118 results about "Subdivision method" patented technology

The invention relates to a rapid three-dimensional dynamic gesture recognition method and system based on a character value subdivision method. To solve the problems existing in the prior art, the invention designs a hardware platform with a small appearance; an accelerometer is used for acquiring the 3-axis acceleration of a user gesture; a gyroscope is used for acquiring the 3-axis angle velocity of the user gesture; a magnetometer is used for acquiring the 3-axis magnetic induction intensity of the user gesture; data error is eliminated through a filtering algorithm, and then attitude algorithm is performed to work out real-time three attitude angles (a course angle, a roll angle, and a pitch angel), three-dimensional dynamic gestures with a certain discrimination such as movement, rotation, ticking, cross making, shaking, and knocking are recognized through the information data by adoption of a character analysis algorithm on the basis of ensuring a certain accuracy and real-time performance, and in this way, corresponding applications can be achieved.

The invention discloses a sine-cosine encoder high precision signal processing system and method. The method is used for processing Asin[alpha] and Bcos[alpha] of a 90-degree phase difference output by an incremental rotary encoder; encoder signals Asin[alpha] and Bcos[alpha] are firstly subjected to differential amplification and shaping filtering; then coarse code counting is carried out to obtain coarse code information, and a real-time fine interpolation position value is obtained through an electronic subdivision method based on the fact that the arc tangent value is directly calculated with the coordinate rotation digital algorithm; and finally, the coarse code information and the real-time fine interpolation position value are integrated to obtain the high-precision angular position and the high-precision angular speed of a motor. The invention solves the technical problem of carrying out refinement of any multiple of an original signal, and achieves any refinement in theory, and thus the rotor real-time angular value (rad) and the rotor real-time speed value (rad / s) higher than the resolution of the original signal are obtained through calculation. The invention also discloses a system employing the method.

The invention relates to a work flow model dynamic segmentation method which supports distributed implementation, which pertains to the technical field of enterprise work flow model segmentation. The invention is characterized in that: the method is realized in each set of work flow system which is composed of a server and a client; the server is provided with a modeling module, a divided piece pool module, a divided piece management module, an activity monitoring module, an activity distribution module and a database, the client is provided with a task client module; wherein, the divided piece pool module sends the set work flow model in the modeling module to the divided piece management module, the divided pieces which are obtained by the dynamic segmentation of the model are sent to the divided piece pool module for the use by the task client module; the function of the activity distribution module is to provide actual implementation station labels for the activities, or designate the actual implementation station labels for the follow-up activities of the new divided piece after segmentation; and the activity monitoring module is used for dynamic recording of the various activities which are implemented by the actual implementation stations. The invention proposes the work flow model dynamic segmentation and implementation method, which has a broad application prospect.

The invention discloses a geological-stratum-structure-based method for generating triangular lattice grids, which is used for a system providing input grids for oil-gas migration and accumulation analog computation. The geological-stratum-structure-based method is characterized by comprising the following steps of: constructing a three-dimensional geological stratum structure model and calculating intersecting lines among stratums based on the geological stratum structure; constructing a triangular grid generation unit for each stratum; cooperatively processing a plurality of units sharing the intersecting lines among the stratums to realize a plurality of lattice grids which continuously transition among stratums; storing the triangular lattice grids by using non-manifold half-side data; and optimizing the local connectivity of a grid vertex point unit by adopting a consistent triangular subdividing method. By using the geological-stratum-structure-based method, the automatic generation of the triangular lattice grids under various geological stratum conditions can be realized and the connectivity of computational grids at stratum junction parts is ensured, so that an attribute value of the oil-gas migration can be used for continuous analog computation on various stratums and furthermore, an effective supplementary means is provided for geological experts to predict the favorable accumulation area of oil gas.

