22 results about "Rigid motion" patented technology

The invention relates to the technical field of face verification, and discloses a face identity verification data processing method based on dynamic feature extraction, which comprises the following steps: acquiring a continuous face video frame sequence and calculating a full-pixel instantaneous velocity vector to generate an original dense optical flow field, selecting rigid region anchor points to calculate a rigid affine transformation matrix and construct a theoretical rigid motion field, performing differential stripping on the theoretical rigid motion field from the original dense optical flow field, and extracting a non-rigid micro-motion residual field; mapping the non-rigid micro-motion residual field to a facial muscle topological grid to generate a time sequence feature tensor, and calculating a geodesic line distance between a covariance matrix of the time sequence feature tensor and a reference dynamic feature in a Riemannian manifold space; when the geodesic distance is smaller than a threshold value, verification is passed, through a rigid-non-rigid orthogonal decomposition mechanism, the special viscoelastic micro-motion and cooperation law of biological soft tissue is captured by utilizing a residual field, and the high-simulation mask is effectively defended.

The invention provides an anti-saturation fault-tolerant control method and device for a double-flexible-arm space robot, relates to the field of robot control, and solves the multiple challenges that in the prior art, a flexible mechanical arm faces continuous and difficult vibration suppression, an actuator is prone to failure, torque is limited, task time is constrained and the like. The dynamic state of the system is unstable; and the control capability is limited. The method comprises the steps that a hypothesis modal method and a momentum conservation theorem are combined to derive a double-flexible-arm space robot kinetic equation in a Lagrange form; a flexible system singular perturbation technology is utilized to obtain a slow-varying subsystem for representing rigid motion characteristics of a mechanical body and a quick-varying subsystem for representing residual vibration characteristics of a rod piece; for the slow-varying subsystem, designing a fixed-time anti-saturation fault-tolerant controller based on an extreme learning machine and a torque output function; and designing an anti-saturation controller based on state variable negative feedback for the quick change subsystem. The method is used in the robot control process.

The invention relates to the technical field of computer vision, and discloses a public resource transaction behavior intelligent supervision and early warning system based on deep learning, which utilizes kinetic energy gating and a speed potential energy field to accurately lock an object keeping rigid motion with a hand, and fundamentally avoids the risk that a static background is misjudged as an operation target; meanwhile, through reverse optical flow reconstruction and gradient direction structured residual verification, a real entity with a three-dimensional volume and a shadow virtual image are accurately distinguished by utilizing a material shielding effect, and interference of dynamic shadow and reflection is thoroughly eliminated; and finally, on the basis of a risk potential field and kinetic energy flux accumulation mechanism, microscopic pixel motion is converted into macroscopic physical action quantity, a substantive transport process along an illegal path is accurately quantified through space-time integration, random jitter and withdrawing actions are effectively filtered, and high-confidence intelligent early warning of high-risk transaction behaviors is realized.

The invention relates to the technical field of three-dimensional graphic processing, and discloses a lunar surface / deep space scene target reconstruction and editing method based on 4D Gaussian structuring. The method is particularly suitable for dynamic target modeling and editing in deep space exploration tasks, including detectors, robots and related equipment on the lunar surface, the Mars or other celestial bodies. The method comprises the following steps: synthesizing a plurality of camera visual angles through a diffusion model based on a monocular video of an object, and reconstructing static three-dimensional Gaussian point representation in a standard space; extracting a skeleton structure of the object based on the surface grid, wherein the skeleton structure comprises a skeleton node position and a topological relation; on the basis of the skeleton structure, each Gaussian point is connected with a plurality of skeleton nodes through a linear hybrid skin mechanism, and skeleton-driven rigid deformation is achieved; non-rigid deformation is compensated through feature extraction and a regression network; and in combination with the rigid deformation and the non-rigid deformation, rendering to generate a dynamic Gaussian point model which can change along with time and can be edited. According to the method, motion modeling of an object is explicitly split into rigid motion driven by a framework and non-rigid correction, so that the definition and interpretability of motion representation are greatly improved, and particularly, higher stability and control precision are shown when challenges such as weak texture, violent illumination change and high noise are faced in a deep space environment. According to the method, object behavior modeling in a detection task is more flexible, and the method can be widely applied to scenes such as task planning, target recognition and dynamic editing.

