Visual data traceability and integrity verification system based on block chain

Abstract

The invention discloses a visual data traceability and integrity verification system based on a block chain, relates to the technical field of data traceability, and is used for solving the problems that a current digital signature mechanism may not cope with potential threats of quantum computing, and particularly, in important data evidence storage, a security evaluation method is generally static, and the security evaluation method cannot cope with the potential threats of quantum computing. In the prior art, a digital signature system cannot respond to the change or risk of data in real time, the digital signature system cannot flexibly deal with the security problem in a long time period, and the problems of single verification mechanism and poor flexibility may exist in the security verification process; by introducing an anti-quantum encryption technology, dynamic security assessment, a rollback re-signing mechanism and intelligent response verification, the problems of vulnerability to quantum computing threats, inaccuracy of data security assessment and poor long-term effectiveness and flexibility of signatures in the prior art are solved; due to the innovations, the system can better adapt to the change of the future technology, and meanwhile, the long-term reliability and safety of visual data are guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data traceability, more specifically, the present application relates to a visual data traceability and integrity verification system based on blockchain. BACKGROUND

[0002] Combined with digital signature, encryption technology, tamper-proof nature of blockchain, smart contract, and quantum-resistant encryption, etc. Advanced technologies can effectively guarantee the authenticity, integrity, traceability and security of visual data. Compared with traditional data storage and verification methods, these technologies have stronger security, reliability and scalability, and can adapt to the development challenges of future technologies, especially under the threat of quantum computing.

[0003] The prior art has the following disadvantages:

[0004] The current digital signature mechanism may not be able to cope with the potential threat of quantum computing, especially in important data storage, the security evaluation method is usually static, and cannot respond to the changes or risks of data in real time, the digital signature system cannot flexibly cope with security problems over a long period of time, and there may be a single verification mechanism and poor flexibility in the security verification process.

[0005] In view of the above problems, the present application provides a solution. SUMMARY

[0006] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a visual data traceability and integrity verification system based on blockchain to solve the problems raised in the background art.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0008] The visual data traceability and integrity verification system based on blockchain comprises a double-sign first storage module, a residual security evaluation module, a rollback and re-sign module and a interlocking verification module:

[0009] The double-sign first storage module: obtains the original visual data, calculates the content identifier CID, signs the original visual data using elliptic curve algorithm and lattice quantum-resistant algorithm respectively, and packs the content identifier and two signatures as initial on-chain records as tamper-proof original storage;

[0010] The residual security evaluation module: obtains the residual security bit of the current original visual data, uses the Bayesian model to calculate the expected security bit corresponding to the security bit index by combining the key frame density, exposure heat and logic gate error of the collected original visual data, labels the risk label of each original visual data, and judges the buffer and dangerous image into the re-sign list;

[0011] Rollbackable Resign Module: Resign the original visual data of the list to generate a new signature with the quantum algorithm, construct a periodic rollbackable resign mechanism, encapsulate the signature and the original content identifier as a hash time lock contract transaction, set a period and write it to the blockchain; If a vulnerability is found during the lock period, it can be revoked, and the old signature will be automatically replaced when the period expires;

[0012] Interlocking Verification Module: Adopt a double-track strategy, preferentially verify quantum-resistant signatures, and verify traditional signatures as a fallback when the former fails, monitor signature status and periodic rollbackable resign implementation, trigger four types of responses: normal, alarm, automatic upgrade or forced switching, and feed back the results to the model.

[0013] In a preferred embodiment, the following is included:

[0014] Obtain the original visual data, calculate the hash of the original visual data to obtain the content identifier CID;

[0015] The device generates a digital signature Sig for the content identifier CID using a classical elliptic curve algorithm ECC , and simultaneously generates a digital signature Sig for the content identifier CID using a lattice quantum-resistant algorithm Dilithium ;

[0016] Combine Sig ECC and Sig Dilithium into a composite signature;

[0017] Write the content identifier CID and the composite signature to the content chain.

