252 results about "3d localization" patented technology

The invention discloses a power transformer partial discharge positioning method based on multi-sensor array fusion, and the method comprises the following steps: S1, selecting a sensor installation point, and laying a multi-sensor array structure; s2, partial discharge signals of the three types of sensors are collected, and primary signal processing is carried out; s3, calculating propagation time differences between the reference channel and other channels by adopting a generalized cross-correlation weighting algorithm, and generating a time difference matrix; s4, constructing a TDOA model in combination with the layout coordinates and the time difference matrix, and solving three-dimensional initial coordinates of a power supply; s5, establishing a structure correction model, compensating the path deviation, and outputting corrected positioning coordinates; s6, calculating an error and generating a confidence score; s7, mapping a positioning result to the three-dimensional model and generating an image; and S8, writing the positioning information into a database for filing management. According to the method, the multi-frequency sensor and the path correction model are fused, and high-precision three-dimensional positioning of partial discharge of the transformer is realized.

The invention relates to the technical field of image processing, and discloses a vehicle surface defect identification method and system based on artificial intelligence. The method comprises the following steps: acquiring a vehicle surface image through polarization filtering multi-angle imaging; processing the standardized image by applying HSV color conversion and texture equalization; constructing three-dimensional deformation features in combination with parallax analysis, and extracting defect feature vectors; and identifying defects by using the position-aware convolutional neural network, and outputting a severity thermodynamic diagram. According to the invention, in a complex illumination environment, tiny defects of vehicle surfaces with various colors and materials are accurately identified and evaluated, and accurate three-dimensional positioning and severity quantification of the defects are realized at the same time.

A localization system comprises: a device; a master unit which wirelessly transmits a first localization signal; a plurality of lateration units distributed about the area within which the device is being localized, wherein each lateration unit of the plurality independently starts its own timer upon its receipt of the first localization signal; and a localization unit. The device receives the first localization signal and responsively wirelessly transmits a second localization signal. Each of the lateration units: independently receives the second localization signal; stops its respective timer responsive to receipt of the second localization signal; and wirelessly transmits a timer count signal to a localization unit. The timer count signal identifies the transmitting lateration unit and a count of its respective timer. The localization unit utilizes the plurality of timer along with respective distances between the master unit and the lateration units to localize the first device via time-of-flight lateration.

The invention discloses an unmanned aerial vehicle monocular vision forest fire positioning method, and aims to realize rapid detection, accurate positioning and three-dimensional reconstruction of an early fire. According to the method, a comprehensive fire monitoring system is constructed by integrating various sensors such as an unmanned aerial vehicle platform, a visible light camera, an infrared camera and an inertial measurement unit (IMU). Through monocular vision simultaneous localization and mapping (SLAM) technology and IMU-GNSS information fusion, the attitude of the camera is recovered in real time. And carrying out three-dimensional positioning on a fire point by utilizing a triangulation method and combining semantic information of visible light and infrared images. According to the method, the SfM (Structural From Motion) technology is adopted, the attitude of a camera and image feature points are combined, and the attitude of the camera and the three-dimensional point cloud are optimized through GBA (Global Bundle Administration), so that the positioning precision is improved. The method has the characteristics of high efficiency, accuracy, real-time performance and remarkable cost effectiveness.

A localization system comprises: a device; a master unit which wirelessly transmits a first localization signal; a plurality of lateration units distributed about the area within which the device is being localized, wherein each lateration unit of the plurality independently starts its own timer upon its receipt of the first localization signal; and a localization unit. The device receives the first localization signal and responsively wirelessly transmits a second localization signal. Each of the lateration units: independently receives the second localization signal; stops its respective timer responsive to receipt of the second localization signal; and wirelessly transmits a timer count signal to a localization unit. The timer count signal identifies the transmitting lateration unit and a count of its respective timer. The localization unit utilizes the plurality of timer along with respective distances between the master unit and the lateration units to localize the first device via time-of-flight lateration.

