43 results about "Data authenticity" patented technology

The application discloses a data verification system implementation method based on a block chain and zero-knowledge proof. In the data verification stage, the data use end pays the corresponding data verification fee to the on-chain smart contract according to the verification requirement, and the demand satisfaction and data real possession are verified through the interaction with the smart contract; in the data declaration stage, the data owner carries out the Merkle tree on the data and generates the characteristic zero-knowledge proof, and the on-chain verification and evidence storage of the data are completed through the interaction with the smart contract, and in the data verification stage, the data owner responds to the challenge of the data real possession, generates the existence proof of the challenge data block set and carries out the on-chain verification and evidence storage, so that the data verification can be independently executed without relying on the third-party service executable environment or the trusted hardware, the independence and safety of the data verification are ensured, and in the whole data real possession verification process, the data is always saved in the local security environment of the data owner, and the complete non-flow of the data is realized.

This invention discloses a multi-agent collaborative scheduling method for smart grids based on reinforcement learning, relating to the fields of power system automation and network security technology. By configuring a twin detection network for each agent, this invention constructs a data credibility active verification layer deeply embedded in the decision-making loop. This twin detection network utilizes the physical laws of power grid operation as a priori knowledge benchmark. It calculates the structured difference between real-time measurement data and expected data predicted based on a physical model to obtain a credibility score, achieving distributed, online detection of false data injection attacks. This mechanism moves data authenticity verification from the periphery, centralized firewall to the front end and integrates it into the real-time decision-making process of each agent, enabling attacks to be identified and alerted before affecting the generation of scheduling instructions. This reduces the path of dangerous decisions driven by false data from the source, solving the security problem of reinforcement learning models blindly trusting contaminated data due to a lack of input verification.

The utility model provides a kind of hydraulic engineering geological survey sampling device, it is related to hydrogeological engineering survey technical field, including cylinder, by coaxial setting upper sampling cavity, middle sampling cavity and lower sampling cavity constitute three-dimensional sampling space;Negative pressure source, it is communicated to each sampling cavity, directional negative pressure is formed when sampling.The utility model can realize high-precision stratified sampling without cross-contamination, by orthogonal distribution upper, middle and lower water inlet channel, make the water body of different depth in geological survey stationary surface water along independent path into corresponding sampling cavity, completely avoid interlayer mixing from physical structure, ensure stratified data authenticity;The utility model can optimize fluid flow, guarantee sample integrity and representativeness, the horn-shaped water inlet of anti-turbulence cover reduces flow velocity, combined with spiral guide flow channel, convert turbulent flow into stable laminar flow, reduce bubble disturbance and particulate matter settlement, significantly improve the originality of sample, meet high-precision water quality analysis demand.

The application discloses a whole-genome DNA cytosine hydroxymethylation modification single-base precision library construction method, relates to the technical field of genomics, epigenetics and molecular biology, and contains the following steps: 5hmC labeling reaction, click chemistry reaction, sample DNA fragmentation, purification of the fragmentation product through magnetic beads, 5hmC fragmented DNA capture, washing of the capture product, reduction reaction, APOBEC enzyme deamination reaction, purification of the deamination product, double-strand conversion reaction, end repair and 3' end A addition reaction, linker ligation, purification of the ligation product and double sorting, library amplification and library purification. The library construction method integrates low data volume, single-base precision and conventional double-strand library construction kit in one under the premise of ensuring data authenticity, and has the technical advantages of high universality and highly repeatable data.

The application relates to a multimedia type-based communication method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: querying each media type and a data transmission sequence between each media type according to user information and service information; finding a resource transmission threshold of each media type in a communication process according to each media type and the data transmission sequence; and controlling data of each media type to communicate through the resource transmission threshold of each media type. The method can enable terminal to synchronously display data of each media type, and improve the authenticity of data received by the terminal. In addition, each service end does not need to be greatly adjusted based on the data transmission sequence, which helps to improve the service deployment range and efficiency.

