90 results about "Database graph" patented technology

The invention belongs to the field of fabric retrieval methods, and relates to a fabric image retrieval method in a complex scene. The method comprises the following steps: establishing a fabric image database; constructing a manual feature extraction model, and obtaining manual feature descriptions of the to-be-queried image and the image in the database image; building an image depth parallel representation model to process the fabric image, and obtaining depth feature descriptions of the image to be queried and the image in the database image; designing an image manual and depth feature fusion mode, fusing the obtained image manual and depth features, and comprehensively representing a fabric image; carrying out similarity measurement on the obtained fusion features of the image to be queried and the image in the database image, and outputting a retrieval result according to the similarity; and calling out fabric product details corresponding to the retrieval result image for guiding fabric production or e-commerce shopping recommendation. The method has high retrieval accuracy and robustness, and has great application potential in complex scenes such as industrial manufacturing and e-commerce platforms.

The invention discloses an image retrieval method and a medium, and the method comprises the following steps: obtaining a to-be-retrieved image, and carrying out the feature extraction of the to-be-retrieved image, so as to obtain a depth feature map corresponding to the to-be-retrieved image; performing weighting processing on each channel of the depth feature map to generate a weighted feature map; extracting local features and global features of the weighted feature map, and performing feature fusion based on the local features and the global features to generate fused features; calculating a to-be-retrieved hash code corresponding to the fusion feature based on a hash-guided measurement loss method; and calculating the similarity between the to-be-retrieved image and the database image based on the to-be-retrieved hash code and the database image hash code, and determining a target image according to the similarity. Loss of important semantic information in the binaryzation process can be effectively prevented, and then the image retrieval precision is improved.

According to the OCR drawing recognition system based on the large model, in the system structure, an access layer provides an access form for a user, and an interface layer provides an access data interface between the access layer and a service layer; the data layer stores data in a system operation process; the access layer comprises data import, data annotation, manual auditing, attribute configuration and system management; the interface layer comprises file processing, data processing and configuration processing; the service layer comprises an OCR (Optical Character Recognition), a large model, authority management and safety management; the data layer comprises a database; the OCR drawing recognition system is based on the OCR of the service layer, feeds back a recognition error correction result through a large model, and provides recognition data for a user of the access layer; the authority of the service layer manages the access authority of the user, and the security management verifies the user session; and data and files generated by the business process are stored in the database. Through simple configuration, efficient identification of a large number of specific drawings and accurate distribution of information verification tasks can be realized.

The invention discloses an automatic report generation method and system, a terminal and a medium. The method comprises the following steps: constructing a database map; constructing a table relation model, and training the table relation model according to the database atlas to obtain a trained table relation model; obtaining a report demand, and inputting the report demand into the trained table relation model to obtain an analysis report; wherein the report demand comprises any one of structured data, semi-structured data and unstructured data; and verifying the analysis report to generate a verification result. According to the method and the device, the required report can be automatically generated from any one or more of the structured data, the semi-structured data and the unstructured data according to user requirements, manual intervention is reduced, and report generation efficiency is improved.

The invention relates to the field of data processing, in particular to a scientific and technological achievement data batch arrangement method based on OCR and semantic processing. Comprising the following steps: scanning paper scientific and technological achievements to obtain an archive image; preprocessing and segmenting the archive image, and converting the archive image into a digital document; extracting corresponding metadata from the digital document; storing the digital document and the corresponding metadata into a database as a paper scientific and technological achievement in a structured data form; generating a knowledge graph according to the digital document and the corresponding metadata, and storing the knowledge graph in a graph database in a knowledge graph form; storing the original scanning copy of the paper scientific and technological achievement into the storage object; correlating the contents of the same paper scientific and technological achievements in the database, the graph database and the object storage. Through the multi-mode OCR recognition, dynamic semantic annotation, intelligent data analysis and knowledge fusion technology, full-process automation of scientific and technological achievement data from collection, processing to storage is achieved.

