529 results about "Collaborative processing" patented technology

The invention provides a multi-agent cooperative processing system and method for multi-scene legal complaint consultation, and relates to the technical field of artificial intelligence, and the method comprises the steps: receiving a legal or complaint consultation text inputted by a user, and generating structured consultation request data; automatically generating an evidence obtaining template, and forming a standardized evidence obtaining data packet; constructing a demonstration tree and outputting a multi-scheme decision matrix; performing multi-objective optimization by combining the success rate, the timeliness, the cost and the regional qualification index, and outputting a comprehensive recommendation result; generating a knowledge unit set, and updating a regulation index and agent prompt template library; predicting a re-complaint risk and generating a remedy and optimization suggestion; and outputting review result data, and reversely updating the review result data to the agent configuration strategy in the step of the knowledge construction module. According to the method and the device, the technical targets of multi-agent cooperative processing and dynamic optimization management in the whole legal complaint consultation process can be realized, and the technical effects of improving consultation processing efficiency, enhancing scheme recommendation accuracy, reducing manual participation cost and ensuring data compliance and risk controllability are achieved.

The invention provides a real-time intelligent agent dynamic scheduling method and system composed of multiple large models, and is applied to the technical field of data processing. The method comprises the following steps: performing dynamic scheduling expansion around a plurality of large-model real-time intelligent agents, firstly receiving task request data of a real-time service scene, performing grouping according to task types, priorities and resource requirements, pre-configuring a scheduling strategy, and generating a pre-scheduling queue and resource pre-planning information; and in combination with the target configuration information, scheduling task information after secondary optimization is obtained through multi-dimensional optimization such as task splitting and priority rearrangement. Dynamically adjusting and generating a target scheduling instruction based on real-time system resources and a model operation state, and cooperatively processing the target scheduling instruction with the large model capability adaptation parameters to form scheduling execution information; and finally, issuing to model nodes according to preset service quality requirements and resource constraints, generating information such as adaptive parameters and priority rules through feature extraction, association processing and the like, and supporting multi-model cooperation to efficiently complete real-time service tasks.

The invention discloses an efficient processing method and system for online troubleshooting of the information department, and relates to the technical field of automatic operation and maintenance, and the system comprises a multi-modal data acquisition module, a dynamic knowledge graph construction module, a reinforcement learning driven self-healing module, an edge co-processing module and a cross-domain privacy calculation module. According to the method, the limitation of traditional single-dimensional monitoring is broken through, multi-source heterogeneous data such as server performance, container state and network flow are fused, a high-timeliness and high-consistency decision information base is constructed through unified time sequence alignment and semantic standardization, and a panoramic view is provided for dynamic analysis; based on an incremental learning mechanism and a causal inference engine, a service topology dependency relationship is updated in real time, a deep root cause is accurately positioned, and a cross-system and cross-level complex fault combination is solved.

Disclosed are a data query method and system, and a related device, relating to the field of artificial intelligence. The method comprises the following steps: a data query system acquires input data; obtains, on the basis of the input data, metrics to be queried as requested by a user; uses, on the basis of said metrics, a multi-agent collaborative processing strategy to confirm a data query result, the multi-agent collaborative processing strategy being used for indicating a manner of implementing data retrieval on the basis of at least one of a graph model and a vector database; and presents the data query result, wherein the graph model expresses, in a graph-based form, correspondences between the metrics and between the metrics and database table data, and the vector database expresses, in a vector-based form, descriptive texts and computational formulas of the metrics. In this way, even if the expression of input data differs from the expression of fields in the database, or if metrics to be read are not present in the database, corresponding SQL statements can still be obtained by means of the graph model or the vector database, thereby reducing the error rate of natural language query, and meeting the query requirements of users.

The invention discloses a delivery task cooperative processing method, device and equipment based on a quadruped robot and a medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: analyzing a delivery task through an upper computer, combining the real-time dynamic information of an office scene, and employing an improved binary particle swarm optimization algorithm to optimize the task distribution of multiple robots, and efficient collaborative distribution in an intelligent office environment is realized. The robot receives a distribution scheme sent by the upper computer, a multi-level navigation strategy is started by utilizing environmental perception data, a topological map and a D * Lite algorithm are fused to carry out path planning, and dynamic obstacle avoidance and efficient distribution are ensured. The target path planning information is monitored in real time through the upper computer, dynamic analysis is carried out through the scene perception algorithm, the scene relevance result is obtained, the collaborative distribution strategy is executed to complete the distribution task, the task response speed and the distribution efficiency are improved, interference to the office environment is reduced, and the work efficiency is improved. And automation and intelligentization of robot distribution in an intelligent office scene are realized.

