27results about How to "Efficient retrieval" patented technology

The invention discloses a video content retrieval method and device based on television system voice interaction and a terminal, and relates to the technical field of video processing, and the method comprises the steps: when a video is played for the first time, extracting a picture frame from the video at a preset frequency, converting the picture frame into a multi-dimensional image feature vector and a corresponding video timestamp, and carrying out hierarchical storage in a database, constructing vectorized data containing visual semantic information; obtaining a voice retrieval instruction, performing intention recognition and semantic understanding, extracting a detection keyword, and generating a multi-dimensional retrieval feature vector; calculating a matching degree between the multi-dimensional retrieval feature vector and a multi-dimensional image feature vector of a video picture frame stored in a database, and screening out picture frames of which the similarity is higher than a preset similarity threshold to form a retrieval candidate matching set; and determining matched picture playing. The video content retrieval method is efficient, accurate and high in interactivity, and retrieval experience and operation efficiency of the user in the video watching process are remarkably improved.

The invention relates to a class C liquid material storage system which comprises an automatic packaging line area, the automatic packaging line area comprises an automatic packaging line system and a warehousing mechanism, and the warehousing mechanism comprises a storage tank supply station, a canning station, a labeling station, a cover sealing station and a tray assembling station which are sequentially connected in the upstream and downstream direction; the automatic packaging line system is in signal connection with the labeling station; the automatic warehouse area comprises a warehouse management system, an automatic warehouse and a pallet supply station, the warehouse management system is in signal connection with the automatic packaging line system and the automatic warehouse, the warehouse management system is used for sending the pallet requirements of the corresponding number to the automatic warehouse, and the automatic warehouse is connected with the pallet supply station; the automatic bin is used for sending pallets to the pallet supply station; and the pallet supply station is connected with the tray assembling station. The automatic packaging line area and the automatic warehouse area can be connected in series, and efficient goods taking can be achieved by means of the automatic packaging line system and the warehouse management system.

The embodiment of the invention relates to the technical field of image retrieval, and provides a secure outsourcing image retrieval method and device, electronic equipment and a medium, and the method comprises the following steps: obtaining an image retrieval request sent by a query user, including retrieving an encrypted image; in response to the image retrieval request, performing ciphertext feature extraction on the retrieved encrypted image to obtain a first orthogonal moment feature vector, and calculating a first ciphertext hash value corresponding to the first orthogonal moment feature vector; determining a target hash bucket in a hash bucket set based on the first ciphertext hash value; screening a preset number of target encrypted images with the minimum distance from the first orthogonal moment feature vector in the target hash bucket; and converting the encryption form of the target encrypted image, and returning the converted target encrypted image to the query user. Therefore, efficient retrieval of the secure outsourcing image is realized, and the security of the scheme is improved.

The application provides an information retrieval prompting method and device, electronic equipment and a storage medium, the method comprising: preprocessing device retrieval information input by a user to determine a retrieval keyword; sequentially matching the retrieval keyword with a prompt word library, a synonym library and a pinyin word library to obtain multiple prompt information corresponding to the device retrieval information; the prompt word library is constructed based on attribute information of various devices, the synonym library is constructed based on synonymous words corresponding to the attribute information of various devices, and the pinyin word library is determined based on the prompt word library; and the multiple prompt information is sent to a front end for display. The application can provide accurate retrieval prompt information for a user, help the user quickly and effectively retrieve attribute information of an Internet of Things device of interest, improve the multi-dimensional prompting and accurate retrieval capability of the Internet of Things device, and provide a good user experience.

The application discloses a method for extracting keywords information of bidding and opening bidding based on regular expression and scrapy, comprising the following steps: S1, determining the analysis field of the bidding and opening bidding project; S2, collecting and sorting the keyword list of the field, wherein the keyword list is created as an ordered keyword list or an unordered keyword list according to the field requirement; and S3, selecting the subsequent processing method according to the specific field condition. By combining the regular expression and the scrapy, the application can extract multiple keywords information of the bidding and opening bidding project, so as to avoid the problem of missing search. The application respectively extracts the keywords information of the bidding, the winning bidding and the opening bidding project, can carry out targeted query extraction according to the different keyword extraction modes of the opening bidding project, realizes the quick extraction of important information in the text, removes the useless data, solves the problems of information redundancy, reading difficulty and the like.

