23 results about "Spatial calibration" patented technology

This invention provides a method for fish passage behavior recognition based on a multimodal large model. First, imaging sonar image sequences and underwater optical video sequences of fish passing through fish passages or facilities are simultaneously acquired, and the multimodal data are aligned through time stamp synchronization and spatial calibration. Then, fish target detection, segmentation, and feature extraction are performed on the sonar image sequences and optical video sequences respectively to obtain the fish's spatial location, body length, water depth, motion state, morphological structure, and posture features. The obtained acoustic and optical features are input into a multimodal coding and fusion model based on the Transformer architecture, and a unified fish behavior representation vector is constructed through a cross-modal attention mechanism. During the training phase, model parameters are adjusted through a multi-task learning approach using cross-entropy loss, mean squared error loss, and contrastive learning loss, ultimately achieving fish passage behavior category recognition and prediction of individual and group passage difficulty scores.

The application discloses a polyp segmentation method based on SAM boundary enhancement and uncertainty guidance, aiming at the problems of high pixel-level labeling cost, blurred polyp boundary, difficult cross-scale feature alignment and poor effect of SAM direct migration to a medical scene in the polyp segmentation task under enteroscopy, fusing the hierarchical feature extraction ability of Res2Net-50, the general segmentation ability of SAM, combining core strategies such as boundary enhancement, uncertainty guidance and dynamic prompt generation, and realizing high-precision polyp weakly supervised segmentation under extremely sparse scribble labeling. The application significantly improves the spatial calibration accuracy of the model in a low-contrast environment, eliminates noise interference, and effectively solves the problem of supervision imbalance caused by the large size difference of lesions in an endoscope image.

The application relates to the technical field of orthopedic surgery navigation, and discloses a high-position tibial osteotomy navigation method, a system and a storage medium, and a communication method which comprises the following steps: generating a 3D bone model of a lower limb of a patient based on medical images and performing preoperative osteotomy planning; performing visual positioning system detection and spatial calibration of a surgical instrument; installing an optical marker array, collecting actual bone point cloud data, and realizing alignment of the model and the bone through a registration algorithm; in-surgery real-time calculation and display of deviation of an operation track and planning parameters; post-surgery reconstruction of a lower limb force line, and comparison and evaluation with preoperative planning and in-surgery navigation records. The system comprises a hardware execution unit and preoperative planning, calibration verification, in-surgery registration, real-time navigation and post-surgery evaluation modules. The application realizes full-process integration of preoperative accurate planning, in-surgery real-time closed-loop navigation and post-surgery quantitative evaluation, and significantly improves the precision, repeatability and predictability of clinical effects of surgery.

The application relates to the technical field of magnetic resonance image analysis, in particular to a DWI-ADC value dynamic atlas construction method and system for bone infection, which comprises the following steps: acquiring DWI and ADC data to establish an image index, screening an anatomical reference point to construct a lesion spatial relationship set, extracting a diffusion value to construct an ADC time sequence set, analyzing a trend to generate a segmented time label, dividing data to generate a node according to the label, and establishing a bone infection DWI-ADC dynamic atlas structure. In the application, parameter deviation is eliminated by image checking consistency and index structure establishment, a stable reference point of bone anatomy is used as a spatial calibration anchor point to overcome soft tissue edema interference, a lesion relative spatial relationship set is established to solidify coordinates, a lesion position is derived to ensure the anatomical identity of longitudinal comparison sampling parts, a lesion pathological turning signal is captured by arranging an ADC change trend label set, discrete data is converted into a continuous pathological model by constructing a dynamic atlas node sequence, and the bone infection evaluation accuracy is improved.

This invention relates to the field of infrastructure engineering technology, specifically to a machine vision-based method and system for locating spraying areas. The method includes: using a depth camera to capture the original point cloud of the beam surface, segmenting and selecting target planes on the beam surface using the RANSAC algorithm; converting the image to two dimensions, extracting line features using Hough transform and edge detection, inversely mapping back to three dimensions, and selecting the uppermost point of the line; moving the camera upwards to capture the point cloud of the upper part of the beam, removing interference to extract two target planes; calculating the normal vectors of the two planes, comparing the point cloud to obtain the coordinates of the calibration points, constructing a new three-dimensional coordinate system and generating a rotation matrix; and planning the robot spraying path based on the beam surface design parameters. The system includes: a depth camera, point cloud segmentation, two-dimensional mapping and feature extraction, three-dimensional coordinate recovery and key point selection, spatial calibration, and path planning modules, which work together to implement the above method, eliminate point cloud blind spots, improve positioning accuracy, and meet the requirements for precise spraying of beam surfaces.

