21results about How to "Improve verification accuracy" patented technology

The invention relates to a multi-client collaborative SAR image target identification method based on federated learning, and the method comprises the steps: collecting SAR image data, carrying out the preprocessing, generating a corresponding target identification tag, and obtaining a data set with a tag; building a federal learning framework, initializing and distributing global model parameters to each client; distributing local private data to each client according to the data set with the label, performing local model training by each client by using the local private data, and uploading updated model parameters to a server; the server aggregates the model parameters uploaded by all the clients, updates global model parameters, and distributes the updated global model parameters to all the clients; repeating the steps until the global model converges; and testing the trained global model, and selecting the tested global model to carry out SAR image target identification. Compared with the prior art, the method has the advantages that the accuracy of the model is effectively maintained while privacy is protected.

This invention belongs to the field of human-computer interaction and discloses a digital human intelligent wake-up interaction device and method that integrates voiceprint recognition and facial recognition. The method includes: responding to a wake-up trigger signal by simultaneously acquiring a user's video stream and audio stream; extracting the temporal trajectory of facial key points during the user's speech from the video stream and encoding the temporal trajectory of the facial key points into a dynamic behavior feature vector; extracting the user's static facial structure feature vector from the video stream and extracting the user's voiceprint feature vector from the audio stream; inputting the dynamic behavior feature vector, the static facial structure feature vector, and the voiceprint feature vector into a pre-trained multimodal fusion model to calculate a comprehensive matching degree; and generating a wake-up command and outputting it to the digital human interaction module when the comprehensive matching degree is greater than or equal to a preset threshold. This invention significantly improves the robustness and reliability of identity verification.

The utility model relates to the technical field of dairy product quality detection, and provides an automatic detection system and an intelligent pH value detection workstation. The automatic detection system comprises a preliminary identification mechanism, an identification rechecking mechanism and a detection mechanism; in the conveying direction of the dairy products to be detected, the preliminary recognition mechanism is located at the upstream of the recognition and rechecking mechanism, and the detection mechanism is located at the downstream of the recognition and rechecking mechanism; the preliminary recognition mechanism is set to obtain first information of a dairy product to be detected; the identification rechecking mechanism is set to obtain second information of the dairy product to be detected and check the correctness of the first information based on the second information; and the detection mechanism is used for detecting the quality of the dairy products to be detected after the verification of the identification and rechecking mechanism is passed. The automatic detection system can effectively solve the problems that in the prior art, dairy product detection only depends on manual work, the detection efficiency is low, errors are prone to occurring and the like, and the dairy product detection efficiency and accuracy are improved.

This invention discloses a security verification method, apparatus, and electronic device. The method includes: responding to a user-initiated verification request by randomly generating an action sequence containing multiple action descriptions; extracting target action samples that match the action descriptions from a pre-built action sample library, and constructing an action sample set based on the extracted target action samples; displaying the target action samples from the action sample set to the user, and obtaining the action trajectory generated by the user performing the verification action under the guidance of the target action samples; detecting the action trajectory based on the action sequence, and confirming that the user has passed the security verification when the action trajectory matches the action descriptions in the action sequence. In this way, by guiding the user to perform the verification action through action samples, there is no need to touch the screen, which avoids verification failure due to accidental touches and improves the accuracy and convenience of verification.

The invention relates to the field of work ticket supervision and inspection, in particular to an intelligent power work ticket supervision and inspection system fusing multi-modal data, and the system comprises a data analysis unit which is used for determining whether to carry out the high-precision optimization of a reference verification mode or not based on a split condition; the parameter optimization unit is used for determining whether to adjust the path enumeration depth or not based on the regional topology complexity corresponding to the regional condition and the dynamic load reference degree; a processing optimization unit to determine whether to perform path order optimization based on the redundancy condition; and the task analysis unit is used for determining and adjusting cache updating duration or executing task circulation based on the range conflict value corresponding to the processing module, and determining a circulation module based on the range discrete value during task circulation execution. According to the invention, the supervision efficiency of the electric power work ticket is improved.

