A blockchain-based AI digital content copyright registration method, device, and medium

Abstract

This application discloses a blockchain-based AI digital content copyright notarization method, device, and medium, relating to the field of copyright authentication technology. The method includes: acquiring data to be authenticated and storing it on a blockchain; wherein the data to be authenticated includes text data, image data, and audio / video data; inputting the data to be authenticated into a preset multimodal feature extraction model to generate target authentication data; verifying the ownership of the target authentication data based on a preset smart contract and generating a verification result; and writing the authenticated target authentication data and the verification result into the blockchain. This application improves the efficiency of AI multimodal digital content copyright notarization by rapidly processing different types of data through a multimodal feature extraction model and automatically executing verification operations through a smart contract.

Description

Technical Field

[0001] This application relates to the field of copyright authentication technology, and in particular to a blockchain-based AI digital content copyright notarization method, device and medium. Background Technology

[0002] In today's digital age, multimodal big data models are flourishing, enabling the collaborative generation of digital content such as text, images, audio, and video, greatly improving content production efficiency and enriching the digital content ecosystem. However, with the rapid development of AI-powered multimodal digital content generation technology, abuses such as deepfake videos and voice imitation are becoming increasingly rampant, seriously threatening the security and stability of the content ecosystem.

[0003] Traditional blockchain-based evidence storage solutions exhibit significant limitations when handling multimodal feature fusion and massive amounts of data. For example, a single-chain structure can only support text hash storage, making it difficult to preserve key ownership identifiers such as video watermarks and voiceprint features. Furthermore, as the number of nodes increases, the evidence storage latency grows linearly, affecting the efficiency of evidence storage.

[0004] Therefore, how to register copyrights for AI-generated multimodal digital content has become an urgent technical problem to be solved. Summary of the Invention

[0005] This application provides a blockchain-based method, device, and medium for copyright registration of AI digital content, which addresses the technical problem of how to register copyright for multimodal AI digital content.

[0006] In a first aspect, embodiments of this application provide a blockchain-based method for AI digital content copyright notarization. The method includes: acquiring data to be certified and storing the data to be certified on a blockchain; wherein the data to be certified includes text data, image data, and audio / video data; inputting the data to be certified into a preset multimodal feature extraction model to generate target certification data; verifying the ownership of the target certification data based on a preset smart contract and generating a verification result; and writing the certified target certification data and the verification result into the blockchain.

[0007] In one implementation of this application, the data to be authenticated is input into a preset multimodal feature extraction model to generate target authentication data. Specifically, this includes: extracting features from the data to be authenticated to obtain multimodal feature vectors; dynamically weighting and aggregating the multimodal feature vectors using a gated fusion network based on a cross-modal attention mechanism to generate a joint feature vector; and performing a hierarchical hash operation on the joint feature vector to obtain the target authentication data.

[0008] In one implementation of this application, feature extraction is performed on the data to be authenticated to obtain a multimodal feature vector. Specifically, this includes: performing natural language feature encoding on text data to generate a text feature vector; performing visual feature encoding on image data to generate an image feature vector; performing spatiotemporal feature encoding on audio and video data to generate a joint audio and video feature vector; and correlating the text feature vector, image feature vector, and joint audio and video feature vector to obtain a multimodal feature vector.

[0009] In one implementation of this application, based on a cross-modal attention mechanism, a gated fusion network is used to dynamically weight and aggregate multimodal feature vectors to generate a joint feature vector. Specifically, this includes: calculating cross-modal attention scores among text feature vectors, image feature vectors, and audio / video joint feature vectors based on the cross-modal attention mechanism, and generating weight coefficients corresponding to the cross-modal attention scores; inputting the weight coefficients into the gated fusion network, and dynamically correcting the weight coefficients based on a preset gating mechanism in the gated fusion network; and performing a weighted summation of the multimodal feature vectors based on the corrected weight coefficients to generate a joint feature vector.

