66 results about "Data aided" patented technology

The invention relates to the technical field of government affair data processing and intelligent decision making, and particularly discloses an index management-based data aided analysis, prediction, research and judgment system, which is characterized in that a multi-level index management system is constructed, an association relationship between indexes is defined, and a dynamic update threshold value is configured; the method comprises the steps of collecting original operation data from multiple government affair data systems, and generating a standardized data set through quality verification and content integration; establishing a multi-index cooperative monitoring network by performing correlation analysis and abnormal feature recognition on the data set; deducing a plurality of development paths in a future time period based on state transition analysis of a historical sequence, and generating a prediction result including trend pre-judgment and risk level; and finally, constructing a decision scheme tree, generating a visual analysis report through multi-dimensional scoring, continuously optimizing parameters based on implementation feedback, and forming a closed loop of analysis prediction and decision support. According to the invention, deep fusion and intelligent research and judgment of cross-domain government affair data are realized.

A compute server of a distributed cloud computing network receives an inference request. The inference request triggers execution of code that is related to an AI application that interacts with the inference request and causes its input to be run through an AI model. Helper code executing on the compute server transmits an inference request to the AI model, where this inference request includes a prompt and one or more function definitions, and where the AI model is executing in the same datacenter as the compute server. The helper code receives a result of the AI model executing the inference request, where it includes structured data for calling a function. The helper code calls the function using the structured data. The helper code receives a response to the called function. The code processes a response to the inference request based at least on the received response to the called function.

Methods, device and data stream are provided to encode unaligned multi-view plus depth images associated with assistance metadata for generating depth maps aligned with the unaligned color views. At the encoding stage, for an unaligned color view, the contribution of each depth map is evaluated, and the objects represented are mapped. Picture patches are cut to be homogeneous in terms of objects and contributing views. Metadata describing this homogeneous cut are encoded together with the atlas. At the decoding stage, to generate a depth map aligned with the unaligned color view, only the depth maps referenced in the assistance metadata are warped.

An assisted driving system includes: inputs configured to receive sensor data, telematics data, and driver response data; an assisted driving module configured to, based on the sensor data and telematics data generate first non-urgent notification messages; and a ranking module configured to rank the first non-urgent notification messages based on the driver response data. The assisted driving module is configured, based on the ranking of the first non-urgent notification messages, to output via at least one output device the first non-urgent notification messages such that saliency of each of the first non-urgent notification messages is based on the respective ranking of that non-urgent notification message.

The application belongs to the field of OFDM channel estimation, and relates to a data-assisted short-wave OFDM sparse channel Turbo iteration estimation method, which comprises the following steps: S1, receiving a signal transmitted through a channel and processing the signal to obtain a receiving end symbol matrix Y; S2, performing channel estimation and equalization on Y to obtain an estimated channel and an OFDM signal pair, and performing Turbo iteration on the estimated channel and the OFDM signal pair to obtain a symbol variance; S3, selecting a virtual pilot in the signal according to the symbol variance; S4, performing channel estimation and equalization on Y according to the virtual pilot to obtain a current estimated channel and an OFDM signal; S5, performing Turbo iteration on the signal to obtain a current symbol variance; and S6, if a preset maximum iteration number is reached, obtaining a final estimated channel; otherwise, returning to step S3 to select a virtual pilot in the signal according to the symbol variance; and the application selects a virtual pilot in the signal according to the symbol variance, and uses the virtual pilot to perform estimation again, so that the spectral utilization rate is improved and the channel estimation performance is improved.

The invention discloses a medical data aided decision-making method and device fusing multi-modal data, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining the original medical data of a to-be-predicted patient; aligning semantics among the data of various modalities, and generating personalized feature vectors of the patient to be predicted; determining an activation weight according to the feature information of the personalized feature vector; according to knowledge data in the original medical data, a clinical rule is converted into a fine-adjustable constraint condition, and according to the constraint condition and the activated identification expert layer, feature analysis is performed on the personalized feature vector to obtain an illness state analysis state vector; assembling cue words, and reasoning the cue words; and generating a plurality of candidate reasoning paths, determining an optimal reasoning path in the candidate reasoning paths, and generating an optimal reasoning result by using the optimal reasoning path. The invention provides an LLM for performing dynamic verification and adjustment by fusing multi-dimensional contrast signals in real time in the process of medical data analysis and auxiliary decision generation.

