72 results about "Motion modeling" patented technology

A method and apparatus for dynamic modeling of an object having material points. The method includes receiving signals from a sensor which correspond to respective material points; providing a volumetric model, having functions as parameters, representative of the object; and adapting the parameters to fit a changing model shape. This method of dynamic shape modeling dynamic motion modeling or both includes shape estimation and motion analysis. The apparatus includes a signal processor for receiving signals from a sensor, a second signal processor for providing a volumetric model having functions as parameters representative of the object and a third signal processor for receiving sensed signals and adapting the model and providing a dynamic representation.

A method for tracking coronary artery motion includes constructing (11) a centerline model of a vascular structure in a base phase image in a sequence of 2D images of coronary arteries acquired over a cardiac phase, computing (12), for each pixel in a region-of-interest in each subsequent image, a velocity vector that represent a change in position between the subsequent image and base phase image, calculating (13) positions of control points in each phase using the velocity vectors, and applying (14) PCA to a P×2N data matrix XT constructed from position vectors (x, y) of N centerline control points for P phases to identify d eigenvectors corresponding to the largest eigenvalues of XXT to obtain a d-dimensional linear motion model {circumflex over (α)}p, in which a centerline model for a new image at phase p+1 is estimated by adding {circumflex over (α)}p to each centerline control point of a previous frame at phase p.

The invention relates to a layered observation vector decomposed hidden Markov model-based method for identifying behaviors. The method comprises the following aspects of: target detection, target tracking, characteristic extraction, motion modeling, behavior identification and the like. The method improves the conventional hidden Markov model by aiming at deficiencies of a present method for identifying behaviors of multi-user, separates an individual state and an interactive state in the model to stand out interactive relation among targets, and reduces calculated amount by decomposing observation nodes. Meanwhile, the model allows the number of the targets taking the movement to be changed, so that the method has more flexibility in characteristic selection problem than the traditional method and allows simultaneously using discrete characteristics and continuous characteristics. The method plays an important role in analyzing the multi-user interactive behaviors in the fields of video monitoring, and content-based video retrieval and the like.

The invention discloses a robot simulation learning method based on a dynamic system model. Simulation of teaching movement of the robot is achieved through learning, specifically, the demonstration motion is modeled into a nonlinear dynamic system model through a gaussian mixture model, in addition, the stability of the motion model is guaranteed through the method with the additional stability constraint conditions, the parameter learning problem of the motion model is converted into a constraint optimization problem, so that a complete description of the motion model is obtained, and finally, the motion model obtained through learning is used as a control strategy to guide the robot to imitate the teaching motion. The method is used for teaching motion of target point fixation, the method has good stability, all the generated motion trails are converged to a target point, and has good expression capacity for simple and complex teaching movement, the generalization ability of the motion model is good, a motion track which is smooth and can be converged to a target can be generated outside the teaching movement range.

The invention discloses a method for calculating an optimal cooperation distance of a D2D fusion network based on file popularity. The method comprises the following steps: cell user movement is modeled as random distribution in a spatial Poisson point process, each user caches a video file, the user can obtain the video file by selecting a cellular mode or a D2D mode, the popularity of the video file is simulated by Zipf distribution, a user number ND2D activated by D2D and the user number NBS activated in the cellular mode in a single cell are obtained according to different features of D2D and cellular communication, then a single user obtains the network capacity of the fusion network by the average throughput in the D2D mode and the cellular mode, and a specific application scene and a service demand are obtained by an existing fast iterative algorithm, so the D2D fusion network obtains an optimal cooperation distance d* with the maximum throughput.

The invention discloses a wind-driven automatic feeding system and an evenly feeding method. An automatic feeding device, a communication device, a central control device and a Beidou navigation device are arranged on an air propeller wind-driven boat. The central control device is respectively connected with the communication device and the Beidou navigation device through signal lines. A communication base station is arranged on the bank and is in communication with the communication device and a monitoring server arranged in a system monitoring room. The steps of modeling of a projectile device of a feeding machine, modeling of oblique projectile motion of feed particles, modeling of accumulation density distribution of feed on the water, solving of the optimal running parameters of even feeding of the automatic feeding system and the like are executed to automatically determine the feeding amount of a feeding unit of the automatic feeding device in unit time, the speed of rotation of a projectile disk, the projectile fan angle of the feeding machine, the driving speed of the wind-driven boat, the width of two adjacent feeding travels in a path plan and other optimal running parameters of the automatic feeding system. By means of a Beidou satellite navigation system, the path is tracked through the Beidou navigation device, and thus automatic and even feeding operation can be completed.