[0010] In one implementation of this application, ownership verification of the target authentication data is performed based on a preset smart contract, and a verification result is generated. Specifically, this includes: sending the target authentication data from the sidechain to the main chain based on a preset main chain-side chain cross-chain evidence storage network transmission protocol; performing basic verification on the format integrity and data source legality of the target authentication data according to the verification consensus rules of the main chain; after the basic verification is passed, activating the preset smart contract in the main chain to send the verification process data back to the sidechain through the cross-chain evidence storage network transmission protocol; and analyzing the verification process based on the sidechain's multi-factor authentication mechanism and determining the ownership status of the target authentication data to generate a verification result.

[0011] In one implementation of this application, a multi-factor authentication mechanism based on a sidechain is used to analyze the verification process and determine the ownership status of the target authentication data to generate a verification result. Specifically, this includes: decrypting the digital signature based on the creator's public key in the target authentication data to generate a signature validity identifier; extracting the biometric hash value from the target authentication data and calculating its similarity with the registration features pre-stored in an off-chain security database to generate a biometric matching identifier; accessing an external copyright registration system through an on-chain oracle module to query the off-chain ownership status based on the content fingerprint of the target authentication data to generate an ownership registration validity identifier; if the signature validity identifier, biometric matching identifier, and ownership registration validity identifier all return a pass status, an ownership validity verification result is generated and the on-chain storage operation is triggered; otherwise, an ownership anomaly alarm result containing failure factor types is generated.

[0012] In one implementation of this application, after generating the verification result, the method further includes: if an ownership anomaly alarm result is received, generating a derivative work identifier and the corresponding derivative work based on the content fingerprint in the target authentication data; storing the derivative work identifier on the blockchain, and sending the derivative work to the copyright detection node for secondary copyright review.

[0013] In one implementation of this application, the target authentication data that has passed authentication is written into the blockchain along with the verification result. Specifically, this includes: extracting the content fingerprint of the target authentication data, generating a random obfuscation factor and calculating anonymized content hash; encrypting the target authentication data using a symmetric encryption algorithm and binding it with the anonymized content hash to generate a privacy storage package; writing the privacy storage package into the sidechain through a cross-chain storage network and simultaneously writing the anonymized content hash into the main chain.

[0014] Secondly, this application also provides a blockchain-based AI digital content copyright registration device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: acquire data to be authenticated and store the data to be authenticated on the blockchain; wherein the data to be authenticated includes text data, image data, and audio / video data; input the data to be authenticated into a preset multimodal feature extraction model to generate target authentication data; perform ownership verification on the target authentication data based on a preset smart contract and generate a verification result; and write the authenticated target authentication data and the verification result into the blockchain.

[0015] Thirdly, this application also provides a non-volatile computer storage medium for AI digital content copyright registration based on blockchain, storing computer-executable instructions. The computer-executable instructions are configured to: acquire data to be authenticated and store the data to be authenticated on the blockchain; wherein the data to be authenticated includes text data, image data, and audio / video data; input the data to be authenticated into a preset multimodal feature extraction model to generate target authentication data; perform ownership verification on the target authentication data based on a preset smart contract and generate a verification result; and write the authenticated target authentication data and the verification result into the blockchain.

[0016] The embodiments of this application provide a blockchain-based AI digital content copyright notarization method, device, and medium, which have the following beneficial effects: Utilizing the decentralized and tamper-proof characteristics of blockchain, the authenticity and integrity of the data are ensured; through a preset multimodal feature extraction model, key features can be extracted from different types of data, and target authentication data can be generated, automatically identifying and extracting important information from the data, avoiding the subjectivity and inaccuracy of manual feature extraction; the multimodal feature extraction model can quickly process different types of data, smart contracts can automatically execute verification operations, and the distributed storage and consensus mechanism of blockchain can quickly complete data writing, effectively improving the efficiency of AI multimodal digital content copyright notarization. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 A flowchart illustrating a blockchain-based AI digital content copyright notarization method provided for embodiments of this application;

[0019] Figure 2 This is a schematic diagram of the internal structure of an AI-based digital content copyright certificate storage device based on blockchain, provided as an embodiment of this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] This application provides a blockchain-based method, device, and medium for copyright registration of AI digital content, which addresses the technical problem of how to register copyright for multimodal AI digital content.