The present disclosure discloses a whole-process traceability system for aquatic products based on blockchain technology, wherein: The data acquisition layer is composed of environmental monitoring unit, feeding monitoring unit and RFID identification unit, which is used to collect aquaculture environmental data; the edge computing layer includes an edge server and an industrial control computer. The edge server and the industrial control computer preprocess the collected data; the blockchain layer is built on the Hyperledger Fabric platform to realize data tamper-proof storage, and the data authenticity is automatically verified through smart contracts. The application layer includes the aquaculture management subsystem, processing management subsystem, logistics management subsystem, and sales management subsystem, and adopts a multi-level authorization mechanism to manage data access rights. This method solves problems of data fraud, high traceability cost, and opaque information, and realizes the whole chain credible traceability of aquatic product quality and safety.

This invention provides a blockchain-based method, system, device, and medium for managing vehicle network vulnerabilities, relating to the field of vehicle network technology. The method utilizes a dynamic and static vulnerability detection mechanism to comprehensively identify potential vehicle security risks and generate structured vulnerability reports. By using blockchain to hash and store key data from the report and subsequent remediation processes, the entire process of vulnerability discovery, analysis, and remediation is tamper-proof and traceable, enhancing data authenticity and trustworthiness. Cloud-based verification of report integrity, combined with an AI analysis engine linking CVE databases and threat intelligence, enables intelligent generation and decision support for remediation solutions. On-vehicle verification of OTA patches via on-chain hash digests ensures the credibility of patch sources and the integrity of content. Finally, feedback on installation results and re-archiving on the blockchain completes a closed-loop management system from vulnerability discovery to remediation verification.

This invention discloses a system and method for reviewing and analyzing motion visual data based on ball-hitting triggers, relating to the field of motion multimodal data synchronization and intelligent analysis technology. The system includes a racket trigger module, a head-mounted visual acquisition module, and a data processing terminal, which work together to form an inseparable core architecture of "ball-hitting trigger - data acquisition - synchronized review and analysis." Specifically, the head-mounted visual acquisition module acquires the athlete's first-person perspective scene in real time, annotates it with a high-precision timestamp, and streams it to the data processing terminal in real time, without setting up a large local cache to reduce device load; the racket trigger module acquires ball-hitting related inertial navigation data, completing ball-hitting detection and timestamp generation and uploading through two independent and non-conflicting methods; the data processing terminal caches the received data, triggers deduplication and time-series alignment, and the time-series alignment adopts the nearest original frame matching strategy to preserve the authenticity of the original data and avoid data distortion. This invention enables high-precision synchronization of hitting action and visual data, supports both automated analysis by the host computer and manual review / autonomous analysis by the user, and features rapid data transfer and stable storage capabilities. It solves the technical problems of multi-source data timing misalignment, chaotic trigger signals, high load on head-mounted devices, and poor adaptability in existing technologies. It can be flexibly adapted to various racket sports such as table tennis, badminton, and tennis, providing data support for professional training in racket sports, while reserving space for parameter adjustment in subsequent research and development.

This invention discloses a blockchain-based system for certifying educational and training personnel qualifications and storing test scores. The core of the system is a consortium blockchain network containing educational supervision nodes, employing an improved and practical Byzantine fault-tolerant consensus algorithm and sharding technology to enhance efficiency. The teacher qualification certification module stores qualification data via IPFS and provides on-chain verification, combined with manual verification to ensure data authenticity. The student score storage module collects answer behavior data in real time, and after anomaly filtering and manual review, generates verifiable digital credentials. Merkel tree roots are anchored to the Bitcoin network to strengthen credibility. Supporting modules include dynamic trusted score calculation, multi-layered anti-cheating verification, and Polkadot cross-chain interoperability, enabling trusted scoring, cheating detection, and cross-domain credit recognition. It can also be extended to include metaverse teaching verification functions. This system, through technological and human collaboration, solves problems such as falsified educational and training qualifications, unreliable scores, and difficulties in cross-domain recognition, enhancing the effectiveness of educational supervision and adapting to diverse teaching scenarios.

