163results about How to "Solve the fusion problem" patented technology

The invention discloses a device, a system and a method for realizing intelligent household application. The device comprises an intelligent household gateway, a family control center and terminal agent equipment, wherein the intelligent household gateway is used for being connected with the terminal agent equipment to finish signaling transfer between the family control center and the terminal agent equipment and transmit the transferred signaling to the family control center or the terminal agent equipment; the family control center is used for executing an intelligent household control logic according to the received user command or the preallocated task, transmitting the command to local terminal equipment through the intelligent household gateway via the terminal agent equipment, and realizing an intelligent household service through the local terminal equipment; and the terminal agent equipment is used for connecting the local terminal equipment to allocate identifiers to the local terminal equipment and finishing underlying protocol transfer between the intelligent household gateway and the local terminal equipment. The invention can realize intelligent household application on the basis of uniformly controlling the terminal equipment.

The invention requests to protect a multifunctional V2X intelligent roadside base station system. The system comprises roadside sensing equipment, an MEC server, a high-precision positioning service module, a multi-source intelligent roadside sensing information fusion module and a 5G / LTE-V communication module. An intelligent roadside device integrating C-V2X communication, environmental perception and target recognition, high-precision positioning and the like is designed, and the problem that multi-device information fusion and integration are inconvenient in intelligent transportation is solved. In the system, a C-V2X intelligent road side system architecture and a target layer multi-source information fusion method are designed. Road side multi-source environment cooperative sensing is combined, real-time traffic scheduling of the intersection is realized by using a traffic scheduling module in the MEC server, and communication and high-precision positioning services are providedfor vehicle driving, and finally the target information after fusion processing is broadcasted to other vehicles or pedestrians through a C-V2X RSU (LTE-V2X / 5G V2X and the like) according to an application layer standard data format, so the driving and traffic safety is improved.

The invention discloses a smart phone-based indoor positioning and pedestrian tracking method. The method comprises the following steps: 1, preliminarily determining the position and the locus of a sensor in a phone; 2, fusing and preprocessing indoor information; 3, assisting the preliminarily positioned position information by using the fused information; and 4, calculating an accurate linear locus according to the fused information by using an improved equalization algorithm, and correcting the angle error of a pedestrian at a corner by using an improved algorithm in order to integrally realize accurate positioning and pedestrian locus tracking. The method solves the partial fusion problem of existing indoor position, integrates the functions of the smart phone, ingeniously fuses all indoor position through the improved algorithm, improves the positioning and locus accuracy of the indoor pedestrian, utilizes the functions of the daily necessity phone in life, and has the advantagesof low consumption and convenience.

The invention discloses a fusion query method and device of heterogeneous multi-source data, and relates to the field of big data. The method includes: receiving a data-source query request, and confirming connection parameters, which correspond to data sources, in a data model according to a query format; determining a data query adapter and data sources corresponding to the data-source query request according to types of the data sources and storage manners of the types of the data sources in a data memory; optimizing query fields in the query format according to a data-source association manner and screening conditions in the data-source query request, and obtaining first query results from the data memory according to the optimized query fields; and carrying out association fusion on the first query results according to data-source association fields and data-source association parameters in the data-source query request to obtain second query results.

The invention discloses a whole bridge deck moving load spatial distribution monitoring system based on dynamic weighing and multi-video information fusion. A dynamic weighing system is arranged at the starting position of the bridge deck to obtain the vehicle weight information of the moving load, A plurality of cameras are arranged along the bridge to obtain video information of traffic flow covering the whole bridge deck, There is an overlap area in the field of view of two adjacent cameras, information collection and network communication system is arranged on site, and deploy the information processing and analysis master control system in the cloud, Combined with Kalman filter and visual field demarcation line method to track vehicle trajectory and position in visual field, Accordingto the time that the vehicle crosses the online slot of the piezoelectric sensor of the dynamic weighing system, the dynamic weighing and monitoring multi-video information are fused to realize the accurate and real-time identification of the magnitude and position of the moving load in the whole bridge deck. The invention has the advantages of fast calculation speed and high identification accuracy, and is suitable for monitoring the full coverage of various bridge decks or pavement moving vehicle loads in a designated area.

