206 results about "Source field" patented technology

An apparatus and method for locating a magnetic implant in a surgical application using the field of a source magnet for the implant guiding field. The source magnet is an electromagnet having a separate calibrated magnetic field component in addition to the guiding field, so that both the magnitude and orientation of the magnetic field as a function of position around the magnet are known. A magnetic implant is provided with a sensor, such as a three-axis Hall effect sensor, to provide an indication of the magnitude and orientation of an applied magnetic field when the implant is surgically implanted in a patient. After implantation, the source magnet is energized with a current having a modulated component. The modulated component is received and filtered from the signal received from the Hall effect sensor in the implant, and provided to a processor that computes the location of the implant.

An apparatus and method for locating a magnetic implant in a surgical application using the field of a source magnet for the implant guiding field. The source magnet is an electromagnet having a separate calibrated magnetic field component in addition to the guiding field, so that both the magnitude and orientation of the magnetic field as a function of position around the magnet are known. A magnetic implant is provided with a sensor, such as a three-axis Hall effect sensor, to provide an indication of the magnitude and orientation of an applied magnetic field when the implant is surgically implanted in a patient. After implantation, the source magnet is energized with a current having a modulated component. The modulated component is received and filtered from the signal received from the Hall effect sensor in the implant, and provided to a processor that computes the location of the implant relative to the electromagnet based upon the detected magnitude and orientation of the modulated component of the magnetic field, and the location and orientation of the electromagnet. Localizers may be used to supply the relative locations of the patient and the electromagnet to the processor. A display may be provided to display a representation of the location of the implant in the patient, which may also be superimposed over a preoperative image of the patient.

An electronic content guide is disclosed. The electronic content guide preferably includes a source field for listing content sources and output devices. The electronic content guide also includes a first content field for listing schedule data corresponding to first content data presently available for viewing from each one of the content sources and for listing display information corresponding to the first content data being displayed for viewing at a present time by each one of the output devices. Preferably, the source field and the first content field are displayable in parallel columns. The electronic content guide can also include a second content field, displayable in a column parallel to the source field and the first content field, corresponding to a second content data. In one embodiment, the first content data from each content source can be displayed simultaneously on more than one of the output devices.

An electronic content guide is disclosed. The electronic content guide preferably includes a source field for listing content sources and output devices. The electronic content guide also includes a first content field for listing schedule data corresponding to first content data presently available for viewing from each one of the content sources and for listing display information corresponding to the first content data being displayed for viewing at a present time by each one of the output devices. Preferably, the source field and the first content field are displayable in parallel columns. The electronic content guide can also include a second content field, displayable in a column parallel to the source field and the first content field, corresponding to a second content data. In one embodiment, the first content data from each content source can be displayed simultaneously on more than one of the output devices.

Embodiments of the invention comprise a method and apparatus for transforming data. Systems implementing the invention allow for loading the data from one or more source tables into one or more destination tables. The system allows for parsing source data fields, defining new source data fields, combining source fields to create source field combinations, and combining destination fields to create destination field combinations. The system allows for mapping source fields and values to destination fields and values, where either the source fields and values or destination fields and values may be field and value combinations, and allows for transforming field values based on destination field type. The tools provided in a system implementing the invention provide a user the ability to intervene at each step during the data transform task. A user may manually input a hierarchy definition, a specific mapping, define rules for combining data fields, or define rules for converting data values.

The present invention provides a method and device for determining data consanguinity based on structural data. The method comprises: parsing a selection statement to obtain a source abstract syntax tree, and organizing the table information and the field information obtained by traversing the source abstract syntax tree into a source list layer by layer; obtaining the target abstract syntax treeby parsing the insertion statement, and organizing the table information and the field information obtained by traversing the target abstract syntax tree into the target list layer by layer; traversing the source manifest to obtain the source table information and traversing the target manifest to obtain the target table information to obtain the table granularity data kinship relationship; extracting the target field information for the target table from the target manifest, starting from the first layer of the source manifest, finding the source field in the source table with the same name as the target field information of the target table layer by layer, and when the source table to which the source field belongs no longer comes from the subquery, determining the corresponding source field to be the source field with blood relationship corresponding to the target field information. The scheme can parse the field-level granularity of data consanguinity based on structured data.

