Performance-lossless watermark credible traceability method and system based on block chain

Abstract

The invention discloses a block chain-based performance-lossless watermark credible traceability method and system. The method comprises the following steps of: embedding extended watermark information based on distribution consistent sampling in a potential space of a generative model to obtain a generation content of an embedded watermark; converting the watermark metadata into a hash value, storing the hash value in a block chain, and executing node trust management and a shared data storage and consensus mechanism; inverting the generated content embedded with the watermark to extract watermark information and calculating a hash value, comparing the hash value with the hash value of the watermark information stored in the block chain, if the comparison is consistent, passing the watermark verification, and if the comparison is inconsistent, failing the watermark verification; and when the watermark verification fails, performing node signature verification, if the node signature verification fails or the data is abnormal, starting an on-chain traceability process, tracking a data generation path according to the watermark metadata, and searching an abnormal node or data transmission point. According to the method, lossless embedding, credible circulation and reliable traceability of watermark information can be realized while high quality of generated content is ensured.

Description

Technical Field

[0001] The present invention belongs to the interdisciplinary technical field of deep learning information security and blockchain technology, and specifically relates to a performance lossless watermark trusted traceability method and system based on blockchain. Background Art

[0002] With the widespread development of blockchain technology and generative models in the digital economy and content generation, copyright protection and trusted traceability of generated content have gradually become urgent issues to be resolved. Blockchain, with its distributed storage, immutability and multi-party consensus mechanism, has shown great application potential in finance, government affairs and supply chain. However, with the popularization of image generation technology similar to diffusion models, its convenient generation capabilities and diverse application scenarios have also brought about a series of problems including infringement and the spread of false information. In recent years, generative models have been abused to generate false news, forged digital artworks or images of fictional events. These behaviors not only disrupt the market order, but may also cause public panic, posing severe challenges to the social trust system and economic stability.

[0003] In order to deal with these problems, higher requirements are put forward for the traceability, copyright protection and content authenticity of generated content. As an important means of copyright protection and content tracing, digital watermarking embeds invisible identification information into the content to achieve the traceability and copyright authentication of generated content. However, existing watermarking technology faces many limitations when applied to deep generative models.

[0004] First, traditional watermark embedding methods are prone to degradation of content generation performance, because watermark embedding often requires certain modifications to the original content, thus affecting the quality of the generated content and user experience. Secondly, most methods require modifying the parameters of the generation model or retraining the model, which not only increases the computational complexity of the system, but may also lead to further degradation of the quality of the generated content. In practical applications, this performance degradation and quality loss will directly affect the user experience and the overall performance of the system. In addition, existing watermarking technologies lack effective trust management and traceability capabilities when facing malicious nodes or complex data flow scenarios in distributed systems. In distributed systems, data flows between multiple nodes, and malicious nodes may tamper with watermark information, making content traceability and verification more complicated. At the same time, watermark information may also be maliciously tampered with by external attacks such as cropping, compression, and noise during cross-platform transmission and storage, further exacerbating the difficulty of traceability and verification. What is more serious is that the existing watermarking methods are not robust enough when dealing with various adversarial attacks or backdoor attacks. In some cases, the watermarking method may even become a means of implanting backdoors, which may not only lead to the loss or tampering of watermark information, but also affect the stability and availability of the system, and lead to instability and increased error rates in the verification process.

[0005] In summary, existing watermark embedding and management methods have significant limitations in meeting the requirements of maintaining performance after content embedding and complex traceability in multi-platform environments. In particular, how to achieve an effective balance between system performance and generated content quality during the watermark embedding process, traditional technical solutions usually rely on the adjustment of system parameters or additional training of models, which not only significantly increases the computational and time costs, but may also lead to a decline in the quality of generated content. Its lack of robustness will also bring serious security risks. Therefore, in practical applications, how to achieve effective copyright protection and trusted traceability while maintaining system performance has become an urgent problem to be solved. Summary of the invention

[0006] In view of the above, the purpose of the present invention is to provide a performance lossless watermark trusted traceability method and system based on blockchain. By combining watermark embedding technology with the distributed ledger system of blockchain, while ensuring the high quality of generated content, lossless embedding of watermark information, trusted circulation and reliable traceability are achieved.

