246 results about "Scene matching" patented technology

The invention discloses a multi-scene term translation model training method and processing method based on a large model, and the method comprises the steps: firstly constructing a term-scene-translation triple data set, and formatting the term-scene-translation triple data set into a training sample containing an instruction, context input and expected output through a structured term translation instruction set; then, the samples are used for fine tuning of the basic large language model, and a preliminary adaptive model is obtained. And finally, introducing a reinforcement learning method based on a verifiable reward mechanism to optimize the fine tuning model, and iteratively improving the performance of the model in the aspects of scene matching degree, specification conformity, cross-scene consistency and the like by calculating a reward value. According to the method, the problems that a traditional term translation model is poor in flexibility and insufficient in specialty are effectively solved, and the multi-scene term translation which is high in accuracy and conforms to specific industry specifications can be automatically generated.

The invention relates to the technical field of dangerous cargo management, in particular to an intelligent compliance verification system for dangerous cargo boxing operation. Comprising a data acquisition preprocessing unit; the law and regulation graph construction iteration unit is used for extracting a'law and regulation term number-verification dimension-constraint condition 'triple of the law and regulation in the dangerous cargo boxing field and introducing an improved three-dimensional dynamic weight conflict resolution algorithm according to the efficacy level, the release subject and the application range indexing; a multi-agent collaborative verification unit; a risk early warning report generation unit; and a man-machine cooperation closed loop unit. According to the method, the problems of lagging of laws and regulations and clause conflicts are solved by adopting triple extraction of'laws and regulations clause numbers-verification dimensions-constraint conditions', a three-dimensional dynamic weight conflict resolution algorithm based on laws and regulations effectiveness-scene matching degree-release time and a technical means of periodically optimizing weight parameters through a reinforcement learning model.

The invention discloses a method for monitoring and identifying dangerous behaviors of a smart park, and relates to the field of intelligent security and protection. The method comprises the following steps: acquiring an environment data stream through a sensing device deployed at a key point location of a park and carrying out standardization processing on the environment data stream; the data stream is sent to an edge computing node, a target is recognized through a lightweight target detection model, the detection confidence coefficient is obtained, a motion track is generated through a multi-target tracking algorithm, the rule matching degree is calculated according to a predefined rule base, and dangerous behaviors are preliminarily recognized; calculating a comprehensive model prediction credibility based on the detection credibility and the rule matching degree, and generating a final comprehensive credibility through a multi-dimensional credibility fusion model in combination with the time sequence consistency credibility and the scene matching credibility; and dangerous behavior trend prediction is carried out according to the comprehensive confidence, and an alarm instruction is generated to trigger disposal linkage, so that advanced prevention and quick response of dangerous behaviors are realized, and the intelligent level and operation and maintenance efficiency of park security and protection are greatly improved.

The invention provides a business travel data management method and system based on cloud computing and scene matching and computing equipment. The method comprises the following steps: acquiring at least one geographic image uploaded by a user in a business travel plan execution process; generating a compliance geographic region set according to the travel plan; matching the geographic image with the compliance geographic region set to obtain a first matching result; identifying and extracting visual scene features in the geographic image, and performing consistency cross validation comparison on the visual scene features and a geographic scene expected based on geographic coordinates to obtain a first comparison result; generating an authenticity verification result of the travel behavior based on the first matching result and the first comparison result; and when the authenticity verification result is that the verification is passed, automatically generating an electronic audit voucher containing a spatial trajectory chain and a geographic image. According to the technical scheme of the invention, the accuracy of authenticity verification of travel behaviors can be improved, a high-credibility electronic audit voucher is automatically generated, and a data closed loop is provided for intelligent travel management.