The application provides an electromagnetic safety restoring force digital twin training method and system, constructs a test twin space, determines a test equipment group based on electromagnetic test data, and associates real interference data; determines interference dimensions based on equipment properties of each test equipment in the test equipment group, generates false interference data according to the interference dimensions; updates the state of the test equipment according to the real interference data, generates actual data, and directionally replaces the running data of the corresponding dimensions based on the false interference data to generate false data; collects calling information of the training end to the auxiliary plug-in, generates error area guide data according to the comparison information of the calling information, the actual data and the false data, and the auxiliary plug-in one-to-one corresponds to the running data of each dimension, so that the real interference data and the false interference data can be fused, the training complexity is improved, and the disposal efficiency of an operation and maintenance personnel for an electromagnetic safety event is improved.

The invention relates to the field of laser communication, and provides a parallel low-complexity frequency offset compensation method, device, equipment and medium, and the method comprises the steps: calculating the M-power variance of N M-PSK signals which are input in parallel, and carrying out the first frequency offset compensation of an original signal through the difference result of N points at intervals; the M-PSK signal is a PSK signal of which the modulation order is M; and solving an N-point M-power spectrum for the signal after the first frequency offset compensation, and performing second frequency offset compensation for the signal after the first frequency offset compensation by using the spectral line position of the M-power spectrum. The two-stage frequency offset compensation method provided by the invention has the characteristics of high dynamic range, low resource consumption, no need of data assistance and support of burst signal processing, and is particularly suitable for a frequency offset compensation scene in a high-speed laser communication link.

The invention discloses a dynamic perception-based maintenance auxiliary equipment monitoring system and method, and relates to the technical field of auxiliary equipment monitoring, and the system comprises the steps: dividing a monitoring region according to a monitoring range, marking a generator set, collecting historical operation data, auxiliary tool data and fault type data, and building a fault database for the monitoring region; calling the fault information of the monitoring area, analyzing the demand condition of the fault type for the auxiliary tools based on historical data, predicting the demand condition of the auxiliary tools in a monitoring period, analyzing the storage condition of the auxiliary tools in the monitoring range, and prompting management personnel to supplement until the auxiliary tools in the monitoring range are sufficient. According to the method, the fault database is established, maintenance records of effective maintenance personnel are integrated, data interference of invalid personnel is eliminated, and the actual requirements of different fault types for various auxiliary tools are objectively judged.

The utility model discloses a monitor operation module based on a voice recognition technology, which comprises a vital sign monitoring module used for collecting vital sign data of a patient and provided with a data transmission interface, and a data processing and display module used for receiving the data of the vital sign monitoring module and displaying the data on a display screen; the interaction module comprises a voice interaction unit, a touch control unit and an audio player; and the response module is connected with the data processing and display module, transmits the vital sign data in real time, monitors data changes, transmits prompt data to the display screen and the audio player, and assists in reminding a user. According to the vital sign monitor, the usability and the response speed of the vital sign monitor are improved through the man-machine interaction technology, the vital sign monitor is compatible with a clinically applied monitor through the data transmission interface, the operation burden of a doctor is reduced, and meanwhile a patient is reminded of key vital sign data.