A latent transformer architecture with latent attention mechanisms and expert processing systems for federated deep learning is disclosed. The system operates entirely within latent space, eliminating traditional embedding and positional encoding layers while maintaining full attention capabilities. Input data is compressed into latent vectors via variational autoencoder encoding, then processed by a latent attention module that computes query, key, and value matrices directly from latent representations. The architecture incorporates expert processing systems including gated latent expert networks for sparse computation and latent mixture of experts for collaborative processing. In the gated approach, a routing network selectively activates specialized expert modules based on latent vector characteristics. The mixture approach enables all experts to contribute through weighted combination, facilitating distributed computation and enhanced model expressiveness.

The invention discloses a multi-agent collaborative application big data management system and method. The system comprises a data acquisition unit, a data preprocessing unit, a multi-agent collaborative processing unit, a data storage unit and a data scheduling unit. The data acquisition unit is used for acquiring multi-source heterogeneous original big data and transmitting the multi-source heterogeneous original big data to the data preprocessing unit; and the data preprocessing unit is used for executing preprocessing operation including de-duplication and format standardization on the original big data. The invention relates to the technical field of big data management. According to the multi-agent collaborative application big data management system and method, through a multi-agent collaborative algorithm, the system can realize a self-adaptive data classification and optimization strategy, a data processing mode is dynamically adjusted, the processing efficiency problem of a traditional static rule during data mode fluctuation is improved, and the data processing efficiency is improved. The system adopts a dynamic resource scheduling strategy based on a data access demand and a system load, efficiently distributes calculation and storage resources, and facilitates priority processing of high-priority tasks.

The invention provides an intelligent progress management and control method and system for information engineering supervision, and relates to the technical field of information, and the method comprises the steps: pushing early warning information to a first-party system and a second-party system through message middleware, collecting a feedback state to complete closed-loop cooperative processing, and updating a processing result to a central database; based on the central database, combining historical project data, performing pattern recognition analysis to extract a progress deviation rule, and dynamically optimizing a plan node threshold and buffer time of a subsequent stage according to the progress deviation rule to obtain an optimization strategy; and on the basis of the optimization strategy, constructing a configurable progress management and control template library, so as to quickly adapt and generate a corresponding progress management and control node and an early warning threshold value in the new-type project. According to the invention, a full-process closed-loop management system from data integration, intelligent optimization to dynamic adaptation is constructed, and the fundamental transformation of project progress management and control from passive response to active prediction and from experience driving to data driving is realized.

The invention provides a complex task multi-agent collaborative disassembling method and system based on a business process, and relates to the technical field of artificial intelligence, and the method comprises the steps: constructing a graph through analyzing a business process document, matching a complex task with a process node, identifying a closed loop dependence path, and selecting an agent based on a constraint logic formula. And establishing a causal tracing chain to process abnormal conditions, and performing semantic alignment to ensure data transmission between subtasks. According to the method, the automation degree and the execution efficiency of multi-agent cooperative processing of complex tasks are improved, and the fault-tolerant capability of the system and the reliability of task completion are enhanced.

The invention relates to the technical field of medical information and block chain crossing, in particular to a medical image co-processing system and method based on a block chain value network, and the system comprises a block chain network, a privacy calculation gateway, an intelligent contract module, a hybrid storage module and a communication module. The intelligent contract module is provided with a collaborative annotation sub-module, a verifiable federated learning sub-module and a dynamic value distribution sub-module; the method comprises the steps of data preprocessing and encryption uploading, collaborative labeling and quality verification, verifiable federated learning model training, dynamic value distribution, consensus evidence storage and supervision auditing. Medical data islands are broken through by means of the block chain network, data availability and invisibility are achieved in combination with the privacy computing gateway multi-encryption technology, the contradiction between data sharing and privacy security is solved, accurate contribution and return matching is achieved through the multi-dimensional contribution degree evaluation model and the intelligent contract, and the accuracy of data sharing is improved. The traceability and the compliance are ensured by full-process operation logs and data uplink evidence storage.

