Product
Patsnap Eureka
For R&D, Patsnap Eureka makes reading and utilizing patents & technical documents easy.
Patsnap Eureka AIR
Designed for self-driven R&D workflows. Generate viable solutions, solve complex R&D challenges, empower your innovation with AI.
Patsnap Eureka Materials
Designed for material experts only. Revolutionize your material R&D, from search, analyze, to developing new materials.
Feature
TechResearch
Generate reliable direction feasibility study reports for your R&D in just a few steps.
TechSeek
Discover and master advanced knowledge NOW. Basics, ideas, possibilities, all at once.
TechMind
As an expert in R&D Theories, TechMind can generates customized viable solutions instantly.
TechRisk
Analyze your overall solution with one click, know your potential R&D risks in advance.
TechMonitor
Get weekly tech updates, stay abreast of the latest tech innovations and key insights.
Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
38 results about "Centering matrix" patented technology
Efficacy Topic
Property
Owner
Technical Advancement
Application Domain
Technology Topic
Technology Field Word
Patent Country/Region
Patent Type
Patent Status
Application Year
Inventor
In mathematics and multivariate statistics, the centering matrix is a symmetric and idempotent matrix, which when multiplied with a vector has the same effect as subtracting the mean of the components of the vector from every component.
Method for controlling a tool
InactiveUS20140160273A1Complex setPrecise positioningProgramme controlCharacter and pattern recognitionCamera imageCentering matrix
A method for controlling a tool, including the steps: providing a reference matrix including reference points and a centering matrix including centering points in a material processing plane; imaging the material processing plane through a camera as a camera image in a size of a camera image field; de-skewing the camera image of the material processing plane by aligning with the reference matrix; scaling a pixel size of the camera image through aligning with the reference matrix; centering the camera image through aligning with the centering points; projecting a processing contour onto the de-skewed and scaled camera image of a workpiece; and aligning the processing contour on the camera image of the work piece and starting the processing, wherein the processing of the work piece is performed though the tool along the processing contour.
Owner:ACSYS LASERTECHN
Support vector machine high-voltage circuit breaker fault diagnosis method based on fuzzy clustering
InactiveCN103345639AEasy to identifyAccurate identificationCharacter and pattern recognitionAlgorithmCircuit breaker
The invention discloses a support vector machine high-voltage circuit breaker fault diagnosis method based on fuzzy clustering. The method comprises the steps of firstly, extracting a feature vector from a stroke-time curve of a circuit breaker moving contact, using the feature vector as a database of fault diagnosis, secondly, conducting fizzy clustering processing on a data sample, generating a new clustering center matrix, thirdly, using the clustering center matrix as a training sample, applying a support vector machine to conduct training, and fourthly applying the high-voltage circuit breaker fault diagnosis method to diagnose test data. According to the support vector machine high-voltage circuit breaker fault diagnosis method based on the fuzzy clustering, the efficiency of the high-voltage circuit breaker fault diagnosis can be effectively improved, the time of the fault diagnosis is reduced, the quality of the fault diagnosis is improved, and the support vector machine high-voltage circuit breaker fault diagnosis method has great significance for research on the safety and the reliability of the power grid.
Owner:HOHAI UNIV CHANGZHOU
Six-dimensional force sensor
ActiveCN106918418AReduced precision requirementsRequirements for realizing force-orthogonal decouplingFluid pressure measurementMeasurement of force componentsCantilevered beamStress concentration
The present invention discloses a six-dimensional force sensor. The sensor comprises a sensor pedestal and a housing arranged on the sensor pedestal, and the sensor pedestal is provided with an elastomer. The elastomer is a rood beam, the rood beam comprises a center matrix and four cantilever beams connected on the center matrix, the sensor pedestal is provided with four positioning parts sliding along the axial direction, the tail end of each cantilever beam sleeves the four positioning parts sliding along the axial direction and slides along the axial direction of the positioning parts, each cantilever beam is provided with a first stress concentration position where stress concentration is generated in the horizontal plane and a second stress concentration position where stress concentration is generated in the vertical plane, strain gages configured to measure the size of the stress at the stress concentration position are respectively arranged at the first stress concentration position and the second stress concentration position. The sensor is simple and compact in structure and high in modularization to realize force decoupling, is accurate in measurement and has an overload protection effect.