The invention belongs to the technical field of aircraft elastic motion modeling, and specifically relates to an elastic motion modeling method of a trailing edge rudder gliding aircraft. The method comprises the following steps: (1) converting an elastic body transverse free vibration into superposition of various mutually independent main vibrations by utilizing modal orthogonality; (2) according to a linear small disturbance hypothesis, under the action of external forces, still describing the transverse vibration of an elastic body approximatively by utilizing modal superposition, and determining a vibration mode function by structural characteristics (rigidity and quality distribution) of the elastic body and a boundary condition of the elastic body; (3) establishing a vibration differential equation according to the d'almbert's principle and other methods; and (4) analyzing the external forces, which comprise an aerodynamic force, a rudder surface control force, a posture jet pipe control force, a rudder surface swinging inertia force and the like, influencing an elastic vibration of the gliding aircraft. A traditional elastic motion modeling method is corrected and improved.

The invention belongs to the field of image processing technologies and relates to a discriminative online target tracking method based on videos in dictionary learning. The method includes the first step of constructing a template of a target through manual calibration or according to an existing tracking result and obtaining the state of the target to be tracked at the same time, the second step of constructing an initial stage matrix of the target, conducting motion modeling and conducting sampling to form candidate samples, the third step of partitioning the candidate samples and the template, conducting sparse coding on all partitions and obtaining expression coefficients of the corresponding partitions, the fourth step of inputting the expression coefficients as features into a classifier of the corresponding partition, obtaining the determination confidence degree of each candidate sample in the partition, randomly selecting K determination confidence degrees of the partitions, conducting summation, traversing all possibilities and selecting the candidate sample with the maximum summation value as the target tracking result of a current frame, wherein the selected frequency of the candidate sample in all the possibilities are maximum. Through the method, the accuracy and robustness of target tracking on the condition of a complicated background and changing of target appearance can be improved.

An apparatus includes a scanner (102, 104) and a scanning motion monitor (100). A motion modeler (116) uses data from the scanning motion monitor (100) and the scanner (102, 104) to generate a motion model which describes motion of a region of interest of an object. A treatment planner (112) uses image data from the scanner (102, 104) to establish a treatment plan for the object. A treatment device (114), which operates in conjunction with a treatment motion monitor (108), uses the motion model to compensate for motion of the object during application of the treatment.

The invention discloses a method for eliminating planar motion blur by using two ordinary images. The method comprises the following specific steps of: 1, computing a foreground image, a background image and transparency; 2, modeling motion within exposure time; 3, establishing a target function; 4, registering two input images to obtain initial motion estimation; 5, evaluating a fuzzy kernel to minimize a target function in the step 3; 6, estimating a binary transparency image by using a Markovian random field; 7, re-estimating the fuzzy kernel to obtain a fuzzy kernel energy function; 8, modifying a Richardson-Lucy algorithm, and combining the modified Richardson-Lucy algorithm with the motion modeling in the step 2 to obtain a clear motion object foreground image; and 9, combining a deblurred foreground with a deblurred background by using deblurred transparency to obtain a deblurred image. In the obtained deblurred image, a boundary is kept clear, and a ''ringing'' effect is suppressed simultaneously.

A system for motion guidance of a motion sensing game is applied to computing devices. A series of standard motion samples included in the motion sensing game and standards for judging whether a motion executed by a user is correct or not are set up, correct motion modeling images of each standard motion are sequentially displayed on a display screen of a computing device according to the motion samples, real-time motion data, relative to spatial X, Y and Z axes, of each body part in each motion execution process are detected by detectors worn on each body part of the user by the aid of an electronic gyroscope, and the detected real-time motion data are received by the system through the network and compared with corresponding standard motion samples to judge whether the motion executed by the user is correct or not according to the set standards. The invention further provides a method for motion guidance of the motion sensing game.

The invention discloses a left and right index finger continuous motor imagery recognition system based on actual motion modeling. The left and right index finger continuous motor imagery recognitionsystem based on actual motion modeling comprises the following steps: performing a left and right index finger continuous keystroke: autonomously performing the left or right index finger continuous keystroke by a user according to a left or right arrow displayed on a computer screen; acquiring electroencephalogram signal data generated by the user when the keystroke is performed, reading electroencephalogram signals and processing and analyzing the data to obtain a classification result; by the characteristic which can achieve relatively high classification accuracy, establishing a classification model for recognizing a continuous imagery keystroke so as to evaluate the method and the ability of motor imagery of the user, wherein a training method and an evaluation criteria are provided for a motor imagery task. The classification performance of imaginary actions is observed by a motion intent recognition model established by real actions; through further research, a more complete brain-computer interface system can be obtained, so that considerable social and economic benefit can be expected to be obtained.