The invention discloses a city form iteration generation system and method based on high-frequency interaction space self-learning, and the system comprises a first data collection module which is used for obtaining a target city space form three-dimensional vector model in a current state; the second data acquisition module is used for constructing a case city spatial form three-dimensional vector model database; the graph network module is used for constructing associated network data of the target city spatial form three-dimensional vector model and an associated network database of the case city spatial form three-dimensional vector model database; a multi-target high-frequency iteration module; and a feasibility scheme extraction and display module. According to the method, multi-target-oriented generation iteration of the urban spatial form is realized, high-frequency interactive self-learning of urban spatial form scheme generation is realized, the rationality of the urban spatial form scheme is improved, and the problems of high manpower and material resource investment, human brain judgment, high randomness and small scale in the traditional urban design are avoided.

The invention discloses a structured query statement generation method and device, equipment and a storage medium, belongs to the field of data processing, and can accurately generate SQL statements according to natural statements. The method comprises the steps of obtaining a to-be-generated statement, and determining intention information in the to-be-generated statement; based on a preset sub-graph determination rule, determining a first sub-graph structure corresponding to the intention information in a preset database graph structure, the database graph structure being formed based on a plurality of first graph nodes and an association relationship between the first graph nodes, and the first graph nodes being determined based on structure information of a plurality of associated databases; a plurality of preset second graph structures are obtained, the second graph structures with the structural similarity with the first sub-graph structures larger than a first threshold value are determined as third graph structures, and different second graph structures are generated based on different types of preset SQL statements; and determining statement generation parameters in the first sub-graph structure based on the third graph structure, so as to generate a target SQL statement based on the statement generation parameters.

The invention relates to the technical field of computer vision and geographic information, and particularly discloses a cross-view geographic positioning acceleration method based on semantic description, which comprises the following steps of: performing instruction fine tuning on an input query image and a database image by adopting a visual language model to generate natural language description; coding the natural language description into a low-dimensional semantic vector by utilizing a word embedding model; constructing a semantic keyword association graph based on a graph embedding technology, and fusing a general semantic knowledge base and a geographic domain ontology to construct a surface feature special semantic tree; screening candidate subsets according to a preset ground feature priority and a saliency detection result of the query image; extracting global features and local key point descriptors of the images in the candidate subsets by adopting a lightweight model; constructing a visual similarity calculation model based on the state space model, and modeling a feature sequence dependency relationship; and fusing the semantic matching score and the visual similarity, outputting a final matching result by adopting a weighted sorting strategy, and cooperatively accelerating according to the matching result in combination with a lightweight model.

A personalized joint skeletal and cartilage morphometry system is provided that uses artificial intelligence. [Solution] The personalized joint skeleton and cartilage morphometry system includes a database 11, an image acquisition module 12, an image analysis module 13, a neural network model 14, a model construction module 15 and a calculation module 16. The database stores a dataset. The image acquisition module acquires a plurality of skeleton images of a subject. The image analysis module processes the skeleton images by a statistical shape model and principal component analysis to generate a 3D model of the skeleton and cartilage. The neural network model generates and outputs skeleton motion parameters. The model construction module generates the personalized joint skeleton and cartilage model. The calculation module simulates the relative positions between the 3D models of the skeleton and cartilage during motion and calculates the cartilage shape and cartilage wear value.

The invention discloses a Text2SQL (Structured Query Language) training and reasoning method based on structured database atlas enhancement, electronic equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: receiving a natural language query request input by a user; splicing a natural language query request input by a user and the map enhancement representation corresponding to the target database into an input sequence, and inputting the input sequence into the trained large language model to generate a group of candidate SQL query statements; performing semantic consistency verification and grammar legality verification on the candidate SQL query statements to obtain final SQL query statements passing verification; and submitting the final SQL query statement to the target database for execution to obtain an execution result.

The present invention provides a multi-dimensional feature selection method based on deep learning. This method learns general features of images through an Inception deep learning model and a feature dimension random truncation model, stores the general features of database images, and selects the feature dimensions of a query image through a carefully designed feature dimension selection model, and queries in the database based on this dimension, so as to reduce the time required for querying.