This invention relates to the field of electromagnetic compatibility technology in power systems, and more particularly to a rail position control and suppression system based on multi-objective optimization. The invention generates a target potential distribution prediction scheme through a potential distribution prediction generation module, which then corrects the scheme using a spatial calibration unit based on real-time train position offset to generate spatial dynamic correction coefficients. The spatial optimization unit then optimizes the scheme based on historical soil resistivity drift data. Simultaneously, a target drainage action timing scheme is generated through a timing scheme generation module, corrected by a timing calibration unit based on real-time rail potential deviation to generate timing dynamic correction coefficients, and then optimized by a timing optimization unit based on historical equipment performance data. Finally, a comprehensive point-based control module achieves coordinated spatial and temporal control.

This invention proposes a spatiotemporally synchronized visualization measurement and analysis method, system, and device for flow-heat-phase. The method includes: acquiring a multimodal sensing dataset and performing spatial calibration and alignment; processing the aligned high-speed visual image to obtain a binary phase diagram, combining temporal information to identify and segment it, dividing and merging the liquid phase region, gas phase region, contact line region, and rewetting region to obtain a partitioned result map; identifying and fusing the infrared temperature image, binary phase diagram, and partitioned result map to output a spatiotemporally synchronized visualization measurement phase diagram for flow-heat-phase. By acquiring data to form a dataset and calibrating and aligning it, this invention achieves, for the first time, a precise spatiotemporally synchronized correlation of gas-liquid distribution, temperature field, and heat flow information; combining binary phase diagram generation, intelligent segmentation, and multimodal data fusion technology, it overcomes the limitations of traditional methods and can quantitatively analyze key physical processes under the interaction of multiple bubbles on complex surfaces; the final output phase diagram can reveal the coupling mechanism between the temperature field of the micro-region of the heated surface and the dynamic behavior of bubbles.

The application relates to the technical field of medical devices and computer-aided surgical navigation, and discloses a virtual positioning CT-guided lung biopsy method, which utilizes real-time acquired body surface point cloud data to drive a biomechanics model, deduces a displacement field of lung internal grid nodes, and generates a virtual lung model which changes with respiratory motion. Through a fiber grating sensing array embedded in a puncture needle, the three-dimensional shape of the needle body is solved in real time, and an axial force strain is separated. The system monitors the first derivative characteristics of the strain, and when a mechanical event representing penetration of an anatomical boundary is identified, a spatial deviation vector of the physical coordinates of the needle tip and the virtual predicted coordinates is calculated. A local space calibration field is constructed by using a radial basis function, the virtual lung model is non-rigidly corrected based on the deviation vector, and a dynamic navigation image is generated. The application improves the positioning accuracy of lung puncture navigation by combining body surface deduction and puncture mechanical feedback correction model prediction error without increasing the radiation dose.

The subject application presents an automated medical procedure system featuring accurate spatial calibration and registration capabilities. The system comprises a calibrated robotic subsystem with a position-accurate arm, a tracking subsystem, and processor-based modules for registration and calibration. Key innovations include the use of factory-defined geometric configurations for targets, systematic spatial recalibration before each procedure, and the ability to perform both rigid point-to-point and point cloud registrations. These features enable the system to maintain accurate spatial relationships between components and patient anatomy, allowing for highly accurate and adaptable automated medical procedures across various specialties.

This invention discloses a twin birth simulation teaching system based on the linkage of mixed reality and physical models, belonging to the field of medical teaching simulation technology. The system includes a multimodal sensor fusion acquisition module, a mixed reality spatial calibration and virtual-real overlay module, and a twin birth physiological dynamics simulation module. By deploying a thin-film pressure sensor array, an inertial measurement unit, and a capacitive proximity switch, the system collects operational data in real time. It uses a weighted iterative nearest point algorithm to achieve sub-millimeter-level spatial alignment between virtual and real data. The system drives the physiological simulation of twin birth based on a finite element contact mechanics model. It combines an expert rule base and a machine learning model to generate a quantitative evaluation report, realizing a closed-loop teaching of twin birth based on visual and tactile data.

