160 results about "Motion correction" patented technology

The invention discloses a multi-frame photoacoustic image reconstruction method based on optical flow alignment and depth feature fusion. The method comprises the following steps: S1, obtaining photoacoustic signal data; s2, reconstructing a photoacoustic cross-sectional image; s3, performing optical flow calculation and image alignment; and S4, training the deep feature fusion network. According to the method, through optical flow motion correction and a potential space learning mechanism, space-time information and complementary features in the aligned multiple frames of images are dynamically integrated, complementary information in the multiple frames of images is adaptively fused, noise is suppressed, and finally reconstruction of high signal-to-noise ratio and high spatial resolution images of biological tissues is achieved. Experimental results show that the method can significantly improve image quality, recover image distortion and detail loss caused by motion and noise, and provide a new effective scheme for promoting robust clinical application of a photoacoustic imaging technology.

The invention discloses a space-time speckle projection three-dimensional imaging method based on multi-frame optical flow alignment. Firstly, a projector based on DLP is used for projecting a space-time speckle pattern to a measured scene, and a binocular camera synchronously collects a three-dimensional space-time speckle image. The calibration parameters of the binocular camera are used to carry out stereo correction on an acquired original speckle image, and a parallax image is generated frame by frame in combination with a coarse-to-fine single-frame speckle matching strategy. And estimating a two-dimensional inter-frame displacement field between continuous disparity maps by using an optical flow method by taking an intermediate frame disparity map as a reference, compensating motion artifacts in a space-time speckle image, and ensuring strict space-time registration of a dynamic target. And based on the speckle image after motion correction, a speckle matching strategy is expanded to a time-space domain, and high-precision multi-frame three-dimensional measurement of a complex dynamic scene is realized. The method is suitable for performing rapid and high-precision three-dimensional modeling on a moving target in an unstructured environment, and can perform accurate three-dimensional measurement on a high-speed dynamic target undergoing any translation or rotation motion.

The invention provides a bridge stay cable video multi-target identification tracking and vibration extraction method based on an unmanned aerial vehicle. The method comprises the following steps: 1, constructing a bridge stay cable refined inclined slender target detection model based on a YOLOv11 model; 2, providing a multi-target tracking algorithm fusing the inclined slender target detection model and a StrongSORT algorithm; step 3, improving a displacement extraction method combining SIFT / ORB feature point matching and a sub-pixel refinement technology; 4, designing an unmanned aerial vehicle motion correction algorithm based on variational mode decomposition and time-frequency domain combined screening; and 5, constructing a joint working modal analysis algorithm for realizing combination of a natural excitation technology and a random subspace recognition algorithm. According to the method, high-precision extraction of the vibration signals of the stay cable and identification of modal parameters of the stay cable are realized, and technical support is provided for health monitoring of the large-span cable-stayed bridge.

The invention provides virtual reality-based dance teaching equipment and a teaching method thereof, and the equipment comprises a VR module which is used for providing multi-view immersive visual experience; the motion capture module is used for capturing the joint angle of the dancer in real time and the limb track feedback module and is used for simulating the ground counter-acting force in dancing through vibration or pressure, enhancing motion perception and providing a touch prompt for motion correction; the virtual scene module is used for creating a dance virtual scene; and the action analysis module analyzes the joint action deviation based on multiple view angles, generates an instant score according to the joint angle deviation and the rhythm synchronization rate, and records historical data to generate a progress curve. According to the method, the problems of resource shortage, feedback lag and other pain points in traditional dance teaching are solved, efficient, personalized and interesting dance learning experience is achieved through multi-modal interaction and a dynamic adjustment mechanism, and the application of the method is expected to promote dance education to develop in the intelligent and universal direction.

The present disclosure provides systems and methods for automated scan preparation and real-time monitoring / adjustment in medical imaging. The automated scan preparation may include positioning a target subject to be scanned by a medical imaging device, determining a rotation scheme of the medical imaging device, targeting a scan region of the target subject with an imaging isocenter of the medical imaging device, determining target position(s) of component(s) of the medical imaging device, performing a virtual scan, generating a reference subject model representing an internal structure of the target subject, or the like. The real-time monitoring / adjustment operations may include achieving automatic brightness stabilization, monitoring a posture of the target subject, adjusting the position of component(s) of the medical imaging device, estimating a dose distribution, monitoring a treatment of the target subject, performing a motion correction, or the like.