[0007] In order to achieve the above-mentioned invention object, the technical solution provided by the present invention is as follows:

[0008] In a first aspect, an embodiment of the present invention provides a blockchain-based performance lossless watermark trusted traceability method, comprising the following steps:

[0009] The watermark information to be embedded is encoded, encrypted and redundantly expanded to form extended watermark information, and the extended watermark information is embedded in the latent space of the generative model based on distribution consistent sampling, and finally the generated content of the embedded watermark is obtained;

[0010] Based on the embedding process of the extended watermark information, watermark metadata including watermark ID, timestamp and node signature is generated and converted into hash values ​​and stored in the blockchain. Node trust management, shared data storage and consensus mechanism are performed on the nodes in the blockchain.

[0011] The generated content embedded with the watermark is inverted to extract the watermark information embedded in the latent space and calculate the hash value, which is compared with the hash value of the watermark information stored on the blockchain. If the comparison is consistent, the watermark verification passes, otherwise the watermark verification fails.

[0012] When watermark verification fails, node signature verification is performed. If node signature verification fails or the data is abnormal, the on-chain traceability process is initiated. The data is tracked based on the watermark ID, timestamp and node signature stored on the blockchain to generate a path, find the node or data transmission point where the abnormality occurred, and handle the abnormality.

[0013] Preferably, encoding, encrypting and redundantly extending the watermark information to be embedded to form extended watermark information includes:

[0014] The embedded watermark information is binary-encoded and encrypted using the ChaCha20 encryption algorithm to generate a pseudo-random binary stream. The redundant design in space and channel is used to replicate the binary stream m multiple times in the latent space to obtain the extended watermark information.

[0015] Preferably, embedding the extended watermark information based on distribution consistent sampling in the latent space of the generative model includes:

[0016] The original latent representation z generated by the generative model in the latent space T ~N(0,I) is a high-dimensional vector that obeys the standard normal distribution, I is the standard deviation, and the expanded watermark information s d Embedded into the original latent representation z T In multiple dimensions, it is expressed as:

[0017]

[0018]

[0019] in, is the potential representation of the watermark after embedding in the i-th dimension, z T [i] is the original potential representation of the i-th dimension, s d [i] is the embedded extended watermark information of the i-th dimension, λ is the weight coefficient controlling the embedding strength, and if(·) is the conditional function;

[0020] In the watermark embedding process, the latent representation after watermark embedding is made uniformly sampled. With the original latent representation z T The distribution of ~N(0,I) remains consistent, and the sampling process is expressed as:

[0021]

[0022] Among them, PPF(·) is the inverse function of the normal distribution, i is the current position of the embedded watermark, that is, the dimension of the potential representation, U(0,1) is the uniformly distributed random noise, 2 l It is 2 raised to the power of l.

[0023] Preferably, after the extended watermark information is embedded, the generated watermark embedded latent representation is Verify the potential representation after watermark embedding With the original latent representation z T The distribution consistency of is expressed as:

[0024]

[0025] Among them, Mean(·) is the mean operation, Var(·) is the variance operation, and after verification, it enters the blockchain storage link.

[0026] Preferably, the embedding process based on the extended watermark information generates watermark metadata including a watermark ID, a timestamp and a node signature and converts it into a hash value and stores it in the blockchain, including:

[0027] Based on the embedding process of the extended watermark information, watermark metadata including watermark ID, timestamp and node signature is generated, and the watermark metadata is stored in the blockchain in the form of a Merkle tree structure. Each leaf node of the Merkle tree stores the hash value H of a piece of data, which is expressed as:

[0028] H=Hash(WatermarkID∥Time∥Sign)

[0029] Among them, Hash(·) is the hash value calculation, WatermarkID is the watermark ID, Time is the timestamp embedded in the watermark, and Sign is the node signature generated by the blockchain;

[0030] When forming a Merkle tree structure, the two adjacent hash values ​​are merged and the calculation is continued until the root hash value H is formed. root , the root hash value H root Stored in the blockchain.

