499 results about "Stream processing" patented technology

The invention relates to the technical field of computer video processing, and discloses a low-delay video stream real-time processing method and device, and the method comprises the steps: obtaining original video stream data, and processing the original video stream data through employing a lightweight motion prediction method; processing the macro block data set and the predicted coding configuration parameter by adopting multi-thread assembly line coding to obtain a coded data block; establishing a data transmission mechanism to perform data flow control on the unified memory access interface; a heterogeneous task scheduling strategy is adopted to distribute task division results; a lightweight neural network is adopted to carry out parameter adaptive adjustment, and an optimized video stream processing result is obtained; according to the method, a zero-copy data transmission technology is adopted, and optimal configuration and efficient utilization of computing resources are achieved.

The invention relates to the technical field of cross-border advertisement putting, and provides a cross-border e-commerce advertisement accurate putting method and system based on a user portrait, and the method comprises the steps: building a unified data flow through multi-source data collection and standardization processing; calculating and fusing the real-time statistical features, the historical interest vectors and the context features in real time by using a stream processing engine to generate multi-dimensional feature vectors; inputting the feature vector into a multi-modal time sequence neural network model which captures a user behavior rule through time sequence coding and an attention mechanism and outputs a user behavior weight and an interest attenuation coefficient; dynamically updating the real-time interest score of the user by adopting an exponential decay model; and when the interest score reaches a threshold value, triggering a reinforcement learning agent to make a decision according to the comprehensive state information, and dynamically outputting an action strategy including release triggering, a bidding coefficient and a creative type. The problems that in the prior art, user interest modeling lags behind, and the self-adaptive capacity of the putting strategy is poor are effectively solved.

The invention relates to the technical field of cognitive inference, and discloses a cognitive inference system fusing a dynamic knowledge graph and deep learning. The system comprises a cognitive event stream processing module which is used for acquiring a real-time event stream and a cognitive target stream, identifying entity state change and extracting a reasoning intention; the graph mapping generation module is used for receiving the entity state change and the reasoning intention, constructing a mapping relation with knowledge graph topology and generating a multi-dimensional graph mapping library; the cognitive path decision-making module is used for extracting key topological characteristics from the multi-dimensional map mapping library, calculating a reasoning path migration probability and generating an initial cognitive path set; the time sequence evolution prediction module is used for carrying out time sequence evolution modeling on the knowledge graph topology based on the initial cognitive path set, identifying node state offset and updating a multi-dimensional graph mapping library; and the reasoning optimization module is used for analyzing a reasoning stage and a reasoning dependency relationship of the cognitive target flow according to the updated multi-dimensional map mapping library, and generating a dynamic cognitive reasoning library.

The invention relates to the technical field of spatial data analysis, and particularly provides a commercial vitality space-time prediction and dynamic evaluation method based on multi-modal feature fusion, which comprises the following steps: collecting geographic space, text semantics, real-time behavior and economic attribute multi-modal heterogeneous data, and cleaning and desensitizing to obtain a structured data set; features are extracted from the single-modal data, and commercial vitality core feature vectors are obtained through cross-modal attention fusion in combination with an emotion-vitality association model; inputting the result into a space-time sequence prediction model, and outputting a commercial activity prediction result in a future preset time period; combining real-time updated data to construct a quantization-quality two-dimensional evaluation system, and performing real-time calculation by using a stream processing framework to obtain a dynamic evaluation result; and displaying and outputting a structured evaluation report through a visual panel. According to the method, timely and reliable decision support is provided for scenes such as business district operation optimization, business investment decision and regional economy monitoring, and the precision and real-time requirements of business vitality analysis are comprehensively met.

The invention discloses a unified streaming processing method, system and device for multi-modal AI interactive content, a medium and a program product, and the method comprises the steps: receiving a request which is sent by a user and comprises an image and a text, and constructing a request context; extracting image features; carrying out image analysis and outputting an image description text; performing intention recognition and outputting an intention; multi-stage reasoning is carried out, and thinking content fragments are generated in a streaming mode; buffering and releasing thinking content fragments; generating text content fragments based on multi-stage reasoning; buffering and releasing the text content fragments; performing recommendation triggering based on the image description text and the intention; recommending content fragments; inserting a recommended position, buffering and issuing; and process event management: issuing event notifications when the beginning, any step fails and the end. The method is a universal method capable of transmitting different types of outputs in a single ordered stream, the analysis cost can be reduced, and the interaction experience and expansibility are improved.