The invention relates to the technical field of information extraction, and discloses a long text news event element extraction method and system based on multi-agent collaboration, and the method comprises the following steps: constructing a large language model multi-agent system architecture; the central coordination agent receives the unstructured news text; the text segmentation agent outputs a logic unit and time description thereof; an event extraction agent extracts an event subject, an action and an object, and complements time and entity role information; the scene matching agent judges the fine-grained news scene category to which the scene matching agent belongs; the information extraction agent generates a structured event record; and the central coordination agent performs quality auditing and exception handling on each step of output to ensure the integrity and consistency of the output data. According to the method, the static relationship between the event main bodies can be constructed according to the structured event information through the relationship construction and context sorting agent, the event situation change is summarized based on the timeline, and the feasibility and the visualization of news event analysis of the system are improved.

The invention belongs to the technical field of data processing, and particularly discloses a computer vision model service scenarized configuration generation method, which comprises the following steps: constructing a basic information database comprising a plurality of computer vision models, the database being used for storing model identification information and feature parameters; extracting structured features of the scene according to a scene analysis engine, wherein the structured features are used for describing operation conditions of the scene; calculating a model matching score according to the structured features and the feature parameters, and obtaining a recommendation model through a multi-condition weight matching function; executing a model selection strategy based on the matching score, and obtaining a model matched with the scene; and generating a configuration file conforming to a predetermined specification according to the recommendation model. The objective of the invention is to solve the problems of low efficiency and error proneness caused by the fact that a computer vision model in the prior art depends on a manual selection model and cannot quickly select and adapt to an optimal model in diversified scenes.

The invention discloses an intelligent storage automatic control optimization method and system based on dynamic environment perception, and relates to the technical field of intelligent storage control. According to the method, a specific scene type of a specified warehouse is firstly obtained, a corresponding dynamic perception allocation scheme is determined through scene matching, and then a collaborative operation association object is determined, so that accurate adaptation of dynamic environment perception and a warehouse scene is realized; then, collecting dynamic environment sensing data of a storage area where a collaborative operation association object is located, and after collaborative association processing is carried out, generating a structured fusion data set adapted to different time sequence scenes; dynamic storage parameter calibration is carried out, an initial control strategy is generated, and effectiveness evaluation is completed; whether the path planning scheme and the operation execution sequence are dynamically adjusted or not is decided according to the evaluation result, dynamic adaptation and accurate regulation and control of intelligent storage automatic control are achieved, and the adaptability of storage control, the smoothness of process operation and the stability of operation execution in a complex dynamic scene are effectively improved.

The invention discloses a spaceborne SAR scene matching curve imaging electromechanical cooperative wave control optimization method based on dynamic programming, and belongs to the technical field of spaceborne SAR. The method comprises the following steps: firstly, acquiring a beam pointing attitude angle time sequence before decomposition by using a satellite-borne SAR non-along-track multi-target imaging satellite-ground configuration joint design and optimization method, and solving a phased array electric scanning angle; secondly, establishing an electromechanical collaborative dynamic planning model comprising a platform mechanical scanning state transition equation, a cost function and a constraint function; initializing parameters such as a mechanical scanning state track, a control input sequence and the like to obtain an initial track and cost; optimal control solution is executed through dynamic planning, and convergence is achieved through inner and outer layer iteration (backward transfer calculation of a control law, forward transfer updating of a track and outer layer correction constraint violation); and finally, outputting a mechanical scanning attitude angle time sequence and a phased array electric scanning angle time sequence, and completing electromechanical cooperative wave control optimization.

The invention discloses a power industry construction unit safety management and control dynamic evaluation method and system, and the method comprises the steps: constructing a full-scene safety standard knowledge base covering power transmission and transformation and power distribution scenes, and building cross-scene semantic association through a knowledge graph; real-time resource information of a construction unit is obtained, and a basic element parameter model is constructed and dynamically updated through a multi-granularity strategy; based on a project scene matching standard security element set, evaluating the security risk of the whole process through a deep Q network algorithm fusing transfer learning and reinforcement learning, and generating a quantitative index and a resource configuration scheme; and generating project-level, process-level and operation-level three-level management and control instructions and pushing the instructions to the terminal. According to the method, scene self-adaption, resource state real-time mapping and risk disposal closed-loop feedback of safety assessment are realized, and the risk identification timeliness, the resource configuration accuracy and the field instruction executable performance are improved.