[0018] In a preferred embodiment, the following is included:

[0019] The ideal security strength derived from the algorithm parameters is the theoretical security bit B0;

[0020] Obtain the current known strongest attack complexity C t required by the latest cracking algorithm, and the cost ratio P t representing the price of computing power of the current unit hash operation t , and calculate the real effective security bit B t at time t by B t = B0-log2!(C t );

[0021] Set the minimum required security bit B req according to the security requirements of the application scenario, and calculate the remaining security bit R t as the difference between the effective security bit B t and the minimum required security bit B req ;

[0022] The key frame density, exposure heat, and logic gate error data of the original visual data are collected to form an observation vector O t A Bayesian model is constructed:

[0023] The input nodes are defined: key frame density, exposure heat, logic gate error, and security bit history data;

[0024] The intermediate nodes are constructed: the data reliability node integrates key frame density and exposure heat, and the system stability node integrates logic gate error and security bit history decay trend;

[0025] The expected security bit is output;

[0026] Quantum threat weakening bit number calculation:

[0027] The current time is t0, and the expected security bit is The security bit at time t0+Δt is estimated by fitting the historical curve Then the basic weakening bit number can be taken as ΔB=B(t0)-B(t0+Δt);

[0028] According to the rate of progress of quantum technology, the additional security bit loss caused by quantum attack is simulated, which is recorded as quantum additional loss ΔQ;

[0029] The sum of the basic weakening bit number ΔB and the quantum additional loss ΔQ is taken as the final quantum threat weakening bit number;

[0030] The calculated current expected security bit is compared with the remaining security bit, and combined with the quantum threat weakening bit number θ, each piece of data is labeled with its security risk level label:

[0031] When the remaining security bit ≥ expected security bit, it is labeled as "abundant";

[0032] When expected security bit > remaining security bit ≥ (expected security bit - θ), it is labeled as "warning";

[0033] When expected security bit - θ > remaining security bit ≥ expected security bit - (1-α)·θ, it is labeled as "buffer", where α∈(0,1) is an adjustable coefficient;

[0034] When the remaining security bit ≤ 0, it is labeled as "dangerous";

[0035] For the data labeled as "buffer" or "dangerous", it is automatically added to the re-sign list.

[0036] In a preferred embodiment, the following is included:

[0037] Constructing a hash time lock contract transaction: containing original CID, new signature, and preset time lock;

[0038] Set double effective condition: reach the predetermined block height to automatically take effect, or provide hash original image revocation transaction within the lock period;

[0039] Maintain old and new signatures within the lock period, and automatically update the valid signature record on the chain after the expiration;

[0040] Generate a new signature for the CID in the re-signing list, and bind and encapsulate the new signature with the original CID.

[0041] In a preferred embodiment, the following is included:

[0042] When there is original visual data that needs to be verified for integrity, the interlocking verification module starts the "double-track" verification process:

[0043] First track verification: preferentially verify Sig with quantum-resistant public key Dilithium ;

[0044] Second track verification: verify Sig with traditional public key ECC ;

[0045] Generate four types of judgment results: both signatures pass, quantum-resistant signature passes but traditional signature fails, quantum-resistant signature fails but traditional signature passes, and both signatures fail, and record quantum-resistant signature passes but traditional signature fails and quantum-resistant signature fails but traditional signature passes as single signature pass;

[0046] According to the double-track verification results and the risk labels provided by the evaluation module, the interlocking verification module can trigger four types of responses to coordinate system actions, namely normal, alarm, automatic upgrade, and forced switching:

[0047] Normal response: when the label is "abundant" and both signatures pass, it is in normal operation and does not require special intervention;

[0048] Alarm response: when the label is "warning" and both signatures pass, the label is "warning" and single signature passes, and the label is "abundant" and single signature passes, keep an eye on it;

[0049] Forced switching response: when the label is "danger" and both signatures fail, the verification module immediately executes the forced switching response, including switching the trust path and switching the algorithm scheme;

[0050] Automatic upgrade response: all other responses are recorded as automatic upgrade responses, and when the re-signing process is needed, the verification module triggers the automatic upgrade response;

[0051] After executing the above responses, relevant information is fed back to the Bayesian model of the remaining security evaluation module. The technical effects and advantages of the visual data traceability and integrity verification system based on the blockchain of the present application are:

[0052] The application ensures the security of data in the present and future by introducing double signatures, uses a Bayesian model combined with multiple data features to evaluate the security of data, and can dynamically evaluate the risk of visual data. By calculating the expected security bit and labeling the risk label, it can accurately identify which data has potential risks and help developers to handle it in time. By using a time lock contract, the validity period of the signature can be set, the signature can be revoked when the vulnerability is found, and the old signature can be automatically replaced after the lock period ends, ensuring the effectiveness of the signature and the long-term security of the data. The double-track mechanism can ensure data security during the transition period before quantum algorithms are widely used, and can effectively respond to abnormal situations by monitoring the signature state and setting automatic responses, improving the reliability and automatic processing capacity of the system. BRIEF DESCRIPTION OF DRAWINGS

[0053] Fig. 1 The figure is a structural diagram of the visual data traceability and integrity verification system based on the blockchain of the application.

