519 results about "Model reconstruction" patented technology

The invention provides a three-dimensional building fine geometric reconstruction method integrating airborne and vehicle-mounted three-dimensional laser point clouds and streetscape images. The method comprises the following steps: (1) a quick modeling method based on airborne laser data; (2) combining the semantic segmentation framework of the vehicle-mounted point cloud and the image; and (3) amodel automatic enhancement algorithm fusing multi-source data. According to the method, airborne laser point cloud, vehicle-mounted laser point cloud and streetscape images are taken as research objects, model reconstruction, model enhancement and updating are taken as targets, joint processing of point cloud and image data of different platforms is realized, and the fusion potential of variousdata is fully mined. The final research result will perfect the fusion and fine modeling framework of vehicle-mounted-airborne data, promote the development of the point cloud data semantic segmentation technology, and serve the emerging application fields of unmanned driving and the like.

The invention relates to a 3D-coordinated power grid energy management system and a power grid evaluation and control method, which belongs to the field of scheduling automation of power system. The system comprises a power grid model reconstruction sub-system, a power grid safety evaluation and alarm sub-system, an active frequency optimization control sub-system and a passive voltage optimization control sub-system, and a background calculation sub-system and a system foreground. The method comprises following steps: tracking isoline of an external power grid and automatically reconstructing a global power grid model; automatically executing online safety analysis integrated with static power grid, dynamic power grid, voltage stabilization, coordination of relay protection setting values, etc., and online recognizing the safety state of the power grid; alarming for normal unsafe condition, and carrying out prevention control; alarming for emergent conditions, and carrying out correction control calculation; and carrying out realtime active scheduling closed-loop control and passive voltage optimized closed-loop control under normal safety condition. The invention can achieve comprehensive and coordinated closed-loop control for the safety of the power grid, the economy and the electric energy quality.

The invention discloses a method of making a partial-anodontia tooth implantation surgical guide plate, which comprises the following steps of: (1) acquiring CT data; (2) reconstructing a three-dimensional model of a skeleton; (3) designing an implantation scheme according to biomechanics and skeleton conditions of a patient; (4) designing an analog column for determining an implantation nest according to a position of an implant; (5) subtracting the analog column of the implant from the three-dimensional model of the skeleton by using Boolean subtraction to obtain a preliminary position of the implantation nest; (6) using a light-cured resin to form a skeleton model with a preliminary implantation nest through a rapid forming method; (7) adjusting the skeleton model with the preliminary implantation nest on the skeleton model; (8) inserting an analog implant into the implantation nest of the skeleton model; (9) making a gypsum model inside the oral cavity of the patient; (10) obtaining an analog soft tissue of an edentulous part; (11) repairing false teeth; and (12) making the surgical guide plate. The invention can be reasonably adjusted by an implantation doctor, and has accurate positioning and good practicability.

The invention discloses an airborne laser radar data-based building three-dimensional reconstruction method and aims to propose a three-dimension reconstruction method combining a building roof boundary with a roof topologic graph so as to realize automatic detection and three-dimensional reconstruction of a building. The method comprises the steps of firstly, filtering airborne LiDAR (Airborne Light Detection And Ranging) data to obtain a ground point and a non ground point; secondly, extracting building point cloud from non ground point cloud in combination with point cloud characteristic information; thirdly, segmenting a roof surface and extracting a boundary contour line; and finally, obtaining key line segments of a roof in combination with the building boundary and the roof topologic graph, and constructing closed polygons of each root surface and combination of the closed polygons so as to obtain a building roof model. The wall surface can be obtained by elevation information of DTM or the ground point, so that the building 3D (three-dimensional) model reconstruction is realized. According to the method, the application bottleneck of the airborne LiDAR data at the present stage is broken through to a certain extent, the complexity of a reconstruction process is lowered, the flexibility of 3D reconstruction is improved, and a breakthrough is provided for city 3D reconstruction.

