Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

908results about How to "Effective prediction" patented technology

Process and system for integrating information from disparate databases for purposes of predicting consumer behavior

InactiveUS7490052B2Powerful and accurate modelWidens” the narrow base of connectivityMarket predictionsDatabase distribution/replicationIntegrated databaseData mining
A process and system for integrating information stored in at least two disparate databases. The stored information includes consumer transactional information. According to the process and system, at least one qualitative variable which is common to each database is identified, and then transformed into one or more quantitative variables. The consumer transactional information in each database is then converted into converted information in terms of the quantitative variables. Thereafter, an integrated database is formed for predicting consumer behavior by combining the converted information from the disparate databases.
Owner:GFK US MRI LLC

Method and system for predictive enterprise resource management

A system and method are disclosed which predict whether a performance problem within a network is likely to be encountered during future operation. Furthermore, a preferred embodiment not only predicts the likelihood of a performance problem, but further determines the appropriate preventative measures to be taken in an attempt to prevent a predicted performance problem from occurring. In a preferred embodiment, a management system (MS) that oversees the operation of a network is implemented to predict likely performance problems within the network, and may determine appropriate preventative measures for preventing predicted performance problems within the network. Polling gateway(s) may be utilized to periodically poll the network resources in order to retrieve status information for such resources, including but not limited to status of disk(s), database(s), memory, CPU(s), and operating system(s) within the network. The gathered status information is then evaluated by the MS by, for example, correlating such status information with known performance rules for the network to predict potential performance problems, and based on such evaluation, the MS may predict whether a future performance problem is likely to be encountered. Once a future performance problem has been predicted, the MS determines an appropriate preventive action for preventing the performance problem from occurring, and the MS may initiate the appropriate preventive action before the occurrence of the predicted performance problem in an attempt to prevent such performance problem. Most preferably, the network management system is implemented to “learn” symptoms of performance problems over time.
Owner:OBJECTIVE SYSTEMS INTEGRATORS

Method for synchronously recognizing identities and expressions of human faces

InactiveCN101620669AFull support for simultaneous recognitionGood recognition characteristicsCharacter and pattern recognitionIdentification rateFacial characteristic
The invention proposes a method for synchronously recognizing identities and expressions of human faces. The method comprises the steps of extracting facial features of each human-face image, defining corresponding semantic features for each image and adopting a feature fusion method of kernel principal component analysis (PCCA) for the facial features so as to enable input image features to have better recognition properties. On the basis, a model of the relation between the facial features and the semantic features is established by use of a partial least-squares regression (PLSR) method, and expression-identity recognition is performed on to-be-recognized human-face images by use of the model. Experiments show that the method proposed by the invention not only can synchronously recognize human faces and expressions, but also can improve the recognition rate of human-face expression recognition.
Owner:南京宇音力新电子科技有限公司

Forecasting and evaluating technologies of three-dimensional earthquake optimum time window river course sand body storage layer

InactiveCN101408624AHigh-resolutionImprove frequency division inversionSeismic signal processingDepth conversionRoot mean square
The invention discloses a prediction and evaluation technology of 3D seismic optimum time window river channel sand body reservoirs, belongs to the technical field of the prediction and evaluation of the 3D seismic reservoirs, and aims at solving technical problem that the river channel predictive resolution of the traditional methods is not enough. The technical proposal is as follows: the 3D visualization scanning is performed on each reservoir at the interval of 1-2ms, the optimum time window is determined according to the scope shown by the target river channel, the corresponding subfield is cut out, clairvoyance and scanning are carried out on the time window properties, such as root-mean-square amplitude, wave impedance and the like, auto-tracing is performed, top surface and bottom surface are picked up, a time isopach map is calculated and converted into a sand body isopach map, a top surface structure diagram is formed by time-depth conversion, and the reservoir physical properties are evaluated by curve reconstruction, thus realizing the prediction and evaluation of the plane morphology, longitudinal thickness and the reservoir physical properties of the river channel sand body. The method adopts the optimum time window to effectively inhibit interference, is applicable to various data volumes, and can effectively predict and evaluate a thin river channel sand body with the thickness far less than 1 / 4 of a wavelength under the condition of frequent interbedding of sand and mudstone, the thin river channel sand body comprises the river channel sand body which is not corresponding to wave crest or wave trough, and the method has good application effect in petroleum exploration and development.
Owner:陶庆学 +2