[0054] Fig. 2 The figure is a flowchart of the visual data traceability and integrity verification system based on the blockchain of the application. DETAILED DESCRIPTION

[0055] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only a part of the embodiments of the application, not all. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.

[0056] EMBODIMENT

[0057] Please refer to Figs. 1-2 As shown in the figure, the application discloses a visual data traceability and integrity verification system based on a blockchain, which comprises a double signature first storage record module, a residual security evaluation module, a rollback and re-signing module, and a interlocking verification module, and the modules are signal connected.

[0058] The double signature first storage record module: obtains the original visual data, calculates the content identifier CID, signs the original visual data using elliptic curve algorithm and lattice quantum resistance algorithm respectively, and packs the content identifier and the two signatures into the initial on-chain record as the tamper-proof original storage record;

[0059] The residual security evaluation module: obtains the residual security bit of the current original visual data, uses a Bayesian model combined with the key frame density, exposure heat and logic gate error of the collected original visual data to calculate the expected security bit corresponding to the security bit index, labels the risk label of each original visual data, and buffers and judges the dangerous image into the re-signing list;

[0060] Rollbackable Resign Module: Resign the original visual data of the list to generate a new signature after the quantum algorithm, construct a periodic rollbackable resign mechanism, encapsulate the signature and the original content identifier as a hash time lock contract transaction, set the periodicity and write it to the blockchain; If a vulnerability is found within the lock period, it can be revoked, and the old signature will be automatically replaced at the end of the period;

[0061] Interlocking Verification Module: Adopt a double-track strategy, preferentially verify quantum-resistant signatures, and verify traditional signatures as a fallback when the former fails, monitor signature status and periodic rollbackable resign implementation, trigger four types of responses: normal, alarm, automatic upgrade or forced switching, and feed back the results to the model.

[0062] In the dual-sign first storage module, the original visual data is obtained, the content identifier CID is calculated, and the original visual data is signed using elliptic curve algorithm and lattice quantum-resistant algorithm respectively, and the content identifier and the two signatures are packaged as the initial on-chain record as the tamper-proof original storage, which includes:

[0063] Obtain the original visual data, calculate the hash of the original visual data to obtain the content identifier CID;

[0064] The device uses a classical elliptic curve algorithm to generate a digital signature Sig for the content identifier CID ECC At the same time, use the lattice quantum-resistant algorithm Dilithium to generate a digital signature Sig for the content identifier CID Dilithium Here you can choose Dilithium-3 parameter set to balance performance and security;

[0065] The digital signatures Sig ECC and Sig Dilithium Two signatures constitute a composite signature, which needs to be verified simultaneously to ensure the integrity and credibility of the message. Through such a hybrid cryptographic design, traditional algorithms and quantum-resistant algorithms are used simultaneously for the same data signature, so that data can be verified in both classical and quantum environments;

[0066] Pack the content identifier CID of the visual content and the above digital signatures Sig ECC and Sig Dilithium Form the first block chain transaction

[0067] The transaction record is on a special "content chain" as the tamper-proof original storage of the original visual data. Due to the tamper-proof and decentralized nature of the blockchain ledger, once the record is successfully chained, it forms an unforgeable and tamper-proof storage. Anyone can subsequently retrieve the hash of the image and the corresponding composite signature from the chain, thereby verifying its integrity and source credibility timestamp. ​

[0068] Through the above process, the system completes the "composite signature + on-chain" operation as soon as the data is generated: both the traditional ECC signature is retained for compatibility with existing systems, and the quantum-resistant signature is introduced to prepare for future security. This step ensures the credibility of the entire process from the generation to the evidence of visual data, laying the foundation for subsequent tracing and verification.

[0069] In the residual security evaluation module, the residual security bits of the current original visual data are obtained, and the Bayesian model is used to combine the key frame density, exposure heat, and logical gate error of the collected original visual data to calculate the expected security bit corresponding to the security bit index. Risk labels are labeled for each original visual data, and the buffer and dangerous images are judged into the re-sign list. The specific content includes:

[0070] Continuously monitor the security threat dynamics faced by cryptographic algorithms, evaluate the number of "residual security bits" of the signature algorithm at the current and future time, and develop risk labels for the content. Specifically, the system combines theoretical analysis and observation intelligence to calculate three key security bit indicators:

[0071] Theoretical security bits B0: the ideal security strength derived from algorithm parameters, expressed in bits;

[0072] Effective security bits B t : Considering the attack capability and cost in reality, the actual number of bits that need to be exhausted may be lower than the theoretical value;

[0073] Get the current strongest attack complexity C t required by the latest cracking algorithm and the cost ratio P t of the current unit hash operation, which represents the computing power price, and approximate the real effective security bit number at time t through B t =B0-log2!(C t / P t );

[0074] Intuitively, if the attack becomes more efficient or the calculation becomes cheaper, the effective security bits will decrease. If a new algorithm reduces the cracking complexity by 1024 times, i.e. C t / P t =2 10 , then the effective security bits B t will be reduced by about 10 bits compared to the theoretical security bits B0.

