Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

113results about How to "Efficient matching" patented technology

Intelligent visualization and text association method for multi-modal knowledge graph

The invention discloses an intelligent visualization and text association method for a multi-modal knowledge graph, and relates to the technical field of information processing. According to the intelligent visualization and text association method for the multi-modal knowledge graph, multi-modal data in an internal database of an enterprise is obtained, and a multi-modal feature vector is generated by using a data preprocessing and feature extraction technology; the feature vectors are mapped to a unified vector space through a cross-modal embedding algorithm, semantic alignment, association and fusion among multi-modal data are realized, and a complete multi-modal knowledge graph is constructed. Through bidirectional association and multi-modal joint query, the information acquisition efficiency is improved, and visual display of the knowledge graph is realized by using the graph neural network. Meanwhile, the knowledge graph is dynamically updated through a graph reasoning algorithm, and the timeliness and accuracy of the knowledge graph are ensured. A more efficient solution is provided for an enterprise in the aspects of management and application of multi-modal data, and the intelligent process of knowledge management is promoted.
Owner:LIAONING DANDELION EDUCATION CONSULTING CO LTD

Large model reasoning scheduling method based on off-network computing power server

The invention relates to the technical field of server scheduling, and discloses a large model reasoning scheduling method based on an off-network computing power server, which comprises the following steps of: collecting hardware information of the off-network computing power server, and constructing a server resource vector; decomposing a large model reasoning task into a plurality of sub-tasks, and constructing task modeling; according to the hardware information of the off-network server and the task dependency relationship, performing task division by using an optimization method; performing dynamic task scheduling on the sub-tasks based on the current load state of the server; according to the real-time calculation load of the server, dynamically migrating part of tasks to other servers to ensure the load balance among the servers; and executing the distributed task, monitoring the running state of the server and the task completion condition, and dynamically adjusting a task distribution strategy according to feedback data. Through modeling task requirements and server resources, in combination with real-time monitoring and optimal scheduling, efficient calculation and resource utilization maximization of large model reasoning in an off-network environment are realized.
Owner:BEIJING QIBU TIANXIA TECH CO LTD

Control system of photovoltaic energy storage inverter

The invention discloses a control system for a photovoltaic energy storage inverter, and the system comprises a maximum power point tracking module which is used for adjusting the output of a photovoltaic panel after the photovoltaic energy storage inverter receives a power input signal, optimizing the energy collection, a dynamic adjustment module which is used for adjusting the output of the photovoltaic panel based on the power grid fluctuation and load change parameters which are monitored in real time, the control system of the photovoltaic energy storage inverter comprises a control module used for dynamically adjusting the output frequency and voltage of the inverter and achieving efficient conversion and matching of electric energy, and a closed-loop feedback module used for continuously adjusting the output parameters of the inverter until the output parameters reach a preset energy conversion efficiency threshold value. The system is ensured to be always in an optimal operation state, and unpredictable factor changes are accurately and intelligently adapted, so that the overall conversion efficiency of the photovoltaic energy storage system is improved.
Owner:GUANGDONG SANRUI POWER CO LTD

Robot vision inertial navigation method and device and storage medium

The invention discloses a robot vision inertial navigation method and device and a storage medium, and belongs to the technical field of mobile robot navigation, and the navigation method comprises the following steps: measuring motion information of a mobile robot by using an IMU sensor; collecting image data of the mobile robot by using a camera; inputting the feature points and the descriptors into a LightGlue network model to obtain a local feature point matching relation of the image; calibrating the offset of the gyroscope by optimizing and recovering visual scale information, and estimating the IMU coordinate system speed, the gravitational acceleration and the offset of the IMU; introducing a visual feature re-projection error, an IMU pre-integration error and marginalized residual constraint, and then integrally converting into a target function to obtain a nonlinear optimization result; further repositioning is carried out to obtain a repositioning result; and optimizing the global pose map to obtain a final global pose map optimization result, and carrying out navigation operation according to the global pose map optimization result. The method can improve the positioning precision and operation efficiency under view angle change and illumination change, and is high in practicality.
Owner:NANJING UNIV OF POSTS & TELECOMM

Dynamic allocation management system for monitoring computing power resources in real time