The invention is suitable for the technical field of virtual surgeries and provides a kidney virtual surgical method and system. The kidney virtual surgical method comprises the following steps of A, performing computerized tomography (CT) scanning on a patient at a preset position to obtain CT scanning data of the patient; B, respectively extracting a plurality of tissues and/or organs relevant to the kidney of the patient from the CT scanning data of the patient; C, performing three-dimensional geometrical model reconstruction on the plurality of tissues and/or organs and establishing a three-dimensional special relation among the plurality of tissues and/or organs; and D, according to the three-dimensional special relation, determining the puncture angle and depth on the skin of the patient to establish a channel from the skin of the patient to the kidney of the patient. Therefore, by means of the kidney virtual surgical method, the accurate channel at the waist of the patient from the skin to the kidney can be established, and a virtual surgery can be operated on the patient.

The invention discloses a point cloud based quick reconstruction STL digital model generation method. The method includes performing simplified filtering and characteristic extraction on input point cloud; performing curve reconstruction on surface point cloud on a digital model; establishing a model topologic relation on the STL digital model and performing grid simplification; performing hole detection and repair on the simplified STL digital model so as to remove defects in the model. According to the invention problems of laser scanning point cloud based STL model reconstruction can be solved effectively. The STL model is generated through steps including point cloud characteristic extraction, model reconstruction and the like and automatic repair of surface defects of the generated STL model is realized through grid simplification and hole repair technology, so that quick reconstruction of the STL digital model is realized and the availability of the reconstructed STL digital model is improved in the premise of keeping basic characteristics of point cloud, and great demands in different fields such as robot offline programming technology, automatic track planning technology and the like in a smart manufacture process can be met.

The invention discloses a method for making oral implant positioning guiding template based on CBCT data. The method comprises the steps: 1) scanning patient jaw bones through CBCT to mainly obtain jaw bone three-dimensional anatomic structure information of a implant region; 2) analyzing and processing CBCT data; 3) using the processed CBCT data to reconstruct a three-dimensional jaw bone model; 4) processing the reconstructed three-dimensional jaw bone model; 5) designing an oral implant positioning guiding template based on the three-dimensional jaw bone model; and 6) carrying out rapid prototype processing; Amount of influence data is effectively reduced through carrying out resampling, anisotropic diffusion filtering to original image data obtained through CBCT scanning, carrying out volume rendering through a ray-casting method, and noise of the data is further removed, thereby accurate reconstruction of the three-dimensional jaw bone model is guaranteed. Furthermore, accuracy of manufacturing the oral implant positioning guiding template is substantially improved through removing model fragments caused by image artifact and non-implant region models in the process of three dimensional jaw bone model reconstruction.

A navigation system for intracerebral hemorrhage puncture surgery based on medical image model reconstruction and localization is disclosed in the embodiment of the invention, Positioned navigation system for intracerebral hemorrhage puncture surgery, including hosts, AR device and naked-eye 3D display device, The host computer receives the medical CT/MRI data of patients with intracerebral hemorrhage, reconstructs the model of multi-group patients with intracerebral hemorrhage, processes the model butt joint, obtains the three-dimensional virtual model of patients' skull and hematoma, and displays the three-dimensional virtual model on the naked eye 3D display device. The host computer receives the medical CT/MRI data of patients with intracerebral hemorrhage. According to the 3D virtualmodel and the angle depth information, the host computer simulates the operation navigation and connection to import the AR device. AR devices are used for feature point detection, spatial registration between virtual hematoma and real hematoma, real-time catheter tracking, holographic display of virtual hematoma and virtual catheter, voice and somatosensory interaction with physicians. The embodiment of the invention can reduce medical cost and provide human-computer interaction function.