The invention belongs to the technical field of clock synchronization, and particularly relates to a high-precision clock synchronization method in an aviation cluster high-speed relative motion scene, which comprises the following steps of: S1, acquiring round-trip time RTT data based on signal transmission among nodes in an aviation cluster, and acquiring a relative radial speed v by combining an inertial navigation module; calculating by using a motion correction formula to obtain a preliminary clock skew epsilon after motion correction; in order to solve the problems of asymmetric high-speed motion path time delay, URT interference, random and nonlinear error coexistence and the like, through a four-stage cooperative scheme of RTT motion correction, URT modeling compensation, Kalman filtering smoothing and BP neural network residual optimization, the clock synchronization precision is improved to 0.1-0.3 ns, the method has strong anti-interference robustness and engineering realizability, and the method is suitable for large-scale popularization and application. The method can be widely applied to aerospace, unmanned aerial vehicle formation, satellite communication and other fields requiring high-precision clock synchronization, and effectively solves the problems of insufficient precision and poor adaptability of a traditional method in a high-speed scene.

The invention relates to a multi-modal sensing fusion-based automatic correction method and system for a welding robot body, the system comprises a three-level distributed architecture of a sensing layer, an edge layer and a control layer, the sensing layer is integrated with a millimeter wave radar, a camera, a six-dimensional force / torque sensor, a welding power supply data acquisition module and an acoustic emission sensor, and multi-modal data acquisition is realized; the edge layer runs a fusion algorithm through an embedded AI computing platform carrying a GPU to complete data processing and decision making; and the control layer receives the deviation correction instruction to drive the robot to execute motion correction. The method comprises the steps of multi-modal data real-time acquisition and synchronization, data preprocessing and feature extraction, multi-source information fusion and decision making, and real-time control and execution. The problems of single sensing dimension, response delay, weak anti-interference capability and the like in the prior art are solved, the robustness, the real-time performance and the deviation correction precision of the system are improved, full-automatic intelligent welding is realized, and the method is suitable for high-precision scenes such as shipbuilding and pressure vessel welding.

The invention discloses a gated image adaptive noise suppression and detail enhancement method based on depth guidance, and the method comprises the steps: carrying out the noise decoupling of an original intensity image and depth map data based on depth-noise joint physical modeling, and generating a noise parameter set and a depth feature map; performing time sequence jitter compensation by using rigid constraint of the depth feature map, and generating a motion correction image through depth weighted interpolation; performing depth Fourier coefficient guided filtering processing on the motion correction image in a frequency domain, separating a base layer and a detail layer, and suppressing noise; according to the depth value, semantic regions are divided, differential enhancement is executed, and a region customization enhancement image is generated; and performing ambient light compensation based on the depth attenuation model, and outputting a final optimized image. According to the method, the image quality can be remarkably improved, the depth-of-field contradiction caused by global enhancement is solved, and the ambient light residue is radically treated.

A method is provided for acquiring a magnetic resonance image dataset of an object, using an imaging protocol in which several k-space lines are acquired in one echo train, wherein the echo train may include one or several sub echo trains, and wherein the imaging protocol includes a plurality of echo trains. Within the method, a set of additional k-space lines within a central region of k-space is acquired a number of times per echo train, wherein the number is greater than 1, in order to detect movement of the object during the echo train. The sets of additional k-space lines are acquired at pre-determined positions within the echo trains. The disclosure is also directed to a method for generating a motion-corrected magnetic resonance image dataset from the dataset thus acquired, a magnetic resonance imaging apparatus, and a computer program.