The application relates to the technical field of artificial intelligence, and discloses a large model expert library construction based on a knowledge graph, which realizes real-time acquisition of multi-source data, marks an update point through a hash fingerprint algorithm and semantic similarity calculation, and stores the update point in a message queue in combination with metadata determination; a low-dimensional semantic vector is generated by extracting features and constructing a fusion network, and cross-modal mapping is established; an attribute graph is built with the semantic vector as a node, and the attribute graph contains basic facts, reasoning rules and time sequence edges; a user request is parsed as a graph traversal task, and a result is returned through counterfactual reasoning; sensitive data and relationships are encrypted and indexed; source field graph features are migrated to optimize target field graph, and the construction of the expert library is accelerated. The application realizes efficient data processing, semantic correlation analysis, safe query and rapid construction of the expert library, and improves data utilization and knowledge reasoning efficiency.

The application discloses an enterprise collaborative office method and system based on intelligent AI, and the method comprises the following steps: collecting and preprocessing the data of each end of the enterprise collaborative office; constructing an AI collaborative engine containing multiple AI models, and training and optimizing each model by using the preprocessed data; the AI collaborative engine analyzes the semantics and intention of the input information of each end, automatically triggers the corresponding process and allocates tasks, adjusts the allocation strategy according to the execution situation; automatically identifying the relevant departments and personnel of cooperation, establishing a temporary collaborative team and generating a cooperation report; performing hierarchical permission management on the data of each end, monitoring abnormal behavior and generating a security audit log; combining the task execution and process operation data, continuously iterating and optimizing each model of the AI collaborative engine. The enterprise collaborative office method and system based on intelligent AI solve the problems of existing collaborative office data islands and insufficient intelligence, improve the collaborative efficiency and management level, and adapt to the needs of enterprise digital transformation.

The embodiment of the invention provides a multi-node database cluster parameter detection method and a related device, and relates to the technical field of databases. The method comprises the following steps: constructing a multi-node parameter value mapping table so as to identify an association relationship between nodes and parameters in the multi-node parameter value mapping table through a unique key; predefining a parameter checking rule set, wherein the parameter checking rule set comprises a parameter consistency requirement and a compliance threshold value; and based on the multi-node parameter value mapping table and the parameter checking rule set, performing comparative analysis on the parameter values of the nodes in the multi-node database cluster, and outputting abnormal information. Based on the method, the detection efficiency and accuracy of multi-node database cluster parameter detection can be improved.

The invention discloses a process knowledge question-answering method based on a domain knowledge graph, which comprises the following steps of: identifying entity information and relation information in a document by adopting a natural language processing technology, and constructing a structured process knowledge graph in a triple form; by training a BiLSTM + CRF model, accurate recognition of subject entities and relationships in user questions is realized, and a plurality of candidate answers are acquired from a knowledge graph in combination with a predefined structured Cyber query statement template library; finally, semantic matching is conducted on the candidate answers and the user questions through a RoBERTa model, and then answers most relevant to the questions are generated. According to the method, the entities and the relations in the question statement can be accurately recognized, matching of a grammar level is considered, and the method is deeper to a semantic level, so that efficient retrieval and accurate question answering of complex process knowledge can be supported.

The application relates to the technical field of radar signal sorting, and provides a jitter PRI sequence retrieval method and device based on a full convolution network. The method introduces a PP sequence as a supplementary feature of a DTOA sequence in pulse sequence retrieval, uses a trained jitter PRI sequence retrieval model based on a full convolution network to retrieve the DTOA sequence and the PP sequence, effectively retrieves the jitter PRI sequence without needing an a priori jitter upper bound, and significantly improves sequence retrieval performance.

The invention discloses a multi-modal large model-based remote sensing image-text cross-modal retrieval method and system, and the method comprises the steps: constructing a retrieval model and carrying out training, and the framework of the model comprises initial modal feature extraction, fine-grained image-text representation enhancement, group perception image-text similarity scoring function design and a multi-view contrast learning strategy; respectively extracting initial representations of the text and the image; respectively enhancing the two types of fine granularity representations; for the image, three learning tasks of feature reconstruction, scene classification and background information alignment are taken as guidance, the background information of each image block is separated from target features, and meanwhile, the relationship between targets is reinforced through image learning; key word information is emphasized for the text; and calculating correlation scores and generating the similarity of the image-text pairs. The system comprises a model building unit, a model training unit and a retrieval unit. By using the method and the device, the cross-modal retrieval precision can be improved. The method can be applied to the field of cross-modal retrieval.