The invention discloses a dynamic allocation management system for monitoring computing power resources in real time, relates to the technical field of computing power resource management, and is used for solving the problem that an existing dynamic allocation management system for the computing power resources cannot efficiently match, optimally utilize, accurately evaluate and dynamically allocate heterogeneous computing power resources. Comprising a central cloud and a plurality of edge nodes in communication connection with the central cloud, and a real-time monitoring prediction unit, a dynamic distribution management unit and a fault detection processing unit are arranged in the central cloud; according to the method, the utilization rate and delay information of the heterogeneous computing power resources are analyzed in detail, so that the resources most suitable for the target processing task can be efficiently matched; by eliminating the computing power resource with a relatively low delay value, the utilization efficiency of the whole resource is improved, and resource waste is avoided; and resource demand prediction is performed in combination with a plurality of parameters, so that the performance and value of the computing power resources to be allocated can be evaluated more accurately. And effective management and optimal utilization of heterogeneous computing power resources are realized.
Owner:JIANGSU CLOUD FACTORY INFORMATION TECH CO LTD

Commercial space lease transaction platform based on site selection demand distribution matching recommendation

The invention discloses a business space lease transaction platform based on site selection demand distribution matching recommendation, and relates to the technical field of business space lease, and the business space lease transaction platform comprises a user demand analysis module, a business space database, an intelligent matching algorithm, a user interface and a transaction processing module. According to the invention, through the integrated user demand analysis module and the intelligent matching algorithm, the matching efficiency and accuracy of commercial space leasing are significantly improved, and the user demand analysis module can accurately collect and analyze detailed site selection demands of users, including key information such as geographic positions, space sizes, budget and industry types. According to the intelligent matching algorithm, the machine learning technology is utilized, user requirements and information in a business space database are efficiently matched, the relevance and accuracy of recommendation results are ensured, the personalized recommendation mechanism based on the specific requirements of the user greatly reduces the workload when the user searches for a proper business space, and the user experience is improved. And the success rate of the lease transaction is improved.
Owner:HANGZHOU LINHUI NETWORK TECH CO LTD

Cultural relic classification and identification system and method based on image recognition and natural language processing

The invention discloses a cultural relic classification and identification system and method based on image recognition and natural language processing, belongs to the technical field of image processing, and aims to solve the technical problems of low efficiency and low accuracy of current cultural relic classification and identification. Comprising an image acquisition module used for acquiring a cultural relic image acquired under a shooting rule; the image processing module is used for performing feature extraction and feature classification on the cultural relics; the feature extraction module is used for performing visual feature analysis on the cultural relics, analyzing historical backgrounds and cultural meanings of the cultural relics, combining visual features and text features into feature vectors with fixed lengths, and outputting the feature vectors as cultural relic feature vectors; the feature matching module is used for performing feature matching on the basis of the cultural relic feature vector of the to-be-tested cultural relic and a cultural relic feature vector in a cultural relic database, and outputting a classification and identification result of the to-be-tested cultural relic; and the result output module is used for generating a cultural relic report based on the cultural relic image of the to-be-tested cultural relic and the classification and identification result and displaying the cultural relic report.
Owner:INSPUR SOFTWARE TECH CO LTD

Cross-regional energy supply and demand big data intelligent matching system

The invention relates to the technical field of energy supply and demand matching analysis, in particular to a cross-regional energy supply and demand big data intelligent matching system, which comprises the following modules: a data acquisition module which is responsible for collecting energy supply data and demand data of each region; the data processing module is used for cleaning, integrating and analyzing the collected data, and predicting future energy supply and demand conditions of each region by utilizing a machine learning algorithm; the intelligent matching module is used for intelligently matching an optimal energy supply and demand scheme according to an analysis result of the data processing module; the scheduling instruction module is used for sending a scheduling instruction to the energy management center of each region to realize cross-region allocation of energy; and the feedback adjustment module is used for collecting feedback information, adjusting algorithm parameters and optimizing a matching scheme according to an actual allocation effect. According to the invention, the optimization of energy distribution is ensured, and the loss and cost in the energy transmission process are reduced.
Owner:BAOYING COUNTY POWER SUPPLY BUREAU +2

Program defect automatic repair method, system and equipment based on retrieval enhancement and readable storage medium