The invention belongs to the computer application field, and discloses an intelligent hospital guide method and system integrated based on multi-source knowledge graphs, for solving the problem of accurate hospital guide of a patient. The intelligent hospital guide method includes the steps: constructing a symptom-disease-department knowledge graph; collecting user symptom description informationand gender and age group information; identifying a symptom set and a patient pre-judging disease set in the user symptom description information through an entity identification and symptom normalization process; calculating the symptom set of the user through an improved TF-IDF algorithm to obtain a possible disease set of the user and the weight of each disease; calculating to obtain the weightof a recommended department by fusing the disease weight and the disease-department correlation degree by utilizing the disease set which the user possibly suffers from, the weight of each disease and the symptom-disease-department knowledge graph; and determining recommended department according to the rank of the department weights.

The invention relates to an electronic product quality automatic detection method based on a knowledge graph, and belongs to the field of electronic product quality detection and knowledge graphs. Themethod comprises the steps of constructing a domain knowledge graph and realizing a quality detection method; constructing a product domain knowledge graph through related standard literature and technical specification of the electronic product. The electronic product quality detection process is detected from four aspects. The process comprises the steps of adopting an ontology modeling methodto manage a mode layer of the knowledge graph; constructing a data layer on the basis of the mode layer; extracting entities and entity relationships from a plurality of heterogeneous data sources byadopting a Bi-LSTM + Attention + CRF method. The data fusion generally adopts a similarity calculation method. A data layer stores knowledge in a Neo4j graph database by taking facts as units. The product quality detection is carried out by utilizing a product standard domain knowledge graph, and a front-end module is added and is used for directly querying quality detection data.

The invention discloses an expansion full-convolution neural network and a construction method thereof. The neural network comprises a convolution neural network, a feature extraction module, and a feature fusion module connected in order; the construction method comprises the following steps: selecting the convolution neural network; removing a full-connection layer and a classification layer forclassifying in the convolution neural network, and only leaving the middle convolution layer and a pooling layer, and extracting a feature map from the convolution layer and the pooling layer; constructing a feature extraction module, wherein the feature extraction module comprises multiple expansion upper-sampling modules connected in series, each expansion upper-sampling module respectively comprises a feature map merge layer, an expansion convolution layer and a deconvolution layer; and constructing a feature fusion module, wherein the feature fusion module comprises a dense expansion convolution block and a deconvolution layer. The expansion full-convolution neural network disclosed by the invention effectively solves the feature extraction and fusion problem in the convolution neuralnetwork, and can be applied to a pixel-level labelling task of an image.

The invention discloses a method for generating an animation based on a human face image, and the method comprises the steps: extracting human face feature points of the human face image; generating a human face three-dimensional model through the human face feature points and a first projection matrix corresponding to the human face three-dimensional model; calculating and obtaining a full-image three-dimensional grid model of the human face image through the human face three-dimensional model; calculating full-image texture coordinates according to the full-image three-dimensional grid model and the first projection matrix; generating a second projection matrix through modifying the parameters of the first projection matrix; carrying out the projection of the full-image three-dimensional grid model and the human face three-dimensional model through the first projection matrix and the second projection matrix, and generating an image after three-dimensional reconstruction; carrying out the texture mapping of the image through the full-image texture coordinates after three-dimensional reconstruction, and generating a processed image; sequentially carrying out the steps of parameter modifying, projection and texture mapping in a repeated manner, and obtaining a plurality of processed images; and generating an animation according to the plurality of processed images. The invention also provides calculating equipment executing the above method.

