Intelligent approval methods, devices, computer equipment, readable storage media, and program products

By using multimodal biometrics to sign smart contracts and store them on the blockchain, the problem of manual annotation required for data flow in traditional methods is solved, realizing automated approval and data self-verification, and improving system efficiency and credibility.

CN123116241APending Publication Date: 2026-11-17YUNNAN NANTIAN INFORMATION EQUIP CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611389557.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-08
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In traditional methods, the identities of the parties involved in resource interaction, the sensor data output by environmental acquisition devices, and the business targets of smart contracts are independent, which requires manual attribution, labeling, and association during data flow, resulting in low system processing efficiency and insufficient data credibility.

Method used

By acquiring the multimodal biometrics of both parties involved in the resource interaction to sign a smart contract, binding environmental data collection devices and generating trusted data packets, storing the data on the blockchain, and generating credit ratings based on the blockchain-stored records and biometrics, automated approval can be achieved.

Benefits of technology

It enables data to self-certify its true source and business affiliation without manual annotation, improving system processing efficiency and data credibility, dynamically reflecting the true performance capabilities of both parties in resource interaction, and automatically outputting approval results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure FT_1
    Figure FT_1
  • Figure FT_2
    Figure FT_2
  • Figure FT_3
    Figure FT_3
Patent Text Reader

Abstract

This application relates to an intelligent approval method, apparatus, computer equipment, computer-readable storage medium, and computer program product. The method includes: acquiring multimodal biometrics of both parties involved in the resource interaction, and signing a smart contract based on the multimodal biometrics to obtain an effective smart contract; binding environmental data collection devices associated with the effective smart contract based on the contract address in the effective smart contract, and acquiring raw environmental data through these devices; generating a trusted data packet based on the raw environmental data and the contract address in a trusted execution environment, and sending the trusted data packet to a blockchain network's notarization node for on-chain notarization to obtain a blockchain notarization record; generating a credit rating based on the blockchain notarization record, the warning threshold in the effective smart contract, and the multimodal biometrics; and obtaining an approval result based on the credit rating. This method enables automated approval.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to an intelligent approval method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Technology

[0002] With the rapid development of e-commerce and cross-border trade digitization technologies, underlying technologies such as blockchain, Trusted Execution Environments (TEEs), and multimodal biometrics are maturing. Blockchain technology, with its decentralized, immutable, and traceable characteristics, is widely used in data storage and smart contract execution. Trusted Execution Environment technology provides isolated secure areas at the hardware level, ensuring the confidentiality and integrity of data during processing. Multimodal biometrics technology, by integrating multiple biometric features such as facial features and fingerprints, significantly improves the accuracy and anti-counterfeiting capabilities of identity authentication.

[0003] However, in traditional methods, the identities of the parties involved in resource interaction, the sensor data output by environmental acquisition devices, and the business targets corresponding to smart contracts are independent of each other. When data flows through each stage, manual attribution labeling and association configuration are required, which not only leads to low system processing efficiency, but also makes it impossible for data to prove its true source and business attribution when entering the credit assessment stage. Overall, there are defects in data credibility breakpoints and insufficient automation of processing. Summary of the Invention

[0004] Therefore, it is necessary to provide an intelligent approval method, device, computer equipment, computer-readable storage medium, and computer program product that can achieve automated approval in response to the above-mentioned technical problems.

[0005] Firstly, this application provides an intelligent approval method, the method comprising: Obtain the multimodal biometrics of both parties involved in the resource interaction, and sign a smart contract based on the multimodal biometrics to obtain an effective smart contract; Based on the contract address in the effective smart contract, bind each environmental acquisition device associated with the effective smart contract, and obtain raw environmental data through each environmental acquisition device; In the trusted execution environment, based on the original environment data and contract address, a trusted data packet is generated and sent to the evidence storage node of the blockchain network for on-chain evidence storage, thereby obtaining a blockchain evidence storage record. A credit rating is generated based on the blockchain evidence records, the early warning thresholds in the effective smart contracts, and the multimodal biometrics. Based on the credit rating, the approval result is obtained.

[0006] In one embodiment, generating a credit rating based on the blockchain-stored evidence, the early warning threshold in the effective smart contract, and the multimodal biometrics includes: Based on the blockchain evidence records and the warning thresholds in the effective smart contract, a warning message is generated; Based on the aforementioned warning information, an on-chain warning log record is generated; A credit rating is generated based on the on-chain early warning log records and the multimodal biometrics.

[0007] In one embodiment, generating a credit rating based on the on-chain alert log records and the multimodal biometrics includes: Based on the on-chain warning log records and the liability determination rules in the effective smart contract, the determination result of the responsible party is determined; Based on the determination of the responsible party and the blockchain evidence record, a responsibility determination report is generated; Based on the aforementioned multimodal biometrics, corresponding historical behavioral data is obtained from the blockchain. The credit rating is generated based on the liability determination report and the historical behavior data.

[0008] In one embodiment, generating warning information based on the blockchain evidence record and the warning threshold in the effective smart contract includes: In the trusted execution environment, the blockchain evidence record is decrypted and restored to plaintext data; If the plaintext data in the sliding window exceeds the warning threshold in the effective smart contract, multiple warning states will be triggered sequentially according to preset priority. Record the triggering action, escalation action, and cancellation action for each level of warning status, as well as the timestamp corresponding to the action, to generate the warning information.

[0009] In one embodiment, the smart contract signed based on multimodal biometrics to obtain an effective smart contract includes: Based on the aforementioned multimodal biometrics, biometric identity anchors are generated; Generate on-chain digital identity keys for both parties involved in the resource interaction based on the biometric identity anchor. The smart contract is electronically signed by the keys of both parties involved in the resource exchange, resulting in an effective smart contract.

[0010] In one embodiment, before generating the trusted data packet based on the original environment data and contract address in the trusted execution environment, the method further includes: The device identifier and device signature of each environmental acquisition device are parsed and extracted from the original environmental data. Obtain the public key of the corresponding environmental data acquisition device based on the device identifier; The device signature is verified using the public key, and if the verification is successful, the original environment data is sent to the trusted execution environment. The process of generating a trusted data packet based on the original environment data and contract address in a trusted execution environment includes: The original environment data is processed for format normalization in the trusted execution environment to obtain environment data with a unified format. Symmetric encryption is applied to the environmental data in the unified format to obtain ciphertext data; Calculate the hash value of the environmental data in the unified format; The encrypted data, hash value, device identifier, and contract address are associated to obtain the trusted data packet.