The invention relates to the field of data processing, in particular to a distributed flow limiting method and device, equipment and a medium, which are used for realizing distributed flow limiting control on a service request. The method comprises the following steps: receiving business requests of business servers connected with a cache server, and clustering the business requests based on multi-dimensional object information of the business requests to obtain request clusters; aiming at each request cluster, respectively executing the following steps: sorting the service requests in the request cluster to obtain a request sequence; aiming at each service request in the request sequence, sequentially executing the following steps: acquiring the count of a counter corresponding to the multi-dimensional object information of the service request, comparing the count with a first quantity threshold value, and judging whether the count is updated or not based on a first comparison result; and sending the first comparison result to a service server corresponding to the service request to enable the service server to analyze the first comparison result, and if a return value in an analysis result is a first set value, performing flow limiting processing on the service request.

An audio synthesis method, an audio synthesis model training method, an apparatus, an electronic device, a computer-readable storage medium, and a computer program product, which relate to artificial intelligence technology. The method includes: invoking an audio synthesis model based on language information and preset style information of a target text to perform following processing, the audio synthesis model including a prior encoder and a waveform decoder: generating audio features corresponding to the target text based on the language information and the preset style information by using the prior encoder; performing normalizing flow processing on the audio features by using the prior encoder, to obtain a hidden variable of the target text; and performing waveform decoding on the hidden variable of the target text by using the waveform decoder, to obtain a synthetic waveform conforming to an audio style described in the preset style information and corresponding to the target text.

The invention relates to the technical field of big data storage, and discloses a multi-level data storage and mixed query method and system, a storage medium and equipment, and the method comprises the steps: collecting original buried point data, carrying out real-time stream processing, storing the processed real-time data in a hot storage layer, carrying out timing offline batch processing on historical buried point data, and storing the processed real-time data in a hot storage layer; converting into an optimized storage format and storing in an object storage system; an extended metadata management system is constructed, and structure information, partition information and access frequency information of data in the object storage system are automatically extracted and managed; executing query, and reducing the data scanning amount through a query optimization strategy; automatically migrating data among different storage layers through a dynamic cold and hot data layering mechanism; a unified query interface is provided, a query request is distributed to a corresponding real-time processing system or an offline query system based on an intelligent routing strategy, and a query result is cached and returned, so that high-performance and low-cost storage and second-level query of data are realized.

The invention discloses a real-time anti-fraud transaction intelligent analysis system and method based on multi-chain collaboration, and the system comprises a dynamic federated learning module which is used for dynamically loading and updating a global model during the operation of a JVM (Java Virtual Machine) by adopting a Java Agent bytecode enhancement technology; the multi-chain collaborative verification engine comprises an alliance chain and a private chain, the alliance chain is formed by taking a plurality of banks as organization nodes and is used for distributed evidence storage of transaction characteristics among financial institutions, and the private chain is maintained by a supervision mechanism and is used for auditing and compliance verification; and the real-time stream processing pipeline processes the transaction request in a stream processing mode to obtain feature data, and makes a decision based on the feature data and the global model output by the dynamic federation learning module. The intelligent risk control system which supports cross-mechanism collaboration, privacy protection, high throughput and low delay is constructed, and the problems that in the prior art, model updating is lagged, a verification mechanism is not transparent, and processing delay is high are solved.

The invention discloses a road crack segmentation model based on Transform enhancement, a construction method and a use method, and relates to the technical field of image processing. According to the method, a multi-scale cross-level module (MSAF) is provided, and overall image global semantics and 2 * 2 local block features are explicitly fused by embedding a multi-head attention mechanism of two-dimensional sine position coding. The module not only retains edge details of the crack, but also captures long-range context information, so that the crack detection capability of the model is remarkably improved, and additional parameters do not need to be added; a global-local double-flow coding architecture, namely MSAFNet, is designed, a parallel double-flow coding architecture is adopted, global flow processing is performed on the whole image, local branch processing is performed on 2 * 2 cutting, and seamless recombination is performed after the features of the whole image and the local branch processing are fused through an MSAF module, so that fine-grained texture information of cracks is reserved, and global context information is captured; therefore, the high-precision crack segmentation capability can be maintained in different scenes.

