45 results about "Model domain" patented technology

A method and an Adaptive Knowledge Management is provided. The Adaptive Knowledge Management System is used for assisting a user with decision making by providing real-time, on-line automated recommendations for actions in a monitored vehicle troubleshooting, performance trend monitoring, health management and preemptive maintenance domain diagnostics and prognostics. The system creates a Structured Knowledge Repository, constructed from models, historical data, and heuristics for organizing a model domain knowledge. It uses a plurality of Analytical and Machine Learning tools for capturing knowledge from data sources and populating cells of the Structured Knowledge Repository. A Mixed-Initiative Planning module is used for interpreting operation goals for the monitored vehicle and utilizing the Structure Knowledge Repository for developing recommendations for user decision making. A plurality of Mixed-initiative Decision Support tools use the feedback from the Mixed-initiative Planning module and query the Structured Knowledge Repository for incorporating the extracted knowledge and information into outputs dealing with current issues and contingencies.

A Unified Debugger that enables the debugging of graphical models with combinations of entities from multiple domains such as time-based block diagrams, statecharts, and physical system models, with additional features for the debugging of hand-written and automatically generated code is disclosed. A view of the entity being executed that is consistent with the type of the modeling domain to which the entity currently being executed belongs is automatically generated for a user. Rather than requiring a user to manually transition from one debugger to the next, the Unified Debugger applies the appropriate debugging features from each modeling domain. The Unified Debugger also enables integration of diagnostic information from a multitude of modeling domains into a central user interface. In addition, a standard diagnostic interface which spans a plurality of modeling domains is defined that enables the customization of debug behavior.

An apparatus performs event processing in a mixed-language simulator. The apparatus comprises logic for generating a list of actions to be performed for a given trigger event occurrence and for processing at least some of the listed actions in a source-code-to-HDL-interface domain before performing a context switch from the source-code-to-HDL-interface domain to a source code model domain. By listing actions to be performed for a given trigger event occurrence and then processing listed actions before switching contexts, the amount of context switching between the HDL model domain and the source code model domain that occurs during the simulation is reduced.

Computing device, computer instructions and method for processing input seismic data d. The method includes receiving the input seismic data d recorded in a data domain, solving a linear inversion problem constrained by input seismic data d to obtain a model domain and its energy, wherein the linear inversion problem is dependent on sparseness weights that are simultaneously a function of both time and frequency, reverse transforming the model domain energy to the data domain, and generating an image of a surveyed subsurface based on the reverse transformed model domain energy.

A system and method for data reconstruction in soft-field tomography are provided. One method includes selecting a model domain for an EIT data set, determining a minimally anisotropic error in the model domain and correcting the model domain. The method also performing isotropization using the determined minimally anisotropic error to recover a boundary shape and isotropic conductivity for the EIT data set.

The invention relates to the technical field of voice recognition and discloses a method and system for eliminating reverberation in remote voice recognition. A novel model domain compensation method in the remote voice recognition comprises the steps: an indoor acoustic environment is simulated and impact response sequences of rooms in different positions are generated through inputting indoor space sizes; clustering analysis is conducted on the generated room impact response sequences, so that the indoor acoustic environment is divided into a plurality of zones, and a corresponding compensation matrix of one impact response sequence of each zone is obtained; in an established recognition network, compensation according to frames is conducted on the recognition network through the compensation matrix of each zone and the optimized compensation is obtained from a plurality of recognition results through the maximum posterior probability thought. Due to the fact that the clustering analysis is conducted on the acoustic environment, model compensation with distinctiveness is conducted on the recognition network, and reverberation resistant performance of the remote voice recognition in the indoor environment is greatly improved.

A method for processing seismic data may include receiving input seismic data (d_i) comprising N spatial coordinates, where the input seismic data is in a first spatial domain, expanding the N spatial coordinates of the input seismic data (d_i) to N′ modified spatial coordinates, where N′ is greater than N, to provide spatially expanded seismic data (d_e) that is in a second spatial domain, transforming the spatially expanded seismic data (d_e) to a model domain to provide model domain data (d_m), and generating a final image (d_f) of a subsurface using the model domain data (d_m).

A non-blended dataset related to a same surveyed area as a blended dataset is used to deblend the blended dataset. The non-blended dataset may be used to calculate a model dataset emulating the blended dataset, or may be transformed in a model domain and used to derive sparseness weights, model domain masking, scaling or shaping functions used to deblend the blended dataset.

The invention relates to a voiceprint recognition device, and particularly relates to an anti-noise voiceprint recognition device with joint treatment of spectral subtraction and dynamic time warping algorithm. The voiceprint recognition device comprises a microphone, a pre-amplification module, an analog-to-digital circuit (ADC), an endpoint detection module, a signal framing module, a spectral subtraction module, a first mel frequency cepstrum coefficient (MFCC) feature extraction module, a second MFCC feature extraction module, a third MFCC feature extraction module, a fourth MFCC feature extraction module, a current template, a user module, a first dynamic time warping (DTW) matching module, a second DTW matching module, a third DTW matching module, a score feedback module, an information compensation module and a result output module. Embedded voiceprint recognition based on isolated words is carried out by using figures from 0 to 9 as passwords, joint treatment is carried out by using signal domain and model domain of voice to prevent recognition performance reduction caused by information loss in the voice enhancing process and the voiceprint recognition performance under the noise background is improved. The system is convenient to implement and low in cost.