The invention discloses an adaptive-filtering target tracking and positioning method based on an embedded platform, comprising the following steps: (1) obtaining observation data of geographical position of a target aircraft at preset time by the data bus of the Advanced RISC Machine (ARM) embedded platform; (2) performing modeling according to the observation data of geographical position in three-order Lagrange's series in the three-dimensional space motion by a motion modeling module; (3) performing residual data processing for the error value output by the motion modeling module by an information extracting module; (4) performing parameter optimizing for the gain matrix according to the evaluated key parameter of the noise space obtained by the information extracting module; (5) obtaining the filtering data at preset time by an iterative filtering module; and (6) performing coordinate system of the filtering data and outputting the data by an output interface module. The adaptive-filtering target tracking and positioning method of the invention achieves high-stability and performance optimization when keeping low computation complex degree and less resource occupation of a core processor.

The invention discloses an action video classification method and system based on multilevel motion modeling, which is used for performing multilevel comprehensive modeling on intra-segment and inter-segment motion information and comprises two neural network branches, the extraction module is used for extracting appearance information and inter-segment motion information of a foreground target; and the intra-segment branch processes a difference value of adjacent video frames in each video segment and is used for extracting intra-segment motion information of the foreground target. The frame difference features extracted by the intra-segment branches are used for weighting the inter-segment branch features according to channels, and finally the convolution features of the two branches are fused and jointly input into a classifier for video classification. The method is easy and convenient to implement, the means is flexible, and the remarkable classification effect improvement is achieved on the action video data set.

The invention provides a vehicle feature deep learning identification trajectory tracking method based on an image system. The system comprises a road monitoring camera, a calculation processing host and a display screen. The method comprises the following steps: firstly, performing deep learning algorithm target recognition on an image acquired by a camera to obtain an original image vehicle target frame data set, and then performing space projection transformation on the original image to obtain a projection image vehicle target frame data set; and associating the projection image vehicle target by Kalman filtering and a Hungary algorithm so as to generate a projection image vehicle target track. The invention provides a vehicle tracking method which can be installed on the ground and is beneficial to modeling of target motion by Kalman filtering.

A method is provided for processing a sequence of sets of 2D projection images of a moving object, wherein the sequence of sets of 2D projection images is obtained by a medical imaging system that is in motion along a trajectory. The method comprises determining a sequence of images which minimize a function dependant on a set of 3D images, a term of fidelity of the sequence of images, a function of spatial and temporal compression of the sequence of images, a compressibility parameter, and a sequence of operators for an approximate modelling of motion. The sequence of operators leads to a compression constraint augmented by partial knowledge of the motion and the minimization comprises defining a decreasing sequence of degrees of compressibility for which an estimation is determined from an initial sequence.

The invention relates to the technical field of synthetic aperture radar measurement, and discloses a method and a device for measuring the baseline length of an airborne synthetic aperture radar. Themeasuring method comprises the following steps of: mounting a measuring fiber engraved with fiber Bragg gratings on a measuring substrate, wherein the measuring substrate is disposed on a wing and connected to the airborne synthetic aperture radar sub-array; and reconstructing a shape function of the measurement substrate by the strain and the curvature of the measurement substrate, thereby calculating the baseline length between the radar sub-array. According to the method and the device for measuring the baseline length of the airborne synthetic aperture radar, modeling without wing motioncan be implemented, the accuracy of wing motion modeling cannot affect, and the problem of baseline length measurement of airborne synthetic aperture radar is successfully solved. The measurement accuracy is high, and the method has the advantages of little influence on the wing structure, light weight, and no weather influence.

The invention discloses an adaptive ship motion modeling method applied to ship motion control. The method comprises the following steps of firstly, determining a six-degree-of-freedom ship motion model according to a ship operation coordinate system and a ship motion description; secondly, obtaining a planar three-degree-of-freedom motion model describing ship motion characteristics according tothe six-degree-of-freedom ship motion model and the ship motion characteristics; thirdly, simulating the ship motion according to a preset high-precision ship motion model, and selecting a support vector regression algorithm as an identification method in combination with structural features of the ship motion model; fourthly, optimizing parameters of the support vector regression algorithm by adopting an artificial bee colony algorithm in order to obtain an optimized support vector regression algorithm; and finally, carrying out model identification by utilizing the optimized support vector regression algorithm in order to obtain an identification result, and selecting a model with the highest precision from the identification result as a target three-degree-of-freedom ship motion model according to a preset evaluation standard. According to the method, a modeling process can be simplified and the precision of the model can be improved.