[0011] Secondly, this application also provides an intelligent approval device, the device comprising: The signing module is used to obtain the multimodal biometrics of both parties involved in the resource interaction, and to sign a smart contract based on the multimodal biometrics to obtain an effective smart contract; The binding acquisition module is used to bind each environmental acquisition device associated with the effective smart contract based on the contract address in the effective smart contract, and to obtain raw environmental data through each of the environmental acquisition devices; The evidence storage module is used to generate a trusted data packet based on the original environment data and contract address in a trusted execution environment, and send the trusted data packet to the evidence storage node of the blockchain network for on-chain evidence storage to obtain a blockchain evidence storage record. The generation module is used to generate a credit rating based on the blockchain evidence record, the early warning threshold in the effective smart contract, and the multimodal biometrics. The approval module is used to obtain the approval result based on the credit rating.

[0012] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.

[0013] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0014] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0015] The aforementioned intelligent approval method, device, computer equipment, computer-readable storage medium, and computer program product first acquire the multimodal biometrics of both parties involved in the resource interaction and sign a smart contract based on these biometrics, resulting in an effective smart contract. Multimodal biometrics are more resistant to interference than single-modal biometrics, establishing a mapping relationship between physical world identities and on-chain digital identities. Second, based on the contract address in the effective smart contract, various environmental data collection devices associated with the effective smart contract are bound, and raw environmental data is acquired through these devices. Each piece of data generated by the devices is inherently linked to the contract at the underlying level, eliminating the need for additional manual labeling or association operations during subsequent data processing. The data is digitally tagged with the business contract from the source of collection, avoiding potential association errors or human fraud that may occur when labeling data after collection in traditional solutions. Third, in a trusted execution environment, a trusted data packet is generated based on the raw environmental data and the contract address, and this trusted data packet is sent to the blockchain network's evidence storage node for on-chain evidence storage, resulting in a blockchain evidence record. The originality and integrity of the data can be self-verified based on the on-chain record. Finally, a credit rating is generated based on blockchain-stored records, early warning thresholds in effective smart contracts, and multimodal biometrics; based on the credit rating, the approval result is obtained. The credit rating dynamically reflects the current actual performance capability of both parties in the resource interaction, rather than a static indicator. The approval result is automatically output based on the credit rating, without human intervention, thus achieving automated approval. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating an intelligent approval method in one embodiment; Figure 2 This is a schematic diagram illustrating the process of generating a credit rating based on blockchain-stored records, early warning thresholds in effective smart contracts, and multimodal biometrics in one embodiment. Figure 3 This is a schematic diagram illustrating the process of generating credit ratings based on on-chain early warning logs and multimodal biometrics in one embodiment. Figure 4 This is a schematic diagram illustrating the process of generating warning information based on the warning threshold in a blockchain-stored record and an effective smart contract, as shown in one embodiment. Figure 5This is a schematic diagram illustrating the process of signing a smart contract based on multimodal biometrics to obtain an effective smart contract in one embodiment. Figure 6 This is a schematic diagram of the verification process in one embodiment; Figure 7 This is a schematic diagram illustrating the process of generating trusted data packets based on raw environment data and contract addresses in a trusted execution environment, as shown in one embodiment. Figure 8 This is a structural block diagram of an intelligent approval device in one embodiment; Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] In one embodiment, such as Figure 1 As shown, an intelligent approval method is provided. This embodiment illustrates the method applied to a terminal, but it is understood that the method can also be applied to a server, and further to a system including both a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps S102 to S110. Wherein: Step S102: Obtain the multimodal biometrics of both parties involved in the resource interaction, and sign a smart contract based on the multimodal biometrics to obtain an effective smart contract.

[0020] Optionally, the camera captures raw facial images of both parties involved in the resource interaction, and the fingerprint sensor captures fingerprint feature point data from both parties. The camera sends the raw facial images to the terminal, and the fingerprint sensor sends the feature point data to the terminal. The terminal performs preprocessing on the raw facial images, including illumination normalization, pose correction, occlusion detection and removal, to obtain a standardized portrait. A standardized portrait feature vector is then extracted from the standardized portrait. The terminal performs feature layer fusion of the standardized portrait feature vector and the fingerprint feature point data to generate unique and uncopyable multimodal biometrics for each party involved in the resource interaction.

[0021] Optionally, the terminal performs a liveness check on the multimodal biometrics of both parties. If the check passes, the public keys, digital signatures, and contract hash values ​​of both parties are cross-validated against a pre-deployed smart contract template on the blockchain. This confirms that both parties' signatures are valid and the contract content has not been tampered with, resulting in a signed and effective smart contract. The effective smart contract includes the identity anchors of both signing parties, the signing timestamp (accurate to milliseconds of block time), and the contract version number.

[0022] Step S104: Based on the contract address in the effective smart contract, bind the environmental acquisition devices associated with the effective smart contract, and obtain the raw environmental data through the environmental acquisition devices.

[0023] Optionally, when deploying a smart contract, the contract code predefines data structures for managing bound devices, including a device mapping table: a mapping structure with device identity (DeviceID) as the key and device key, device type, binding time, status (bound / unbound) as the value; and a contract-device association table: a list of all device IDs that the current contract has bound.

[0024] After the smart contract takes effect, the terminal scans the connected temperature and humidity sensors and BeiDou GPS modules through physical interfaces (such as serial port, Bluetooth, ZigBee, etc.) to read the device hardware information of each environmental acquisition device, such as the unique hardware serial number (burned by the manufacturer and cannot be changed), device model, measurement range, calibration date, and other metadata. Based on the acquired device hardware information, the terminal generates a unique device identification identifier (DeviceID) and corresponding key for each environmental acquisition device.

[0025] The device ID, key, type, binding time, and other information are written into the contract's device mapping table. The device ID is appended to the contract's "Bound Devices List," generating an on-chain event log entry that records "Device XXX has been bound to Contract XXX." This log entry includes the device ID, contract address, and binding timestamp. Subsequently, all data collected by this device will carry this contract address when uploaded to the blockchain. When receiving data, the smart contract will verify whether the contract address in the data matches the bound record.