The invention discloses a grid free-form surface toroidal cutter path planning method based on improved Butterfly subdivision. The method comprises the steps of toroidal cutter surface modeling and triangular grid free-form surface modeling. Besides, the cutter surface of a cutter contact point (CC point) and a workpiece surface are geometrically analyzed with combination of the geometrical features of the toroidal cutter. The minimum front inclination angle of the cutter is calculated according to the principle of cutting cutter interference so as to obtain the maximum width of a cutting path, based on which subdivision is performed on the triangular grid surface by adopting the improved Butterfly subdivision method so that the line space of the cutter path is enabled to meet the width requirement of the cutting path. Planning of the spiral cutter path cutter contact point (CC point) is performed line by line along the boundary of the triangular grid with a surface contour curve acting as an initial cutter path so that the position of a cutter center (CL point) is finally solved according to the toroidal cutter geometrical model. The method is applied to high-speed machining so that continuity of the formed cutter path can be enhanced, cutter load fluctuation can be reduced, and cutting stability can be enhanced and the service life of the cutter can be prolonged under the premise of guaranteeing the cutting quality.

The invention relates to a differential capacitance displacement conversion and subdivision method and a capacitive linear displacement measuring system. First, the displacement is solved with the use of a conversion reference of differential capacitance displacement, no modulation, demodulation, amplifier or A / D is needed, and measurement and subdivision are simple and accurate; second, differential capacitance variation is converted into pulse width, a circuit needs no zeroing, no noise, parasitic or zero drift interference exists, and the range is not limited and is only related with the series resistance; third, the capacitive linear displacement measuring system is provided by the use of the method and the circuit, the system is large in range and can be used for performing absolute position measurement, a sensor and a circuit of the system are simple, and the system is fully digital, high in precision and good in stability, can be used in harsh environments with water, oil or dust pollution and is better than an optical grating, a ball grid or the like; and fourth, the sensor, the circuit, a display and the like are integrated into a micro (MEMS) device, the precision is at the nanometer level, motion parameters of the MEMS such as mechanical position, displacement, velocity, amplitude and frequency can be accurately measured, and parameter transformation is remotely controlled.

The provided geometrical data refine method based on triangular interpolation surface comprises: with topology rule, splitting the triangular topological mesh to add new vertex points and form new mesh topology; with geometrical rule, calculating position for all vertex points; cycling former steps till meeting terminal condition. This invention can form thick body data point set.

The invention provides an adaptive Loop subdivision surface drawing method based on an irregular region. According to the drawing method, based on an adaptive mode of the irregular region, the position of a control vertex is computed by a compute shader, an irregular original surface patch is divided into a plurality of regular sub surface patches according to the relationship between the current original surface patch and an adjacent original surface patch, a triangulation grid is reconstructed and drawn by a tessellation shader to obtain corresponding surface patches, and T cracks can be effectively eliminated. The position of the vertex of each layer in the triangulation grid can be accurately computed by the aid of the tessellation shader through division, so that all computed vertexes and normal vectors are as same as those of extreme Loop subdivision surfaces, hardware is drawn by the tessellation shader, the problems of huge memory and bandwidth of a global subdivision method are solved, and the Loop subdivision surfaces can be accurately drawn in real time.

The invention discloses a feature subdivision method for call center customer service staff in mass data. The feature subdivision method includes: acquiring a preset number of workers; determining a cluster index, and determining data corresponding to the cluster index, of each worker as cluster index data of the worker; performing clustering through a K-Means algorithm according to the cluster index data of each worker so as to obtain a cluster result; and dividing the workers into different classes according to the cluster result to obtain a group result. A call center work management method can automatically adapt worker subdivision work in the mass data, is high in accuracy, and can avoid the situations that classification based on subjective experience does not confirm with actual task performance of the workers.

The invention discloses a tile type large file organization storage method of three-dimensional model data. The method comprises the following steps: tiling the three-dimensional model data according to a global grid subdivision method; carrying out tightening and reconstruction on the original discrete tile data according to a calculation rule so as to form data block large files; arranging the divided data block large files from low to high according to resolution so as to form a tile pyramid ; defining disc organization structures and internal memory organization structures of the data block large files; decreasing the hierarchical division of the tile pyramid, and merging every two adjacent layers of data block large files of the tile pyramid into one layer according to a proportion of 1: 4; reconstructing the merged data block large files; re-mapping the vision field range of the hierarchy of the tile pyramid and storing the vision field range as a new data layer file; and when dividing one layer at each time, dynamically writing a file by utilizing an internal memory mapping technology, so as to write the tile pyramid into a disc layer by layer and then form large files of three-dimensional model data in which organizations are effectively stored in a tile manner.