[0031] Preferably, the performing of node trust management and shared data storage and consensus mechanism on nodes in the blockchain includes:

[0032] Node trust management is performed on nodes in the blockchain. The data processing behavior of the nodes is monitored in real time and their trust scores are calculated based on compliance, consensus participation rate, and anomaly detection results. Malicious nodes are identified and their permissions are restricted based on the trust scores. The trust scores are expressed as:

[0033] T n =α·P n +β·C n +γ·A n

[0034] Among them, T n is the trust score of the nth node, P n , C n and A n are the compliance score, consensus participation rate, and abnormal behavior detection rate of the nth node, respectively. α, β, and γ are the weight coefficients, respectively;

[0035] The alliance chain architecture is used to realize shared data storage and consensus mechanism among multiple nodes. Each node saves a complete copy of the ledger to synchronize and make the watermark information consistent across the entire network. The PBFT consensus algorithm is used to ensure the reliability of the data on the chain, and to keep the system running normally in the event of failures or malicious behavior on some nodes.

[0036] Preferably, when watermark verification failure is detected, the node signature data is tracked and the node signature is verified, which is expressed as:

[0037] Verify(Sign k ,PublicKey k ,H k )=True

[0038] Among them, Verify(·) is the verification operation, Sign k 、PublicKey k and H k They are the node signature, public key and hash value of the kth node respectively. If the verification is True, it means the verification is successful. Otherwise, if the verification fails, the current node is marked as suspicious and the exception handling mechanism is triggered.

[0039] Preferably, the real-time monitoring node is used to determine whether the node submits data on time and frequently, and whether the data submitted by the node is consistent with the data of other nodes. If the data is submitted frequently or inconsistently, it is marked as an abnormal node and the abnormality handling mechanism is triggered.

[0040] In the second aspect, in order to achieve the above-mentioned invention object, the embodiment of the present invention also provides a blockchain-based performance lossless watermark trusted traceability system, which is implemented by the above-mentioned blockchain-based performance lossless watermark trusted traceability method, including: a watermark embedding module, a blockchain trusted storage module, an on-chain and off-chain watermark verification module, and an anomaly detection and traceability module;

[0041] The watermark embedding module is used to encode, encrypt and redundantly expand the watermark information to be embedded to form extended watermark information, embed the extended watermark information based on distribution consistent sampling in the latent space of the generative model, and finally obtain the generated content of the embedded watermark;

[0042] The blockchain trusted storage module is used to generate watermark metadata including watermark ID, timestamp and node signature based on the embedding process of extended watermark information and convert it into a hash value and store it in the blockchain, and perform node trust management and shared data storage and consensus mechanism on the nodes in the blockchain;

[0043] The on-chain and off-chain watermark verification module is used to invert the generated content embedded with the watermark, extract the watermark information embedded in the latent space and calculate the hash value, and compare it with the hash value of the watermark information stored on the blockchain. If the comparison is consistent, the watermark verification passes, and if the comparison is inconsistent, the watermark verification fails;

[0044] The anomaly detection and tracing module is used to perform node signature verification when watermark verification fails. If the node signature verification fails or the data is abnormal, the on-chain tracing process is started. The path is generated according to the watermark ID, timestamp and node signature tracking data stored on the blockchain, and the node or data transmission point where the anomaly occurs is found for exception handling.

[0045] On the third aspect, in order to achieve the above-mentioned purpose of the invention, an embodiment of the present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to implement the above-mentioned blockchain-based performance lossless watermark trusted traceability method when executing the computer program.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] (1) The present invention combines watermark embedding technology with the distributed ledger system of blockchain to achieve lossless embedding and verification of watermark information while ensuring the high quality of generated content. By embedding watermark information in the latent space of generated content and using distributed consistent sampling technology, the accuracy index of the content after watermark embedding is ensured to be consistent with that of the original content. The immutability of blockchain technology ensures the security of watermark metadata during transmission and storage, and improves the transparency and credibility of traceability management. Through the organic combination of blockchain and watermark embedding technology, an efficient traceability system covering the entire life cycle of generated content is constructed.

[0048] (2) The present invention performs node trust management and shared data storage and consensus mechanisms on the nodes in the blockchain, and performs real-time evaluation and management of the node behavior, ensuring that all nodes in the system can participate in data transmission and processing according to established rules. After the watermark embedding of the generated content is completed, the watermark information is automatically uploaded to the blockchain for archiving, forming a trust management system that links the chain and the off-chain. The interoperability of the blockchain in cross-platform and multi-system scenarios provides security guarantees for the secure flow of data and the trusted management of watermarks.

