203results about How to "Good accuracy" patented technology

A paraphrase model of a question text inputted by a user is learned, and a paraphrase expression is generated in real time. When information in text set storage unit is updated, text pair extracting unit extracts a paraphrase text pair from the text set storage unit and stores it in text pair storage unit. Model learning unit learns a question text paraphrase model from the paraphrase text pair in text pair storage unit, and stores it in model storage unit. Text pair extracting unit extracts a paraphrase text pair again from the text set storage unit by using the question text paraphrase model which the model storage unit possesses, and stores it in the text pair storage unit. In case where the stored paraphrase text pair is the same as the paraphrase text pair stored in the text pair storage unit, learning of the question text paraphrase model is ended. Candidate creating unit reads the question text paraphrase model from the model storage unit and generates a paraphrase candidate of the inputted question text.

Disclosed is a concept for computer-assisted procedures of surgery and diagnosis that target rigid, non-deformable anatomical parts such as bone, tissue, or teeth. The disclosure describes attaching small visual markers to instruments and anatomy of interest (e.g. bone surface), with each marker having a printed known pattern for detection and unique identification in images acquired by a free-moving camera, and a geometry that enables estimating its rotation and translation with respect to the camera using solely image processing techniques.

An instant messaging interaction system and method work by: analyzing communicative information sent by a remote user to create emotional messages and analyzing information about the remote user's identity; storage in a storage module behavior weight value preset and corresponding to the information about the remote user's identity; determining, by a learning module, interactive responses according to the emotional messages and the behavior weight values; outputting, by an output module, the interactive responses; detecting if receiving a feedback signal from an local user; if the feedback signal is not received, the learning module stores the behavior weight value in the storage module; if the feedback signal is received, the feedback module generates a modification value corresponding to different levels of the feedback signal; generating, by the learning module and according to a detection result, modification values for modifying the behavior weight values. As a result, the messaging interaction system is capable of presenting rated interactive responses in a personalized, artificial intelligence-based manner that meets the local user's expectation.

Light from a light source is made to be reflected by a light source mirror and to pass through a variable collimator, an irradiation field shape shaped by the variable collimator is projected on a photographing screen by the light passed through the variable collimator, a projection portion of the photographing screen is photographed by a photographic device, and an image photographed by the photographic device is analyzed by an image processor.

In one aspect of the invention, roughly stated, Applicants have discovered that a compensation material within slot elongated in a direction parallel to a segment of waveguide in an arrayed waveguide grating apparatus can compensate for both first and second order change in refractive index of the base waveguide material over temperature. Unlike the transverse slots of conventional linear athermalization techniques, the elongated slot generally parallel to the base material defines a composite waveguide section having a second order effective index of refraction temperature dependency which can be utilized to accurately minimize the temperature dependence of the overall optical path length to both the first and second order. The techniques described herein are also generalizeable to neutralization of the optical path length temperature dependence to any order.

By repeatedly executing a predetermined measurement at set intervals, data on a predetermined performance (a best focus position) of a predetermined apparatus and data on variation factors of the performance are obtained (Steps 204 to 214). Based on the obtained data, multivariate analysis is performed and a model equation that is used to predict a variation amount of the performance and includes at least one of the variation factors as a variable is derived (Step 214). Therefore, after deriving the model equation, a variation amount of the performance can be predicted using the model equation by obtaining data on the variation factor that serves as the variable (Step 238). Accordingly, it becomes possible to maintain the performance described above with good accuracy in accordance with the prediction results and also optimize the implementation timing of maintenance and the like.

The invention relates to a field programmable gate array (FPGA)-based joint test action group (JTAG) test method. In the method, an FPGA is connected with a circuit board to be tested by a JTAG interface and a probe; when a test is performed through the JTAG interface, the probe led out of the FPGA is connected with a monitoring point of the circuit board to be tested so as to read the data change of each point during the test; and finally, the FPGA sends the data change to a personal computer (PC) or a microprocessor for data analysis and judgment through an asynchronous series-port circuit. By the FPGA-based JTAG test method, a welding defect of a ball grid array (BGA) chip and the welding condition of the circuit around the chip can be precisely judged.

