the unit will use natural language processing algorithms to parse the requirement keywords (e-commerce, women's clothing, product images, high definition, minimalist, 800*800px) and generate standardized generation instructions to ensure the accuracy of requirement parsing.

[0071] The multimodal generation unit has built-in image generation, audio generation, text generation, and virtual asset generation models, supporting the simultaneous generation of multiple types of digital assets. For example, if a user needs to generate an image of women's clothing with product description text, the multimodal generation unit can generate both the image and text simultaneously and achieve accurate matching between the two. At the same time, the multimodal generation model uses a transfer learning algorithm, which can optimize the performance of the generation model based on the user's historical generation data, improve the matching degree between the generated assets and the user's needs, and also supports users to upload custom models to adapt to the generation needs of special scenarios.

[0072] The dynamic parameter optimization unit is based on reinforcement learning algorithms. It collects data in real time during the generation process (including generation speed, quality feedback, resource utilization, etc.) and dynamically adjusts the parameters of the generation model. For example, when the generation speed is detected to be slow, the generation accuracy is automatically reduced (provided that the quality standards are met) and the generation speed is increased; when the generation quality is detected to be unsatisfactory, the generation accuracy is automatically increased to ensure a balance between generation efficiency and generation quality.

[0073] The quality inspection and correction unit establishes a multi-dimensional quality assessment index system (including clarity, completeness, compliance, relevance, etc.). It uses computer vision technology to detect the clarity and completeness of images, text review technology to detect the compliance of text, and semantic analysis technology to detect the relevance of generated assets to requirements. If assets that do not meet quality standards are detected (such as blurry images, text violations, or irrelevance to requirements), a correction instruction is automatically triggered to adjust the generation parameters and regenerate the asset. No manual intervention is required, and the detection accuracy rate reaches over 95%.

[0074] like Figure 3 As shown, the collaborative management module includes an asset classification unit, a standardization processing unit, a version control unit, an intelligent retrieval unit, and a collaborative operation unit.

[0075] The asset classification unit adopts a semantic graph-based intelligent classification algorithm, combined with user-defined classification rules, to automatically classify and archive the generated digital assets. For example, the generated women's clothing product images are classified under the "e-commerce-women's clothing-product images" directory. At the same time, asset association relationships are established, associating women's clothing product images with corresponding product description text and audio assets, so as to realize quick association and query of similar and related assets.

[0076] The standardization processing unit automatically performs standardization processing on digital assets of different formats, such as format unification, size adjustment, and data compression. For example, it converts product images of different formats (JPG, PNG) into JPG format, adjusts the size to 800*800px, compresses data, optimizes storage structure, and reduces storage costs.

[0077] The version control unit uses incremental backup and version marking technology to record every modification operation of digital assets. For example, after a user modifies the color of a women's clothing product image, the system automatically generates a new version and marks the modifier, modification time, and modification content. It supports version rollback and version comparison to avoid version confusion. At the same time, version permissions can be set to prevent unauthorized modifications.

[0078] The intelligent search unit supports keyword search, semantic search, image similarity search, and cross-modal search. For example, if a user enters "simple style women's clothing product images", relevant images can be quickly retrieved; if a user uploads a women's clothing product image, other product images with similar styles can be retrieved; if a user enters "women's clothing product introduction audio", relevant audio assets and corresponding image and text assets can be retrieved; at the same time, combined with intelligent recommendation algorithms, digital assets that may be needed can be pushed according to the user's usage habits, reducing the asset search time from an average of 30 minutes to 2 minutes.

[0079] The collaborative operation unit supports multi-user and cross-departmental collaborative operations. For example, after creators generate product images, they can share the assets with reviewers. After the reviewers approve the images, they can then share them with operations staff. Users can view the processing progress and operation records of the assets in real time, thereby improving the efficiency of cross-departmental collaboration.

