130 results about "Structure diagram" patented technology

An evaluation system, method and program product based on a large language model relate to the technical field of artificial intelligence. The system obtains an evaluation object and an evaluation task configuration, constructs an evaluation structure graph containing theme segments, argument segments, basis segments and reference segments and their support, reference and conflict relations, filters key semantic units, and constructs a coupling representation between key semantic units, evaluation dimensions, dimension rules, candidate evidence segments, evidence positions and unit-evidence relationship markers; input the coupling representation into a large language model to generate an initial evaluation result; construct a statement disturbance set around the key semantic units that keeps the evaluation theme and core facts unchanged, obtain consistency indicators through re-evaluation, and calculate joint confidence by combining evidence coverage, evidence conflict and evidence chain drift; and perform evidence weighting and local calibration on low-confidence target evaluation dimensions. The scheme can improve the accuracy, stability, explainability and traceability of the evaluation result.

This invention belongs to the field of cashmere raw material sorting and processing technology, specifically disclosing a method for intelligent sorting and processing quality control of cashmere raw materials. The method includes: acquiring multi-dimensional sensing data of the cashmere raw materials to be processed, including fiber scale structure images, surface gloss images, and component spectral data; performing feature-level fusion processing on the multi-dimensional sensing data to generate quantitative feature parameters; inputting the quantitative feature parameters into a machine learning hierarchical decision model to generate sorting instructions; controlling a nozzle array actuator that combines negative pressure adsorption and positive pressure jetting to perform non-contact automatic sorting according to the sorting instructions; and collecting real-time process quality data of subsequent combing and spinning processes through an IoT platform, and dynamically adjusting the sorting threshold and raw material ratio strategy based on this data. This invention achieves quantitative evaluation and closed-loop quality control of cashmere raw material quality, improving the accuracy of sorting, the adaptability of the production system, and the quality stability of the final product.

The application provides a substation civil engineering large model training method and system integrating design specifications, and belongs to the field of substation civil engineering design. The method comprises the following steps: constructing a unified substation structure diagram; obtaining a substation civil engineering design specification text and parsing it into design rule embeddings, including trigger functions, constraint functions, rule weights, and repair templates; inputting the structure diagram into a multi-modal large model to generate an initial design output representation; performing a differentiable constraint projection process on the initial design output representation based on the trigger functions, constraint functions, and rule weights to obtain a projected design output representation; verifying the compliance of the projected design output representation with the design specifications based on a symbolic verifier, and generating corresponding rule violation counterexample information when the verification fails; performing a repair generation operation on the rule violation counterexample information based on a repair generator to obtain a repair candidate design, and backflowing the repair candidate design for large model training. The application can enable the design specifications to play a constraint role in the model generation and training stages.

This invention discloses a wellbore structure design method and apparatus based on the minimum business unit. The wellbore structure design method includes: obtaining the minimum business unit corresponding to the wellbore structure, wherein the minimum business unit includes configured data resource retrieval rules, tool resource invocation rules, and application scenario affiliation rules; retrieving the corresponding mapping component based on the tool resource invocation rules; extracting the corresponding dataset from a preset data lake based on the data resource retrieval rules; drawing a wellbore structure atlas in a preset drawing area using the mapping component based on the dataset, wherein the wellbore structure atlas includes at least well depth lines and annotations, a formation distribution map, a drilling fluid density distribution map, and a wellbore structure map; and pushing the wellbore structure atlas to a designated file library in the data lake based on the application scenario affiliation rules. This method can solve the problem of difficulty in sharing results and improve the efficiency of wellbore structure design.

This invention discloses a document-level debiased sentiment analysis method, system, and storage medium. The method includes: pre-constructing a structured causal model; extending the structured causal model by encoding comment text and aspect words, and calculating interaction node representations; forming a front-door path based on the aspect words through the interaction node representations; introducing shared confusion factor nodes, and generating causal structure edges for the shared confusion factor nodes, comment text, aspect words, and front-door paths; constructing a causal structure graph using the causal structure edges; transforming the causal structure graph into a heterogeneous graph neural network; calculating the total effect on sentiment using an intervention procedure on the heterogeneous graph neural network; analyzing the polarity input of the comment text, aspect words, and interaction node representations based on the structured causal model to obtain a task loss function; and converging the extended structured causal model based on the task loss function.