Owner:SOUTHEAST UNIV
Method and device for multi-party joint dimension reduction processing of private data
ActiveCN111401479AGuaranteed privacy and securityEnsure safetyCharacter and pattern recognitionDigital data protectionNon symmetricAlgorithm
The embodiment of the invention provides a method and device for carrying out dimension reduction processing on private data in a multi-party joint mode. The method comprises the steps: under the condition that the private data is longitudinally distributed, a first holding party performs zero equalization on a first original matrix to obtain a first center matrix, obtains an N*N asymmetric orthogonal matrix, multiplies the asymmetric orthogonal matrix by the first center matrix to obtain a first secret matrix, and sends the first secret matrix to a trusted third party; the trusted third partysplices the secret matrixes to obtain a global secret matrix, multiplies the global secret matrix by the transposed matrix thereof to obtain a covariance matrix, performs eigenvalue solving on the covariance matrix to obtain a dimension reduction transformation matrix, splits the dimension reduction transformation matrix to obtain split matrixes, and sends the split matrixes to a holder; the first holder processes the first original matrix by using the first split matrix to obtain a first dimension reduction matrix, wherein the first dimension reduction matrix is used for performing businessprediction analysis on the business object in a machine learning mode.
Owner:ALIPAY (HANGZHOU) INFORMATION TECH CO LTD
RCS conversion method based on scattering center matrix
InactiveCN103487791ALarge applicable angle rangeReduce data volumeWave based measurement systemsPhysical opticsCentering matrix
The invention relates to an RCS conversion method based on a scattering center matrix. The method comprises the steps of adopting a moment method or an iterative physical optical method to obtain the approximation relation between an induction source and an exposure field according to a set target rough geometric model, and setting the scattering center conversion matrix between the exposure field and the RCS with a target to be measured comprising an undetermined coefficient; according to the observation direction required to conduct RCS conversion, using the direction of an observation center as a fixed incident angle, and obtaining double-station scattering performance data under the different scattering angles of the target to be measured; adopting the least square method to determine the undetermined coefficient in the conversion matrix; achieving the RCS conversion of the target. The method is not limited in the high-frequency area, the suitable angle range is larger, the required data size is small, and engineering realization is easy.
Owner:SHANGHAI RADIO EQUIP RES INST
Method and device for jointly determining object feature correlation in private data by multiple parties
PendingCN112818290ADigital data protectionComplex mathematical operationsCorrelation coefficientAlgorithm
The embodiment of the invention provides a method and device for jointly determining object feature correlation in private data by multiple parties. The privacy data are distributed in a plurality of holders, and the first holder performs zero mean on feature values of multiple features in the first original matrix to obtain a first center matrix; and a first fragment matrix of the covariance matrix is determined based on the first central matrix and respective central matrixes of other owners by using multi-party security calculation. For the ith feature in the first holding party and the jth feature in the second holding party, the first holding party obtains data from the local covariance matrix fragment and the locally stored feature data, and determines a first correlation coefficient fragment between the ith feature and the jth feature based on the local feature data of the second holding party by using multi-party security calculation.
Owner:ALIPAY (HANGZHOU) INFORMATION TECH CO LTD
Abnormal service request identification method and terminal equipment
PendingCN109492394AGuaranteed uptimePlatform integrity maintainanceTerminal equipmentTime distribution
Owner:PING AN TECH (SHENZHEN) CO LTD
Item recommendation method and system based on collaborative filtering algorithm
InactiveCN105824942ASolve the speed bottleneckReduce workloadRelational databasesSpecial data processing applicationsAlgorithmCentering matrix
The invention relates to the technical field of automatic recommendation, and in particular relates to an item recommendation method and system based on a collaborative filtering algorithm. The method comprises the following steps: performing an cluster analysis on item scoring behaviors of m basic users offline to obtain k user clusters; determining a cluster center matrix Ckxn and a class subordination degree matrix Vmxk offline; carrying out an item recommendation on a target user by utilizing the cluster center matrix Ckxn and the class subordination degree matrix Vmxk online. By obtaining the k user clusters offline in advance, the cluster center matrix Ckxn and the class subordination degree matrix Vmxk are further obtained, and the working amount of online recommendation is reduced, so that the aim of guaranteeing a recommendation generation speed is realized, and a speed bottleneck problem caused by a traditional collaborative filtering algorithm in a recommendation generation speed is solved.
