344results about How to "Improve accuracy and efficiency" patented technology

Methods and systems for large-scale airframe assembly are disclosed. In one embodiment, a method includes measuring a plurality of discrete point positions at least one of on and adjacent to at least one of a first and a second component, and measuring at least one surface position on the at least one of the first and second components. The measured positions are compared with a desired position information (e.g., a computer aided design model). The comparison may include applying a fitting routine to the measured positions and the desired position information. Next, a transformation matrix for improving the comparison between the measured positions and the desired position information is computed. At least one of the first and second components is then moved according to the transformation matrix. During movement, the plurality of discrete point positions may be monitored and provided to the position control system by a feedback loop.

An apparatus, method, and medium for dialogue speech recognition using topic domain detection are disclosed. An apparatus includes a forward search module performing a forward search in order to create a word lattice similar to a feature vector, which is extracted from an input voice signal, with reference to a global language model database, a pronunciation dictionary database and an acoustic model database, which have been previously established, a topic-domain-detection module detecting a topic domain by inferring a topic based on meanings of vocabularies contained in the word lattice using information of the word lattice created as a result of the forward search, and a backward-decoding module performing a backward decoding of the detected topic domain with reference to a specific topic domain language model database, which has been previously established, thereby outputting a speech recognition result for an input voice signal in text form. Accuracy and efficiency for a dialogue sentence are improved.

The invention discloses a hub form-position parameter detection method based on image recognition which comprises the following steps: a digital image of a hub with a calibration template is taken, and the distortion correction coefficient of the digital image is obtained; the digital image of the hub is orderly made grey conversion, image compression and image binarization processing; the binarized image is made an area partition based on the connected area lookup to obtain the effective area of each mounting hole of the hub; the effective area of each mounting hole of the hub is made an edge detection to extract the edge information of the mounting hole to locate the boundary of the mounting hole, meanwhile the boundary of the mounting hole is made a geometric distortion correction; random points on the boundary of the mounting hole is made a sub-pixel interpolatory subdivision to obtain the more accurate boundary of the mounting hole; the circle center and the diameter of each mounting hole are fit according to the random points of the boundary of the mounting hole, and a pitch circle parameter of the hub is fit according to the circle center position of each bolt hole. The method of the invention ensures that the efficiency and the accuracy of the hub detection process are greatly improved and also alleviates the labor intensity of the workers.

The invention is applicable to the field of assembly technologies for screws and nuts, and discloses an automatic screw and nut combination locking device which comprises a fixing assembly and a sliding assembly. The sliding assembly comprises a sliding plate, the sliding plate is slidably connected to the fixing assembly, and a sliding plate driving assembly is arranged on the fixing assembly; a screw locking assembly and a nut locking assembly are connected onto the sliding plate, the screw locking assembly comprises a screwdriver, a screw rotation driving assembly and an axial driving assembly, the screw rotation driving assembly is used for driving the screwdriver to peripherally rotate, and the axial driving assembly is used for driving the screwdriver to vertically slide; the nut locking assembly comprises a nut locking sleeve and a nut rotation driving assembly, the nut rotation driving assembly is used for driving the nut locking sleeve to rotate, and the screwdriver and the nut locking sleeve are coaxially arranged. The automatic screw and nut combination locking device has the advantages that a screw and a nut can be assembled at one station, so that the product assembly working hours can be greatly reduced, the assembly efficiency and the assembly precision can be improved, the cost can be reduced for an enterprise, and the core competitiveness of the enterprise can be improved.

The invention relates to a building engineering computation and evaluation system and method. The system comprises an associating module, a parameter reading module, an analysis module and a computation and evaluation module, wherein the associating module is used for identifying a building information model object from a building information document and associating a corresponding building system project in a master database with the building information model object; the parameter reading module is used for acquiring the structural parameters of the building information model object according to the structural item corresponding to the building system project and generating a parameter list file of the acquired structural parameters; the analysis module is used for analyzing the parameter list file and determining the definitions of the structural parameters according to the building system project; and the computation and evaluation module is used for estimating the engineering quantity according to the definitions of the structural parameters, evaluating the price of required building resources by querying a unit price catalogue database, and further generating an engineering quantity and evaluation list by combining the engineering quantity and the evaluated result. According to the invention, the efficiency and accuracy of engineering computation and evaluation are greatly increased.

