40results about How to "Improve accuracy and performance" patented technology

A voice activity detection apparatus (1) comprising: a signal condition analyzing unit (3) which analyses at least one signal parameter of an input signal to detect a signal condition SC of said input signal; at least two voice activity detection units (4-i) comprising different voice detection characteristics, wherein each voice activity detection unit (4-i) performs separately a voice activity detection of said input signal to provide a voice activity detection decision VADD; and a decision combination unit (5) which combines the voice activity detection decisions VADDs provided by said voice activity detection units (4-i) depending on the detected signal condition SC to provide a combined voice activity detection decision cVADD.

The present invention provides a projectile device and a method of manufacture of a projectile device and in particular, to a pistol bullet and a rifle bullet and method of manufacture of same. In one embodiment, the projectile apparatus has a cylindrical body portion having a diameter, a front nose section tapering from a most proximal point of the projectile to the cylindrical body portion, and a rear tail section connected to the body portion and extending to the most distal point of the projectile, in which the front nose portion comprises a plurality of twisting depressions forming troughs.

A thin film device and a method of providing thermal assistance therein is disclosed. Accordingly, a heater material is utilized to thermally assist in the operation of the thin film device. By utilizing a heater material to thermally assist in the operation of the thin film device, a substantial improvement in the accuracy and performance of the thin film device is achieved. A first aspect of the present invention is a thin film device. The thin film device includes at least one patterned thin film layer, a heater material coupled to the at least one patterned thin film layer for providing thermal assistance to the at least one of the patterned thin film layers and a conductor coupled to the heater material for supplying energy to the heater material.

A method of inspecting full-chip mask data to locate layout pattern design induced defects and weak points that cause functional failure or performance degradation for integrated circuits (ICs) manufactured in subwavelength technology. Given the pre-OPC integrated circuit design layout data, the method of present invention refers to available post-OPC data or generates post-OPC data condition to do the inspection based on the modeling of integrated circuit wafer patterning. Build-in direct checks of specified electrical functional defects and a multilayer pattern-centric approach are used to improve the accuracy and performance. A technique of adaptive search is used to speed up the critical dimension search during the process of optical proximity correction data verification. A defect synthesis capability is supplied for defect disposition to facilitate systematic correction and prevention of the defects in integrated circuit layout design.

The invention includes a gas valve having an annular body with an inner surface defining a gas chamber and first and second annular end surfaces defining first and second openings of said gas chamber, the gas valve further having outer surface and at least one gas channel extending between the inner surface and the outer surface providing a gas communication path from the outer surface to the gas chamber, wherein said at least one gas channel is orientated to direct fluid egressing from the channel into the chamber along the inner surface. The invention further includes a regulator occupying a portion of the chamber to define a chamber operating volume and flowrate, the regulator having a at least one outer diameter corresponding to an inner diameter of the passage to substantially inhibit gas flow from passing through the second opening.

The invention provides a detection method and system for WEB application system content change. The detection method and system stores grasped website webpage content into a detection page storage repository with a page file as a unit and records the frequency that the website is detected; by utilizing the stored detection frequency and the existence of the grasped page file, whether the website is a new website and whether the webpage is a new page are judged; whether storage elements contained inside the page file are the same as storage elements contained in a webpage file stored in the past is judged, and if the detection result is that change occurs, the detection result is output. The detection method and system for WEB application system content change carries out all-around detection on the WEB application system content, and accurately positions the specific reason why the webpage content changes on the aspects of webpage feature library codes, webpage content elements and webpage dynamic or static content, thereby obtaining the accuracy of webpage detection and bringing convenience to the detection of WEB application system content change.

The invention provides a static correction method and static correction equipment for a converted wave. The method comprises the following steps: acquiring three-component earthquake data; performing preprocessing on the three-component earthquake data and obtaining preprocessed converted wave data; obtaining field shot static collection quantity of a longitudinal wave; according to the preprocessed converted wave data and the field shot static collection quantity of the longitudinal wave, obtaining common receiver gather of the converted wave subjected to dynamic correction; according to the common receiver gather of the converted wave subjected to dynamic correction, obtaining converted wave data after residual shot static correction; according to the converted wave data after residual shot static correction, obtaining converted wave data after static correction of shot points and receiver points. Problems of strict application conditions and poor calculation precision due to relatively complicated near surface of a converted wave in the prior art are solved, a situation that the time for static correction influence on the converted wave receiver points is greater than 1/2 of cycle is adapted to a certain extent, and the static correction precision and efficiency of the converted wave are effectively improved.