[0075] Residual security bits R t : Set the minimum required security bits B req according to the security requirements of the application scenario, and calculate the residual security bits R t as the effective security bits B t and the minimum required security bits B reqThe difference between the calculated B t The difference between the calculated B req If the difference is positive, the residual security bit is positive, otherwise it is negative.

[0076] The key frame density, exposure heat, and logical gate error data of the original visual data are aggregated to form an observation vector O t , which reflects the improvement of quantum attack ability in real time.

[0077] Key frame density: refers to the frequency of occurrence of key frames in the original visual data, reflecting the structural characteristics and content change frequency of the data.

[0078] High key frame density means that the video content changes quickly and has little inter-frame redundancy, so the system needs to record the integrity of the data more frequently, but it also means that there are more checkpoints on the chain, making tampering easier to detect. On the contrary, sparse key frames may indicate that the content is continuous for a long time, and the interval between the recorded checkpoints on the chain is longer, and the granularity of tampering detection is coarser. The key frame density is input into the model as a factor affecting data reliability: high key frame density usually improves the reliability of data integrity verification, while too low density may reduce the sensitivity to tampering.

[0079] Exposure heat: refers to the exposure degree or attention heat of the original visual data in the real world, quantified by the number of times the data is accessed, the range of distribution, the heat index on social media, etc.

[0080] The higher the exposure heat, the more subjects the data is acquired and concerned by, and the higher the probability of being maliciously tampered with or attacked by counterfeiters. For example, a widely disseminated video is more likely to be targeted by attackers than an archived video that is not publicly disclosed. Therefore, exposure heat can be input as a risk factor: high exposure data tends to reduce the predicted security bit value in the model, prompting early strengthening of protection; on the contrary, low exposure private data is less likely to be attacked and is relatively safe.

[0081] Logical gate error: refers to the average error rate of the current quantum computer logical gate operation, which is considered as an indicator of the maturity of quantum computing capability.

[0082] The lower the logical gate error, the more gate operations the quantum computer can perform in series while maintaining high reliability, thereby having the ability to run more complex and deep quantum algorithms (such as Shor decomposition or Grover search on large keys). Therefore, low logical gate error means that quantum attack capability is improving, posing a greater threat to traditional cryptographic algorithms. The model takes the logical gate error as a key observation reflecting quantum progress: when the error rate decreases, the model tends to predict faster decay of the security bit. This parameter is usually obtained from public data in the quantum computing field, such as the average two-bit gate error rate published by major quantum chip providers.

[0083] Security bit history data: record the history curve data of the number of security bits of each signature algorithm changing over time, that is, the evolution trajectory of the theoretical / effective security bits in the past period;

[0084] This module uses a Bayesian network model to infer the above observation vectors, and outputs the current expected security bits of each algorithm. The model structure is designed as follows:

[0085] Node definition: the key frame density, exposure heat, logic gate error, and security bit history data are used as input nodes;

[0086] These input nodes receive observation values independently of each other, and the model sets two intermediate hidden nodes: data reliability and system stability. The data reliability node reflects the degree to which visual data is not easily tampered with, mainly affected by key frame density and exposure heat. The system stability node represents the robustness of the underlying cryptographic system, mainly based on logic gate error and historical security bit decay trend. Through these two intermediate layers, the model can express the influence of "content characteristics" and "algorithm environment" on security. The output node of the model is the current predicted expected security bit, which is the mathematical expectation of the actual security bit that the algorithm can achieve under the influence of the current factors.

[0087] In the Bayesian network connection relationship, the key frame density and exposure heat in the input node point to the data reliability DR node, and the logic gate error and historical data point to the system stability SS node; DR and SS nodes jointly affect the output expected security bit node;

[0088] On the one hand, the reliability of the data itself will affect the final security strength evaluation, and on the other hand, the stability of the external environment of the cryptographic system has a direct impact on the security strength. Through the learning of historical data, the model parameters are trained or adjusted by a large amount of simulation and real data, thereby forming the prediction ability of the possible security bit changes in the future. When new situations are observed, the Bayesian model can infer the corresponding changes in the expected security bit based on existing knowledge, realizing real-time inference.