[0049] (3) The present invention embeds watermarks by forming extended watermark information, which has high robustness and can maintain the integrity and effectiveness of watermark information in the face of malicious attacks such as strong noise, cropping and compression. The watermark embedding process not only does not damage the generation performance of the model, but also ensures the detectability of the watermark in certain adversarial environments, thereby achieving stable traceability of the generated content. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0051] Figure 1 It is a flowchart of a performance lossless watermark trusted traceability method based on blockchain provided by an embodiment of the present invention;

[0052] Figure 2 It is a schematic diagram of the framework of a blockchain-based performance lossless watermark trusted traceability method provided by an embodiment of the present invention;

[0053] Figure 3 It is a structural diagram of a blockchain-based performance lossless watermark trusted traceability system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0054] To make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific implementation methods described herein are only used to explain the present invention and do not limit the scope of protection of the present invention.

[0055] The inventive concept of the present invention is: in view of the problem that the existing watermark embedding and management methods have significant limitations in meeting the performance maintenance after content embedding and the complex traceability requirements in a multi-platform environment, the embodiments of the present invention provide a performance-lossless watermark trusted traceability method and system based on blockchain. Under the premise of not affecting the content generation performance, the watermark information is embedded into the latent space of the generated content, and the distributed consistent sampling technology is combined to ensure the natural integration of the embedding process and the quality maintenance of the generated content. At the same time, relying on the tamper-proof characteristics of the blockchain and the multi-party consensus mechanism, the transparent storage and dynamic management of the watermark information on and off the chain are realized, ensuring the trusted flow of data and the reliable traceability of the watermark information in a multi-node, multi-platform environment, and by enhancing the robustness of the system, even in complex application scenarios, such as noise interference, data clipping or compression operations, the integrity and verifiability of the watermark can still be ensured.

[0056] Figure 1 It is a flowchart of a performance lossless watermark trusted traceability method based on blockchain provided by an embodiment of the present invention. Figure 2 Schematic diagram of the framework of the blockchain-based performance lossless watermark trusted traceability method provided by the embodiment of the present invention. Figure 1 and Figure 2As shown, the embodiment provides a performance lossless watermark trusted traceability method based on blockchain, including the following steps:

[0057] S1, encode, encrypt and redundantly expand the watermark information to be embedded to form extended watermark information, embed the extended watermark information in the latent space of the generative model based on distribution consistent sampling, and finally obtain the generated content of the embedded watermark.

[0058] S1.1, input model and data. Input the generative model and the watermark information to be embedded. In the embodiment, the generative model adopts a diffusion model and is used to generate images. The latent space of the diffusion model is used to carry the watermark information to ensure that the embedding process does not affect the quality of the generated content. The latent representation z generated by the diffusion model T ~N(0,I) is a high-dimensional vector that obeys the standard normal distribution, and I is the standard deviation. Before watermark embedding, the original sample is first input into the diffusion model to generate an image, and the generated image is segmented based on the Gaussian Shading method. The segmented main area is used as the target area for watermark embedding to improve the efficiency and concealment of watermark embedding.

[0059] S1.2, encoding and encryption of watermark data. The embedded watermark information s is binary encoded to form a watermark bit stream. The encoded watermark information is encrypted using the ChaCha20 encryption algorithm to generate a pseudo-random binary stream m, which is expressed as:

[0060] m=ChaCha20(s,K)

[0061] Among them, ChaCha20(·) is the ChaCha20 encryption algorithm, K is the key, and the encrypted m is a uniformly distributed binary stream, which is used to prevent the watermark information from being easily guessed by malicious attackers.

[0062] S1.3, Diffusion and redundancy design. In order to enhance the robustness of the watermark, the redundancy design in space and channel is adopted to copy the binary stream m multiple times in the latent space to obtain the extended watermark information S d , expressed as:

[0063]

[0064] Among them, Repeat(·) is a copy operation, f c is the channel redundancy factor, is the height-width redundancy factor. In the diffusion process, multiple iterative replication operations are performed to ensure that part of the watermark information can still be fully extracted even under cropping, compression and noise attacks.

[0065] S1.2, watermark embedding and fusion based on distribution consistent sampling. The extended watermark information sd Embedded into the original latent representation z T In multiple dimensions of , the watermark is embedded bit by bit using the following formula:

[0066]

[0067]

[0068] in, is the potential representation of the watermark after embedding in the i-th dimension, z T [i] is the original potential representation of the i-th dimension, s d [i] is the embedded extended watermark information of the i-th dimension, if(·) is the conditional function, and λ is the weight coefficient controlling the embedding strength. By adjusting λ, we ensure that the embedding process does not affect the distribution of the latent representation and the quality of the generated content.