The invention discloses a positioning method of a working object and a target of a substation based on infrared and visible light fusion. The technical scheme is as follows: by using a visible light and infrared sensor system deployed in the substation site, real-time images are collected, and a distributed image processing platform is used to remove the noise, fuse and enhance the pretreatment, then the target and the background are segmented, target features are extracted to detect a site intrusion target, and then dynamic target identification, lactation and tracking are performed. It is divided into three stages: dynamic target detection and identification based on 3D scene and multi-spectral image fusion, target location based on a multi-camera and target tracking based on a 3D scene,and a corresponding algorithm prototype is given. The invention adopts a fusion method based on a three-dimensional scene model, utilizes the spatial corresponding relationship between visible lightand infrared images to transform, solves the fusion problem under the condition of large field of view and multiple angles, obtains a better effect, and obtains more complete and reliable image information.

The invention belongs to the technical field of big data, and discloses a multi-source data aggregation sampling method and system based on a big data environment, and the method comprises the steps:collecting a plurality of original data sources, wherein each original data source comprises a data source name and at least one association domain; cleaning and identifying the acquired data source,and removing redundant of the acquired data source; obtaining an original strategy list by utilizing a construction program according to the original data source, and sorting the original strategies in the original strategy list to form a strategy list between the data sources; carrying out fusion processing on different source data sets by utilizing a fusion program; carrying out word segmentation on the fused file to form a two-dimensional word frequency matrix of file words; setting a balance verification numerical value, circularly matching each word, and carrying out snowball sampling; and displaying the acquired multi-source data by using a display. According to the method, distributed computing is completed by scheduling the computing nodes by the Spark through the preprocessing module, more efficient data preprocessing can be achieved, practicability is high, and the application range is wide.

The invention discloses a visibility measuring method based on image study, which includes preprocessing a scene image shot by an existing camera, extracting local contrast fitting human eye vision characteristics in the image to use the local contrast as image characteristics for machine learning, proving training data, removing large-error values in the training data, and resetting up a model. Current visibility value is automatically calculated through one scene image. Errors caused by noise of a transmission channel in the image and the like and deviation in the study training data are considered in the method, so that accuracy of visibility measuring results is improved. The visibility measuring method is applicable to real-time measurement of meteorological visibility under any weather conditions (fogging, raining, snowing, dust blowing and the like), and is applicable to places such as roads, meteorological stations and the like requiring visibility monitoring.

The invention discloses a system and a method for multi-platform network video monitoring based on an SIP protocol. The system comprises front-end equipment, front-end monitoring stations, sub monitoring and management centers and a main monitoring and management center, wherein the front-end equipment is accessed to the sub monitoring and management centers through the front-end monitoring stations, the main monitoring and management center monitors and manages the sub monitoring and management centers and the directly-managed front-end equipment based on the SIP protocol, and the sub monitoring and management centers monitor and manage the accessed front-end equipment based on the SIP protocol and through the front-end monitoring stations. The problem that existing video monitoring systems have a single monitoring range and the monitoring systems are difficult to be integrated and compatible is solved.

The invention discloses a big data loading system. The big data loading system comprises a client side for receiving user data, a data storage system for storing the user data, and a meta-database for storing management information of the user data, wherein the management information includes the data structure, the data property and the data storage position. The big data loading system further comprises a loading server, wherein the loading system is connected with the client side, receives the user data from the client side and loads the user data into the data storage system according to state information of the data storage system and the management information of the meta-database. The invention further discloses a big data loading method. According to the big data loading system and method, effective combination and unified storage management of mass structured data, semi-structured data and unstructured data are achieved, and the possibility is provided for linkage, analysis and lookup of the mass-structured data, the semi-structured data and the unstructured data.

The invention discloses a complex product heterogeneous model fusing method. The method is mainly prepared through the following steps of firstly, extracting geometric feature and processing shape and position error information of a product model at different stages; secondly, establishing a voxel model of product uniform expression through a voxelization method; thirdly, establishing mapping of distance field intersection, subtraction and complementation operation and speed field operation functions and heterogeneous model primitive feature, shape and position error and high-grade feature operation; fourthly, fusing product heterogeneous primitive feature and processing shape and position error information into the voxel model; fifthly, reestablishing a voxel model surface model through a movable cuboid method so that the visual expression of the fused model can be achieved. The unified framework of product heterogeneous information processing is provided, heterogeneous model fusing integration and uniform visual expression in different CAX systems are achieved, the effective approach of product multi-major and multi-performance analysis and evaluation is provided, complex product design efficiency can be effectively improved, the complex product development cycle can be shortened, and product competitiveness can be greatly improved.