An embodiment of the invention provides a data table migrating method and device. The method comprises steps as follows: target data in a currently to-be-migrated target source field are acquired from a source data table of a source database; field conversion information configured by a user in a configuration file is called and comprises the correspondence relation of a source field and a target field as well as a conversion function of the data format of the source field and the data format of the target field, wherein the target field is a field in a target data table; on the basis of the field conversion information, the data format of the target data is converted into the data format of the target field corresponding to the target source field; the target data in the converted data format are migrated to the target field corresponding to the target source field in the target data table. With the adoption of the method and the device, data of codes required to be written for data migration can be reduced, and the code writing complexity can be reduced.

Methods are described for assessing material condition. These methods include the use of multiple source fields for interrogating and loading of a multicomponent test material. Source fields include electric, magnetic, thermal, and acoustic fields. The loading field preferentially changes the material properties of a component of the test material, which allows the properties of the component materials to be separated. Methods are also described for monitoring changes in material state using separate drive and sense electrodes with some of the electrodes positioned on a hidden or even embedded material surface. Statistical characterization of the material condition is performed with sensor arrays that provide multiple responses for the material condition during loading. The responses can be combined into a statistical population that permits tracking with respect to loading history. Methods are also described for measuring the stress in the material by independently estimating effective electrical properties, such as magnetic permeability or electrical conductivity, using layered models or predetermined spatial distributions with depth that are then correlated with the stress.

Gallium nitride material devices and methods associated with the same. In some embodiments, the devices may be transistors which include a conductive structure connected to a source electrode. The conductive structure may form a source field plate which can be formed over a dielectric material and can extend in the direction of the gate electrode of the transistor. The source field plate may reduce the electrical field (e.g., peak electrical field and / or integrated electrical field) in the region of the device between the gate electrode and the drain electrode which can lead to a number of advantages including reduced gate-drain feedback capacitance, reduced surface electron concentration, increased breakdown voltage, and improved device reliability. These advantages enable the gallium nitride material transistors to operate at high drain efficiencies and / or high output powers. The devices can be used in RF power applications, amongst others.

The invention discloses a method for establishing a large-scale cross-field text emotion orientation analysis framework. The method comprises the steps that precise word segmentation is carried out on sample documents in a source field and sample documents in a target field to form two word vector tables; word vectors are clustered, and the fields are aligned; primary sentence modeling is carried out on calibration samples in the source field through the word vectors and serves as input of DCELM, and interlayer abstraction features of text vectors are extracted through convolution operation; convolution layer parameters obtained when the classification effect of a verification set is best are recorded and serve as parameters of a DCELM network convolution layer; finally, interlayer abstraction features of calibration samples, extracted through DCNN, of a small number of target fields are used for training implicit layer parameters of a classifier ELM to establish the large-scale cross-field text emotion orientation analysis framework. By means of the technical scheme, the difference of words for expressing emotion polarities among the fields is eliminated on the sample layer, the defects that on a full-connection layer, local optimization is easily caused and the generalization ability is weak are effectively solved, and the anti-interference performance of a model is improved.

The invention discloses a semiconductor device and a manufacturing method thereof. The device comprises a substrate, a multi-layer semiconductor layer positioned on the substrate, a source electrode and a drain electrode positioned on the multi-layer semiconductor layer, a grid electrode positioned between the source electrode and the drain electrode, as well as a dielectric layer positioned on at least one part of the surface of the multi-layer semiconductor layer arranged between the grid electrode and the drain electrode and a source field plate positioned on the dielectric layer, wherein a groove is formed in the dielectric layer, the source field plate is electrically connected to the source electrode through at least one conductive path, and the source field plate completely or partially covers the groove in the dielectric layer. The semiconductor device and the manufacturing method thereof provided by the invention have the advantages that the source field plate is given a full play, grid source capacitance Cgs is reduced, a peak value electric field at the edge of the grid electrode is reduced, the breakdown voltage of the device is improved, the grid leakage current of the device is reduced, the frequency characteristic of the device is improved, and the advantage of the high output power of the device is fully played.