The invention discloses a universal configurable, compatible multi-model and domain-migratable multi-round dialogue system. The system comprises: a configurable module used for carrying out self-defined configuration according to professional knowledge of a current domain; a dialogue understanding module used for carrying out language coding on user statements through dialogue understanding, and carrying out intention recognition and joint extraction of slot value pairs by replacing a compatible dialogue understanding model; a dialogue state tracking module used for acquiring current multi-round historical dialogue state information; a strategy learning module used for performing action decision through the dialogue strategy model; a dialogue generation module used for generating dialoguesaccording to corresponding actions; and an online training module used for serving as a user simulator to be provided for developers, performing online iterative updating of the model and generatingavailable data. The whole architecture module of the system is universal, the system adapts to different application scenes, all field parameters are universal and configurable, corpora and dialogue understanding can be accepted at will, and a dialogue strategy model can be flexibly expanded and replaced.

The invention discloses a sound insulation simulation calculation method of an acoustic metamaterial plate. The sound insulation simulation calculation method comprises the following steps: establishing a three-dimensional model of the acoustic metamaterial plate; selecting an acoustic-solid coupling module from commercial software COMSOL; importing the geometric model into a COMSOL and setting aunit; establishing a cavity and forming a combined body with the imported model; setting global parameters; setting a variable and an integral operator; using the solving module for adding materials to the corresponding model domains and setting the corresponding domains; setting a load and a boundary condition; dividing the model into grids; adding research, and setting the frequency range and step length of solving; and calculating and adding a one-dimensional drawing in commercial software, drawing a structure sound insulation index curve, and exporting data. According to the sound insulation simulation calculation method, complex theoretical calculation is avoided, and the design efficiency is optimized, and the sound insulation simulation calculation method is of great significance toresearch of the acoustic metamaterial.

The invention includes a method for placement of sidelobe inhibitors on a radiation-patterning tool. Elements of the tool are represented by design features in a modeling domain. The modeling domain is utilized to generate vectors spanning between edges of design features within a threshold spatial distance of one another. Locations of vector midpoints are identified. The locations are utilized in identifying areas where sidelobe overlap is likely. The areas are shifted from the modeling domain to a real domain, and in the real domain correspond to regions of the tool where sidelobe overlap is likely to occur. Sidelobe inhibitors are formed across at least some of the identified regions of the tool. The invention includes computer readable code that can enable a computer to determine locations for placement of the sidelobe inhibitors.

A method, apparatus and computer-program product for performing a system analysis of a system is disclosed. The system is represented as a plurality of independent model domains, wherein the model domain represents selected ones of a plurality of components and the relationships among the components and the method comprises the steps of representing, as proxies, selected ones of the plurality of components of a first domain, wherein the relationships among the components are limited to direct relationships among adjacent ones of the components, wherein the first domain representation is available to a second domain, providing events occurring in the first domain to the second domain, providing a mapping between a plurality events and a plurality of observable events occurring among the proxy components of the first domain and a plurality of observable events occurring among the components of the second domain, wherein the mapping is represented as a value associating each event with each observable event, and performing the system analysis based on the mapping of events and observable events.

The invention discloses a cross-domain three-dimensional model retrieval method based on a multi-level feature alignment network, and the method comprises the steps: carrying out virtual photographingof a three-dimensional model in a three-dimensional model database through a virtual camera, and generating multi-view data; constructing a multi-level feature alignment network; performing domain-level feature alignment on an image and the features of the three-dimensional model by s discriminator through an alignment network; calculating centroids of all features of each class in the two domains of the image and the three-dimensional model based on the alignment network, taking the distance between the centroids of the same class as a part of a loss function, and minimizing the loss function by using a back propagation algorithm; and when the loss function is minimized, respectively extracting features of the image and the three-dimensional model by utilizing the trained multi-level feature alignment network, and carrying out cross-domain three-dimensional model retrieval. According to the method, a new network framework, namely a multi-level feature alignment network, is provided,feature alignment is carried out on an image domain and a three-dimensional model domain on the domain level and the class level, and the cross-domain three-dimensional model retrieval precision is improved.

In an apparatus for identifying software components having an object and use-case information related to a model domain in a software component developing system, the object and use-case information is inputted in an input section. A UDA table creation section makes use-relation between an object and a use-case transmitted from the input section into a UDA (Use-case Data Access) table. A clustering engine creates at least one identified component by normalizing affinity values between use-relation factors in a UDA table prepared in the UDA table creation section and by performing a clustering only for a value greater than or equal to a threshold among the normalized affinity values.