The invention discloses a data processing method and device in agent dialogue, equipment, a medium and a product, and relates to the technical field of big data. According to the method, the user attribute, the service rule, the scene information and other structured knowledge are extracted in response to the dialogue data and stored in the dynamic cache layer, and the timeliness tag is added, so that the problem of dialogue memory loss of an existing agent dialogue system is effectively solved, and the interaction efficiency and the user experience are improved. Then, target structured knowledge is screened in combination with dialogue relevance and business relevance and then classified and stored according to a three-level hierarchical storage architecture, and then the value of the memory unit is evaluated based on timeliness, scene matching degree and cross-session relevance strength and retrieved to generate response data. Therefore, physical isolation of user features and business knowledge is realized through three-level hierarchical storage, mutual interference of different types of data during retrieval is reduced, retrieval accuracy and efficiency are improved, and timely response of an intelligent agent dialogue system to user requirements is guaranteed.

The invention relates to a scene matching navigation adaptation area selection method based on self-supervised learning, which comprises the following steps of: firstly, dividing an obtained remote sensing image into a plurality of sub-images, and constructing a similarity degree between self-supervised quantity representation sub-images by utilizing a difference value between the remote sensing image sub-images and enhanced sub-images of the remote sensing image sub-images; secondly, calculating the proportion of stable ground features in the remote sensing image sub-graph, and taking the proportion of the stable ground features as the weight of the self-supervision quantity; carrying out self-supervised learning on the network through the similarity between the images, and carrying out self-supervised training by using a ResNet network model, so that the model can evaluate the similarity of the images; and finally, the trained network model is utilized to evaluate the similarity between the sub-graphs of the candidate adaptation area, the suitability of the candidate adaptation area is represented through similarity evaluation, and the area with the best suitability is obtained to serve as the adaptation area.

The invention relates to the technical field of robot design, and provides a robot design method and system based on agent collaboration. The method comprises the steps that a design agent generates a plurality of initial robot models based on robot demand information of a user; the simulation agent constructs a virtual scene matched with the robot demand information; a verification agent performs performance verification on each initial robot model in the virtual scene, and feeds back abnormal event information generated in the verification process of each initial robot model to the design agent; the design agent generates a target robot model based on the abnormal event information corresponding to each initial robot model; and the deployment agent generates manufacturing information of the entity robot required by the user based on the target robot model, guides production of the entity robot based on the manufacturing information, and continuously monitors the entity robot after the entity robot runs. The method is helpful for improving the design efficiency of the robot.

This invention relates to the field of information extraction technology, and discloses a method and system for extracting elements from long-text news events based on multi-agent collaboration. The method includes the following steps: constructing a large language model multi-agent system architecture; a central coordinating agent receiving unstructured news text; a text segmentation agent outputting logical units and their time descriptions; an event extraction agent extracting the event subject, action, and object, and completing time and entity role information; a scene matching agent determining the fine-grained news scene category to which it belongs; an information extraction agent generating structured event records; and a central coordinating agent performing quality review and anomaly handling on each step of the output to ensure the integrity and consistency of the output data. This invention, through relationship building and contextual analysis agents, can construct static relationships between event subjects based on structured event information, and summarize event situation changes based on a timeline, improving the feasibility and visualization of the system's news event analysis.

The application discloses a multi-modal data fusion vehicle-mounted T-BOX communication optimization method and system, relates to the technical field of data processing, and comprises the following steps: performing credibility evaluation, constructing a scene-credibility comparison table, and acquiring an adaptive credibility sequence according to a current traffic scene; acquiring a multi-source sensing data sequence, performing key feature extraction, and acquiring a key sensing data sequence set; performing fluctuation analysis on the key sensing data sequence set, determining a data transmission type according to an analysis result and the adaptive credibility sequence; extracting the key sensing data sequence set according to the data transmission type, acquiring a key transmission data sequence set, and performing communication transmission to a cloud server for recommended speed prediction and then downlinking to a vehicle-mounted T-BOX. The application solves the technical problem that the prior art cannot dynamically adjust a vehicle-mounted data communication strategy according to a traffic scene and sensor data fluctuation characteristics, and achieves the technical effect of improving the effectiveness of T-BOX communication data.