The invention discloses an anti-jitter data acquisition and obstacle recognition method for a rail transit vehicle, and relates to the technical field of rail transit vehicle data acquisition and processing, continuous four frames of original images are acquired through a binocular camera by adopting synchronous and staggered shooting, and global motion estimation is performed in combination with acquired sensor data; local pixel level alignment is realized by using a preset image layer and pixel grid division, a stable image is generated through weighted fusion, local features of the image are optimized based on a multi-head attention mechanism, the stable image is mapped to a high-dimensional vector space by using a depth encoder, motion correction information is fused, and finally, the motion correction information is obtained. A deep reconstruction network is combined with binocular parallax to generate a three-dimensional point cloud, three-dimensional coordinate changes are monitored in real time, obstacle recognition and early warning are achieved, high-quality images can be obtained in the violent vibration environment of a vehicle, the data acquisition precision and the real-time performance and accuracy of obstacle detection are improved, and the operation safety of rail traffic is effectively guaranteed.

The invention belongs to the technical field of image processing, and discloses a three-dimensional radar sub-aperture imaging deformation monitoring device and method, and the device comprises a radar which is used for collecting echo signals passing through a preset position, processing the echo signals according to frequency mixing and analog-to-digital conversion, obtaining a three-dimensional echo matrix, and outputting the three-dimensional echo matrix; the three-dimensional echo matrix and a sampling ending signal are sent to a motion compensation module; the motion compensation module is used for resampling the azimuth direction of the three-dimensional echo matrix by adopting an interpolation algorithm; the distance matching filtering module is used for performing pulse compression on the distance direction of the three-dimensional echo matrix; the migration correction module is used for carrying out migration correction on the distance direction of the three-dimensional echo matrix after pulse compression; the sub-aperture imaging module is used for calculating a three-dimensional echo matrix after migration correction according to a sub-aperture imaging algorithm; and the deformation monitoring module is used for performing interference difference processing on the three-dimensional complex image to obtain deformation data. The imaging quality and the deformation monitoring precision can be improved.

The present description relates to an image sequence motion correction method and apparatus based on intensity correction optical flow registration. The method comprises: determining a reference image of each deformed image in a deformed-image sequence, and performing motion correction on the deformed image by means of adjusting the weight of a mechanical regularization term between the reference image and the deformed image, so as to obtain a displacement field of the deformed image relative to the reference image; on the basis of the displacement field and an intensity correction value of a finite element node, performing intensity correction on the reference image; adjusting the weight of the mechanical regularization term, performing motion correction on the deformed image, performing intensity correction on a first corrected reference image on the basis of an optimized displacement field, repeatedly executing the process until an image residual function converges, determining that the deformed image has been corrected, and recording the optimal solution of a weight value of the mechanical regularization term of the deformed image; and on the basis of the optimal solution of the weight value, determining the actual displacement of the deformed image. In the present description, the mechanical model behind physiological motion and an intensity correction technique are integrated into an image registration algorithm to reduce image noise, thereby improving the algorithm accuracy.

A method for acquiring a magnetic resonance image dataset of an object includes using an imaging protocol in which a number of k-space lines are acquired in one shot. The imaging protocol includes a plurality of shots. A plurality of additional k-space lines are acquired in at least a subset of the shots, such that movement of the object is detected throughout the imaging protocol. A method for generating a motion-corrected magnetic resonance image dataset from the dataset thus acquired, a magnetic resonance imaging apparatus, and a computer program are also provided.

The invention relates to a radar data correction method and device, computer equipment and a storage medium. The method comprises the steps of determining an installation deviation angle of an external meteorological station according to a fitting result of an azimuth angle and a radial wind speed in multiple groups of first radar data measured by a continuous wave radar in a preset period; according to the installation deviation angle, symbol information of radial wind speed in each group of first radar data is determined; and performing motion correction on the first radar data according to a preset turbulence correction algorithm and inertia measurement data measured by an inertia measurement unit to obtain target radar data. The method can improve the accuracy of wind direction and wind speed measurement results.

Examples are disclosed that relate to motion blur corrections for time-of-flight (ToF) depth imaging. One example provides a depth camera comprising a ToF image sensor, a logic machine, and a storage machine storing instructions executable by the logic machine to receive depth image data from the ToF image sensor, the depth image data comprising phase data and active brightness (AB) data, determine a first two-dimensional (2D) AB image corresponding to a first modulation frequency, and determine a second 2D AB image corresponding to a second modulation frequency. The instructions are further executable to determine a 2D translation based upon a comparison between the first 2D AB image and the second 2D AB image, determine corrected phase data based on the 2D translation to form corrected phase data, perform phase unwrapping on the corrected phase data to obtain a three-dimensional (3D) depth image, and output the 3D depth image.