The invention discloses a property asset digital operation and intelligent security linkage cooperative processing system and an implementation method, and relates to the technical field of Internet of Things, and the system comprises a protocol adaptive gateway layer which solves the problem of compatibility of multi-manufacturer equipment, and outputs a standardized data stream based on a dynamic protocol sniffing and fingerprint feature library, SDK-free access control, fire protection, monitoring and IoT equipment; the method comprises the following steps: constructing a millimeter-level precision three-dimensional space model by a digital twinborn middle table layer, fusing a security event and operation data, and ensuring that the operation is compliant and audible through blood relationship tracking; the intelligent decision execution layer combines a rule engine and machine learning to generate a linkage instruction, drives a work order to be automatically distributed and positions an equipment fault responsible party; and the edge cooperative computing node locally realizes video millisecond-level analysis, and starts localized linkage according to an emergency plan when the network is disconnected, so as to ensure that the core security function is not interrupted.

The invention relates to the technical field of community management, in particular to an intelligent community service management method and system based on cloud computing, and the method comprises the following steps: S1, collecting community multi-dimensional data in real time through a heterogeneous data collection terminal; s2, constructing an edge-cloud cooperative processing framework; s3, implementing intelligent video analysis; s4, realizing data-driven resource scheduling; and S5, establishing an intelligent storage system. According to the invention, omnibearing perception of a community environment is realized through a multispectral camera and a distributed sensor network, data processing efficiency is improved by adopting an optimized lightweight algorithm and an edge-cloud collaborative architecture, and a visual decision platform is constructed in combination with adaptive resource scheduling and an intelligent analysis technology. A closed-loop management system integrating real-time monitoring, intelligent early warning and accurate operation and maintenance is formed, the community management efficiency is improved, the operation cost is reduced, stable operation of the system is guaranteed through multiple safety mechanisms, and an intelligent solution is provided for modern community services.

The invention provides a task processing method and system based on multiple agents, electronic equipment and a medium, and the task processing method comprises the steps: responding to a received processing request of a target task, and splitting the target task into a plurality of subtasks; processing any subtask by adopting a preset hybrid retrieval mechanism to obtain a target agent set matched with each subtask; wherein the preset mixed retrieval mechanism comprises at least two types of preset retrieval strategies; processing the corresponding sub-task based on the target agent set to obtain a sub-processing result; and aggregating each sub-processing result to obtain a target processing result corresponding to the target task. According to the method and the device, the cooperative task processing among multiple agents is realized based on a multi-agent dynamic arrangement mechanism, and a cross-agent call link tracking mechanism, a resource metering and dynamic allocation mechanism and a charging and profit sharing mechanism are provided, so that the support of cross-domain call link tracking, dynamic charging and the like is realized.

The invention provides a distributed cooperative processing method and system based on event driving, and relates to the technical field of data processing, and the method comprises the steps: analyzing event stream data, screening the event stream data into time sequence features and semantic features, and fusing codes to construct event multi-dimensional feature vectors; constructing a load situation prediction model on the basis, and generating a comprehensive load descriptor; grouping the processing nodes according to the descriptors and reconstructing a resource topology; mapping to an optimization space to generate a scaling decision scheme; and constructing a risk assessment function to realize co-evolution of the prediction model and resource configuration. The system processing efficiency can be effectively improved, the resource consumption is reduced, and the load fluctuation adaptability is enhanced.

The invention relates to the technical field of data processing, can be applied to the field of science and technology finance, and discloses a credit evaluation method and device for multi-source data co-processing and related equipment. The method comprises the steps of receiving credit evaluation data of at least two types of preset data sources, and performing cleaning, time scale unification and numerical standardization processing on the credit evaluation data to form standardized multi-source feature data; constructing credit portrait data based on the standardized multi-source feature data; performing scoring reasoning on the credit portrait data through a pre-trained credit extension evaluation model, and generating credit extension evaluation result data through a limit mapping rule; receiving the use behavior data and the performance behavior data, performing time sequence analysis and deviation identification, and generating credit use behavior analysis data; and updating credit extension evaluation model parameters based on the credit use behavior analysis data, loading the credit extension evaluation model parameters to a model operation state, outputting credit extension evaluation result data, and recording model parameter updating information. According to the method, the dynamic adaptive capacity of multi-source credit assessment is improved.