[0080] like Figure 4 As shown, the blockchain traceability module includes an information collection unit, a blockchain writing unit, a traceability query unit, and an auditing unit;

[0081] The information collection unit collects real-time information on the entire process of digital assets from generation to destruction, including the source of generation (generator, generation device, generation time), modification records, circulation path (recipient, circulation time), usage (usage scenario, number of times used), destruction records, etc., to ensure the integrity of traceability information;

[0082] The blockchain writing unit adopts a consortium blockchain architecture and combines the elastic scalability of Alibaba Cloud BaaS platform. After encrypting the collected traceability information, it writes it into the blockchain node. By utilizing the immutable and distributed storage characteristics of blockchain, the authenticity and security of traceability information are ensured, and the traceability information is prevented from being tampered with. At the same time, the traceability nodes can be expanded as needed to reduce deployment costs.

[0083] The traceability query unit allows users to quickly query the full-process traceability information of digital assets by keywords such as asset ID, generation time, and asset name, and generate a visual traceability report that clearly shows the asset's circulation trajectory and operation records. For example, when a user queries the traceability information of a women's clothing product image, they can clearly see information such as the person who created the product image, the generation time, modification records, and the sharing objects.

[0084] The audit unit automatically performs audit analysis on traceability information, identifies abnormal operations (such as unauthorized modification or illegal transfer), generates audit reports, provides data support for asset security management, and enables accountability. For example, when an unauthorized user modifies a product image, the audit unit records the abnormal operation, generates an audit report, and notifies the administrator to handle it promptly.

[0085] like Figure 5 As shown, the intelligent value mining module includes a data acquisition unit, a correlation analysis unit, a value assessment unit, and a reuse recommendation unit.

[0086] The data acquisition unit collects real-time usage data of digital assets (including usage frequency, usage scenarios, usage effects, user reviews, etc.), storage data (including storage duration, storage costs, etc.), and related data (including relationships with other assets and matching degree with business scenarios, etc.).

[0087] The correlation analysis unit uses graph neural network algorithms to analyze the correlation between digital assets and the matching degree between assets and business scenarios, and to explore potential asset reuse scenarios. For example, if the analysis shows that a simple style women's clothing product image is frequently used in multiple promotional activities and is highly matched with the summer promotion scenario, the reuse potential of the product image in subsequent summer promotion activities can be explored.

[0088] The value assessment unit establishes a digital asset value assessment model, which combines indicators such as usage frequency, reuse potential, creation cost, and business contribution to conduct quantitative value assessment of digital assets and generate value assessment reports. For example, a women's clothing product image with high usage frequency and high reuse potential is assessed as having a high value, while redundant assets with low usage frequency and no reuse potential are assessed as having a low value, helping companies understand the distribution of asset value.

[0089] Based on the value assessment results and the user's business needs, the reuse recommendation unit pushes high-value digital assets with high reuse potential. For example, when a user needs to generate product images for a summer promotion, the system pushes women's clothing product images that have been previously assessed as high-value and match the summer promotion scenario for the user to reuse, avoiding repetitive creation and improving the efficiency of the enterprise's creative workflow by more than 300%.

[0090] like Figure 6As shown, the adaptive module includes a requirement identification unit, a parameter adjustment unit, and a module expansion unit;

[0091] The demand identification unit collects users’ business data and operating habits, and combines them with industry characteristics to identify the personalized needs of different industries. For example, it identifies the needs of the e-commerce industry as product image management, batch generation, and fast retrieval, and the needs of the media industry as video asset management, version tracking, and copyright protection.

[0092] The parameter adjustment unit automatically adjusts various system parameters based on the identified personalized needs. For example, for the e-commerce industry, the classification rule is adjusted to "industry-category-product type", and the quality assessment indicators are adjusted to focus on clarity and relevance. For the media industry, the classification rule is adjusted to "industry-content type-publication time", and the quality assessment indicators are adjusted to focus on compliance and completeness, without the need for manual adjustment.

[0093] The module extension unit adopts a microservice architecture, which supports the rapid addition and deletion of functional modules. For example, when an enterprise needs to add a virtual asset generation function, it can quickly add a virtual asset generation module through the module extension unit without making large-scale modifications to the system. At the same time, it supports cloud-native deployment and achieves elastic scaling through K8s + container architecture to cope with sudden peak asset demand.