The invention discloses an auxiliary management method and system for rattan plaiting skill data based on image analysis, and relates to the technical field of rattan plaiting image processing.The method comprises the steps that an initial plaiting topological graph is constructed by obtaining an original rattan plaiting image, and high-confidence and interference topology first-level sub-graphs are divided through pixel quality evaluation; taking the adjacent high-confidence topology sub-graphs as reference, calculating the similarity between the adjacent boundary length and the topological structure and coding a structure prior vector, and selecting weighted fusion according to vector dispersion or dividing into interference topology secondary sub-graphs according to weighted distance so as to determine generation prior; topological sub-graphs are repaired through a conditional generation model, a complete woven topological graph is output through fusion, and precision can be improved through optimization of time sequence, symmetry and boundary consistency. According to the method, the topological structure of the interference area can be accurately repaired, and the integrity and reliability of the rattan-plaited woven topological graph are remarkably improved.

The disclosure relates to a 5G or 6G communication system for supporting a higher data transmission rate. A user equipment (UE) includes a transceiver configured to: transmit, to a base station (BS), UE capability information that indicates support for data-aided modulation constellations, wherein the data-aided modulation constellations support a first transmission mode utilizing data symbols for channel estimation, and receive, from the BS, a modulation and coding scheme (MCS) indication. The UE also includes a processor operably coupled to the transceiver and configured to: determine, based on the MCS indication, a first data-aided modulation constellation, and generate first modulation symbols from input bits according to the first determined data-aided modulation constellation. The transceiver is further configured to transmit, to the BS, the first modulation symbols.

The invention provides a method for performing athlete trajectory tracking and motion data analysis by using an unmanned aerial vehicle cluster, and belongs to the technical field of physical training. The leading unmanned aerial vehicle collects wide-angle trajectory data, the assisting unmanned aerial vehicle obtains action videos at a high frame rate, and the collection accuracy is cooperatively guaranteed in combination with a local area network; the global trajectory is fused through Kalman filtering, action parameters such as joint angles are obtained by means of CNN and multi-view geometry, and the problem that in the prior art, the data dimension is single is solved; a visual report containing historical and standard data comparison is generated through multi-layer calculation of a presentation layer, a physiological layer and a tactical layer subsequently, fatigue early warning can be pushed to a coach in real time, comprehensive and deep data analysis support is provided for training of athletes, and the training precision is improved.

The invention provides a deep learning DD channel estimation system and method combined with an adaptive CIR truncation mechanism, and the system comprises a pilot frequency estimation module which carries out the channel estimation according to a received pilot frequency sequence, and outputs a pilot frequency assisted channel estimation value; the deep learning prediction module is used for predicting the channel state of the next moment according to the historical channel estimation sequence; the data auxiliary channel estimation module is used for balancing and decoding the data symbols based on the channel state, and obtaining data auxiliary channel estimation after reconstruction; the self-adaptive CIR truncation module takes data-assisted channel estimation as input and outputs a channel estimation value after truncation; and the periodic pilot frequency correction module is used for triggering pilot frequency updating and correcting the pilot frequency assisted channel estimation value. According to the method, the influence of a data judgment error is actively considered, the tap weight is dynamically output by using the lightweight neural network, the historical sequence is intelligently cleaned and cut off, error propagation is effectively inhibited, and the stability and robustness of prediction are improved.

The invention belongs to the field of supply chain auxiliary decision-making, and particularly relates to a supply chain data auxiliary decision-making analysis system and method. The supply chain data aided decision analysis method comprises the following steps: S10, acquiring supplier nodes selling parts to a local node in a supply chain, marking the sold parts as main parts, and capturing a current selling price; s20, dividing the industries for purchasing the main parts into reference industries belonging to the same genera as the local nodes and contrast industries belonging to different genera from the local nodes; and based on historical sales data of all suppliers, calculating the total sales amount of the main parts as the total sales amount. According to the scheme, market risks of suppliers are accurately measured through industry residual analysis and supplier sales proportion mapping, and the problem that risk scores of the suppliers are inaccurate is solved; and then a financial tool is connected to realize integration of purchasing and sleeve protection, and parts are replaced through residual error cross-correlation identification, so that supply interruption is finally avoided, and synchronous amplification of the purchasing price and the capital occupation cost is prevented.