The invention discloses a convex hull and OBB (Oriented Bounding Box)-based three-dimensional grid model framework extracting method, which comprises the steps of: partitioning a three-dimensional grid model to generate a plurality of sub grid models; solving a minimal convex hull for a point set of each sub grid, constructing convex hull approximation for the sub grids by using convex hull apexes and calculating a centroid of each convex hull point set to construct original framework points; then solving intersected point sets and generating joint framework points by using OBB for repeated calculation; and finally forming a complete framework by connecting the original framework points and the joint framework points. The convex hull and OBB-based three-dimensional grid model framework extracting method has the advantages of accurate data, convenience for operating, high calculating efficiency and the like. The invention also has an important application value in fields, such as computer animation, object identifying and matching, model reconstruction, virtual navigation and the like.

The invention discloses a fusion method of building BIM model and live-action three-dimensional model. The fusion and conversion process mainly comprises model reconstruction, geometric information conversion, model three-dimensional registration and semantic mapping. The fusion method comprises the steps that firstly, a building BIM model is split, and json attribute files and ifc geometric filesof all components are generated; then, intermediate format exchange is carried out, and ifc is converted into obj; obj is converted into glTF, and three-dimensional geometric figure data transmissionis further achieved; then attributes and geometric data are connected, and glTF and json generate b3dm; and finally, a data description file tileset.json are added, and the data description file tileset.json and b3dm are converted into 3DTiles data. According to the method, the semantic attribute characteristics in the BIM original model are reserved while the geometric figures are completely expressed, and the spatial positions of the two models are unified in the same environment, so that an internal and external integrated display effect is achieved.

The invention discloses a method for reconstructing a model after drilling a surface grid model of a rigid object, which aims to simplify a three-dimensional Boolean operation problem into a two-dimensional Boolean operation problem and reduce the computational complexity of an intersecting detection process of in model Boolean operation so as to achieve the real-time performance of model reconstruction. The method comprises the following steps of: firstly detecting triangular surface sheets intersected with cutting curved surfaces, moving grid peaks outside the intersected curved surfaces of the triangular surface sheets onto the cutting curved surfaces, then preprocessing model surface grids to realize the classified division of the triangular surface sheets on the surface of the grid model, and finally dynamically reconstructing the grid model in Boolean operation geometric processing. A two-dimensional Boolean operation processing method can be directly adopted to realize the Boolean operation geometric processing for the grid model in three-dimensional space, thereby saving the process of space division on the surface of the model, simplifying an intersecting detection algorithm, improving the operation efficiency and achieving the effect of real-time processing.

The invention provides a three-dimensional model reconstruction method based on grid deformation, which is used for constructing a training sample set and comprises the following steps of: manufacturing discrete view angle pictures of a plurality of models and corresponding three-dimensional point cloud data; setting a deep learning network model based on the graph convolutional neural network, wherein the deep learning network model based on the graph convolutional neural network comprises a discrete view angle feature fusion module and a grid deformation module, and the output of the discrete view angle feature fusion module is connected with the input of the grid deformation module; setting a loss function, and based on the training sample set, training a deep learning network model based on the graph convolutional neural network; and inputting the discrete view angle picture of the to-be-reconstructed object to the trained network model, performing automatic reconstruction of the three-dimensional grid model, and evaluating the precision. According to the method, the discrete view angle picture and the three-dimensional point cloud data set of the object are learned and trained, and automatic three-dimensional grid model reconstruction can be carried out on objects of different types and sizes stably and accurately.

The invention discloses a complex fractured rock mass physical model manufactured based on 3D printing production and a modeling method. By using 3D printing, a complex fractured network model entity is reconstructed and a model suitable for an indoor analog simulation experiment is established; the complex fractured rock mass physical model comprises a discrete fracture network DFN model and an entity discrete joint network DFN model. Based on the 3D printing, the complex fractured network model entity can be reconstructed and the model suitable for the indoor analog simulation experiment is established, and an effective method can be provided for an indoor laboratory research on complex fractured rock mass mechanical behaviors. A reliable and effective method is provided for the rock mechanical test field by adopting fractured rock mass model reconstruction and fractured network entity model construction through the 3D printing; with the help of the 3D printing technology, the problems that modeling is difficult to conduct and the cost is high in the current complex fractured rock mass physical model indoor experiment can be effectively solved.

