76 results about "Time series generation" patented technology

Trying on the clothes is simulated by processing an image sequence by using a posture dictionary, and a three-dimensional figure, obtained in advance, of a person who wishes to try on clothing. The process includes estimating the posture of the person, and simulating the motions of the clothes, in agreement with estimated time series posture data. A virtual modeling apparatus comprises an imaging device configured to take an image sequence, a posture-estimator which estimates the posture of a person in the image sequence from the three-dimensional figure data of the person and from data for various postures, a posture corrector which generates smooth and natural posture time series data from the time series of estimated postures, a clothes simulator which calculates the motions of the clothing being modeled, and a display device which displays images of the person as if they were actually wearing the clothing.

A method an apparatus of automatically controlling a wind turbine re provided. of: A time-series of measurement values of the aerodynamic flow property of the wind turbine rotor blade is determined; a predictive wind field model representing a structure of a wind field acting on the wind turbine rotor blade is generated based on the time-series of measurement values, and a control value is generated based on the wind-field model.

Radiosurgical treatment of tissues of the heart to mitigate arrhythmias such as atrial fibrillation or the like. Radiosurgical targeting of the relatively rapid movement of heart tissues may be enhanced by generating a moving model volume using a time-sequence of three dimensional acquired tissue volumes. A digitally reconstructed radiograph (DRR) may be generated from the model at a desired cardiac and / or respiration motion phase and compared to an X-ray or the like taken immediately before or during treatment. When a series of radiation beams will be directed to a heart tissue to alleviate an arrhythmia, the treatment system may alter the radiation beam series in response to the type of the arrhythmia.

An imaging device includes an image unit that captures a subject and generates a plurality of consecutive captured images in time series, a synthesis unit that performs synthesis using at least a part of each of the plurality of generated captured images and generates a plurality of synthesized images having an order relationship based on a predetermined rule, and a control unit which performs control for displaying information about the progress of the generation of the synthesized images by the synthesis unit on a display unit as progress information, after the process of generating the plurality of captured images by the imaging unit is finished.

An interleaver employs a generalized method of generating a mapping. The mapping is generated for interleaving bits of a data block and associated error detection / correction information. The data block is of length N, and the length of the error detection / correction information is P. An (N+P)×(N+P) square matrix is formed and divided into sub-blocks, where one portion of the matrix is associated with error detection / correction information and another portion is associated with data of the data block. New positions in the matrix are generated in a time sequence on a sub-block by sub-block basis based on a generator seed pair and an original position seed pair. The time sequence also corresponds to positions in an output interleaved block. Once the new position sequence is generated, the matrix is populated with data and error detection / correction information based on the corresponding time sequence. A de-interleaver performs the inverse mapping of the interleaver.

Trying on the clothes is simulated by processing an image sequence by using a posture dictionary, and a three-dimensional figure, obtained in advance, of a person who wishes to try on clothing. The process includes estimating the posture of the person, and simulating the motions of the clothes, in agreement with estimated time series posture data. A virtual modeling apparatus comprises an imaging device configured to take an image sequence, a posture-estimator which estimates the posture of a person in the image sequence from the three-dimensional figure data of the person and from data for various postures, a posture corrector which generates smooth and natural posture time series data from the time series of estimated postures, a clothes simulator which calculates the motions of the clothing being modeled, and a display device which displays images of the person as if they were actually wearing the clothing.

A method an apparatus of automatically controlling a wind turbine re provided. of: A time-series of measurement values of the aerodynamic flow property of the wind turbine rotor blade is determined; a predictive wind field model representing a structure of a wind field acting on the wind turbine rotor blade is generated based on the time-series of measurement values, and a control value is generated based on the wind-field model.

The invention provides an identity authentication method based on keystroke behaviors. The method includes following steps: S1, building a user model of legal user keystroke characteristics according to legal keystroke behaviors during training of legal users; S2, acquiring to-be-estimated keystroke behaviors during keystroke of logged-in users, and generating a to-be-estimated keystroke time sequence; S3, generating a to-be-estimated pole adjacent character sequence according to the to-be-estimated keystroke time sequence; and S4, matching the to-be-estimated pole adjacent character sequence and the user model, and determining whether the to-be-estimated keystroke behaviors of the logged-in users are legal according to a matching result; generating an alarm and enabling the logged-in users to log in again if illegal, and allowing the logged-in users to log in if legal; and storing the to-be-estimated keystroke behaviors and updating the user model. According to the method, a few representative keystroke characteristics of the users are extracted, the authentication accuracy is guaranteed, the response capability of continuous authentication is improved, the keystroke behaviors of the users are monitored in real time, and continuous account safety guarantee is provided for the users.