This application relates to a method and device for environmental perception and 3D mapping based on multi-sensor fusion. The device performs unified time synchronization and spatial calibration on multi-source perception data acquired by LiDAR, inertial measurement unit, image acquisition unit, and satellite navigation system. It then performs multi-sensor data fusion processing on the main control computing platform to obtain temporally and spatially consistent environmental perception information, achieving high-precision positioning and 3D environmental mapping for the device itself. Specifically, the time synchronization module uniformly triggers or aligns timestamps during the multi-sensor data acquisition process, and determines the spatial pose relationships between the sensors through extrinsic parameter calibration. During device movement, LiDAR acquires 3D point cloud data of the environment, and combines this with motion constraints provided by the inertial measurement unit, visual information acquired by the image acquisition unit, and absolute position information provided by the satellite navigation system to construct a device motion model and an observation model. A fusion algorithm is used to jointly optimize the multi-source sensor data to obtain the optimal pose estimation results of the platform at each moment, and the environmental spatial information is cumulatively updated to gradually construct a 3D environmental map. Based on the fusion and mapping results, a three-dimensional reconstruction and visualization of the surrounding environment structure is achieved, and a modular mechanical structure provides reliable support for the mobile platform's environmental perception, positioning, and spatial understanding.

This invention provides a microwave radar-camera spatial calibration method and system based on pose joint optimization, comprising: Step S1: constructing a calibration scene, in which a radar, a camera, and a calibration reference are arranged; the radar is rigidly connected to the camera; Step S2: defining a radar coordinate system and constructing an optimization problem, solving the optimization problem to obtain the optimal radar pose, and adjusting the radar; Step S3: acquiring images of the calibration reference through the camera, obtaining the pixel coordinates of the calibration reference, solving for the optimal pose of the camera, and adjusting the camera; Step S4: transforming the coordinates of the radar and the camera, outputting the final radar-camera spatial calibration result, and reducing errors. This invention, by introducing pitch angle constraints, creatively solves the ill-posed problem of pose estimation for linear antenna array radars, providing a practical solution for the precise pose calibration of this type of sensor.

This invention discloses a method for detecting defects on track surfaces, comprising the following steps: Step 1: Multi-source data acquisition: acquiring three-dimensional point cloud data and two-dimensional texture images of the track surface through a multi-source data acquisition system mounted on a detection vehicle, maintaining spatial calibration between the sensor coordinate system and the track coordinate system during acquisition; Step 2: Data preprocessing: denoising, downsampling, and normal vector estimation are performed on the point cloud data, and distortion correction, illumination equalization, and region of interest extraction are performed on the image. Spatiotemporal registration of the point cloud and the image is achieved through feature point matching; Step 3: Defect feature extraction: the registered multi-source data is input into a YOLOv5-DeepLabV3+ hybrid neural network to simultaneously extract the geometric and texture features of the defects. This invention's multi-source data fusion compensates for the shortcomings of a single sensor, greatly improving the accuracy of detecting micro-cracks and micro-peeling; the entire process is automated, reducing manual intervention and supporting detection under nighttime and complex lighting conditions.

This invention relates to the field of visual inspection and discloses an online inspection method for the appearance quality of traditional Chinese medicine pills. The method includes controlling a multi-ring LED light source to perform asynchronous structured light field modulation and synchronously triggering a camera to acquire a two-frame high-light sequence image; extracting the high-light connected components of the images and refining and reducing their dimensionality to obtain an odd-even ring high-light skeleton network; addressing the outermost high-light features of the skeleton network for spatial calibration, obtaining the instantaneous physical center coordinates of the spheres, calculating the kinematic displacement transformation matrix, and reconstructing the dynamic reference curvature distribution function; calculating the local actual curvature of the skeleton lines and comparing it with the reference curvature to extract candidate distortion nodes; performing an inverse affine transformation based on the displacement transformation matrix, and performing multimodal orthogonal verification on the distortion nodes in a unified coordinate system to distinguish defect types and filter out environmental noise. This invention overcomes the influence of dynamic production line mechanical translation errors, pill size tolerances, and environmental dust interference, improving the accuracy of surface morphology defect identification.