[0022] The technical solutions proposed in the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0023] Figure 1 This document presents a flowchart of a blockchain-based AI digital content copyright registration method. Figure 1 As shown in the figure, the AI ​​digital content copyright notarization method based on blockchain provided in this application embodiment specifically includes the following steps:

[0024] Step 10: Obtain the data to be authenticated and store it on the blockchain.

[0025] In this step, the data to be authenticated can be digital content in various forms, such as text, images, audio, or video. This data may come from different creators, such as individual creators, enterprise users, or content generation platforms. The core characteristics of blockchain are decentralization, immutability, and traceability. Decentralization means there is no single control center; the security and integrity of the data are maintained collectively by the entire network. Immutability means that once data is written to the blockchain, it cannot be modified or deleted. Traceability means that the source and history of the data can be traced through the blockchain's chain structure. Blockchain is a distributed ledger technology that stores data on multiple nodes through cryptographic algorithms and consensus mechanisms. Each node holds a complete copy of the ledger, and these copies are interconnected and synchronized through the network. Storing the data to be authenticated on the blockchain prevents modification or deletion, providing strong guarantees for data security and trustworthiness.

[0026] Step 20: Input the data to be authenticated into the preset multimodal feature extraction model to generate target authentication data.

[0027] As an optional embodiment, the data to be authenticated is input into a preset multimodal feature extraction model to generate target authentication data, which may specifically include: Step 201: extracting features from the data to be authenticated to obtain a multimodal feature vector.

[0028] As an optional embodiment, feature extraction is performed on the data to be authenticated to obtain a multimodal feature vector, which may specifically include: Step 2011: Natural language feature encoding is performed on the text data to generate a text feature vector.

[0029] In this step, natural language feature encoding refers to transforming the semantic information in text data into numerical feature vectors using specific algorithms or models. The purpose of this process is to represent the complex semantics in the text in a form that computers can understand and process. Since text data usually contains rich semantic information, such as the meaning of words, sentence structure, and the topic of the text, preprocessing the text data using natural language processing techniques, including word segmentation, stop word removal, and part-of-speech tagging, helps to extract key information from the text and remove irrelevant noise. The preprocessed text data is then feature-encoded using the BERT language model to capture the semantic information in the text and transform it into a high-dimensional feature vector, which facilitates subsequent fusion and analysis with data from other modalities.

[0030] Step 2012: Encode the image data visual features to generate an image feature vector.

[0031] In this step, visual feature encoding refers to converting the visual information in image data into numerical feature vectors using specific algorithms or models. The purpose of this process is to represent information such as color, texture, shape, and spatial relationships in the image in a form that a computer can understand and process. Since image data usually contains rich visual information, computer vision technology is used to normalize, crop, scale, and perform other operations on the image data. The preprocessed image data is then feature-encoded using the ViT visual model to capture the visual information in the image and convert it into a high-dimensional feature vector. The generated image feature vector can reflect the visual content of the image, which is convenient for subsequent fusion and analysis with data from other modalities.

[0032] Step 2013: Perform spatiotemporal feature encoding on the audio and video data to generate a joint audio and video feature vector.

[0033] In this step, spatiotemporal feature encoding refers to converting the temporal and spatial information in audio and video data into numerical feature vectors using specific algorithms or models. Since audio and video data not only contains audio signals but also time-related dynamic visual information, it is necessary to consider features in both time and space dimensions simultaneously. The audio signal is denoised, the video signal is extracted from video frames, and the preprocessed audio and video signals are feature encoded using the Wav2Vec2 audio and video model to capture the temporal and spatial features and convert them into a joint audio and video feature vector, reflecting the comprehensive content of the audio and video data, which facilitates subsequent fusion and analysis with data from other modalities.