The application is specifically a network security identification method and system based on multi-modal fusion features, relates to the technical field of network security, and comprises: a flow reorganization and sequence extraction module, a multi-dimensional entropy feature calculation module, a phase space reconstruction and trajectory analysis module, and a topological geometry judgment module.In the application, the adjustable parameter Renyi entropy can accurately capture the content confusion degree difference of different types of encrypted traffic, the box dimension fractal dimension quantifies the self-similarity of the traffic, and the maximum Lyapunov index is combined to realize accurate mining of time sequence chaotic characteristics; and the flow reorganization and sequence extraction module supports multiple network access modes, adapts to different bandwidth scenes, and rearranges out-of-order packets to guarantee data authenticity.

This invention discloses a method for sharing regional rain gauge data in a smart stormwater and sewage drainage system, relating to the field of smart stormwater and sewage drainage monitoring technology. It includes steps such as multimodal data acquisition, protocol normalization processing, multi-level data cleaning, rainfall feature construction, and data publishing and sharing, achieving reliable sharing of regional rain gauge data across the entire area. This invention is compatible with multiple types of rain gauges, unifies data standards, and eliminates heterogeneity and timing deviations; it removes anomalies and identifies faults through three levels of cleaning to ensure data authenticity and validity; it uses dynamic weighted fusion to generate three levels of standard rainfall values ​​at the station, sub-region, and system levels, and constructs high-order rainfall features to adapt to drainage scheduling needs; relying on a low-latency shared bus for hierarchical data opening, coupled with a fault self-healing mechanism, it provides high-precision, high-consistency, and high-real-time data support for stormwater and sewage network scheduling, pump station linkage, flood warning, and interception and pollution control, significantly improving the collaborative scheduling and flood prevention capabilities of the smart drainage system.

This invention relates to the intersection of carbon emission data management and blockchain technology, specifically a dynamic carbon data ownership confirmation method and system based on a carbon-locked dual-chain Merkle twin. It includes: constructing a dual-chain collaborative architecture of a public blockchain main chain and industry consortium chain side chains, achieving cross-chain interoperability through dual Merkle tree cross-anchoring. The main chain uses an MPT Merkle tree to store the side chain root hash and ownership confirmation information, reconstructing only changed nodes. The side chains deploy ordinary Merkle trees according to industry to store all carbon data. After outlier removal and clustering preprocessing, the carbon data generates a side chain root hash, which is then signed by consensus nodes and atomically cross-chain anchored through a hash time-lock contract. Ownership confirmation records implement access control based on DID, and a version chain is formed through the preceding hash value to support historical traceability. Verification nodes complete the verification of data authenticity and ownership legality through dual Merkle tree cross-anchoring, significantly improving evidence storage efficiency and solving the problems of missing cross-chain interoperability and delayed ownership confirmation.

Techniques are disclosed relating to determining whether input data is authentic. A system detects input data, that includes text data and typing data, at a computing device. The system may generate, using a string model, a string-level prediction for the input data, where the string model is trained to increase a similarity between embeddings of authentic text data and corresponding sequences of typing data. Using a character model, the system may generate a character-level prediction for the set of input data, where the character-level model predicts an intended sequence of characters based on the text data and a sequence of typing actions included in the input data. Using machine learning, the system determines, based on the string-level prediction and the character-level prediction, whether the input data is authentic input. The system transmits, to the device, a decision that is generated based on determining whether the input data is authentic.

The application relates to the technical field of functional egg production management, and discloses a blockchain traceability method and system for non-antibiotic production of functional eggs, which records multi-source traceability data of functional eggs from breeding to the whole chain of sales, generates a hash value and a digital identity identifier by using the multi-source traceability data, associates the digital identity identifier with the multi-source traceability data, binds the hash value with the digital identity identifier by using a blockchain, generates a two-dimensional code picture through the digital identity identifier, and traces the source by scanning the two-dimensional code. The multi-source traceability data of functional eggs from breeding to the whole chain of sales is recorded and collected, information is more comprehensive, the hash value and the digital identity identifier are generated by using the multi-source traceability data and are associated, the data can be prevented from being modified by setting the hash value, the data authenticity is improved, the data can be better mastered, and all the information is recorded by using the regional chain, so that information islands are avoided.

