265results about How to "Reduce training" patented technology

A system, method, and device for recording and tracking resistance training parameters in connection with a resistance training workout. A specially configured portable recording device has a very compact user interface having an input device with very few inputs. A user may record resistance training parameters directly on the recording device both quickly and easily with few and simple keystrokes, resulting in an improved recording process. A personal trainer or strength coach may track the workouts of multiple exercisers with increased convenience. In a particularly preferred embodiment, the recording device is provided in the body of a sport watch.

Methods and systems for communication traffic control rule generation are provided. Configuration information for communication equipment, default information stored for the communication equipment, or both, is accessed. One or more parameters which affect processing of communication traffic by the communication equipment are determined from the accessed information and used to generate a communication traffic control rule to be applied to communication traffic at the communication equipment. The generated communication traffic control rule is applied at interfaces of the communication equipment to communication traffic being terminated by the communication equipment.

The present invention describes a magnetic targeting system suitable for guiding a biocompatible device to a target area within the body (in vivo) and method of using the same. The system includes a targeting member having a steering material and is attached to the biocompatible device. The system also includes at least one anchoring member constructed and arranged for the inclusion of a magnetic material effective for influencing the traversal of the steering material, in vivo. The magnetic material is configured and sized so as to positionable external of the anchoring member, in vivo. The magnetically influenced anchoring member interacts with the targeting member such that the biocompatible device is positionable relative to the target area. An extender and connector have threads indexed to a securing set screw to facilitate positioning and affixation of the biocompatible material

State-of-the-art speech recognition systems are trained using transcribed utterances, preparation of which is labor-intensive and time-consuming. The present invention is an iterative method for reducing the transcription effort for training in automatic speech recognition (ASR). Active learning aims at reducing the number of training examples to be labeled by automatically processing the unlabeled examples and then selecting the most informative ones with respect to a given cost function for a human to label. The method comprises automatically estimating a confidence score for each word of the utterance and exploiting the lattice output of a speech recognizer, which was trained on a small set of transcribed data. An utterance confidence score is computed based on these word confidence scores; then the utterances are selectively sampled to be transcribed using the utterance confidence scores.

An optical proximity corrected mask design is generated from a given a target mask design by processing the target mask design through a feature trained neural network, configured to perform an optical proximity correction of geometric features, to obtain a representation of a first corrected mask design. The target mask design is processed in parallel through a rule processor, configured to perform placement of sub-resolution geometric features relative to geometric features in the target mask design, to obtain a representation of a second corrected mask design. A layout reassembler operates to generate a corrected mask design through an overlaid composition of said first and second corrected mask designs.

A steady-state visual evoked brain-computer interface method is based on the stimulation of two frequencies of left and right visual fields, and relates to a visual evoked presentation method of a visual brain-computer interface. The brain-computer interface method is characterized in that four different targets are combined through two blinking frequencies by utilizing a principle of the crossing of the left and the right visual fields; and a correlational analysis method is used to identify which target is watched by a user after brain electrical acquisition equipment records a multi-brain electrical signal in the visual brain area. The equipment and the method of the invention comprise a recording mode of the brain electrical signal, lead selection and signal preprocessing, and a feature extraction method. The visual target presentation method of the invention can be realized through a screen display of a common computer, and the size, the brightness and even the color of the target can be controlled easily. Compared with other steady-state visual evoked brain-computer interfaces, the presentation method can increase the number of the targets under a condition that the usable frequencies are limited, thereby improving the brain-computer interaction ability.

The logistics operation training system includes manipulator assembling and conveying operation unit, automatic cargo distinguishing and buffering operation unit, stacking operation unit, automatic warehouse operation unit, visual detection operation unit, stacking operation unit, cargo flow direction controlling operation unit and container packing operation unit. Each of the operation units is provided with monolithic computer circuit and logic circuit with modular electric interface, and different units may be selected to form stepped configuration. The present invention is suitable for different teaching and training modes.

Methods and systems for building and using a parameter isolation model to isolate measurement signal information associated with a parameter of interest from measurement signal information associated with incidental model parameters are presented herein. The parameter isolation model is trained by mapping measurement signals associated with a first set of instances of a metrology target having known values of a plurality of incidental model parameters and known values of a parameter of interest to measurement signals associated with a second set of instances of the metrology target having nominal values of the plurality of incidental model parameters and the known values of the parameter of interest. The trained parameter isolation model receives raw measurement signals and isolates measurement signal information associated with a specific parameter of interest for model-based parameter estimation. The number of floating parameters of the measurement model is reduced, resulting in a significant reduction of computational effort.

The invention comprises, in one form thereof, an operator interface system for a touch screen device having a simplified graphic interface. The operator interface system includes a start screen having fewer than five start screen buttons. The touch screen device is generally a kiosk located within a store. A store map screen is linked to one of the start screen buttons and includes several department buttons integrated into a store layout. A products screen is linked to one of the start screen buttons and provides access to information about the plurality of products sold by the store organized into an intuitive hierarchy. A store services screen is linked to one of the start screen buttons and provides access to one or more service description screens.

