98 results about "Multimodal analysis" patented technology

The invention relates to the technical field of video generation, in particular to an intelligent commodity video generation method based on multi-modal analysis and a dynamic narrative architecture, which comprises the following steps: inputting original commodity data and a user behavior log, and outputting a dynamic selling point weight vector through a selling point value evaluation engine; s2, inputting the dynamic selling point weight vector output in S1 into a narrative flow generator, and executing the following steps: intercepting first K core selling points according to the weight vector to form a narrative trunk chain; based on the historical preference data of the user, inserting an emotion enhancement node to generate a branch enhancement narrative flow; generating a multi-version narrative flow instruction set in combination with the real-time network bandwidth data; retrieving a preset video clip library according to the instruction set to generate an initial video sequence; detecting semantic fault regions between adjacent segments to generate compensation animation parameters; and injecting compensation animation parameters and outputting a continuous narrative video stream. According to the invention, the jamming feeling and the frame skipping phenomenon caused by the change of a narrative structure or different sources of video clips are reduced, and the overall watching experience of a user is improved.

The invention relates to an intelligent ship repair quotation method and system based on multi-modal analysis and a large language model, and the method comprises the steps: firstly receiving a ship repair inquiry sheet file uploaded by a user, and carrying out the feature enhancement of image data through a DocEnTR model, and obtaining a binary image; automatically extracting unstructured text data, semi-structured table data and text information in the binarized image by adopting an OCR (Optical Character Recognition) technology, and performing multi-modal analysis on image features of the binarized image and the extracted text information by adopting a large language model to generate an initial data stream; synchronously generating a standard semantic tag and a non-standard content tag by adopting semantic association of the large language model, automatically generating a dynamic cue word according to the standard semantic tag, inputting the dynamic cue word into the large language model, and outputting a standardized engineering description; performing manual auditing on the non-standard content mark to correct the engineering item field so as to generate structured engineering data; and finally, calculating the total price based on the structured engineering data and the material quantity parameters, and further generating a final quotation list to complete ship repair quotation.

The invention discloses an intelligent verification system and method for full-process review based on multi-modal analysis, and belongs to the technical field of intelligent review. Comprising a processing analysis module, an experiment extraction module, a risk early warning module, a reproduction verification module, a chart comparison module, a dispute tracing module, a feedback analysis module, a load monitoring module, a report generation module and a preference analysis module. According to the invention, comprehensive and multi-angle intelligent review can be realized, more detailed risk early warning and positioning information can be generated, the experiment reproducibility can be quickly verified at low cost, the manual verification burden is greatly reduced, the innovation and risk of the thesis viewpoint are scientifically evaluated, the fatigue of a manuscript reviewer is relieved, the manuscript review efficiency and quality are improved, and the method is suitable for popularization and application. And dynamically adjusting the review content recommendation sequence.

The invention relates to the technical field of artificial intelligence, and discloses a dialogue text mining method and system based on code scanning consumption cash back of big data, and the system comprises a consumption data collection module, a multi-modal analysis module, a strategy decision engine and a dynamic execution module. When code scanning consumption cash-back strategy decision making is carried out, real-time dialogue text semantics and consumption time-space behavior data are fused, a deep association mechanism of consumption intentions and payment scenes is established, the matching deviation of user language expression and actual consumption behaviors can be eliminated in real time, the consistency of a cash-back strategy and real consumption appeals is ensured, and the payment efficiency is improved. The problem of strategy mismatching caused by semantic understanding isolation in a traditional system is effectively avoided, the precision of personalized marketing is improved, double feedback of dialogue emotion tendency and historical strategy effects is dynamically analyzed, the deviation value of strategy feature matching is corrected in real time, and the system can actively adjust weight distribution of feature combinations in a strategy library.