The invention provides a big data advertisement putting method and system based on a cloud platform. The method comprises the following steps of performing multi-source data collection and preprocessing; dynamically constructing a user portrait; performing multi-dimensional analysis on the behavior pattern; target audience subdivision is based on a Gaussian mixture model; dynamically generating personalized advertisement creativity; a step of optimizing a multi-target delivery strategy; a step of real-time bidding and bidding optimization; a step of putting execution and real-time monitoring; a step of effect evaluation and multi-dimensional feedback; and carrying out adaptive system adjustment and iterative optimization. The big data advertisement putting method based on the cloud platform has the advantages that putting precision and efficiency are improved, audiences are subdivided through complex statistics, dynamic creativity is generated, advertisement correlation is enhanced, the predicted click rate is increased by 15-20%, and budget waste is reduced. And the cloud platform supports stream processing and automatic adjustment, so that the system can quickly respond to market changes, and the delay is reduced to a millisecond level.

The invention discloses a data verification method and system based on a Drools rule engine, and belongs to the technical field of data analys.The method comprises the steps that a data stream is accessed in real time, and the data stream is preprocessed to form a structured data object; receiving a verification rule configured by a user, and performing dynamic compiling and hot loading on the verification rule to update a rule base in the Drools rule engine; and inserting the structured data object as a fact object into a Drools rule engine, and executing the verification rule configured in the rule base by the Drools rule engine based on a ReteOO algorithm to generate a verification result. According to the method, real-time data stream processing and large-scale field association verification are supported, so that the data quality, the service efficiency and the overall reliability of the system in the aviation field are improved.

The invention relates to the technical field of electronic information data processing, in particular to an electronic information data AI anomaly detection and repair method, which comprises the steps of preprocessing multi-modal data, automatically identifying multi-format data and adaptively encoding and normalizing; fusing a convolutional neural network, a bidirectional long-short-term memory network and a Transform architecture to extract multi-modal features, and judging abnormality through an improved deep isolation forest model; the exceptions are subdivided through a lightweight convolutional neural network in combination with a rule base, and service scene labels are automatically matched; based on the abnormal type, grade, scene and semantic four-dimensional degree, a reinforcement learning model intelligently matches a repair strategy such as filling, correction or manual intervention; a data, model and repair three-level parallel architecture and a sliding window mechanism are adopted to realize real-time stream processing; incremental learning closed-loop optimization is driven through validity and rationality dual verification, and the method is suitable for high-reliability data governance scenes such as finance and communication.

The invention relates to the technical field of video compression, in particular to an AI-based video compression platform, which comprises a key target detection module, an image region division module, a coding level setting module, a parameter adjustment and analysis module and a coding result integration module. According to the method, key region extraction is completed by collecting pixel color combination and contour change information in a video frame image, background and non-background region boundaries are delimited, a region attribute labeling result is constructed, a region distribution classification relation mapping coding level is established, and reference frame and prediction interval configuration is extracted. Real-time adjustment of compression levels is realized by performing normalized combination analysis on textures and motion change trends of different regions, and region fragments under different compression levels are recombined and subjected to unified code stream processing, so that coding resources can be dynamically allocated on the basis of accurately identifying contents in a compression process; the problems of inaccurate target positioning, fixed compression configuration, delayed content response and the like in an existing compression system are effectively solved.

The invention discloses a remote OTA firmware upgrading method for multiple MCUs. The remote OTA firmware upgrading method comprises the steps that a master control MCU receives OTA firmware data issued by a cloud server in real time and stores the OTA firmware data; the main control MCU carries out protocol analysis on the received OTA firmware data, carries out shunting processing on the OTA firmware data and Modbus data, and carries out data analysis and verification on the OTA firmware data to obtain a data packet passing verification; the master control MCU subpacks the data packets passing the verification and forwards the data packets to the target sub-MCUs; the sub-MCUs receive the data packets and carry out subpackage recombination and sequence control on the data packets to form a complete OTA upgrade packet; and the sub MCU executes the Bootloader upgrade according to the OTA upgrade package.