The invention discloses a pathological voice subdivision method. A model training module and an identification module are included. The model training module is used for modeling of input voice signals, getting corresponding likelihood, calculating and comparing matching probabilities, and finding out voice signals meeting the conditions. The identification module is used for matching the voice signals which meet the conditions. The lengths of the input voice signals are not required, the voice signals can have characteristic parameters of any types, and different weights are distributed to different characteristics. Therefore, advantages of all parameters can be utilized, and the number of dimensions of the characteristic parameters is not limited. Multiple times of trainings can be carried out, retraining is carried out on voice signals which are difficult to identify, and threshold values, finishing conditions and identification conditions can be flexibly set in the training. By means of the pathological voice subdivision method, types of pathological voices can be automatically set and subdivided precisely, pre-diagnosis of voice diseases and timely-tracking of recovery condition of a patient are achieved, and meanwhile the pathological voice subdivision method is suitable for being used for health self-inspection of teachers and singers and the like.

The invention provides a method for directly dividing on sphere by combining meridian and parallel and adopting thinking of regular polyhedron progressive recursion. The method utilizes meridian and parallel, the meridian and parallel are directly connected on sphere by virtue of arc, and multi-stage recursive rhombic dividing is carried out. Firstly 0-stage dividing is carried out, meridians of 0 degree, 90 degrees, 180 degrees and 270 degrees are utilized to divide the earth into four equal parts, and the four equal parts are respectively numbered to be 0, 1, 2 and 3 from east to west from the meridian of 0 degree. Then multi-stage recursive rhombic dividing of 1-n stages are carried out. Each divided rhombic sphere is numbered. The method is operated on sphere directly, no inscribing on the regular polyhedron is required, and no projection is required, thus operation is simple. Random resolution can be realized and seamless zero-overlap global dividing is basically realized.

The invention relates to an SRG (sphere rhombus grid) subdivision code and geographic coordinate converting algorithm, which is based on the SRG subdivision method. The SRG subdivision code and geographic coordinate converting algorithm inherits the advantages of the SRG subdivision method, SRG address codes have constant directivity and are favorable for proximity searching, no projection conversion is involved in the SRG subdivision method and no projection is involved during conversion, simple arithmetic operations of addition, subtraction, multiplication and division are involved during computing, and computing speed is high. Besides, rhombus blocks hard to differentiate are differentiated by the aid of the coordinate system, and conversion accuracy is increased.

The invention discloses a method for automatically recognizing a lip color based on a support vector machine. The method comprises the following steps of: (1) establishing a lip color training sample set: firstly, establishing a lip image library; and secondly, extracting the histogram characteristic vector of a lip image by using a histogram interval subdivision method to form a lip color training sample set; (2) training the lip color training sample set by using the support vector machine to establish a lip color classifier; and (3) recognizing the lip color of a new lip image by using the lip color classifier. By adopting the histogram characteristic vector obtained with the method, fine difference information on different lip colors can be captured; the lip color is recognized by simulating traditional Chinese medical science, so that the defects of strong subjectivity and poor repeatability of the conventional method are overcome; and a training lip image is judged in advance by a well-known traditional Chinese medicine expert and the lip image is recognized by learning, so that the method is the conclusion and the inheritance of lip color recognition experience in traditional Chinese medicine.

The invention discloses a point cloud mesh refinement system and a point cloud mesh refinement method, which are applied to a computer device. The method comprises steps: a point cloud for a target product is acquired, and triangulation is carried out on the point cloud for the target product; triangular meshes in need of refinement are selected, and a needed iteration number N is inputted according to the precision requirement for the mesh refinement; according to the shape and the position of each triangular mesh, a mesh subdivision method is used for subdividing each triangular mesh into different types; a loop triangular mesh subdivision algorithm is used for carrying out N times of refinement processing on the triangular meshes of different types, and triangular meshes after refinement are obtained; point cloud smoothing processing is carried out on the triangular meshes after refinement, a three-dimensional mesh model for the target product is obtained, and the three-dimensional mesh model for the target product is displayed on a display device. When the system and the method of the invention are executed, quick and effective refinement can be carried out on the point cloud mesh of the target product, the model features of the target product can be kept, and a refiner and smoother three-dimensional mesh model for the target product can be acquired.