Disclosed is an apparatus for speech recognition and automatic translation operated in a PC or a mobile device. The apparatus for speech recognition according to the present invention includes a display unit that displays a screen for selecting a domain as a unit for a speech recognition region previously sorted for speech recognition to a user; a user input unit that receives a selection of a domain from the user; and a communication unit that transmits the user selection information for the domain. According to the present invention, the apparatus for speech recognition using an intuitive and simple user interface is provided to a user to enable the user to easily select/correct a designation domain of a speech recognition system and improve accuracy and performance of speech recognition and automatic translation by the designated system for speech recognition.

There is provided a computer implemented data driven classification and troubleshooting system and method. The system has an interface application enabled to receive data. The system has an associative memory software in communication with the interface application via an API. The associative memory software has an associative memory and a machine learning algorithm. The system has one or more individual areas, within the associative memory, requiring one or more troubleshooting actions to improve accuracy of the individual areas. The system has at least one troubleshooting tool enabled by the interface application. The at least one troubleshooting tool enables or performs the troubleshooting actions. The system has a quality rating metric (QRM) that measures a strength and an assurance that one or more predictions of the associative memory are correct. The one or more troubleshooting actions results in improving the accuracy and the performance of the associative memory.

The embodiments described herein provide devices and methods that facilitate improved input device performance. Specifically, the devices and methods provide improved resistance to the effect of interference on input devices, and in particular, to the effect of unison noise on proximity sensors that use capacitive techniques to generate images of sensor values. The devices and methods provide improved resistance to the effects of interface by using multiple images of sensor values, where the multiple images of sensor values are captured at different times along two different axes. The multiple images of sensor values are combined to produce a modified image of sensor values, the modified image having reduced errors due to noise. This reduction in errors due to noise can improve the accuracy and performance of the input device.

A differential delay cell is disclosed. The delay cell includes a voltage bus and a differential pair of MOS transistors having respective source terminals coupled to define a current node, and respective drain terminal outputs that cooperate to form a differential output. A current source is disposed at the current node while a differential diode-connected load is disposed between the differential pair and the voltage bus. The differential diode-connected load comprises at least one n-channel MOS transistor configured as a diode.

The present invention discloses a data classification method. The method comprises the steps of: S1, obtaining training set samples for training a classifier, and averaging the obtained training set samples according to the number of iterations required for training to obtain a plurality of training subset samples; S2, based on the Adaboost algorithm, training each of the training subset samples respectively by using a plurality of weak classifiers, when each weak classifier performs training, selecting some training subset samples and some error samples obtained by the previous weak classifier to constitute and form a final training sample, and obtaining a final ADB strong classifier from various weak classifiers after completing training; and S3, using the trained ADB strong classifier to classify to-be-classified data, and outputting a classification result. According to the method disclosed by the present invention, the data during classification training is complete, the trainingdata can be prevented from multiplying and over-fitting, and the method has the advantages of a simple implementation principle, high classification efficiency and precision, and the like.

A body temperature detection method is characterized by comprising the steps of firstly, calculating and calibrating the parameter values of a fitting function in real time according to preset parameters and the AD values, read by high-precision reference resistors, of two resistors, and obtaining the AD values of the two resistors; secondly, measuring the AD value of a resistor, and obtaining a calibrated AD value through fitting function calibration; thirdly, obtaining a calibrated measured resistance value according to the calibrated AD value; fourthly, obtaining a calibrated measured temperature value according to the resistance value. According to the method, the two high-precision resistors serve as the reference resistors, the fitting function is designed according to the characteristics of an AD sampling device, the parameter values of the fitting function are calibrated dynamically in real time, and the parameter values are calculated dynamically each time measurement is conducted, so that the calibrated temperature value is obtained and the precision and performance of the electronic thermometer device are improved.