[0034] Step 2014: Correlate the text feature vector, image feature vector, and audio / video joint feature vector to obtain a multimodal feature vector.

[0035] In this step, the transformed text feature vector, image feature vector, and audio-video joint feature vector are fused into multimodal features. This fully utilizes the complementary information between different modal data to improve the overall understanding and representation of digital content, providing strong feature support for subsequent operations such as ownership authentication, infringement detection, and content tracing.

[0036] Step 202: Based on the cross-modal attention mechanism, use a gated fusion network to dynamically weight and aggregate multimodal feature vectors to generate a joint feature vector.

[0037] As an optional embodiment, based on the cross-modal attention mechanism, the multimodal feature vectors are dynamically weighted and aggregated using a gated fusion network to generate a joint feature vector. Specifically, it may include: Step 2021: Based on the cross-modal attention mechanism, calculate the cross-modal attention score between the text feature vector, image feature vector and audio-video joint feature vector, and generate the weight coefficients corresponding to the cross-modal attention score.

[0038] In this step, the cross-modal attention mechanism is a method for processing multimodal data. Its purpose is to determine the importance of each modal feature by calculating the relationship between different modal feature vectors. This scheme inputs text feature vectors, image feature vectors, and audio-video joint feature vectors into the cross-modal attention mechanism to evaluate the semantic correlation between different modal feature vectors and obtain the attention score of each modal feature. The weight coefficient corresponding to the cross-modal attention score is obtained through normalization. This weight coefficient is used in the subsequent feature fusion process to determine the contribution of each modal feature in the fusion result.

[0039] Step 2022: Input the weight coefficients into the gated fusion network and dynamically adjust the weight coefficients based on the preset gating mechanism in the gated fusion network.

[0040] In this step, the gated fusion network is a neural network structure used to dynamically adjust the weights of feature fusion. Its core idea is to dynamically adjust the weights of each modality feature based on the characteristics of the current input data through a gating mechanism, thereby achieving more flexible feature fusion. In this scheme, the generated weight coefficients are input into the gated fusion network, and the gating mechanism adjusts the weight coefficients of each modality feature according to the characteristics of the current input data (such as content type, task requirements, etc.). For example, if the current task focuses more on image information, the gating mechanism will increase the weight coefficients of image features while appropriately decreasing the weight coefficients of other modality features. In complex multimodal data processing scenarios, this scheme can better capture the complex relationships between various modal data by dynamically adjusting the weight coefficients, making the feature fusion process more adaptable to different scenarios and needs.

[0041] Step 2023: Based on the corrected weight coefficients, perform a weighted summation of the multimodal feature vectors to generate a joint feature vector.

[0042] In this step, the text feature vector, image feature vector, and audio / video joint feature vector are weighted and summed according to the corrected weight coefficients. This can highlight the important features of different data to be authenticated, while retaining information from other features, thereby generating a more representative and adaptive joint feature vector.

[0043] Step 203: Perform a hierarchical hash operation on the joint feature vector to obtain the target authentication data.

[0044] In this step, hierarchical hashing is a process that gradually simplifies complex data structures into fixed-length hash values. Through multi-level hashing, the joint feature vector is gradually transformed into a simpler and more unique hash value. This hierarchical processing method can effectively handle high-dimensional feature vectors while ensuring that the final hash value can uniquely identify the original data. It not only preserves the key information of the original data but also ensures the uniqueness and immutability of the data, providing a reliable foundation for subsequent blockchain storage and data authentication.

[0045] Step 30: Based on the preset smart contract, verify the ownership of the target authentication data and generate the verification result.

[0046] As an optional embodiment, based on a preset smart contract, the ownership of the target authentication data is verified and a verification result is generated. Specifically, it may include: Step 301: Based on a preset main chain-side chain cross-chain evidence storage network transmission protocol, the target authentication data is sent from the side chain to the main chain.