The invention relates to a three-dimensional visual monitoring system based on a novel RGB-Depth camera. The RGB-Depth camera is adopted, a static three-dimensional model rebuilding and dynamic personnel restoring method is used, three-dimensional scene information in a monitoring place is reproduced, and a user can watch a highly-simulated three-dimensional virtual scene at any visual angle. The three-dimensional virtual scene comprises static three-dimensional models and further comprises dynamic three-dimensional models of moving personnel and the like. The system comprises data obtaining devices of the RGB-Depth camera and the like, client computational nodes for obtaining data, a network communication platform, an information comprehension integration service platform and a display platform of the three-dimensional virtual scene. A specific implement method comprises the steps of rebuilding a static three-dimensional visual scene, configuring cameras of the RGB-Depth camera and the like, setting up a three-dimensional visual monitoring system body, generating three-dimensional templates of moving personnel entering a monitoring area, generating dynamic three-dimensional models of personnel in the monitoring area in an on-line mode, displaying a three-dimensional visual monitoring scene in a real-time mode, and carrying out intelligent monitoring services.

The invention relates to a computer aided viscera operation simulation system which comprises a medical database module (1) for storing the original medical image data of a human body, a medical image processing module (2) for generating a three-dimensional organ model from the original medical image data of the human body in the medical database module (1), and a simulating operation platform (3) for carrying out simulating operation to the three-dimensional organ model generated by the medical image processing module (2); wherein the medical database module (1) is connected with the simulating operation platform (3) through the medical image processing module (2). The invention provides the computer aided viscera operation simulation system which can perform three-dimensional model reconstruction according to the medical picture of a true patient and carry out operation simulation repeatedly.

The invention provides a 3D face grid reconstruction method based on a video image. The method mainly includes: inputting a 2D face image in video; building a linear shape model; estimating a face mark point through cascade regression; matching through a 3D model; performing 3D reconstruction on a face. The linear model is built, position and clarity of the mark point are estimated, and face shape is reconstructed by matching a local 3D model; 3D cascade regression is adopted, so that marks in faces in action change are maintained uniform; by selecting a compact 3D point set, the face is completely marked, so that position change of the marks is avoided, and the problems of nonuniformity and self-closing of action change locating points are solved. In addition, 3D marks are adopted for 2D projection, so that calculation expense is reduced greatly. A 3D grid does not contain a background, so that the method is high in universality and good in real-time effect.

The invention discloses a three-dimensional hairstyle design method based on a somatosensory sensor, comprising the following steps: collecting front, left-side, right-side and upward-looking depth image information of a person; establishing a three-dimensional model of the face of the person; combining the three-dimensional model of the face of the person with a standard head model; improving the number and precision of grids of the model by using a method of 'interpolation'; fusing color image information and a standard skin template to generate a model map; superposing a hairstyle three-dimensional model and a head three-dimensional model; and displaying the effect of the three-dimensional model of hairstyle design of the person. The method is simpler and more flexible in operation, and burr, dislocation and other phenomena in the process of model reconstruction are avoided. Meanwhile, the number and precision of grids of the model are improved greatly, the quality of three-dimensional models is improved to the image level, and a new effect is achieved in appearance and experience.

The invention relates to a measured-landform-feature-based real-time adaptive adjusting method and apparatus for airborne laser radar parameters. A plane flies straightly at a constant speed with the altitude unchanged. When an elevation and a two-dimensional space frequency of a measured landform change, optimal computation is carried out to adjust setting values of a scanning field angle, a pulse repetition frequency, and a scanning frequency in real time, so that the laser scanning point cloud bandwidth is unchanged; meanwhile, the point cloud density meets the sampling theorem requirement all the time, thereby eliminating difficulty of distortion and data storage and processing for three-dimensional landform model reconstruction. Flight parameters and measured landform features are measured in real time by a laser scanner, a plane flight speed measurement instrument, and an aerial camera; with integrated computation, an elevation change and a two-dimensional space frequency of the landform are obtained; and then optimal computation is carried out to obtain optimal adjustment values of the three parameters including the pulse repetition frequency, the scanning frequency, and the scanning field angle of the laser scanner and real-time adjustment is carried out, thereby realizing real-time adaptive optimal measurement based on the measured-landform change.