The invention discloses a network information security access control system based on dynamic trust evaluation, and the system comprises a behavior collection module which collects original behavior log data, and builds a behavior feature sequence reflecting an entity security state through a preset streaming processing engine; the behavior analysis module is used for carrying out long and short term behavior modeling analysis on the behavior feature sequence to obtain long and short term behavior pattern representation, outputting potential causal chain data based on the long and short term behavior pattern representation and an abnormal causal inference model combined with context perception, and constructing a dynamic trust portrait according to the potential causal chain data; the long and short-term behavior modeling analysis comprises long behavior modeling analysis and short behavior modeling analysis; and the strategy generation module is used for driving a strategy engine in real time based on the dynamic trust portrait, carrying out attribute extraction and dynamic attribute coding based on a behavior sequence in a software defined boundary frame, and dynamically generating and issuing an optimal dynamic access strategy with attributes according to a minimum permission principle.

The invention discloses a data processing method and device, electronic equipment and a nonvolatile storage medium. The method comprises the following steps: dividing a data matrix to be calculated into a plurality of data blocks with target sizes; data corresponding to the data blocks are allocated to a plurality of stream processors for data reading operation, the data reading operation is used for loading the data from a global memory of the edge computing device to a shared memory, and the data reading and writing efficiency of the shared memory is higher than that of the global memory; and loading the data in the memory block with the preset vector length in the shared memory into a register by taking the preset vector length as a unit, and operating the data loaded into the register by adopting a stream processor to obtain an operation result. The technical problem that the computing efficiency is not high due to the fact that the related technology is limited by the particularity of the memory access and computing rule of the correction operator and the hardware performance of the edge computing equipment cannot be fully exerted is solved.

The invention relates to the field of energy management, in particular to an energy consumption settlement method based on multi-source data, and the method comprises the following steps: S1, obtaining original multi-source heterogeneous data from an intelligent meter, an environment sensor and a building equipment management system through an edge computing gateway, and completing protocol conversion and data aggregation; s2, performing streaming processing on the original multi-source heterogeneous data, executing validity verification, missing value marking and timestamp alignment, and generating regular data subjected to time alignment; and S3, performing abnormal mode recognition on the regular data based on an unsupervised machine learning algorithm, repairing abnormal data in combination with a time sequence prediction technology, and outputting a clean data set subjected to data quality enhancement. According to the invention, through the load state identification and reinforcement learning optimization algorithm, the adaptive adjustment of the data acquisition strategy is realized, the key state monitoring precision is ensured, the system operation overhead is effectively reduced, and the overall energy efficiency management level is improved.

A device includes one or more processors configured to detect single-stream data and generate multi-stream augmented data that includes one or more modified versions of the single-stream data. The one or more processors are configured to process the multi-stream augmented data to generate multiple output channels. The one or more processors are also configured to reduce the multiple output channels to produce single-stream output data.

The invention discloses an energy storage battery system fault prediction method and system based on machine learning, and the method comprises the steps: collecting historical operation data from an energy storage battery system, and carrying out the denoising of the historical operation data, and obtaining a denoised operation data sequence; performing multi-dimensional spatial-temporal feature extraction on the denoised operation data sequence by using a convolutional neural network, and determining a key feature vector related to a fault; constructing a long short-term memory network, and training the long short-term memory network based on the key feature vector to obtain a fault prediction model; acquiring real-time operation data of the energy storage battery system, processing the real-time operation data by using the fault prediction model to obtain initial fault probability distribution, and performing risk assessment according to the initial fault probability distribution to obtain a fault risk assessment value; and for the fault risk assessment value, a real-time stream processing framework is adopted to monitor a data stream, current charging and discharging state information is obtained, and whether an early warning is sent out or not is determined based on the current charging and discharging state information.