The invention relates to a cooperative frequency spectrum sensing method based on a fuzzy integral and optimization theory, which comprises the following steps of: respectively carrying out local frequency spectrum sensing by all sensing users and performing two assumptions according to the existence condition of a main user signal; respectively transmitting detecting statistical data to a fusing center; respectively setting two fuzzy integrators, four fuzzy measuring degrees and corresponding fuzzy evaluating methods for the two assumptions by the fusing center, respectively transmitting two different training sequences of the existence of the main user signal, obtaining an optimal numerical value of the four fuzzy measuring degrees by an optimization method and respectively calculating fuzzy integral values of all the sensing users in the two fuzzy integrators on the basis of the obtained optimal fuzzy measuring degree; and finally performing final judgment according to a preset judging policy. The invention leads a final judging result of the fusing center to fully consider the uncertainty of detecting results of all the sensing users by the fuzzy integral and optimization theory and greatly improves the detecting accuracy of the whole sensing radio system frequency spectrum.

The invention relates to a distributed data processing method and device, computer equipment and a storage medium, and relates to the technical field of artificial intelligence. The method comprises the steps that model training information sent by at least two edge node devices is acquired, wherein the model training information is transmitted in a plaintext form, and the model training information is obtained by training a sub-model by the edge node equipment in a differential privacy mode; on the basis of the model training information sent by the at least two pieces of edge node equipmentrespectively, sub-models obtained by training the at least two pieces of edge node equipment respectively are obtained; and on the basis of a specified model integration strategy, model integration iscarried out on the sub-models obtained by training the at least two pieces of edge node equipment to obtain a global model. By means of the scheme, on the premise that data privacy safety is guaranteed, the model integration mode is expanded, and the model integration effect is improved.

The invention discloses a method for connecting kernel network to MPLS VPN, firstly configuring a VLAN logic subinterface on GGSN external interface, and configuring a VLAN logic subinterface on opposite-end PE, making their IP and ARP intercommunicate, then establishing APN-VLAN mapping relationship on GGSN and simultaneously establishing VLAN-VPN mapping relationship on the PE, establishing PDP context on control layer according to the APN-VLAN-VPN mapping relationship, and interacting with MPLS VPN on user layer according to the PDP context and corresponding VLAN information, so as to connect the kernel network to MPLS VPN. By VLAN technique, it establishes APN-VLAN-VPN mapping relationship in GGSN, and connects a user to corresponding MPLS VPN by this mapping relationship.

The invention provides a panoramic vision system with a synchronous shot parameter adjusting function. The panoramic vision system with the synchronous shot parameter adjusting function comprises a camera, an installing disc, a rotating device and an upholder, the camera and the rotating device are installed on the installing disc, the upholder is connected with the rotating device, the rotating device controls the installing disc to rotate relative to the upholder, the installing disc is provided with a synchronous focus adjusting gear ring, a synchronous focal length adjusting gear ring and a synchronous aperture adjusting gear ring, the synchronous focus adjusting gear ring, the synchronous focal length adjusting gear ring and the synchronous aperture adjusting gear ring can rotate on the installing disc in a sliding mode, the camera is provided with a focus gear, a focal length gear and a aperture gear, and the synchronous focus adjusting gear ring of the installing disc, the synchronous focal length adjusting gear ring of the installing disc and the synchronous aperture adjusting gear ring of the installing disc respectively mesh with the focus gear of the camera, the focal length gear of the camera and the aperture gear of the camera. The panoramic vision system with the synchronous shot parameter adjusting function overcomes error accumulation caused by independent control of a camera parameter and guarantees that parameter errors of all cameras are within an acceptable range.