The application discloses an artificial intelligence generated video authentication method based on multi-view consistency. The method comprises the following steps: constructing a large-scale real-world simulation video dataset, using the previous frame as a prompt to generate a future frame video with high consistency in semantics, color and physical laws, and improving data authenticity; designing a detection model based on multi-view consistency physical prior, extracting video features through real-time multi-view matching, combining a time sequence memory module to dynamically store and update inter-frame stereo reconstruction features, and using an attention mechanism to enhance long-time consistency analysis; using a multi-modal large model to generate unified prompts and label data, and improving the model generalization ability; scoring each frame feature through a scorer and globally averaging to output a video authenticity judgment result. The application uses multi-view consistency prior and memory enhancement mechanism to effectively capture the subtle defects in AI generated videos that violate the real physical laws, and significantly improves the detection robustness and accuracy of high-quality generated videos.

The present application relates to the field of intelligent agriculture technology, and provides a lightweight intelligent production management system for multiple breeding varieties, which comprises a three-layer lightweight architecture using a hardware acquisition end, an edge operation end and a local interaction end. Each end cooperates to realize local data acquisition, analysis, operation, storage and interaction without cloud dependence. The system is based on the core data flow logic of real-time sharing of environmental monitoring data and precise feeding data linkage, feeding and weighing data and inventory accounting data, and integrally realizes intelligent regulation and control of breeding environment, precise feeding management, full-dimension production data acquisition, inventory digital accounting, equipment state monitoring and remote visualized supervision. By accurately implementing the hash value evidence mechanism into specific breeding business scenarios such as feed weight, weighing at slaughter, and feed into warehouse, and combining with multiple technical means such as device unique identification, timestamp, database encryption storage and write protection, the data is prevented from being tampered with throughout the whole process, and the authenticity and traceability of the data are guaranteed.

The application provides a method for identifying data authenticity based on action time sequence logic of an operator, and relates to the field of on-site detection. The method comprises an action recognition module, an instrument data acquisition module, a data fusion and labeling module and a data anomaly judgment module. During detection, the instrument data acquisition module generates a detection result data set in time sequence; the action recognition module collects wrist and waist action data streams through a wrist type and a waistband type wearable device equipped with an IMU, extracts features and recognizes actions by using an SVM algorithm, and generates a personnel action time sequence data set; after time synchronization calibration, the data fusion and labeling module fuses the two types of data through a time sequence alignment algorithm, and generates labeled data by using AES-128 encryption; after decoding, the data anomaly judgment module compares the time sequence matching degree (threshold value 90%) of the action and the detection data based on a time sequence logic model trained by a random forest, and identifies false data. The application realizes unmanned closed-loop judgment, the recognition accuracy is greater than or equal to 95%, and on-site detection cheating can be effectively eliminated.

The application discloses a kind of based on edge computing's construction engineering management method, system and storage medium, belong to construction engineering digital technology field.For mobile terminal in the service break problem under the environment without network, the application uses cloud edge collaborative architecture, and preposition lightweight model and business rules in smartphone APP, realize online and offline mode seamless switching.Material acceptance whole process controllable is achieved;Realize worker / mechanical operation whole process continuous management: before work, generate semi-structured form;Many nodes generate intermediate semi-structured form record state in the process of operation;After work, all node data are generated to complete structured form and carry out offline electronic signature.Network is recovered, and system automatically incremental synchronization data, and through digital thread, structured form is distributed to supplier, project account book, enterprise account book, cost early warning etc.system by key, realize whole process visualization and real-time early warning.The application realizes the service continuity under the environment without network, data authenticity, process visualization and digital collaboration.