A closed loop system is presented for selecting samples for labeling so that they can be used to generate classifiers. The sampling is done in phases. In each phase a subset of samples are chosen using information collected in previous phases and the classification model that has been generated up to that point. The total number of samples and the number of phases can be chosen by the user.

An automatic, or “one touch,” filtering system for filtering a cooking medium in a cooking apparatus, e.g., a fryer, includes a cooking vessel, e.g., a frypot; a drain valve disposed on a drain path between the cooking vessel and a filter container; a fill valve disposed on a fill path between the filter container and the cooking vessel; a pump disposed on one of the first path and the second path for conveying the cooking medium from the drain path to the fill path through the filter container; a controller configured to selectively operate the drain valve and the fill valve to regulate a flow of cooking medium to and from the cooking vessel; and a single switch configured to send a control signal to the controller in response to actuating the switch. The systems and methods of the invention provide automatic filtering that may reduce or eliminate incorrect valve operation, so that the filtering process may be performed correctly each time.

The invention provides a speech recognition method based on a convolution neural network, which is more good at extracting high-level features, has simple modeling process, is easy to train, has better generalization performance of the model, and can be more widely applied to various speech recognition scenes. The method comprises the following steps: S1, preprocessing the input original speech signal; S2, extracting the key feature parameters reflecting the characteristics of the speech signal to form a feature vector sequence;S 3, base on that DCNN network model, taking the connected time classifier CTC as a loss function, constructing an acoustic model of an end-to-end mode; S4, training the acoustic model to obtain the trained acoustic model; S5, inputting the feature vector sequence to be recognized obtained in the step S2 into the trained acoustic model to obtain a recognition result; and S6, performing a subsequent operation on the basis of the recognition result obtained in step S5, that is, obtaining a word string capable of outputting the speech signal with a maximum probability, that is, a language character after the original speech is recognized.

The invention provides a modular manufacture training unit device and a modular combined manufacture training system. A training modular unit consists of a processing and assembling modular unit, a straight walking manipulator modular unit, a fixed manipulator modular unit, a stereoscopic warehouse modular unit, a cargo loading table modular unit, a conveying belt detection modular unit, an inclined slideway modular unit, a horizontal slideway modular unit, a well-type feeding modular unit, a motor modular unit and a multi-piece processing / assembling modular units, wherein the modular training units are respectively and independently assembled to operate for training and are arranged on a combined installation training platform and respectively combined to form the modular combined manufacture training systems; and the modular combined manufacture training systems are respectively and independently assembled to operate for training and are mutually combined to form a manufacture training system. The device and system provided by the invention are scientific in design, reasonable in structure, low in construction cost, small in occupied area and multi-functional in training; each modular unit can operate independently; and the modular units can be combined to constitute new training systems, thereby realizing the processing, assembly, carrying, detection, classification, recognition, transport, warehousing, management and other processes of various cargoes and performing monitoring and managing and the like on each operation link.

The invention discloses a traveling vehicle vision detection method combining laser point cloud data, belongs to the field of unmanned driving, and solves the problems in vehicle detection with a laser radar as a core in the prior art. The method comprises the following steps: firstly, completing combined calibration of a laser radar and a camera, and then performing time alignment; calculating anoptical flow grey-scale map between two adjacent frames in the calibrated video data, and performing motion segmentation based on the optical flow grey-scale map to obtain a motion region, namely a candidate region; searching point cloud data corresponding to the vehicle in a conical space corresponding to the candidate area based on the point cloud data after time alignment corresponding to eachframe of image to obtain a three-dimensional bounding box of the moving object; based on the candidate region, extracting a direction gradient histogram feature from each frame of image; extracting features of the point cloud data in the three-dimensional bounding box; and based on a genetic algorithm, carrying out feature level fusion on the obtained features, and classifying the motion areas after fusion to obtain a final driving vehicle detection result. The method is used for visual inspection of the driving vehicle.

The invention relates to the technical field of surface defect detecting of cold-rolled strip steel, and provides an online detecting method for surface defects of the cold-rolled strip steel. The method is characterized by comprising the following steps of: S1, irradiating the cold-rolled strip steel by using light and receiving an image reflected by the cold-rolled strip steel; S2, preprocessingthe image; S3, constructing a deep learning network for the surface defect detecting of the cold-rolled strip steel, extracting defect features of the defect image by utilizing the deep learning network for the surface defect detecting of the cold-rolled strip steel and identifying defect types of the defect features; and S4, determining the quality grade of the cold-rolled strip steel accordingto the defect types, the severity degree, the quantity of the defect features and the distribution situation of the detects. The online detecting method for the surface defects of the cold-rolled strip steel, which is provided by the invention, is capable of quickly identifying the essential defect types of the surface defects of the cold-rolled strip steel and is good in adaptability.