[0047] In this step, the proposed solution constructs a dual-chain evidence storage network consisting of a main chain and side chains. The main chain is responsible for global supervision and storage of core data, while the side chains are used to process specific business data and evidence storage information. The cross-chain evidence storage network transmission protocol is a pre-defined communication mechanism used to securely and efficiently transmit data between the main chain and side chains. It defines the data transmission format, security mechanism, verification process, and specific rules for cross-chain interoperability, ensuring that the data transmission between the two blockchains is reliable and consistent, and preventing data loss, tampering, or errors during transmission.

[0048] Step 302: Based on the main chain's verification consensus rules, perform basic verification on the format integrity and data source legality of the target authentication data.

[0049] In this step, the main chain's verification consensus rules are a mechanism to ensure the authenticity and integrity of data. These rules define the conditions that target authentication data must meet before it is accepted and stored. After receiving the target authentication data, the main chain performs format integrity verification, including checking whether the data conforms to predefined structures and encoding rules. At the same time, the main chain also verifies whether the source of the target authentication data is legitimate. This involves checking whether the data generation process conforms to preset rules and whether the data is sent by authorized sidechain nodes. The legitimacy verification can prevent unauthorized data from entering the main chain.

[0050] Step 303: After the basic verification is passed, the pre-set smart contract in the main chain is invoked to send the verification process data back to the side chain through the cross-chain evidence storage network transmission protocol.

[0051] In this step, the smart contract pre-set in the main chain is specifically designed to process the target authentication data. It includes logic for verifying the authenticity and integrity of the data, as well as the function of recording the verification process and results. When the target authentication data passes the basic verification of the main chain, the smart contract on the main chain will be triggered. The intermediate data and results generated by the smart contract during the execution of the verification operation record the verification process of the target authentication data, including the verification time, verification steps, verification results, etc., and securely transmit the verification process data back to the side chain. This mechanism not only improves the automation and efficiency of copyright authentication, but also ensures the credibility and traceability of the data through the combination of cross-chain protocols and smart contracts, providing solid technical support for the confirmation and protection of digital content rights.

[0052] Step 304: Based on the sidechain multi-factor authentication mechanism, analyze the verification process and determine the ownership of the target authentication data to generate verification results.

[0053] As an optional embodiment, a sidechain-based multi-factor authentication mechanism is used to analyze the verification process and determine the ownership of the target authentication data in order to generate a verification result. Specifically, it may include: Step 3041: Decrypting the digital signature based on the creator's public key in the target authentication data to generate a signature validity identifier.

[0054] In this step, when generating the target authentication data, the creator uses their private key to sign the data. The signed data can be verified during transmission to ensure that the data has not been tampered with and indeed comes from the creator. During the verification process, the creator's public key is used to decrypt the digital signature in the target authentication data. The public key and private key are a key pair; the public key is used for decryption, and the private key is used for signing. By decrypting the signature, the validity of the signature can be verified. If the decrypted data matches the target authentication data, it means that the signature is valid, and a signature validity identifier is generated.

[0055] Step 3042: Extract the biometric hash value from the target authentication data and perform similarity calculation with the registration features pre-stored in the off-chain security database to generate a biometric matching identifier.

[0056] In this step, the biometric hash value is a unique identifier generated by hashing the creator's biometric features (such as fingerprints, voiceprints, facial recognition, etc.) to verify the creator's identity and ensure that the target authentication data matches the creator's biometric features. The biometric hash value is extracted from the target authentication data and its similarity is calculated with the registration features pre-stored in the off-chain security database. The off-chain security database stores the creator's biometric hash value. If the biometric hash value in the target authentication data is highly similar to the registration features in the database, it indicates a biometric match, and a biometric match score identifier is generated, indicating that the creator's identity verification is successful.