The invention discloses a dynamic field retrieval method and system based on an extensible retrieval processor mechanism. The method comprises the steps that multiple retrieval types are defined, and corresponding retrieval processors are configured; constructing a dynamic mapping template based on regular expression matching, and automatically distributing data types for dynamic fields; configuring a search field pool which can be dynamically loaded, and defining a retrieval attribute and an interaction mode of a field; receiving and analyzing a dynamic retrieval request generated by a front end according to the field pool, and obtaining a retrieval parameter of each field; according to a retrieval type in the parameters, dynamically scheduling a corresponding retrieval processor through a processor factory; query conditions are automatically constructed by the processor based on the parameters and integrated into the primary query. Full-link configuration driving from field definition, type mapping, interface generation to query execution is achieved, non-intrusive expansion of retrieval logic is supported, and the flexibility, maintainability and development efficiency of a system facing business changes are remarkably improved.

The utility model discloses a kind of winding storage racks for waterproofing membrane, including U-shaped base and waterproofing membrane body, further include: multilayer storage mechanism: the multilayer storage mechanism includes pivot being arranged on the U-shaped base left and right uprights, first placement plate is connected on the pivot on left side, second placement plate is connected on the pivot on right side, limit device is arranged on the top outer wall of first placement plate and second placement plate, the bottom of first placement plate and second placement plate is provided with pivot, connecting rod is connected on the pivot, the connecting rod is used in cooperation with the limit device. The utility model discloses winding storage rack for waterproofing membrane has reduce separating force, antiskid, prevent sticking, prevent extrusion, multilayer storage improves space utilization, improves operation convenience, makes storage and taking thing more efficient, intelligent, improves the accuracy and efficiency of overall work technical effect.

The embodiment of the invention provides a multi-mode retrieval method and device and electronic equipment. According to the embodiment of the invention, the method comprises the steps: obtaining a question-answer pair matched with a target question from cached question-answer pairs; then, a judgment model with a small model parameter quantity is introduced to be used for judging whether the answers in the obtained question and answer pairs are matched with the target questions or not, if yes, the target answers of the target questions can be directly determined based on the answers in the question and answer pairs, and if not, the user questions are further analyzed through an RAG large model to obtain the target answers. According to the method, the cached question and answer pairs are used for retrieval, the judgment model with the small model parameter quantity is combined for judgment, and the RAG large model with the large model parameter quantity is further introduced for analysis only under the condition that the retrieved question and answer pairs do not meet the requirements, so that the retrieval efficiency can be effectively improved, efficient retrieval is realized, and the retrieval efficiency is improved. And the calling frequency of the RAG large model can be reduced as much as possible, and resource consumption is reduced.

The application provides a question and answer method and system based on a large language model, which comprises the following steps: determining a current question of a user and a natural language understanding prompt word; inputting the current question and the natural language understanding prompt word into a pre-trained large language model to obtain a question understanding result and a question to be answered output by the large language model according to a natural language understanding strategy; in the case that the question understanding result is a factual question, retrieving an answer from a knowledge graph based on semantic analysis; inputting the retrieved answer and an answer verification prompt word into the pre-trained large language model to obtain an answer verification result; in the case that the verification is reasonable, taking the retrieved answer as the final answer to the current question; and in the case that the verification is unreasonable, generating an answer by the large language model. In the multi-round question and answer task, the application can deeply understand the question of the user based on the context, effectively retrieve and integrate information from different knowledge sources, and thus accurately and efficiently provide an answer to the current question.

The invention discloses a method for searching and identifying road diseases by using persistent coherence, and relates to the technical field of road defect detection. According to a topological data analysis method based on persistent coherence, path finding is carried out by using persistent coherence. Comprising the following steps: step 1, zooming an image again; 2, finding out a topological structure by using persistent coherence; step 3, bonding the area near the position; and 4, analyzing the shapes and sizes of different segmented regions to detect and identify the road disease. According to the topological fingerprint generation method based on continuous coherence, a novel and powerful technical path is provided for solving the essential feature extraction problem of a two-dimensional object, and path diseases can be found and recognized more accurately.