The invention provides a pretreatment pipeline system for online monitoring of water quality and quantity, which is used for pretreatment before the sampling analysis of a water sample and is provided with an air back-blow unit used for providing air pressure for cleaning, a water-taking unit used for pumping source water, an automatic cleaning unit used for cleaning algae and impurities of pipelines indoor and outdoor, a pretreatment unit used for pretreating before the sampling of source water and a water distribution unit used for taking water and distributing samples for sampling analysis, wherein the water distribution unit comprises a filtering unit; and the air back-blow unit, the water-taking unit, the automatic cleaning unit, the pretreatment unit and the water distribution unit are respectively connected with each other. The pretreatment pipeline system has reasonable structure and tight system, realizes the technical effect of online monitoring by full automation and all weather, and can ensure the representation and accuracy of water sample analysis.

The invention discloses a device and a method for predicating a fan load index of a wind powder plant, which belong to the technical field of wind power generation. The method for predicating the fan load index of the wind powder plant adopts a control unit and a mathematical model, wherein the control unit comprises a sensor, a data acquisition chip, a central processing unit, an industrial personal computer and a wireless communication module. The predication comprises the following steps of: 1) establishing input amount; 2) forming a training sample set; 3) inputting an input sample data set into a least-square-method support vector machine model; 4) resolving the least-square-method support vector machine model; and 5) inputting the sample data set to obtain a load index of the wind power plant. The device and the method have the advantages that by realizing the fan state monitoring in a predication process and establishing the least-square-method support vector machine model, the prediction on the fan load index has high precision, high accuracy and high efficiency.

A vehicle controller for controlling the air-fuel ratio of an engine is provided. In one embodiment, the controller comprises a first exhaust gas sensor provided downstream of the catalyst for detecting oxygen concentration of exhaust gas, a first decimation filter connected to the first exhaust gas sensor, and a control unit connected to the first decimation filter. The control unit determines a manipulated variable for manipulating the air-fuel ratio. The first decimation filter oversamples, low-pass filters and then downsamples the output of the first exhaust gas sensor. The first decimation filter can remove chemical noise from the output of the exhaust gas sensor. In another embodiment, a second decimation filter is connected to a second exhaust gas sensor provided upstream of the catalyst for detecting the air-fuel ratio of the exhaust gas. The second decimation filter oversamples, low-pass filters and then downsamples the output of the second exhaust gas sensor. The second decimation filter can compensate the shortage of resolution of the air-fuel ratio sensor.

A complex device includes: a substrate having a thick portion, a cavity and a membrane for bridging the cavity; and multiple piezoelectric elements having a lower electrode, a piezoelectric film and an upper electrode. A part of the piezoelectric elements has a projecting portion arranged on the upper electrode. The part of piezoelectric elements (30) provides a vertical pressure detection element. The piezoelectric elements further have an ultrasonic element other than the vertical pressure detection element. The ultrasonic element is arranged over at least the cavity of the substrate in a horizontal direction.

The invention designs a machine translating device based on intelligent matching and a method of the machine translating device. The method comprises the following steps of: A step, carrying out a linguistic data pre-processing step: carrying out word frequency counting and marking on sub-domain linguistic data; B step, carrying out a sentence database establishing step: establishing the sentence database by using sub-domain English-to-Chinese linguistic data; C step, carrying out a word database establishing step: establishing a word database according to the linguistic data pre-processing step and the sentence database establishing step; and D step, carrying out a machine translating step: firstly, manually making a phrase framework consisting of Chinese phrase nodes and a translation rule of the phrase framework according to the sentence database establishing step and the word database establishing step, and identifying a rule of a Chinese sentence sub point; automatically generating a sentence framework template base by using the sentence database, grammar identification and the phrase framework; and finding a framework matching sentence.