The present application relates to the field of work ticket supervision, and particularly relates to a power work ticket intelligent supervision system fusing multi-modal data, comprising: a data analysis unit, used to determine whether to perform high-precision optimization for a benchmark verification mode based on a split condition; a parameter optimization unit, used to determine whether to adjust path enumeration depth based on a regional topology complexity corresponding to a regional condition and a dynamic load reference degree; a processing optimization unit, used to determine whether to perform path order optimization based on a redundancy condition; and a task analysis unit, used to determine whether to adjust a cache update duration or perform task flow conversion based on a range conflict value corresponding to a processing module, and determine a flowable module based on a range discrete value in the task flow conversion. The present application improves the supervision efficiency of the power work ticket.

The invention provides a cross-border data verification system and method, and the system comprises a data main body, an overseas receiver, and a domestic agent provided with a domestic trusted third party. And the data subject sends the first data security certification file of the target cross-border data and the data subject identity identifier to an overseas receiver. And the overseas receiver performs data encryption processing on the first data security certificate file and the data subject identity identifier to obtain a second data security certificate file and a cross-border service identifier, and sends at least one of the second data security certificate file and the cross-border service identifier to the domestic agent and the data subject. The domestic agent performs data verification on the target cross-border data based on a first data security certification file, a second data security certification file, a data subject identity identifier and a cross-border service identifier stored by the domestic trusted third party to obtain a target verification result, and sends the target verification result to the overseas receiver; therefore, the effect of improving the verification security of the cross-border data is achieved.

The invention provides a nuclear power station industrial steam supply model parameter verification method and a computer device, and the method comprises the steps: completely copying a unit main instrument control picture and logic related to a nuclear power station industrial steam supply power plant side heat exchange system to a simulation platform, communicating with the existing logic of the simulation platform, and generating an instrument control simulation model; supplementing and updating communication interfaces of the newly-added process system simulation model and the newly-added instrument control system simulation model; establishing a full-range operation condition, carrying out parameter static comparison by referring to the corresponding condition of the unit, and analyzing the difference; if the parameters are adjusted according to the design parameter adjustment scheme, the adjusted parameters are updated to the simulation platform process / instrument control simulation system. According to the method, the existing full-range simulation platform of the nuclear power plant is used for building the nuclear power plant industrial steam supply power plant side heat exchange system related model and carrying out parameter verification, so that the problem that the risk of field debugging in an operation unit is relatively large as the related model cannot cover a complete process system and an instrument control system of the power plant is solved.

The application belongs to the technical field of engineering physics verification, and discloses an engineering physics verification system and method based on a unified field theory and a national geographic coordinate system, which comprises a GIS physical base map module, provides unified geographic reference and basic physical field data, a multi-physical field acquisition module, collects physical parameters of engineering components and environment in multiple dimensions to provide real-time input data, a physical verification engine module, internally built with a unified field equation, performs multi-physical field coupling calculation, a core calculation unit, outputs total correction stress, safety factor and compliance score, generates a virtual pre-assembly report, automatically issues the correction parameters to a black light factory for execution, receives execution feedback, a social collaboration module, supports socialized collaboration and data notarization in the whole verification process, and a mobile terminal interaction module, which realizes code scanning positioning, lightweight verification and early warning pushing. The application adopts the above system and method to provide a real-time, automatic and high-precision physical compliance verification tool for construction projects and industrial products.

The disclosure provides a verification method, device, equipment, medium and program product, relating to the field of information security or the field of finance. The verification method comprises: in response to an operation of a first user, determining to-be-verified information, wherein the to-be-verified information comprises information of M characters, and first verification data of N sensors is collected within a first time period during which the first user moves an electronic device according to the information of M characters; the first verification data is classified by using a pre-trained verification model, and a classification result is output; and in a case where the classification result represents that a trajectory of the first user moving the electronic device matches the M characters, verification is passed. The disclosure also provides a model training method, device, equipment, medium and program product.

The invention relates to the technical field of intelligent financing, in particular to a general financial service platform and method based on big data, artificial intelligence and RPA, and the platform comprises a message push system, a data receiving and collection module, a data storage module, a pre-loan admission credit approval module, and a post-loan management module. The message pushing system is used for issuing credit product information and carrying out customer drainage and data interaction; the data receiving and collecting module is used for receiving a credit request of a customer, automatically inputting information of the customer and establishing a customer label; the data storage module is used for storing data information; the before-loan admission credit granting approval module is used for verifying credit materials provided by the customer, judging whether the customer meets a credit admission condition or not, and then performing risk assessment and limit calculation to complete customer credit granting approval; and the post-loan management module is used for performing post-loan management, and performing customer operation data tracking, customer interest settlement and repayment pre-reminding.