The invention relates to the technical field of document intelligent review, in particular to a multi-mode digital publishing intelligent review system and method based on a large model, and the system comprises a sample construction module, a multi-mode analysis module, a structure recognition module, a term verification module and an annotation output module. According to the method, a model configuration data set is generated by analyzing a subject-object combination relationship in a text and jointly screening part-of-speech, sentence pattern and semantic structure, the accuracy of semantic recognition and structure judgment is improved, and local features of semantic offset and image-text expression separation are recognized by combining a comparison relationship between an image action path and a text behavior object; a paragraph logic structure and a theme connection mode are analyzed based on semantic offset information, the problems of content dislocation and theme jump between paragraphs are effectively revealed, matching change tracks and part-of-speech continuation fluctuation in term cross-paragraph contexts are recognized, label overlapping and coverage redundancy conditions are analyzed from sentence group hierarchy, and the term cross-paragraph contexts are obtained. And forming a label combination suggestion and constructing an annotation structure record.

The invention discloses a geographical intelligent data base station, which comprises a data sensing layer, a communication transmission layer, a management and calculation layer and a storage layer, and relates to the technical field of data communication. According to the geographical intelligent data base station, the intelligent data base station is arranged, the function boundary of a traditional base station is broken through, data localization storage, preprocessing and multi-modal analysis are achieved at the hierarchy of the base station, real-time decision feedback and data collaboration between distributed nodes are supported, and through a built-in self-adaptive algorithm, the intelligent data base station is obtained. Different scene requirements can be dynamically matched to complete tasks such as predictive analysis, anomaly detection and intelligent optimization, closed-loop data processing ecology is formed, response efficiency and system robustness are remarkably improved, infrastructure-level support is provided for intelligent internet-of-things application in a complex environment, various environmental factors are associated to establish an evaluation model, and the evaluation efficiency is improved. Data analysis and resource allocation are efficiently realized, and an execution strategy is timely generated for regulation and control.

The invention relates to the technical field of network information extraction, in particular to a visual webpage data crawling method and system based on a large model, and the method comprises the following steps: automatic browsing and screenshot, image preprocessing, OCR (Optical Character Recognition), multi-modal large model analysis, result fusion, large language model analysis, and data storage and output. The method has the beneficial effects that various picture or dynamic webpages can be handled through screenshot and multi-modal analysis, and dependence on a DOM (Document Object Model) structure is not needed; oCR and a multi-mode large model are complementary, and the extraction accuracy of data in different fonts, complex typesetting and charts is remarkably improved; a real browsing behavior is simulated and manual verification is combined, so that a common anti-crawling strategy can be effectively avoided; the large language model intelligently recombines page information according to Prompt, and the integrity and logicality of output data are guaranteed.

The invention discloses a fault repairing method of a server single board and electronic equipment, and relates to the field of server hardware reliability, and the method comprises the steps: scanning a plurality of target signal links of the server single board through a time domain reflectometer, correspondingly comparing the new impedance value and the new reflection waveform characteristic in the scanning result with the initial impedance value and the initial reflection waveform characteristic, determining at least one fault link of the server single board when the comparison result is inconsistent, and performing multi-modal analysis by combining the operation state data of the server single board. And determining the fault level of at least one fault link in the server single board and correspondingly generating a fault repair strategy to switch the fault link to the standby link, thereby solving the problems of single detection dimension, complex fault monitoring blind area, high-frequency signal monitoring failure and the like in the related technology, and improving the detection accuracy. Through time domain reflectometer monitoring, multi-modal data fusion analysis and redundancy switching, fault prediction, accurate positioning and lossless self-repairing of a server single board target signal link are realized.

The invention discloses a children's language narrative ability assessment method and tool based on multi-modal analysis, and relates to the technical field of natural language processing, and the technical scheme is characterized in that audio data and video data of a testee during a narrative process are collected, and voice rhythm features, text language features and visual behavior features are extracted for fusion; the method comprises the steps of inputting a multi-modal scoring model to generate multi-dimensional quantitative scores of a macrostructure, a microstructure, a language organization and a pragmatic function of the language narrative ability of a testee child, generating a structured evaluation report for the testee child according to the scores, and giving personalized intervention suggestions. According to the method, multi-modal data such as voice, language texts and visual behaviors are collected, so that the narrative ability of the children can be comprehensively analyzed from multiple dimensions such as macrostructures, microstructures, language organizations and pragmatic functions, and the method is closer to the real natural expression state of the children; therefore, the comprehensive language ability and communication performance of children can be reflected more comprehensively and truly.