[0026] Step S106: In the trusted execution environment, a trusted data packet is generated based on the original environment data and contract address, and the trusted data packet is sent to the evidence storage node of the blockchain network for on-chain evidence storage to obtain the blockchain evidence storage record.

[0027] Optionally, after receiving the raw environment data, the trusted execution environment verifies the device signature carried in the raw environment data. If the verification passes, the raw environment data is moved into a secure memory area; if the verification fails, the data is discarded and an exception log is generated. The raw environment data includes at least one of the following: temperature data and its acquisition timestamp, humidity data and its acquisition timestamp, and location coordinate data and its acquisition timestamp.

[0028] The terminal generates a trusted data packet based on the original environment data and the contract address. The evidence storage node then stores this trusted data packet on the blockchain, resulting in a blockchain evidence storage record. This blockchain evidence storage transaction record contains the contract address, enabling the blockchain evidence storage record to be linked to the active smart contract on-chain via the contract address.

[0029] Step S108: Generate a credit rating based on blockchain-stored records, early warning thresholds in effective smart contracts, and multimodal biometrics.

[0030] Among them, the credit rating is used to dynamically reflect the current actual performance capability of both parties in resource interaction, rather than a static indicator.

[0031] Optionally, the terminal uses the on-chain digital identity corresponding to the multimodal biometrics as an index to retrieve current and historical data information from the blockchain's evidence storage records of both parties involved in the resource interaction. Historical data includes historical behavioral data, environmental compliance rate data, and external compliance data. Examples of historical behavioral data include historical order fulfillment rate data.

[0032] The system compares the current data information in the blockchain evidence record with the warning threshold in the effective smart contract to generate warning information. Then, based on the warning information and the blockchain evidence record, it generates the credit rating of the two parties involved in the resource interaction.

[0033] Step S110: Based on the credit rating, the approval result is obtained.

[0034] Optionally, the terminal matches the corresponding credit multiple from a preset credit parameter configuration table based on the credit rating. For example, a credit rating of A corresponds to the first credit multiple, B corresponds to the second credit multiple, and C corresponds to the third credit multiple, with the first credit multiple being greater than the second credit multiple, which in turn is greater than the third credit multiple.

[0035] Obtain the transaction amount data for the current transaction, and calculate the credit limit based on the credit multiple and the transaction amount data. Compare the credit limit with the financing application amount for the current transaction. If the credit limit is greater than or equal to the financing application amount, an approval result is generated; otherwise, an approval result is generated.

[0036] Optionally, when the credit line is greater than or equal to the financing application amount, the application is approved and a disbursement instruction is generated. When the smart contract executes the disbursement instruction, it verifies whether the payee's account information is consistent with the supplier's account information bound in the effective smart contract. If they are inconsistent, the disbursement is rejected. After the disbursement is completed, the logistics evidence data in the blockchain evidence record, the contract information in the effective smart contract, and the disbursement record are aligned on a timeline to generate a three-stream integration closed-loop report and uploaded to the blockchain for evidence storage.

[0037] In the aforementioned intelligent approval method, firstly, multimodal biometrics of both parties involved in the resource interaction are acquired, and a smart contract is signed based on these multimodal biometrics to obtain an effective smart contract. Multimodal biometrics are more resistant to interference than single-modal biometrics, establishing a mapping relationship between physical identities and on-chain digital identities. Secondly, based on the contract address in the effective smart contract, various environmental data collection devices associated with the effective smart contract are bound, and raw environmental data is obtained through these devices. Each piece of data generated by the devices is inherently linked to the contract at the underlying level, eliminating the need for additional manual labeling or association operations during subsequent data processing. The data is digitally tagged with the business contract from the source of collection, avoiding potential association errors or human fraud that may occur when labeling data after collection in traditional solutions. Thirdly, in a trusted execution environment, a trusted data packet is generated based on the raw environmental data and the contract address, and this trusted data packet is sent to the blockchain network's evidence storage node for on-chain evidence storage, resulting in a blockchain evidence record. The originality and integrity of the data can be self-verified based on the on-chain record. Finally, a credit rating is generated based on the blockchain evidence record, the warning threshold in the effective smart contract, and the multimodal biometrics; the approval result is obtained based on the credit rating. Credit rating is used to dynamically reflect the current actual performance capability of both parties in resource interaction, rather than a static indicator. Approval results are automatically generated based on credit rating, without human intervention, thus achieving automated approval.

[0038] In one exemplary embodiment, such as Figure 2 As shown, a credit rating is generated based on blockchain-stored records, early warning thresholds in effective smart contracts, and multimodal biometrics, including the following steps S202 to S206. Wherein: Step S202: Generate warning information based on the blockchain evidence records and the warning threshold in the effective smart contract.

[0039] Optionally, the monitoring subroutine of the smart contract listens to newly generated blocks in the blockchain network, extracts blockchain evidence records that match the contract address of the effective smart contract from the new blocks, obtains the ciphertext data corresponding to the blockchain evidence records from the off-chain distributed storage system through an oracle, and decrypts the ciphertext data to restore it to plaintext data in a trusted execution environment; compares the plaintext data with the preset warning threshold in the effective smart contract, and generates a warning message when the plaintext data exceeds the warning threshold.

[0040] The plaintext data includes at least one of temperature data, humidity data, and location coordinate data; the warning thresholds include temperature thresholds, humidity thresholds, and location deviation thresholds that correspond one-to-one with each type of environmental data; different types of plaintext environmental data are independently compared with their respective warning thresholds, and when any type of plaintext data exceeds its corresponding warning threshold, the warning information corresponding to that type is triggered independently.

[0041] Step S204: Based on the warning information, generate on-chain warning log records.

[0042] Among them, on-chain early warning log records refer to the immutable evidence records generated after the early warning information is written into the blockchain ledger.

[0043] Optionally, after the terminal writes the warning information into the blockchain ledger, it generates an immutable evidence record, namely an on-chain warning log record.

[0044] Step S206: Generate a credit rating based on on-chain early warning log records and multimodal biometrics.