The application provides an application window switching method and device, a storage medium and an electronic device, and relates to the technical field of computers. The method comprises the following steps: in the case that a first application scene changes into a second application scene, obtaining a target application window matched with the second application scene which is created by a user on a target level; adding the target application window and an original application window to the same image rendering engine; if a target condition is met, copying image data in the original application window to the target application window, and rendering the original application window and the target application window by using the image rendering engine; and closing the original application window after the image data of the target application window and the original application window is displayed on a target screen for a first time. The application can reduce the phenomenon of flashing screens and improve the stability of pictures.

The invention discloses a switch cabinet abnormal behavior processing method, device and equipment, a medium and a product. The method comprises the following steps: acquiring real-time monitoring parameter data of a switch cabinet to be detected, and determining a parameter change rate according to the real-time monitoring parameter data; according to the real-time monitoring parameter data and the parameter change rate, judging whether the to-be-detected switch cabinet meets a preset abnormal threshold judgment condition or not; if yes, according to the real-time monitoring parameter data and the parameter change rate, whether a target abnormal matching rule meeting a corresponding triggering condition exists or not is judged based on the triggering condition in a pre-constructed abnormal scene matching library; the abnormal scene matching library stores at least one reference abnormal matching rule and an association relationship between the triggering condition and the execution action corresponding to the reference abnormal matching rule; if yes, determining the rule number of the target abnormal matching rules; and if the number of the rules is only one, executing an execution action corresponding to the target abnormal matching rule so as to solve the abnormal behavior of the switch cabinet to be detected.

The embodiment of the invention provides a humanoid robot self-inspection method and related equipment, and belongs to the technical field of robot control. The method comprises the steps that the power-on condition of the humanoid robot is monitored, and if it is determined that the humanoid robot is powered on, an environment signal and an equipment interaction signal are obtained; determining whether a current application scene is an industrial scene or a household scene through a scene recognition model according to the environment signal and the equipment interaction signal; if it is detected that a preset scene triggering condition is met, starting a scene triggering self-checking task set matched with the current application scene; if a task starting instruction input by a user is detected, starting a pre-task precise self-inspection task set matched with the current application scene; wherein the user perceptibility of the scene triggering self-checking task set is weaker than the user perceptibility of the precise self-checking task set before the task. According to the embodiment of the invention, the scene adaptability, the operation reliability and the user experience of the humanoid robot can be improved.

The invention relates to an anti-highlight adjusting system of a high-altitude operation cab and a control method of the anti-highlight adjusting system. The adjusting system comprises a light environment sensing module, a scene recognition module, a partitioned dimmable glass assembly, a central control module and a feedback correction module. The control method comprises the following steps: S1, initializing a system; s2, scene matching; s3, carrying out partitioned dimming; and S4, performing dynamic correction. The system does not need manual intervention, can adapt to multi-scene strong light changes in real time, does not shield the operation view, is safe and reliable in design, is high in automation degree, is high in universality, can achieve batch production, can effectively reduce the labor intensity, and improves the operation safety and production efficiency.

The invention discloses a voice recognition method and device, a storage medium and computer equipment, relates to the technical field of voice recognition, and can be applied to the field of digital medical treatment and finance. The main purpose is to solve the problem of low speech recognition efficiency. The method mainly comprises the steps of obtaining to-be-recognized voice and interaction scene information of the to-be-recognized voice; recognizing the voice to be recognized to obtain an initial conversion text and a multi-dimensional confidence feature of the initial conversion text; using the interaction context information as a scene matching basis, calculating a text credibility parameter of the initial conversion text together with the multi-dimensional confidence feature, and determining a target path for executing a text error correction task from three-level routing according to the text credibility parameter, an interaction category and an error correction routing strategy; and performing text error correction processing on the initial conversion text based on the target path to generate a target conversion text of the to-be-recognized voice. The method is mainly used for text error correction based on dynamic routing so as to improve the speech recognition efficiency.