In order to provide a technology which allows efficient understanding of images of a disease locus and diagnosis supporting information for the images, an information processing apparatus comprises: an input unit which inputs object identification information for identifying an object; an acquiring unit which acquires one or more schemas related to the object and medical image data related to the schema, an identification unit which identifies a disease locus region in medical image data respectively related to each of the one or more schemas, a time-series schema generating unit which generates a time-series schema of the disease locus, a time-series image data generating unit which generates time-series image data of the disease locus, and a display output unit which synchronizes and outputs the time-series schema of the disease locus and the time-series image data of the disease locus.

The invention provides an edge time reading circuit. Some sensors carry output information by time quantum, and particularly by generation time of an analog edge. Unit sensing time is long, so that when sensor arrays are arranged in groups, time cost of reading out a serial line is high. The edge time reading circuit disclosed by the invention comprises an analog-to-digital conversion module, an edge-pulse conversion module, a time sequence generation module, a snapshot module and a reading module, and can be used for reading analog edge generation time quantum in parallel and converting data read in parallel into serial output, thereby providing an important technology for grouping of the sensors with a time quantum output manner.

The invention relates to a network information detection method and apparatus, a computer readable storage medium and a computer device. The method comprises the following steps: obtaining a propagation tree structure of network information, and publish contents and feature information of propagation nodes in the propagation tree structure; based on the hierarchical relationship between the propagation nodes in the propagation tree structure, generating a prediction model corresponding to the propagation tree structure according to a time sequence of propagation, wherein the hierarchical relationship between neural network nodes of the prediction model corresponds to the hierarchical relationship between the propagation nodes; and inputting the publish contents and the feature informationof the propagation nodes to the neural network nodes corresponding to the prediction model, and obtaining a detection result of the network information according to the parameters of the prediction model and the prediction model. According to the method, whether the publish contents are rumors is predicted from the propagation tree structure of the network information, and meanwhile, the publish contents and the feature information of the propagation nodes are considered in the detection of the rumors, thereby further improving the information amount necessary for the rumor detection, and thusimproving the accuracy of the rumor detection.

A multi-modal emotion recognition system is disclosed. The system includes a data input unit for receiving video data and voice data of a user, a data pre-processing unit including a voice pre-processing unit for generating voice feature data from the voice data and a video pre-processing unit for generating one or more face feature data from the video data, a preliminary inference unit for generating situation determination data as to whether or not the user's situation changes according to a temporal sequence based on the video data. The system further comprises a main inference unit for generating at least one sub feature map based on the voice feature data or the face feature data, and inferring the user's emotion state based on the sub feature map and the situation determination data.

The presented method comprises: (1) with satellite cloud chart sequence, generating the differential cloud chart sequence for wavelet transformation to select wavelet domain as the bottom data source and information extraction source; (2) according to the space correlation, taking adaptive state block division with EHMM to the target; (3) according to the time correlation, dynamic tracking and scaling motion energy with the cross entropy; (4) constructing the energy function to primary find out the typhoon center by the least energy value on the low-resolution layer. This invention has strong versatility and high location precision.

The invention provides an industrial data generation method based on a neural network model, and the method comprises the following steps of generating a large-scale data set by taking a time sequencegeneration process as a continuous decision-making process through an identification feedback mechanism based on the idea of a generative adversarial network; wherein the discriminator extracts timesequence characteristics and evaluates the importance of each characteristic to the sequence, and the quality of the generated data is measured by training a real sample and a generated sample, and the discriminator feeds back a corresponding reward value of a data probability generated in each step in the generator through time difference learning, the generator based on the LSTM is trained by astrategy gradient of reinforcement learning, and the reward value is provided by a return value of the discriminator. According to the industrial data generation method based on the neural network model, a new mode of generating big data from small data is realized, so that the mining analysis effect of data mining in a big data environment is improved, and the whole framework has the reliable performance and the excellent expandability.

A method is provided that is performed in a network having nodes that generate heterogeneous logs including performance logs and text logs. The method includes performing, during a heterogeneous log training stage, (i) a log-to-time sequence conversion process for transforming clustered ones of training logs, from among the heterogeneous logs, into a set of time sequences that are each formed as a plurality of data pairs of a first configuration and a second configuration based on cluster type, (ii) a time series generation process for synchronizing particular ones of the time sequences in the set based on a set of criteria to output a set of fused time series, and (iii) an invariant model generation process for building invariant models for each time series data pair in the set of fused time series. The method includes controlling an anomaly-initiating one of the plurality of nodes based on the invariant models.