[0094] like Figure 7 As shown, the security control module includes an access control unit, a data encryption unit, an anomaly monitoring unit, and a security early warning unit;

[0095] The permission management unit adopts a role-based fine-grained permission management mechanism, which divides users into different roles (such as administrators, creators, reviewers, and ordinary users) and assigns different operation permissions to each role. For example, administrators have all operation permissions, creators have the permission to generate and modify assets, reviewers have the permission to review assets, and ordinary users only have the permission to query assets, so as to realize hierarchical protection of assets and avoid unauthorized operations.

[0096] The data encryption unit uses a combination of symmetric and asymmetric encryption to encrypt the digital assets themselves, traceability information, and user data, ensuring data security during storage and transmission and preventing data leakage.

[0097] The anomaly monitoring unit monitors user behavior and system operation status in real time, identifies abnormal behaviors (such as multiple failed login attempts, unauthorized asset access, batch download of a large number of assets, etc.), and records anomaly logs.

[0098] Based on anomaly monitoring results, the security early warning unit automatically triggers security alerts (such as SMS alerts and system pop-up alerts) to notify administrators to handle the situation promptly, preventing security risks such as asset theft, tampering, and leakage, while reducing the brand compliance violation rate by more than 85%.

[0099] In this embodiment, the intelligent generation module, targeting AIGC video creation scenarios, achieves precise generation based on three main modules: characters, props, and scenes.

[0100] Character generation: Users can choose to authorize characters from the asset library or submit customized requests. The system verifies the compliance of the assets through authorization comparison, generates character assets from different angles using multi-view rendering, and ensures that the style of characters throughout the project is consistent through a style unification algorithm, thus solving the problem of character drift.

[0101] Prop generation: Based on authoritative reference image library, high-precision assets are generated according to film and television prop standards, supporting multi-angle view output, automatically adapting to the style requirements of different projects, and enabling cross-project reuse;

[0102] Scene generation: For cultural and tourism landmarks, historical buildings and other scenes, high-precision digital scene assets are generated by using real-scene replication and AI restoration technology, supporting the restoration of ancient buildings, scene restoration and asset management;

[0103] Meanwhile, the system integrates five major tools: intelligent image generation, multi-image fusion creation, perspective switching, style transfer drawing, and professional camera language, providing creators with a one-stop capability for generating video materials.

[0104] The workflow of this embodiment is as follows:

[0105] 1. Generation stage: Users input their digital asset generation requirements through the system. The requirement parsing unit parses the requirements and generates standardized instructions. The multimodal generation unit generates digital assets according to the instructions. The dynamic parameter optimization unit adjusts the generation parameters in real time. The quality detection and correction unit automatically detects and corrects the generated assets to ensure generation quality.

[0106] 2. Management Phase: The collaborative management module classifies, standardizes, and manages the versions of the generated digital assets. Users can quickly retrieve assets through the intelligent search unit and achieve cross-departmental collaboration through the collaborative operation unit.

[0107] 3. Traceability Phase: The blockchain traceability module collects asset information throughout the entire process in real time and writes it to the blockchain node. Users can query traceability information through the traceability query unit, and the audit unit performs anomaly audits.

[0108] 4. Value Mining Stage: The intelligent value mining module collects asset usage data, analyzes correlations, evaluates asset value, pushes reuse suggestions, and activates idle assets;

[0109] 5. Security Management Phase: The security management module ensures asset security through fine-grained access control, data encryption, and anomaly monitoring and early warning.

[0110] 6. Adaptation and Expansion Phase: The adaptive adaptation module automatically adjusts system parameters according to industry needs, supports the expansion of functional modules, and enables low-latency response from the edge computing module.

[0111] This embodiment achieves intelligent, collaborative, and secure management of multimodal digital assets throughout their entire lifecycle through the above technical solution. It solves the core defects of existing technologies, has outstanding creativity and practicality, and can be widely applied to multiple industries such as e-commerce, media, healthcare, and metaverse.