The invention provides an industrial operation area multi-view video collaborative analysis method and system based on artificial intelligence. The method comprises the following steps: firstly, acquiring multi-view video data; secondly, based on the working area and the equipment layout, deducing a semantic association relationship among the multi-view video data; according to the semantic association relationship, dynamically adjusting the analysis priority of the multi-view video data through an artificial intelligence model to determine an analysis priority sequence; and according to the determined analysis priority sequence, executing interactive co-processing of the multi-view video data, and finally generating a co-analysis result. According to the technical scheme provided by the invention, the physical layout and the workflow of the industrial site are converted into the semantic association relationship, and the dynamic scheduling analysis resources and the interactive collaborative analysis are performed according to the semantic association relationship, so that the spanning from simple monitoring of a plurality of isolated video streams to intelligent and coherent understanding of the whole operation flow is realized; and the intelligent level and decision support capability of monitoring analysis are improved.

The invention relates to the field of inspection, and particularly provides a public inspection operation safety and abnormal behavior real-time early warning data processing method and system, a terminal and a medium, and the method comprises the steps: collecting the position and behavior data, field visual data, facility state data and task data of inspection personnel in real time; realizing dynamic synchronization of a physical scene and a virtual model based on a data-driven digital twin model; identifying personnel behavior safety abnormity and polling object and environment abnormity in the digital twinborn model, performing risk assessment in combination with the context of abnormity occurrence, and generating graded early warning information; and finally, the early warning information is synchronously pushed to the field mobile terminal and the back-end command platform, and a cooperative processing flow is triggered. Through multi-modal data fusion and digital twinning dynamic synchronization, accurate identification of personnel safety and equipment abnormity, context risk assessment and front-end and rear-end collaborative early warning in inspection operation are realized, and the safety guarantee level and the emergency response efficiency are effectively improved.

The invention discloses an intelligent voice interaction method and system based on an intelligent agent and tool extension, and relates to the technical field of artificial intelligence. Terminal equipment captures a user request and then evaluates the user request as a simple task or a complex task by a cloud end, field intention logic processing is directly performed on the simple task, and the complex task is marked as needing cooperative processing; the cloud dynamically arranges the complex tasks in combination with the tool list packaged by the terminal to generate a tool calling instruction; the terminal equipment performs protocol analysis and mapping on the tool calling instruction, and performs tool distribution and task execution; and after the task is executed, the terminal equipment returns an execution result to the cloud, and resource recovery and result presentation are carried out. According to the invention, flexible, efficient and extensible intelligent voice interaction is realized through a highly decoupled instrumentalized architecture, a dynamic arrangement engine based on multi-model fusion and a collaborative execution and management mechanism.

The invention relates to the technical field of data processing and risk monitoring, and discloses a performance risk intelligent monitoring and collaborative response method and system, and the method comprises the steps: an edge sensing node calculates the compression characteristics of performance flow data through employing a mean aggregation algorithm according to a current sliding window parameter, and transmits the compression characteristics to a cloud end; the cloud decision-making platform executes data standardization management, calculates performance risk probability by using a model reasoning algorithm, and calculates feature uncertainty indexes by using a game theory interpretation method and a variance statistical method; a feedback control strategy module generates a self-adaptive sampling control instruction according to the index, drives an edge node to dynamically update a sliding window parameter, and constructs an acquisition feedback closed loop; and the collaborative response module calculates a response priority by utilizing multi-dimensional weighted logic, calculates desensitized data by utilizing a differential privacy algorithm and then executes response. According to the invention, adaptive adjustment of data acquisition granularity and safe cooperative processing of risk events are realized.

The invention discloses a cloud edge collaborative information interaction processing device, method and system with an endogenous security architecture. The equipment comprises a main processor, a hardware security chip, a programmable logic unit, a communication unit and an equipment management unit. A trusted execution environment is constructed in the hardware security chip and is in isolated communication with a common execution environment of the main processor through a security bus, a security computing engine, a remote attestation module and a trusted key warehouse are arranged in the trusted execution environment, and the programmable logic unit is directly controlled by the trusted execution environment to realize dynamic hardware acceleration. The communication unit is provided with independent interfaces which are connected with the two execution environments respectively, unification of hardware-level security isolation and efficient cooperative computing is achieved, and the problems of privacy disclosure and untrusted process in cloud edge collaboration are solved.