The invention relates to a communication engineering full-process safety production risk dynamic assessment method, in particular to the technical field of communication engineering, and effectively overcomes the hysteresis quality and one-sidedness of a traditional static assessment method. According to the method, real-time data stream processing and an online learning mechanism are fused, continuous perception and quantitative evaluation of safety production risks are realized, short-term sudden risk response and long-term law research and judgment can be intelligently balanced, the timeliness of risk early warning and the accuracy of decision support are remarkably improved, the engineering field safety management and control capability is enhanced, and the engineering field safety management and control efficiency is improved. The occurrence probability of safety accidents is reduced.

The invention discloses a network security situation real-time evaluation method and system based on urban rail transit, and the method comprises the steps: collecting original data from a plurality of heterogeneous data sources of an urban rail transit network, carrying out the standardization processing, and generating a standardized security event; constructing a digital twin map; calculating the risk probability of nodes in the graph by using a time-space diagram neural network model, predicting an attack path and evaluating the service influence; a stream processing engine is adopted to consume a standardized security event in real time, and model reasoning, node risk probability updating, attack path prediction and service influence evaluation are performed on a local sub-graph related to the current event in the digital twin graph; and performing real-time parameter fine tuning on the space-time diagram neural network model through an online learning mechanism, and updating the space-time diagram neural network model through an incremental learning mechanism. According to the technical scheme, real-time dynamic prediction and active defense of urban rail network security threats are achieved, and the situation awareness capability and the autonomous protection level of key infrastructures are remarkably improved.

The invention relates to the technical field of data processing, and particularly discloses a financial information management system and method based on big data analysis. The system comprises a multi-source heterogeneous data acquisition module, a data semantic understanding and mapping module, a dynamic rule engine module, a real-time stream processing module and a financial decision support module, realizes data semantic unification through deep learning and a knowledge graph, and dynamically adapts to business change by means of a self-learning rule engine. And real-time calculation and anomaly detection are guaranteed based on a distributed stream processing framework, and finally the intelligence level and the operation efficiency of financial decision making are improved through visualization and predictive analysis.

The invention provides a data stream processing method for binding a kernel and a network card queue, and relates to the technical field of data stream processing, and the method comprises the following steps: setting the number of queues of each gigabit network card and each 10-gigabit network card as the maximum number of queues supported by each network card and the minimum number of idle CPU (Central Processing Unit) cores; binding each queue of each gigabit network card and each 10-gigabit network card to an idle CPU core; the number of queues of each gigabit network card is set to be the minimum value of the maximum number of queues supported by each gigabit network card and the number of idle CPU cores, and the number of queues of each 10-gigabit network card is set to be the minimum value of the maximum number of queues supported by each 10-gigabit network card and the number of idle CPU cores; enabling each queue to process a corresponding data stream through an idle CPU; according to the invention, the problems of non-intelligent and unbalanced resource allocation and possible configuration failure in a mixed rate network card environment are effectively solved.

The invention discloses an e-commerce data intelligent processing method and system based on dynamic flow optimization. The method comprises the following steps: acquiring and analyzing multi-platform e-commerce data, and converting the multi-platform e-commerce data into a standardized format; constructing a matching rule by adopting a decision tree algorithm, and performing classification and priority marking on the data; the data flow is dynamically scheduled based on the priority, so that the high-priority data is executed preferentially; data are processed through parallel nodes, performance indexes such as network delay and throughput are monitored in real time, and data stream processing efficiency is optimized in combination with a dynamic scheduling algorithm. The system correspondingly comprises a data standardization module, a rule matching module, a dynamic marking module, a parallel processing module and a monitoring scheduling module, efficient and self-adaptive processing of the multi-source e-commerce data is achieved, and the data processing real-time performance and the system throughput are remarkably improved.

The invention discloses a Beidou inertial cooperative operating vehicle multi-source fusion high-precision positioning method, and relates to the technical field of vehicle-mounted navigation positioning, and the method comprises the steps: S1, completing GNSS / IMU / wheel speed collection, PPS alignment, external parameter calibration and quality control at an end side, and carrying out the uplink at unified event time; s2, constructing a short window factor graph and an IMM, carrying out adaptive switching between RTK / PPP / DR, and outputting HPL by adopting logarithm domain covariance update and RAIM; s3, an out-of-order is rearranged according to event time, and probability association and double-time-axis smoothing are implemented; s4, performing deduplication shaping on the cloud edge stream, introducing an HPL perception fence and performing multi-channel convergence; s5, driving mirror image updating by scene reweighting, soft minimum and tail risk; centimeter / decimeter-level positioning, tunnel exit convergence, fence low false alarm and end-to-end time delay P95 < = 1s can be realized under complex shielding, and playback and statistics are kept consistent.