[0045] Optionally, the terminal uses the on-chain digital identity corresponding to the multimodal biometrics as an index to retrieve the party's historical data information and on-chain warning log records from the blockchain's notarized records. From the retrieved historical data, the terminal extracts the party's historical order fulfillment rate data and environmental data compliance rate data; wherein, the historical order fulfillment rate data is determined based on the proportion of the number of smart contracts in which the party has historically participated that did not trigger warning information to the total number of contracts; the environmental data compliance rate data is determined based on the proportion of the number of data collection points in the environmental data associated with the party's currently active smart contracts that did not trigger warning information to the total number of data collection points.

[0046] External compliance data is obtained by connecting to external systems. Historical order fulfillment rate data, environmental compliance rate data, and external compliance data are standardized to generate three-dimensional standardized data. This three-dimensional standardized data is then input into a dynamic credit scoring model for weighted fusion calculation to generate a credit rating. The credit ratings are A, B, and C. A rating corresponds to a dynamic credit score ≥80 points, B rating corresponds to a dynamic credit score ≥60 points and <80 points, and C rating corresponds to a dynamic credit score <60 points.

[0047] In this embodiment, the entire process can be automated by automatically comparing the data with the warning threshold.

[0048] In one exemplary embodiment, such as Figure 3 As shown, a credit rating is generated based on on-chain early warning log records and multimodal biometrics, including steps S302 to S308. Wherein: Step S302: Based on the on-chain warning log records and the liability determination rules in the effective smart contract, determine the determination result of the responsible party.

[0049] Optionally, the terminal extracts abnormal event information from the on-chain warning log. This abnormal event information includes the environmental data type that triggered the warning, the environmental data value and its real-time collection timestamp, the warning duration, and the location coordinates corresponding to the time the warning was triggered. Using this abnormal event information as input, the terminal calls the responsibility determination rule engine in the active smart contract. The responsibility determination rule engine determines the abnormality type based on the environmental data type and value in the abnormal event information and activates the corresponding responsibility determination matrix based on the abnormality type. The responsibility determination matrix contains multiple determination factors, each with a weight value associated with different time stages. The responsibility determination rule engine determines the target time stage where the abnormal event occurred based on the real-time collection timestamp and location coordinates in the abnormal event information. Based on the target time stage, it obtains the weight values ​​of each determination factor corresponding to the target time stage from the responsibility determination matrix. Based on the weight values ​​of each determination factor and the actual parameter values ​​corresponding to each determination factor in the abnormal event information, it calculates the responsibility weight score for each candidate responsible party. The candidate responsible party with the highest responsibility weight score is determined as the responsible party.

[0050] The time stages include loading, transit, unloading, and customs inspection; the judgment factors include at least one of temperature, humidity, location, and time factors; in the responsibility judgment matrix, the weight value of the same judgment factor is different in different time stages; when the target time stage is transit, the weight value of the temperature factor is the first weight value; when the target time stage is unloading, the weight value of the temperature factor is automatically adjusted to a second weight value lower than the first weight value, and the weight value of the location factor is adjusted to a third weight value, which is greater than the weight value of the temperature factor corresponding to the unloading stage; the first weight value is greater than the second weight value.

[0051] For example, if an on-chain warning log records a temperature exceedance warning: the temperature is 9.2℃ (threshold 8℃), the duration is 45 minutes, the time period corresponding to the collection timestamp is the transit stage, and the location coordinates are on the sea transport route, the liability determination rule engine, upon receiving the above abnormal event information, identifies the abnormality type as "temperature exceedance" and activates the corresponding liability determination matrix. In this matrix, the weight of the temperature factor is 0.8, the weight of the location factor is 0.1, and the weight of the time factor is 0.1 during the transit stage; the carrier's actual temperature factor score is 100 points (because the exceedance is within the agreed range), the location factor score is 95 points (GPS is within the route), and the time factor score is 90 points (45 minutes of duration complies with the contract). The carrier's liability weight score = 100 × 0.8 + 95 × 0.1 + 90 × 0.1 = 98.5 points; the supplier's corresponding score is calculated in the same way as 30 points. 98.5 points > 30 points, therefore the carrier is determined to be the responsible party.

[0052] Step S304: Based on the determination of the responsible party and the blockchain evidence record, generate a responsibility determination report.

[0053] The liability determination report refers to a comprehensive electronic document that records all evidence information, liability determination process information, and liability attribution conclusions for an abnormal event. The liability determination report includes at least the identity of the responsible party, the proportion of liability, the basis for determination, the set of transaction hash values ​​of the blockchain evidence records cited, and the generation timestamp.

[0054] Optionally, the terminal determines the time interval corresponding to the abnormal event based on the warning trigger timestamp and warning cancellation timestamp in the on-chain warning log; based on the contract address and time interval of the effective smart contract, it extracts all blockchain evidence records associated with the contract address and within the time interval from the blockchain. All blockchain evidence records include the hash fingerprints of all environmental data collected during the warning trigger time period and the corresponding collection timestamp sequence; it obtains the encrypted data corresponding to the hash fingerprint from the off-chain distributed storage system through an oracle, and decrypts the encrypted data to restore it to plaintext environmental data in a trusted execution environment, generating a complete evidence dataset containing at least one of temperature change time series, humidity change time series, and GPS trajectory data; it packages and assembles the complete evidence dataset, the determination result of the responsible party, the weight configuration and actual parameter values ​​of each determination factor in the responsibility determination matrix, and the set of transaction hash values ​​of the referenced blockchain evidence records to generate a responsibility determination report, and signs the responsibility determination report with the private key of the contract account of the effective smart contract before storing it on the blockchain.

[0055] Step S306: Obtain corresponding historical behavior data from the blockchain based on multimodal biometrics.

[0056] Optionally, the terminal acquires the first multimodal biometric feature currently collected by the resource interaction party. The first multimodal biometric feature includes a multimodal biometric feature vector generated by fusing facial image features and fingerprint features. The terminal then compares the first multimodal biometric feature with a stored multimodal biometric feature template, which is generated when a smart contract is signed and established with a mapping relationship to the on-chain digital identity. After successful comparison, the terminal acquires the on-chain digital identity corresponding to the multimodal biometric feature template. Using the on-chain digital identity as an index, the terminal retrieves a list of contract addresses for all smart contracts in which the enterprise has historically participated from the blockchain. Based on the contract address list, the terminal retrieves the historical data of each contract from the blockchain. The system includes historical blockchain evidence records and historical on-chain warning log records corresponding to smart contracts; environmental data compliance rate data for each historical contract is extracted from the historical blockchain evidence records, and the environmental data compliance rate is determined based on the proportion of data collection points in each historical contract that did not trigger warning information to the total number of data collection points; the number of warning triggers and warning cancellation records for each historical contract are extracted from the historical on-chain warning log records; the performance status of each historical contract is determined based on the number of warning triggers and warning cancellation records, and the proportion of contracts that did not trigger warnings and completed transactions on time to the total number of contracts is calculated as the historical order performance rate data.