The invention discloses a shipboard aircraft cluster motion modeling simulation method. The method comprises the following steps of: motion solution space construction: establishing a motion solutionspace of a shipboard aircraft cluster; motion modeling: establishing a motion control model of the shipboard aircraft cluster; and motion simulation: inputting parameters required for simulating motion of the shipboard aircraft cluster on the basis of the motion solution space and the motion control model, and carrying out shipboard aircraft clyster motion simulation. Through the shipboard aircraft cluster motion modeling simulation method, effective motion solution spaces and motion control models can be provided for motion simulation of shipboard aircraft clusters, so that motion laws of theshipboard aircraft clusters can be researched under various simulation conditions. The method is capable of better simulating rising / landing and cruising behaviors of the shipboard aircraft clustersunder various environments, so as to observe operation data of shipboard aircrafts in real time and provide comprehensive simulation data for real large-scale shipboard aircraft cluster flight experiments.

The invention discloses a human body posture real-time estimation method based on RGBD image feature fusion. According to the method, a method based on event triggering is adopted to obtain human bodyposture fusion estimation of depth and color images. According to the invention, through fusion of color and depth image features, error identification of a hand-held tool is effectively reduced, andthe robustness of illumination change is improved; and in order to solve the problem that human body joint point motion modeling is difficult, a BP neural network is used for carrying out human bodyjoint point motion modeling, so that the uncertainty of each joint point motion model is reduced, and the accuracy and robustness of human body posture estimation are effectively improved.

The invention discloses a method and a system for modeling the equivalent satellite attitude motion of vertical rods in the deployment process of a loop antenna. The method comprises the following steps: establishing a coordinate system describing each quadrilateral unit in a ring antenna main body structure and a ring antenna coordinate system; establishing the relative position equations of eachvertical bar in the main structure of the loop antenna in the loop antenna coordinate system; combining all the vertical bars with the satellite motion equations to form the coupled motion equationsof the loop antenna and the satellite; establishing the equations of motion constraint between the radius of large loop antenna and the vertical bar; establishing the governing equation between motionconstraint and control force; obtaining motion information of the satellite and the loop antenna by solving the coupled equations of motion of the loop antenna and satellite and the governing equations of motion constraint and control force by numerical integration method. The method guarantees the calculation precision of the antenna deployment process, greatly improves the calculation rapidityand reduces the modeling complexity.

A method is provided for processing a sequence of sets of 2D projection images of a moving object, wherein the sequence of sets of 2D projection images is obtained by a medical imaging system that is in motion along a trajectory. The method comprises determining a sequence of images which minimize a function dependant on a set of 3D images, a term of fidelity of the sequence of images, a function of spatial and temporal compression of the sequence of images, a compressibility parameter, and a sequence of operators for an approximate modelling of motion. The sequence of operators leads to a compression constraint augmented by partial knowledge of the motion and the minimization comprises defining a decreasing sequence of degrees of compressibility for which an estimation is determined from an initial sequence.

The invention relates to a motion modeling and decoupling method for an air-launched cruise aircraft with a brake parachute. The method comprises the following steps: step 1, carrying out dynamics andkinematics modeling on a parachute ejection system; 2, solving the acceleration of a connection point by using rigid body kinematics and dynamics knowledge; 3, carrying out equal decoupling by usingthe acceleration of the connection point of the projectile body and the brake parachute; 4, carrying out connection comparison simulation with a spring. The invention has the advantages that 1, the parachute ejection system is regarded as spherical hinge connection, an original simplified six-degree-of-freedom parachute ejection model connected through a spring is expanded to nine degrees of freedom, the simulation calculation precision and the calculation efficiency are improved, and higher numerical calculation stability is achieved; 2, the analytical solution precision of the rigid hinge interaction force of the missile-parachute system model is higher, more actual conditions can be simulated, and reference can be provided for parameter design of an air-launched weapon pilot with a brake parachute;.