The invention belongs to the technical field of wireless electricity measurement and particularly relates to a distance-Doppler matching method for measuring a radar cross section of a moving target. The invention discloses a high-precision distance-Doppler matching method for measuring the radar cross section (RCS) of the moving target. By utilizing a matching filter with time shifting and a frequency domain subdivision method, the high-precision distance-Doppler matching is realized, and therefore, a high-precision RCS measurement value is obtained. The method has the advantages that compared with a traditional pulse compression method, by utilizing the matching filter with the time shifting, the fluctuation of energy in a distance domain is obviously reduced; compared with a traditional Doppler filter set, by utilizing the frequency domain subdivision method, the fluctuation of energy in a Doppler domain is obviously reduced; and compared with traditional schemes, by utilizing a signal processing method, errors of matching between the distance domain and the Doppler domain are reduced, and therefore, the high-precision RCS measurement value is obtained.

An embodiment of the invention discloses a recursive curved surface generating method and device on the basis of quadrangle segmentation. The method comprise the steps of scanning an object to obtain corresponding discrete three-dimensional point cloud data; utilizing a Delaunay triangulation algorithm to preprocess the three-dimensional point cloud data and obtaining a good data structure; merging an interpolation subdivision method and iterating according to a recursive equation to generate new vertexes and a new quadrangle surface piece; conducting iteration for corresponding times according to a needed fine degree; drawing every new quadrangle surface piece after the iteration is completed, and displaying a three-dimensional model of the object. According to the embodiment, the recursive display method merging the classical Catmull-Clark method and the interpolation subdivision method is used. A smooth three-dimensional model created by the novel interpolation subdivision method through subdivision of a rough three-dimensional model is a little bit smaller than a real model. The novel method enables accurate shape-preserving display closest to the real model and better achieves gradual quick accurate display of a three-dimensional model.

The invention provides a parametric modeling method of a folded plate reticulated shell based on grasshopper, which is used for solving the problems of modeling difficulty, low modeling efficiency andlow repetition utilization rate of the existing space folded plate reticulated shell with complex modeling. The invention comprises the following steps: expressing the geometrical parameters of a lattice shell model to be established by corresponding input plug-ins; generating a flap surface or quasi-spherical surface by using the application plug-ins according to geometrical relation; selectingone of the planes and using an isoparametric method or subdivision method to mesh a folded surface to generate local meshes; gridding all the planes by the circumferential array plug-in, and generating the single-layer folded latticed shell by connecting the bars with the plug-in. The invention has the advantages of simple design method, new model can be established by changing the parameter valuein the model, the modeling efficiency can be improved, the force analysis and the structural design optimization of the complex space reticulated shell structure under different types and different parameters are improved greatly and conveniently, and the use of the complex space reticulated shell structure is promoted.

The invention discloses a finite difference time domain electromagnetic calculation carrier meshing method. According to the method, the mesh size is jointly controlled by the discrete electrical dimensional accuracy and the carrier surface approximation accuracy, the geometrical approximation degree between the divided mesh model and the carrier model is high especially under the situation of low frequency with large wavelength, finite difference time domain electromagnetic calculation is performed on the basis of the mesh model, and the calculation result is relatively accurate; surface approximation accuracy controlling parameters for meshing the mesh model are provided, the blindness of meshing precision control of the conventional direct subdivision method is avoided, and the calculated amount for meshing can be effectively controlled; when mesh refinement is required, all that is needed is to perform octree recursive subdivision to the black nodes in the bottommost layer, and mesh refinement is easy to realize.

The invention provides a nanometer laser measurement rule and its realization close method and belongs to laser fine measurement technique field and is characterized by the following: it is based on the displacement sensor HeNe laser system using dopal as chamber lens and adds a fine micro displacement piezoelectricity sensor PZT which can alter the chamber length together with the subject to be measured. The output crossed polarized light goes through the polarized splitter lens and shoots on two photoelectric detectors and is under the large number and small number processes by the following circuit to get and output the measurement results. It eliminates the current errors in current technique and counts the power polarized curve cycles number so to achieve the high accuracy and resolution.