The invention discloses a point-cloud-model-based precise detection method for a surface profile. The method comprises: a theoretical point cloud model of a to-be-measured workpiece is established; a practical point cloud model of the to-be-measured workpiece is measured; registering of the theoretical point cloud model and the practical point cloud model of the to-be-measured workpiece is completed; feature extraction is carried out on the theoretical point cloud model to obtain a normal feature; and a profile error of the theoretical point cloud model and the practical point cloud model is calculated. According to the invention, the method that is simple and practical and can be executed rapidly can be applied to profile detection of any model; and the universality is high. The industrial profile detection precision and efficiency are improved simultaneously; and the demand of high-precision profile error evaluation can be satisfied. In addition, the invention also discloses a point-cloud-model-based precise detection system for a surface profile.

The invention discloses a method for levelling metal plate with stress deformation by utilizing vacuum annealing. The metal plate required to be levelled is placed in a fixture comprising an upper cover plate and a lower cover plate, a sealed air pocket is arranged on the metal plate, the elastic thin wall of the air pocket is contacted with the metal plate, the air pocket is fastened onto the metal plate required to be levelled, and the metal plate is placed into a vacuum annealing furnace, vacuumizing, heating and annealing are carried out, and levelling on the metal plate is carried out after annealing is finished. The method of the invention changes the levelling method from local deformation into global deformation while the metal plate is in uniform stress after yield stress is reduced, thus finishing levelling in one step and improving efficiency and accuracy of levelling; and meanwhile the method of the invention combines 'levelling' and 'annealing' processes, thus shortening production period and reducing production cost.

The embodiment of the invention discloses one method and one device for regulating optical power. The method comprises the following steps of: determining a first power attenuation value of a power regulation unit in an optical network, which is related to new service wavelength, according to the path of the new service wavelength and the structure of the optical network; calculating the performance parameter of a channel corresponding to the new service wavelength according to the first power attenuation value; and judging whether the performance parameter of the channel corresponding to the new service wavelength meets the preset tolerance and flatness requirements or not, and regulating the power attenuation value of the power regulation unit related to the new service wavelength to reach the first power attenuation value if a judgment result indicates that the performance parameter meets the preset requirements. The invention also provides another method and another device for regulating the optical power. Through the technical scheme in the embodiment of the invention, network cost can be effectively reduced, and the efficiency and accuracy of regulating the optical power are improved.

The invention discloses a self-adapting sightline tracking system and a tracking method. A human eye image is collected to a computer by a charge coupled device (CCD) camera, and eye fixation point coordinates can be obtained through steps of image preprocessing, pupil periphery point extracting, ellipse fitting and the like. A theoretical model and an experimental device of the self-adapting sightline tracking system are set up, wherein the theoretical model and the experimental device are combined with software and hardware and are suitable for the condition that the head rotates, and high-precision sightline tracking under a non-interfering condition for a user is achieved. In order to overcome affection of movements of the head in sightline tracking technology, the self-adapting sightline tracking method based on multi-scale geometric analysis and faced with the condition of head moving is utilized. A multi-scale geometric analysis tool is introduced to the field of sightline tracking, a high-precision detecting algorithm of image characteristics of human eyes and faces based on Directionlet change is utilized so as to improve detecting efficiency and precision of human eye edges. A neural network and a genetic algorithm are combined so as to study the relation of the image characteristics of the human eyes and the faces and nonlinear mapping of eye fixation points and raise a new method of sightline direction detection.