[0112] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A digital asset generation and management system, characterized in that, include: The system comprises an intelligent generation module, a collaborative management module, a blockchain traceability module, an intelligent value mining module, an adaptive adaptation module, and a security control module. These modules communicate and collaborate through a distributed bus architecture, forming a closed-loop system of "generation-management-traceability-value mining-security control". The intelligent generation module is used to receive user generation requests, realize the intelligent generation of multiple types of digital assets based on multimodal generation models and dynamic parameter optimization algorithms, and complete the automated detection and correction of generation quality in real time. The intelligent generation module provides unified, compliant and reusable digital materials for AIGC image creation. As the core tool layer of image production, it ensures the stability of work quality from the source. The collaborative management module is used to manage the generated digital assets throughout their entire lifecycle, including asset classification, standardization, version control, multi-dimensional retrieval, and cross-departmental collaborative operations. The blockchain traceability module is used to record the entire process information of digital assets from generation, modification, circulation to destruction. Based on the immutability of blockchain, it achieves the accuracy and security of asset traceability and supports full-process operation auditing and accountability. The intelligent value mining module is used to intelligently analyze the usage data and relationships of digital assets, mine the value potential of the assets, and generate value assessment reports and reuse suggestions. The adaptive module is used to automatically adjust system parameters and functional modules according to the business needs of different industries, adapt to the management needs of multimodal digital assets, and achieve functional expansion without large-scale modification. The security management module adopts a role-based fine-grained permission management mechanism, combined with data encryption and abnormal behavior monitoring technologies, to achieve hierarchical protection, access control and security early warning of digital assets.

2. The digital asset generation and management system according to claim 1, characterized in that, The intelligent generation module includes a demand parsing unit, a multimodal generation unit, a dynamic parameter optimization unit, and a quality detection and correction unit. The requirement parsing unit is used to receive the generation requirements input by the user, parse the requirement keywords through natural language processing algorithms, and generate standardized generation instructions. The generation instructions support structured requirement parsing for professional scenarios such as film and television production, character design, and script visualization. The multimodal generation unit has built-in image generation model, audio generation model, text generation model, and virtual asset generation model, which supports the synchronous generation of multiple types of digital assets, and the models can communicate with each other. The layer relies on mainstream multimodal large model APIs such as nanobanana, gptimage2, and gpt5, and adopts a modular architecture, designed according to three major modules: characters, props, and scenes, which can efficiently generate high-precision digital assets. Character generation unit: Supports dual modes of material library access and customization, ensures compliance through authorization comparison and element blending, and solves the problems of character drift and poor stability with multi-view output and unified style rendering; Item generation unit: Covers the generation of all types of items, relies on authoritative reference image library to ensure accurate details, supports multi-view output and style consistency, and can be reused across projects; Scene Generation Unit: Focuses on high-precision restoration and rendering of cultural and tourism landmarks, historical buildings, and regional culture, realizing real-scene replication, ancient building restoration, and asset management; Meanwhile, the multimodal generation unit integrates five major creative tools: intelligent image generation, multi-image fusion creation, perspective conversion, style transfer, and professional lens language, providing AIGC image creation with one-stop material generation capabilities; The dynamic parameter optimization unit, based on reinforcement learning algorithm, collects data in real time during the generation process and dynamically adjusts the parameters of the generation model to achieve a balance between generation efficiency and generation quality. The quality inspection and correction unit establishes a multi-dimensional quality assessment index system. Through technologies such as computer vision, audio recognition, and text review, it automatically inspects the generated digital assets. If an asset that does not meet the quality standards is detected, a correction instruction is automatically triggered to adjust the generation parameters and regenerate the asset. For video creation scenarios, the quality inspection and correction unit focuses on verifying the consistency of characters, the precision of prop details, the scene reproduction, and the uniformity of style to ensure that the generated assets meet film and television production standards.

3. The digital asset generation and management system according to claim 1, characterized in that, The collaborative management module includes an asset classification unit, a standardization processing unit, a version control unit, an intelligent retrieval unit, and a collaborative operation unit; The asset classification unit adopts a semantic graph-based intelligent classification algorithm, combined with user-defined classification rules, to automatically classify and archive digital assets, while establishing asset association relationships. For video creation assets, it supports a multi-level classification system based on "characters / props / scenes-themes-projects" to achieve rapid classification and retrieval of materials. The standardization processing unit automatically performs standardization processing on digital assets of different formats, such as format unification, size adjustment, and data compression, to generate asset files that conform to industry standards. The version control unit uses incremental backup and version marking technology to record every modification operation of digital assets, supports version rollback and version comparison, and can set version permissions. The intelligent retrieval unit supports keyword retrieval, semantic retrieval, image similarity retrieval, and cross-modal retrieval. Combined with intelligent recommendation algorithms, it pushes digital assets that users may need based on their usage habits. The collaborative operation unit supports multi-user, cross-departmental collaborative operations, including functions such as asset sharing, permission allocation, task assignment, and progress tracking.