The invention provides an RFID-based swimming safety monitoring system. The system comprises an exercise modeling module, an exercise assessment module and a safety monitoring module, wherein the exercise modeling module is used for defining an activity in a single period as an exercise unit according to a periodical characteristic result of a swimming exercise, dividing the state of the swimmingactivity of an object into a non-swimming state and a swimming state, completing state division and transfer according to law analysis on received signals and establishing a personal stroke model by completing exercise unit division; the exercise assessment module is used for realizing stroke recognition and classification and stroke proficiency assessment on the object at least on the basis of the exercise modeling module; and the safety monitoring module is used for completing swimming activity monitoring and safety early-warning according to the received corresponding signals. Through the system, the demand for privacy in a swimming scene and the demand for low interference on the exercise of an RFID tag wearer are both considered, action recognition of the wearer is realized, and therefore the purpose of safety monitoring is achieved.

A method for tracking coronary artery motion includes constructing (11) a centerline model of a vascular structure in a base phase image in a sequence of 2D images of coronary arteries acquired over a cardiac phase, computing (12), for each pixel in a region-of-interest in each subsequent image, a velocity vector that represent a change in position between the subsequent image and base phase image, calculating (13) positions of control points in each phase using the velocity vectors, and applying (14) PCA to a P×2N data matrix XT constructed from position vectors (x, y) of N centerline control points for P phases to identify d eigenvectors corresponding to the largest eigenvalues of XXT to obtain a d-dimensional linear motion model {circumflex over (α)}p, in which a centerline model for a new image at phase p+1 is estimated by adding {circumflex over (α)}p to each centerline control point of a previous frame at phase p.

The invention discloses a target tracking method based on target and camera motion modeling. The target tracking method comprises the following steps: S1, predicting the real speed of a target at thecurrent moment through a filter algorithm; S2, predicting the speed of the camera at the current moment; S3, synthesizing the real speed of the target predicted in the step S1 and the speed of the camera predicted in the step S2 to obtain a predicted value of the resultant velocity of the target in the pixel plane at the current moment, and selecting a target search area according to the predictedvalue of the resultant velocity of the target in the pixel plane; and S4, carrying out a target tracking algorithm in the target search area selected in the step S3, and obtaining the position of thetarget in the pixel plane at the current moment. According to the target tracking method based on target and camera motion modeling provided by the invention, the accuracy and robustness of target tracking are effectively improved.

The invention discloses a panoramic dance motion modeling method and a dance teaching auxiliary system, belongs to the technical field of computers, and particularly relates to a panoramic dance motion modeling method. The method comprises the following steps: establishing a virtual human body model and a skeleton model, and setting skeleton key points; configuring motion capture devices for a dancer, enabling key points on the motion capture devices to be matched with skeleton key points, and enabling the motion to be synchronous; acquiring shape and motion information of the dancer equipped with the motion capture devices; calculating performance parameters according to data transmitted and acquired by the motion capture devices, and deploying the motion capture devices to work; performing visualization processing on the acquired movement information of the dancer, the virtual human body model and the skeleton model, and displaying and storing a virtual human body dance leader image on a visual terminal. According to the technical means based on virtual reality, dance motions can be comprehensively and completely displayed, and dance teaching and dance learning are facilitated.

The invention relates to the field of robot motion modeling and parameter identification methods, in particular to the field of motion modeling and parameter identification methods based on an airportrunway detection robot. A motion modeling and parameter identification method for the airport runway detection robot comprises the following steps of setting preconditions of motion modeling for theairport runway detection robot and establishing a relevant coordinate system; establishing a kinematic model and a transfer matrix of the airport runway detection robot by adopting a coordinate transformation method; establishing a newton dynamical equation of the system; solving the relation between an input voltage and an wheel angular velocity according to the established newton dynamical equation, and identifying unknown parameters of the robot system model by adopting a least square method; and establishing a final robot motion system dynamics model by adopting a lagrangian method in combination with the identification result. The motion modeling method for the airport runway detection robot can achieve the purpose of establishing a more accurate system model and has the characteristics of being accurate in modeling and the like.

The invention discloses a crowd motion modeling simulation method in a queuing scene. The method comprises the steps of scene modeling, emotion infection modeling and crowd queuing simulation. A queuing scene is constructed, a composition module and crowd part categories are determined, emotion parameters and personality parameters are given to agents participating in queuing, a cellular automatonmodel is adopted as a dynamic model of queuing crowd behaviors, and behavior modeling of the agents is achieved through the external environment, the psychological states of the agents and the physical states of the agents. According to the method, the influence of influence factors of individual emotion fluctuation on behaviors of individuals in crowd queuing is built through modeling, administrator emotion parameters are further added, a queuing crowd disorder prevention model is built, the simulation method is stable and effective, and a vivid queuing crowd disorder prevention simulation result is well generated.