The invention relates to the technical field of financial data processing, and particularly discloses an AI-based financial text data auxiliary processing system, which comprises a data acquisition module, a data preprocessing module, an AI intelligent analysis module, a data verification module, an intelligent decision-making auxiliary module, a data storage and management module and a safety management and control module. The system can automatically collect multi-format and multi-source financial text data in batches, after the data quality is optimized through preprocessing, intelligent text classification, key information extraction and structured conversion are achieved through an improved deep learning algorithm, the accuracy and compliance of the data are guaranteed in combination with multi-dimensional verification, then decision support such as financial analysis and risk early warning is provided, and the accuracy and the reliability of the system are improved. And data management and security are guaranteed through distributed storage and an omnibearing security mechanism. According to the method, full-process automation of financial text processing is realized, the processing efficiency and data accuracy are improved, the problem of data dispersion of an existing system is solved, the intelligent level of financial decision is improved, and compliance requirements are met.

The application discloses a kind of semi-supervised image semantic segmentation methods based on similarity weighting multi-teacher network model, comprising the following steps: preparing training set and verification set data for semantic segmentation model training;Construct similarity weighting based multi-teacher network model SW-MMTNet;The unlabeled data of training set is respectively subjected to strong data enhancement and weak data enhancement, to realize the data disturbance of unlabeled data;The training of SW-MMTNet model is carried out using the labeled data and unlabeled data of training set;The effect of segmentation model is evaluated using verification set, and the final image semantic segmentation model is obtained.The method can integrate the prediction information of multiple teacher networks, generate high-quality and stable pseudo-labels for unlabeled data, thereby effectively utilizing unlabeled data to assist the training of semantic segmentation model and reducing the demand for large amounts of data for semantic segmentation.In addition, the method also enables the student network of SW-MMTNet to learn more information through data disturbance.

The invention relates to an electric power dispatching auxiliary system and method. The system comprises a login authentication module, a data collection module, an auxiliary dispatching operation ticket to-be-written module, an auxiliary terminal calling module, an operation progress supervision module and a data output module. The system realizes secure access through identity recognition and authentication; acquiring and transmitting a scheduling application form, an operation order and equipment running state data by using a data collection module; the auxiliary writing module determines a writing requirement through similarity calculation and generates a corresponding scheduling operation ticket; the auxiliary terminal calling module calls a related unit according to the operation order and records an unsuccessful contact condition; the operation progress supervision module counts timeout conditions caused by equipment defects or low operation speed and calculates timeout rates in a classified manner; and finally, the data output module uniformly outputs various results. According to the method, the scheduling operation ticket generation efficiency can be improved, a scheduling terminal notification and supervision mechanism is optimized, and the intelligence and safety of power scheduling work are enhanced.

It can sometimes be difficult to perform signal switching correctly. [Solution] The switching device includes a receiving unit that receives IP (Internet Protocol) signals from multiple sources, which include video data, audio data, ancillary data, and time information, and a switching unit that uses the time information contained in the IP signals to switch the video data, audio data, and ancillary data to be output from among the video data, audio data, and ancillary data corresponding to each received IP signal.

The invention discloses a digital anchor power grid data auxiliary analysis method based on a multi-modal large model, and the method comprises the following steps: collecting and preprocessing multi-source power grid data, and generating a multi-modal data package; multi-modal feature extraction is executed, and a multi-modal feature structural body is constructed; performing matching screening to obtain a pre-screening expert set; performing matching degree calculation to generate a matching score, and forming a target expert sub-network set; an improved DeepAR network is introduced to predict key parameter trends, and reasoning output results of all expert sub-networks are generated; performing weighted fusion to obtain a power grid operation comprehensive analysis result, and generating inference chain data; inputting into a digital anchor generation module to obtain a power grid operation analysis video stream; periodical updating is carried out on expert sub-network specialist vectors and routing weight parameters, visualization and real-time performance of power grid data are achieved, and the efficiency and accuracy of a power system in the aspects of power grid fault prediction, trend analysis, equipment defect diagnosis and the like are improved.