The invention provides an unsupervised field adaptation method and system based on adversarial learning and a medium, and the method comprises the steps: a feature extraction step: employing a featureextraction network to extract the features of images in a source field and a target field, and obtaining the image features of the source field and the image features of the target field; a categoryprediction step: predicting the probability that the image belongs to each category according to the obtained image feature of the source field and the image feature of the target field, and obtaininga category prediction probability; and a field discrimination step: predicting the probability that the image feature comes from the source field and the target field through a field discrimination network according to the obtained image feature of the source field and the image feature of the target field, and obtaining a field prediction probability. According to the method, the characteristicsthat the image extraction fields in the source field and the target field are invariable and have relatively high discrimination capability can be realized, so that the unsupervised field adaptationis realized.

The present invention provides a vehicle for the exchange of data between a database and an independent spreadsheet. Specifically, the present invention may be used to import data from a database to a spreadsheet and / or to export data to a spreadsheet. To facilitate this exchange, one or more controls are entered into cells in the spreadsheet. The controls specify the manner in which the data is exchanged between the database and the spreadsheet. For example, a control may specify a data type, a formatting parameter, a source field, a destination field, and / or a formula for a designated cell in the spreadsheet. The controls are executed by a linker that links the data in the database and the data in the spreadsheet to effect the exchange of data between the two.

An improved magnetic torque transducer arrangement for self-compensating effects of external magnetic sources and temperature offset comprises a shaft with at least one magnetized zone, at least one active magnetic field sensor and at least one passive magnetic field sensor disposed in such a way that active field sensor always in a position with higher magnetic field strength arise from applied torque than that of passive sensor. Passive field sensors may also be placed in both sides of the active field sensor, or on one side of active field sensor only. The transducer output is obtained by subtract the output of passive field sensors from that of active field sensor thus cancel out the effect of interfering magnetic field flux and temperature offset on the torque transducer, and partially filter out temperature sensitivity drift and rotational dependant signal. The sensitivity of active and passive field sensors can also be electrically matched by calibrating them in a uniform magnetic field, thus a completely common mode rejection can be achieved. The sensor arrangements may also be utilized in other type of sensors that extract changes in magnetic fields to indirectly detect direction, speed, presence, force, linear position, or angle to cancel out interfering magnetic field and temperature offset effect.

A method for measuring a compact range antenna by three reflectors comprises the following steps of: determining a system mapping function according to the distribution of a feed source field and the distribution of a system outlet field; tracking and analyzing the electromagnetic wave emitted by the feed source by adopting a dynamic waveband tracking theory; and determining all mirror surface parameters of a first forming auxiliary reflector and a second forming auxiliary reflector by the system mapping function, the feed source field wavefront parameters and the main reflector surface parameters. The embodiment of the invention also discloses a system for measuring compact range antenna by three reflectors. The system and the method can be applicable to the feed source higher than 300GHz, and lead the system outlet field to have large quiet zone utilization coefficient (achieving more than 70%), high cross polarization isolation degree, compact structure and wide measurement frequency bands.

The invention discloses a pedestrian re-recognition method based on transfer learning and aims at solving a problem that pedestrian samples to be re-recognized are less. The method has the following steps of (1) pedestrian prospect segmentation: acquiring known pedestrians in different cameras under different scenes; (2) feature extraction: extracting features of the known pedestrians; (3) improved nerve network model learning in a source field: learning through the improved nerve network model and acquiring a model parameter which is shown in the description; (4) transfer learning from the source field to a target field: through transferring the model parameter to an improved nerve network model parameter in the target field, learning and acquiring a target field model parameter; (5) pedestrian re-recognizing: using the improved nerve network model in the target field to carry out pedestrian re-recognition discrimination.

The invention relates to a sentiment classification method based on multi-source field instance migration. For the phenomenon that the migration efficiency of a TrAdaBoost algorithm in migration learning may be lowered, multi-source learning is introduced, by trying sample migration from different source fields or combining the features of multiple source fields, migration learning is stable and effective, much initial weight is distributed to a target sample, so that weight mismatching is relieved, and the phenomenon of quote imbalance is improved by resampling data at each step; and for the phenomenon of source field weight premature convergence in the TrAdaBoost algorithm, dynamic factors are added, and the problem that weight entropy is transferred from a source sample to the target sample is solved. According to the method, premature convergence of the weight of the source field sample which is small in the correlation with the target field is avoided, a learning target task is helped together, and knowledge of all source fields is fully used.