The invention provides a full-link management method and device for business data, and the method comprises the steps: collecting first metadata covering the whole process of a target business, the first metadata comprising at least one type of metadata; storing the first metadata into a relational database; a matched target analysis logic is determined according to the type of the first metadata, the target analysis logic is called to analyze the first metadata, the blood relationship among the metadata is obtained, and the target analysis logic comprises at least one of SQL statement analysis logic, ETL script analysis logic, API interface analysis logic and exchange operation analysis logic; constructing a corresponding blood relationship map in a map database in a layering manner based on the blood relationship among the metadata, and storing the blood relationship map; and in response to the change of the structure of the second metadata, searching a blood relationship link related to the change in the relational database and the graph database so as to trace and obtain all nodes influenced by the change.

The application belongs to the technical field of computers and discloses a general metadata operation method and system based on a graph database, which comprises the following steps: S1, using a configuration file yml of SpringBoot to configure to create a connection pool SessionPool of a NebulaGraph Database graph database based on a NebulaPool custom implementation, containing definitions of maxCountSession, minCountSession, hostAndPort, userName and passWord parameters; S2, defining parameter models for accepting different interfaces, including a model for creating a graph space, a model for creating a point, a model for creating an edge, a model for creating an index and a general query model; S3, using an independently developed NebulaUtil tool to automatically arrange nGQL statements for the accepted parameter models according to the input models; S4, defining different Controller layers to provide an open and unified Api interface externally; S5, using an execution service of a GraphCommonService graph database to obtain an effective connection of the NebulaGraph Database graph database from the SessionPool; and S6, executing the nGQL statements automatically arranged by the NebulaUtil tool and analyzing and returning a result model, and uniformly returning the result.

The application discloses a kind of multi-modal information comprehensive retrieval street view image positioning method and system, by establishing multi-modal information database, using visual feature extraction model and scene text recognition model establish multi-modal information database;By fusion feature retrieval, global feature and local feature are fused and preliminary recall search is carried out;By geometric verification rearrangement, the geometric verification algorithm based on local feature is used to rearrange preliminary recall search result, and obtain refined retrieval result;By scene text retrieval, the scene text set overlap degree of query picture and database picture is calculated, and the final retrieval result is obtained based on this score ranking. The application considers the comprehensive retrieval scheme of multi-modal information, can greatly improve the image positioning accuracy in practical application, and to a certain extent, solves the complete positioning failure problem of single modal information in some scenes.

1. Name of the product of this design: Electronic device with wing-shaped database graphical user interface. 2. Purpose of this design product: an electronic device. 3. The key design point of this design product lies in the graphical user interface content on the screen. 4. The picture or photo that best illustrates the key points of the design: main view. 5. Purpose of the graphical user interface: The graphical user interface of this appearance design is a display interface for airfoil selection in wind turbine modeling and simulation calculations; the main view interface is the initial interface; in the main view interface, when you hover the mouse over the third icon from the upper right corner of the interface, the name of the icon "Airfoil Database" will be displayed, as shown in the interface change state diagram 1; after clicking the "Airfoil Database" icon in the interface change state diagram 1, the "Airfoil Database" pop-up window will pop up, as shown in the interface change state diagram 2; in the interface change state diagram 2, click on one of the airfoils in the "Airfoil Database" pop-up window and click the "View Aerodynamic Characteristics" icon, the "Airfoil Database - Aerodynamic Characteristics" pop-up window will pop up, as shown in the interface change state diagram 3; click the close button in the upper right corner of the "Airfoil Database - Aerodynamic Characteristics" pop-up window to return to the main view interface.