The invention relates to a grating subdivision device and a grating subdivision method based on an FPGA (field programmable gate array), and belongs to the technical field of grating subdivision processing. An output signal of an FPGA device controls the clock and the chip selection terminal of an A / D conversion circuit; after an input signal passes through a differential amplification circuit I and a differential amplification circuit II: a two-position level signal is generated by a zero-crossing comparison circuit I and a zero-crossing comparison circuit II; besides, an absolute value signal is obtained by an absolute value circuit I and an absolute value circuit II: a one-position level signal is obtained from the absolute value signal through a comparator II, and each cycle of a sine signal output by a reading head is divided into 8 linear regions from the absolute value signal through a comparator I, an analog selector, a following circuit and the A / D conversion circuit, and the 8 regions are subdivided one by one to obtain an 8-position level signal. According to the device and the method, the calculating speed of a subdivided data processing algorithm is increased, the defect of low calculating speed caused by a singlechip microcomputer and an DSP (digital signal processor) is overcome, and the subdivision multiple is improved.

The invention provides a method for measuring small facula intensity distribution base on a two-dimension subdivision method, relating to a method for measuring the small facula intensity distribution, solving the problem that the intensity distribution measurement can not be carried out on the small facula with nonuniform energy distribution or extremely small size currently. The measuring method completely irradiates the incident facula on an image element of the photosensitive detecting surface of a CCD detector, records gray level images in various scanning states in a snakelike scanning mode, and obtains the intensity distribution of the incident facula finally through calculation. The invention can carry out direct measurement on the facula with nonuniform energy distribution or extremely small size, and is suitable for the filed of tiny facula measuring.

The invention discloses a welding seam recognition method based on piecewise linear fitting. The method comprises the following steps: (1) obtaining a series of discrete points of a laser centerline and segmenting the discrete points by adopting a recursive subdivision method; (2) searching the interior points of the discrete points in the segment by means of random sampling consistency in each segment of discrete points; (3) judging the ratio w of the number of the searched interior points to the number of the discrete points of the segment and outputting the searched interior points and fitting the searched interior points into straight lines when the ratio w is greater than the threshold we, otherwise returning to the step (2); (4) using the distance proportion judgment method to mergethe adjacent straight line segments with similar slope; (5) and comparing and matching the merged straight line segments and the intersection points with the set weld type and extracting the welding seam recognition point position Ps under the corresponding welding seam type. The non-planar situation of the surface on both sides of the welding seam can be coped with so that adaptability is high and the recognition stability of the welding seam tracking system can be improved.

The invention discloses a driving method for a three-dimensional probe in a three-dimensional / four-dimensional ultrasonic imaging system. The driving method is used for driving and controlling a stepping electric motor of the three-dimensional probe, and comprises the following steps of selecting the model of the three-dimensional probe of the three-dimensional imaging system; setting electrical and mechanical structure parameters of the three-dimensional probe; setting application parameters of the three-dimensional imaging system; initializing a control model; calculating parameter values needed in the control process according to the parameters of the system and the probe; and carrying out operational control according to the model of a control state machine. The driving method for the three-dimensional probe in the three-dimensional / four-dimensional ultrasonic imaging system has good compatibility and control performance, and can be suitable for a three-dimensional probe of a stepping electric motor in any model, by means of sine control impulse and an electrical control subdivision method, vibration of the probe is greatly reduced, and interference of relatively-large power driving harmonic waves on system signals is greatly reduced.

The invention discloses a spatial subdivision method for intersecting linear geographic elements. The method includes the following steps that: intersecting linear elements are dispersed so as to form point sets based on the principle of a Voronoi distribution model; point addition processing is performed on intersecting positions of every two intersecting linear elements; Delaunay triangulation and Voronoi diagram generation are performed based on the point sets; and a generated Voronoi diagram is clipped, and sub regions generated by point sets which belong to the same linear element are merged, so that sub regions of different linear elements can be obtained, and therefore, spatial subdivision of the intersecting linear geographic elements can be realized.