Owner:SHANGHAI TRUELAND INFORMATION & TECH CO LTD
An internet data clustering method and system
InactiveCN106897276AAvoid restrictionsReduce instabilityCharacter and pattern recognitionSpecial data processing applicationsFeature setThe Internet
The invention relates to an internet text data clustering method and system. The internet text data clustering method comprises the steps of: a, training text data by using a topic model to obtain a probability distribution matrix of all key words in each topic, and grouping the key words in a text set; b, rearranging feature sets of the text data according to the grouping of the key works to obtain new document data containing key word grouping feature information; c, operating a double-layer soft subspace clustering algorithm on the new document data containing the key word grouping information to generate a clustering center matrix and a sample ownership matrix; d, repeating the steps a-c for n times to obtain a plurality of clustering results; e, operating a clustering integration algorithm on a model set to integrate the multiple clustering results to obtain a final clustering result. The method and the system can reduce the instability of an FG-k-means algorithm effectively.
Owner:SHENZHEN INST OF ADVANCED TECH CHINESE ACAD OF SCI
Method and system of object recommendation
ActiveCN108073578AReduce the impactReduce the number of timesMarketingSpecial data processing applicationsCentering matrixData mining
The application relates to the field of computer technology, and particularly to a method and a system of object recommendation. The method and the system are used for solving problems that price interval dividing existing in the prior art is liable to divide close prices into different sub-intervals, results of dividing are liable to be impacted by data smoothing methods, and thus objects recommended for users can be impacted. According to the embodiment of the application, iteration is carried out on a preset clustering center matrix and a clustering coefficient matrix according to an indexfeature data matrix, then target index intervals are determined, the target index intervals of a category which is the same as that of the object associated with a user are determined, an interval towhich a target index of the target associated with the user belongs is determined in the determined target index intervals, and recommendation is carried out to the user. According to the embodiment of the application, occurrence frequency of a case of dividing the close prices into the different sub-intervals can be reduced, impacts from the data smoothing methods can be reduced, and impacts on recommending the objects for the users can be reduced.
Owner:浙江飞猪网络技术有限公司
Clustering cluster calculation method and device, terminal and storage medium
InactiveCN110825826AAccurately obtain the number of clustersRelational databasesSpecial data processing applicationsEstimation methodsEngineering
The embodiment of the invention discloses a clustering cluster calculation method and device, a terminal and a storage medium. The method comprises the following steps: acquiring target feature data of communication equipment contained in a sample equipment set at a preset time magnitude within a preset duration, and constructing a standard initial matrix according to the target feature data; extracting a preset number of communication equipment samples from the sample equipment set, and constructing a clustering center matrix according to the target feature data corresponding to the preset number of communication equipment samples; calculating a target distance value between each row of elements in the clustering center matrix and all rows of elements in the standard initial matrix, and constructing a distance vector; and calculating the number of class clusters contained in the target feature data by adopting a parameter-window-free function probability density estimation method based on the distance vector. In addition, the embodiment of the invention further discloses a multi-dimensional time series clustering cluster calculation device, a terminal and a computer readable medium. According to the invention, the method can achieve the precise obtaining of the number of multi-dimensional class clusters.