[0089] Introduce "quantum threat weakening bits" θ to further combine the impact of quantum computing and provide a corresponding estimation method;

[0090] The quantum threat weakening bits θ is defined as: in the case of considering the potential breakthrough of quantum computing, the number of bits that the security strength of the algorithm may be additionally weakened, the ability of quantum attack has improved significantly in recent years, and the existing security bits may decrease sharply. The estimation of θ can be modeled by integrating the following factors:

[0091] Using the aforementioned algorithm security bit history data, the downward trend of security bits in the future period is inferred, and the current time is t0, and the expected security bit is Select a prediction time window, estimate the security bit at t0+Δt by fitting the historical curve Then the base weakening bit number can be taken as ΔB=B(t0)-B(t0+Δt), which reflects the natural decay amplitude of the security bit under the evolution speed of the conventional attack.

[0092] Considering the acceleration effect brought by quantum computing, the historical curve may not yet reflect the sudden impact of future quantum breakthroughs, so we introduce a quantum technology development index to assess potential additional security bit loss. If quantum technology progress is observed to be faster than expected, an additional loss value ΔQ is added based on historical data, obtained by simulating the impact of quantum attacks on the current algorithm.

[0093] In summary, the two parts can be added to obtain the quantum threat weakening bit number: θ=ΔB+ΔQ, and a slightly higher value is taken according to the conservative principle to ensure security redundancy;

[0094] Risk label division and early warning: the evaluation module compares the current expected security bit with the remaining security bit, and combines the quantum threat weakening bit number θ to label the security risk level of each piece of data:

[0095] Ample: when the remaining security bit ≥ expected security bit, the data is in a very safe state, and the effective security bit not only meets the minimum requirement, but also exceeds the model prediction requirement;

[0096] Warning: when the expected security bit > remaining security bit ≥ expected security bit-θ, the data is in a safe critical range. At this time, the remaining security bit is positive but lower than the model expectation, meaning that the security margin is insufficient and attention is needed but not completely over-standard;

[0097] Buffer: when expected security bit-θ> remaining security bit ≥ expected security bit-(1-α)·θ, where α is an empirical coefficient, 0<α<1, used to adjust the buffer width, the data security almost touches the unsafe edge;

[0098] Danger: when the remaining security bit ≤ 0, it means that the effective security bit has fallen below the application scenario requirement, and the algorithm security cannot be guaranteed or the algorithm has already failed;

[0099] It should be noted that the threshold values used in the above determination process can be adjusted and set by those skilled in the art according to specific application requirements, to ensure that the upgrade operation is not triggered too frequently, and the security turning point is not missed. Through real-time evaluation and graded early warning, when quantum technology makes progress and the cracking cost decreases, the model will reflect the decrease in security bit in time, thereby triggering the subsequent steps in advance to ensure the long-term security of on-chain data.

[0100] Output and re-sign trigger: for data marked as "buffer" or "danger", automatically add it to the re-sign list, which requires the signature algorithm upgrade operation to be performed as soon as possible, and is handled by the rollbackable re-sign module; for data marked as "warning", generate a warning log or notification to remind relevant personnel to pay attention, but do not change its signature; for data in "ample" state, maintain normal and do not need any operation.

[0101] In the rollbackable re-sign module, a new signature is generated for the original visual data in the re-sign list by using another quantum algorithm, a periodic rollbackable re-sign mechanism is constructed, the signature and the original content identifier are packaged into a hash time-locked contract transaction, and the transaction is set to be periodically written into the blockchain; if a vulnerability is found within the lock period, it can be revoked, and the old signature is automatically replaced at the end of the period. The specific content includes:

[0102] According to the re-sign list from the evaluation module, a new digital signature is generated for the corresponding image data, and the old signature is smoothly transitioned to the new signature through a blockchain transaction. In the signature update process, a rollbackable mechanism is used to ensure that if there is a problem with the new algorithm, it can be safely revoked, avoiding affecting the credibility of the original data.

[0103] Listen to the re-sign list, take out the visual data identifier CID and related information that needs to be re-signed, each task contains the CID of the original content, the currently used signature algorithm type, the corresponding public key, and the recommended new algorithm type, etc. The task is pushed by the previous module according to the risk level, for example, "buffer" state usually requires re-signing as soon as possible, while "danger" state belongs to emergency re-signing.