The application discloses an improved method for fault alarm of an electric vehicle braking system, and relates to the technical field of new energy vehicles, comprising the following steps: obtaining an anti-lock braking system fault light signal, an electronic brake force distribution fault light signal and a fault diagnosis code of an electronic stability controller, judging the validity of the fault light signal, and excluding the self-checking state within the power preset time of the electronic stability controller. The application realizes accurate identification of the braking system fault by constructing a double-condition checking mechanism of the fault light signal and the fault diagnosis code, effectively avoids false alarms and improves data authenticity; at the same time, an independent fault grading system is established to realize accurate matching of the fault grade and the braking capacity influence and reduce the risk of missed reports; further, a full-link closed-loop mechanism from fault triggering to remote reporting is constructed to realize standardized fault management without adding new hardware, and the application has good universality and engineering application value.

The present application relates to the technical field of intelligent sports monitoring, and in particular to a running data authenticity evaluation method and system based on biometric recognition, which comprises the following steps: collecting motion data and biometric data during running and generating timestamp-aligned collected data records, filtering and denoising the collected data, removing outliers, completing missing data and standardizing the format, obtaining standardized data, extracting kinematic features and biometric features from the standardized data, constructing a multi-modal feature set, generating a fusion feature vector through a deep learning algorithm, performing time consistency detection, data integrity verification and physiological and motion data rationality analysis based on the fusion feature vector, generating a credibility score, and generating an authenticity verification report based on the credibility score. The present application can effectively improve the credibility of running data and provide accurate risk assessment and recommendations for athletes and data users.

This invention relates to the field of labor education evaluation technology, specifically disclosing a system and method for evaluating the effectiveness of labor education practices. The system includes a data acquisition module, a blockchain evidence storage module, an AI behavior analysis module, a closed-loop control module, and a central database. The data acquisition module collects multi-source data, including IoT check-in data, videos, multi-subject evaluations, and results. The blockchain evidence storage module encrypts and distributes the data using the SHA-256 algorithm. The AI ​​behavior analysis module uses the YOLOv8 model to identify labor behaviors and quantify parameters. The closed-loop control module generates evaluation results and outputs task adjustment and teaching optimization instructions, forming a periodic closed loop. This invention solves the problems of difficulty in ensuring the authenticity of existing labor education evaluation data, lack of process evaluation, and disconnect between evaluation and teaching, achieving full-chain traceability of evaluation, intelligent analysis, and integrated teaching.

The invention provides a data resource processing method, device and system. The method belongs to the technical field of data processing. According to the method, firstly, a to-be-scored data set is constructed according to data resource information from different users, and the construction process of the scored data set does not involve cross-domain circulation of original data, so that the risk of data leakage caused by data out-of-domain is fundamentally avoided. Then calculating the to-be-scored data set based on a privacy calculation algorithm to obtain a privacy calculation result, and calculating to obtain a multi-dimensional scoring factor of the to-be-scored data set based on the privacy calculation result and the data resource information of the target user, thereby covering different dimensions such as data magnitude, data validity, data authenticity, data security and the like; the comprehensive value of the data can be better reflected, and the accuracy of the scoring result is improved. And finally, sending the to-be-scored data set and the corresponding multi-dimensional scoring factor to a server, so that the server calculates the score of the to-be-scored data set according to the multi-dimensional scoring factor.

A method and system for secure recording and reliable storage of UAV bus data are disclosed. The method includes: capturing raw interactive communication data frames of the UAV's internal core bus in real time through a data acquisition interface; parsing and filtering the captured data frames to extract key information and attach timestamps and serial numbers to form standardized data elements; transmitting the standardized data elements to a hardware security module, generating a device-specific key using a true random number generator and storing it in a one-time programmable storage area, generating a hash digest using a hash algorithm, and digitally signing the hash digest using an asymmetric algorithm with the device's private key; encapsulating the standard data elements, hash digest, and digital signature into a secure data credential, and writing it to different non-volatile storage media according to business priority; during data synchronization, encapsulating the credential using a symmetric encryption algorithm and transmitting it to the ground station, where the ground station performs signature verification and hash comparison to verify the data authenticity.