This application relates to the field of gimbal technology, and provides a method, device, gimbal, and storage medium for automatic supplementary lighting during gimbal tracking and shooting. The method includes: performing temporal alignment and spatial calibration fusion on the real-time acquired raw image data stream to obtain a color point cloud sequence; performing target detection and illumination resampling on the current frame point cloud data in the color point cloud sequence to obtain target environment light field data; performing difference comparison and inverse solution of illumination compensation on preset illumination target distribution data and the target environment light field data to generate a supplementary lighting control parameter set; and performing supplementary lighting according to the supplementary lighting control parameter set. This application constructs a color point cloud sequence, performs spherical resampling and anchor point matching on the target area to form target environment light field data, and then combines this with a preset illumination target distribution to perform inverse solution to generate supplementary lighting control parameters, ultimately achieving precise automatic supplementary lighting during gimbal tracking and shooting.

The application relates to an image reconstruction method, a magnetic resonance imaging method and a computer device, which comprises the following steps: acquiring K-space calibration data sets corresponding to a target part, the K-space calibration data sets being fully sampled in a center region of K-space; acquiring a plurality of undersampled K-space data sets corresponding to the target part, each undersampled K-space data set being data collected in one excitation of the target part; based on the K-space calibration data sets, performing first fitting recovery on unsampled points in the center region of K-space of each undersampled K-space data set to acquire a plurality of intermediate K-space data sets; performing second fitting recovery on unsampled points in a non-center region of K-space of the plurality of intermediate K-space data sets to acquire a plurality of target K-space data sets; and reconstructing the plurality of target K-space data sets to acquire magnetic resonance images corresponding to multiple excitations of the target part. In this way, the magnetic resonance images of the target part generated according to the target K-space data sets are of better quality.

Line-field optical coherence tomography (OCT) systems and methods provide enhanced imaging accuracy through k-linearization of interferometric data. The system comprises a swept-frequency laser source, a line-field sensor, and a single-board computer (SBC) with processing capabilities. The system uses a frequency reference to generate a periodic reference pattern detected by a subset of pixels in the line-field sensor. The SBC determines a resampling curve based on the reference pattern to correct for nonlinearities in the laser's wavelength tuning. Applying the resampling curve to k-linearize the interferometric data enables the generation of high-resolution transform-limited depth profiles via inverse Fourier transform. Methods for periodically, continuously, or adaptively recalculating the resampling curve based on a linearity threshold are disclosed. The system can also dynamically update the laser's tuning function to ensure consistent performance. These advancements enable accurate and efficient OCT imaging for applications such as ophthalmology, angiography, and other diagnostic uses.

The application discloses a fusion-based adaptive pupil interaction method and belongs to the technical field of pupil positioning, and the steps of the method comprise the following steps: HOG feature extraction cooperates with an SVM model to predict key points, lock the iris position and calculate the centroid; based on the screen ratio of a display, a spatial calibration model with the same ratio as the display screen is constructed, interpolation algorithms are used to optimize pupil data, the corrected coordinate data is projected to the display, the actual position of the line-of-sight point in the screen is determined, the parameter vector of the model is iteratively updated through an L-BFGS-B optimization algorithm, the coordinate data is iteratively optimized through an HBG algorithm after mapping, and the fixation center point is determined. The application has the beneficial effects that: in view of the uncertainty of the pupil position, a calibration model is constructed by using device parameters, the data is finely calibrated by using interpolation and optimization algorithms, the error is greatly reduced, the prediction deviation is further reduced through the iterative optimization of the HBG algorithm, and the system has the advantages of simple structure, high lightweight degree, high positioning accuracy and fast response speed.