The present invention discloses a design method and device for an agricultural cultural heritage landscape space database. The method includes the steps of: adopting an object-oriented landscape gene classification mode to define multiple landscape categories and category elements under each landscape category; dividing the agricultural cultural heritage landscape data into spatial data and non-spatial data, and defining corresponding data storage types; digitizing various category elements; establishing an E-R diagram; establishing multiple database layers representing spatial data; constructing a physical model of the agricultural cultural heritage landscape space database, and its data layer is provided with a basic ground object information database and an agricultural cultural heritage landscape special database. The present invention can solve the problems existing in the prior art, such as the inability to conduct systematic dynamic monitoring and management, data systematic analysis, scientific planning and design, etc. for agricultural heritage landscapes.

The invention provides a method and a system for comparing and positioning database images by a chip, and belongs to the field of semiconductors. The method for comparing and positioning the database images by the chip comprises the following steps of: providing a GDS original file, wherein the GDS original file comprises an original graph when an integrated circuit is designed; obtaining the key sizes of the original graphs, and screening the graphs with the key sizes of the original graphs being a first set value and a second set value respectively; etching the wafer to obtain an etched pattern; and measuring the etched pattern, obtaining the variable quantity of the critical dimension of the first pattern and the second pattern, and obtaining the compensation value of the critical dimension of the original pattern based on the variable quantity. According to the method, the key size compensation value of the graph in the GDS file is obtained in an internal difference mode, and the situation that due to different contraction or expansion sizes of the etched image of the wafer, positioning of a detection machine table fails can be avoided.

The invention discloses an in-plant implementation method and system for upgrading and reconstruction of an intelligent substation monitoring system, and the method comprises the steps: restoring field data of a visual chemical industry to an in-plant same-type environment, and guaranteeing the consistency of data replication through online comparison of large-version marks of backup and recovery data; the upgrading and reconstruction process is realized through an intermediate file, and old system export and new system import of module information such as a database, a graph, channel data and a forwarding point table are supported; a factory simulation test environment is constructed, so that the monitoring system can complete simulation test verification and automatic point alignment after upgrading and reconstruction in a factory. Therefore, potential problems can be found and solved in advance by utilizing abundant software and hardware resources and technical support strength in a factory before the field debugging stage of an upgrade and reconstruction project, the field power failure debugging acceptance time is remarkably shortened, and the efficiency and quality of the upgrade and reconstruction work of the intelligent substation monitoring system are improved.

The application provides a multi-source spatial data topology consistency automatic processing method for quick update of a vector database, and belongs to the field of topology relation checking. The method comprises the following steps: pre-processing a fusion data layer graph spot, processing the topology consistency of the fusion data layer graph spot and a basic database graph spot, and post-processing the fusion data layer graph spot; setting a topology consistency processing threshold value of the fusion data layer graph spot and the basic database graph spot, and obtaining a processing result according to the processing. The application uses an R-tree spatial index method to search for geometric graph spots with common points, common edges or overlaps, so that the accuracy is not sacrificed while meeting the quick search. The application processes and aligns the boundaries of points, lines and surfaces of the geometric graph spots in all directions, and has high correction accuracy and high speed.

The invention discloses a cross-platform data fusion method in the field of cross-border trade, and the method specifically comprises the following steps: A1, constructing three types of data storage modes of a modular trusted data middle table, a support relational database, a graph database and a block chain, and relates to the technical field of data processing. According to the cross-platform data fusion method in the cross-border trade field, by constructing a modular data platform and integrating a relational database, a graph database and block chain storage, the problem of unified storage and management of multi-source heterogeneous data in cross-border trade is effectively solved. Based on a micro-service architecture and a unified data access layer, flexible deployment of an algorithm module and efficient access of multiple data sources are realized, and the automation level of data fusion processing and the system flexibility are remarkably improved. Through the introduction of the block chain credible mutual recognition system, the authenticity, traceability and non-tampering performance in the cross-border data sharing process are ensured, a credible foundation is laid for multi-party cooperation, and the credibility of the platform is greatly enhanced.