To provide a horological movement including a device that can simplify the coupling between a rigid element and an elastic element.SOLUTION: A method for coupling a rigid element 10 with a spring 16 includes the steps of: rigidly attaching a first end of the spring 16 to a support; and positioning the support and the rigid element 10 in an initial relative position. The method, in which a coupling member 20 carried by a second end of the spring 16 is located on one side of a stressing ramp 40 which is opposite a specific recess 36 in the rigid element 10 and for which the rigid element 10 and the support can undergo relative movement MR in a first direction D1 from the initial relative position, then includes a coupling step. Here, the relative movement in the first direction is applied between the support and the rigid element 10 so that the coupling member 20 comes to bear against the stressing ramp 40 and then follows the stressing ramp 40, which is configured to move the coupling member 20 towards a rotation axis 22 by stressing the spring 16.SELECTED DRAWING: Figure 3B

The embodiment of the invention provides a modeling and analysis method for suspension four-wheel positioning angle tolerance sensitivity analysis, and the method comprises the steps: obtaining the structural information of a suspension system and the geometric data of system parts, and building a multi-body dynamic model based on the structural information and the geometric data; replacing a rigid kinematic pair in the multi-body dynamic model with a bushing pair, and setting a rigidity parameter of the corresponding bushing pair according to the type of the replaced rigid kinematic pair; collecting process information of system parts, designing a hard point tolerance range based on the process information and an error range of the suspension system, and adjusting a pre-offset value of a bushing pair according to the hard point tolerance range; the rigidity parameters and the pre-offset values of different lining pairs are modified for simulation, simulation results of each time are recorded, and the simulation results are compared to determine the sensitivity influence factors of the four-wheel positioning angle.

The invention relates to the technical field of machine vision and image processing, and discloses a non-ferrous metal intelligent impurity removal analysis method based on machine vision, and the method comprises the steps: collecting and correcting the hyperspectral time sequence data of the melt surface; constructing a spectrum prior weighted optical flow energy functional to calculate an accurate motion field capable of retaining a material boundary; combining information of two physical dimensions of spectrum and motion to identify a region of interest which is different from a background melt; extracting multi-dimensional features such as rheology, relative motion and spectral stability representing rigidity, motion independence and material stability of the region for the region; and carrying out collaborative analysis based on the characteristics to judge whether the artifacts are impurities or dynamic fluid artifacts. Through multi-dimensional feature collaborative analysis and physical judgment logic simulation, the problem that targets are difficult to distinguish only by means of single visual features is solved, and the accuracy and reliability of machine vision in intelligent impurity removal analysis application of non-ferrous metals are remarkably improved.

The embodiment of the invention provides a robot non-sensing rigid-flexible coupling track control method and system, and relates to the technical field of robot control technologies. The method comprises the following steps: acquiring actual joint state information and motor input information of the robot; based on the actual joint state information and the motor input information, the real-time interaction force between the robot end effector and the external environment is determined; based on a preset rigid motion track, decomposing the real-time interaction force into an effective interaction force component in the tangential direction of the rigid motion track; determining an expected motion displacement along the rigid motion trajectory based on the effective interaction force component; the expected movement displacement is mapped to the rigid movement track, and the expected three-dimensional space position of an end effector of the robot is determined; and converting the expected three-dimensional space position into an expected joint angle of the robot, and performing driving control on the robot based on the expected joint angle. According to the invention, the problem of poor interaction effect is solved, and the effect of improving the man-machine interaction efficiency and precision is achieved.