Provided is a velocity calculating apparatus including: a horizontal-angular-velocity-detecting portion disposed to a main-body portion attached to a movable object moving along a movement plane to detect a horizontal-axis angular velocity around a horizontal axis perpendicular to a moving direction generated according to an inclination angle of the movement plane; a vertical-acceleration-detecting portion disposed to the main-body portion to detect a vertical acceleration generated according to a shape of the movement plane; a time-difference-calculating portion configured to calculate a time difference between characteristic components occurring in the horizontal-axis angular velocity and the vertical acceleration due to a position of the movement plane; and a velocity-calculating portion configured to calculate a time-differential velocity based on a ratio between an installation distance from a position to which the horizontal angular velocity is exerted in the overall movable object to an installation position of the main body portion and the time difference.

The invention discloses a display control method and device, a computer device and a computer readable storage medium. The display control method is applied to a terminal provided with a first touch display screen and a second touch display screen and comprises the steps that awakening operation information of a user for the terminal is received; whether the first touch display screen is controlled for display or the second touch display screen is controlled for display is judged according to the awakening operation information, and a judgment result is obtained; when the judgment result represents that the first touch display screen is controlled for display, the first touch display screen is controlled for display; or, when the judgment result represents that the second touch display screen is controlled for display, the second touch display screen is controlled for display. Through the method, the touch display screen used by the user can be controlled for display while the terminal is awakened, and the intelligence and accuracy of operation of a double-screen mobile terminal are improved.

A general-purpose system for enabling the user to easily enter characters whose contensts are hard to understand at a glance, such as bar codes, by using a general-purpose application program such as a wordprocessor is provided. For this purpose, a printer driver comprising a font file storing a display font and a bar code record font corresponding to the display font and a conversion function of converting display font codes into record font codes is provided. When a bar code is entered in an application program, a legible display font corresonding to the bar code can be used. On the other hand, when the bar code is printed, the priner driver compatible with a given printer converts the display font into a bar code record font and supplies the bar code record font. Thus, a high-accuracy bar code is printed in accordance with the standard.

A method for determination of the inclination angle and the anteversion angle of an acetabulum of a patient in an anatomical reference system whose orientation is determined by the pelvic bone of the patient, taking account of any pelvis tilt, the method comprising the following steps: a first image of the pelvic bone is provided, wherein the first image shows a frontal plane including the pelvic bone; a first inclination angle of the acetabulum is determined using the first image; a pelvis rotation angle is determined, which indicates the rotation angle of the pelvic bone about the normal to the frontal plane relative to an optimum patient location position; a second image of the pelvic bone is provided, wherein the plane of the second image is rotated by an image rotation angle relative to the frontal plane about a common section line of the two planes, preferably about the body longitudinal axis of the patient; a second inclination angle of the acetabulum is determined using the second image; a pelvis tilt angle is determined, which represents a rotation of the frontal plane of the first image relative to an anterior-pelvic plane in the optimum patient location position; a linear equation system is set up having at least two linear equations as a function of the previously determined angles; the normal vector no is expressed in polar coordinates of the anatomical reference system; the at least two linear equations are solved for the azimuth and polar angle of the normal vector no in order to determine the inclination and the anteversion in this way.

The invention discloses an application of methyl salicylate. The compound can be used for preparing a bursaphelenchus xylophilus attractant. The results of trapping effect experiments against bursaphelenchus xylophilus in a forest indicate that the compound methyl salicylate has a strong attracting effect against bursaphelenchus xylophilus; and with an appropriate price, the product is nontoxic to human bodies, livestock, trees and crops, has good social and ecological benefits and is suitable for popularization and application.