The invention provides a major hazard source risk assessment method and system based on multi-modal analysis, and the method comprises the steps: determining a certain major hazard source as a to-be-assessed target, obtaining the multi-modal monitoring data of the to-be-assessed target, and determining the comprehensive risk value of the to-be-assessed target according to the multi-modal monitoring data; therefore, the risk level of the to-be-assessed target is determined according to the comprehensive risk value of the to-be-assessed target, the risk level of the major hazard source can be rapidly and accurately determined, real-time assessment analysis and display of the security risk of the major hazard source are realized, and sending, checking, feedback and supervision of early warning information can be timely and automatically completed according to the early warning level. That is to say, the method provided by the invention can quickly and accurately realize risk assessment of major hazard sources, improve accuracy and timeliness, and meet risk assessment requirements of users in different usage scenarios.

This invention discloses an intelligent alarm and control method for personnel safety in power plants, relating to the field of power industry safety control technology. The method includes: acquiring multimodal sensing data related to personnel based on the existing sensing system of the power plant; using a multimodal analysis model to fuse and analyze the collected multimodal sensing data to identify personnel safety risks; generating an alarm trigger signal when the risk level reaches a preset threshold; combining intelligent agent technology to generate a personalized alarm strategy after receiving the alarm trigger signal, simultaneously finding the nearest alarm device based on the real-time location of the personnel for continuous alarm, and establishing a continuous monitoring mechanism to track personnel status in real time; continuously monitoring the behavior and location changes of the alarmed personnel until the risk is confirmed to be completely eliminated. This invention enables cross-system correlation analysis, improves the accuracy of risk identification, effectively shortens response time, and improves the safety level of power plants.

This invention relates to the field of video inspection technology and discloses a multimodal model AI video inspection system and method based on cameras. The system includes: a video acquisition and processing module, an embedded vector calculation module, a difference judgment and triggering module, a multimodal analysis module, a work order management integration module, and a strategy dynamic adjustment module. During video inspection, by dynamically adjusting the screenshot acquisition interval, the detection frequency can be adaptively optimized according to the degree of scene change, avoiding resource waste caused by fixed intervals. This allows the system to reduce unnecessary analysis operations when there are no obvious changes in the image, thereby improving the efficiency of computing resources and the real-time performance of system operation, ensuring energy-saving and efficient inspection. During video inspection, by combining a multimodal large model with text prompting technology, visual features and scene context information are integrated for anomaly recognition, which can reduce the risk of misjudging subtle changes by a single model and improve the adaptability to complex scenes.

The invention provides a safety early warning method, system and equipment for a galvanizing process production line and a medium. The safety early warning method comprises the following steps: acquiring multi-source data of the galvanizing process production line; performing multi-modal analysis and vectorization processing on the multi-source data, constructing a galvanizing process production line knowledge graph, and storing the knowledge graph to an AI data lake; reading a knowledge graph from an AI data lake to adjust the general large model; transversely splitting the galvanization process into a plurality of process thinking chain nodes according to a process sequence; longitudinally splitting each process thinking chain node into an analysis node and an auxiliary node executed by the potential safety hazard analysis model to form a tree structure thinking chain; traversing the tree structure thinking chain based on a preset sequence, and analyzing the potential safety hazard of each process; and summarizing the potential safety hazard analysis results of all the processes, and generating a potential safety hazard report and rectification measures for safety early warning. The method helps managers to master potential safety hazards possibly existing after operation of a production line and corresponding rectification measures in the galvanization industry.

The invention discloses a software testing method based on multi-modal analysis and a terminal. The method comprises the following steps: monitoring a picture of equipment running to-be-tested software in real time, obtaining a frame-dropping picture in the picture and context data thereof, and analyzing the frame-dropping picture by using a multi-modal picture analysis model based on the context data to obtain a picture analysis text; extracting keywords in the picture analysis text, performing clustering operation on the extracted keywords, generating a question label corresponding to each cluster, and determining all check points according to the question labels; and checking the frame drop picture according to all the check points, marking the check points which do not pass the check result as problem data, and generating a test report, so that picture-level BUG detection can be realized, and the software test quality and efficiency are improved.