Multi-target tracking method based on multi-model fusion and data association

ActiveCN107292911AReduce distractions from light and background noiseGood real-time and robustnessImage enhancementImage analysisOptical flowBackground noise
The invention discloses a multi-target tracking method based on multi-model fusion and data association; the tracking method comprises the following steps: firstly using an interframe difference method to detect a motion target contour and center of mass coordinates; fusing a pyramid optical flow method with Kalman filtering so as to predict the center of mass coordinates of the motion target in the next moment; using Euclidean distances between the center of mass coordinate predicted value and the center of mass coordinate detection value at next moment to form a benefit matrix, and using a Hungary algorithm to obtain the optimal matching through data association; finally removing certain portion unable to satisfy requirements in a tracker, and building a tracking unit for non-assigned detections, thus realizing multi-target tracking. The tracking method can be less affected by light changes and background noise interferences, thus solving the tracking failures caused by target blocking or mutual interferences between targets, providing multi-target tracking accuracy, and providing well instantaneity and robustness.
Owner:NANJING UNIV OF POSTS & TELECOMM

Quick chemical leakage predicating and warning emergency response decision-making method

ActiveCN103914622AAppropriate layoutSimple optimization of concentration distributionSpecial data processing applicationsDistributed control systemModel parameters
The invention relates to a quick chemical leakage predicating and warning emergency response decision-making method which combines diffusion model simulation with a neural network and a gas sensor system and is applied to quick warning and aid decision making of leakage of harmful gas in an industrial park. The method includes park risk factor identification, numerical simulation, data screening, neural network training and sensor system and neural network model integration, wherein the park risk factor identification is used for identifying various possible leakage accidents, the numerical simulation includes simulating all the possible accidents to obtain a range of influences of the harmful gas, the data screening includes extracting and reconstructing an effective part in a numerical simulation result according to actual sensor layout, the neural network training includes training specific neural network models by the aid of screened data so as to acquire model parameters aiming for the specific industrial park and surrounding conditions and using redundant data for parameter validation, and sensor system and neural network model integration includes combining the models with a sensor DCS (distributed control system).
Owner:TSINGHUA UNIV

Coupling simulation method of flow and sediment process of distributed watershed

The invention relates to a coupling simulation method of a flow and sediment process of a distributed watershed. The method comprises the steps of digital terrain processing, building of a contour strip, building of a contour strip erosion landform data file, data collecting and processing, watershed hydrological process computation, slope erosion and sediment transport process computation, flow and sediment process computation of a channel or a river or a reservoir, judgment, variable parameter transferring, model parameter calibration and model validation, and ending. According to the coupling simulation method, the contour strip is taken as a computing platform for the slope scale, and the real flow and sediment physical process of a slope is taken as a picture, so that the coupling simulation computing capability for rainfall-runoff and erosion-sediment processes of the contour strip slope is achieved; the computational accuracy is ensured on the premise of reducing the computing amount; and the flow and sediment process of the slope with any grid scale is effectively reduced and predicted. Compared with the prior art, the coupling simulation method has the advantages that the terrain adaptability and the scale adaptability are achieved by taking the contour strip as a basic computing unit; meanwhile, deeper recognition of the law of the flow and sediment process of the slope can be achieved on the basis of physical flow and sediment process simulation of the slope and a more complicated practical application is supported.
Owner:CHINA INST OF WATER RESOURCES & HYDROPOWER RES