The invention relates to an automatic program defect repairing method based on retrieval enhancement, which comprises the following steps of: S1, collecting a repairing template, and constructing a repairing template database; step S2, retrieving a matched repair template from a repair template database; s3, calculating the lexical similarity and the semantic similarity of the context code line and the error code line; s4, based on a cross-correlation ranking fusion strategy, generating a comprehensive similarity ranking list of the code lines; s5, based on the comprehensive similarity ranking list, candidate identifiers are extracted from the similar code lines; s6, constructing a restoration prompt based on context enhancement; inputting the repair prompt into the language model to generate candidate patches; and S7, verifying the candidate patches. According to the method, historical repair template retrieval and local related code retrieval are creatively combined, the multi-dimensional similarity analysis and static analysis technologies are utilized, a context-enhanced repair prompt is generated, and the effectiveness of patches is ensured through an automatic verification mechanism.
Owner:HARBIN INST OF TECH

IB network transmission rate adaptive system and method

The invention discloses an IB network transmission rate adaptive system and method, and the system comprises a link training module which is used for building and maintaining a link; the data sending module sends a signal according to the training data sent by the link training module; the data receiving module receives and decodes the ordered set sent by the opposite end from the high-speed interface, and obtains the maximum rate supported by the devices at the two ends; the rate adjusting module is used for acquiring corresponding clock and DRP serial numbers according to the link rate adjusting signal sent by the link training module and outputting corresponding clock and DRP configuration to a high-speed interface; and the link test module is used for sending the pseudo-random code and counting the bit error rate of the received pseudo-random code after the high-speed interface is switched to the maximum rate supported by the two ends, and the link test module also reports the bit error test result to the link training module. According to the IB network transmission rate adaptive system and method provided by the invention, a scheme supporting rate coding ordered set and link quality testing is adopted, so that the problem of synchronization mismatching with opposite terminal equipment is avoided.
Owner:ZHEJIANG UNIV

Dual-fuel combustion nozzle for liquid hydrocarbon fuel and hydrogen fuel

The invention provides a dual-fuel combustion nozzle which comprises a nozzle shell connected with a cooling plate. A hydrogen fuel gas collecting cavity and a liquid hydrocarbon fuel pipe which are coaxial are arranged in the nozzle shell; a first air channel is formed between the inner wall of the hydrogen fuel gas collecting cavity and the liquid hydrocarbon fuel pipe, and a swirler is arranged in the first air channel; a second air channel is formed between the outer wall of the hydrogen fuel gas collecting cavity and the inner shell of the nozzle shell; a liquid hydrocarbon spray hole is formed in the outlet end of the liquid hydrocarbon fuel pipe and is positioned in a space formed by the expansion section of the venturi structure; a hydrogen spray hole is formed in the side wall, close to the top end, of the hydrogen fuel collecting cavity. The invention further provides an engine combustion chamber comprising the combustion nozzle. According to the invention, efficient mixing of hydrocarbon fuel and hydrogen fuel can be realized, the phenomenon of backfire spontaneous combustion is avoided, the flameout boundary is widened, and stable switching of multiple working modes is realized.
Owner:BEIHANG UNIV

Fast self-adaptive impedance matching method and system and wireless power transmission system

The invention discloses a fast self-adaptive impedance matching method and system and a wireless power transmission system, an MCU enables a pi-type impedance matching network to be in a short circuit state by controlling a relay switch, two groups of reflection coefficients of the system are obtained through measurement of an amplitude phase measurement circuit, and two groups of input impedance are calculated; and calculating to obtain a theoretical optimal capacitor combination corresponding to each input impedance. And actually measuring the reflection coefficients of the two input impedances in the corresponding theoretical optimal capacitor combination, wherein the capacitor combination with the smaller reflection coefficient is the actual optimal capacitor combination. Furthermore, according to the method, a local optimal solution is searched around an actual optimal capacitor combination, and re-impedance matching is completed. And dynamically monitoring the input impedance of the system under the given frequency, and repeating the steps when the change of the input impedance exceeds a set threshold value. According to the method, theoretical calculation is combined with hardware testing, the matching efficiency of the system is greatly improved, in addition, the influence of parasitic parameters on the matching precision of the system can be effectively avoided through a local optimization control algorithm, the transmission efficiency and stability of the system are further improved, and good practicability is achieved.
Owner:CHENGDU UNIV OF INFORMATION TECH

Intelligent hospital guide method and system based on symptom matching and department recommendation