The application provides a data transmission aid and relates to the field of data transmission.The data transmission aid comprises a data aid transmitter, a power line and a device power supply interface, one side of the data aid transmitter is provided with two USB interfaces and three network interfaces, and the other side of the data aid transmitter is provided with three male serial ports and three female serial ports.The data transmission aid realizes the functions of a serial port distributor, a serial port server, software distribution, data characteristic display, data mutual transmission between different interfaces and the like through the cooperation of internal hardware of the data aid transmitter and internal software of a display screen, so that the data aid transmitter and the display screen are provided with several most commonly used and most general data transmission interfaces, the special-shaped interface and the shell of the different models of the backup interface can be installed and dismounted in the installation slot at any time, the use needs of the data aid transmitter for some backup interface shells which are not commonly used can be ensured, and the application range is ensured.

The invention provides a vibration signal-based multi-source domain adaptive fusion transfer learning method, which comprises the following steps of: (1) acquiring vibration signals of mechanical equipment through various physical sensors, constructing multi-dimensional time sequence data, introducing data from different sources, and assisting feature learning of a target domain by using the data to obtain a multi-source domain adaptive fusion transfer learning model; the dynamic characteristics of the equipment under various working conditions are captured; (2) automatically extracting discriminative features from the mechanical vibration signals by using a deep residual shrinkage feature extraction network; (3) designing a multi-source domain feature adaptive weighting module to process and map source domain features; (4) inputting the weighted multi-source domain features and target domain features into an adversarial migration network framework for training; and (5) identifying a fault state. The method is reasonable in conception, consistency alignment of cross-domain features and strengthening of key features can be achieved, and therefore a high-precision fault diagnosis effect can still be obtained under the condition that target domain samples are rare and few.

The invention discloses a multi-source bill automatic identification and compliance verification method based on deep learning, and relates to the technical field of finance and tax bill processing. According to the technical key points, feature-level deep coupling of multi-source heterogeneous bill data is carried out through a multi-modal feature collaborative fusion recognition model and a bidirectional cross verification mechanism, and the information splitting problem of traditional single-modal recognition is eliminated; the identification deviation is effectively corrected through correction of homologous information bidirectional comparison and business data auxiliary verification, so that finance and tax accounting errors caused by core information identification errors are avoided, the workload of manual rechecking is reduced, and the overall efficiency of bill information entry is improved; besides, by means of implementation of a compliance verification standard library integrating laws and regulations and a rule dynamic adaptation engine, the system can be directly connected with a finance and tax laws and regulations official port, real-time capture of laws and regulations revision information and automatic conversion of verification rules are achieved, and the hysteresis problem of a static rule library is solved.

The application relates to a lost behavior judgment and alarm method and device based on data analysis, which comprises the following steps: obtaining accompanying persons of target lost persons from original data, calculating intimacy data of the accompanying persons, constructing an intimacy database corresponding to each target lost person based on the intimacy data; obtaining real-time monitoring data, combining trajectory information in the original data to determine whether there is a lost behavior corresponding to the target lost person; when the lost behavior exists, combining the intimacy database to determine the lost level of the lost behavior, and performing an alarm operation corresponding to the lost level. By introducing the intimacy data of the accompanying persons related to the target lost person, the lost behavior level is determined based on the intimacy data, the alarm operation corresponding to the lost behavior level is performed, the accuracy of the lost behavior judgment is improved, and the possibility of the target lost person being lost is reduced.

The present application aims at the problem that the performance of the diagnosis model is unstable in the auxiliary transfer training using single working condition source domain sample data, and proposes a reducer fault diagnosis method under multiple working conditions. The method uses the probability latent semantic analysis model to mine the implicit features between the reducer fault vibration signals under different working conditions, introduces the multi-source domain labeled auxiliary feature samples under different working conditions, constructs the Fisher kernel function based on the fault feature dictionary, establishes the multi-source domain transfer LSSVM fault diagnosis model based on the probability latent semantic analysis, improves the stability of the diagnosis effect, and realizes the fault diagnosis of the reducer under multiple working conditions in the actual production environment.