The invention discloses an automatic sorting device for defective products of pole pieces of timers. As sorting devices in the prior art are easily affected by inspectors, environment elements and the like, the inspection is instable and the detection efficiency and precision are hard to ensure. According to the invention, a rectangular rack is used as a support, four supporting feet are arranged at the bottom of the rectangular rack, adjusting wheels are arranged on the supporting feet, a material plate is fixed on the rack, eight proximity switches are arranged on the material plate, a movable material clamping plate is arranged on the side surface of the material plate, a material clamping cylinder and a feeding cylinder are connected on the material clamping plate, and a first image acquisition system, a first sorting mechanism, a turn-over mechanism, a second image acquisition system, a second sorting mechanism, a third image acquisition system and a third sorting mechanism are arranged above the material plate from left to right. The automatic sorting device is high in detection precision and large in detection scope and greatly reduces the labor intensity of workers.

The invention discloses an application layer-based data segmenting method which is applied to depth detection equipment. The method comprises the following steps of: detecting whether the data flow of the user is a data flow with a dialogue by the depth detection equipment and if so, acquiring the segmental characteristics corresponding to the data flow; segmenting the data flow according to the received segmental characteristics by the depth detection equipment to acquire a depth detection unit; and detecting the depth of the depth detection unit by the depth detection equipment. The method can highly effectively solve the segmental problem of the depth detection data of the application layer equipment, match the depth detection unit with the resolving process of the majority of protocols, prevent the data packet of the protocols from being analyzed and be used for depth detection-based reorganization of an engine and flow detection-based establishment and destruction of the detected flow of the engine, thus significantly improving the precision and efficiency of the detection.

Provided is a fine method for global vegetation classification. The method comprises: firstly, obtaining spectroscopic data and remote sensing data of vegetation with a first main component; secondly, obtaining an area where land classification changes through the remarkable statistical test of average values; thirdly, performing minimum range monitoring classification of the area where land classification changes by taking an area where land classification doesn't change as a monitoring classification sample; fourthly, performing minimum range monitoring classification of a vegetation part of obtained data of new land classification data by utilizing spectroscopic data to obtain fine vegetation classification data; and finally, combining the obtained fine vegetation classification data with the non-vegetation data to obtain a fine vegetation classification result map. The method is applicable to departments of territorial resources, agriculture, forestry, and the like for fast updating a huge amount of land and vegetation classification data, and for fast accurately monitoring condition of vegetation growth and situation of change of land use.

The invention relates to an acoustic positioning system, in particular to an acoustic positioning system of a shallow sea wave detector for secondary positioning of a seabed wave detector in shallow sea petroleum earthquake exploitation. The system mainly comprises a carrier-borne part and a seabed part, wherein the carrier-borne part consists of processing software, a host machine and a receiving and transmitting combined transducer; and the seabed part consists of a plurality of seabed transponders. Because the technology for manufacturing the receiving and transmitting combined transducer is adopted, the designed source level and the sensitivity meet the positioning requirement of a complex exploitation area of the shallow sea; and the underwater working posture of the transducer is kept and the turbulence and reverberation are reduced by designing streamline turning vanes. Therefore, the system meets the requirements of high recovery ratio and high positioning efficiency of secondary positioning of a secondary positioning earthquake exploitation water detector in the Bohai Bay area.

The invention relates to a method for measuring an optical three-dimensional contour of space coding, which belongs to the field of photoelectric measurement. The method comprises the following steps: a strip in the prior binary coding is replaced by an idea for subdividing a binary coding strip by a color information strip; color information is easy to recognize, the contradiction between strip distance and strip recognition is solved, and a space point in a color strip distance in the previous step is measured in such a way of strip displacement multi-measurement, thereby being equivalent to further subdivide the measurement distance and preventing the problem of difficult strip recognition caused by denser projection light in such a way of displacement measurement; and finally the three-dimensional contour information of an object is obtained by integrating several measurement results. The invention can effectively solve the contradictions between strip coding and speed as well as strip distance and recognition in the three-dimensional contour measurement of structural light and ensure that the measuring speed reduction is prevented when the measuring precision is improved in the coding measurement of the structural light or the measuring speed can be improved under the condition of ensuring a certain precision.