Owner:SHENZHEN UNIV
Large-area?topographic map?segmentation method based on random probability?sampling and multi-level fusion
The invention discloses a large-area?topographic map?segmentation method based on random probability?sampling and multi-level fusion. The method comprises steps: 1) an original?topographic map image is imputed; 2) random probability?sampling is carried out on the original?topographic map image; 3) the number of topographic map?segmentation types is determined; 4) a clustering center is calculated and a clustering center matrix is obtained; 5) a membership matrix of the original?topographic map is calculated and a membership matrix is obtained; 6) the membership matrix is used for carrying out fuzzy classification on the original?topographic map and segmented split images are obtained; 7) multi-level fusion is carried out on the segmented images and a fused image is obtained; and 8) the segmented split images are outputted. According to the method, random probability?sampling and multi-level image fusion are combined to carry out topographic map?segmentation, segmented split images of the topographic map can finally be more accurately acquired, the segmentation efficiency is greatly improved, and the method is particularly suitable for large-area?topographic map?segmentation.
Owner:西安电子科技大学重庆集成电路创新研究院
Combined friction stir welding mixing head aiming at different thicknesses of sheets
ActiveCN102151979AEven by forceImprove mechanical propertiesNon-electric welding apparatusEngineeringCentering matrix
The invention relates to a combined friction stir welding mixing head aiming at different thicknesses of sheets, which is characterized in that a center matrix is coaxially inserted into a sleeve; internal threads and external threads which are matched are respectively formed on a threaded connection column and a threaded connection hole; after the threaded connection column is in threaded fit with the threaded connection hole, a shaft shoulder of the center matrix is connected with a clamping body of the sleeve; and a matrix mixing needle is coaxially connected with a sleeve mixing needle. In the machining process, the center matrix is universal, and a series of sleeves with different dimensions and parameters only need to be machined, and a plurality of entire mixing heads are not required to be machined, so that the machining materials are saved. If the mixing head is abraded in the welding process, the sleeve is only required to be replaced, and the center matrix can be continuously used, so that the assembly and disassembly time of the mixing head is saved, and the welding efficiency is improved. Meanwhile, the center matrix and the sleeve can select dissimilar materials to adapt to the welding of various materials.
Owner:NORTHWESTERN POLYTECHNICAL UNIV
Market partitioning method and device based on nodal electricity price
ActiveCN110137951AAnd the application value is goodAc network with energy trading/transmission rightsData setElectricity price
The invention discloses a market partitioning method and device based on nodal electricity price. The method includes the following steps: S1, acquiring electricity price data of a target area in a preset period to generate a data set; S2, acquiring a clustering center matrix and a random matrix, initializing and normalizing the random matrix to generate a membership matrix, and setting the numberof iterations to 1; S3, judging whether the current number of iterations is less than a preset threshold, executing S5 if the current number of iterations is not less than the preset threshold, and if the current number of iterations is less than the preset threshold, updating the clustering center matrix, updating the membership matrix according to the updated clustering center matrix and the data set, and executing S4; S4, judging whether the updated membership matrix meets the convergence condition, adding 1 to the number of iterations and executing S3 if the updated membership matrix doesnot meet the convergence condition, and executing S5 if the updated membership matrix meets the convergence condition; and S5, outputting the updated clustering center matrix and the updated membership matrix, classifying each node according to the updated membership matrix, and partitioning the target area to generate a partitioning result. The method can be used to reasonably partition multipleareas.