The invention discloses an intelligent hospital guide method and system based on symptom matching and department recommendation, and the method comprises the following steps: receiving a multi-modal symptom description inputted by a user, and carrying out the combined processing, thereby obtaining multi-modal symptom information; inputting the multi-modal symptom information into a multi-modal symptom deep learning model to obtain text key symptom features and voice key symptom features, and performing feature fusion to obtain comprehensive symptom feature representation; performing multi-dimensional matching according to the comprehensive symptom feature representation to generate a candidate department list; acquiring the real-time position of the user, inputting the real-time position of the user, the candidate department list and the voice key symptom features into a department recommendation model together, outputting a recommended department by calling a # imgabs0 # minimum spanning tree path algorithm, and generating a department information graph for the recommended department; and displaying department information, a doctor-seeing path and an appointment mode for the user through the department information atlas, thereby realizing intelligent hospital guide. According to the invention, the symptom information of the user is accurately identified and efficient matching with related departments is realized.
Owner:ZHUHAI HUARAN TECH CO LTD

Intelligent traffic asset data management method and system

The invention discloses an intelligent traffic asset data management method and system, and relates to the technical field of data management, and the method comprises the steps: carrying out multi-dimensional analysis to determine a data classification target label, constructing a machine learning model, obtaining an intelligent data classification model, and carrying out the recognition and classification of asset data. Generating a data label to label data in the traffic asset database; establishing a response association relationship between the traffic asset data and a data application scene, and performing data label dependency and transparency analysis to obtain a data calling constraint; and searching and filtering in the traffic asset database according to the data label in the data calling constraint, and outputting scene application data. The technical problems that in the prior art, due to the fact that the incidence relation between data and application scenes cannot be preset in the data management process, search condition setting is complex, data matching is not accurate, and omission exists during data query and application are solved, and the technical effect of improving data management efficiency is achieved.
Owner:SHANDONG ZONGYUN INFORMATION TECH CO LTD

Video identification method based on dynamic space-time full-scale feature fusion

The invention discloses a video recognition method based on dynamic space-time full-scale feature fusion, which comprises the following steps: a video data set used for training a deep neural network model is acquired, the video data set comprises a plurality of video sequences, and each video sequence is composed of a plurality of time frames; training a pre-defined video recognition deep neural network model by using the video data set, wherein the video recognition deep neural network model comprises a parallel space-time adapter and a self-prompt generator; and applying the trained video recognition deep neural network model to a video recognition task, inputting a to-be-recognized video sequence, and outputting a classification prediction result of the video. The method has the advantages that through cooperative work of the parallel space-time adapter and the self-prompt generator, efficient modeling of dynamic changes in the video is achieved, the accuracy and robustness of video recognition are remarkably improved, and the method is suitable for various video recognition tasks such as supervised learning, few-sample learning and zero-sample recognition.
Owner:ZHEJIANG UNIV

Multi-view vision system for brain magnetic automatic registration

The invention provides a multi-view vision system for brain magnetic automatic registration, and the system comprises a rigid helmet which is provided with a plurality of known coordinate marking points at preset positions, and is used for providing initial coordinates of a sensor and assisting helmet-head registration; the multi-view vision camera system is arranged around the subject and is used for capturing feature points of the helmet and the head and the face of the subject; and the registration system is used for realizing registration from the helmet to the head based on the helmet mark points and registration from the head to the MRI based on the face feature points. The accuracy and reliability of brain magnetic signal acquisition can be improved, and powerful technical support is provided for brain function research and clinical diagnosis.
Owner:BEIHANG UNIV

Geofence generation and matching method and device, electronic equipment and storage medium

The embodiment of the invention discloses a geo-fence generation and matching method and device, electronic equipment and a storage medium. The method comprises the steps that intersecting hash rectangles of a target area are searched step by step according to the sequence of levels from high to low; for the searched intersecting Hash rectangles of each level, after the intersecting Hash rectangles of the current level are found, determining target intersecting Hash rectangles based on the intersecting Hash rectangles of the current level and target accuracy; and determining a fence hash code corresponding to the target area based on the target intersection hash rectangle, and generating a geo-fence of the target area based on the fence hash code, wherein the intersected Hash rectangles are geographical Hash rectangles intersected with the boundary of the target area, the area represented by the highest-level geographical Hash rectangle is the largest, and the area represented by the lowest-level geographical Hash rectangle is the smallest. According to the embodiment of the invention, the geo-fences of the target area with relatively small data volume can be quickly and accurately generated, and the matching process of the geo-fences can be quickly and efficiently carried out with relatively small calculation amount.
Owner:SHANGHAI HUACE NAVIGATION TECH +1