The invention discloses an artificial intelligence-assisted standard artificial temporomandibular joint automatic placement method. The method comprises the following steps: training a three-dimensional reconstruction model used for extracting a mandible and zygomatic bone zygomatic arch joint prosthesis placement area; based on an improved 3D U-Net neural network, carrying out segmentation extraction on the interested areas of the glenoid fossa and the mandibular branch, and obtaining a space geometric structure for prosthesis installation; the method comprises the following steps: extracting outer bone surface point cloud data of a zygomatic arch and a mandibular ramus on the basis of a space geometric structure of prosthesis installation, identifying anatomical key points, and generating a standard prosthesis fitting surface through geometric fitting and constrained optimization; determining the relative position relation between the maxillary joint fossa prosthesis and the mandibular branch prosthesis, and determining the installation direction, the hole position definition and the assembly reference point; and determining a contact area, a posture direction and a placement position of the prosthesis, and completing three-dimensional positioning and posture matching of the prosthesis in the patient model as a surgical planning scheme.

The invention belongs to the technical field of unmanned aerial vehicles, and particularly relates to an intelligent aerial 3D printing and spraying system and method based on an unmanned aerial vehicle cluster, and the system comprises a spatial three-dimensional laser ranging scanning modeling module for generating a real-size scene model, a three-dimensional positioning network module, and a three-dimensional positioning network composed of a plurality of laser radar arrays. Wherein the laser radar or the wireless positioning beacon array is arranged on a working plane to construct a three-dimensional space coordinate reference system; the printing module is integrated on the unmanned aerial vehicle, compared with traditional high-altitude operation, the construction efficiency is improved, the constraint of tedious procedures such as scaffold building and manual high-altitude operation is thoroughly eliminated, the unmanned aerial vehicle cluster can be rapidly deployed in a short time, and operation tasks such as large-area single-color or full-color spraying, repairing, modeling and printing are efficiently completed with high precision; the project construction period is greatly shortened, the comprehensive construction cost is reduced, the safety construction level is improved, the engineering income is improved, and the construction accident risk is reduced.

The invention provides an anti-shielding head three-dimensional positioning method and system based on instance segmentation, and relates to the technical field of image processing. Inputting the low-resolution video frames into the feature extraction network to obtain shallow features and deep features; inputting the deep feature into a feature fusion detection network to obtain a fusion feature and a low-resolution detection result comprising a sheltered head instance and an unsheltered head instance; inputting the shallow features and the fusion features into a feature decoding segmentation network to obtain a low-resolution feature map and a low-resolution head segmentation mask; based on the high-resolution video frame, the non-occlusion head instance, the occlusion head instance, the low-resolution feature map and the low-resolution head segmentation mask, obtaining a high-resolution non-occlusion head instance segmentation result and an occlusion-removed head instance segmentation result; and inputting a result into a three-dimensional positioning module to obtain a three-dimensional coordinate of the head. Through multi-network processing, feature fusion and other operations, high-precision human head segmentation is realized, and three-dimensional positioning errors are reduced.

The invention provides a three-dimensional positioning method and system for a satellite-borne multi-base radar system based on weighted linear iteration, and the method comprises the steps: obtaining radar echo data of a plurality of receivers, carrying out the effective detection of a target, and carrying out the searching of a pitch angle and a space taper angle of a target signal through a two-dimensional adaptive filtering technology, obtaining two-dimensional angle estimation information of the target at different bases and different visual angles; constructing a nonlinear equation set according to two-dimensional angle information and slope distance information of the target under different visual angles, and performing first-order Taylor expansion on the nonlinear equation set to obtain a corresponding linear overdetermined equation set; and forming a precision matrix according to the known measurement precision of the radar system, and carrying out normalization operation and iterative solution on the linear overdetermined equation to obtain the three-dimensional coordinate of the target in the earth-centered earth-fixed coordinate system, thereby obtaining the latitude, longitude and height information of the target, and realizing the high-precision three-dimensional positioning of the multi-static radar system.

The invention provides a three-dimensional scanning processing method and device based on prior data and computer equipment, and relates to the technical field of three-dimensional scanning, and the method comprises the steps: obtaining an initial reference model and three-dimensional positioning data of a scanned object when the scanned object is subjected to three-dimensional scanning; wherein the initial reference model is prior data used for representing object surface information of a scanned object; aligning the initial reference model with the three-dimensional positioning data to obtain a target reference model; and performing three-dimensional scanning processing on the scanned object based on the target reference model to obtain surface information of the target object. When the scanned object is subjected to three-dimensional scanning processing, the prior data representing the object surface information of the scanned object is introduced, so that the scanning precision and the scanning efficiency can be improved by utilizing the prior data, the complexity of a scanning data processing algorithm is reduced, and the user experience is improved.