The application discloses a communication fault root cause intelligent diagnosis positioning system and method, and belongs to the technical field of communication, comprising: a collection module, a feature extraction module, an abnormality recognition module and a diagnosis module; the collection module is used for acquiring the topological structure diagram of a target network, and constructing a diagnosis data set; the feature extraction module is used for classifying and normalizing the diagnosis data set, combining feature extraction and topological association labeling, and constructing a diagnosis feature vector; the abnormality recognition module is used for performing multi-channel parallel abnormality recognition on the diagnosis feature vector, performing confidence filtering and association aggregation on the abnormality recognized by each channel, and constructing an abnormal event list; the diagnosis module is used for constructing a fault propagation graph, labeling abnormal events, tracing an abnormal propagation path along the fault propagation graph in a reverse direction through a neural network algorithm, generating a root cause candidate set, performing confidence scoring on the root cause candidate by calculating a deviation abnormality degree, and screening a root cause diagnosis result.

The application discloses a method and system for predicting protein drug binding sites based on a multi-modal dynamic graph, comprising: obtaining amino acid sequence data and three-dimensional structure data of a protein, and generating evolutionary conservation features, structure graph topology features and sequence features respectively; inputting the features into a multi-modal fusion encoder to generate first fusion features; inputting the first fusion features into a prediction decoder to generate initial prediction probabilities; iteratively updating the structure graph topology features according to a preset three-graph update rule according to the initial prediction probabilities and residue dynamic communication scores; re-inputting the updated structure graph topology features into the multi-modal fusion encoder and the prediction decoder, and repeatedly executing until a preset iteration number is reached, to generate final prediction probabilities; and generating a residue importance heat map and outputting a prediction report based on the final prediction probabilities through a gradient weighted class activation mapping algorithm. The application realizes the cooperative optimization of prediction and graph structure, and significantly improves the accuracy of binding site prediction.

The application discloses a kind of based on high-efficiency structure diagram clustering realizes the method for urban traffic network planning, first using the characteristics that core intersection node in high threshold under SCAN algorithm and collaborative traffic function area are necessarily contained in low threshold result, the traffic area division result under different intersection correlation degree threshold and traffic area scale threshold is pre-organized into a kind of urban traffic road network hierarchical index model;Second, in online query stage, by parameter analysis and index positioning processing to the intersection correlation degree threshold input by user, quickly locate to corresponding target traffic hierarchical tree;Subsequently, using the hierarchical relationship and threshold screening processing of target traffic hierarchical tree, replace traditional complex full road network traversal operation, quickly extract traffic subnet node set;Finally, based on the initial clustering member list extracted, accurately determine pivot node and outlier by label mapping and extraction processing in closed neighborhood structure, to realize the efficient clustering of large-scale urban road network data.

The application relates to the technical field of circumferential road paving topology reconstruction, and particularly discloses a circumferential road paving topology reconstruction method for a grounding grid drawing missing working condition, which injects a pulse sequence signal carrying a unique pseudo-random code feature, accurately extracts a target response from a strong interference background by using cross-correlation operation, and improves detection anti-interference capability; meanwhile, a conductor intersection point or a branch point is pre-judged by using external stray current information, and the requirement for detection equipment resolution is reduced; a preliminary conductor path diagram is generated by connecting path tracking coordinate points, and a topology node to be verified is identified based on an interference field intensity diagram; finally, a branch path is confirmed by multi-directional detection, and a complete grounding grid topology structure diagram is generated; the method aims to solve the problems of grounding grid drawing missing or inconsistency with the actual due to historical reasons or poor data management, and has high precision, high efficiency and high safety, and provides a solid data basis for grounding grid operation and maintenance.

The application provides an interactive video script generation method and device, electronic equipment and a storage medium. The application obtains a plot structure diagram corresponding to an interactive video, wherein the interactive video includes a plurality of video clips, and the plot structure diagram is used to describe the playing logic of playing the plurality of video clips. The plot structure diagram is subjected to image recognition processing to obtain a plurality of first image elements and a second image element used to represent the association relationship between the plurality of first image elements. The first element information corresponding to each first image element is determined, and the second element information corresponding to each second image element is determined. The script file corresponding to the interactive video is generated based on the first element information and the second element information, wherein the script file is used to record the playing logic. In this way, the production efficiency of the script file is improved.