Owner:广东电力交易中心有限责任公司
Unknown radar target recognition method based on radiation source feature subspace knowledge
ActiveCN113095354ASmall amount of calculationIncrease computing speedCharacter and pattern recognitionNeural architecturesFeature vectorSingular value decomposition
The invention discloses an unknown radar target recognition method based on radiation source feature subspace knowledge, and belongs to the field of radar and artificial intelligence, and the method comprises the following steps: S1, employing known radar target data to construct a training set, employing the training set to train a classification recognition model, and obtaining a trained classification recognition model; s2, acquiring a high-dimensional feature vector output by processing an input radar target test sample by the trained classification recognition model, constructing a center matrix based on the high-dimensional feature vector, and performing singular value decomposition on the center matrix to obtain a zero-value domain subspace; s3, calculating the ratio of the vector length of the projection of the test sample in the value domain subspace to the vector length of the projection of the test sample in the zero domain subspace, and determining a judgment threshold; s4, judging whether the new radar target to be identified belongs to a known radar target or an unknown radar target based on the judgment threshold. According to the invention, the problem that the radar unknown target cannot be accurately identified in the prior art is solved.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA
Image dimension reduction clustering method based on fuzzy theory
PendingCN114863151AImprove efficiencyReduce lossCharacter and pattern recognitionAlgorithmCentering matrix
The invention discloses an image dimension reduction clustering method based on a fuzzy theory, which comprises the following steps of: firstly initializing a projection matrix U, a membership matrix Y, a clustering center matrix M, a projected sample matrix V and a regularization parameter lambda, then alternately updating V, M and Y by adopting an alternate optimization algorithm, and repeatedly iterating until an objective function is converged to realize unsupervised data dimension reduction. According to the invention, the unsupervised method, namely the fuzzy principal component dimension reduction and clustering method (FPCPC), which can be used for carrying out dimension reduction and clustering at the same time, is realized. According to the method, the dimensionality reduction of the image data and the clustering in the subspace are simultaneously realized in one method, the efficiency is improved, and the loss of category information in the dimensionality reduction process of the image is reduced.
Owner:NORTHWESTERN POLYTECHNICAL UNIV
Method for identifying wide-area adjustable load
ActiveCN108155639AOvercoming complexityOvercome errorAc network circuit arrangementsWide areaFeature extraction
Owner:NORTH CHINA ELECTRIC POWER UNIV (BAODING) +4
Cluster analysis method of gene expression data, and terminal equipment
PendingCN110211641AGood clustering resultExcellent speed and precisionBiostatisticsInstrumentsComputation complexityRound complexity
The invention discloses a cluster analysis method of gene expression data, and terminal equipment. The cluster analysis method comprises the steps of determining a gene expression data coding plan, acquiring the gene expression data, a constitutive gene cluster center matrix and a cluster membership grade matrix; initializing the position and the speed vector of an intelligent single particle, performing segmenting, generating a corresponding sub-vector, setting a related parameter in an iterative calculating process; according to a learning intelligent strategy, performing continuous iterative computation, and updating the position vector of the intelligent single particle; reconstructing a clustering center matrix according to the final position sub-vector of the intelligent single particle, and obtaining an optimal gene clustering center. Compared with a traditional method, the cluster analysis method according to the invention has advantages of reducing at least one order of magnitude in computing complexity, greatly reducing computing time, and realizing important meaning in processing high-flux gene data of which the data scale increase explosively.
Owner:纪震
Unsupervised parasite classification method and system based on artificial intelligence
ActiveCN110032973AHigh similarityImprove classification accuracyCharacter and pattern recognitionNeural architecturesData setArtificial general intelligence
The invention discloses an unsupervised parasite classification method and system based on artificial intelligence. The classification method comprises the steps of obtaining a training data set of ato-be-detected sample; extracting feature information of the training data set by using a deep convolutional neural network VGG network; classifying the feature information by using a fuzzy C-means clustering FCM algorithm, and determining a clustering center matrix of each category; determining a clustering center vector of each category according to the clustering center matrix; determining a membership matrix according to the clustering center vector; determining an FCM loss function according to the clustering center matrix and the membership matrix; using an FCM loss function to train theVGG network, and determining the trained VGG network; and classifying cells and parasites in the training data set according to the trained VGG network and an FCM algorithm. By adopting the classification method and system provided by the invention, parasitic cells and host healthy cells can be accurately identified and classified, and the classification accuracy is improved.