An interactive response system directs input to a software-based router, which is able to intelligently respond to the input by drawing on a combination of human agents, advanced speech recognition and expert systems. The system utilizes human “intent analysts” for purposes of interpreting customer utterance. Automated speech recognition subsystems are trained by coupling utterances with IA-selected meanings for those utterances, using non-real time systems to develop the training information for the automated speech recognition subsystems.

A neural network 80 for aligning a source word in a source sentence to a word or words in a target sentence parallel to the source sentence, includes: an input layer 90 to receive an input vector 82. The input vector includes an m-word source context 50 of the source word, n−1 target history words 52, and a current target word 98 in the target sentence. The neural network 80 further includes: a hidden layer 92 and an output layer 94 for calculating and outputting a probability as an output 96 of the current target word 98 being a translation of the source word.

A wind energy installation control device includes a wind rotor, a generator driven by the wind rotor, a torque control unit configured to control a torque of the generator, and a control system. The control system includes a detector configured to identify a grid dip and an end of the grid dip, a residual torque transmitter configured to provide a preset value for a torque of the generator after identification of the grid dip, and an initializer configured to initialize a component of the torque control unit at the preset value. Accordingly, upon return of grid power after a grid dip, the vibration behaviour of a wind power system can be significantly improved. Overload of a drive train upon return of grid voltage can thus be reduced.

The invention discloses a transverse federation and longitudinal federation combination method and device, equipment and a medium. The transverse federation and longitudinal federation combination method comprises the steps of obtaining available public information, inputting the available public information into a preset longitudinal federation service party; obtaining vector information, and training a longitudinal federation model of the preset longitudinal federation service party based on the vector information, updating the network weight of each preset reinforcement learning model, periodically inputting each updated preset reinforcement learning model into a preset transverse federation server, and iteratively updating each updated preset reinforcement learning model. The technicalproblem of high resource consumption of a computing system of a reinforcement learning model in the prior art is solved.

The invention discloses an industrial equipment control system based on multi-terminal personal hotspot control. The industrial equipment control system comprises a controller mounted on CNC (computer numerical control) machining equipment, when the controller is interconnected with mobile intelligent equipment, data of the CNC machining equipment can be sent to the mobile intelligent equipment, and a control command sent by the mobile intelligent equipment can be received; the mobile intelligent equipment is connected with the controller wirelessly, performs two-way data transmission with a cloud server through a mobile data network, performs control operation on the CNC machining equipment according to a user graphic interface provided by an App, and displays the machining state in real time; the cloud server supplies data service to the CNC machining equipment through the internet. A wireless network is opened through App software to be connected with the equipment, the cloud server is connected through the mobile data network, and version detection and upgrading are performed on software, firmware, equipment parameter configuration information and the like; the command is sent to the machining equipment for controlling the machining equipment, and meanwhile, machining information such as videos, images, characters and the like of the machining equipment can be fed back and displayed.

The invention discloses a multi-shaft mechanical arm visual control system and a multi-shaft mechanical arm visual control method based on somatosensation and touch. The system comprises a handheld control terminal, a controller and a multi-shaft mechanical arm in sequential connection, wherein the handheld control terminal and the controller are communicated through a network, and the handheld control terminal comprises a space posture measuring module, a touch screen, a processing module and a voice input module. The multi-shaft mechanical arm visual control system and the multi-shaft mechanical arm visual control method based on somatosensation and touch have the advantages that good man-machine interaction characteristics can be realized, so the process for operating the multi-shaft mechanical arm can be simplified, and the flexibility and the humanization of the multi-shaft mechanical arm control system are further improved; operators controls the multi-shaft mechanical arm through visual modes such as somatosensation, touch and voice, the controlling and teaching programming can be completely only through using few press keys, users can be familiar with and start to use the system through simple and short-time learning, in addition, the stored work tasks can be shared among different multi-shaft mechanical arms through networks, and the training and programming cost is reduced.

The invention provides an MIMO system channel estimation method and device in a 60GHz indoor scene, and relates to the field of communications. The method comprises the following steps: obtaining a first channel estimation value of each user equipment; carrying out two-dimensional discrete Fourier transform on the first channel estimation value to obtain a spatial pattern point set; grouping all user equipment according to the spatial pattern point set, using the same pilot frequency on the same group of users, and obtaining a second channel estimation value of each user equipment in each group to realize the MIMO system channel estimation in the 60GHz indoor scene. According to the MIMO system channel estimation method in the 60GHz indoor scene provided by the invention, the discrete Fourier transform is used as a spatial basis expansion model of orthogonal basis to parameterize a user equipment channel, thereby reducing the dimensionality of an unknown channel parameter, reducing training and feedback costs and easing pilot frequency pollution.