Candidate geographical location information and retail store-based person and post matching system

The invention discloses a person and post matching system based on candidate geographical location information and retail stores, and relates to the technical field of post matching systems. The candidate geographic position information and retail store-based person and post matching system comprises a candidate information management module, a retail store information management module, a commuting data analysis and path planning module, a candidate and store matching module and a result display and feedback adjustment module. By collecting and processing commuting modes and time period preferences of candidates, a commuting path is planned in combination with a map service API, and commuting time considering congestion and congestion influence is calculated, so that the matching accuracy and practicability are improved. In addition, the system also combines factors such as working experience, skills and the like to carry out matching result optimization, so that bidirectional selection is realized, and the matching satisfaction degree is improved. And the result display and feedback adjustment module enables the user to visually see the matching result.
Owner:BEIJING WEI LAI XIAO LIE TECH CO LTD

Method and system for evaluating matching degree of production research and development project and park

The invention relates to the technical field of park investment attraction, in particular to a production research and development project and park matching degree evaluation method and system, and the method comprises the steps: obtaining the target demand data of a production research and development project, and constructing a production research and development project demand map; obtaining park space feature data, and constructing a park space feature map; obtaining the number of isomorphic sub-graphs by adopting a sub-graph isomorphism algorithm; and according to the number of the isomorphic sub-graphs and a matching degree scoring formula, evaluating the matching degree between the park spatial features and the production research and development project requirements. According to the method, accurate matching is carried out by utilizing a subgraph isomorphism algorithm through the constructed production research and development project demand map and the park space feature map, a quantitative matching degree score is provided, and the matching degree between park space features and production research and development project demands is evaluated, so that an enterprise is helped to efficiently match a proper park space, and the production research and development project demand matching efficiency is improved. And more scientific and reasonable decision support is provided for enterprises.
Owner:BEIJING YIZHUANG TECHNOLOGY INNOVATION CO LTD

Photovoltaic energy storage air conditioning system and power supply control method thereof

The invention provides a photovoltaic energy storage air conditioning system and a power supply control method thereof. The power supply control method comprises the steps that the generating capacity W of a photovoltaic module, the stored electric quantity Z of an energy storage device and the total load demand electric quantity Q of a unit are obtained; when W is smaller than Q, according to the ratio of W / Q, the energy storage device is selected to supply power to a preset high-power load and a preset low-power load in the unit, or the photovoltaic module is selected to supply power to the preset low-power load in the unit; the photovoltaic technology is applied to the air conditioning unit, clean energy solar energy is used for supplying power, dependence on traditional energy is reduced, and energy consumption is remarkably reduced; and meanwhile, electric energy generated by the photovoltaic module is reasonably utilized, efficient operation of the air conditioning unit under different working conditions is ensured, and maximum use of the electric energy is achieved.
Owner:GREE ELECTRIC APPLIANCE INC OF ZHUHAI

Intelligent voice interaction recruitment matching method and system based on large language model

The invention relates to the technical field of intelligent recruitment matching, and particularly discloses an intelligent voice interaction recruitment matching method and system based on a large language model, and the method comprises the steps: receiving recruitment demands and interview questions uploaded by a recruitment end, and storing the recruitment demands and interview questions in a knowledge base; receiving a job hunting request which is uploaded by a job hunter and contains resume information, retrieving the knowledge base based on the resume information, and screening out a matched recruitment end; sending the matched recruitment end as a recommended post to the job seeker, and receiving a selection request of the job seeker; an interview environment is created based on interview questions in the knowledge base, interview is performed on the job seeker, and an evaluation result is generated; and generating an interview condition oriented to the recruitment end and job hunting suggestions oriented to the job hunters based on the evaluation result. According to the invention, the matching degree between the job seeker and the post is improved, the occurrence of invalid interviews is reduced, the job seeker is helped to better prepare interviews, the interview passing rate is improved, the recruiter can more quickly screen out proper candidates, and the recruitment efficiency is improved.
Owner:ZHEJIANG WANYOUMALI NETWORK TECHNOLOGY CO LTD