The invention relates to a control method, device and equipment for reciprocating rescue of an elevating fire truck and a storage medium. The method comprises the following steps: planning a periodic reference trajectory of rigid motion of a boom system according to target point location information of a reciprocating rescue task based on preset motion constraints, and extracting motion state characteristic parameters from the periodic reference trajectory; on the basis of the instantaneous equivalent length of the boom and the instantaneous posture of the boom, obtaining an instantaneous inherent frequency estimated value by inquiring a pre-stored inherent frequency fitting data table; constructing a parameter pair according to the instantaneous motion frequency and the instantaneous inherent frequency estimation value, and generating a stability early warning signal of a corresponding grade according to the position of the parameter pair in a preset parameter unstable area map and the distance between the parameter pair and the unstable area boundary; and selecting a corresponding active control strategy according to the level of the stability early warning signal, and performing control adjustment according to the active control strategy to generate a final control instruction. By adopting the method, the movement speed and the operation efficiency of reciprocating rescue can be improved.

The method of snapshot compressive imaging three-dimensional Gaussian splatting reconstruction belongs to the field of computer vision and computational photography. A reconstruction network based on the hybrid architecture of stochastic gradient Langevin dynamics (SGLD) and explicit three-dimensional Gaussian representation is built. Then, multiple physical perception modules are combined to achieve the deep decoupling and reconstruction of the spatiotemporal information of dynamic scenes. Driven by the embedded decoupling deformation field, the deformation field contains a coarse-grained trajectory prediction branch and a fine-grained integral fitting branch: the coarse-grained branch combines low-frequency time coding with latent embedding features, enabling Gaussian primitives to capture global rigid motion trajectories and serving as a temporal regularizer; the fine-grained branch uses high-frequency time coding to predict non-rigid deformation and physically fits the motion blur stripes generated by temporal integration through anisotropic scale stretching. An adaptive density control strategy compatible with physics is introduced. Using the physical integral imaging simulation module, high-fidelity and multi-view consistent high-speed dynamic three-dimensional scene reconstruction results are obtained.

The invention relates to the technical field of underground structures, in particular to a replaceable node of a bearing-damping decoupling regulation and control mechanism and a working method. The replaceable joint comprises a supporting column, a bearing beam and a decoupling regulation and control component, and the decoupling regulation and control component is located between the supporting column and the bearing beam. The decoupling regulation and control component comprises a connecting beam, energy dissipation pieces and a connecting plate, the energy dissipation pieces are located on the upper side and the lower side of the connecting beam, the outer end of the connecting beam and the outer ends of the energy dissipation pieces are fixedly connected with the supporting column, the inner ends of the energy dissipation pieces are rotationally connected with the outer end of the connecting plate, and a sliding groove is formed in the connecting plate and arranged horizontally. The inner end of the connecting beam is connected with the sliding groove in a sliding mode, and the inner end of the connecting plate is rotationally connected with the bearing beam. According to the invention, bearing-damping decoupling is independently regulated and controlled, and the rigid motion of the support column is utilized to realize relative displacement between the energy consumption piece and the connecting plate to achieve deformation coordination, so that cooperative work is realized.

The application discloses a motion-robust non-contact heart rate detection method, an electronic device and a storage medium. The method comprises the following steps: firstly, extracting a pixel mean time sequence from a selected video face region of interest, including the selection and positioning of the region of interest, tracking the region of interest and extracting the pixel mean time sequence; secondly, removing non-rigid motion or rigid motion, including signal preprocessing, time delay, conversion from a single channel to a multi-channel, extraction of a bottom layer shared source, selection of a motion noise signal and a pulse signal; finally, heart rate estimation, Fourier transform of the pulse signal and corresponding operation to obtain a heart rate value. The application extracts a heart rate related signal from a video image, removes motion noise by using a denoising algorithm, detects the heart rate in a motion state, provides a new direction for non-contact heart rate detection research, and further promotes the application of the non-contact detection technology in a life scene.