The invention provides a fingerprint identification system comprising a plurality of first pixel circuits, a first sensing circuit, at least one second pixel circuit, a second sensing circuit, and a differential amplification circuit. The first sensing circuit is used for sensing a contact capacitor and outputting a first output signal, wherein the first output signal comprises a large signal component and a small signal component, and the small signal component is related to a variation of the contact capacitor. The second sensing circuit is used for outputting a second output signal, and the second output signal is equal to the large signal component. The differential amplification circuit is used for amplifying the difference between the first output signal and the second output signal in order to generate an amplified output signal, and the amplified output signal is related to the small signal component.

An adjustable gas valve having an annular body with an inner surface defining a gas chamber and first and second annular end surfaces defining first and second openings of said gas chamber, the gas valve further having outer surface and at least one gas channel extending between the inner surface and the outer surface providing a gas communication path from the outer surface to the gas chamber, wherein said at least one gas channel is orientated to direct fluid egressing from the channel into the chamber along the inner surface. The invention further includes a regulator occupying a portion of the chamber to define a chamber operating volume, the regulator having at least one outer diameter corresponding to an inner diameter of the passage to substantially inhibit gas flow from the pass through the second opening.

The invention provides a fishery culture monitoring system and method based on fish collection class acoustic monitoring and positioning. The system comprises a culture facility, an aquatic independent patrol inspection platform, a plurality of positioning arrays and a control center, wherein the culture facility is a cuboid; the plurality of positioning arrays are independently fixed on differentpositions of the culture facility; the control center is in communication connection with each positioning array; and the aquatic independent patrol inspection platform and the control center are incommunication connection and are arranged in the culture facility. By use of the fishery culture monitoring system and method based on the fish collection class acoustic monitoring and positioning, anunmanned vehicle is adopted to serve as an acoustic detection carrier, and the unmanned vehicle is positioned through a detection signal, the trajectory planning of the unmanned vehicle is finished through positioning information, an acoustic signal combines detection and positioning functions into a whole, so that the bearing capacity of the unmanned vehicle is greatly saved, the integration level and the stability of equipment are improved, and equipment cost is lowered.

The invention provides a method and a device for improving the figure recognition precision of an indoor monitoring system. The method comprises the following steps: setting a face feature library and an appearance feature library; extracting face features and appearance features of the figure target entering the monitoring image range, and recognizing the figure target; extracting face features and appearance features of the figure targets appearing at other positions in the monitoring image range, and recognizing the figure targets; when the number of people in the monitoring image range is large, extracting the face feature and the appearance feature of each person, and recognizing each person; and when the face features of the person in the monitoring image range cannot be extracted, performing detection and recognition by utilizing the appearance features of the person. The figure recognition precision and performance of the monitoring system can be remarkably improved, the operation efficiency is high, the calculation burden of the system is not increased while the performance of the system is improved, deployment, expansion and system upgrading are easy, and the method can be applied to intelligent video monitoring systems of office places, shopping malls and the like.

Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may transmit a sounding reference signal (SRS) that indicates that the UE is configured with a quantity of receive elements that is greater relative to a quantity of transmit elements with which the UE is configured. The UE may receive, based at least in part on transmitting the SRS, a first channel state information (CSI) reference signal (CSI-RS) set and a second CSI-RS set. The UE may transmit a CSI report that is based at least in part on the first CSI-RS set and the second CSI-RS set. Numerous other aspects are provided.

The invention discloses a construction method of a breast ultrasound image focus intelligent identification missed judgment prevention system. The construction method comprises the steps: S1, data preparation, S2, deep neural network model design, S3, missed diagnosis prevention data consistency design, S4, missed diagnosis prevention model training, and S5, missed diagnosis prevention model testing. According to the construction method of the breast ultrasound image focus intelligent identification missed judgment prevention system, the constructed system is additionally provided with the missed diagnosis prevention model module, the missed diagnosis prevention model module is designed by using a generative adversarial neural network, and the missed diagnosis rate of an original computer intelligent diagnosis system can be reduced, so that the effect and the precision of auxiliary diagnosis are improved; and the film viewing cost of doctors in hospitals is reduced and the efficiency is improved.