An automatic money transaction machine, which is small in size to be able to detect a state of a tear in a bill with good accuracy and gives good service and no discomfort to a customer, comprises a receipt / payment opening 4, a discriminating unit 6, a temporary accumulation section 8, and a bill discriminating device 20 for discriminating a state of a bill. The bill discriminating device 20 comprises an energizing drive device provided on a conveyance path between the receipt / payment opening 4 and the temporary accumulation section 8 for application of tensile or compressive forces in a direction horizontally perpendicular to a direction of conveyance in a plane of a bill, and a bill image acquiring device, such as image sensors, imagers or the like, provided close to the energizing drive device for acquiring an image of a bill, to which tensile or compressive forces are applied by the energizing drive device. A state of a bill is distinguished on the basis of bill image data acquired by the bill image acquiring device, and gates 7A, 7B are switched over in accordance with the distinguishing result and the result of the discrimination from the discriminating unit 6.

A temperature-measuring substrate (50) for use in a heat treatment apparatus (2) for performing heat treatment of a substrate W to be processed, includes: a substrate body (62); an oscillator (64) including a piezoelectric element (68) and provided in the substrate body; and an antenna portion (66) connected to the oscillator and provided beside or around the periphery of the substrate body. The temperature-measuring substrate can reduce the attenuation of radio waves emitted by the oscillator.

By inserting a light source bulb into a reflector from a side thereof at a position apart downwardly from an optical axis, it is made possible to use an area situated on a side of the optical axis effectively for controlling the light distribution of a vehicle headlamp. In addition, by providing an additional reflector between the light source bulb and a shade for reflecting light from a light source laterally outwardly and diagonally forward of a vehicle through a gap between the reflector and a projection lens, a diffused area of a low beam light-distribution pattern is illuminated brightly and the diffused area is expanded laterally outwardly of the vehicle.

Employed is an object information acquiring apparatus including an irradiation unit irradiating an object with light; a detection unit including a plurality of elements each obtaining an acoustic wave generated from the object and converting the acoustic wave into a detection signal; a detection unit moving mechanism moving the detection unit to the object relatively; an adding unit selecting, from detection signals converted from acoustic waves obtained by the plurality of elements at respective positions as the detection unit is moved, detection signals converted from acoustic waves obtained from an identical position on the object, and adding together these detection signals to output a resultant summed signal; an initial sound pressure computing unit computing an initial sound pressure of a region of interest from the summed signal; a light quantity computing unit computing a quantity of light of the region of interest.

The invention discloses an industrial production process target data prediction method of a multi-feature fusion deep neural network. Collecting other variable time sequence data related to the key variable in the industrial equipment through equal-interval sampling by utilizing a sensor, and performing predictive analysis on the time sequence data of the key variable in the process industry; inputting into a pre-designed and pre-constructed deep convolutional neural network for training; segmenting historical data of the key variables according to time steps and then inputting the segmented historical data into a deep gated recurrent neural network for learning; a multi-feature fusion method is utilized to fuse output features obtained by two networks and then input the output features into a full connection layer, network parameters are optimized through back propagation, and prediction precision is improved. According to the method, reliable and effective target variable parameter prediction is provided for process monitoring in industrial production, and the hysteresis quality of measurement of key variables such as the silicon content of molten iron in industrial production isrelieved.

A drive system determines a combination of basic patterns (type of basic patterns to be generated and the number of pulses of each basic pattern) based on design data of patterns and pattern combination information stored in a memory. Then, based on the determined result, each micro mirror of a variable molding mask is individually controlled such that a plurality of basic patterns are sequentially generated according to each of the number of pulses, and each basic pattern generated by the variable molding mask is sequentially image-formed on a plate via a projection optical system. Thus, a pattern with a desired line width corresponding to the design data is formed at a desired position on the object with a good accuracy.

A control method for a PWM cyclo-converter is provided in which a voltage can be accurately generated even when a voltage command is small.In the PWM cyclo-converter, the turning on and off operations of two-way semiconductor switch are repeated at intervals of short time. As switching patterns, within the intervals of short time, a first terminal of output side terminals outputs in order a maximum potential phase P, an intermediate potential phase M and the maximum potential phase N, a second terminal of the output side terminals outputs in order the maximum potential phase P, the intermediate potential phase M, a minimum potential phase N, the intermediate potential phase M and the maximum potential phase P, and a third terminal of the output side terminals outputs in order the intermediate potential phase M, the minimum potential phase N and the intermediate potential phase M.