[0069] During the watermark embedding process, the latent representation after watermark embedding is ensured by distribution consistent sampling. With the original latent representation z T The distribution of ~N(0,I) is kept consistent to avoid the embedding process affecting the performance of the generated model. The sampling process is expressed as:

[0070]

[0071] Among them, PPF (·) (Percent Point Function) is the inverse function of the normal distribution, i is the current position of the embedded watermark, that is, the dimension of the potential representation, U (0,1) is the uniformly distributed random noise, 2 l is the power of 2. Through this sampling method, it can be ensured that the embedded potential representation is statistically consistent with the representation without watermark embedding.

[0072] S1.3, after the extended watermark information is embedded, the generated watermark is embedded into the latent representation Verify the potential representation after watermark embedding With the original latent representation z T The distribution consistency of is expressed as:

[0073]

[0074] Among them, Mean(·) is the mean operation, and Var(·) is the variance operation. After verification, you can enter the next blockchain storage link.

[0075] S2, based on the embedding process of the extended watermark information, generates watermark metadata including watermark ID, timestamp and node signature and converts it into a hash value and stores it in the blockchain, performs node trust management and shared data storage and consensus mechanism on the nodes in the blockchain.

[0076] S2.1, Generate watermark metadata. Based on the embedding process of the extended watermark information, generate watermark metadata including watermark ID, timestamp and node signature to ensure the uniqueness and integrity of the data. The watermark ID is used to track the source of the generated content, and the timestamp ensures the timing accuracy of data generation.

[0077] S2.2, blockchain storage. The watermark metadata is stored in the blockchain in the form of a Merkle tree structure. Each leaf node of the Merkle tree stores the hash value H of a piece of data, expressed as:

[0078] H=Hash(WatermarkID∥Time∥Sign)

[0079] Among them, Hash(·) is the hash value calculation, WatermarkID is the watermark ID, Time is the timestamp embedded in the watermark, and Sign is the node signature generated by the blockchain. In this way, the integrity of the data can be quickly verified when needed, and the metadata can be ensured to be tamper-proof.

[0080] When storing the Merkle tree structure, the hash value is calculated for each metadata (including watermark ID, timestamp and node signature):

[0081] H k =Hash(WatermarkID k ∥Time∥Sign k )

[0082] Among them, the subscript k is the index of the hash value, watermark ID and node signature.

[0083] By merging two adjacent hash values ​​and continuing the calculation until the root hash value H is formed root :

[0084] H root =Iteration(Hash(H j ∥H j+1 ))

[0085] Where, Iteration(·) is the iteration, and the subscripts j and j+1 are the indexes of two adjacent hash values.

[0086] The root hash value H root Stored in the blockchain, it ensures the integrity of the data during subsequent verification.

[0087] S2.3, Node trust management, node behavior monitoring and trust score calculation. Perform node trust management on the nodes in the blockchain, monitor the data processing behavior of the nodes in real time and calculate their trust scores based on compliance, consensus participation rate and anomaly detection results. Identify malicious nodes based on the trust score and restrict their permissions. The trust score is expressed as:

[0088] T n =α·P n +β·C n +γ·A n

[0089] Among them, T n is the trust score of the nth node, P n , C n and A n are the compliance score, consensus participation rate, and abnormal behavior detection rate of the nth node, respectively. α, β, and γ are the weight coefficients, respectively;

[0090] S2.4, cross-node consensus and synchronization of on-chain data. The consortium blockchain architecture is used to implement shared data storage and consensus mechanisms among multiple nodes. Each node saves a complete copy of the ledger to synchronize and maintain watermark information across the entire network. The PBFT (Practical Byzantine Fault Tolerance) consensus algorithm is used to ensure the reliability of on-chain data and keep the system running normally in the event of failures or malicious behavior on some nodes.

[0091] S3, invert the generated content embedded with the watermark to extract the watermark information embedded in the latent space and calculate the hash value, which is compared with the hash value of the watermark information stored on the blockchain. If the comparison is consistent, the watermark verification passes, otherwise the watermark verification fails.