The invention belongs to the technical field of digital identification, and relates to a digital instrument test data auxiliary system, which comprises a data capture module, a data processing module, a data storage and judgment module and a human-computer interaction interface, the data capturing module collects the real-time state of an instrument data area through a monitoring camera, converts collected videos into pictures and transmits the pictures to the data processing module. The data processing module is used for carrying out region selection and image processing on the transmitted picture through an image processing system in computer software, and identifying digits in the picture; the data storage judgment module is used for transmitting the digits converted by the image processing system to a general data storage format or a database and marking the recording time of the digits; a user operates the human-computer interaction interface to realize various video digital capture requirements. According to the system, collected character images can be converted into digits to be stored in a computer, and test monitoring visually perceived by testers on site is replaced.

The application discloses a kind of maintenance auxiliary equipment monitoring system and method based on dynamic perception, it is related to auxiliary equipment monitoring technical field, according to monitoring range division monitoring area, mark generator set, collect historical operation data, auxiliary tool data and fault type data, establish fault database for monitoring area, call the fault information of monitoring area, based on historical data, analyze the demand condition of auxiliary tool of fault type, predict the demand condition of auxiliary tool of a monitoring period, analyze the auxiliary tool storage condition of monitoring range, prompt management personnel to supplement, until the auxiliary tool of monitoring range is sufficient, analyze the auxiliary tool storage condition of monitoring area, carry out scheduling to the monitoring area auxiliary tool in monitoring range, the application is by establishing fault database, integrates the maintenance record of effective maintenance personnel, eliminates invalid personnel data interference, objectively judges the actual demand of different fault types to various auxiliary tools.

A medical visualisation system including a medical visualisation device for performing a medical visualisation procedure and a monitor device. The medical visualisation system includes an auxiliary component including an auxiliary memory and an auxiliary communication interface. The auxiliary memory storing initial data and the auxiliary component being configured to transmit the initial data to the monitor device using the auxiliary communication interface, and wherein a monitor processing unit of the monitor device is configured to receive the initial data and adjust one or more parameters of the medical visualisation system based on the initial data.

The invention relates to a system for realizing operator base station drive test data acquisition and analysis based on WEB, which comprises a data acquisition subsystem and a distributed data analysis subsystem, and is characterized in that the data acquisition subsystem is used for automatically acquiring frequency spectrum data and base station demodulation data according to a preset data acquisition task and observing the acquisition condition of real-time data in real time through a network; and the distributed data analysis subsystem is used for receiving the data acquired by the data acquisition subsystem, processing and analyzing the data, analyzing the data according to user requirements, generating a corresponding report form, and checking and exporting an analysis result. By adopting the system, the method, the device, the processor and the computer readable storage medium for realizing operator base station drive test data acquisition and analysis based on the WEB, data acquisition and system analysis are separated, drive test data can be flexibly and quickly acquired and analyzed at low cost, the work of frequency utilization rate of a public mobile communication system is assisted to be completed, and the work efficiency is improved. The application range is wide.

There are disclosed an apparatus for generating an encoded audio scene, and an apparatus for decoding and / or processing an encoded audio scene; as well as related methods and non-transitory storage units storing instructions which, when executed by a processor, cause the processor to perform a related method. An apparatus for processing an encoded audio scene may include, in a first frame, a first soundfield parameter representation and an encoded audio signal, wherein a second frame is an inactive frame, the apparatus including: an activity detector for detecting that the second frame is the inactive frame; a synthetic signal synthesizer for synthesizing a synthetic audio signal for the second frame using the parametric description for the second frame; an audio decoder for decoding the encoded audio signal for the first frame; and a spatial renderer for spatially rendering the audio signal for the first frame using the first soundfield parameter representation and using the synthetic audio signal for the second frame, or a transcoder for generating a meta data assisted output format including the audio signal for the first frame, the first soundfield parameter representation for the first frame, the synthetic audio signal for the second frame, and a second soundfield parameter representation for the second frame.