This application provides a screenshot method, apparatus, device, and storage medium. The screenshot method can reconstruct a three-dimensional game scene matching the game screenshot captured by the player in the screenshot editing interface. It supports the player to adjust the reconstructed three-dimensional game scene, thereby generating a target screenshot matching the adjusted three-dimensional game scene. This allows the player to directly modify the actual image content of the game screenshot, effectively improving the fault tolerance of the game screenshot function and the editing freedom of the game screenshot.

This invention discloses a USV berthing and unberthing pose estimation method based on ship-shore LiDAR collaboration, comprising: acquiring shipborne and shore-based point clouds and shore-based fixed reference poses; performing odometer calculations on continuous shipborne point clouds to generate LiDAR odometer factors; constructing a global descriptor for the ship-shore point clouds and performing scene matching to generate candidate ship-shore point cloud pairs; extracting key points and calculating FPFH descriptors, solving for high-precision ship-shore relative poses through coarse / fine registration, obtaining the final ship-shore relative poses through asynchronous reconstruction and generating correlation factors; constructing a collaborative factor graph model containing odometer and ship-shore correlation factors; solving the model and minimizing the overall residuals to obtain a globally consistent optimal pose, thereby achieving berthing and unberthing control. This invention improves ship-shore data transmission efficiency and matching accuracy through global descriptor matching and FPFH feature compression; and eliminates the cumulative error of single-ship odometers by leveraging shore-based global pose constraints, achieving high-precision continuous pose estimation under long-term navigation and ensuring the robustness of berthing and unberthing navigation.

The application provides a camera parameter configuration method and an electronic device, relates to the terminal technical field, and improves the accuracy of enabling an out-picture mode. The specific scheme is that a first configuration file is stored in the electronic device, the first configuration file includes a corresponding relationship between first scene information and a first screening condition, and the first screening condition is a condition for screening the out-picture mode. When the electronic device is started, the first configuration file is parsed and loaded. In response to an operation of indicating opening of a first application, a first interface is displayed, the first interface is a shooting preview interface of a first camera mode. When the first scene information matches a shooting scene of enabling the first camera mode, the first screening condition is acquired. Configuration parameters corresponding to a first out-picture mode matched with the first screening condition are sent to a camera sensor, so that the camera sensor collects a first image frame in the first out-picture mode. In the first interface, the first image frame is displayed.

This invention discloses a photovoltaic power generation prediction method based on multidimensional data analysis, specifically relating to the field of new energy power prediction technology. The method includes: multi-source data fusion and field construction: integrating multi-source heterogeneous meteorological data, and generating a spatiotemporally continuous comprehensive meteorological influence factor field through spatiotemporal alignment and dynamic weight fusion; adaptive scene matching prediction: performing multidimensional similarity matching between the current meteorological field sequence and a historical scene database, retrieving similar historical scenes, adaptively selecting and fine-tuning the prediction model online, and generating a preliminary power prediction sequence; and quantified uncertainty output: quantifying three types of uncertainty—input data, model selection, and real-time deviation—and fusing them using a Bayesian model averaging method. This invention improves the quality of prediction input data, enhances the adaptive prediction capability for complex weather, and provides prediction results with quantifiable uncertainty, effectively improving the accuracy and practicality of photovoltaic power prediction.

The application provides a method and an electronic device for vertical search, the method comprising: configuring different search templates for different scenes in advance, that is, configuring search rules for different scenes in advance, so that when a search word is received, a pre-configured search template matching the scene of the search word can be found, and a DSL statement is packaged based on the matching version and a search engine associated with the search engine is called to search. The scheme can meet different search requirements of different scenes in the whole search process, and the pre-configured search template is more conducive to subsequent updating and maintenance, and compared with a traditional scheme, the development time is shortened and the development cost is saved.