[0057] Step 3043: Access the external copyright registration system through the on-chain oracle module to query the off-chain ownership status based on the content fingerprint of the target authentication data and generate an ownership registration validity identifier.

[0058] In this step, the on-chain oracle module is a mechanism for interaction between the blockchain and external copyright registration systems. It allows blockchain smart contracts to access external data sources, such as copyright registration systems. By accessing the external copyright registration system through the on-chain oracle module, the content fingerprint in the target authentication data is used for querying. The content fingerprint is a unique identifier for the target authentication data and is used to find the corresponding ownership information in the copyright registration system. Based on the query results, an ownership registration validity identifier is generated. If the copyright registration system confirms that the content fingerprint of the target authentication data has been successfully registered and the ownership is clear, an ownership registration validity identifier is generated, indicating that the ownership status of the target authentication data is valid.

[0059] Step 3044: If the signature validity identifier, biometric matching identifier, and ownership registration validity identifier all return a pass status, then generate an ownership validity verification result and trigger the on-chain storage operation; otherwise, generate an ownership anomaly alarm result containing failure factor types.

[0060] In this step, if the signature validity identifier, biometric matching identifier, and ownership registration validity identifier all return a pass status, it means that the target authentication data has passed verification in terms of signature, biometrics, and ownership registration. At this time, an ownership validity verification result is generated. After the ownership validity verification result is generated, the evidence storage and on-chain operation is triggered. If any identifier fails verification, an ownership anomaly alarm result containing failure factor types is generated. The failure factor types may include invalid signature, biometric mismatch, or invalid ownership registration. The ownership anomaly alarm result is used to notify the administrator or relevant parties for further investigation and processing.

[0061] Step 305: If an ownership anomaly alarm result is received, generate a derivative work identifier and the corresponding derivative work based on the content fingerprint in the target authentication data.

[0062] In this step, if an ownership anomaly alarm is received, it indicates that the target authentication data has failed any verification step, such as signature validity, biometric matching degree, or ownership registration validity. In this case, the content fingerprint in the target authentication data is further analyzed. The content fingerprint is the unique identifier of the target authentication data, which can uniquely identify the content of the target authentication data and is the basis for copyright review and source tracing. A derivative work identifier is generated based on the content fingerprint to uniquely identify derivative works related to the target authentication data. Derivative works refer to new works formed by adaptation, translation, annotation, and collation on the basis of the original work.

[0063] Step 306: Store the derivative work identifier on the blockchain and send the derivative work to the copyright detection node for secondary copyright review.

[0064] In this step, the generated derivative work identifiers are stored on the blockchain as the basis for subsequent copyright review and tracing, providing support for resolving copyright disputes. The copyright detection node is a dedicated node for copyright review, responsible for further copyright review of the target authentication data and its derivative works to confirm their ownership status. If the review passes, it indicates that the ownership of the derivative work is legal; if it fails, a new ownership anomaly alarm result will be generated, prompting further processing. This solution can not only verify the copyright ownership of the work to be certified, but also, for data that fails verification, generate new derivative works through adaptation for secondary certification. Adaptation and innovation under the premise of legality and compliance can stimulate creators' creativity and promote the production of more excellent derivative works.

[0065] Step 40: Write the verified target authentication data and verification results into the blockchain.

[0066] As an optional embodiment, the target authentication data that has passed authentication is written into the blockchain along with the verification result. Specifically, this may include: Step 401: Extracting the content fingerprint of the target authentication data, generating a random obfuscation factor and calculating the anonymized content hash.

[0067] In this step, the random obfuscation factor is a mechanism used to enhance privacy protection. By introducing randomness, it makes the content fingerprint impossible to be directly identified during storage and transmission, thereby protecting data privacy. The content fingerprint is extracted from the target authentication data. The content fingerprint is combined with the random obfuscation factor and an anonymized content hash is generated by calculating a hash function. The anonymized content hash retains the uniqueness of the content fingerprint while enhancing privacy protection through the random obfuscation factor.