[0092] S3.1, off-chain watermark information extraction. The watermark information in the latent space is extracted from the generated content through the DDIM inversion (Denoising Diffusion Implicit Models Inversion) algorithm. The inversion process is as follows:

[0093] z0=DDIM_Inverse(X s )

[0094] Among them, DDIM_Inverse(·) is the DDIM inversion calculation, X s is the generated image embedded with watermark, z0 is the potential representation obtained by inversion calculation, the embedded watermark information contained in z0 is extracted, and the embedded watermark information is hashed to obtain the hash value Hextracted , H extracted It will be used to compare with the data stored on the chain during the watermark verification phase.

[0095] S3.2, on-chain and off-chain watermark verification. Use the watermark ID as an index to quickly locate the corresponding watermark metadata on the chain. By comparing the metadata stored on the chain with the watermark information extracted off-chain, ensure that the content has not been tampered with. The specific verification is based on the watermark ID of the watermark information extracted off-chain. s , find the corresponding hash value H on the blockchain s , H s With H extracted Make a comparison.

[0096] S4, when watermark verification fails, node signature verification is performed. If node signature verification fails or data is abnormal, the on-chain traceability process is started. The path is generated based on the watermark ID, timestamp and node signature tracking data stored on the blockchain, and the node or data transmission point where the abnormality occurs is found for abnormality handling.

[0097] S4.1, anomaly detection mechanism. The core of the anomaly detection part is to monitor the behavior of all nodes in the blockchain, such as data tampering and malicious data release. Real-time monitoring of whether the node submits data on time, whether it submits frequently, and whether the data submitted by the node is consistent with the data of other nodes. If the data is submitted frequently or the data is inconsistent, it will be marked as an abnormal node and the exception handling mechanism will be triggered.

[0098] S4.2, traceability mechanism. When watermark verification fails, the node signature data will be tracked to analyze whether there are malicious nodes involved in data tampering. The verification formula of the node signature is expressed as:

[0099] Verify(Sign k ,PublicKey k ,H k )=True

[0100] Among them, Verify(·) is the verification operation, Sign k 、PublicKey k and H k They are the node signature, public key and hash value of the kth node respectively. If the verification is True, it means the verification is successful. Otherwise, if the verification fails, the current node is marked as suspicious and the exception handling mechanism is triggered.

[0101] In the event of a node signature verification failure or data anomaly, the system will automatically start the on-chain traceability process and track the data generation path based on the timestamp and node signature on the blockchain. The system checks the signature and data transmission link of each node one by one along the historical record of the watermark ID until the node or data transmission point where the anomaly occurs is found for anomaly handling.

[0102] In summary, the embodiment of the present invention provides a high-performance lossless watermark trusted traceability method based on blockchain. By introducing Gaussian Shading watermark embedding technology and combining it with the distributed ledger system of blockchain, it realizes lossless embedding and verification of watermark information while ensuring the high quality of generated content, realizes the security of watermark metadata during transmission and storage, improves the transparency and credibility of traceability management, and ensures the high robustness of watermark embedding. By constructing an efficient traceability system covering the entire life cycle of generated content, it is possible to maintain the integrity and effectiveness of watermark information in the face of malicious attacks such as strong noise, cropping and compression, thereby achieving stable traceability of generated content.

[0103] Based on the same inventive concept, Figure 3 As shown, an embodiment of the present invention also provides a performance lossless watermark trusted traceability system 300 based on blockchain, including: a watermark embedding module 310, a blockchain trusted storage module 320, an on-chain and off-chain watermark verification module 330 and an anomaly detection and tracing module 340.

[0104] The watermark embedding module 310 is used to encode, encrypt and redundantly expand the watermark information to be embedded to form extended watermark information, embed the extended watermark information based on distribution consistent sampling in the latent space of the generative model, and finally obtain the generated content of the embedded watermark;

[0105] The blockchain trusted storage module 320 is used to generate watermark metadata including watermark ID, timestamp and node signature based on the embedding process of the extended watermark information and convert it into a hash value and store it in the blockchain, and perform node trust management and shared data storage and consensus mechanism on the nodes in the blockchain;

[0106] The on-chain and off-chain watermark verification module 330 is used to invert the generated content embedded with the watermark, extract the watermark information embedded in the latent space and calculate the hash value, and compare it with the hash value of the watermark information stored on the blockchain. If the comparison is consistent, the watermark verification passes, and if the comparison is inconsistent, the watermark verification fails;

[0107] The anomaly detection and tracing module 340 is used to perform node signature verification when watermark verification fails. If the node signature verification fails or the data is abnormal, the on-chain tracing process is started. The path is generated based on the watermark ID, timestamp and node signature tracking data stored on the blockchain, and the node or data transmission point where the anomaly occurs is found to handle the anomaly.