The invention relates to the field of intelligent recognition, in particular to an intelligent sign language ability identity recognition system based on multi-modal data analysis, which comprises the following steps of: S10, constructing a sign language animation resource library, and decomposing a sign language expression system in a national general sign language dictionary into 500 basic gesture units and 1200 composite action templates, combining an artificial intelligence technology to synthesize sign language animations; s20, randomly extracting four sign language animations and corresponding text vocabularies from the resource library, performing out-of-order processing on the text vocabularies, and performing distortion interference imaging processing; s30, pushing an interactive webpage containing the four sign language animations and the vocabularies subjected to graphical processing, starting 60-second countdown, and requiring a user to click matched vocabulary images according to an animation sequence; and S40, performing multi-modal analysis based on the user operation time difference and the coordinate position, and verifying the sign language ability of the user. Therefore, more resources can provide efficient services for the hearing-impaired speech disabled while the safety problem is solved through the system.

This invention discloses an automated data security assessment method and system based on intelligent agents, belonging to the field of data security technology. It includes defining and constructing multiple intelligent agents, determining agent attributes, and generating capability profiles for the agents based on these attributes, providing a reliable foundation for subsequent modal analysis and resource scheduling. Multimodal analysis is performed on data assets, and resource scheduling and allocation are performed based on the agent capability profiles. Output conflicts are handled through a confidence-flow arbitration mechanism. Traditional automated tools often generate numerous false positives or false negatives due to rigid rules. This system no longer simply takes the union or intersection of values, but uses a dynamically flowing confidence algorithm for automatic adjudication. The risks of data assets are quantified based on attack graphs and Bayesian networks. Reasonable scheduling and coordination of multiple intelligent agents resolve conflicts between them, improving the efficiency and reliability of automated data security assessment and meeting complex and ever-changing security assessment needs.

The invention relates to an apparatus for multimodal analysis of sample material, for example from a tissue, which detects molecular image information from the sample material in a spatially resolved manner, for example using a MALDI time-of-flight mass analyser, records light microscopy image information from the sample material in a spatially resolved manner, and associates the two with improved accuracy to give spatially resolved co-registered overall image information.

The application discloses a kind of legal dossier intelligent retrieval method and system based on multimodal heterogeneous big model, comprising: obtaining original legal dossier data containing text, image and table, carrying out multimodal analysis and structured preprocessing;Through multimodal heterogeneous big model, each modal depth feature is extracted and is fused using multimodal cross attention mechanism;Case element sparse graph is constructed and is optimized by optimization algorithm;Robust learning real value multi-index model algorithm is applied to weight adjustment and vector optimization;Through random recovery sampling technology, candidate case subset is screened and multidimensional matching calculation is carried out;Similar historical case list is visualized and interactive analysis and processing are carried out.The application improves the accuracy and efficiency of legal dossier intelligent retrieval by multimodal fusion understanding, sparse graph optimization and robust learning technology, provides accurate case reference and decision support for legal workers.

The present invention discloses a multimodal analysis system for smart wearable cardiopulmonary exercise clothing. This system belongs to the technical field of smart wearables and includes a multimodal analysis algorithm, a multisensor integration system, and wearable clothing. The wearable clothing is equipped with an inertial measurement unit sensor, an electrocardiogram sensor, a rotation axis goniometer, and a data integration center module. The multimodal analysis algorithm includes human behavior modeling and cardiopulmonary health analysis modeling. The human behavior modeling is based on data processed by the inertial measurement unit sensor and uses a hierarchical network to identify the user's specific activities: dynamic and static classification is first performed through thresholds, and then specific activities are identified using pre-trained dynamic deep learning models and static deep learning models. At the same time, the system also adaptively updates the model based on the user's uniqueness. The present invention has the effect of improving the stability of multimodal modeling and achieving accurate monitoring of cardiopulmonary exercise function.