This application provides a guidance measurement system for acoustic wave drilling measurement and control, comprising: acquiring a sequence of reflected acoustic signals; extracting initial waveform features from the sequence; processing the features using Fourier transform to obtain a frequency domain distribution spectrum for subsequent anomaly detection; performing multi-path propagation simulation on the reflected signal path using a changing timestamp sequence to obtain a propagation time difference matrix for interface geometry estimation; extracting a set of spatial coordinate points from the optimized anomaly region boundary description; processing the point set using a least squares fitting algorithm to obtain position and shape parameters for real-time tracking and updating; updating the overall rock layer layer structure model using refined spatial calibration results; smoothing the model using a Kalman filter algorithm to obtain the final anomaly region location output.

This invention discloses an AR multi-person collaborative interactive projection system and method for public spaces, belonging to the field of AR multi-person collaborative interactive projection technology. It includes a multi-source spatial calibration module configured to extract static physical feature point clouds in the public space based on real-time environmental images and depth data collected by multiple AR devices, and to perform global registration of the feature point clouds using a weighted nearest neighbor iterative algorithm to generate a unified world coordinate system. This invention utilizes the multi-source spatial calibration module to perform global registration of static physical feature point clouds collected by multiple AR devices using a weighted nearest neighbor iterative algorithm to generate a unified world coordinate system, and detects the relative pose deviation between each AR device. When the deviation exceeds a preset threshold, a dynamic recalibration process is triggered, effectively solving the problem of inconsistent coordinate systems among multiple AR devices in public spaces, and significantly improving the spatial consistency and registration accuracy of multi-person collaboration.

The application relates to a multi-target motion synchronization detection method and system in group dance performance. The method comprises the following steps: performing spatial calibration and time synchronization on a multi-view video of group dance performance to obtain a standard multi-view video and camera calibration parameters; inputting the camera calibration parameters and the multi-view image into a skeleton coordinate extraction neural network to obtain a three-dimensional skeleton coordinate sequence and a three-dimensional skeleton motion sequence; extracting a dance action phase sequence based on the three-dimensional skeleton motion sequence, and calculating a phase synchronization index; mapping the three-dimensional skeleton coordinate sequence to a standard spatial coordinate system to obtain a spatially aligned three-dimensional skeleton coordinate sequence, and calculating a spatial coincidence index; integrating the phase synchronization index and the spatial coincidence index to generate a multi-target motion synchronization detection result. By using the method, the state phase difference of time synchronization can be quantified by extracting the instantaneous phase of the action, the spatial coincidence of the group can be quantified by calculating the posture alignment, and the action deviation can be accurately positioned.

This invention provides an automatic method for extracting the shock layer and boundary of a blunt-nosed body based on high-speed camera images. It utilizes high-frequency imaging of the shock tube or shock tunnel experiment process using a high-speed camera, maps pixels to physical coordinates using spatial calibration, selects the region of interest in the image sequence and calculates the light intensity time series, and automatically identifies the effective time window of the experiment through rate of change analysis. Within the effective time window, representative frames are selected or averaged, and after dark flat-field correction, a one-dimensional light intensity profile is extracted. The location of the shock layer boundary is determined using a combination of gradient maxima and relative brightness thresholds. Finally, sub-pixel-level fitting and physical quantity conversion of the continuous boundary are achieved through conformal piecewise cubic interpolation. This invention, by constructing an automated boundary extraction process based on high-speed camera images, enables real-time identification and visualization of the shock layer location, providing reliable support for flow field structure diagnosis and quantitative analysis of experimental data.

This application belongs to the field of Building Information Modeling (BIM) and digital preservation of cultural heritage, specifically involving a refined BIM modeling method for historical buildings based on multi-source data fusion. This method includes: collecting multi-source heterogeneous data such as 3D laser point clouds, UAV oblique photography, and survey drawings; performing spatial calibration and constructing a structured metadata database under a unified geographic coordinate and time benchmark; performing parametric BIM modeling based on building component classification and precision grading standards to realistically restore details such as the platform, brackets, and roof; simultaneously carrying out parametric reconstruction of the vegetation system and environmental integration; and finally importing the data into Unreal Engine, combining physical rendering, dynamic weather, day-night cycle, and AI interaction logic to construct a high-fidelity, highly immersive digital cultural scene. This application achieves a full-element, high-precision, and interactive digital twin of historical buildings, significantly improving the level of protection, display, and revitalization of cultural heritage.