[0104] According to the task requirements, a new signature algorithm or parameter is selected to generate a replacement signature for the data to be re-signed. The new algorithm generates a new digital signature for the CID of the original visual data. The new signature is a signature of the original visual data content identifier CID, not a re-signature of the old signature. This ensures that no matter how many times the re-signing process is experienced, the CID recorded on the chain always corresponds to the original visual data itself, and the verification is always directly against the data content for hashing and comparison, without the need to trust the old signature layer by layer, thereby maintaining the simplicity and reliability of the chain evidence. The private key corresponding to the new algorithm is used to generate the signature.

[0105] The module packages the "original content CID" and "new signature" together as a special blockchain transaction to realize the delay effect and revocable mechanism of signature update;

[0106] Specifically, the transaction condition is constructed in the form of a hash time-locked contract, which is implemented as follows: the transaction contains an output locking a "signature update statement", which can be unlocked only when one of the two conditions is met: one is that the transaction automatically takes effect when the predetermined block time / height arrives, and the new signature is officially written on the chain; the other is that the transaction is revoked and invalidated by providing a pre-agreed hash value image before the time lock takes effect;

[0107] With this design, the introduction of the new signature is placed in a time window: if any vulnerabilities or problems are found during this window, the pre-set secret value is provided to trigger the hash lock, and the update transaction is revoked, so that the old signature continues to be valid, avoiding the introduction of a defective signature; if no problems occur until the end of the lock period, the transaction is automatically unlocked, and the new signature officially takes effect to replace the old signature. The entire transaction is equivalent to a conditional escrow: during the lock period, the old and new signatures are in a "dual-track coexistence" transition state, and the old and new signatures are recorded on the chain to prevent possible risks during the update process.

[0108] The constructed re-signing transaction with lock conditions is submitted to the blockchain network, and the parameters of the lock period are set. When the re-signing transaction enters the lock period, the module enters a monitoring state;

[0109] If the revocation condition is triggered during the monitoring process, the revocation right in the HTLC contract is exercised. If the lock period ends without triggering the revocation, the blockchain automatically executes the time lock condition, making the re-signing transaction officially effective. At this time, the new signature is recorded on the chain and accepted, indicating that the old signature is replaced: the current valid signature under the content CID is updated to the new algorithm signature, and the original old signature record is invalidated or archived. After that, the interlocking verification module and other verification parties should preferentially use the new signature to verify the content integrity. Since the CID does not change and the data itself does not change, the verifier only needs to know which public key / algorithm to use for verification. After taking effect, the old signature is still in the blockchain history, but it no longer participates in the subsequent regular verification process.

[0110] Through the design of the above-mentioned rollback re-signing module, smooth upgrade of the digital signature algorithm is achieved: it not only ensures the continuous trustworthiness of historical data on the blockchain, but also provides a safety net for problem rollback, avoiding new risks caused by hasty upgrade.

[0111] In the interlocking verification module, a dual-track strategy is adopted, which preferentially verifies the quantum-resistant signature and fails to verify the traditional signature. The signature state is monitored and the periodic rollback re-signing implementation is implemented, triggering four types of responses: normal, alarm, automatic upgrade, or forced switching, and the results are fed back to the model. The specific content includes:

[0112] When there is original visual data to be verified for integrity, the interlocking verification module starts a "double-track" verification process:

[0113] The current quantum-resistant digital signature of the data recorded on the chain and the corresponding algorithm public key are extracted, and the quantum-resistant algorithm is used to verify whether the signature matches the original content CID. Since the quantum-resistant signature is the core of future security, it is the primary trusted credential. If the quantum-resistant signature verification is passed, it proves that the data has not been tampered with and the quantum-resistant algorithm is currently valid. At this time, the integrity of the data is guaranteed at the quantum security level;

[0114] Regardless of the result of the quantum-resistant signature verification, the corresponding classical elliptic curve signature can be selected as a fallback verification, meaning that the traditional signature must be checked in the case of quantum-resistant signature verification failure;

[0115] Through double-track parallel / sequential verification, the module can produce the following several judgment cases:

[0116] Both signatures pass: indicating that the data is intact and both types of algorithms are working properly, the data state is "normal". At this time, the verification module returns a positive result of data integrity, allowing the application layer to trust the use of the data.