The invention provides a deep migration learning method based on entropy minimization, and the method comprises the steps: S1) dividing a source field and a target field according to different migration learning tasks, constructing a migration learning network, and initializing hyper-parameters of the migration learning network; S2) inputting respective data samples of the source domain and the target domain into a transfer learning network and carrying out forward propagation to obtain a network prediction label; training the whole network by using a stochastic gradient descent method according to the proposed loss function, completing updating of network parameters by using back propagation, and stopping training until the model converges or reaches the maximum number of iterations; S3)storing the network model and a training result; according to the target domain label prediction method based on the entropy minimization technology. By maximizing the distribution diversity of each batch of data prediction categories in the target domain, a common solution result occurring when only the entropy minimization technology is used is avoided, and the reliability of transfer learning is guaranteed.

The invention discloses a method for reducing dimensions of hyper-spectral data on the basis of pairwise constraint discriminate analysis and non-negative sparse divergence, and belongs to methods for processing hyper-spectral remote sensing images. The method aims to solve the problem of deterioration of the classification performance of most advanced algorithms for classifying hyper-spectral data on the basis of machine learning when source hyper-spectral data and target hyper-spectral data are distributed differently. The method includes firstly, performing pairwise constraint discriminate analysis according to pairwise constraint samples; secondly, designing a non-negative sparse divergence criterion to create a bridge among source-field hyper-spectral data and target-field hyper-spectral data which are distributed differently; thirdly, combining the pairwise constraint discriminate analysis with the bridge to transfer knowledge from the source hyper-spectral data to the target hyper-spectral data. The pairwise constraint samples containing discriminate information can be automatically acquired. The method has the advantages that the knowledge can be transferred among the hyper-spectral data acquired at different moments, in different areas or by different sensors; the information of the source-field hyper-spectral data can be effectively utilized to analyze the target-field hyper-spectral data, and high integral classification precision and a high Kappa coefficient can be acquired.

The invention discloses a self-attention multi-core maximum mean difference-based transfer learning speech enhancement method, which comprises the following steps: extracting GFCC features from original speech and using the GFCC features as input features of a deep neural network; calculating ideal floating value masking of Fourier transform domain by using noisy speech and clean speech information, and using the ideal floating value masking as the training target of the deep neural network; building a speech enhancement model based on deep neural network; building a self-attention multi-coremaximum mean difference-based transfer learning speech enhancement model; training the self-attention multi-core maximum mean difference-based transfer learning speech enhancement model; inputting frame-level features of the noisy speech in the target domain, and rebuilding enhanced speech waveform. By adding a self-attention algorithm to the front end of the multi-core maximum mean difference andby minimizing the multi-core maximum mean difference between features noticed by the source field and features noticed by the target domain, transfer learning of the unlabeled target domain is realized, and the speech enhancement performance is improved. The method of the invention has a good application prospect.

Method for redatuming seismic data to any arbitrary location in the subsurface in a way that is consistent with the internal scattering in the subsurface. Direct arrival times are estimated from every point to every point on the edges of a virtual box in the subsurface (102). Green's functions are estimated by iterative optimization (103), using the direct arrival times as initial guesses (102), to minimize error in the source field reconstruction, which consists of the multidimensional auto-correlation of the Green's functions. The estimated Green's functions are the used to determine simulated internal multiple reflections (104). The measured data may be corrected by subtracting the simulated internal multiple reflections, or the Green's function may be used to do local imaging or local velocity model building, particularly advantageous in full wavefield inversion (104).

The invention discloses a large-scale dynamic evolution dust type aerosol retrieval method. The large-scale dynamic evolution dust type aerosol retrieval method comprises step 1 training a classifier; acquiring a source field sample and a cloud layer sample of a dust type aerosol layer and a target field sample of the dust type aerosol layer according to passive type satellite sensor remote sensing data; acquiring characteristic vectors of the samples according to active type satellite sensor laser radar profile data; training a support vector machine classifier according to the characteristic vectors of the source field sample and the cloud layer sample; optimizing the classifier according to the characteristic vector of the target field sample and the transfer learning theory; step 2 performing retrieval on the dust type aerosol through an optimized classifier. The large-scale dynamic evolution dust type aerosol retrieval method can improve the cloud-sand classification accuracy after the dust is subjected to long-distance transmission, effectively reverse the situation that only low-accuracy data provided by NASA (National Aeronautic And Space Administration) can be used currently passively, solve the bottleneck problems that, in China, large-area dust vertical profile data is lost and the data usability is poor and the like and satisfy the observation requirement of dust aerosol source regions and diffusion transmission regions in China and even the globe.