The application discloses a kind of vehicle trajectory similarity calculation methods based on dynamic graph alignment and POI enhancement, comprising the following steps: trajectory is preprocessed and map matching;Adopt the adaptive grid division of curvature-density double constraint to generate multi-scale grid representation, in combination with POI distribution density and function weight injection area semantics;Road representation is learned using the graph attention network guided by transition probability and co-occurrence probability counted from trajectory data;Dynamic graph is constructed with GPS, grid, road as node, based on multi-modal correlation index dynamic calculation edge weight, cross-modal semantic alignment and fusion are realized through graph attention network;Finally, the model is optimized through the joint loss function of fusion local structure preservation, global contrast learning and modal regularization, and the trajectory similarity is output;The application improves the geometric fidelity, semantic richness, robustness to noise and adaptability of multi-modal fusion of trajectory representation, significantly improves the accuracy of trajectory similarity calculation.

The application provides an intelligent power data retrieval system, which comprises a data acquisition module, a data processing module, a retrieval interaction module, an intelligent matching module, a retrieval output module and a retrieval security module; the data acquisition module is used for acquiring real-time and historical multi-source power data and storing the multi-source power data in a power database; the data processing module is used for performing a preprocessing operation on the acquired multi-source power data and extracting power data features; the retrieval interaction module is used for receiving retrieval request information of a user and analyzing the retrieval request information; the intelligent matching module is used for performing similarity matching between the analyzed retrieval request and the power data features; the retrieval output module is used for feeding back the matching result of the intelligent matching module to the user in a preset format; and the retrieval security module is used for guaranteeing the safety of the multi-source power data in the acquisition, transmission, storage and retrieval processes. The application can realize efficient retrieval of power data and improve retrieval accuracy and response speed.

The invention relates to a dynamically reconfigurable table item storage and retrieval method, and belongs to the field of network communication processing. According to the method, the to-be-stored table item content is mapped into the index value with the uniform length, and the index value is used for achieving storage and retrieval of the stored table item content. Wherein the mapping algorithm and the unified length can be dynamically reconfigured and online replaced according to the actual storage capacity and retrieval rate requirements, and index value conflicts are reduced by adopting a linear expansion exploration method. According to the method, efficient storage and retrieval of dynamically reconfigurable table item data can be achieved, high-speed retrieval of table item content is achieved, the requirement for table item storage capacity is lowered, and the utilization rate of storage space is increased.

The invention discloses a honey pomelo supply chain full-process tracing management system based on big data storage, and the system comprises a feature construction module which is used for carrying out the slicing and vectorization of each stage of data time sequence, and generating a batch feature sequence; the stage graph modeling module is used for constructing a directed graph, mapping the features into nodes and generating connections; the graph embedding calculation module is used for generating node high-dimensional representation through multi-layer feature propagation and writing and storing the node high-dimensional representation; the weight dependence module is used for calculating an attention weight and generating a stage weight matrix; the feature tracing module is used for weighting aggregation node representation, generating a tracing vector and verifying the tracing vector; the index mapping module is used for generating index codes for the tracing vector hash and establishing mapping storage; and the tracing reconstruction module is used for traversing the reconstruction path by the reverse graph and generating a whole-process tracing result. According to the method, continuous collection and full-process tracing of supply chain data are achieved, batch feature vectors and index codes are generated, data integrity is guaranteed, and retrieval efficiency is improved.

The invention relates to a nuclear power operation and maintenance data cross-modal retrieval method, device and equipment and a medium, and belongs to the technical field of nuclear power operation and maintenance management.The nuclear power operation and maintenance data cross-modal retrieval method comprises the steps that nuclear power operation and maintenance image data and text data are obtained; performing feature extraction on the image data and the text data, performing modal similarity calculation on image features, global pairing text features, pairing generated text features and non-pairing generated texts to obtain a similarity distance, determining a target loss function based on the similarity distance, and adjusting parameters of a cross-modal retrieval model by using the target loss function to obtain a cross-modal retrieval result. Inputting to-be-retrieved operation and maintenance data into the trained cross-modal retrieval model to obtain an image feature matrix and a text feature matrix, performing quantification processing on the image feature matrix and the text feature matrix to obtain a hash code, performing retrieval based on the hash code to obtain a text or image corresponding to the hash code, and storing the text or image in the cross-modal retrieval model. And the retrieval precision of the nuclear power operation and maintenance data is improved.