The invention discloses a dual-machine target positioning method based on semantic prior and spatial intersection constraint. The method comprises the following steps: 1, acquiring images of a camera 1 and a camera 2, and determining the spatial orientation of a camera coordinate system relative to a reference world coordinate system; step 2, obtaining a target normalized coordinate under a coordinate system of the camera 1; step 3, obtaining a target normalized coordinate under a coordinate system of the camera 2; 4, converting the normalized coordinates of the target into geometric rays in a three-dimensional space; 5, constructing an optimization function with the minimum distance sum of squares; step 6, performing semantic-level foreground and background division on the image content through a YOLOv5 network, and constructing a three-dimensional space feature point library of a semantic background; and 7, performing single-view semantic neighborhood retrieval and depth compensation by using the three-dimensional space feature point library of the semantic background constructed in the step 6, and realizing target three-dimensional positioning in a high-altitude long-distance complex scene. According to the invention, the detection robustness and the positioning precision of the high-altitude long-distance target in the environment are improved.

The invention discloses a digital twinning-based three-dimensional visual management system and a digital twinning-based three-dimensional visual management method for an anode assembly workshop. The system comprises a triple binding index module, a unified time axis alignment module, a view cone inverse solution positioning module, a safety linkage arrangement module and a resume playback module, and is provided with a process template library and a network isolation / main / standby module. Three-dimensional positioning and PTZ linkage of an AI / threshold event are realized by establishing mapping of a three-dimensional object, a camera and a PLC point location; the batch / bracket / station events, the PLC time sequence and the video slices are indexed in a unified mode; shadow verification, interlocking check, instruction issuing, read-back and video consistency confirmation and trace leaving are executed according to the process, and a safe closed loop from warning to disposal is formed. The method comprises the steps of access and calibration, triple binding, time axis alignment, event positioning and stream taking, linkage processing and resume filing. The positioning and handling efficiency can be improved, the misoperation risk is reduced, and cross-production-line reuse and tracing evidence obtaining are supported.

The invention discloses a physical object three-dimensional positioning method and system based on video data, and relates to the technical field of three-dimensional positioning. The method is used for solving the problems of low object space positioning precision, unstable pose estimation and poor time sequence continuity in a video scene. Firstly, an input video stream is analyzed, an object segmentation mask is generated, feature points are extracted, camera motion parameters are calculated through inter-frame matching, and a scene sparse three-dimensional point cloud is reconstructed; a candidate three-dimensional bounding box is generated according to the segmentation mask and the point cloud, an optimal bounding box is selected by combining geometric matching degree and feature similarity evaluation, and a preliminary three-dimensional positioning result of the object is obtained; a positioning result and historical frame motion data are fused, a space-time constraint optimization model is constructed, and the six-degree-of-freedom pose of the object is solved; and finally, neural radiation field representation is established based on the pose, the pose and neural parameters are optimized through combination of micro rendering and back propagation, continuous and accurate updating of the three-dimensional position of the object is realized, and the positioning stability and robustness in a dynamic scene are improved.

The present invention provides a visual guidance system for a mobile grasping device, comprising a data acquisition and preprocessing module, the data acquisition and preprocessing module being used to acquire image data and preprocess the acquired image data, and transmitting the preprocessed image data to an instance segmentation module, the instance segmentation module being used to identify and segment a target object in the preprocessed image data, and outputting the pixel coordinate position and mask information of the target object to a three-dimensional positioning module, the three-dimensional positioning module being used to calculate the spatial position of the target object, and transmitting the spatial position information of the target object to an execution module, which accurately guides the grasping of the target object through the spatial position information of the target object. The visual guidance system for a mobile grasping device in the present invention can more accurately perform spatial positioning of the target object, and due to the use of technologies such as deep learning, instance segmentation, and spatial point cloud, the time required for the visual guidance system to adapt to the mobile grasping device is shorter.