4. The digital asset generation and management system according to claim 1, characterized in that, The blockchain traceability module includes an information collection unit, a blockchain writing unit, a traceability query unit, and an auditing unit. The information collection unit collects information on the entire process of digital assets from generation to destruction in real time, ensuring the integrity of traceability information. For image assets, it simultaneously collects compliance information such as authorization source, material call records, and modification traces, providing data support for copyright tracing. The blockchain writing unit adopts a consortium blockchain architecture, which encrypts the collected traceability information and writes it into the blockchain node, and combines it with the Alibaba Cloud BaaS platform to achieve elastic scaling. The traceability query unit allows users to quickly query the full-process traceability information of digital assets using keywords such as asset ID, generation time, and asset name, and generate a visual traceability report. The audit unit automatically performs audit analysis on traceability information, identifies abnormal operations, generates audit reports, and provides data support for asset security management.

5. A digital asset generation and management system according to claim 1, characterized in that, The intelligent value mining module includes a data acquisition unit, a correlation analysis unit, a value assessment unit, and a reuse recommendation unit; The data acquisition unit collects usage data, storage data, and associated data of digital assets in real time. The association analysis unit uses a graph neural network algorithm to analyze the relationship between digital assets and the matching degree between assets and business scenarios, and to explore potential asset reuse scenarios. For video assets, it can analyze the reuse potential of characters / props / scenes in different themes and projects, thereby reducing the cost of repeated creation. The value assessment unit establishes a digital asset value assessment model, combines indicators such as usage frequency, reuse potential, creation cost, and business contribution to conduct quantitative value assessment of digital assets, and generates a value assessment report. The reuse recommendation unit pushes high-value, high-reuse-potential digital assets based on the value assessment results and the user's business needs.

6. A digital asset generation and management system according to claim 1, characterized in that, The adaptive module includes a requirement identification unit, a parameter adjustment unit, and a module expansion unit. The demand identification unit, by collecting users' business data and operating habits, combined with industry characteristics, identifies the personalized needs of different industries. It can identify the image creation needs of different themes such as film and television, cultural tourism, intangible cultural heritage, and commerce, and automatically adapt the corresponding generation strategies and quality standards. The parameter adjustment unit automatically adjusts various system parameters based on the identified personalized needs, without requiring manual adjustment. The module expansion unit adopts a microservice architecture, supports the rapid addition and deletion of functional modules, supports cloud-native deployment, and achieves elastic expansion through Kubernetes + container architecture.

7. A digital asset generation and management system according to claim 1, characterized in that, The security control module includes an access control unit, a data encryption unit, an anomaly monitoring unit, and a security early warning unit. The permission management unit adopts a role-based fine-grained permission management mechanism, which divides users into different roles and assigns different operation permissions to each role. The data encryption unit uses a combination of symmetric and asymmetric encryption to encrypt the digital assets themselves, traceability information, and user data. The anomaly monitoring unit monitors user actions and system operating status in real time, identifies abnormal behaviors, and records anomaly logs. The security early warning unit automatically triggers a security early warning based on the abnormal monitoring results, and notifies the administrator to handle the situation in a timely manner.

8. A digital asset generation and management system according to claim 1, characterized in that, It also includes an edge computing module, which is used to offload some generation, detection and retrieval tasks to edge nodes, reduce the pressure on cloud servers, improve task processing speed, achieve low latency response, and synchronize data with the cloud system to ensure data consistency.

9. A digital asset generation and management system according to claim 2, characterized in that, The multimodal generation model employs a transfer learning algorithm, which can optimize the performance of the generation model based on the user's historical generation data. It also supports users uploading custom models to adapt to the generation needs of special scenarios.

Images