A talent competency verification method and system based on multimodal baseline calibration belongs to the fields of artificial intelligence and human resources technology. It solves the problems of inconsistent evaluation standards, difficulty in ensuring data authenticity, and disconnect between assessment and job requirements in the existing technology. The method includes: loading a feature vector template containing multiple data components; collecting multimodal data through a microphone sensor (1) and a camera sensor (2), and cross-validating the encoding behavior telemetry data and Git commit records collected through the IDE plugin client; extracting personal physiological baseline calibration in the first 30 seconds of the preset period; generating a 7-dimensional combat capability vector and calculating the confidence factor; constructing a multi-dimensional space model of job requirements and performing dimension mapping; calculating the matching degree score and outputting an interpretability assessment report. When the sensor or network is unavailable, switch to plain text mode: if the sensor is unavailable but the network is normal, only the resilience confidence factor decreases when combining text semantics and IDE telemetry data; if the network is interrupted, both the resilience and driving force confidence factors decrease based only on local text semantics. This invention corrects individual differences through physiological baseline calibration and ensures data authenticity through IDE cross-validation, and is used for talent recruitment.

A generation unit creates a second image file containing second image data by performing image editing on a first image file containing first image data. If the first image file contains first authentication assurance information configured to ensure the authenticity of the first image data, a verification unit verifies the authenticity of the first image data based on this first authentication assurance information. If the verification of the first image data's authenticity is successful, a control unit includes second authentication assurance information in the second image file, configured to ensure the authenticity of the second image data.If the verification of the authenticity of the first image data fails, or if the first image file does not contain the first authentication information, the control device refrains from including the second authentication information in the second image file.

The application discloses a kind of forensic botanical evidence packaging transport traceability tracking management systems, the present application relates to traceability information management and judicial expertise field, including: evidence packaging information acquisition module, for collecting the unique identification code of botanical evidence, the unique identification code of seal, the identity information of packaging personnel, packaging time stamp, packaging location positioning information and packaging mode information in packaging site, and the information collected is bound with the unique identification code of evidence, to generate the first re-packaging confirmation record.The judicial expertise botanical evidence packaging transport traceability tracking management system, each link confirmation record is linked as an unalterable traceability chain in generation order by block chain storage, which eliminates the possibility of single-link data modification or bypass verification from the mechanism, and improves the data authenticity and responsibility traceability reliability of the whole process from extraction to reception of judicial expertise botanical evidence.

The application discloses a water body detection sampling device, which comprises a support frame, a sampling cover box assembly located below the support frame, a lifting driving assembly connected with the support frame to provide a stable water environment for sampling, a sampling detection assembly arranged above the support frame and extending into the sampling cover box assembly, a propelling assembly connected with the support frame to provide power for the device to walk underwater, and a moving auxiliary assembly connected with the lower ends of the two sides of the support frame to assist the device to walk on the water bottom. The sampling cover box assembly comprises a cover box, and one mounting frame is connected with each side of the cover box to connect the lifting driving assembly. The water body detection sampling device can realize in-situ real-time detection underwater, and has the advantages of high integration, high efficiency and high data authenticity.

The application belongs to the technical field of chip testing, and particularly relates to a multi-interface compatible USB chip comprehensive testing method, which comprises the following steps: S1, chip ID intelligent identification and historical data block chain matching: obtaining chip parameter information through a USB interface protocol, constructing a digital fingerprint of the chip, and intelligently matching with a distributed chip database based on a block chain to ensure the authenticity and non-tamperability of testing data; introducing a multi-objective genetic algorithm to automatically generate testing parameters, so that the testing efficiency is improved by more than 40%; secondly, a multi-dimensional virtualization parallel testing channel is adopted to ensure comprehensive testing coverage and improve the problem discovery rate by 30%; the block chain is used to ensure data authenticity, AI is used to realize testing parameter self-adaptation, and multi-dimensional parallel testing is used to improve efficiency, so that key problems such as low testing efficiency, incomplete coverage and inaccurate problem positioning are fundamentally solved, and a revolutionary solution is provided for the USB chip testing field.