Owner:HARBIN INST OF TECH SHENZHEN GRADUATE SCHOOL
Space target posture extrapolation method based on three-dimensional reconstruction
ActiveCN110082765ASolve the problem that HRRP samples cannot be obtained for a long timeGuaranteed accuracyRadio wave reradiation/reflectionSingular value decompositionTime domain
The invention discloses a space target posture extrapolation method based on three-dimensional reconstruction and mainly solves the problem that in the prior art, a target identification rate is low when the number of multitask radar samples is relatively low. The method comprises the realization processes of 1), carrying out uniform framing on data and carrying out ISAR imaging; 2), carrying outcross-range scaling on each ISAR image; 3), extracting a scattering center parameter set of each ISAR image; 4), associating scattering centers of an ISAR image set to obtain a track matrix; 5), carrying out singular value decomposition on the track matrix to obtain a projection matrix and a position matrix; 6), carrying out the three-dimensional reconstruction on the projection matrix and the position matrix to obtain a scattering center matrix; and 7), extrapolating data in each frame according to the scattering center matrix, thereby obtaining a time domain real radar range profile sample of each frame after extrapolation, wherein the time domain real radar range profile sample is used for follow-up target identification. According to the method, through adoption of a data extrapolationmethod, the target identification rate can be remarkably improved. The method can be used for identifying radar range profiles.
Owner:XIDIAN UNIV
Production method for improving grain size of 6063 aluminum alloy center matrix
The invention discloses a production method for improving the grain size of a 6063 aluminum alloy center matrix. The method comprises the following steps of casting, stirring, cast rod homogenizationtreatment, extrusion aluminum bar temperature, mold design, extrusion speed and cooling. Through multiple experiment detection of the grain size of the 6063 aluminum alloy, the structure is the one level, the grain size of the part of the center matrix is small, and after anodic oxidation, the piebald phenomenon cannot appear.
Owner:江苏南铝创佳金属股份有限公司
A six-dimensional force sensor
ActiveCN106918418BReduced precision requirementsRequirements for realizing force-orthogonal decouplingFluid pressure measurementMeasurement of force componentsCantilevered beamStress concentration
The present invention discloses a six-dimensional force sensor. The sensor comprises a sensor pedestal and a housing arranged on the sensor pedestal, and the sensor pedestal is provided with an elastomer. The elastomer is a rood beam, the rood beam comprises a center matrix and four cantilever beams connected on the center matrix, the sensor pedestal is provided with four positioning parts sliding along the axial direction, the tail end of each cantilever beam sleeves the four positioning parts sliding along the axial direction and slides along the axial direction of the positioning parts, each cantilever beam is provided with a first stress concentration position where stress concentration is generated in the horizontal plane and a second stress concentration position where stress concentration is generated in the vertical plane, strain gages configured to measure the size of the stress at the stress concentration position are respectively arranged at the first stress concentration position and the second stress concentration position. The sensor is simple and compact in structure and high in modularization to realize force decoupling, is accurate in measurement and has an overload protection effect.
Owner:SOUTHEAST UNIV
Reservoir classification method and device
ActiveCN108596383AImprove classification accuracyEnables continuous forecastingForecastingResourcesReservoir typeNMR - Nuclear magnetic resonance
The invention provides a reservoir classification method and device. The method comprises the following steps that: according to the nuclear magnetic resonance transverse relaxation time T2 spectrum and the radial basis function of N pieces of rock samples, obtaining a first center matrix; according to the nuclear magnetic resonance T2 spectrum and the mercury penetration experiment data of N pieces of rock samples, obtaining a first reservoir classification comprehensive index; according to the inverse matrix of the first center matrix and the first reservoir classification comprehensive index, obtaining a weight coefficient matrix; and then, according to the nuclear magnetic resonance T2 spectrum of points to be classified and the weight coefficient matrix, determining a second reservoirclassification comprehensive index, wherein the second reservoir classification comprehensive index is used for indicating the reservoir type of the points to be classified. By use of the reservoir classification method and device, an underground reservoir type can be continuously predicted, a plurality of rock samples or a plurality of underground points to be classified can be subjected to reservoir classification, operation speed is high, and reservoir classification accuracy is high.