The invention relates to the technical field of security and protection monitoring, in particular to an AI-based face recognition security and protection method, which comprises the following steps: recognizing a channel based on monitoring an entrance and exit, collecting and analyzing a multispectral face image, screening feature points and skin color areas, optimizing spatial feature matching and structural distribution, and fusing head actions and face behavior parameters. And judging structural consistency and linkage features, and outputting a face recognition mapping result. According to the method, facial detail level discrimination is realized through multi-spectral information fusion, parallel matching of continuous spatial features and time sequence behavior parameters and synchronous screening of behavior linkage and facial structure features are carried out based on regional pixel distribution and a structure sequence in combination with dynamic spatial trajectory analysis and abnormal motion correction, and therefore, the accuracy and the reliability of the method are improved. According to the multi-parameter joint recognition strategy in the dynamic scene, interference caused by environment change and human intervention is reduced, the reliability of target identity judgment in a complex place is improved, and the practicability and the real-time recognition capability for the multi-source scene are enhanced.

The invention discloses a sonar image geometric correction method based on image optimization, and relates to the technical field of underwater sonar imaging, and the method comprises the steps: collecting a multi-source underwater observation data packet including sonar image original data, platform motion state data and underwater environment data, and carrying out the preprocessing; extracting sonar image original data from the multi-source underwater observation data packet, performing geometric coarse correction to obtain a pre-corrected sonar image, and performing motion compensation on the obtained pre-corrected sonar image to generate a motion corrected sonar image; constructing a joint optimization objective function, carrying out geometric fine tuning on the motion correction sonar image by adopting a sliding time window strategy, and outputting a geometric fine correction sonar image; geometric distortion caused by non-uniform motion of a platform, attitude disturbance and acoustic propagation distortion is effectively eliminated, inter-frame consistency and submarine topography continuity of sonar images are remarkably improved, and the method is suitable for high-precision operation scenes such as underwater surveying and mapping, target recognition and three-dimensional reconstruction.

The invention belongs to the technical field of artificial intelligence and medical rehabilitation, particularly relates to an intelligent training optimization system for orthopedic rehabilitation, and aims to solve the problems of insufficient individuation, feedback lag, subjective evaluation and the like in traditional rehabilitation training. The system collects motion and physiological data through a wearable sensing unit, and generates and adjusts a training scheme in combination with a self-adaptive optimization algorithm driven by cloud individualized modeling, dynamic risk assessment and reinforcement learning; real-time motion correction is achieved through multi-modal feedback, doctor-patient cooperation and group intelligent analysis are supported through a clinical management platform, and rehabilitation safety, effectiveness and accessibility are improved.

The embodiment of the invention provides an angular velocity correction method and device and a robot, and relates to the technical field of motion correction, and the method comprises the steps: determining an angular velocity zero offset according to motion data measured by an inertial measurement unit, and taking the angular velocity zero offset as a zero offset observation value; determining the weight of the zero offset observation value and the weight of the zero offset prediction value based on the discrete degree information of the historical zero offset observation value and the uncertainty information of the zero offset prediction value; updating the zero offset predicted value through the zero offset observed value, the zero offset predicted value, the weight of the zero offset observed value and the weight of the zero offset predicted value; and correcting the angular velocity included in the motion data by adopting the updated zero offset predicted value. The angular velocity zero offset adopted for correcting the angular velocity is determined based on the motion data measured by the inertial measurement unit instead of a pre-calibrated fixed angular velocity zero offset, so that the accuracy of the corrected angular velocity can be improved.

In a case where motion information of a subject (3) during scanning is acquired from an image pair, and a CT image is generated by performing motion-corrected image reconstruction, erroneous recognition of a motion component and excessive smoothing are suppressed to improve accuracy of motion correction. A motion information acquisition unit (330) includes a filtering unit (331) that reduces noise of an image pair, a registration unit (333) that performs registration processing by using the image pair after filtering, and a processing condition adjustment unit (335) that adjusts conditions of filtering processing and the registration processing in accordance with a tube current ratio of the image pair. The processing condition adjustment unit refers to a table in which a relationship between the tube current ratio and a parameter of each processing is predetermined, selects parameters corresponding to a tube current ratio of a target image pair, and determines the parameters for filtering and registration. The motion information acquisition unit acquires motion information by performing filtering and registration by using the determined parameters, and the image reconstruction unit performs motion-corrected image reconstruction by using the motion information.