The invention discloses a power distribution network verification method, system and device based on multi-path intelligent switching and a storage medium, and relates to the technical field of power distribution network verification, and the method comprises the steps: building and initializing a verification platform, and configuring related rules. And continuously monitoring the operation of the power distribution network, judging switching conditions and evaluating indexes. And selecting an optimal path for switching after the conditions are met. And a test case is injected after switching, so that protocol compatibility is ensured. And monitoring a test response, and recording data. And after the test is completed, switching back equipment and performing closed-loop control. Analyzing data, diagnosing faults, monitoring own indexes, evaluating coverage rate, and adaptively optimizing internal parameters according to a result; according to the method, the switching efficiency of the power distribution network can be greatly improved, the equipment compatibility is enhanced, the verification precision is improved, the fault recovery time is shortened, the operation complexity is reduced, and reliable operation of the power distribution network is guaranteed.

The present application discloses an intelligent tongue image data acquisition system and method, which includes: an image sensor is provided in the intelligent terminal, which is used to obtain the user's face image and send it to the server; the server verifies the current user according to the face image and the user's basic information, and sends a tongue image acquisition instruction to the intelligent terminal to determine the tongue position in the face image, calculate the histogram features according to the pixel information corresponding to the tongue position, generate the user feature vector, and determine the color correction matrix according to the Euclidean vector distance between the feature vector and the sample feature vector; the intelligent terminal collects the user's tongue image according to the tongue image acquisition instruction, and sends the tongue image to the server; the server performs color correction on the tongue image according to the color correction matrix, and stores the corrected tongue image. Through the technical solution in the present application, the problem that the existing tongue diagnosis instrument is expensive and not convenient and intelligent enough is solved.

The invention discloses an electric power operation checking method and device and a storage medium. The method comprises the steps that operation associated data of multiple electric power operations in an electric power operation list in a preset time period are acquired, and the operation associated data comprise data generated by executing each electric power operation, actual operation starting time data, actual operation ending time data, operation plan starting time data and operation plan ending time data; for each power operation, determining standard operation data corresponding to the operation associated data of the power operation, and checking the operation associated data based on the standard operation data to obtain an initial checking result corresponding to the power operation; and determining a target check result corresponding to the power operation list based on the plurality of initial check results, generating prompt information based on the target check result, and displaying the prompt information. The checking accuracy of the electric power operation is improved, and the safety and normalization of the electric power operation are improved.

The application relates to an application service deployment method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring a plurality of service messages, a blue-green application stack and a gray application stack, and dividing the service messages into a plurality of message partitions; identifying asynchronous service consumption and a target message partition of pulled target messages in each service message of each message partition; performing consumption verification processing on each synchronous service message in the target message partition based on the blue-green application stack to obtain a first verification result of each synchronous service message, and performing consumption verification processing on each asynchronous service message in the target message partition based on the gray application stack to obtain a second verification result of each asynchronous service message; and deploying consumption code logic of each service message to complete an application service deployment task in the case that the first verification result and the second verification result both satisfy system stable operation conditions. The method can improve the deployment stability effect of the system in asynchronous consumption logic change deployment.

The invention discloses a transmission line key point identification method and a hidden danger treatment method, and the method comprises the steps: based on the data of rivers, roads, bridges and administrative regions in a GIS map and transmission line resource data, employing an Oracle Spatial space component function of an Oracle database to identify cross-road, cross-river / bridge-hanging, along-river and outgoing route key points; and then, marking a service reloading type transmission line so as to grade and file the key points of the transmission line. And filing, scheme making, on-site renovation and renovation auditing are further carried out, a closed-loop management and control process is established, and key point on-site hidden danger renovation work is effectively implemented. The whole key point filing process and the hidden danger renovation process are IT online, automatic and intelligent, transmission lines of the whole province are covered by 100%, and the key point troubleshooting efficiency is at least improved by 35%. According to the method, the checking work efficiency of the key points of the transmission line is improved, the checking accuracy of the key points of the transmission line is improved, the output result can be checked anytime and anywhere, the resource accuracy of the key points of the transmission line is improved, the number of group faults is reduced, the security of a transmission network is improved, and good practicability is achieved.