The invention discloses a semiconductor equipment part shipment report generation method and system based on a serial number, and relates to the technical field of intelligent manufacturing, image information, helium detection information, granularity information, size detection information and order information are collected in real time through a data acquisition module which runs through a whole production process, and the information is packaged into a reference data set after integrity verification; a unique serial number is generated by fusing the customer identifier, the material number and the completion timestamp with the UUID, the unique serial number is bound with the reference data set and stored in a shipment database as a key value record, and an image link points to an independent file server; according to a client identifier, a template library is intelligently matched, a corresponding field is dynamically extracted, an image and a detection result are automatically embedded, a multi-language visual shipment report is generated according to needs, a storage path is returned, single-piece-level data tracing, quality inspection closed loop and personalized report second-level output are achieved, and delivery transparency and client satisfaction are improved.

The invention discloses a picture loading method and system based on network speed, and the method comprises the following steps: constructing a picture database which comprises low-resolution pictures, medium-resolution pictures and high-resolution pictures; receiving a network request, and adding a current network speed parameter in the network request; obtaining a value of a current network speed parameter in the network request; and returning a corresponding resolution picture from the picture database according to the value of the current network speed parameter. The method solves the problems that under the current complex and changeable network environment, a traditional picture loading mode often adopts a fixed picture resolution ratio, the mode can lead to overlong loading time and overlarge data flow consumption when the network condition is poor, and under the condition that the network condition is good, the loading time is too long and the data flow consumption is too large. And the requirement of a user on a high-quality visual effect cannot be met due to too low resolution of the loaded picture.

The invention discloses an image retrieval method and system based on cross-X semantic hypergraph learning. The method comprises the following steps: importing an image to be detected; extracting global features and local region features in a cross-layer and multi-scale manner through a pre-trained convolutional neural network, and outputting feature spectrums of the last two layers of the convolutional neural network; inputting the feature spectrum into a pre-constructed hypergraph neural network to generate a semantic hypergraph so as to simulate high-order correlation between global features and local region features in different layers, different scales and different images, and outputting a multi-layer logic element vector; and fusing the multiple layers of logic element vectors, inputting the fused multiple layers of logic element vectors into a multi-codebook quantizer, carrying out calculation through a quantization codebook in the multi-codebook quantizer, finding a quantization code corresponding to a database image closest to the quantization embedding of the image to be detected, and taking the corresponding database image as a retrieval result. According to the method, the precision of fine-grained image retrieval can be improved.

To provide an image retrieval system and method with high retrieval accuracy.SOLUTION: The database has a function of storing a plurality of pieces of database image data, and a database tag is associated with each of the plurality of pieces of database image data. The processing unit has a function of acquiring database image feature amount data representing a feature amount of database image data for each piece of database image data, a function of acquiring query image feature amount data representing a feature amount of query image data, a function of calculating a first degree of similarity that is a degree of similarity of the database image data to the query image data for each piece of database image data, and a function of acquiring a query tag that is a tag associated with the query image data using a part of the database tag.SELECTED DRAWING: Figure 11

The invention relates to a source tracing method and system for a fire site and a fire point, and a storage medium, and belongs to the technical field of information source tracing based on space correlation and fire scene trace image intelligent analysis. Comprising the following steps: data acquisition and database construction, and standardized processing and tagged management are carried out; extracting fire scene trace features in a three-dimensional space, wherein the fire scene trace features comprise spreading direction trace feature extraction, fire site local feature extraction and fire point local trace feature extraction; constructing a mapping model from a three-dimensional space to a two-dimensional image, and establishing a cross-modal feature correlation model; carrying out machine learning analysis, training a large amount of collected fire scene data and laboratory simulation data by using a deep learning algorithm, comparing a scene image with a database image, calculating the similarity and judging the position of a fire point; and model optimization and dynamic learning: through a dynamic incremental learning mechanism, continuously introducing new data, optimizing model performance, and improving accuracy and generalization ability. According to the invention, the scientificity and accuracy of fire investigation are greatly improved.