The invention provides a feces analyzing device which comprises a turntable assembly. The turntable assembly comprises a turntable. A plurality of detection card case placement parts are uniformly distributed on the turntable, a detection card case is placed on each of the detection card case placement part, and detection cards are successively placed in the case bodies of the detection card cases from top to bottom; a card pulling assembly is arranged on one side of the turntable assembly, the detection cards are pulled out through the card pulling assembly and penetrate through a guide slot of a guide assembly to arrive at a sample application position, the sample application position is arranged on the front portion of a card conveying tray of a card conveying assembly, a specimen is subjected to sample application at the sample application position of each detection card, each detection card can be transferred to a reaction rack of a reaction rack assembly through the card conveying assembly and a rack feeding assembly to react, the reaction rack has a plurality of detection card placement parts, the card conveying assembly is located on one side of the reaction rack assembly, a photographing assembly is arranged on the other side of the reaction rack assembly, and each detection card can be pushed into a photographing tray of the photographing assembly through a rack feeding arm of the rack feeding assembly.

A board inspection device includes an irradiation device for irradiating light on a printed circuit board, a CCD camera for imaging the irradiated part of the circuit board. First image processing is performed for a first exposure time such that an inspection target region is free of brightness saturation, and second image processing is performed using a second exposure time corresponding to the insufficiency of the first exposure time relative to a certain exposure time appropriate for measurement of a measurement standard region. Thereafter, image data for three-dimensional measurement is prepared for the inspection target region using the value of image data obtained by the first image processing, and image data for three-dimensional measurement is prepared for the measurement standard region using a value obtained by summing the image data value acquired by the second image processing and the image data value acquired by the first image processing.

A fluorescence spectrophotometry to test optical self-cleaning materials and a method of photochemical catalysis for activity for file materials with nano TiO2 coated surface. Its steps are: (1) Intercept 10-50cm2 optical self-cleaning materials into reactor. (2) Add organic compound solution and take out 2mLto test fluorescence intensity of solution without illumination. (3) Open violet lamp, take out 2mL every 10min to test fluorescence intensity like above for 4~6 times. (4) Graph based on fluorescence intensity of solution and illumination time is a line, whose absolute value of slope K great indicates strong photochemical catalysis and oxidation of tested optical self-cleaning materials. The invention has merits of high sensitivity, good accuracy, shortcut and simple, which can be used to test photochemical catalysis capability of optical self-cleaning materials.

The present invention provides a malicious webpage discovery method based on feature detection. The method comprises the steps of: reading a URL file, and extracting URL related network behavior features; reading a DNS file to extract domain name related network behavior features; reading an NetFlow file to extract flow related network behavior features; and for the URL related network behavior features, the domain name related network behavior features and the flow related network behavior features, performing rule matching, and identifying malicious URLs according to a matching result. Basedon a network flow captured in real time, a system having detection and identification functions online to achieve the method mentioned above is constructed. The system implements online webpage identification to identify the URLs of the malicious webpage in real time from the real-time network flow.

An illuminator emits light obliquely to an end of an adhesive film. A CCD camera is provided in a direction perpendicular to the adhesive film to capture an image of the light reflected irregularly from the end of the adhesive film. A black plate is provided on a back side of the adhesive film. In the image obtained by the CCD camera, therefore, only the end of the adhesive film is displayed whitely in a line shape. A position of the adhesive film is determined based on a line corresponding to the end of the adhesive film. Since the light reflected from the end of the adhesive film is utilized, the position of the adhesive film can be determined accurately irrespective of a color of the adhesive film, a color of a base film to which the adhesive film is joined, and a depth of each color.