The invention belongs to the technical field of intelligent nursing, and relates to a full-dimension intelligent nursing inspection system and method for a patient based on a virtual reality technology, and the system comprises a plurality of different types of sensors, virtual reality intelligent glasses, a mobile calculation unit, and a server. The sensor is arranged on an inspection site and is used for acquiring images and data of the inspection site; the virtual reality intelligent glasses are used for establishing a three-dimensional simulation display scene according to the image and data obtained by the sensor; the mobile calculation unit is arranged on the virtual reality intelligent glasses and is used for carrying out single-mode data calculation through an artificial intelligence model according to the image and data obtained by the sensor; and the server is used for fusing the calculation result of the single-mode data through a multi-mode analysis model to generate a full-dimensional intelligent nursing inspection result. The system can realize objectification and standardization of nursing evaluation, establish continuous and preventive monitoring, and achieve full-dimensional comprehensive evaluation.

The present invention discloses a triage method and system based on multimodal triboelectric analysis, which belongs to the field of medical triage technology, and specifically includes: providing an interactive model with a surface covered with triboelectric material, the interactive model dividing a number of anatomical mapping areas; detecting user touch events and parsing the specific anatomical mapping area positions; selecting a corresponding target sensor combination and feature analysis mode according to the type of touched area; activating corresponding sensors according to the target sensor combination to collect the user's biophysical signal data stream; executing the calculation process specified by the feature analysis mode to extract a multidimensional feature parameter set; inputting the multidimensional feature parameter set into a clinically verified symptom classification decision structure, generating a triage priority code, and transmitting it to a medical system terminal through an encrypted communication protocol; the present invention uses triboelectric technology to locate the touch position, combines biophysical signal acquisition and multimodal analysis, and quickly generates a preliminary assessment result of the condition.

The application discloses a kind of medical image repeated detection methods based on deep learning and multimodal analysis, belong to artificial intelligence and image processing field, including: obtaining the DICOM file of the medical image to be detected and preliminary verification;To the AI detection model that has been trained, parse out target image pixel array and fusion text vector;Target image pixel array is carried out gray scale conversion and standardization processing and with fusion text vector after verification standardization image tensor is obtained, feature is extracted to obtain target feature vector;Retrieve its multiple nearest neighbor historical feature vectors and associated historical DICOM file data;Based on the obtained fusion text vector, calculate the difference between the medical image to be detected and each historical medical image retrieved in feature vector dimension, image pixel dimension and multimodal check dimension, obtain similarity score, determine that there is repeated anomaly when exceeding threshold value and alarm, output detection result.The application can improve the accuracy of medical image repeated detection.

The invention discloses a multi-modal analysis traditional Chinese medicine knowledge mining method and system, and the method comprises the steps: obtaining the multi-modal data, such as a tongue image, a face image, a pulse image and an inquiry text, of the same patient, extracting features, and mapping the features to a unified traditional Chinese medicine syndrome semantic space; mining a support, supplement or contradictory relationship among different modal semantic vectors by using an attention mechanism, and constructing a cross-modal evidence collaborative network; generating a comprehensive syndrome representation based on the network fusion information, and completing preliminary differentiation; and finally, generating a final dialectical conclusion and report with credibility evaluation by quantitatively analyzing the consistency degree of the network and the integrity of the evidence chain. The system effectively simulates the thinking process of comprehensive judgment of traditional Chinese medicine doctors, not only improves objectivity and accuracy of syndrome differentiation, but also realizes quantitative evaluation of reliability of diagnosis conclusions and automatic identification of evidence contradictions, and provides an intelligent tool for clinical aid decision making and traditional Chinese medicine experience inheritance.

This invention relates to the field of multimodal analysis technology and discloses an AI-based method for stakeholder influence analysis and communication strategy optimization in IT projects. The method includes: constructing a unified terminology for an IT domain knowledge graph; achieving multi-source data fusion through anonymization and feature correlation calculation; outputting fused data; realizing comprehensive identification and dynamic role mapping of IT project-specific stakeholders; mining latent stakeholders through trajectory influence values; constructing a dynamic mapping model to track role changes; reusing trajectory influence values ​​and role suitability to complete priority ranking; extracting four types of IT-specific influence features; binding influence and technical risk through correlation strength; calculating a dynamic influence index based on a priority index; constructing and calibrating an IT project-specific communication profile; generating personalized communication strategies; performing quantitative verification of strategy suitability; and outputting communication strategies; collecting dual-channel feedback data according to the influence update cycle and simultaneously adjusting the dynamic influence index and communication strategies.