The application provides a vehicle damage assessment method and system, electronic equipment and storage medium, which are applied to the technical field of computer vision and deep learning. The method comprises the following steps: acquiring accident images and vehicle identification information, and determining a corresponding component topology graph according to the identification information; extracting visual features of the images, and mapping the visual features to nodes of the topology graph to obtain initial node features; calculating dynamic weights of component connection edges based on the features, and constructing a dynamic accident topology graph reflecting mechanical conduction strength; performing cross-domain interaction between the dynamic accident topology graph and the visual features to obtain enhanced node features; predicting component damage states based on the features, and performing consistency verification on the results according to a preset physical logic constraint; and finally outputting a structured damage assessment result containing damage information and mechanical conduction logic. In this way, the physical topology structure and the mechanical conduction logic of the vehicle can be integrated into the deep learning network, and intelligent damage assessment with causal reasoning and structural consistency is realized.

The application discloses a three-dimensional human body posture estimation method based on dynamic modeling and bone topological consistency, multi-modal input construction and feature extraction, feature extraction is respectively performed on input RGB images, two-dimensional joint coordinates and skeleton structure diagrams, and each mode feature after extraction is uniformly mapped to the same dimension, thereby providing a basis for subsequent multi-modal fusion; by sequentially executing five core processes of multi-modal input construction and feature extraction, dynamic credibility modeling and adaptive fusion, bone topological guided repair, multi-scale detail enhancement and three-dimensional regression, and loss function joint optimization, the problem that a static strategy is adopted for multi-modal fusion of an existing three-dimensional human body posture estimation method and the mode contribution proportion cannot be dynamically adjusted according to a scene is effectively solved.

The application discloses a large language model-based visual large-screen intelligent generation method and system, relates to the field of visual processing, and comprises the following steps: receiving a visual large-screen generation demand instruction, calling a large language model to generate an initial large-screen item containing multiple visual graph elements; receiving a modification instruction, determining the physical coordinates, layer depth and logical association relationship of all visual graph elements on the current canvas, and converting the same into a spatial semantic tree; inputting the combination of the modification instruction and the spatial semantic tree into the large language model to obtain a differential operation instruction set; analyzing the differential operation instruction set and applying the same to the current visual large screen; receiving a graph element redrawing instruction, determining whether there is a spatial conflict between the visual graph elements in the current visual large screen, and adaptively adjusting the positions of the visual graph elements that have the spatial conflict to eliminate the spatial conflict. The application can effectively guarantee the stability of the layout structure, graph element position and hierarchical relationship during the generation of the visual large screen.

The present application relates to the technical field of database construction, and particularly relates to a construction method of a foraminifera morphological feature database, comprising the following steps: S1, collecting foraminifera images from multiple sources, identifying scale drift and shooting deviation based on principal axis normalization and local vorticity field, performing structure-preserving image enhancement, and outputting an enhanced image set consistent in shell substance, shell arrangement, shell opening, suture line and micro-pore features; S2, extracting a multi-scale shell structure atlas, constructing a structure-semantic coupled label nested tree, performing polysemy recognition and ambiguity resolution, and generating a structured annotation set; S3, fusing the image set and the annotation set, and constructing a three-layer database model containing a part layer, an attribute layer and a variation layer according to the label nested tree. Through image structure enhancement, semantic ambiguity resolution and three-layer database construction, the present application effectively improves the recognition accuracy of foraminifera morphological features, the label organization logicality and the information retrieval efficiency.

The application discloses a ship lock concrete structure durability analysis method and system under the coupling action of multiple factors, which comprises the following steps: acquiring internal structure images of different erosion areas of a ship lock concrete test piece; identifying and quantitatively analyzing the internal structure images of different erosion areas by a deep learning image segmentation algorithm to generate an erosion structure factor set; calculating environmental characteristics corresponding to multiple environmental parameters by a multi-physical field coupling inversion algorithm; constructing a three-dimensional microstructure model of different erosion areas; dynamically simulating the three-dimensional microstructure model according to the environmental characteristics; performing spatio-temporal correlation analysis on simulation result data by a hypergraph neural network to generate a prediction result; and comparing an expert knowledge base and performing construction maintenance based on text information of the prediction result by an Euclidean distance similarity algorithm. The application introduces high-precision three-dimensional model multi-factor coupling simulation, and improves the ship lock concrete structure analysis precision through high-precision coupling prediction of the hypergraph neural network.

The application provides an artificial intelligence-based dialogue text generation method and system. The method comprises the following steps: obtaining a predicate and an argument in a historical dialogue sentence and adding a semantic label; taking the argument and the predicate as nodes to construct a dialogue structure graph, generating an adjacent correlation vector corresponding to the nodes in the dialogue structure graph, and then obtaining a fusion vector corresponding to the nodes to form an enhanced dialogue sentence corresponding to the historical dialogue sentence; and obtaining a latest reply text based on the enhanced dialogue sentence, the semantic label and a self-recurrence language model. The semantic label provides a basis for role recognition for subsequent text generation, avoiding the occurrence of role misplacement problems. The latest reply text can dynamically focus on important parts in the enhanced dialogue sentence, and the enhanced dialogue sentence has fused graph information, so that the generation of the latest reply text has strong dynamic adaptability and deep understanding of the structure and logic of multi-round dialogue.