[0068] Step 402: Encrypt the target authentication data using a symmetric encryption algorithm and bind it with the anonymized content hash to generate a privacy storage package.

[0069] In this step, symmetric encryption algorithms are a type of encryption technology that uses the same key for both encryption and decryption. They are efficient and secure, and suitable for encrypting large amounts of data. The encrypted target authentication data is then bound to anonymized content hash to generate a privacy-preserving evidence package. This package contains both encrypted data and anonymized content hash, which is used for subsequent storage and verification. Through encryption and anonymization, the privacy-preserving evidence package ensures the privacy and security of the target authentication data during storage and transmission.

[0070] Step 403: Write the privacy-preserving data package to the sidechain via the cross-chain evidence storage network, and simultaneously write the anonymized content hash to the main chain.

[0071] In this step, the privacy-preserving data package is encapsulated according to the protocol requirements of the cross-chain evidence preservation network. The encapsulated privacy-preserving data package is then sent from the main chain to the side chain. Upon receiving the privacy-preserving data package, the side chain stores it in its distributed ledger. Simultaneously with writing the privacy-preserving data package to the side chain, the anonymized content hash is written to the main chain. Upon receiving the anonymized content hash, the main chain stores it in its distributed ledger. The anonymized content hash serves as a unique identifier for the target authentication data, ensuring the security and privacy of the data during storage and transmission. This achieves distributed storage and trusted verification of data, providing solid technical support for the copyright protection of digital content.

[0072] The above are embodiments of the method proposed in this application. Based on the same inventive concept, embodiments of this application also provide a blockchain-based AI digital content copyright storage device, the structure of which is as follows: Figure 2 As shown.

[0073] Figure 2 This is a schematic diagram of the internal structure of an AI-based digital content copyright registration device based on blockchain, provided as an embodiment of this application. Figure 2 As shown, the device includes:

[0074] At least one processor 201;

[0075] And a memory 202 that is communicatively connected to at least one processor;

[0076] The memory 202 stores instructions executable by at least one processor. These instructions are executed by at least one processor 201 to enable the processor 201 to: acquire data to be authenticated and store the data on the blockchain; wherein the data to be authenticated includes text data, image data, and audio / video data; input the data to be authenticated into a preset multimodal feature extraction model to generate target authentication data; perform ownership verification on the target authentication data based on a preset smart contract and generate a verification result; and write the authenticated target authentication data and the verification result into the blockchain.

[0077] Some embodiments of this application provide corresponding to Figure 1 A non-volatile computer storage medium for AI digital content copyright registration based on blockchain is disclosed. The storage medium contains computer-executable instructions, which are configured to: acquire data to be authenticated and store the data on the blockchain; wherein the data to be authenticated includes text data, image data, and audio / video data; input the data to be authenticated into a preset multimodal feature extraction model to generate target authentication data; perform ownership verification on the target authentication data based on a preset smart contract and generate a verification result; and write the authenticated target authentication data and the verification result into the blockchain.