The invention provides a precise workpiece positioning method and system based on three-dimensional vision, and the method comprises the steps: obtaining a scene three-dimensional vision data set containing a target workpiece, carrying out the multi-modal data fusion processing of the scene three-dimensional vision data set, generating a fusion data set with aligned space coordinates, extracting a workpiece feature identifier set from the fusion data set, and carrying out the recognition of the workpiece feature identifier set. And feature matching association processing is carried out on the workpiece feature identifier set and a pre-stored standard workpiece feature template, a feature matching corresponding relation set is generated, three-dimensional positioning coordinate information of the target workpiece in the scene is calculated based on the feature matching corresponding relation set, and the three-dimensional positioning coordinate information comprises workpiece centroid coordinates and space attitude angle parameters. The precision of precision workpiece machining is improved, and the stability of the positioning process is ensured.

According to the plug cord injection molding on-line real-time detection method, an industrial camera array is integrated at key nodes of an injection molding machine, technological parameters of mold clamping force and screw rotating speed are synchronously collected, and a comprehensive signal collection system is constructed. By developing a defect recognition algorithm and combining a copper sheet three-dimensional positioning model and a plastic filling dynamic simulation database, micro-offset detection and early warning of bubble formation are realized. An image feature extraction and quantification technology and a three-dimensional reconstruction and micro offset extraction method are utilized to carry out high-precision analysis on the injection molding process. And a parameter adjustment mapping relation is established based on the defect type and severity, an injection molding parameter PID controller is developed, and online correction of the mold locking force and the screw rotating speed is achieved.

The invention relates to the technical field of GIS partial discharge high-precision three-dimensional positioning, and discloses a multi-stage iteration GIS partial discharge high-precision three-dimensional positioning method, and the technical scheme is characterized in that S1, an electromagnetic simulation model is constructed based on a GIS equipment structure, and the installation position of an ultrahigh frequency sensor is optimized through electromagnetic simulation; s2, performing wide-band adaptive sub-band reconstruction and multi-resolution feature enhancement processing on the acquired signals; s3, calculating three-dimensional focusing energy distribution by adopting a multi-band time inversion method, and extracting candidate discharge positions; s4, screening candidate positions; s5, performing multi-stage hybrid optimization on the screening result to obtain refined three-dimensional positioning coordinates; and S6, outputting a three-dimensional positioning result, and performing performance evaluation based on a confidence index. According to the method, a time delay consistency screening mechanism is introduced, false candidate points which do not conform to an actual signal propagation rule are effectively eliminated, and the accuracy and reliability of a positioning result are remarkably improved.

The invention relates to the technical field of unmanned aerial vehicle detection and computer vision, in particular to an application method of a multi-scene throwing object recognition FOD algorithm carried by an unmanned aerial vehicle, and the method comprises the steps: dynamically fusing data through a multispectral camera and a laser radar, carrying out the scene self-adaptive defogging and illumination equalization preprocessing, inputting an improved twin network-FPN model, and extracting features; then, a target frame is screened by adopting a DIOU-NMS containing a scene complexity coefficient; and three-dimensional positioning is realized by combining RTK-GPS and a laser radar, the data is transmitted to a ground terminal through 5G and is subjected to secondary verification, and the models can be dynamically switched according to the flight height. The method solves the problems of missing and false detection, poor environmental adaptability and the like of complex scene FOD recognition, data integrity is improved through multi-sensor fusion, robustness is enhanced through scene self-adaption, the model and dynamic strategy balance precision and real-time performance are improved, multi-scene adaptability is high, and the recognition effective rate exceeds 98%.

The invention discloses a 3D digital asphalt paving system and a paving quality monitoring method. The 3D digital asphalt paving system comprises an asphalt paving machine and a 3D paving control system, the 3D paving control system comprises a three-dimensional positioning system, a signal acquisition system and a vehicle-mounted control system; the three-dimensional positioning system comprises a data control system, a GPS base station, a GPS moving station, a laser receiver and a laser transmitter. The signal acquisition system comprises a vibration sensor mounted on the screed, a data line connected with the vibration sensor, a data memory and a signal processing module; the vehicle-mounted control system comprises a vehicle-mounted analyzer, a three-dimensional quality parameter processing module, a quality parameter calculation module, a slope sensor and a handheld controller; the output end of the data control system and the output end of the quality parameter calculation module are connected to the three-dimensional quality parameter processing module, and the three-dimensional quality parameter processing module and the handheld controller are both connected with the vehicle-mounted analyzer. The problems that the construction efficiency is low, the influence of human factors is large, and paving quality control lags behind are solved.