[0057] Step S308: Generate a credit rating based on the liability determination report and historical behavior data.

[0058] Optionally, historical behavioral data, such as historical order fulfillment rate data and environmental data compliance rate data, are standardized to generate two-dimensional standardized data. External compliance data of the party is obtained by connecting to external systems and standardized to generate third-dimensional standardized data. The three-dimensional standardized data is input into a dynamic credit scoring model for weighted fusion calculation to generate a basic credit score. After weighted fusion calculation, the determination result of whether the party currently responsible in the responsibility determination report is itself is obtained. If the responsibility determination report determines that the party is responsible for the current warning event, a first penalty coefficient is determined based on the company's responsibility ratio in the responsibility determination report, and a second penalty coefficient is determined based on the number of warning records that the company has not lifted in the on-chain warning log. The corresponding points are deducted from the basic credit score based on the first and second penalty coefficients. If the responsibility determination report determines that the company is not responsible for the current warning event, no penalty points are deducted. The corresponding credit rating is generated based on the final dynamic credit score after adjustment of the penalty factors.

[0059] In this embodiment, since all the evidence data for liability determination comes from the immutable evidence records on the blockchain, and the liability determination process is automatically executed by smart contracts, the entire chain from liability determination to credit assessment does not require human intervention, the system processing delay is significantly reduced, and the objectivity and timeliness of the assessment results are improved.

[0060] In one exemplary embodiment, such as Figure 4 As shown, based on the blockchain's stored evidence records and the early warning thresholds in the effective smart contracts, early warning information is generated, including steps S402 to S406. Wherein: Step S402: In a trusted execution environment, the blockchain evidence record is decrypted and restored to plaintext data.

[0061] Optionally, the terminal listens for new block generation events in the blockchain network through a smart contract monitoring subroutine. It extracts a blockchain evidence record from the new block that matches the contract address of the effective smart contract. This blockchain evidence record includes the hash value of the environmental data, the collection timestamp corresponding to the environmental data, and the device identification of the environmental data collection device. Based on the hash value of the environmental data in the blockchain evidence record, an oracle queries and retrieves the ciphertext data corresponding to the hash value from the off-chain distributed storage system. The oracle is a data relay service deployed between the blockchain and the off-chain storage system, used to respond to data retrieval requests from the blockchain, retrieve and return the corresponding ciphertext data from the off-chain storage system. The retrieved ciphertext data is then passed to the trusted execution environment. The trusted execution environment calls its built-in hardware decryption engine and uses a dynamically generated symmetric decryption key during the session cycle to decrypt the ciphertext data, obtaining the restored plaintext data. The symmetric decryption key and the symmetric encryption key used to encrypt the ciphertext data are generated by the same key derivation function, and encryption and decryption use the same session cycle key.

[0062] Step S404: If the plaintext data in the sliding window is greater than the warning threshold in the effective smart contract, trigger the multi-level warning state in sequence according to the preset priority.

[0063] Optionally, the terminal uses a monitoring subroutine to compare the environmental data value in the plaintext data with the warning threshold corresponding to the data type of the environmental data in the effective smart contract in real time. If the environmental data value is greater than the warning threshold, the collection point is recorded as an over-threshold point. The monitoring subroutine continuously slides along the collection time axis with a preset time length as the window width and a preset sliding step size, counting the number of collection points marked as over-threshold points within the current sliding window. If the number of over-threshold points is greater than or equal to a preset number threshold N, where N is the number of collections within a preset duration threshold in the warning threshold, then it is determined that the plaintext data in the sliding window is greater than the warning threshold in the effective smart contract, and an over-threshold event is generated. In response to the generation of an event exceeding the threshold, a multi-level warning state is triggered sequentially in order of severity from low to high: First-level warning state is triggered by sending a first-preset notification to a first-preset notification recipient. The first-level warning notification includes at least the contract address corresponding to the event exceeding the threshold, environmental data values, and a collection timestamp. If the event exceeding the threshold is not resolved within a first-preset time period after the first-level warning state is triggered, a second-level warning state is automatically triggered by sending a second-preset notification to a second-preset notification recipient, carrying complete temperature change trend data since the event exceeding the threshold has been triggered. If the event exceeding the threshold is not resolved within a second-preset time period after the second-level warning state is triggered, a third-level warning state is automatically triggered by sending a third-preset notification to a third-preset notification recipient, carrying a complete evidence dataset since the event exceeding the threshold has been triggered. The complete evidence dataset includes at least the temperature change time series, GPS trajectory data, and a set of transaction hash values ​​of the referenced blockchain evidence records.

[0064] Optionally, the condition for releasing the over-threshold event is: within the sliding window, the environmental data values ​​of M consecutive collection points are all lower than the warning threshold, where M is the preset number of release confirmation points.

[0065] Step S406: Record the triggering action, escalation action, and cancellation action for each level of warning status, as well as the timestamp corresponding to the action, and generate warning information.

[0066] The warning information refers to the on-chain evidence data unit containing the entire lifecycle state change record of the warning; the warning information includes at least: the unique identifier of the warning event, the contract address that triggered the warning, the transaction hash value of the blockchain evidence record corresponding to the warning, the warning trigger timestamp, the warning upgrade timestamp, the warning cancellation timestamp, the environmental data value corresponding to each action, and the level identifier of the current warning status.

[0067] Optionally, in response to the triggering action of the first-level warning state, the smart contract creates a warning record entry in the contract state and records the following fields: unique identifier of the warning event, contract address, transaction hash value of the first blockchain notarized record that triggered the warning, trigger timestamp corresponding to the triggering action, environmental data value corresponding to the triggering time, and current warning level identifier.