An emergency alert message is included in a cable broadcast signal. The emergency alert message includes major and minor details channel numbers associated with a details channel. A details channel number is configured based on the major details channel number when the minor details channel number is set to a null value. Next, it is tuned to a channel frequency of a digital cable channel corresponding to the details channel number when the details channel number exists in a digital channel number list. Alternatively, the emergency alert message includes a details channel source field which indicates whether a details channel is a digital or an analog channel. Then, it is tuned to a channel frequency of one of a digital cable channel and an analog cable channel based upon a value of the details channel source field.

The invention discloses a fire alarming application layer data transmitting method based on CAN2.0B, it constructs the information pack according to the length of the transmission information, at the same time, constructed data frame labels include information pack field, transmitting source field, receiving target address field, and information pack data frame number field, the priority of the transmitting information can be defines easily, overcomes the deficiencies that there has no transmitting source address and target address field in the CAN protocol, it can be applied to fire alarming control system; the replay mechanism of the receiving joint guarantees the reliability of the data transmission; it uses multi-frame transmission, or multi-pack, multi-frame transmission, accords the transmission of pack whose length more than single information layer, and of the file whose length more than the maximal permissible length of the single information layer pack; it uses time-out repeating mechanism, joint failure detecting mechanism, buffer area repeating mechanism in the data transmission process, thus it can guarantee the accuracy and effectiveness of data transmission.

The invention discloses statement template generation system, development tools, development methodologies and client statements, including statements template development tools and templates generate reports associated data source field information, generate the corresponding template statements. In accordance with user needs , and generate reports from templates associated with the data source information in all fields, filtering out statements template adjustment; As adjusted information in the template statements include statements template not only in the field, and other data sources in the word paragraph, but shielding swap with the adjustment process irrelevant statements template temporary fields, therefore, statements, client statements. In accordance with information on the template adjustment statements template adjustment, not only to the use of data in the source field, but also at the same time shielding swaps and reporting templates temporary adjustment process irrelevant fields.

On a semiconductor substrate, a well is formed. In the well, one MOS transistor including a gate electrode, a source region, a source field limiting layer and a source / drain region, and another MOS transistor including a gate electrode, a drain electrode, a drain field limiting layer and a source / drain region are formed. The one and another MOS transistors are connected in series through the source / drain region common to the two transistors. Accordingly, a semiconductor device can be provided in which increase in pattern layout area is suppressed when elements including a high-breakdown voltage MOS transistor are to be connected in series.

The invention discloses a source field plate transistor with high electron mobility. The transistor comprises, from bottom to top, a substrate (1), a transition layer (2), a barrier layer (3), a source electrode (4), a drain electrode (5), a gate electrode (6), a passivation layer (7), a protection layer (10) and a source field plate (8); the source field plate (8) is electrically connected with the source electrode (4), wherein, n floating field plates (9) are deposited on the passivation layer (7) to increase the depletion region area of the barrier layer (n is not less than 1). The floating field plates and the source field plate are positioned on the same passivation layer, all the floating field plates have the same size and are mutually independent, and the distance between two adjacent floating field plates increases based on the number of the floating field plates arranged along the direction from the source field plate to the drain electrode. The n floating field plates are in a floating state and are completed together with the source field plate on the passivation layer by one-time process. The source field plate transistor has the advantages of simple process, good reliability and high breakdown voltage, and can be used for fabricating high power devices based on an III-V group compound semiconductor heterojunction structure.

Embodiments of the present invention disclose an interface conversion method, an interface converter, a storage medium, and an electronic terminal. The method comprises: obtaining a system source field set by a user based on a visual configuration interface; obtaining a matching relationship between a field in an interface document set by the user based on the visual configuration interface and the system source field, and generating an interface file based on the matching relationship; and when receiving a request field from a cooperating party, converting the request field into the system source field according to the interface file for the system to perform a request or response operation according to the business logic. By virtue of the technical scheme of the embodiments of the present invention, the cooperating party can be quickly connected without any modification to the code of the business system so as to realize the seamless connection with the cooperating party; and at the same time, the interface management of the cooperating party is carried out based on the visual configuration interface, the interface maintenance operation process is simplified, and the interface maintenance costs of multiple cooperation parties are reduced.