[0117] Single signature passes, including: quantum-resistant signature passes but traditional signature fails and quantum-resistant signature fails but traditional signature passes;

[0118] Quantum-resistant signature passes but traditional signature fails: indicating that the data itself is complete, as the quantum-resistant signature is trusted to pass verification, but the classical signature verification fails;

[0119] Quantum-resistant signature fails but traditional signature passes: meaning that the data integrity cannot be verified at the quantum security level, but the classical verification believes that the data has not been tampered with;

[0120] Both signatures fail: indicating that the data has not been verified by either trusted signature, and it is likely that the data has been tampered with and the signature cannot be matched, indicating that the data integrity is severely damaged, and the upper application will be notified that the data is not trusted;

[0121] Continuously monitor the signature status of each data item on the chain, including signature validity and update progress, according to the double-track verification result and the risk label provided by the evaluation module, the interlocking verification module can trigger four types of responses to coordinate system actions, namely normal, alarm, automatic upgrade and forced switching. These responses correspond to the aforementioned risk labels and further take actual measures:

[0122] Normal response: when the label is "abundant" and both signatures pass, it is in normal operation and no special intervention is needed;

[0123] Alarm response: when the label is "warning" and both signatures pass, the label is "warning" and a single signature passes, and the label is "abundance" and a single signature passes, the alarm response is triggered when a slight anomaly is found in the dual-track verification;

[0124] Forced switching response: when the label is "danger" and both signatures fail, the verification module immediately performs a forced switching response. Forced switching includes two meanings: one is to switch the trusted path, that is, to immediately stop trusting the algorithm signature that has failed in the verification logic to prevent attackers from bypassing the verification, and the other is to switch the algorithm scheme, that is, to quickly deploy a new safer algorithm to replace the original algorithm. For a single data, forced switching is essentially triggering an emergency re-signing, that is, re-signing with a new algorithm of the highest level, and the process is performed by the forced mode of the aforementioned rollbackable re-signing module;

[0125] Automatic upgrade response: in addition to the normal response, the alarm response, and the forced switching response described above, the automatic upgrade response is triggered when the verification module needs to enter the re-signing process;

[0126] After the above responses are executed, relevant information is fed back to the Bayesian model of the remaining security assessment module, which is used to update the input and parameters of the model to make the next round of assessment more accurate.

[0127] Through the mechanism of dual-track verification + active response + feedback learning, the functional modules are organically connected to form the operation center of the application. Under its coordination, the blockchain storage, risk assessment, and signature update are closely linked and controlled in a closed loop, which not only ensures the efficient and reliable daily data verification, but also ensures that the strategy can be quickly adjusted in an emergency state to maintain data security and credibility.

[0128] The above formulas are dimensionless values calculated, and the formulas are obtained by software simulation of a large amount of data to obtain a formula of the latest real situation. The preset parameters in the formula are set by a person skilled in the art according to the actual situation.

[0129] The above embodiments can be realized wholly or partially by software, hardware, firmware, or any other combination. When realized by software, the above embodiments can be realized in the form of a computer program product.

[0130] Those skilled in the art can realize that the modules and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application of the technical solution and the constraints of the application. A person skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0131] In addition, each functional module in each embodiment of the present application can be integrated in one processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0132] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0133] Finally: the above is merely preferred embodiments of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be covered in the protection scope of the present application.

Claims

1. The blockchain-based visual data traceability and integrity verification system is characterized by: include: The dual-signature first evidence storage module, the remaining security assessment module, the rollback re-signature module, and the interlocking verification module have signal connections between each module; Double-signature initial evidence storage module: obtains the original visual data, calculates the content identifier CID, signs the original visual data using the elliptic curve algorithm and the lattice-based quantum-resistant algorithm respectively, and packages the content identifier and the two signatures into the initial on-chain record as the original and tamper-proof evidence; Remaining Safety Assessment Module: This module obtains the remaining safety bits of the current raw visual data and uses a Bayesian model to combine the key frame density, exposure heat, and logic gate error of the aggregated raw visual data to calculate the expected safety bit corresponding to the safety bit index. It then labels each raw visual data with a risk label and places the buffered and dangerous images into a re-signing list. Rollback re-signature module: Generate a new signature for the original visual data of the re-signature list using a post-quantum algorithm, construct a periodic rollback re-signature mechanism, encapsulate the signature and the original content identifier into a hash time-locked contract transaction, set a period and write it to the blockchain; If a vulnerability is found during the lock period, it can be revoked, and the old signature will be automatically replaced upon expiration; Interlocking verification module: adopts a dual-track strategy, giving priority to verifying quantum-resistant signatures, and falls back to verifying traditional signatures in case of failure. It monitors signature status and regularly rolls back and re-signs, triggering four types of responses: normal, alarm, automatic upgrade, or forced switch, and feeding the results back to the model.

2. The blockchain-based visual data traceability and integrity verification system according to claim 1 is characterized by: Obtaining original visual data, and calculating a hash on the original visual data to obtain a content identifier CID; The device uses the classic elliptic curve algorithm to generate a digital signature Sig for the content identifier CID ECC , and simultaneously use the lattice-based quantum-resistant algorithm to generate a digital signature Sig for the content identifier CID Dilithium ; Sig ECC With Sig Dilithium Combined into a composite signature; Write the content identifier CID and the composite signature into the content chain.