[0068] In response to an upgrade action in the second or third level of alert status, the upgrade timestamp, the upgraded alert level identifier, and the corresponding environmental data value at the time of upgrade are added to the alert record entry. The log recording function in the effective smart contract is called every time an upgrade action occurs to record the upgrade action information on the blockchain.

[0069] In response to the threshold event cancellation action, the cancellation timestamp, the corresponding environmental data value at the time of cancellation, and the warning level identifier are recorded in the warning record entry and updated to the cancelled state. The condition for the cancellation action is that the monitoring subroutine detects that the environmental data values ​​of M consecutive collection points are all lower than the warning threshold.

[0070] The triggering actions, all upgrade actions, and deactivation actions recorded in the warning log are arranged in chronological order, assembled to generate warning information, and then signed with the private key of the effective smart contract account before being stored on the blockchain.

[0071] In this embodiment, the triggering, escalation, and cancellation actions of each level of warning are timestamped and stored on the blockchain, forming an immutable record of the complete lifecycle of the warning event. This provides accurate and traceable event timeline evidence for subsequent responsibility determination, improving the data credibility and processing efficiency of the entire warning-to-responsibility determination chain.

[0072] In one exemplary embodiment, such as Figure 5 As shown, a smart contract is signed based on multimodal biometrics to obtain an effective smart contract, including steps S502 to S506. Wherein: Step S502: Generate biometric anchor points based on multimodal biometric features.

[0073] Optionally, the camera captures the original facial images of both parties involved in the resource interaction, and the fingerprint sensor captures the fingerprint feature point data of both parties. The camera sends the original facial images to the terminal, and the fingerprint sensor sends the feature point data to the terminal. The terminal performs preprocessing on the original facial images, including illumination normalization, pose correction, occlusion detection and removal, to obtain a standardized portrait. A standardized portrait feature vector is extracted from the standardized portrait. The terminal fuses the standardized portrait feature vector and the fingerprint feature point data at the feature layer to generate unique and uncopyable multimodal biometrics for each party involved in the resource interaction. The terminal binds the multimodal biometrics one-to-one with the identification information of the parties involved in the resource interaction and performs a hash operation on the bound data to generate a biometric identity anchor.

[0074] Step S504: Generate on-chain digital identity keys for both parties involved in the resource interaction based on biometric identity anchors.

[0075] Among them, the on-chain digital identity key refers to the asymmetric key pair used for digital signing and identity verification of transactions in the blockchain network, including the on-chain digital identity private key and the on-chain digital identity public key; the private key is used to sign transactions to prove identity, the public key is used to verify the transactions signed by the private key, and the public key, after being hashed, serves as the identity identifier address of the resource interaction party on the blockchain.

[0076] Optionally, generating on-chain digital identity keys for both parties in resource interaction based on biometric identity anchors specifically includes: using the biometric identity anchor as a seed input, combining the terminal's unique hardware identifier and the user-defined PIN code, and generating an on-chain digital identity private key through a key derivation function after a preset number of iterations; generating a corresponding on-chain digital identity public key based on the on-chain digital identity private key using elliptic curve multiplication; storing the on-chain digital identity private key in the secure area of ​​a trusted hardware module and never exporting it; and uploading the on-chain digital identity public key to the blockchain network as the on-chain identity identifier address of the resource interaction parties.

[0077] Step S506: Electronically sign the preset smart contract based on the keys of both parties involved in the resource interaction to obtain an effective smart contract.

[0078] Among them, a pre-set smart contract refers to executable code that is pre-deployed on the blockchain network and includes trade terms, warning threshold configuration, liability determination rules and credit parameter configuration; an electronic signature refers to the signature data generated by the resource interaction parties after performing digital signature calculation on the hash value of the contract terms of the pre-set smart contract using their respective on-chain digital identity private keys; and an effective smart contract refers to a smart contract instance whose contract status has changed to effective after all resource interaction parties have verified the signatures.

[0079] Optionally, the terminal obtains the hash value of the contract terms of a preset smart contract. The hash value is obtained by hashing the bytecode and configuration parameters of the preset smart contract. The first resource interaction party uses its on-chain digital identity private key to digitally sign the hash value of the contract terms, generating a first electronic signature. Using the first electronic signature and the first resource interaction party's on-chain digital identity public key as input, the terminal calls the signing function in the preset smart contract. The signing function executes the following verification logic: it verifies whether the first electronic signature matches the hash value of the contract terms based on the first on-chain digital identity public key; if the verification passes, it marks the signing status of the first resource interaction party as signed in the contract state of the preset smart contract and records the signing timestamp. The second resource interaction party repeats the above electronic signature and verification operations. When the signing status of all signatories of the preset smart contract is "signed", the contract status of the preset smart contract is automatically changed to "effective", and a contract effectiveness record is generated. The contract effectiveness record includes at least the smart contract address, the biometric identity anchor of all signatories, the on-chain digital identity public key of all signatories, the hash value of the contract terms, and the bank account information of all signatories.

[0080] In this embodiment, a single trusted anchor point is used to connect identity authentication, device binding, data storage, credit assessment, and fund settlement. This solves the technical defects of inconsistent data identification and lack of automatic association in the various stages of identity authentication, data collection, credit assessment, and fund payment in traditional solutions, and provides a unified cryptographic trust foundation for the entire cross-border trade data processing process.

[0081] In one exemplary embodiment, the verification process, such as Figure 6 As shown, before generating a trusted data packet based on the original environment data and contract address in the trusted execution environment, steps S602 to S606 are also included. Wherein: Step S602: Parse and extract the device identifier and device signature of each environmental acquisition device from the original environmental data.

[0082] Among them, the device identifier is the identity information used to uniquely identify the environmental data acquisition device, including the device serial number or device ID; the device signature refers to the signature data generated by the environmental data acquisition device using its device private key to perform digital signature operation on the collected raw environmental data, which is used to verify whether the raw environmental data does indeed come from the legitimate device corresponding to the device identifier and has not been tampered with during transmission.

[0083] Optionally, the terminal receives raw environmental data packets sent by the environmental acquisition device. The raw environmental data packets are encoded according to a preset data format, which includes a device identification field, a device signature field, and an environmental data field. The terminal parses the raw environmental data packets according to the preset data format, extracts the device identifier from the device identification field, extracts the device signature from the device signature field, and extracts environmental data from the environmental data field. The environmental data includes the environmental data values ​​collected by the environmental acquisition device and the collection timestamp.