Sea area management and control early warning evidence obtaining method based on video intelligent analysis

The invention relates to a sea area management and control early warning evidence obtaining method based on video intelligent analysis, and the method comprises the steps: collecting equipment parameters: collecting internal and external parameters of equipment and environment parameters; equipment screening control is carried out, controllable and controllable equipment with a good detection effect is selected through equipment state analysis and target detection effectiveness analysis, and follow-up monitoring of a target area is realized through pan-tilt control; performing video intelligent analysis, determining a follow-up concerned target through mapping matching of a video intelligent analysis model and different coordinate system positioning information, and performing accurate linkage and video snapshot by fine tuning equipment; according to early warning and evidence obtaining association, structured attribute analysis is performed on a target and an area detected in a video capturing process, and an evidence obtaining result is added to early warning information with an association relationship, so that repeated evidence obtaining is avoided. According to the invention, the accuracy and efficiency of early warning evidence obtaining can be effectively improved, the utilization efficiency of data and resources is improved, and powerful technical support is provided for sea area management and control.
Owner:CHINA CHANGFENG SCI TECH IND GROUPCORP

Digital twinborn rule engine method and system based on intelligent oil field station

The invention relates to the technical field of intelligent oil fields, in particular to a digital twinborn rule engine method and system based on an intelligent oil field station yard, and the method and system comprise the steps of design and implementation of a rule engine system, rule configuration and testing, and system integration and deployment. The design and implementation of the rule engine system are as follows: A1, digital twin platform integration: virtual and real mapping and synchronization: the digital twin platform collects real-time data of an oil field station through Internet of Things equipment and a sensor; independent configuration and dynamic adjustment of business rules are achieved through a rule engine, limitation of a hard coding mode is avoided, the flexibility and maintainability of the system are enhanced, multi-dimensional rule management supports construction and representation of multi-dimensional rules, the scheduling requirement in a complex business scene can be met, and the scheduling efficiency is improved. And the accuracy and efficiency of production management are improved.
Owner:SHENZHEN JIAYUNTONG ELECTRONICS

Remote sensing image matching method and device based on meta-learning

The invention discloses a remote sensing image matching method and device based on meta-learning, and the method comprises the steps: selecting a first preset number of image pairs from an existing data set, carrying out the pairing marking of salient region points on each pair of images, enabling each pair of images to have a second preset number of matching pairs, and constructing a metadata set; executing a transformation operation on one image in each image pair from the existing data set to construct a coarse annotation data set; constructing a feature pairing network and a significance judgment network; and iteratively executing a double-layer optimization step through different data sets until the approximate optimal parameter values of the feature pairing network and the significance judgment network do not change significantly or reach a preset number of iterations, so as to obtain corresponding optimal parameter values. The method solves the problems that a traditional feature extraction operator is difficult to effectively extract significant features between different-source remote sensing images and is difficult to deal with complex and changeable practical application scenes.
Owner:XI AN JIAOTONG UNIV

Mobile phone antenna with high port isolation

The invention provides a mobile phone antenna with high port isolation, which comprises a bottom substrate, a left substrate, a right substrate, an upper substrate and a lower substrate, the left substrate, the right substrate, the upper substrate and the lower substrate are vertically arranged on the bottom substrate, a metal floor covers the outer side of the bottom substrate, and metal patches electrically connected with the metal floor cover the outer sides of the left substrate, the right substrate, the upper substrate and the lower substrate; the metal patches of one of the left substrate and the right substrate are provided with slits, and the metal patches are divided into long metal patches and short metal patches. A first transmission line and a second transmission line which are connected with the short metal patch are arranged on the metal floor; a first impedance matching element is arranged on the first transmission line, a second impedance matching element is arranged on the second transmission line, a neutralizing line is connected between the first transmission line and the second transmission line, and an isolating element is arranged on the neutralizing line, so that high isolation of the antenna among different frequency bands is realized, coexisting interference in equipment is reduced, and the service life of the antenna is prolonged. The antenna can be well matched with signals of different frequency bands, and the transmission efficiency of the signals is improved.
Owner:FUDAN UNIVERSITY

New media anchor matching recommendation system and method based on deep learning and data analysis