The invention discloses an underwater robot deep learning adaptive data processing method and system, and relates to the technical field of deep learning. Sonar data, a visual image and a robot motion state of an underwater environment are collected; enhancing the definition of a visual image by using a convolutional neural network, marking autonomous moving elements, and constructing a two-dimensional coordinate system to complete preliminary positioning; the element depth is calculated based on the pixel value and the time delay of the sonar data, and the two-dimensional coordinates are fused to construct a three-dimensional coordinate system; and unifying the track in the three-dimensional space, calculating the distance between the elements and the robot, and planning the moving route and speed. The system comprises a data acquisition module, a visual processing module, an acoustic fusion module and a path planning module, and realizes multi-source data integration, three-dimensional positioning and path planning. Through visual and acoustic data fusion and a deep learning model, the underwater environment stereoscopic perception precision and the path planning efficiency are improved, and the adaptability and the operation safety of the underwater robot in a complex water area are enhanced.

The invention relates to a personnel identification and behavior detection method and device and a security robot, and the method achieves the whole-process intelligent processing from threat perception to precise intervention through a multi-module cooperative working mechanism and an innovative algorithm design. According to the method, the problem of balance between threat detection precision and real-time performance in the prior art is solved, and accurate three-dimensional positioning and tracking of a threat target are realized through effective integration of depth information. According to the method, through technologies of deep learning algorithm optimization, three-dimensional space positioning and the like, the threat detection capability, the response speed and the intervention accuracy of the security robot in a complex environment are comprehensively improved, powerful technical support is provided for high-risk scene safety prevention and control, and the method has remarkable practical value and popularization prospects.

Disclosed herein are systems and methods related to a capsule endoscopy, where a patient can swallow an ingestible capsule that records images of digestive tract, and a new in-body positioning system can precisely localize the capsule's position. Implementations include a new frequency-division multiplexing-based magnetic localization (FDMML) approach which leverages a higher frequency carrier in the low MHz range. The approach significantly reduces the reference excitation coil sizes and decreases the required excitation current by three orders of magnitude compared to prior work, making it practical for wearable systems. A fully integrated wireless receiver prototype is implemented in 180 nm bulk CMOS and packaged in an ingestible pill form factor. The new scheme achieves the best experimentally demonstrated tracking accuracy in both 2D and 3D localization experiments, achieving a sub-mm mean absolute position error and consuming only 247 pW while running at 100% duty cycle.

The invention provides a WiFi and UWB fused positioning method, and relates to the field of indoor positioning, a data set is constructed and preprocessed through a WiFi-RTT technology and a UWB-TWR technology, the data set is processed by fusing a fingerprint positioning regression model and a Hessian-PSO three-dimensional positioning algorithm, accurate positioning of a to-be-measured point is realized, and the accuracy of the to-be-measured point is improved. And further optimization is carried out based on an HMMK joint algorithm to obtain an optimal state sequence, so that the problems of low positioning precision, poor positioning stability and the like in a complex dynamic environment are solved, and the positioning precision and the positioning stability are effectively improved.

The invention provides an automatic calibration method and system based on joint point searching of a total station and a camera, and the method comprises the steps: respectively building a two-dimensional coordinate system and a three-dimensional coordinate system with an industrial camera and the total station as original points, and extracting the two-dimensional positioning information of a to-be-monitored mark point from a two-dimensional image shot by the industrial camera; then calculating three-dimensional positioning information of point location data obtained by dotting relative to three-dimensional coordinates of the industrial camera by using an Euclidean distance formula, then performing information matching on the two-dimensional positioning information and the three-dimensional positioning information to obtain two-dimensional-three-dimensional positioning information pairing of the same mark point to be monitored, and then performing two-dimensional-three-dimensional positioning information pairing according to the two-dimensional-three-dimensional positioning information pairing. And solving a pose parameter of the industrial camera, back-projecting the three-dimensional positioning information to the two-dimensional image, calculating to obtain a back-projection error, and if the back-projection error is smaller than an error threshold value, converting the two-dimensional positioning information into a three-dimensional physical coordinate in a three-dimensional coordinate system according to the pose parameter, thereby completing two-dimensional-three-dimensional joint calibration.