The invention relates to a multi-client collaborative SAR image target recognition method based on a semi-decentralized architecture, and the method comprises the steps: collecting SAR image data, carrying out the preprocessing, generating a corresponding target recognition tag, and obtaining a data set with a tag; building a semi-decentralized federated learning framework, and initializing global model parameters; each client performs model training by using local private data, and obtains a local parameter consensus through communication between devices; the server randomly selects the clients for parameter transmission, aggregates model parameters, updates global model parameters, and broadcasts the updated global model parameters to all the clients; iterative training is carried out until global model parameters reach a convergence state, and a trained global model is obtained; and testing the trained global model, and selecting the tested global model to carry out SAR image target identification. Compared with the prior art, on the premise of ensuring model performance and data privacy protection, the communication overhead between the server and the client can be effectively reduced.

The invention relates to the technical field of remote sensing imaging, in particular to a high-temperature optical window medium-wave infrared imaging performance detection device and detection method.The detection device comprises a target generator, a light adjusting assembly, a rapid temperature impact high-temperature experiment box, a transfer device and a photoelectric load system, and the target generator is used for generating a medium-wave infrared target; the light adjusting assembly is used for guiding light of the medium-wave infrared target to the optical window; the rapid warm-punching high-temperature experiment box is used for unilaterally heating the optical window to a target set temperature; the transfer device is used for transferring the optical window to a light path of the detection device; and the photoelectric load system is used for receiving the light penetrating through the optical window and forming a medium-wave infrared target image. Therefore, the photoelectric load imaging performance change of the optical window caused by the aerodynamic thermal effect in a high Mach number complex flight environment can be accurately simulated in a ground laboratory, and the medium-wave infrared imaging performance of the optical window at the target set temperature can be comprehensively detected.

The invention provides a neural network formal verification method and system based on node importance and adaptive scheduling. The method comprises the following steps: firstly, executing symbol interval propagation on a target neural network, and identifying unstable nodes with uncertain input; then, integrating the layer depth position of the node in the network and the propagation upper bound of the node to output disturbance, and dynamically calculating the importance score of each unstable node; further, according to the current hardware resource state and node importance, a quality and budget two-level constraint decision mechanism is adopted, key nodes are intelligently screened, an input field is divided, and sub-fields are generated; and finally, under real-time resource monitoring, dynamically scheduling each sub-domain verification task to a computing unit for parallel processing, and reducing a result to obtain a final verification conclusion. Through importance-driven resource allocation, the problem of insufficient resource utilization of an existing verification method is solved, and the verification efficiency and precision are remarkably improved.

The invention relates to the field of power systems, and discloses a verification test method for increasing output of a unit and improving emergency guarantee capability of a power grid, which comprises the following steps: initializing a unit simulation environment after loading a unit model; obtaining a plurality of expert strategies by training a reinforcement learning model based on a unit simulation environment; a plurality of expert tracks are sampled through interaction between an expert strategy and a unit simulation environment; constructing a generator and discriminators of which the number is the same as that of the expert strategies; in the training process, a simulator track is sampled through interaction between a generator and a unit simulation environment; respectively training a discriminator with each expert track by utilizing the simulator track, calculating a reward value according to a discriminant value output by each discriminator, and training a generator by adopting a PPO algorithm according to the simulator track and the reward value; and obtaining a verification test model for increasing the output of the unit according to the trained generator. The output of the unit and the emergency response capability of the power grid can be improved, and the operation stability and the performance improvement are ensured.

The application discloses a DOC engine exhaust pollution conversion efficiency verification method and an engine bench, and belongs to the technical field of engines. The DOC engine exhaust pollution conversion efficiency verification method and the engine bench refer to two main factors affecting the DOC efficiency, i.e. the bed temperature of the DOC and the volume flow of the exhaust gas entering the DOC, so that the DOC efficiency verification precision is higher and more comprehensive; not only the conversion efficiency of the DOC on THC can be verified, but also the conversion efficiency of the DOC on CO and NO can be verified, the performance verification of the DOC is more comprehensive, and the DOC selection and bench fine calibration process can be used.