[0108] Based on the same inventive concept, an embodiment of the present invention also provides an electronic device, including a memory and a processor, the memory is used to store a computer program, and the processor is used to implement the above-mentioned blockchain-based performance lossless watermark trusted traceability method when executing the computer program.

[0109] It should be noted that the blockchain-based performance lossless watermark trusted traceability system and electronic device provided in the above embodiments belong to the same inventive concept as the blockchain-based performance lossless watermark trusted traceability method. The specific implementation process is detailed in the blockchain-based performance lossless watermark trusted traceability method embodiment, which will not be repeated here.

[0110] The specific implementation methods described above provide a detailed description of the technical solutions and beneficial effects of the present invention. It should be understood that the above is only the most preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, supplements and equivalent substitutions made within the scope of the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A performance lossless watermark trusted traceability method based on blockchain, characterized in that: The following steps are involved: The watermark information to be embedded is encoded, encrypted and redundantly expanded to form extended watermark information, and the extended watermark information is embedded in the latent space of the generative model based on distribution consistent sampling, and finally the generated content of the embedded watermark is obtained; Based on the embedding process of the extended watermark information, watermark metadata including watermark ID, timestamp and node signature is generated and converted into hash values ​​and stored in the blockchain. Node trust management, shared data storage and consensus mechanism are performed on the nodes in the blockchain. The generated content embedded with the watermark is inverted to extract the watermark information embedded in the latent space and calculate the hash value, which is compared with the hash value of the watermark information stored on the blockchain. If the comparison is consistent, the watermark verification passes, otherwise the watermark verification fails. When watermark verification fails, node signature verification is performed. If node signature verification fails or the data is abnormal, the on-chain traceability process is initiated. The data is tracked based on the watermark ID, timestamp and node signature stored on the blockchain to generate a path, find the node or data transmission point where the abnormality occurred, and handle the abnormality.

2. The blockchain-based performance lossless watermark trusted traceability method according to claim 1 is characterized in that: The step of encoding, encrypting and redundantly extending the watermark information to be embedded to form extended watermark information includes: The embedded watermark information is binary-encoded and encrypted using the ChaCha20 encryption algorithm to generate a pseudo-random binary stream. The redundant design in space and channel is used to replicate the binary stream m multiple times in the latent space to obtain the extended watermark information.

3. The blockchain-based performance lossless watermark trusted traceability method according to claim 1 is characterized in that: The embedding of extended watermark information based on distribution consistent sampling in the latent space of the generative model includes: The original latent representation z generated by the generative model in the latent space T ~N(0,I) is a high-dimensional vector that obeys the standard normal distribution, I is the standard deviation, and the expanded watermark information s d Embedded into the original latent representation z T In multiple dimensions, it is expressed as: in, is the potential representation of the watermark after embedding in the i-th dimension, z T [i] is the original potential representation of the i-th dimension, s d [i] is the embedded extended watermark information of the i-th dimension, λ is the weight coefficient controlling the embedding strength, and if(·) is the conditional function; In the watermark embedding process, the latent representation after watermark embedding is made uniformly sampled. With the original latent representation z T The distribution of ~N(0,I) remains consistent, and the sampling process is expressed as: Among them, PPF(·) is the inverse function of the normal distribution, i is the current position of the embedded watermark, that is, the dimension of the potential representation, U(0,1) is the uniformly distributed random noise, 2 l It is 2 raised to the power of l.

4. The blockchain-based performance lossless watermarking trusted traceability method according to claim 3 is characterized in that: After the extended watermark information is embedded, the generated watermark is embedded into the latent representation Verify the potential representation after watermark embedding With the original latent representation z T The distribution consistency of is expressed as: Among them, Mean(·) is the mean operation, Var(·) is the variance operation, and after verification, it enters the blockchain storage link.