The invention provides a mesh subdivision method based on laplacian coordinates, relates to the technology of three-dimensional mesh model modeling and subdivision, and particularly relates to the technology of maintenance of enhancement of fineness of three-dimensional mesh model specific characteristics. With increasing development of computer hardware, multiple computer graphical applications use complex models with high-fineness detail information so as to realize high sense of reality, and the construction of the high-fineness mesh models cannot meet the requirements. The mesh subdivision method is put forward to facilitate work of the aspect. Based on a low-fineness mesh model, the degree of interpolation of a vertex and the topological structure of a new mesh are confirmed by a user, and then an extended laplacian matrix and the corresponding expected laplacian coordinates are constructed to form a linear system so that the position of the vertex of the new mesh is solved. Finally, a new high-fineness mesh model is constructed through combination of the topological structure. The concrete realization process is shown in the appended drawings.

The invention discloses a K-Means-algorithm-based direct broadcast platform user subdivision method and system, relating to the field of data mining. The method comprises the steps of (A) collecting the basic information data and behavior information data of users from a server, selecting the indexes of users in a sampling time as sample data, wherein the sampling data comprises M indexes of N users, (B) carrying out preprocessing and standardization processing on the sample data, and (C) using K-Means algorithm to carry out clustering, calculating the class to which each user belongs, and constructing a user subdivision group according to the class to which each user belongs in the sample data. The user subdivision group can be obtained according to the indexes of the users, and the division precision is high.

An absolute coding raster phase subdivision method based on a charge coupling device (CCD for short) is disclosed, comprising the following sequential steps of: (1) reading code channel graphics of the absolute coding raster with a CCD reading head; (2) subjecting a periodical discrete space function sequence formed by a row of the lowest code channel graphics in the obtained code channel graphics of the absolute coding raster to Flourier conversion so as to result in spectral distribution thereof; (3) taking out a fundamental component of the spectral distribution obtained in the step (2); (4) subjecting the obtained fundamental component to inverse Flourier conversion so as to result in a fundamental wave signal and a phase thereof; (5) generating the obtained fundamental wave signal in step (4) into a continuous phase by using N periodical phases; (6) expressing measurement resolution of the lowest code channel graphics in the absolute coding raster by using the average of the continuous phase. Such a method can realize the subdivision of pixels and remarkably enhance resolution and measurement accuracy of the absolute coding raster.

The present invention discloses a final subdivision method of a loading instrument of a bulk carrier. The method comprises the following steps: inputting initial data into the loading instrument; determining an objective function, a design variable and a constraint condition of final compartment according to floating state equations; using a differential evolutionary algorithm to resolve and optimize the objective function; using the differential evolutionary algorithm to perform operations of variating, crossing and selecting; and and traversing an individual set that meets the condition and outputting a result. In the present invention, as values of xF, TPC and MTC are not used to perform calculation, and instead floating state equilibrium equations are used to set the objective function and the constraint condition; and the differential evolutionary algorithm is used to perform calculation, thereby giving a more precise result and effectively reducing errors. According to the final subdivision method of a loading instrument of a bulk carrier disclosed by the present invention, not only can a small amount of goods be used to adjust a mean draft and a trim of a vessel, but also a large number of goods can be used to adjust the mean draft and the trim; and adjustment ranges of the mean draft and the trim are wide. According to the method disclosed by the present invention, as related data needs not to be manually inquired and manually input, the manpower can be effectively reduced, and the convenience is improved.

The invention relates to the technical field of computer graphics, and provides a self-adaptive subdivision method and device. The method includes the steps that the dihedral angle value of any two adjacent triangular patches participating in area subdivision is detected; when the dihedral angle value is larger than a preset threshold value, whether the two adjacent triangular patches are both inner triangular patches or not is judged; if the two adjacent triangular patches are both the inner triangular patches, the control states of the two adjacent triangular patches are both set as Inner_1; if one of the two adjacent triangular patches is the inner triangular patch and the other triangular patch is a boundary triangular patch, the control state of the inner triangular patch is set as Inner_1, and the control state of the boundary triangular patch is set as Bound_1 or Bound_2. According to the self-adaptive subdivision method and device, a local area can be subdivided conveniently and rapidly, and huge time consumption for global subdivision is avoided.