Numerical control machine tool thermal error real-time compensation modeling method based on time series algorithm

The invention relates to a numerical control machine tool thermal error real-time compensation modeling method based on a time series algorithm, which belongs to the technical field of precision machining. The method comprises the steps of (1) carrying out data zero mean pretreatment, namely employing an inverted sequence test method and a kurtosis and skewness test method to judge the stationarity and the normality of the data; (2) using an autocorrelation function, a partial correlation function, and the censored results as judgment criteria to carry out the pattern recognition of a thermal error mathematical model; (3) employing a least square estimation method or a long autoregressive residual calculating method to realize the parameter estimation of the thermal error mathematical model; (4) determining the order of the thermal error mathematical model, namely employing a judgment method that combines an AIC order determination criterion, an F test order determination criterion, and a whiteness test order determination criterion to realize the order determination of the thermal error mathematical model; (5) and carrying out integration processing of synthesizing judgment conditions, namely constructing a complete forecasting mathematical model formula. The modeling method provided by the invention has the advantages that less hardware is required, the applicability is wide, and the established model has high prediction precision and reliability.
Owner:上海睿涛信息科技有限公司

Learning-based visual attention prediction system and method thereof

ActiveUS20130021578A1Effectively predictPredict visual attention effectivelyAcquiring/recognising eyesEye exercisersPrediction systemPrediction methods
A learning-based visual attention prediction method is disclosed. The method includes a correlation relationship between the fixation density and at least one feature information being learned by training, followed by a test video sequence of test frames being received. Afterward, at least one tested feature map is generated for each test frame based on the feature information. Finally, the tested feature map is mapped into a saliency map, which indicates the fixation strength of the corresponding test frame, according to the correlation relationship.
Owner:NAT TAIWAN UNIV +1

Ship conflict early warning method

The invention relates to a ship conflict early warning method. The ship conflict early warning method comprises the following steps that the real-time position information and the historical position information of a ship are obtained through maritime radar and are processed primarily; ship trajectory data are preprocessed at each sampling moment, clustering is conducted on the ship trajectory data at each sampling moment, parameter training is conducted on the ship trajectory data at each sampling moment by means of a hidden Markov model, the hidden state q corresponding to an observation value at the current moment is obtained at each sampling moment according to parameters of the hidden Markov model by means of the Viterbi algorithm, a predicted value of the position of the ship in the future time period is obtained at each sampling moment through a set prediction time domain W based on the hidden state q at the current moment, the dynamic behaviors of the ship are monitored in real time by establishing a safety rule set, and a warning message is sent out in time. According to the ship conflict early warning method, the trajectory of the ship is predicted in real time in a rolling mode, effective early warning of maritime conflicts is achieved, and the safety of marine traffic is improved.
Owner:JIANGSU UNIV OF TECH

Particle filtering-based pupil tracking method in sight tracking system

The invention discloses a particle filtering-based pupil tracking method in a sight tracking system and belongs to the field of human-computer interaction. Aiming at a problem that the pupil tracking efficiency in the conventional infrared image is poor, the method creates a concept of a tri-channel pseudo color image (TCPCM) which fully uses information of each channel, enables the characteristics of a pupil to be obvious, and improves the stability and the accuracy of tracking. Aiming at a problem that a background interferes in a pupil tracking process, the method creates a pupil object model which accords with the morphological characteristics and change rules of the pupil to fully differentiate the front background from the back background so as to decrease the interference of the background to the object model. Aiming at a problem that the prediction effect of the position and the form of the pupil is poor, the method creates a state transition equation embodying the motion rules of the pupil, fully considers various conditions of the state change of the pupil and can effectively predict the position and shape of the pupil. Through high-quality characteristic detection and tracking, the method enables the precision of a developed sight tracking system to reach a level of real-time human-computer interaction.
Owner:UNIV OF SCI & TECH BEIJING
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products