A three-dimensional positioning and motion-detecting system for human body and object status in mixed reality and a method thereof are disclosed. A virtualization computing device performs gesture recognition for scaling, moving or 3D rotating a virtual posture model, to make posture model aligning points of the virtual posture model correspond to the posture model positioning points one by one, a virtual-and-real display device displays the adjusted virtual posture model in real time, when the virtual posture model is fixed, the virtualization computing device recognizes a hand position on the physical environment image, and detects a motion on the hand position based on the posture model positioning points, to generate operating position information and hand motion information. The virtual-and-real display device displays virtual-and-real integration of operating position information, hand motion information and a physical space.

A method for in-vehicle localization of a mobile device includes receiving, in real time, sensor data from a plurality of UWB sensors inside a vehicle. The plurality of UWB sensors includes a plurality of UWB anchors and a UWB tag, which is part of the mobile device. The method further includes determining a plurality of location candidates of the UWB tag based on the sensor data received, determining a plurality of sensed current locations of the UWB tag and a plurality of probabilities for each of the plurality of sensed current locations of the UWB tag using a Gaussian Kernel Density Estimation (KDE), tracking a motion of the UWB tag, and determining a real-time position of the UWB tag using a Bayesian estimation.

An automatic end effector tip position initialization mechanism for a micromanipulation system is provided. The method includes: S1, positioning a macro end effector tip, calculating a camera pose by using ArUco Tag, and realizing a three-dimensional positioning of the end effector tip by using a triangulation method, and moving the end effector to a area near a center of a focal plane based on a visual servo; S2, positioning a micro end effector tip; S3, acquiring calibration data based on a visual servo, calibrating the end effector tip and the petri dish, and obtaining a left calibration matrix and a right calibration matrix reflecting the transformation relationship between left and right micromanipulator coordinates and image coordinates; generating an intuitive human-computer interaction interface based on a mouse and a keyboard according to acquired position information. The method has obvious advantages in terms of operation accuracy, operation efficiency and repeatability.

The invention discloses a single-sensor type robust high-precision three-dimensional positioning method and system, and aims to improve the positioning precision and stability of a moving target in a three-dimensional scene. According to the method, a geometric constraint structure formed by a multi-label sensor is constructed, rich observation information is fused, and simultaneous acquisition of three-dimensional position and attitude information is realized. The optimal layout coordinates of a base station sensor and a label sensor are calculated by adopting a particle swarm optimization algorithm, and high-precision three-dimensional position and attitude calculation is realized by combining a weighted least square algorithm of single-point positioning, a weighted least square algorithm of rigid body positioning, robust M estimation, gross error detection and a Kalman filtering backward smoothing algorithm. According to the method, in a complex environment of non-line-of-sight signal interference, the positioning robustness can be remarkably enhanced, the elevation error is reduced, and the application feasibility and the precision reliability of a single type of sensor in an actual scene are improved.

The invention discloses a hidden ore body three-dimensional positioning prediction method and system based on artificial intelligence, and relates to the technical field of hidden ore body positioning prediction. Geological information variables and three-dimensional data of known ore bodies are collected and preprocessed; dividing the geological space range into a plurality of cubic information units, and obtaining geological information variables in each unit and a space position relation of a known ore body; analyzing the geological information variables by using a neural network algorithm, and calculating to obtain relevance and weight values among the geological information variables; according to the relevance and the weight value, calculating a comprehensive evaluation value of each cubic information unit, and delineating a metallogenic prediction target region; through comparison and verification with actual exploration data, a neural network algorithm model is continuously optimized and adjusted. Based on the artificial intelligence algorithm, the three-dimensional positioning prediction precision and the prospecting effectiveness of the concealed ore body are improved. The invention provides an effective method for objectively determining the weight, and the reliability and the accuracy of a prediction result are improved.

The present invention relates to a systems and methods to perform 3D localization of target objects in point cloud data using a corresponding 2D image. According to an illustrative embodiment of the present disclosure, a target environment is imaged with a camera to generate a 2D panorama and a scanner to generate a 3D point cloud. The 2D panorama is mapped to the point cloud with a 1 to 1 grid map. The target objects are detected and localized in 2D before being mapped back to the 3D point cloud.