The present invention belongs to the interdisciplinary technical field of medical image processing and artificial intelligence. Disclosed is a medical image reconstruction method based on deep learning. The method comprises: acquiring a medical image data set, and labeling corresponding key anatomical structure information; on the basis of a feature encoder and a graph structure processor, constructing a medical image reconstruction model; the feature encoder performing feature extraction on the medical image data set by means of a convolutional neural network, so as to obtain a multi-scale feature map; the graph structure processor constructing an initial landmark point set of the multi-scale feature map by means of a graph attention network, and analyzing topological relationships between landmark points, so as to obtain an optimized landmark point arrangement; on the basis of the optimized landmark point arrangement, constructing an anatomical structure graph; on the basis of the anatomical structure graph, designing an adaptive loss function to train the medical image reconstruction model; and on the basis of the trained medical image reconstruction model, obtaining a corresponding reconstructed image. The present invention achieves the objective of improving the detail representation of a reconstructed image while maintaining the accuracy of an anatomical structure.

The application discloses a kind of micro-light image enhancement methods, including obtaining target micro-light image;Based on gamma correction, RTV correction and histogram matching scheme, the structure diagram and texture diagram of target micro-light image are calculated;According to the structure diagram and texture diagram obtained, based on Retinex model and structure / texture guide scheme, the brightness diagram and reflection diagram of target micro-light image are calculated;The obtained brightness diagram is adjusted and adaptively corrected;Based on RTV correction scheme, the enhanced image of target micro-light image is calculated to complete the image enhancement of target micro-light image.The application also discloses a kind of mine image enhancement method comprising the micro-light image enhancement method.The application not only realizes the enhancement of micro-light image and mine image, but also has higher reliability, better accuracy and better enhancement effect.

The application discloses a robot synchronous semantic positioning and mapping method and system based on a human-simulated thought, which comprises the following steps: acquiring environment pictures and spatial structure point cloud data of each frame; recognizing the environment pictures to obtain observed straight lines of positioning road marks; performing cluster division on all the observed straight lines, and dividing the observed straight lines belonging to the same positioning road mark into a cluster; analyzing the observed straight lines in each cluster by means of a least square method to determine the spatial coordinates of each positioning road mark; optimizing the spatial coordinates of the positioning road marks to obtain optimized coordinates of the positioning road marks; obtaining a positioning road mark map according to the optimized coordinates of all the positioning road marks; projecting each frame of point cloud data on a local path to a horizontal direction to generate a line segment map; associating all the line segment maps to obtain an environment structure map; and combining the positioning road mark map and the environment structure map to form a semantic map. The robot synchronous semantic positioning and mapping are realized.

The application provides a house source three-dimensional modeling method, equipment and a storage medium, wherein the method comprises using a camera to shoot a first video of the interior of a house, and splitting the first video into multiple indoor images. The indoor image set is subjected to semantic segmentation, and movable obstacles in the indoor image set are removed according to K labels. The active structure in the house interior structure image set is marked according to the K labels, and the two sides of the active structure are subjected to connected component analysis. The house interior structure image set is converted into M three-dimensional point clouds, and the M three-dimensional point clouds are reconstructed based on the spatial connection relationship. N reconstructed point clouds are subjected to Poisson reconstruction, and the interior three-dimensional model of the house source is obtained. The application adopts the semantic segmentation mode to exclude the influence of furniture and other movable obstacles on the house structure, uses the active structure to improve the accuracy of the spatial relationship between different regions in modeling, and represents the space separated by the active structure, so that the house is comprehensively and accurately three-dimensionally modeled.

This invention relates to an automatic generation method and system for control models in PSCAD simulation, belonging to the field of electromagnetic transient simulation and intelligent modeling technology for power systems. This method aims to solve the problems of low efficiency and error-proneness in manual modeling. Its technical solution includes: acquiring a transfer function system structure image and using a pre-trained deep learning model for detection; reading the type, coordinates, and parameter information of components in the image; automatically constructing the directed signal flow network relationship of the system; automatically generating Fortran code segments and model appearance based on this; compiling the code into a dynamic link library; integrating all generated content into a configuration file conforming to PSCAD specifications; and automatically generating a Web service interface providing upload and download functions. This method achieves end-to-end automatic generation from structure diagrams to simulation models, significantly improving modeling efficiency and accuracy, ensuring model consistency and reliability, and possessing strong engineering practicality.