The invention provides a BIM and GIS integration method based on semantic fusion. The BIM and GIS integration method comprises the steps: S1, building a unified semantic framework of BIM and GIS integration, and forming a shared domain ontology; S2, obtaining instance information from the BIM data model to fill the shared domain ontology; S3, semantizing the GIS data and filling the GIS data intoa shared domain ontology in an ontology alignment mode; and S4, realizing information query and reasoning based on the fused domain knowledge model. The BIM and GIS integration method researches a method for converting IFC and CityGML into an RDF graph model, so as to enable the heterogeneous data to be expressed by a semantic-based knowledge model, thus solving the fusion problem of heterogeneousinformation of BIM and GIS, classifies, maps and hierarchically organizes classes and relations in the field to form concept integration under a unified framework, well processes and displays representation of knowledge, and infers implicit knowledge through establishment of logic rules.

The invention discloses a service data transmission method and a service data transmission device. The method comprises the following steps of acquiring service data from an interface of a line side, wherein the service data comprises interface information which is marked by the line side and used for identifying the interface for acquiring the service data; according to a pre-configured corresponding relation between the interface information and the service tag information used for identifying service data type, stripping the interface information on the service data, and making corresponding service tag information, wherein the service data type comprises PTN (Packet Transport Network) service data and / or OTN (Optical Transport Network) service data; and transmitting the service data marked with the service tag information to a system side. With the method and the device provided by the invention, a problem of failure in fusion of devices of a PTN and an OTN in a related technology is solved, effective combination of the PTN service and the OTN service is completed, and efficient and flexible transmission of the born service in the PTN and the OTN is realized.

The invention discloses a method for fusing multi-physical-domain feature information, and the method comprises the following steps of: (1) obtaining required apriori information in an engineering event through a measuring tool, removing redundant information and obtaining an optimized feature information set; (2) converting each feature information value Xi in the optimized feature information into an interval form; (3) solving a broad sense Hidden Markov initial model; (4) obtaining an optimal model according to the obtained broad sense Hidden Markov initial model alpha; and (5) solving theoptimum broad sense Hidden Markov model of each single-physical-domain, after synthesizing combined broad sense interval, checking the propability distribution according to a coupling relationship among various single-physical-domain optimum model parameters, and finishing and obtaining the fusion of the multi-physical-domain feature information. The method can be used for establishing coupling relationship among the multi-physical-domain parameters in different engineering and effectively solving the fusion problem of the multi-physical-domain feature information.

The invention provides a mechanical arm impedance control method. A position-based impedance control method and a force-based impedance control method are fused; a joint position controller is arranged in the mechanical arm; the control architecture of an admittance controller combined with the inverse kinematics of the mechanical arm in the position-based impedance control method is replaced by the control architecture of a virtual impedance controller combined with a forward dynamics solver, an impedance controller in the force-based impedance control system is replaced by a control loop, sothat a reference impedance force Fimp obtained by a pose error of a Cartesian space through a reference impedance model is converged to an actually measured external force Fs, and then an FFC is output through a closed-loop force controller. According to the method, the control precision and the stability during interaction with the hard environment are ensured simultaneously in mechanical arm impedance control, and a new method is provided for motion control of the industrial mechanical arm.

The embodiment of the invention relates to the technical field of software testing, and discloses a test processing method and device, electronic equipment and a computer readable storage medium, and the method comprises: obtaining a service instance list of a target test suite service through a control center service of a test control plane of a cloud platform, wherein address information of each service instance is stored in the instance list, each test suite service and the corresponding sidecar Proxy are located in the same Pod, each Pod is located in a test data plane of the cloud platform, the Pods are mutually independent, and each Pod is a test grid in a micro-service based on a Service Mesh; and then, based on the address information of each service instance in the service instance list, performing test processing on the target test suite service, wherein the test processing comprises at least one of the following items: traffic interception, load balancing, traffic management, service monitoring and access control.