[0084] Step S604: Obtain the public key of the corresponding environmental acquisition device based on the device identifier.

[0085] The public key is the public part of the asymmetric encryption key pair corresponding to the device's private key, and is used to verify the digital signature generated by the device's private key.

[0086] Optionally, the terminal queries the local cache to see if a public key corresponding to the device identifier exists based on the device identifier; if a corresponding public key exists in the local cache, the terminal retrieves the public key from the local cache; if a corresponding public key does not exist in the local cache, the terminal queries the blockchain network for the device registration record associated with the effective smart contract based on the device identifier, extracts and retrieves the public key corresponding to the device identifier from the device registration record, and stores the retrieved public key in the local cache.

[0087] Furthermore, the device registration record is an on-chain evidence record generated when the environmental data collection device is bound after the effective smart contract takes effect. The device registration record contains the device identifier, device public key, device type, binding time, and contract address.

[0088] Step S606: Verify the device signature using the public key. If the verification is successful, send the original environment data to the trusted execution environment.

[0089] The Trusted Execution Environment (TEE) is a hardware-secure, isolated environment deployed within the terminal, where external operating systems and applications cannot access data.

[0090] Optionally, the terminal uses the public key corresponding to the device identifier to perform a signature verification operation on the device signature to verify whether the device signature matches the environmental data in the original environmental data packet. If the signature verification passes, it confirms that the original environmental data comes from the legitimate environmental acquisition device corresponding to the device identifier and has not been tampered with during transmission. The original environmental data is then sent to the entry point of the trusted execution environment through a secure channel. If the signature verification fails, the original environmental data packet is discarded and an exception log record is generated.

[0091] In this embodiment, by pre-verifying the device signature of the environmental acquisition device, only data that passes the verification is allowed to enter the TEE, ensuring that all data entering the TEE is from a legitimate source and has not been tampered with, thus avoiding the pollution of the TEE's internal processing flow by forged or tampered data.

[0092] In one exemplary embodiment, such as Figure 7 As shown, in a trusted execution environment, a trusted data packet is generated based on the original environment data and the contract address, including steps S702 to S706. Wherein: Step S702: In the trusted execution environment, the original environment data is processed to standardize the format to obtain environment data in a unified format.

[0093] Format normalization refers to converting heterogeneous raw data from different types of environmental acquisition devices into a unified data format with the same data structure.

[0094] Optionally, the trusted execution environment parses the raw environmental data and identifies the environmental data type, which includes temperature, humidity, and location coordinates. Based on the environmental data type, the trusted execution environment maps the environmental data values ​​in the raw environmental data to the corresponding numerical fields of a unified data structure, maps the collection timestamps in the raw environmental data to the timestamp field of the unified data structure, and maps the device identifier carrying the raw environmental data to the device identifier field of the unified data structure. The contract address is then written into the contract address field of the unified data structure to generate environmental data in a unified format.

[0095] Step S704: Perform symmetric encryption on the unified format environment data to obtain ciphertext data, and calculate the hash value of the unified format environment data.

[0096] Symmetric encryption refers to an encryption method that uses the same key for both encryption and decryption; hash values ​​are data fingerprints that uniquely identify data in a uniform format environment.

[0097] Optionally, the trusted execution environment invokes the built-in hardware encryption engine to symmetrically encrypt the unified format environment data using a symmetric encryption key dynamically generated during the session, thereby obtaining ciphertext data; the trusted execution environment then uses a hash algorithm to calculate the hash value of the unified format environment data before encryption, thereby obtaining the hash value of the unified format environment data.

[0098] Step S706: Associate the encrypted data, hash value, device identifier, and contract address to obtain a trusted data packet.

[0099] The trusted data packet is a data unit containing encrypted data, hash value, device identifier and contract address, which is used to transmit to the blockchain network for on-chain evidence storage.

[0100] Optionally, the trusted execution environment obtains the sequence number of the current data packet and the hash value of the previous data packet; the trusted execution environment associates and assembles the encrypted data, hash value, device identifier, contract address, sequence number of the current data packet and hash value of the previous data packet to obtain a trusted data packet; the trusted execution environment uses its built-in private key to digitally sign the trusted data packet, and the signature is used to verify the authenticity of the source of the trusted data packet at the blockchain evidence storage node.

[0101] In this embodiment, by associating encrypted data, hash value, device identifier and contract address to assemble a trusted data packet, the data stored on the blockchain not only includes the environmental data itself, but also the source device information and the business information to which the data belongs. Any missing information in any dimension can be traced through the associated fields in the data packet, providing a structured data foundation for subsequent retrieval of data from blockchain storage records by device dimension or contract dimension.

[0102] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0103] Based on the same inventive concept, this application also provides an intelligent approval device for implementing the intelligent approval method described above. The solution provided by this device is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more intelligent approval device embodiments provided below can be found in the limitations of the intelligent approval method described above, and will not be repeated here.

[0104] In one exemplary embodiment, such as Figure 8 As shown, an intelligent approval device is provided, including: a contract signing module 801, a binding and data collection module 802, an evidence storage module 803, a generation module 804, and an approval module 805, wherein: The signing module 801 is used to obtain the multimodal biometrics of both parties involved in the resource interaction, and to sign a smart contract based on the multimodal biometrics to obtain an effective smart contract.

[0105] The binding acquisition module 802 is used to bind the environmental acquisition devices associated with the effective smart contract based on the contract address in the effective smart contract, and to obtain the raw environmental data through the environmental acquisition devices.

[0106] The evidence storage module 803 is used to generate a trusted data packet based on the original environment data and contract address in a trusted execution environment, and send the trusted data packet to the evidence storage node of the blockchain network for on-chain evidence storage to obtain a blockchain evidence storage record.

[0107] The generation module 804 is used to generate credit ratings based on blockchain-stored records, early warning thresholds in effective smart contracts, and multimodal biometrics.

[0108] Approval module 805 is used to obtain approval results based on credit rating.

[0109] In an exemplary embodiment, the generation module 804 is further configured to generate warning information based on blockchain evidence records and warning thresholds in effective smart contracts; generate on-chain warning log records based on the warning information; and generate a credit rating based on the on-chain warning log records and multimodal biometrics.