The invention discloses a self-adaptive interpolation restoration device and method for automatic driving continuous time-space data transmission, belongs to the technical field of automatic driving data transmission and processing, and solves the problems of high bandwidth occupation, insufficient restoration precision, poor scene adaptability and the like of an existing transmission scheme. The device comprises a data acquisition module, a vehicle end edge calculation module, a self-adaptive interpolation reduction module and an error verification module. The method comprises the steps that a vehicle end collects multiple types of spatio-temporal data and adds identifiers, dynamic characteristic parameters are extracted to divide low, medium and high dynamic driving scenes, a starting and ending point interval is adaptively configured, data are packaged in a customized mode, a cloud end restores the data according to a scene matching interpolation algorithm, error verification is conducted through original spatio-temporal data cached by the vehicle end, and the data are stored in the cloud end. If the threshold value is exceeded, triggering supplementary transmission and reducing the interval between the starting point and the ending point. According to the method, the data transmission quantity and bandwidth occupation are greatly reduced, the restoration precision and the data reliability are improved, the vehicle end energy consumption is reduced, and the real-time data interaction requirements of all working conditions of automatic driving are met.

The invention relates to a clothing hanging line push rod protection method and system and a storage medium, and the method comprises the steps: obtaining a feature signal of a data collection module in real time; performing feature extraction on the feature signal to obtain a multi-dimensional feature of the feature signal; the multi-dimensional features comprise traditional basic features, harmonic features, transient impact features and dynamic trend features; determining a scene matching degree based on the multi-dimensional features and a preset feature weight value; taking the scene with the highest scene matching degree as a target working scene; determining a protection strategy corresponding to the target working scene based on a preset scene strategy mapping table; and controlling the push rod assembly to execute the protection strategy. According to the method and the device, the problem that complex scenes cannot be effectively distinguished in the prior art is solved, and the complex scenes can be effectively distinguished under the real working condition.

The invention relates to the technical field of photovoltaic operation and maintenance management, and discloses an operation and maintenance management system for a new energy power station, and the system comprises a monitoring unit which is used for setting a plurality of monitoring sub-regions, and arranging a monitoring module on an equipment point for collecting the monitoring data of each equipment point; the central control unit is used for establishing an operation and maintenance model and generating the deterioration speed of each equipment point; the central control unit comprises a first processing module which is used for setting a plurality of working scenes and matching each monitoring sub-region with the working scenes; the second processing module is used for generating an operation abnormal value according to the working scene and the monitoring data and judging the working state of the equipment point; the third processing module is used for setting a preset image library and generating anchor point similarity; setting a comparison time period according to the anchor point similarity; generating a similarity number sequence in the comparison time period, and calculating a deterioration speed based on the similarity number sequence; and the maintenance unit is used for dividing maintenance priorities of the monitoring sub-regions according to the working states and the deterioration speeds of the equipment points, and determining a maintenance path according to the priorities.

The invention relates to the technical field of vehicle thermal management control, and discloses a thermal management assembly energy efficiency optimization system driven by vehicle networking data. The system comprises a data acquisition module, an energy efficiency analysis module, a communication interface module, a cloud strategy module and a vehicle-mounted control module. The data acquisition module initializes an energy efficiency monitoring environment and continuously acquires an operation parameter sequence; the energy efficiency analysis module compares the parameter sequence with an energy efficiency reference library to generate an energy efficiency deviation index, and extracts a current scene feature vector when the index exceeds a dynamic threshold value; the communication interface module packages the feature vector into a data packet and sends the data packet to the cloud; the cloud strategy module starts a strategy derivation engine after receiving the data packet, and generates an optimization instruction sequence based on scene matching; the vehicle-mounted control module receives and executes the instruction sequence to adjust thermal management assembly parameters. According to the invention, online monitoring, intelligent diagnosis and cloud collaborative optimization of the energy efficiency of the thermal management assembly are realized, and the energy utilization efficiency of the system under different working conditions is improved.