Owner:CHINA UNIV OF PETROLEUM (BEIJING)
Market partition method and device based on node electricity price
The invention discloses a market partition method and device based on node electricity prices. The method includes: S1, obtaining electricity price data in a preset period in a target area to generate a data set; S2, obtaining a cluster center matrix and a random matrix, and performing random The matrix is initialized and normalized to generate a membership matrix, and the number of iterations is set to 1; S3, judge whether the current number of iterations is less than the preset threshold, if not, S5, update the clustering center matrix, according to the updated clustering The center matrix and the data set update the membership matrix, execute S4; S4, judge whether the updated membership matrix meets the convergence condition, if not, add 1 to the number of iterations, execute S3, and S5 if it is satisfied; S5, output the updated clustering center matrix and the updated membership matrix, each node is classified according to the updated membership matrix, and the target area is partitioned to generate a partition result. This method can reasonably partition multiple regions.
Owner:广东电力交易中心有限责任公司
Center convergence optimization in a projection display apparatus
Progressive rounding error and convergence error encountered due to multiple use of a center correction adjustment (820) of an image of a video projection display is reduced by calculating a 3 x 3 matrix (770) for a moved color signal where the non center matrix values represent the difference between the initially measured sensor values (Si1,...,Si8) stored at initial alignment (720), and stored most recently measured sensor values (Sp1,...,Sp8). The matrix center value (860) is the sum of the averaged values calculated from the edge center errors, the stored sum of previous moves (MS) and the current move (CP). Rerun of the sensor finding routine (745) resets the stored move sum to zero.
Owner:THOMSON LICENSING SA
Traffic state identification method based on binocular camera
InactiveCN110610608AMake full use ofEasy to controlImage enhancementImage analysisSimulationCentering matrix
The invention discloses a traffic state identification method based on a binocular camera, and belongs to the technical field of intelligent traffic guidance. The method comprises the following threesteps: (1) establishing a reference data a, determining a traffic state clustering center, and transmitting and storing a traffic state clustering center matrix to the information board of each traffic section through a traffic system management control platform, and b, setting a road background image, and sending the road background image to the information board of a corresponding traffic intersection through the traffic system management control platform; (2) calculating relevant parameters for determining a road traffic state: a, calculating a road space occupation rate S, and b, calculating two parameters of vehicle flow q and vehicle speed v; and (3) traffic state identification: judging to which traffic state each traffic section belongs through the two parameters of the vehicle flow q and the vehicle speed v of each traffic section by the information board, and issuing the two parameters. The method can provide real-time traffic conditions for travelers, and road resourcesare fully and reasonably utilized.
Owner:JIANGSU GENTURE ELECTRONICS INFORMATION SERVICE CO LTD
Method for estimating location of terminal in wireless communication system and apparatus therefor
ActiveUS11346931B2Accurate estimateElectromagnetic wave reradiationNeural learning methodsTelecommunicationsCommunications system
A method of estimating, by a user equipment (UE), a location of the UE in a wireless communication system is disclosed. The method includes measuring distances between a plurality of anchor nodes and the UE; creating a first matrix using values of the measured distances; creating a third matrix based on the first matrix and a second matrix, the second matrix being a centering matrix; and estimating the location of the UE based on a result of comparing a first value generated based on eigenvalues of the third matrix with a second value that is a pre-defined reference value, wherein when the first value is greater than the second value, all the measured distances are distances measured through a line of sight (LOS) path, wherein when the first value is less than the second value, some of the measured distances are distances measured through a non-line of sight (NLOS) path.
Owner:LG ELECTRONICS INC
An Unknown Radar Target Recognition Method Based on Subspace Knowledge of Emitter Characteristic
ActiveCN113095354BSmall amount of calculationIncrease computing speedCharacter and pattern recognitionNeural architecturesFeature vectorTest sample
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA
Large Format Topographic Map Segmentation Method Based on Random Probability Sampling and Multilevel Fusion
Owner:西安电子科技大学重庆集成电路创新研究院
Combined friction stir welding mixing head aiming at different thicknesses of sheets
ActiveCN102151979BEven by forceImprove mechanical propertiesNon-electric welding apparatusEngineeringCentering matrix
Owner:NORTHWESTERN POLYTECHNICAL UNIV
Who we serve
- R&D Engineer
- R&D Manager
- IP Professional
Why Patsnap Eureka
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com