A system and method automates the creation of business models via generators that generate data structures and models based on general assumptions regarding business models. A time series generator automatically generates a time series model suitable for creating a spreadsheet, even though the input description of the business model may be time-independent. A cross-category generator creates a cross-category hierarchy, even though the business model is described using independent categorizations, such as market categories, product-line categories, organizational categories, and so on. In this manner, the creator of the business model is freed of the tedium generally associated with creating a business model, and the likelihood of errors in the resultant models is substantially reduced. Further, the same input description of the business model can be used as the source of alternative models, depending upon the requirements of the intended application of the model.

The invention provides a method for generating time series based on generating antagonistic network. Based on the idea of generating antagonistic network, the invention designs a time series generating antagonistic network, and generates large-scale initial training data through limited time sequence data. The method includes designing a generator to generate data based on a long-term and short-term memory model LSTM, wherein the LSTM is used to learn potential temporal dependencies between the data for generating data with time dependencies that conform to the data distribution. A feature discriminator is designed based on scalable hypothesis testing to authenticate the quality of the generated data, the discriminator trains the LSTM-based generator by a policy gradient of reinforcement learning through a corresponding reward value of the data probability generated in each step of the temporal differential learning feedback generator, wherein the reward value is provided by the discriminator return value, thereby measuring the availability of the generated data.

This invention provides a sort of metrical method which realizes the forecast of the angular displacement of the numerical control rotating floor by the time-bar displacement sensing device. Add an intelligent interface unit between the time-bar and the numerical control system, the import of the intelligent interface unit is the absolute value of the angle which is measured by the time-bar; the export is the increment type impulse signal which can connects with the currency numerical control system. The intelligent interface unit progresses the forecast to the angular displacement figures in the unborn spell according to several discrete absolute value of the measured angle of the time-bar which is in the nonce and in the past. It corrects the predictive error figures which are in the former time slice according to the real measured angle figures which is in now. It realizes the predictive measure of the angular displacement of the numerical control rotating floor. This invention applies to the displacement sensing device with the timing sampling that the time-bar is used to be the deputation, and it applies to the position feedback of the whole closed loop which is from the function unit of the numerical control division likes the numerical control rotating floor, and so on.

A system for object recognition in which a multi-dimensional scanner generates a temporal sequence of multi-dimensional output data of a scanned object. That data is then coupled as an input signal to a trainable dynamic system. The system exemplified by a general-purpose recurrent neural network is previously trained to generate an output signal representative of the class of the object in response to a temporal sequence of multi-dimensional data.

The method for simulation generation of wind power times series based on an improved Markov chain comprises: dividing historical wind power time series into different data slots according to different months and time frames, and calculating the state-transition matrix of each data slot; fitting the probability distribution of the historical wind power fluctuation quantity, and generating a set of random numbers satisfying the distribution; taking a historical wind power average value as a zero moment value prior to the beginning of simulation of the wind power series, in the calculated state-transition matrixes of the improved Markov chain model, according to the months and time frames of the previous time of the simulation of the wind power, determining corresponding state-transition matrixes, generating accumulative state-transition matrixes, and obtaining the state of the wind power at the current moment, namely an interval where the wind power is located; and extracting the fluctuation quantity, overlapping the fluctuation quantity on the wind power value of the previous time as the wind power value of the current moment, and generating the wind power value at the next moment in a similar way until generating the wind power time series requiring the number of data.

The invention discloses a prediction method and device for traffic data of an electric power information network and a network upgrading method and device. The prediction method for the traffic data of the electric power information network comprises the steps that the traffic data of the electric power information network is acquired; a time sequence of the traffic data is generated according tothe traffic data; and traffic trend predication data, associated with time, of the electric power information network is generated according to the time sequence. Accordingly, the technical problem that in the prior art, the traffic data of the electric power information network cannot be predicted is solved.

The invention discloses a bridge circuit for an EMIF interface and an AHB / APB time series and a control method of the bridge circuit. The bridge circuit is connected with an external DSP through the EMIF interface, and comprises a DSP signal synchronization module, a DSP operation detection module, an address mapping controller, a data bit width matcher, an AHB / APB time series generation state machine and a configuration register. The bridge circuit is integrated in a SoC, and is connected with the external DSP through the EMIF interface to achieve the functions of analyzing, splicing and converting DSP access time series, and then finally realize the efficient operation for on-chip resources. The bridge circuit can support the access control, for 16 paths of AHB / APB slaves, of the DSP at most.