Disclosed herein is a method of magnetic resonance imaging. The method comprises receiving (200) groups of k-space data descriptive. The groups of k-space data comprise group specific imaging k- space data and group specific navigator k-space data. The method further comprises determining (202) group specific subject positions using the group specific navigator k-space data of the groups of k- space data. The method further comprises initializing (204) group specific motion parameters of a joint optimization using the group specific subject positions. The method further comprises reconstructing (206) a motion corrected magnetic resonance image by performing the joint optimization on the group specific imaging k-space data of the groups of k-space data. The joint optimization is configured to simultaneously reconstruct the motion corrected magnetic resonance image while determining the group specific motion parameters. The method further comprises providing (208) the motion corrected magnetic resonance image descriptive of the field of view.

Described herein is a medical system (100, 300) comprising a memory (110) storing machine executable instructions (120) and a motion estimating neural network (122, 700, 800, 900, 1000) configured for outputting trajectory data (130) in response to receiving a trial motion trajectory (128) as input. The execution of the machine executable instructions causes a computational system (104) to: receive (200) measured k-space data (124) descriptive of a subject (318); perform (202) motion estimation of the subject between the sequence of discrete acquisitions by solving an optimization problem to determine a calculated motion trajectory of the subject in the predefined coordinate system, wherein the optimization problem is modified using the trajectory data; and reconstruct (204) a final motion corrected magnetic resonance image (136) from the measured k-space data and the calculated motion trajectory in the predefined coordinate system.

Examples are disclosed that relate to motion blur corrections for time-of-flight (ToF) depth imaging. One example provides a depth camera comprising a ToF image sensor, a logic machine, and a storage machine storing instructions executable by the logic machine to receive depth image data from the ToF image sensor, the depth image data comprising phase data and active brightness (AB) data, determine a first two-dimensional (2D) AB image corresponding to a first modulation frequency, and determine a second 2D AB image corresponding to a second modulation frequency. The instructions are further executable to determine a 2D translation based upon a comparison between the first 2D AB image and the second 2D AB image, determine corrected phase data based on the 2D translation to form corrected phase data, perform phase unwrapping on the corrected phase data to obtain a three-dimensional (3D) depth image, and output the 3D depth image.

The invention discloses a dynamic tracheal stenosis diagnosis system based on dynamic DR combined with deep learning, which relates to the technical field related to medical image intelligent analysis and comprises a dynamic DR imaging module, a time sequence data processing module, a deep learning analysis module and a visual report module. The invention further discloses a tracheal stenosis dynamic diagnosis method based on dynamic DR combined with deep learning. The tracheal stenosis dynamic diagnosis method comprises the steps of dynamic DR imaging, data processing, deep learning analysis and report generation. According to the method, 3D convolution and time sequence modeling are fused, real-time segmentation and tracking of a dynamic image are achieved, the performance is superior to that of a traditional 2DCNN or an independent time sequence model, data collection is optimized through a pulse perspective mode, respiratory gating and motion correction are combined, double challenges of dynamic image quality and algorithm robustness are solved, key data in a structured report can be filed in batches, and real-time segmentation and tracking of the dynamic image are achieved. A standardized database is provided for tracheal stenosis cause distribution, a thermodynamic diagram and a curve graph generated by the system provide high-quality annotation data, and iterative optimization of an algorithm is promoted.