The application belongs to the technical field of video coding, and discloses an indoor monitoring video coding method based on human-computer collaborative perception, which comprises the following steps: 1, building a human-computer collaborative video coding framework; 2, compact feature representation of specific semantic objects; 3, adaptive background updating; 4, residual coding resource allocation.The application proposes an adaptive background updating framework to dynamically update the background according to the characteristics of indoor monitoring video, considers that specific foreground semantic objects in the indoor monitoring video can be represented by compact semantic features, extracts compact semantic feature representation for specific foreground semantic objects through a feature extraction network, formulates a residual coding resource allocation strategy for a traditional encoder, avoids useless coding resource allocation, improves the human eye visual quality of the decoded video, greatly improves the compression performance of the indoor monitoring video, and realizes efficient specific semantic information retrieval and analysis.

The application provides a blockchain-based verifiable keyword Top-K non-fuzzy query and fuzzy query method, and a scheme combines RSA accumulators, inverted indexes and prefix trees and other related technologies to construct an ADS, and realizes ranking search verification of non-fuzzy queries and fuzzy keywords. The framework of the application can enable light nodes to effectively verify the correctness of query results, and performs well in search, verification, storage and update operations, and has good practical value.

The application discloses a cross-document type retrieval method and system based on differential nested coding. The method includes six steps: document preprocessing, document type identification, differential coding, index construction and optimization, online retrieval and result fusion. First, the document is standardized and preprocessed to extract structured information. Second, the BERT classifier is used to analyze the document features and identify the document type. Then, according to the document type, the corresponding encoder is selected to generate the original coding vector. Next, the differential nested coding index is constructed and optimized. In the online retrieval stage, the user query is preprocessed, the relevant document type is predicted, and the approximate nearest neighbor retrieval is performed. Finally, the retrieval results are sorted by a stepwise weight fusion strategy, and the top-100 retrieval results are returned. Through differential coding and index optimization, the application realizes efficient retrieval of different types of documents, and has significant practical value and application prospect.

This invention relates to the field of information retrieval and data visualization technology, and discloses a visual text search method based on a hybrid data model. The method includes: constructing a hypergraph hybrid model by associating a graph database and a vector database using global identifiers; identifying the intent of user queries; generating structured search instructions for the graph database and semantic similarity search instructions for the vector database; fusing and expanding the two types of search results based on global identifiers to form a search result set containing an extended knowledge graph, a list of semantic vectors, and traceable links between them; and finally, converting the result set into a dynamic and interactive visual view according to predefined mapping rules and a hybrid mechanics layout algorithm. This invention effectively unifies semantic retrieval and relational retrieval, improves the comprehensiveness and accuracy of retrieval, and significantly enhances the depth of information understanding and user experience through intelligent visualization and interaction mechanisms.

The embodiment of the application discloses a private domain data federated search method and system for data space, relates to the technical field of computers, and can realize effective and accurate query of scattered private domain data in a data space while meeting the requirement of a data owner in local control and management of private domain data. The method calls the search capability of a local private domain data source of a federation, thereby realizing effective search of scattered private domain data in the data space without the need of establishing centralized indexes. Based on a probability local sensitive hash table, the method can fuzzily select a target data source in which stored private domain data is consistent with a user query intention without knowing original data in the private domain data source, and sends a search request to the target data source, thereby reducing the risk of leakage of the original data in the private domain data source, and realizing accurate query of scattered private domain data in the data space.

The application discloses a local archive search management system and method based on digital humanities, and particularly relates to the field of archive search management, and comprises an archive information region division module, an archive information collection module, an archive information identification module, an archive information database establishment module, a text information acquisition module, a text information extraction module, an information matching analysis module, an archive search evaluation module, a search engine optimization module and an archive resource security supervision module. The application collects archive information, extracts archive information therefrom, further establishes an archive information database, extracts key word information and search volume from text information input by a user, analyzes to obtain an archive priority matching coefficient, compares the archive priority matching coefficient with a preset archive priority matching coefficient threshold value, automatically searches in the database, and accurately and rapidly outputs archive information searched by the user, so that the digital technology is scientifically and reasonably used and plays a greater role in modern archive management.