The invention relates to the technical field of image recognition, in particular to an intelligent solid waste recognition system based on multi-modal fusion, which comprises a data acquisition module, a visual guidance module, a recognition module and a cooperative processing module, and is characterized in that the system quickly analyzes surface information such as visible light images, three-dimensional point clouds and infrared spectrums of materials through the visual guidance module; and coding to generate a dynamic vector for guiding a subsequent refined analysis task. The identification module constructs a two-dimensional pixel map on the internal density map according to the guidance of the dynamic vectors, and utilizes a map neural network to deeply analyze local density textures so as to identify high-value components such as an internal circuit board; and the co-processing module performs template matching of the internal components and external surface anomaly detection according to the instruction so as to assess the security risk. According to the method, the external form and internal structure information of the material are deeply fused through the strategy of coarse first, fine second and dynamic guidance, and intelligent recognition of the complex electronic waste is achieved.

A power transformation equipment oil temperature prediction and abnormity early warning method based on a fuzzy algorithm comprises the following steps that real-time operation parameters and oil temperature historical data of power transformation equipment are collected, the data are preprocessed, and a high-quality input data set is constructed; constructing a fuzzy inference system considering multiple input factors, and setting a fuzzy membership function and an inference rule base; according to the fuzzy prediction result and the real-time oil temperature change, a self-adaptive dynamic safety threshold model is compared, a potential abnormal trend is identified, and automatic grading early warning is carried out according to the overtemperature grade; and corresponding exception type identifiers and disposal suggestions are generated and are pushed to the operation and maintenance platform through the communication module, so that remote operation and maintenance scheduling and response control are supported, and intelligent cooperative processing is realized. According to the invention, intelligent prediction of the oil temperature change trend under the influence of multiple factors is realized, the capability of sensing temperature rise abnormity in advance is improved, the accuracy of sensing the operation state of the equipment and the timeliness of early warning response are also remarkably improved, and the risk of equipment failure caused by the abnormal oil temperature is reduced.

The invention discloses a video recognition-based man-machine cooperative processing method for ditches and rivers in an urban area, and relates to the field of video recognition or understanding, and the method comprises the steps: constructing a data collection and analysis closed loop covering the whole periods before, during and after heavy rainfall: collecting images before and after treatment through an unmanned aerial vehicle and a camera before rainfall, generating a treatment operation influence map for quantitatively evaluating the treatment effect; in rainfall, the fixed camera is used for continuously monitoring the water level and the flow velocity; after rainfall, images are collected again, and the abnormal river reach is recognized. According to the method, camera data is taken as a boundary condition, a hydrological model is driven to perform inversion in combination with a governance influence map, missing heavy rainfall process data is deduced, and finally, accurate attribution of abnormal reasons is realized, and a differentiated governance scheme is generated. According to the method, the problem of monitoring data missing in severe weather is effectively solved, and quantitative evaluation of the treatment effect and scientific diagnosis of the abnormal cause of the river channel are realized.

The invention discloses a distributed cooperative processing method, system and equipment for mass spectrum data analysis, and relates to the field of biological information distributed processing, and the method comprises the steps: receiving original mass spectrum data submitted by a user, and analyzing flow configuration and global parameter configuration; analyzing the analysis process configuration into a directed acyclic graph taking the mass spectrum data processing step as a node; splitting the directed acyclic graph into a plurality of parallel atomic tasks according to the quantity of the original mass spectrum data and the space occupation amount; each atomic task is allocated to a Worker node storing corresponding associated data, and the atomic tasks are executed in parallel; when it is detected that the atomic tasks fail to be executed, the Worker nodes are distributed to the atomic tasks which fail to be executed again; different data storage systems are adopted to store original mass spectrum data, intermediate results and final analysis results. According to the invention, efficient parallel processing of mass spectrum data is realized, and continuity and reliability of an analysis process are guaranteed.

The invention discloses a log node double-robot collaborative processing system and a collaborative processing method thereof. The method comprises the steps that firstly, a log conveyed to a bearing platform is subjected to coarse positioning through a fixed vision unit; then a first vision unit at the tail end of the grabbing robot is driven to conduct multi-view-angle three-dimensional scanning on the log, a high-precision point cloud model is reconstructed and registered with a pre-stored digital model containing process information, and the grabbing pose and the initial machining path are synchronously planned; then the grabbing robot grabs and carries the logs to the cooperative position; secondly, a second vision unit at the tail end of the machining robot conducts secondary scanning on the log in the holding state, the pose deviation is calculated through matching, the initial machining path is compensated in real time, and a self-adaptive path is generated; and finally, the machining robot executes milling according to the compensation path. Full-process automation and self-adaptive high-precision machining from sensing, positioning, grabbing to machining are achieved, and the flexible machining problem of multi-scale logs is effectively solved.