[0110] In an exemplary embodiment, the generation module 804 is further configured to determine the determination result of the responsible party based on the on-chain warning log records and the responsibility determination rules in the effective smart contract; generate a responsibility determination report based on the determination result of the responsible party and the blockchain evidence record; obtain corresponding historical behavior data from the blockchain based on multimodal biometrics; and generate a credit rating based on the responsibility determination report and the historical behavior data.

[0111] In an exemplary embodiment, the generation module 804 is further configured to decrypt and restore the blockchain evidence record to plaintext data in a trusted execution environment; if the plaintext data in the sliding window is greater than the warning threshold in the effective smart contract, trigger multiple warning states in sequence according to preset priority; record the triggering action, escalation action and deactivation action of each warning state, as well as the timestamp corresponding to the action, and generate warning information.

[0112] In an exemplary embodiment, the signing module 801 is further configured to generate a biometric identity anchor based on multimodal biometric features; generate on-chain digital identity keys for both parties involved in the resource interaction based on the biometric identity anchor; and electronically sign a preset smart contract based on the keys of both parties involved in the resource interaction to obtain an effective smart contract.

[0113] In one exemplary embodiment, an intelligent approval device further includes: a verification module, configured to parse and extract device identifiers and device signatures of each environmental acquisition device from the raw environmental data; obtain the public key of the corresponding environmental acquisition device based on the device identifier; verify the device signature using the public key; and, if the verification is successful, send the raw environmental data to a trusted execution environment.

[0114] In an exemplary embodiment, the evidence storage module 803 is further configured to perform format normalization processing on the original environment data in the trusted execution environment to obtain environment data in a unified format; perform symmetric encryption on the environment data in the unified format to obtain ciphertext data; calculate the hash value of the environment data in the unified format; and associate the ciphertext data, hash value, device identifier and contract address to obtain a trusted data packet.

[0115] Each module in the aforementioned intelligent approval device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0116] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 9 As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores associated data between resource interaction parties. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements an intelligent approval method.

[0117] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0118] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0119] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0120] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0121] It should be noted that the user information (including but not limited to multimodal biometrics) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0122] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0123] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0124] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An intelligent approval method, characterized in that, The method includes: Obtain the multimodal biometrics of both parties involved in the resource interaction, and sign a smart contract based on the multimodal biometrics to obtain an effective smart contract; Based on the contract address in the effective smart contract, bind each environmental acquisition device associated with the effective smart contract, and obtain raw environmental data through each environmental acquisition device; In the trusted execution environment, based on the original environment data and contract address, a trusted data packet is generated and sent to the evidence storage node of the blockchain network for on-chain evidence storage, thereby obtaining a blockchain evidence storage record. A credit rating is generated based on the blockchain evidence records, the early warning thresholds in the effective smart contracts, and the multimodal biometrics. Based on the credit rating, the approval result is obtained.

2. The method according to claim 1, characterized in that, The process of generating a credit rating based on the blockchain-stored records, the early warning threshold in the effective smart contract, and the multimodal biometrics includes: Based on the blockchain evidence records and the warning thresholds in the effective smart contract, a warning message is generated; Based on the aforementioned warning information, an on-chain warning log record is generated; A credit rating is generated based on the on-chain early warning log records and the multimodal biometrics.

3. The method according to claim 2, characterized in that, The process of generating a credit rating based on the on-chain early warning log records and the multimodal biometrics includes: Based on the on-chain warning log records and the liability determination rules in the effective smart contract, the determination result of the responsible party is determined; Based on the determination of the responsible party and the blockchain evidence record, a responsibility determination report is generated; Based on the aforementioned multimodal biometrics, corresponding historical behavioral data is obtained from the blockchain. The credit rating is generated based on the liability determination report and the historical behavior data.

4. The method according to claim 2, characterized in that, The generation of warning information based on the blockchain-stored evidence records and the warning thresholds in the effective smart contract includes: In the trusted execution environment, the blockchain evidence record is decrypted and restored to plaintext data; If the plaintext data in the sliding window exceeds the warning threshold in the effective smart contract, multiple warning states will be triggered sequentially according to preset priority. Record the triggering action, escalation action, and cancellation action for each level of warning status, as well as the timestamp corresponding to the action, to generate the warning information.

5. The method according to claim 1, characterized in that, The smart contract signed based on multimodal biometrics, resulting in an effective smart contract, includes: Based on the aforementioned multimodal biometrics, biometric identity anchors are generated; Generate on-chain digital identity keys for both parties involved in the resource interaction based on the biometric identity anchor. The smart contract is electronically signed by the keys of both parties involved in the resource exchange, resulting in an effective smart contract.

6. The method according to claim 1, characterized in that, Before generating the trusted data packet based on the original environment data and contract address in the trusted execution environment, the method further includes: The device identifier and device signature of each environmental acquisition device are parsed and extracted from the original environmental data. Obtain the public key of the corresponding environmental data acquisition device based on the device identifier; The device signature is verified using the public key, and if the verification is successful, the original environment data is sent to the trusted execution environment. The process of generating a trusted data packet based on the original environment data and contract address in a trusted execution environment includes: The original environment data is processed for format normalization in the trusted execution environment to obtain environment data with a unified format. Symmetric encryption is applied to the environmental data in the unified format to obtain ciphertext data; Calculate the hash value of the environmental data in the unified format; The encrypted data, hash value, device identifier, and contract address are associated to obtain the trusted data packet.

7. An intelligent approval device, characterized in that, The device includes: The signing module is used to obtain the multimodal biometrics of both parties involved in the resource interaction, and to sign a smart contract based on the multimodal biometrics to obtain an effective smart contract; The binding acquisition module is used to bind each environmental acquisition device associated with the effective smart contract based on the contract address in the effective smart contract, and to obtain raw environmental data through each of the environmental acquisition devices; The evidence storage module is used to generate a trusted data packet based on the original environment data and contract address in a trusted execution environment, and send the trusted data packet to the evidence storage node of the blockchain network for on-chain evidence storage to obtain a blockchain evidence storage record. The generation module is used to generate a credit rating based on the blockchain evidence records, the early warning threshold in the effective smart contract, and the multimodal biometrics. The approval module is used to obtain the approval result based on the credit rating.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.