The present invention provides an information processing system comprising: a device for collecting audio data, video data, and text data; the analysis device is used for carrying out multi-modal analysis on the data collected by the equipment; a diagnosis means for diagnosing an applicable support policy using an analysis result obtained by the analysis means; the notification device is used for notifying a user of the diagnosis result of the diagnosis device; and the communication device is used for sharing the diagnosis result with a local government.

Disclosed herein are system, apparatus, article-of-manufacture, method and / or computer program product embodiments, and / or combinations / sub-combinations thereof, for personalized multimodal analysis for content item recommendation. An embodiment operates by identifying playback of a first content item by a user device, and simulating playback of a second content item with a modality feature that matches the first content items. Affinity for a modality of the second content item is identified based on weights assigned to the different modalities according to the simulated playback. Respective similarity scores are generated for a plurality of content items based on a similarity between a vector for an embedding indicative of the modality for the second content item and a respective vector for an embedding indicative of the modality generated for the content items. Indication of a set of content items with respective similarity scores that satisfy a similarity score threshold is sent to the user device.

This invention relates to the field of wearable medical device technology, specifically to a flexible wristband for health monitoring of the elderly integrating multiple sensors. It includes a smartwatch body with four symmetrically arranged grooves on the bottom. Each groove has a first electromagnetic plate installed on its inner top wall, and a column is fixedly connected to the bottom of the first electromagnetic plate. A sleeve is fitted onto the outer wall of the column. The integrated blood glucose monitoring microneedle module of this invention uses a micron-level painless needle that can painlessly penetrate the stratum corneum to obtain interstitial fluid. Combined with multimodal analysis technology, the measurement accuracy reaches medical grade, solving the problem of insufficient accuracy in traditional non-invasive devices while avoiding the discomfort of invasive blood collection. This results in high compliance among the elderly. It integrates multiple monitoring functions such as blood glucose, blood pressure, heart rate, blood oxygen, arterial stiffness, and fall detection, achieving full coverage of chronic disease management and health risk warning. It eliminates the need to wear multiple devices, meeting the needs of the elderly for joint prevention and control of multiple diseases.

The present invention discloses a rock fragment lithology analysis method based on a multi-level fusion image algorithm and X-ray diffraction. The method includes: S1: acquiring rock fragment image data at different resolutions; S2: generating a multi-level rock fragment image feature matrix; S3: preliminarily generating a lithology classification result; S4: generating supplementary data for X-ray diffraction analysis; S5: fusing the supplementary lithology analysis data with the preliminarily generated lithology classification result to ultimately generate a comprehensive lithology classification result; S6: inputting the comprehensive lithology classification result into a dynamic adaptive analysis system, which adaptively adjusts the analysis strategy based on the complexity of the rock fragment sample and optimizes the parameters of the dual-path convolutional neural network model and multimodal analysis algorithm through real-time feedback and learning mechanisms; S7: outputting an analysis report containing mineral distribution, crystal structure, and comprehensive lithology classification results. The present invention improves the classification accuracy of different types of rock fragment samples by approximately 15%.

A content retrieval system may receive a query associated with a plurality of content items in a repository. For each content item of the plurality of content items: a respective first and second similarity score may be generated based on a similarity between embeddings indicative of a first and second data type generated from the query and for the content item; and a respective normalized similarity score may be generated based on a combination of the respective first and second similarity scores. A set of content items with respective normalized similarity scores that satisfy a similarity score threshold may be identified. An exact-match (lexical) search may yield respective mapping scores for content items that may also be ranked. An output indicative of content items that are identified in the set of content items with high-ranking similarity scores and identified in the set of content items with high-ranking mapping scores.