The present application relates to the technical field of mechanical arm, and particularly relates to a flexible mechanical arm motion stability adaptive control method and system, the method comprising: decomposing a reference vibration signal composed of vibration data of each target joint of a target mechanical arm obtained in advance at a current time and within a preset time interval before the current time; determining rigidity motion coherence corresponding to each current signal component; reconstructing the reference vibration signal and determining a mutation vibration time corresponding to each target joint; screening out a vibration transmission starting joint; and adaptively controlling a torque at each target joint according to a purified vibration signal and the mutation vibration time corresponding to the vibration transmission starting joint and target joints on both sides of the vibration transmission starting joint. The present application improves the rationality of torque adjustment at the joints of the mechanical arm, thereby improving the motion stability of the mechanical arm.

The invention relates to the technical field of mechanical arms, in particular to a flexible mechanical arm motion stability self-adaptive control method and system.The method comprises the steps that reference vibration signals formed by vibration data, obtained in advance, of each target joint point of a target mechanical arm at the current moment and within a preset time interval before the current moment are decomposed; determining rigid motion coherence corresponding to each current signal component; reconstructing the reference vibration signal, and determining a sudden change vibration moment corresponding to each target joint point; a vibration transmission starting joint point is screened out; and according to the purification vibration signals and the abrupt change vibration moments corresponding to the vibration transmission starting joint point and the target joint points on the two sides of the vibration transmission starting joint point, self-adaptive control is conducted on the torque of each target joint point. The rationality of torque adjustment at the joint points of the mechanical arm is improved, and therefore the movement stability of the mechanical arm is improved.

The invention belongs to the technical field of industrial robot precision measurement and kinematics calibration, and particularly relates to a mobile robot base elevation precision calibration method based on a tail end joint. According to the method, a laser tracker and a robot tail end short chain motion strategy are combined, and local high-rigidity motion of a fifth shaft and a sixth shaft of a robot tail end joint is utilized; according to the calibration method, tail end actual measurement pose data under a laser tracker coordinate system and nominal kinematics data under a robot base coordinate system are subjected to alignment resolving through a decentration singular value decomposition algorithm. According to the method, the requirements of efficient data acquisition speed, extremely high parameter identification stability and final positioning precision for base calibration in the field rapid deployment of the mobile robot are met, and the problems in traditional full-axis linkage calibration, especially in the traditional full-axis linkage calibration are solved. The problems of large data dispersion and low precision caused by substantial amplification of basic joint angle measurement error and geometric parameter deviation through a long kinematic chain are solved.

A computing system and method can be used to render a 3D shape from one or more images. In particular, the present disclosure provides a general pipeline for learning articulated shape reconstruction from images (LASR). The pipeline can reconstruct rigid or nonrigid 3D shapes. In particular, the pipeline can automatically decompose non-rigidly deforming shapes into rigid motions near rigid-bones. This pipeline incorporates an analysis-by-synthesis strategy and forward-renders silhouette, optical flow, and color images which can be compared against the video observations to adjust the internal parameters of the model. By inverting a rendering pipeline and incorporating optical flow, the pipeline can recover a mesh of a 3D model from the one or more images input by a user.

The invention discloses a high-precision reconstruction method and system for a three-dimensional model of a sportsman, relates to the technical field of computer vision and three-dimensional modeling, and aims to solve the problems of dependence on mark points, inconsistent model topology and detail missing in the prior art. According to the system, a depth image and inertia measurement data are synchronously acquired through a multi-modal data acquisition module, deep fusion of vision and inertia information is realized through a time sequence feature fusion module, and a symbol distance value and a non-rigid motion deformation amount are output based on an implicit dynamic field construction module; topological consistency constraint is combined to ensure that model deformation accords with biomechanical characteristics, and finally, a high-fidelity model generation module extracts contour surfaces and corrects the contour surfaces frame by frame to generate a three-dimensional human body model sequence which is coherent in time sequence and has submillimeter-level details.