One embodiment of the description provides a method, device and processing device for correcting abnormal data in a time series. The method comprises: decomposing a time series, and at least decomposing out a first time series and a second time series, wherein the first time series reflects the change trend of the time series, and the second time series reflects the irregular change of the time series; correcting the first time series and / or the second time series respectively; and generating the corrected time series according to the corrected first time series and / or the corrected second time series. The technical scheme of the invention can automatically recognize the abnormal data in the time series and automatically correct the abnormal data, and can improve the efficiency and accuracy of correcting the abnormal data.

A system for object recognition in which a multi-dimensional scanner generates a temporal sequence of multi-dimensional output data of a scanned object. That data is then coupled as an input signal to a trainable dynamic system. The system exemplified by a general-purpose recurrent neural network is previously trained to generate an output signal representative of the class of the object in response to a temporal sequence of multi-dimensional data.

The invention provides an MIPI (Mobile Industry Processor Interface) module ESD (Electro-Static Discharge) protection device. The MIPI module ESD protection device comprises a data and clock abnormal state MIPI state machine reset circuit, a combined data line frequency dividing circuit, and an MIPI integral reset circuit for data abnormal width exceeding a certain time; the three circuits are connected in parallel with one another. For the data and clock abnormal state MIPI state machine reset circuit, a low speed switching to high speed time series detection module is used for detecting whether a low speed switching to high speed time series exists on an MIPI data line or not, a low speed and high speed detection module is used for detecting whether the MIPI clock line is in a low speed state or not, and if the two conditions are satisfied at the same time, a first external MIPI switching-off module is used for switching off an MIPI module input interface of a display screen driver chip; an MIPI state machine reset module is used for resetting an MIPI state machine; an MIPI clock line low speed switching to high speed time series can be automatically generated by a first clock line low speed switching to high speed time series generation module; a first MIPI switching-on module is used for switching on the MIPI input interface again according to the reset result of the MIPI state machine and the MIPI clock line low speed switching to high speed time series, so that a function of receiving MIPI clock and data signals of a mobile phone primary board is recovered.

The embodiment of the invention discloses a multi-time-series generation circuit and a liquid crystal display. The multi-time-series generation circuit can comprise a PWM chip, N voltage input ends, N voltage output ends, N switching circuits and N delaying circuits, wherein N is an integer larger than or equal to 2. The PWM chip comprises N PWM output pins. Each PWM output pin is connected with the corresponding switching circuit and the corresponding delaying circuit so that the closing time series of at least N switching circuits can be controlled respectively. The N switching circuits are connected with the N voltage input ends and the N voltage output ends respectively. When at least one of the N switching circuits is closed, the voltages of the voltage input ends connected with the switching circuits are transmitted to the voltage output ends connected with the switching circuits through the switching circuits. By the adoption of the multi-time-series generation circuit and the liquid crystal display, multiple time series signals can be generated through one PWM chip, the occupied area of the multi-time-series generation circuit on a circuit board can be reduced, and the cost of the multi-time-series generation circuit is reduced.

Performance information which is a possible generation cause of a fault is extracted accurately.A fault cause extraction apparatus 10 includes a storage unit 12 and a correlation-destruction-propagation detecting unit 25. Here, the storage unit 12 stores a correlation model including one or more correlation functions, each of which is generated based on a time series of performance information including a plurality of types of performance values in a system and transforms a performance value for one of the types being an input to a performance value for another one of the types being an output. The correlation-destruction-propagation detecting unit 25 calculates a degree of influence of a performance value for a base point that is one of the types upon the system, based on the number of base point propagation functions, the base point propagation functions being the correlation functions used to transform a performance value for the base point to a performance value for one of other the types being transformable from the base point directly or indirectly and not being input of the correlation function.

The embodiment of the invention provides a power supply with a switch control function. The power supply comprises an automatic switch control device. The automatic switch control device comprises a switch state editing unit, a time sequence generating unit, and an output state control unit, wherein the switch state editing unit is used for generating a switch sequence composed for a plurality of groups of switch states and a switch state table is obtained according to the switch sequence; the time sequence generating unit is used for generating a time sequence composed of a plurality of groups of delay time and a time sequence table is obtained according to the generated time sequence; mapping relationship exists in each switch state group and each delay time group; and the output state control unit is used for controlling on or off of power supply output signals in the corresponding time according to each switch state group and each delay time group in mapping relationship in the switch state table and the time sequence table. The power supply has the advantages of saving manpower, avoiding errors of manual operation, and realizing accurate delay opening and delay closing control.

A server includes a time-series generator to receive a sequence of unlabeled data records for a first user equipment. The unlabeled data records include values of measurements performed by the first user equipment on signals received from at least one base station. The server also includes a localization engine to estimate locations of the unlabeled data records based on the values of the measurements, a labeled dataset representing a channel model of a geographic area, and a map representative of the geographic area.