5. The blockchain-based performance lossless watermark trusted traceability method according to claim 1 is characterized in that: The embedding process based on the extended watermark information generates watermark metadata including watermark ID, timestamp and node signature and converts it into a hash value and stores it in the blockchain, including: Based on the embedding process of the extended watermark information, watermark metadata including watermark ID, timestamp and node signature is generated, and the watermark metadata is stored in the blockchain in the form of a Merkle tree structure. Each leaf node of the Merkle tree stores the hash value H of a piece of data, which is expressed as: H=Hash(WatermarkID∥Time∥Sign) Among them, Hash(·) is the hash value calculation, WatermarkID is the watermark ID, Time is the timestamp embedded in the watermark, and Sign is the node signature generated by the blockchain; When forming a Merkle tree structure, the two adjacent hash values ​​are merged and the calculation is continued until the root hash value H is formed. root , the root hash value H root Stored in the blockchain.

6. The blockchain-based performance lossless watermark trusted traceability method according to claim 1 is characterized in that: The node trust management and shared data storage and consensus mechanism for nodes in the blockchain include: Node trust management is performed on nodes in the blockchain. The data processing behavior of the nodes is monitored in real time and their trust scores are calculated based on compliance, consensus participation rate, and anomaly detection results. Malicious nodes are identified and their permissions are restricted based on the trust scores. The trust scores are expressed as: T n =α·P n +β·C n +γ·A n Among them, T n is the trust score of the nth node, P n , C n and A n are the compliance score, consensus participation rate, and abnormal behavior detection rate of the nth node, respectively. α, β, and γ are the weight coefficients, respectively; The alliance chain architecture is used to realize shared data storage and consensus mechanism among multiple nodes. Each node saves a complete copy of the ledger to synchronize and make the watermark information consistent across the entire network. The PBFT consensus algorithm is used to ensure the reliability of the data on the chain, and to keep the system running normally in the event of failures or malicious behavior on some nodes.

7. The blockchain-based performance lossless watermark trusted traceability method according to claim 1 is characterized in that: When watermark verification failure is detected, the node signature data is tracked and the node signature is verified, which is expressed as: Verify(Sign k ,PublicKey k ,H k )=True Among them, Verify(·) is the verification operation, Sign k 、PublicKey k and H k They are the node signature, public key and hash value of the kth node respectively. If the verification is True, it means the verification is successful. Otherwise, if the verification fails, the current node is marked as suspicious and the exception handling mechanism is triggered.

8. The blockchain-based performance lossless watermark trusted traceability method according to claim 1 is characterized in that: Real-time monitoring of whether the node submits data on time, whether it submits data frequently, and determines whether the data submitted by the node is consistent with the data of other nodes. If the data is submitted frequently or the data is inconsistent, it will be marked as an abnormal node and the exception handling mechanism will be triggered.

9. A blockchain-based performance lossless watermark trusted traceability system, implemented using the blockchain-based performance lossless watermark trusted traceability method according to any one of claims 1 to 8, characterized in that: include: Watermark embedding module, blockchain trusted storage module, on-chain and off-chain watermark verification module, and anomaly detection and tracing module; The watermark embedding module is used to encode, encrypt and redundantly expand the watermark information to be embedded to form extended watermark information, embed the extended watermark information based on distribution consistent sampling in the latent space of the generative model, and finally obtain the generated content of the embedded watermark; The blockchain trusted storage module is used to generate watermark metadata including watermark ID, timestamp and node signature based on the embedding process of extended watermark information and convert it into a hash value and store it in the blockchain, and perform node trust management and shared data storage and consensus mechanism on the nodes in the blockchain; The on-chain and off-chain watermark verification module is used to invert the generated content embedded with the watermark, extract the watermark information embedded in the latent space and calculate the hash value, and compare it with the hash value of the watermark information stored on the blockchain. If the comparison is consistent, the watermark verification passes, and if the comparison is inconsistent, the watermark verification fails; The anomaly detection and tracing module is used to perform node signature verification when watermark verification fails. If the node signature verification fails or the data is abnormal, the on-chain tracing process is started. The path is generated according to the watermark ID, timestamp and node signature tracking data stored on the blockchain, and the node or data transmission point where the anomaly occurs is found for exception handling.

10. An electronic device comprising a memory and a processor, wherein the memory is used to store a computer program, wherein: The processor is used to implement the blockchain-based performance lossless watermark trusted traceability method described in any one of claims 1 to 8 when executing the computer program.

Images