The application discloses a kind of customer relationship audit analysis method and system based on graph database, and heterogeneous data is synchronized by distributed lock, customer, account is converted into graph node, and is converted into relationship edge such as fund transfer, legal person, mobile phone association, and dynamic construction customer relationship graph model;Adopt relational database+graph database hybrid storage, utilize KCore, shortest path, community discovery and other graph algorithms to complete depth audit analysis;With JSON / Schemaless dynamic structure storage result, without modifying database table structure, can extend audit dimension;With standardization interface and visual interaction realize with audit business platform depth integration.The application can significantly improve the efficiency of complex relationship query under large-scale customer data, guarantee data consistency, enhance system scalability, reduce audit operation threshold, and be suitable for bank customer relationship audit, anti-money laundering investigation, associated transaction identification, risk penetration verification scene.

Embodiments of the present application relate to systems and methods of inspecting reticles using machine learning. Near-field reticle images are generated via a deep learning process based on reticle database images generated from a design database, and far-field reticle images at an image plane of an inspection system are simulated via a physics-based process based on the near-field reticle images. The deep learning process includes training a deep learning model based on minimizing differences between the far-field reticle images and a plurality of corresponding training reticle images acquired by imaging training reticles fabricated from the design database. Test regions of test reticles fabricated from the design database are inspected for defects via a die-to-database process that includes comparing a plurality of reference images from reference far-field reticle images to a plurality of test images acquired by the inspection system from the test reticles. The reference far-field reticle images are simulated based on reference near-field reticle images generated by the trained deep learning model.

The invention discloses a personnel re-recognition system and method based on attitude guide generation and reasoning stage enhanced retrieval, and the method comprises the steps: carrying out the retrieval of a query image, selecting the first k results with the highest cosine similarity from a database image, and selecting m reference images based on the identity recognition accuracy; performing attitude extraction on the reference image; pixel features and attitude features are extracted and fused to obtain an attitude normalized generated image; identifying identity consistency of the generated image and the query image and attitude consistency of the generated image and the reference image; and in a reasoning stage, combining the query image with the generated image set, performing feature matching, and calculating weighted cosine similarity to obtain a retrieval-enhanced personnel re-identification task sorting result. According to the method, the target attitude is extracted by using the reference image in the reasoning stage, the generated image with the normalized attitude is guided to be obtained, and the query image and the generated image are combined for retrieval, so that the personnel re-recognition performance is improved.

The present invention discloses a scatter plot rendering method and system based on non-sampling big data, and the method comprises: a front-end browser initiates a rendering request to an RPC gateway according to a scatter plot type selected by a user; the RPC gateway routes the rendering request to a corresponding business system; the business system optimizes data of a relational database, records optimized filtering conditions and optimized data; the business system loads a data analysis strategy, obtains a script of a database query language in combination with configuration parameters of a set of data obtained after optimization, and transmits the script to the front-end browser via the RPC gateway after processing; the front-end browser grammar analyzes the processed script, and forwards the processed script to a proxy gateway for verification; the proxy gateway splices the processed script and converts it into a new HTTP request and sends it to a ClickHouse database; an image rendering component stores the data of the ClickHouse database returned by the proxy gateway; and a rendered image is obtained.

The invention discloses a depth variable-length Hash retrieval method for adaptive bit selection, and relates to the technical field of image retrieval, an image passes through a trained retrieval model to generate a query image binary code, the Hamming distance between the query image binary code and an image binary code in an image database is calculated, and the Hamming distance between the query image binary code and the image binary code in the image database is calculated; the database images are returned according to the distance from small to large to serve as retrieval results; the training process of the retrieval model comprises the following steps of: 1, sequentially inputting each image in a constructed image training set into a deep hash network so as to output a binary code corresponding to each image, and training the deep hash network through optimization measurement so as to generate a bit pool; step 2, inputting the bit pool into a bit selection agent for bit selection, and taking the optimization measurement which is the same as the optimization measurement in the step 1 as an environment reward; step 3, alternately iterating the step 1 and the step 2 until the retrieval model converges; according to the retrieval method, variable-length compact feature expression is realized, and the retrieval precision is improved.