The invention discloses a dual-binocular infrared and visible light fused stereo imaging system and relates to the field of binocular stereo imaging system. The system comprises a left field visible light camera and a right field visible light camera which are arranged symmetrically and align to an object to be measured to guarantee parallel optical axes and co-plane imaging planes. A left field dichroscope and a right field dichroscope are arranged symmetrically and are arranged in front of the left field visible light camera and the right field visible light camera respectively. A left field infrared camera and a right field infrared camera are arranged symmetrically and are arranged on the outsides of the left field visible light camera and the right field visible light camera respectively, and lenses face towards the left field dichroscope and the right field dichroscope respectively. By means of the system, by the design of a dual-binocular infrared and visible light image fusion technology, the defect of the single-wave-band stereo imaging system is compensated, and the problem of accuracy reduction of images obtained due to matching among systems is solved.

The disclosure relates to a near field communication NFC card calling method and device. The near field communication NFC card calling method comprises the following steps: obtaining location information of an electronic device in response to an NFC card calling request; determining an NFC card that matches the location information as a card to be called in the NFC cards of the electronic device;and calling the card to be called. According to the near field communication NFC card calling method and device provided by the disclosure, the operation of a user can be simplified, and the efficiency and accuracy of calling the NFC card can be improved.

A computer-implemented system and method are described for integrating a series of auxiliary spam detection models with a base spam detection model, thereby improving the accuracy and efficiency of the overall spam detection engine. For example, a system according to one embodiment of the invention comprises: a base spam filter feature extractor to detect a first set of features from incoming email messages; one or more auxiliary model feature extractors, each of the auxiliary model feature extractors to detect a different set of features from the incoming email messages; one or more auxiliary detection modules, each of the auxiliary detection modules to receive an indication of the different sets of features detected by a corresponding one of the auxiliary model feature extractor modules and to apply weights to the detected features; and a base spam filter module to receive an indication of the first set of features from the base spam filter feature extractor and the weights generated by the auxiliary detection modules, the base spam filter module to assign base spam filter weights to the first set of features and to determine whether an email message is spam based on the weights of the first set of features and the weights of the different set of features identified by the auxiliary model feature extractors.

The invention discloses a bathymetric surveying method without tide observation based on quasigeoid refinement. The method comprises the steps of firstly carrying out height fitting for WGS-84 geodetic height measured by a network RTK by virtue of a sectional secondary curve surface and a BP neural network model so as to convert the measured geodetic height to 85 normal height, subtracting the height from a GPS antenna to the water from the obtained 85 normal height to obtain the normal elevation of the water surface, and obtaining the water bottom 85 elevation data according to the water surface normal elevation and the water depth data measured by a bathymetric surveying energy transducer without observing the water level data of a tidal station. By adopting the method, the water depth surveying precision and efficiency can be obviously improved, and the manpower and the material resource can be saved.

The invention provides a welding device for titanium alloy sheets. Key pressing plates are arranged at the two ends of an assembling bottom plate respectively, the assembling bottom plate and the key pressing plates are combined to form a workpiece clamping mechanism, and a back cushion plate is detachably arranged on the assembling bottom plate. A U-shaped groove is formed in the center of the upper surface of the back cushion plate, and an air inlet pipe opening is formed in the U-shaped groove. Two blocking lumps are movably arranged in the U-shaped groove and used for blocking argon in the U-shaped groove in the welding process so as to protect the back face of a weld joint. A welding gun nozzle is connected with a welding gun through a separation sleeve. The front end of a tail protecting dragging cover is connected with the welding gun nozzle in a matched mode, and the welding gun is used for driving the tail protecting dragging cover to move together. Meanwhile, the invention provides a welding method of the welding device. The welding device is simple and reliable in structure, the welding method is good in stability, and the welding quality meets the batch production requirement.

A method and system parses natural language in a unique way, determining important words pertaining to a text corpus of a particular genre, such as tax preparation. Sentences extracted from instructions or forms pertaining to tax preparation, for example are parsed to determine word groups forming various parts of speech, and then are processed to exclude words on an exclusion list and word groups that don't meet predetermined criteria. From the resulting data, synonyms are replaced with a common functional operator and the resulting sentence text is analyzed against predetermined patterns to determine one or more functions to be used in a document preparation system.