This invention relates to a precise motion correction control system and method for an unmanned workboat, belonging to the field of unmanned vessels. The system comprises a workboat hull, an accelerometer, a spirit level, and a central processing unit installed on the hull. The method comprises the following steps: S1: initialization; S2: receiving client instructions to perform operations; S3: real-time sensing of the hull state using the installed accelerometer, spirit level, and pressure sensor; S4: force analysis of the hull based on empirical formulas and the sensed hull state, calculating the theoretical action of the water pump; S5: selecting a candidate experience pool that meets the error limit requirements using a KNN network; S6: calculating the optimal action using a DQN network; S7: sending corresponding control instructions to the water pump according to the optimal action and executing them; S8: repeating steps S3 to S7 until the operation is completed. This invention can effectively reduce the action solution space, reduce the complexity of the solution, and improve control accuracy, thereby enhancing the adaptability of unmanned workboats to complex environments.

Disclosed herein is a method of medical imaging and an associated medical system. The method comprises: receiving (200) k-space data (114, 114') by subjecting at least a portion of a subject (318) to a magnetic resonance imaging protocol, wherein the k-space data comprises sequential groups (500, 502, 504) defined based on acquisition time of lines of k-space data, wherein navigator lines for one pair of adjacent sequential groups are different from the navigator lines (510, 512, 514) of at least one other pair of adjacent sequential groups; calculating (202) a motion detection metric (116) for the adjacent pairs of the sequential groups using the respective navigator lines of k-space data; combining (204) adjacent pairs of the sequential groups if the respectively calculated motion detection metric is below a predetermined threshold to reform the sequential groups; and reconstructing (206) a motion corrected magnetic resonance image (122) from the reformed sequential groups (118) of k-space data using a joint optimization algorithm (120) that simultaneously provides the motion corrected magnetic resonance image and motion transformations between the reformed sequential groups, and wherein the motion corrected magnetic resonance image is at least partially reconstructed using the navigator lines of k-space data.

The embodiment of the invention provides a multi-target tracking method and device based on camera motion correction, equipment and a medium, and the method comprises the following steps: estimating a camera motion matrix through a previous frame gray level image I1 and a current frame gray level image I2 containing a detection frame; through Kalman prediction, prediction frames of all tracking targets are obtained, and the prediction frames are corrected based on the camera motion matrix; performing one-to-one association matching on all the corrected prediction frames of all the tracking targets and all the detection frames of the current frame gray level image I2 to generate an association matching result; and dynamically adjusting Kalman filtering parameters according to the updated tracking frame information and the detection frame information matched with the updated tracking frame information, and taking the prediction frame as the updated tracking frame information at the k moment for each tracking target which fails in association matching. According to the scheme, the prediction result of Kalman filtering is corrected through the camera motion matrix, and the efficiency and accuracy of multi-target tracking are improved.

A method, system, processing circuitry, and computer program product for providing initial motion correction in magnetic resonant imaging (MRI) data that enables additional image correction to be performed on subsequently processed MRI data in the same imaging set. One such method receives k-space data including a first set of motion corrupted k-space data and a second set of k-space data (different than the first set); generates motion correction data based on the first set of motion corrupted k-space data; and generates an image based on the second set of undersampled k-space data and the motion correction data.

The invention discloses a cardiac magnetic resonance report generation method and system based on a multi-dimensional tissue feature knowledge graph. The method comprises the following steps: performing motion correction and multi-view feature extraction on a cardiac magnetic resonance multi-sequence image to obtain an original numerical attribute of an instance node; constructing a dynamic knowledge graph containing tissue characteristics and medical priori, and generating a target reasoning chain with a clinical basis in combination with neural symbol reasoning; then, in a controlled generation stage, an inference chain is injected into a prefix tree, in a decoding step, mask physical shielding logic conflict description is utilized, and a pointer network is synchronously started to directly copy original numerical value attributes into a text sequence; and finally, executing consistency logic verification through a text restoration technology, and identifying and intercepting factual errors. Through the atlas constraint and numerical unvarnished transmission mechanism, the factual illusion problem of a generative model is effectively solved, and the accuracy and reliability of a heart diagnosis report are remarkably improved.

An exemplary AI system and method are disclosed for reconstructing high-fidelity motion-corrected MRI scans from low-field, motion-corrupted MRI scans (e.g., neonatal MRI data) using a neural network (NN) model in combination with a physics-based motion model. The exemplary AI system and method can enhance image quality and remove motion artifacts which can allow for scans to be performed with significantly reduced scan time, a particular benefit for neonatal subjects as well as all patients when employing low-field MRI.