The invention discloses a method for realizing network identification conversion, an apparatus and a system thereof. The method, the apparatus and the system are applied on border gateway. The method comprises the following steps: after a head message of a terminal is received, determining whether source network identification of the terminal is an IPv6 address; if the source network identification of the terminal is not an IPv6 address, setting an identification code for the terminal and generating objective network agent identification for the terminal; establishing mapping relation information of the identification code of the terminal and the objective network agent identification. According to the invention, the source network identification can be translated into an objective network identification agent so that the source network identification can be reused. Theoretically, infinite expansibility is possessed. On the border gateway, the reused source network identification can be clearly distinguished. By using the method, the apparatus and the system of the invention, foreign identification translation and a reuse expansion ability of the source network identification can be effectively performed. Intercommunication and fusion of different network identification and different addressing mode networks are effectively achieved. A service is provided maximumly for the user.

The invention discloses a powder metallurgy spring steel composite material and a preparation method thereof. The composite material comprises the components with the content by mass percent: 85-95% of 45CrMoV and 5-15% of carbonized niobium powder; and the powder metallurgy spring steel composite material is prepared by the method combining high-energy ball-milling and thermal isostatic pressing sintering. The invention adopts the high-energy ball-milling to lead NbC to be evenly dispersed on a 45CrMoV spring steel base body, and greatly improves the wear resistance of the material under the condition of ensuring the mechanical property of the material; and furthermore, by adopting the method combining the high-energy ball-milling and the thermal isostatic pressing sintering, the invention effectively solves the problem of enhancing the fusion and interface of particles and the base body, thus leading the enhanced particles and base body to be more closely combined with each other. The spring steel composite material is especially applicable for manufacturing an elastic element.

This invention discloses an E1 simulation system includs an E1 interface circuit, E1 simulating module, a mono-chip processor, a control module, a DC / DC power supply and an adaptive clock resume module. The control module also includes an ARP protocol process module, ICMP protocol process module, a head process module, serial communication protocol process modules and a clock correction process module. The invented E1 simulation system can be used in edge / access layer Ethernet exchanger, using single Ethernet exchange can constitute a uniform accessed total service access network of IP and EI so as to optimize the access network.

The invention relates to a control system for a digital home, in particular to a middleware system for the data and information management of the digital home. The middleware system for the digital home comprises a physical connection module, a drive adaptation module, a transmission module, a service module, a template module and an application module, wherein the physical connection module is connected with the transmission module through the drive adaptation module; the transmission module is connected with the template module through the service module; and the template module is connected with the application module. By the middleware system for the digital home, information service and device function service are seamlessly combined in the digital home, so that problems about the standardization and fusion of home devices and the information service are solved in a service level.

The invention discloses a detection method for position of a rotor in a full-speed domain of a permanent magnet synchronous motor without a position sensor. The detection method mainly comprises a high-frequency signal injection method, an extended back electromotive force method and a switching method. When the motor is stationary or operates at a low speed, the high-frequency signal injection method is used for estimating the position of the rotor, and the switching method is designed, so that when back electromotive force is large enough, the high-frequency signal injection method is switched to the extended back electromotive force method to estimate the position of the rotor. According to the detection method disclosed by the invention, the two methods are effectively integrated through the designed switching method, so that the detection method for the position of the rotor of the permanent magnet synchronous motor without the position sensor can be used in a full-speed range and is high in detection precision and easy to implement.

The invention discloses a pecan-milk compound protein beverage and a processing method thereof. The pecan-milk compound protein beverage is characterized by comprising the following raw materials: pecan kernels, whole milk powder, a compound stabilizer, white granulated sugar, concentrated jujube juice, edible salt, starch, sodium caseinate, sodium tripolyphosphate, sodium citrate, trichlorosucrose, sodium hydrogen carbonate, edible essence and water. The processing method comprises the following steps: screening, rinsing, finely grinding, dissolving milk powder, dissolving white granulated sugar, dissolving ingredients, blending with the jujube juice, homogenizing, sterilizing, filling and sterilizing. The pecan-milk compound protein beverage has a peculiar fragrance of pecans and rich nutrition of the whole milk powder, has a unique milk red color of a pecan beverage, and has good color, fragrance and taste; the pecans and the milk powder are mixed in one beverage product, so that the pecan-milk compound protein beverage is more balanced in nutrition and has a gentle mouth feel at the same time.