The invention discloses an indoor scene positioning method based on a hybrid camera. The indoor scene positioning method includes taking a depth image and a color image of an indoor scene by the hybrid camera; tracking position of the camera; utilizing a standard greed forest algorithm to train the depth image and the color image of the indoor scene to build a regression forest; when positioning the indoor scene, utilizing the depth image and the color image of a current frame photographed by the hybrid camera and the regression forest which is trained well to calculate world coordinates corresponding to the current camera to complete positioning. In the indoor scene positioning method, the hybrid camera is adopted for photographing, a regression forest algorithm is adopted to train each pixel point in a known scene, and sparse feature points or dense feature points do not need to be used for scene positioning, so that time for detecting, describing and matching the feature points is saved, camera positioning is enabled to have nothing to do with state, accumulated errors in camera tracking are avoided, and accuracy and efficiency of scene positioning are improved effectively.

The invention discloses an online decision method for emergency control and correction control of an ultrahigh voltage DC receiving-end grid. The method is characterized in that coordinate control measures include direct current emergency boosting control, pumped-storage pump switching control and accurate load control for a DC blocking fault of causing large power shortage of the ultrahigh voltage DC receiving-end grid; and the online decision method comprises the steps of before the fault, considering stability constraints of relevant AC sections according to a DC operating state, calculating the emergency boosting capacity, as shown in the specification, of DC power in real time, monitoring the operating conditions of various pumped-storage units, recording the total capacity, as shownin the specification, of switchable pumped-storage units, collecting interruptible load data and calculating the total amount, as shown in the specification, of interruptible loads; and after the fault, coordinating direct current emergency boosting control, pumped-storage pump switching control and accurate load control from two aspects of frequency emergency control decision and section overloadcorrection control decision.

The invention is applicable to the field of mobile communication and provides a different-frequency measuring method and a different-frequency measuring device. The method comprises the steps of transmitting a preset parameter to a micro eNB, wherein the preset parameter is used by the micro eNB for transmitting a finding signal at the working frequency of a macro eNB; transmitting the preset parameter to UE (User Equipment) served by the macro eNB, wherein the preset parameter is used by the UE for detecting the finding signal transmitted by the micro eNB; receiving signal quality information reported by the UE of the detected finding signal; and configuring carrier frequency for executing different-frequency measurement related to the micro eNB for the UE according to the signal quality information. By configuring corresponding finding signal parameters for the micro eNB through the macro eNB and informing the served UE of the configured finding signal parameters to enable the served UE to perform detection on the basis of the corresponding finding signal parameters and then perform different-frequency measurement according to the detection result, the micro cell finding accuracy and efficiency of the UE are improved and the electricity consumed by the UE is effectively saved.

The invention discloses a system for monitoring wear of a pantograph sliding plate. The system comprises a three-dimensional (3D) camera and a laser, wherein the three-dimensional camera and the laser are connected with a PLC (Programmable logic controller), and are controlled to be switched on and off by the PLC; the PLC is also connected with an industrial control computer; and the industrial control computer is connected with the 3D camera through a broadband network, and can be used for receiving and storing photos transmitted by the 3D camera through the broadband network. During the passage of a pantograph, the 3D camera can be used for transmitting shot photos of the top of the pantograph through the broadband network, the industrial controllable computer is used for analyzing 3D photos, and the 3D coordinate values of the sliding plate on the top of the pantograph can be obtained clearly, so that the wear condition of the sliding plate can be obtained clearly and accurately, monitoring of the wear condition of the pantograph sliding plate can be automatically realized, and high efficiency and high accuracy are achieved.

The invention discloses a method and a circuit for testing a multi-clock domain. The multi-clock domain comprises a plurality of working clocks. The method for testing the multi-clock domain comprises the following steps that: each working clock and each testing clock in each clock domain provide a corresponding clock for a working circuit through a clock switching control circuit; a selection signal of each clock domain is set to provide a working enable signal for the clock switching control circuit in each clock domain; the effectiveness or ineffectiveness of the working enabling of different clock switching control circuits is controlled by the selection signal of each clock domain; and a test state of the clock domain corresponding to the clock switching control circuit is controlled. By the method and the circuit for testing the multi-clock domain, simultaneous test on circuits in the multi-clock domain and test on circuits in a cross-clock domain simultaneously are realized.