The utility model belongs to 3D printing technical field especially is high -speed 3D printing device of fused deposition technology of artificial limb orthosis manufacturing, in view of the existing existing 3D printing equipment exists big, high noise, Z axis rigid motion mechanism is easy to cause the influence to the quality of printing product etc. problem, present and propose the following scheme, including the box, the bottom fixed mounting of box has the base, the inside of base is used for the placement of multiple electrical parts, the side hinged of box has the movable door, the bottom inner wall fixed mounting of box has the printing platform, in the utility model, this artificial limb orthosis manufacturing high -speed 3D printing device, through the compact design of bottom integrated electrical bin effectively reduced the volume of equipment, through adopting the movement part lifting frame structure and four axle belt -guide rail cooperative drive system, combine spatial flexible constraint network, can improve Z axis positioning accuracy, effectively eliminate big height model layer grain defect, ensure the printing quality of product good.

The invention relates to the technical field of unmanned aerial vehicle vision monitoring, and discloses a displacement monitoring method and system based on unmanned aerial vehicle vision. According to the method, a space-time mode is separated from an unmanned aerial vehicle time sequence image to form a coupling descriptor, and the coupling descriptor is input into a network after time domain unwrapping to obtain a preliminary trend vector field; adaptive space sampling is driven by the field, and a multi-scale structure block sample is collected in an original image; identifying a rigid motion element and a nonlinear deformation element from the sample, and establishing a kinematics constraint chain between the rigid motion element and the nonlinear deformation element; and carrying out physical guidance iterative correction on the initial trend field in combination with the constraint chain to generate an optimized displacement field. And performing multi-modal confidence evaluation on the optimization field, fusing high and low confidence region information, and generating a final full-field displacement distribution diagram. According to the invention, through physical constraint and adaptive sampling, the accuracy, reliability and efficiency of displacement monitoring are improved.

A method for motion-compensated reconstruction of a target vessel by providing a primary 3D reconstruction of a volume including the target vessel, identifying an image of the target vessel and an image of a secondary vessel different from the target vessel in the primary 3D reconstruction, forward-projecting the image of the target vessel and the image of the secondary vessel onto forward projection images, erasing the image of the secondary vessel in the forward projection images, and generating a secondary 3D reconstruction from the forward projection images with automatic motion compensation.

The invention relates to a method for motion-compensated reconstruction of a target tube (13), comprising: providing (S1) a first 3D reconstruction of a volume having the target tube (13); identifying (S2) a mapping of a target tube (13) and a mapping of a second tube (14) different from the target tube in the first 3D reconstruction; forward-projecting (S3) the mapping of the target tube (13) and the mapping of the second tube (14) onto the forward-projected image; deleting (S4) the mapping of the second tube (14) in the forward projection image; a second 3D reconstruction is created (S5) from the forward projection image with automatic motion compensation. The invention further relates to an image processing device for motion-compensated reconstruction as well as to a computed tomography device, to a computer program product and to a computer-readable storage medium.

The invention discloses a self-supervised heart dynamic magnetic resonance imaging reconstruction method based on implicit neural expression and optical flow constraint, which does not depend on any external annotation data set, realizes self-supervised learning through joint optimization of an implicit neural expression network and a physical sampling model, remarkably reduces the cost and data dependence of model training, and improves the reconstruction efficiency. And the clinical expandability is relatively high. And meanwhile, in combination with a physical model, inter-frame motion consistency is enhanced. According to the invention, an optical flow constraint module is introduced, and motion field information between adjacent frames is used for modeling myocardial non-rigid motion, so that artifacts and jumps in time dimension are effectively inhibited, the time sequence smoothness and motion authenticity of a dynamic image sequence are improved, the intra-frame structure continuity is maintained, and the method adapts to various sampling strategies and imaging conditions; an efficient and extensible network structure and an optimization strategy are adopted, the model parameter quantity is small, the expression ability is high, and deployment and acceleration are facilitated.