The invention discloses a computing power scheduling method for multi-satellite collaborative application. The computing power scheduling method comprises the steps that 1, a ground station generates an undirected connection graph according to a satellite constellation topological structure and a resource state; step 2, the ground station receives a user instruction and carries out task computing power demand estimation; step 3, the ground station adopts a sliding window mode on the undirected connection graph to determine an execution area for executing the task; 4, calculating a central node of the execution area, and taking the central node as a source satellite for executing an information acquisition task; 5, executing an information acquisition task by the source satellite; step 6, completing task data distribution from the source satellite to other satellite nodes in the area by adopting a double-factor sorting scheduling strategy; 7, the satellite node returns a result to the source satellite after completing task processing; and step 8, the source satellite collects the result and sends the result to the ground station. The invention provides a strategy of multi-satellite cooperative processing and resource dynamic scheduling under the condition that the single-satellite capability is limited, and the processing time delay of on-satellite data is reduced.

The invention discloses a storage material checking system, belongs to the technical field of intelligent storage and Internet of Things, and is used for solving the problem that an existing storage material checking method is difficult to give consideration to dual requirements of efficiency and credibility. The system comprises a sensing layer which comprises an RFID reader-writer network and is used for collecting material label data in real time to obtain real-time sensing data; the block chain layer comprises a BID identity authentication node which is used for generating a unique block chain identity label for the Internet of Things equipment and realizing equipment identity verification, and also comprises an intelligent contract engine which is used for realizing business rule verification; the co-processing layer comprises a double-chain data verification module used for comparing the real-time sensing data with block chain historical evidence storage data and generating a verification state, and further comprises an exception handling unit used for performing hierarchical response under the condition that the verification state is abnormal; and the application layer comprises a WMS interface and is used for synchronizing the material checking result with the warehouse management system. According to the system, efficient, accurate and credible warehousing material checking is realized.

The invention discloses a multi-role collaborative transportation safety management and responsibility tracing method and system. The method comprises the following steps: S1, reporting an event, collecting travel data, identifying a security event through AI, and generating a unique event number for associated storage; s2, role and flow distribution: grading based on event information, determining participation roles by means of a built-in rule engine, and generating a processing flow and a task list containing logic association through a workflow engine; s3, multi-role cooperative processing and recording are carried out, the roles are operated according to a list, and the system automatically records the operation and forms a corresponding event processing chain; and S4, automatic responsibility tracing: based on the processing chain, generating a responsibility chain and a responsibility affiliation result through a hybrid mechanism. The system comprises a database, an event report and other functional modules for supporting the method. According to the invention, full-process closed-loop management from event grading and role allocation to cooperative processing and automatic tracing is realized, and the problems of low efficiency and difficult responsibility determination of multi-role cooperation are solved.

The invention discloses a cross-platform teaching tool data collaboration and personalized resource recommendation system based on an MCP protocol, and belongs to the technical field of education. According to the system, a multi-platform teaching tool is in unified butt joint through the MCP protocol adaptation layer, and the problem of data splitting is solved; the data co-processing layer cleans and fuses multi-tool data, generates a teacher / student / parent data view and stores an evidence through a block chain; the personalized recommendation engine generates role-based resource recommendation based on a teaching knowledge graph and multi-dimensional behavior analysis; the user interaction layer supports multi-terminal access, and the data security module guarantees compliance. According to the system, the cross-platform data collaboration efficiency is improved by 70% or above, the resource recommendation matching degree reaches 92%, the system is suitable for K12 education, higher education and vocational education scenes, teaching tool data barriers are effectively broken through, and the teaching data utilization rate and the resource recommendation accuracy are improved.

The invention discloses a multi-task fixation point estimation method, which relates to the technical field of computer vision and deep learning, and comprises the following steps of: firstly, constructing a data set for providing associated data from tasks to labels and supporting multi-task cooperative training, and then constructing an MTHGaze model, through a semi-structured framework, deep learning, a geometric constraint fitting model and key tasks of cooperative processing are fused, a loss function is designed and trained, defects of each task are solved through targeted loss function design, and a scientific training process enables the model to be iteratively optimized. Finally, the targets of improving the estimation precision of the fixation point, ensuring the physical rationality and enhancing the environmental robustness are achieved; and finally, performing fixation point mapping. According to the estimation method, through data set construction, data consistency can be guaranteed, training noise can be reduced, a real reference can be provided for loss function calculation, error back propagation can be realized, and Euclidean distance errors can be minimized through L2 norms of real fixation point coordinates and prediction coordinates.