The invention discloses an ore resource consumption assessment method based on the three-dimensional laser scanning technology. The method comprises the steps that early-stage preparations are conducted; mining area range space point cloud data is collected and preprocessed; according to the mining area range space point cloud data, an employed ore body morphology model is corrected and built according to the mining area range space point cloud data; a block morphology data model is built through space division of the employed ore body morphology model; a digital surface model is built to estimate the simulated point cloud data distribution corresponding to block sections; the block range space volume is decompose and estimated through the block morphology model; the ore resource consumption in the employed ore body range is assessed. The invention further provides a high-density continuous uniform distribution data point space range limit estimation method and a space point cloud data simulated distribution quadric surface fitting method. The assessment efficiency and the precision of the ore resource consumption are obviously improved.

The invention discloses a high-precision secondary processor mainly comprising a processor table, a base, at least one group of cutter shell seat, a material transmitting mechanism, a main shaft mechanism, a material blocking mechanism and a material returning mechanism. In addition, a 90-degree horizontal cutting mechanism, a 90-degree transverse cutting mechanism and a 90-degree drilling mechanism can be arranged as required. The high-precision secondary processor has simple and compact structure; PLC is used for controlling and mechanical parts can be processed automatically and secondarily with high efficiency and precision and the production cost for the secondary processing of the mechanical parts is largely reduced.

The embodiment of the invention discloses a virus prevention method and a device, the method includes detecting a network connection request message transmitted from a client, the request message includes a first network address requested by the client for connection; judging if the first network address is utilized as the network address by the virus for updating data; if so, setting the network parameters to lead the client to be connected to a second network address, a server corresponding to the second network address stores a program for killing the virus. The virus prevention method provided by the embodiment utilizes the automatic updating mechanism of the virus to lead the virus to download virus-killing software during the automatic updating of the virus by setting the network parameters, thereby, the trouble of finding virus-killing programs by the customer can be saved, and at the same time, the efficiency and accuracy of the virus-killing software can be improved considerably.

The invention discloses an intelligent control system for water and fertilizer integrated drip irrigation. The intelligent control system is characterized by comprising a drip irrigation tape band body, emitter nodes, soil property sensor module nodes near emitters, an in-drip-irrigation-tape water and fertilizer pressure sensor, a drip irrigation water input pipeline, a drip irrigation water input pipeline solenoid valve, a drip irrigation fertilizer input pipeline, a drip irrigation fertilizer input pipeline solenoid valve, an embedded module and a far-end decision host. According to the intelligent control system, related sensors are mounted at the drip irrigation tape emitters to acquire soil conditions in real time and transmit information back to the host to assist a decision maker in making the best judgment for parameters such as the water-to-fertilizer ratio, the drip irrigation amount and the like, so that the water-to-fertilizer ratio and the flow velocity are adjusted automatically, and purposes of water-saving irrigation, precise water and fertilizer integrated irrigation and the like are achieved.

The invention provides a defect detection method and device for a circuit board. The method comprises the following steps of: obtaining a shot image of a tested circuit board; carrying out pixel information difference on corresponding pixels of the shot image and the standard image; obtaining a difference value of the corresponding pixel; synthesizing the pixel information of each pixel of the shot image with the differential value of the corresponding pixel; obtaining an input image according to the input image of a tested circuit board, inputting the input image into a trained classificationmodel to determine defect information of the tested circuit board according to the output of the classification model, and learning the classification model to obtain image features of the input image corresponding to the circuit boards with various defects. Because the classification model learns to obtain the image features of the input images corresponding to the circuit boards with various defects, the differential value is added to the pixel information of the shot image for image feature extraction, the input data of the classification model can be enriched, and the accuracy and efficiency of defect detection of the detected circuit board can be improved.