The disclosure provides a layered material identification detection method for oil painting restoration process, which comprises multi-scale imaging and local micro-area section detection of the oil painting surface to generate a pollution deposition space distribution map; based on the pollution deposition space distribution map, an interface continuity expression of the smoke deposition layer and the material layer is constructed to distinguish the penetrating diffusion structure, the interface adhesion structure and the transition mixed structure; based on the interface continuity expression, the penetration path of the pollution deposition to the varnish and pigment interface is identified to generate a pollution penetration path model; based on the pollution penetration path model, layered discrimination of the pollution pseudo-layer and the real restoration layer is performed to generate an oil painting real material layer sequence structure diagram and output a layered identification result, which can accurately distinguish the smoke deposition pseudo-layer and the real material layer, restore the real layer sequence structure of the oil painting, and improve the accuracy of layered judgment in the oil painting restoration process.

1. The name of the design product: cabinet lifting parts cover box. 2. The use of the design product: used as a cover box installed on the cabinet as a lifting part. 3. The design points of the design product: in shape. 4. The picture or photo that best shows the design points: perspective view. 5. Other circumstances that need to be explained: the reference drawing is the internal structure diagram of the product after removing the side cover.

The application discloses a kind of power equipment fault identification method, system and equipment, belong to computer vision and artificial intelligence technical field, the method is: to the original structure image of power equipment is respectively carried out space standardization processing, obtain target structure image, and target structure image is carried out structure division processing, obtain component area label chart;Target structure image is input into the hybrid coding network of preestablished, obtain semantic feature map, combine semantic feature map and component area label chart to obtain multiple structure tokens, wherein, structure token and the component area of power equipment one-to-one correspond;Structure token is fused with preestablished task prompt vector, obtain query vector, and query vector is input into decoder, under the guidance of semantic feature map, carries out fault identification, obtains fault identification result, therefore, by implementing the application, the accuracy of fault identification can be improved.

This invention provides a method and system for detecting and visualizing cross-border data transmission. The method includes: parsing an input network data source to extract all contained IP addresses; querying the location information of the extracted IP addresses and, based on preset location rules, filtering out cross-border IP addresses from the extracted IP addresses; based on the filtered cross-border IP addresses, extracting data units from the network data source whose source or destination IP address contains any cross-border IP address, and saving the data units as independent cross-border data transmission record files; constructing a topology diagram reflecting the communication relationship between domestic and cross-border entities based on the data flow information in the network data source and the cross-border IP addresses, and then visualizing and interacting with the diagram. This method, by integrating automated IP location determination, intelligent cross-border data segmentation, and interactive topology visualization, effectively overcomes the shortcomings of traditional manual analysis methods and achieves significant technological progress.

The application discloses a kind of self-media promotion content heat prediction methods based on deep learning, comprising: S1, acquisition multi-modal data and pre-process output text word vector sequence and standard size image;S2, construct semantic structure atlas and output text semantic feature vector;S3, carry out saliency target detection and output visual saliency feature map;S4, feature aggregation and interaction are carried out using MLP-Mixer model, and output graphic-text coupling feature vector;S5, construct improved DCCA network, introduce covariance reconstruction loss function to generate psychological resonance index feature;S6, carry out time series evolution prediction and output heat prediction result.The application realizes graphic-text feature deep coupling and audience psychological resonance quantization, effectively improves the accuracy and explainability of heat prediction.

This invention discloses a method for constructing a dynamic question bank for safety training based on big data analysis, relating to the field of safety data processing technology. The method includes: acquiring safety event text data and accident context label data; the safety event text data includes multi-role composite behavioral narrative text and implicit causal sentence chain text; the accident context label data includes role category labels, spatial scene labels, and consequence risk labels; based on the safety event text data and accident context label data, constructing a syntactic dependency graph and a temporal logic graph, generating a nested event structure graph, and extracting role behavior semantic chains from the nested event structure graph; and constructing a causal path graph based on the role behavior semantic chains. This invention achieves the coordinated and unified structured extraction of multi-dimensional risk elements in composite safety scenarios and the construction of a dynamic question bank without relying on manual rule setting and while maintaining semantic integrity, by constructing a standardized expression system for risk factors.