ActiveCN119557344Areflect diversityRealize precise excavationData miningSpecial data processing applicationsMediaFLOEngineering
The invention relates to a new media anchor matching recommendation system and method based on deep learning and data analysis, which has the effect of improving the matching accuracy and effectiveness of a product and a new media anchor, and comprises the following steps: screening historical purchase orders about to-be-marketed product information from a commodity marketing database; the to-be-marketed product information comprises a product name; calculating a first proportion of the number of times that each feature tag appears in the historical purchase orders to the number of all the historical purchase orders; generating a key feature tag set by using the feature tags of which the first proportions exceed a preset proportion; distributing an initial weight value for each feature tag in the key feature tag set based on the size of the first proportion of each feature tag; generating a real weight value of the reference feature tag; calculating the matching degree of the concerned user information of each new media anchor based on the real weight value of each feature tag and the concerned user information of each new media anchor in a new media anchor database; and sorting and generating a recommended new media anchor sequence according to the matching degree.
Owner:SHENZHEN QIANHAI HOFAN TECH CO LTD

GPU-based medical endoscope image processing method, system and medium

The present invention belongs to the field of medical image processing, and provides a method, system and medium for processing medical endoscopic images based on GPU. The present invention utilizes the powerful parallel processing capability of GPU to significantly improve the speed and accuracy of endoscopic image acquisition, preprocessing, feature point detection, image correction and pathological feature recognition. The image quality is enhanced by executing GPU-accelerated filtering algorithms and histogram equalization, while the GPU-optimized SIFT algorithm effectively improves the accuracy of feature point matching. The estimation of the affine transformation model and the transformation correction of the image further correct the distortion and ensure the accuracy of the image data. Finally, the pre-trained deep learning model is used to implement fast convolutional neural network forward propagation on the GPU, identify pathological features in real time, and assist clinical diagnosis. The efficiency and accuracy of clinical diagnosis are greatly improved, providing doctors with instant diagnostic assistance, thereby improving the quality of medical services.
Owner:GUANGZHOU SOFTLINK MEDICAL INNOVATION CO LTO

Comprehensive guarantee order dispatching method and system for large-scale service scene

PendingCN121961036AAccurately depict space-time constraintsAccurately characterize dynamic disturbancesInstrumentsRoad networksOperations research
The invention discloses a comprehensive guarantee order dispatching method and system for a large-scale business scene, and relates to the task distribution and data processing technology, and the method comprises the steps: obtaining a to-be-dispatched order, and enabling any order to be distributed to a corresponding vehicle; establishing a dynamic vehicle path planning problem DDVRPTW model with a time window according to the vehicle condition, the road network condition and the user demand when the vehicle executes the order task; establishing constraint conditions according to actual orders and vehicle conditions; for different order dispatching scenes, selecting a corresponding adaptation algorithm to solve the DDVRPTW model; and according to a solving result, the order to be dispatched is dispatched. According to the method, the dynamic vehicle path planning problem model with the time window is constructed, the spatial-temporal constraint and the dynamic disturbance of the guarantee task are accurately described and displayed, and efficient matching of resource demand and supply is realized.
Owner:NO 15 INST OF CHINA ELECTRONICS TECH GRP

Pico-meter surface shape measuring device and method for large-aperture optical element

PendingCN121953865ARealize switchingAchieve displacement measurement resolutionUsing optical meansAcousto-opticsMeasurement precision
The invention relates to a picometer-level surface shape measuring device for a large-aperture optical element. The picometer-level surface shape measuring device comprises a laser, a first polarization beam splitter, a first acousto-optic modulator, a second polarization beam splitter, a quarter-wave plate, a beam expander group, a local gating measuring surface unit, a 45-degree reflector, a second acousto-optic modulator, a non-polarization beam splitter and a photoelectric detector. The invention further provides a picometer-level surface shape measurement method for the large-aperture optical element, high-precision phase information is obtained based on the double-frequency heterodyne interference principle, regionalized scanning is achieved on the surface of the measured optical element through the local light gating module, and full-aperture high-resolution surface shape measurement can be completed without depending on a mechanical moving platform. The device has the advantages of high mechanical error resistance, high measurement precision, simple system structure and the like, and is suitable for surface shape detection of ultrahigh-precision optical elements.
Owner:SHANGHAI INST OF OPTICS & FINE MECHANICS CHINESE ACAD OF SCI