Claims

1. A blockchain-based method for copyright registration of AI digital content, characterized in that, The method includes: The process involves acquiring data to be authenticated and storing it on a blockchain; the data to be authenticated includes text data, image data, and audio / video data. The data to be authenticated is input into a preset multimodal feature extraction model to generate target authentication data; Based on a preset smart contract, the ownership of the target authentication data is verified, and a verification result is generated. The verified target authentication data and the corresponding verification results are written into the blockchain; Based on a preset smart contract, the ownership of the target authentication data is verified, and a verification result is generated, specifically including: Based on the preset main chain-side chain cross-chain evidence storage network transmission protocol, the target authentication data is sent from the side chain to the main chain; Based on the main chain's verification consensus rules, basic verification is performed on the format integrity and data source legality of the target authentication data; After the basic verification is passed, the pre-set smart contract in the main chain is invoked to send the verification process data back to the side chain through the cross-chain evidence storage network transmission protocol. Based on the multi-factor authentication mechanism of the sidechain, the verification process is analyzed, and the ownership of the target authentication data is determined to generate a verification result. Based on the multi-factor authentication mechanism of the sidechain, the verification process is analyzed, and the ownership of the target authentication data is determined to generate a verification result, specifically including: Decrypt the digital signature based on the creator's public key in the target authentication data to generate a signature validity identifier; Extract the biometric hash value from the target authentication data and perform similarity calculation with the registration features pre-stored in the off-chain security database to generate a biometric matching degree identifier; Access the external copyright registration system through the on-chain oracle module to query the off-chain ownership status based on the content fingerprint of the target authentication data and generate an ownership registration validity identifier. If the signature validity identifier, the biometric matching degree identifier, and the ownership registration validity identifier all return a pass status, an ownership validity verification result is generated and the on-chain storage operation is triggered; otherwise, an ownership anomaly alarm result containing a failure factor is generated.

2. The method for storing copyright of AI digital content based on blockchain according to claim 1, characterized in that, The data to be authenticated is input into a preset multimodal feature extraction model to generate target authentication data, specifically including: Feature extraction is performed on the data to be authenticated to obtain a multimodal feature vector; Based on the cross-modal attention mechanism, the multimodal feature vectors are dynamically weighted and aggregated using a gated fusion network to generate a joint feature vector; Perform a hierarchical hash operation on the joint feature vector to obtain the target authentication data.

3. The method for storing copyright of AI digital content based on blockchain according to claim 2, characterized in that, Feature extraction is performed on the data to be authenticated to obtain a multimodal feature vector, specifically including: Natural language feature encoding is performed on the text data to generate a text feature vector; The image data is visually encoded to generate an image feature vector; The audio and video data are spatiotemporally encoded to generate a joint audio and video feature vector; The text feature vector, the image feature vector, and the audio-video joint feature vector are correlated to obtain the multimodal feature vector.

4. The method for storing copyright of AI digital content based on blockchain according to claim 3, characterized in that, Based on a cross-modal attention mechanism, a gated fusion network is used to dynamically weight and aggregate the multimodal feature vectors to generate a joint feature vector, specifically including: Based on the cross-modal attention mechanism, the cross-modal attention score among the text feature vector, the image feature vector, and the audio-video joint feature vector is calculated, and the weight coefficients corresponding to the cross-modal attention score are generated. The weight coefficients are input into the gated fusion network, and the weight coefficients are dynamically modified based on the preset gating mechanism in the gated fusion network. Based on the corrected weight coefficients, the multimodal feature vectors are weighted and summed to generate the joint feature vector.

5. The method for storing copyright of AI digital content based on blockchain according to claim 1, characterized in that, After generating the verification result, the method further includes: If the ownership anomaly alarm result is received, a derivative work identifier and the corresponding derivative work are generated based on the content fingerprint in the target authentication data; The derivative work identifier is stored on the blockchain, and the derivative work is sent to the copyright detection node for secondary copyright review.

6. The method for storing copyright of AI digital content based on blockchain according to claim 1, characterized in that, The verified target authentication data and the corresponding verification result are written into the blockchain, specifically including: Extract the content fingerprint of the target authentication data, generate a random obfuscation factor, and calculate the anonymized content hash; The target authentication data is encrypted using a symmetric encryption algorithm and then bound to the anonymized content hash to generate a privacy storage package. The privacy-preserving data package is written to the sidechain via the cross-chain evidence storage network, and the anonymized content hash is simultaneously written to the main chain.

7. A blockchain-based AI digital content copyright registration device, characterized in that, The device includes: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform a method as described in any one of claims 1-6.

8. A non-volatile computer storage medium for storing AI digital content copyright certificates based on blockchain, storing computer-executable instructions, characterized in that, When the computer-executable instructions are executed, they implement a method as described in any one of claims 1-6.

Images