The invention relates to a JPEG image mismatch steganography analysis method based on heterogeneous characteristic subspace migration. The method provided by the invention takes characteristics of different fields as a combination of domain-independent and domain-related characteristic subspaces, and the method comprises the following steps: firstly providing a low-rank constrained domain-independent characteristic migration method, and realizing local information migration by utilizing local data characteristics between different fields; secondly, based on sparse representation of modeling domain-related characteristics, measuring influence produced by field change on the modeling domain-related characteristics; and finally, by virtue of the steps, constructing an objective function, andsolving by virtue of an imprecise augmented Lagrange multiplier method. Unique global information of the domain-related characteristics is considered while local information migration is realized in adomain-independent characteristic subspace, so that characteristic differentiation degree between a cover image and a stego image can be improved, detection effect of mismatch steganography analysiscan be beneficially improved, and great significance is produced to the mismatch steganography analysis.

The invention provides an intelligent manufacturing-oriented field modeling method and system, and the method comprises the steps: defining a field meta-model, and specifying the inheritable field elements and field specifications of field modeling; constructing a modeling primitive, performing unified modeling on the domain elements of the to-be-modeled domain by using the modeling primitive, and constructing an initial application domain model; unifying the modeling primitives and the domain context concept, and clarifying the context boundary of the initial application domain model; constructing a knowledge graph based on the field specification of the field meta-model, and performing knowledge completion on the initial application field model with the clear context boundary to obtain a complete application field model; during modeling, on the basis of a field meta-model defined in advance, in combination with field knowledge reasoning, facing specific scenes, complementing and describing demands, and perfecting incomplete demands; performing context division on a field, the range of uncertainty is limited by establishing a clear boundary, and the uncertainty is gradually reduced through data transmission.

The invention discloses an image recognition model migration method and device, equipment and a storage medium, and relates to the technical field of artificial intelligence. The method comprises the following steps: acquiring head feature maps of a source domain image sample and a target domain image sample; determining object prediction labels of the two samples based on the head feature maps of the two samples; determining model training loss based on difference information between the feature maps of the two samples and difference information between an object prediction label and a true value label of the source domain image sample; and adjusting parameters of the image recognition model according to the model training loss to obtain a migrated image recognition model. Single-stage domain migration is realized based on the head feature map, and the migration efficiency and generalization of the model can be improved. In addition, the model can be migrated to a real scene for application by utilizing simulation data with true value annotation and real scene data without true value annotation. For example, the method and device can be used for a model domain migration task of an automatic driving scene.

Methods and apparatus provide for an event identity manager that defines a topology of resources. The topology of resources represents relationships amongst the resources in a data center and each of the resources comprises resource attributes described in a data model. The event manager acquires event data describing an event that occurred amongst the resources. In addition, the event data comprises a format that is incompatible with the data model. The event manager correlates the event data with the data model in order to determine a locus of the event data within the topology of resources, the locus of the event data being a resource where the event is most likely associated.

The invention relates to a failure test case generation method for a model domain, and the method comprises the steps: firstly defining a minimum suspicious set, and extracting common features from all failure test cases; and then, for each failed test case in the model domain, on the basis of reserving the common characteristics of the failed test cases, mutating the information of the non-common characteristics so as to generate a new failed test case of the model domain; and finally, fusing the failure test cases of the newly added model domain into the original test case set, and carrying out defect positioning. According to the method, the variant nearest neighbor algorithm is used for generating the failure test case vector of the model domain, so that the purpose of improving the defect positioning effect is achieved, and compared with a traditional method for generating the test case from the input domain, the method is simpler and more effective, and a failure label does not need to be obtained through program execution.

A noise robustness method operates jointly in a signal domain and a model domain. For example, energy is added in the signal domain for frequency bands where an actual noise level of an incoming signal is lower than a noise level used to train models, thus obtaining a compensated signal. Also, energy is added in the model domain for frequency bands where noise level of the incoming signal or the compensated signal is higher than the noise level used to train the models. Moreover, energy is never removed, thereby avoiding problems of higher sensitivity of energy removal to estimation errors.

The invention discloses a scheduling task execution method and system. The method comprises the following steps of: presetting a model domain, a scheduling domain and a technical domain; enabling themodel domain to extract and store information data of various facilities; wherein the scheduling domain has an execution method for various scheduling tasks; driving technologies corresponding to various facilities are arranged in the technical field; by acquiring intiation information of the scheduling task, creating the scheduling task according to a corresponding preset task scheme, obtaining and analyzing the scheduling task through the scheduling domain, sending task information to the technical domain, obtaining information data from the model domain by the technical domain according toa target facility identifier in the task information, and selecting a target driving technology to execute the scheduling task according to the task information. Compared with the prior art, the invention has the advantages that the model domain, the scheduling domain and the technical domain are divided, so that the execution of the scheduling task can avoid the limitation of facility types, various facilities can be uniformly operated according to the requirements of the scheduling task, and the execution complexity of the scheduling task is reduced.