3. The blockchain-based visual data traceability and integrity verification system according to claim 1 is characterized by: The ideal security strength derived from the algorithm parameters is the theoretical security bit B0; Get the number of operations required by the latest cracking algorithm to get the currently known strongest attack complexity C t And the current cost ratio of unit hash operation is expressed as the computing power price P t , through B t =B0-log2! (C t / P t ), approximately calculate the real effective security bit B at time t t ; Set the minimum required security level B based on the security requirements of the application scenario req , calculate the remaining security bit R t Valid security bit B t With minimum required security bit B req difference.

4. The blockchain-based visual data traceability and integrity verification system according to claim 3 is characterized by: The key frame density, exposure heat and logic gate error data of the original visual data are collected to form the observation vector O t , build a Bayesian model: Define input nodes: key frame density, exposure heat, logic gate error, and safety bit history data; Construct intermediate nodes: Data reliability nodes integrate key frame density and exposure popularity, and system stability nodes integrate logic gate errors and historical security bit attenuation trends; Output expected security bit; Calculation of the number of bits weakened by quantum threats: The current time is t0, and the expected security bit is Fitting the historical curve to estimate the safety position at time t0+Δt Then the basic weakening bit number can be taken as ΔB=B(t0)-B(t0+Δt); The additional security bit loss caused by quantum attacks is simulated according to the rate of quantum technology advancement, which is recorded as the quantum additional loss ΔQ; The sum of the basic weakening bit ΔB and the quantum additional loss ΔQ is taken as the final quantum threat weakening bit θ.

5. The blockchain-based visual data traceability and integrity verification system according to claim 4 is characterized by: Compare the calculated current expected security bits with the remaining security bits, and combine the number of quantum threat weakening bits θ to label each piece of data with its security risk level: When the remaining security bits ≥ the expected security bits, mark "sufficient"; When the expected safety bit>remaining safety bit≥(expected safety bit-θ), mark "warning"; When expected security bit - θ > remaining security bit ≥ expected security bit - (1-α)·θ, mark "buffer", where α∈(0,1) is an adjustable coefficient; When the remaining safety bit is ≤0, mark "Danger"; For data marked as "buffered" or "dangerous", it is automatically added to the re-signing list.

6. The blockchain-based visual data traceability and integrity verification system according to claim 1 is characterized by: Construct a hash time-lock contract transaction: including the original CID, new signature, and preset time lock; Set dual validation conditions: automatically take effect upon reaching a predetermined block height, or cancel the transaction by providing a hash preimage within the lock-up period; During the lock-up period, the old and new signatures will coexist, and the valid signature records on the chain will be automatically updated after the expiration of the period; Generate a new signature for the CID in the re-signature list and bind and encapsulate the new signature with the original CID.

7. The blockchain-based visual data traceability and integrity verification system according to claim 1 is characterized by: When there is raw visual data that needs to be verified for integrity, the interlocking verification module initiates a "dual-track" verification process: First track verification: Prioritize using quantum-resistant public keys to verify Sig Dilithium ; Second track verification: using traditional public key to verify Sig ECC ; Four types of judgment results are generated: both signatures pass, the quantum-resistant signature passes but the traditional signature fails, the quantum-resistant signature fails but the traditional signature passes, and both signatures fail. The quantum-resistant signature passes but the traditional signature fails, and the quantum-resistant signature fails but the traditional signature passes are recorded as single signature passes.

8. The blockchain-based visual data traceability and integrity verification system according to claim 7 is characterized by: Based on the dual-track verification results and the risk tags provided by the assessment module, the interlock verification module can trigger four types of responses to coordinate system actions: normal, alarm, automatic upgrade, and forced switch: Normal response: When the tag is "sufficient" and both signatures pass, the system is operating normally and no special intervention is required. Alarm response: Pay attention when the label is "Warning" and both signatures pass, the label is "Warning" and a single signature passes, or the label is "Abundant" and a single signature passes. Forced switching response: When the label is "dangerous" and both signatures fail, the verification module immediately performs a forced switching response, including switching the trust path and switching the algorithm scheme; Automatic upgrade response: All responses other than the above-mentioned normal response, alarm response, and forced switching response are recorded as automatic upgrade responses. When the re-signing process needs to be prepared, the verification module triggers the automatic upgrade response; After executing the above responses, the relevant information is fed back to the Bayesian model of the remaining security assessment modules.

Images