The invention belongs to the technical field of electric digital data processing, and particularly relates to an AI chip performance data acquisition method which comprises the following specific steps: deploying hardware-level distributed monitoring nodes: in an AI chip design stage, integrating light-weight distributed monitoring units, comprising a calculation module monitoring point, a storage module monitoring point, a power consumption monitoring point and a time synchronization unit. According to the invention, through hardware-level monitoring nodes and microsecond-level sampling frequency, it is ensured that data acquisition and AI tasks are synchronously carried out, offline log analysis, dynamic sampling adjustment and real-time stream processing are replaced, 1ms-level performance fluctuation can be captured, dynamic optimization is realized through real-time feedback, and the deficiency of real-time performance and dynamic performance is solved; according to the modular power meter, synchronous acquisition of total power consumption and sub-module power consumption is realized, the energy consumption ratio of a specific module is determined, and a timestamp and a correlation analysis engine are unified.

The invention discloses an unmanned vending equipment state monitoring method and system based on streaming and batch integration, and the method comprises the steps: collecting the multi-dimensional state data of equipment through a data collection layer, and carrying out the unified structural processing based on a data model; the streaming processing layer acquires a structured equipment state event stream from a Kafka data source, performs real-time processing through timestamp allocation, data grouping and a window aggregation strategy, and calls a multi-dimensional health scoring algorithm to calculate a real-time health score; the storage layer adopts a streaming-batch integrated architecture, stores the data to corresponding data tables respectively, and meets the requirements of real-time query and historical analysis; the analysis query layer performs data aggregation calculation based on the aggregation table to obtain an analysis result; and the intelligent decision-making layer calls an anomaly detection algorithm to carry out anomaly judgment, and equipment health assessment and predictive maintenance are realized. The problems that an existing monitoring system is insufficient in real-time processing, poor in storage adaptability, lack of intelligent prediction, weak in multi-dimensional analysis and limited in expansibility are effectively solved.

The present application relates to a multi-modal large model-based industry electricity abnormality intelligent analysis method and system, comprising the following steps: S1: obtaining power system multi-source data and preprocessing to obtain standardized time series data set; S2: based on the standardized time series data set, feature extraction is carried out, and a feature data set is constructed; S3: based on the feature data set, an industry electricity abnormality knowledge graph is constructed in combination with a multi-level correlation relationship network; S4: a multi-modal abnormality detection model is constructed by fusing time series Transformer, graph neural network and knowledge embedding module, and is trained based on the feature data set and the power knowledge graph; S5: the trained multi-modal model is deployed on a distributed stream processing platform to realize real-time abnormality detection and multi-dimensional intelligent analysis; S6: the real-time abnormality detection and multi-dimensional intelligent analysis results are visualized. The present application can realize all-round intelligent monitoring of industry electricity abnormalities facing complex industry actual demands.

The invention discloses a data center system based on decision analysis. The data center system comprises a data acquisition layer, a data storage layer, a data analysis layer, a decision engine layer and a data service layer, the data acquisition layer is used for acquiring various heterogeneous data sources inside and outside an enterprise, supporting access of multiple data sources and calculating a null rate and a numerical range compliance rate of a data set; the data storage layer carries out data organization by adopting a mixed architecture combining a data lake and a data warehouse, and is used for storing various data from the data acquisition layer in an original format; and the data analysis layer works cooperatively with the real-time analysis unit through the batch processing unit. According to the invention, a flexible data access channel is constructed by integrating Flume, Sqoop and Kafka tools, efficient acquisition and cleaning of structured, semi-structured and unstructured data are realized, and meanwhile, the architecture design that Spark is adopted for batch processing analysis and Flink is adopted for real-time analysis is adopted, so that the problem of off-line batch processing and real-time stream processing splitting is solved.