A fast reverse modeling method based on three-dimensional feature extraction is proposed, which utilizes the characteristics of large-scale structural parts splicing measurement and realizes the rapidreverse modeling of large-scale structural parts at the same time of splicing measurement. As the splice measuring probe moves, the noise of measuring point cloud in single field of view of large-scale structure is reduced, filter preprocessing is performed, then the 3D features are extracted by using the curvature and normal vector information of point cloud, the point cloud data is segmented, the stitching parameters are calculated according to the segmentation results, and several single-field data are quickly stitched, and the point cloud data processing is synchronized with the stitchingmeasurement of multiple fields of view, so as to realize the rapid reverse modeling, and guide the production processing and process optimization. The point cloud data processing method of the invention not only has high speed, but also retains the three-dimensional characteristics of the point cloud, can realize the simultaneous processing of the point cloud data and the splicing measurement, and meets the requirements of rapid model reconstruction.

The invention discloses a cuboid automatic detection and three-dimensional reconstruction system and method. The system comprises a driving displacement module, a laser scanning module, a texture acquisition module, and a central processing module. The driving displacement module comprises a multi-axis stepping motor driver, an X-axis translation table, a Y-axis translation table, a Z-axis translation table, a rotation table, and an objective table. The laser scanning module is disposed over the objective table and is used for acquiring the laser point cloud of the surface of a to-be-tested cuboid. The texture acquisition module is disposed on the Z-axis translation table and is used for acquiring the texture image of the surface of the to-be-tested cuboid. The central processing module is connected with the multi-axis stepping motor driver, the laser scanning module, and the texture acquisition module. The cuboid automatic detection and three-dimensional reconstruction system and method achieves automation in laser point cloud acquisition analsysis, image capturing and processing, and three-dimensional model reconstruction, may rapidly achieve the automatic detection and three-dimensional texture reconstruction of the cuboid, and solves low efficiency and large work intensity in the prior art.

The application discloses a three-dimensional pore scale structured grid model reconstruction method based on a rock micro CT image. According to the method, by digital image denoising and binarization processing, partitioning and extraction of pores and skeletons in the rock micro CT image are completed, and storage of image data is implemented in a form of a three-dimensional array matrix; and on the basis of a corresponding relationship of model grid unit bodies and image pixels in the aspect of space positions, independently developed codes are utilized to complete search and space position calibration of the pixels of pore phases and skeleton phases, a structured network mode of the skeletons and the pores is constructed by adopting the grid unit bodies with the same sizes with the pixels, and a final model obtained by surface grid growth and geometric remodeling can be directly used for numerical calculation. The three-dimensional pore scale structured grid model reconstruction method disclosed by the invention improves the defects of poor grid quality, distorted topological structure and the like of a conventional reconstructed model, can meet numerical simulation requirements of problems of seepage, deformation, heat transfer and the like, and has the important significance for promoting researches in the related fields of energy mining and geotechnical engineering.

The present invention relates to a novel means to use a fundamental model of the parts of the brain that are dysfunctional in Parkinson's disease, and related dynamical diseases of the brain, as part of a feedback control system to modulate the signs and symptoms of disease. Fundamental computational models that embody our knowledge of the anatomy, neurons, and dynamics of the parts of the brain we wish to control, and use those models to reconstruct what is inaccessible to our measurements. Through emulation the controller synchronizes to the parts of the brain we wish to observe and track. By passing simultaneous control pulses to both the model controller, as well as the brain, we control both the model and the brain. The detailed framework to embed fundamental models of the brain within a control scheme to control symptoms of Parkinsons and related dynamical diseases of the brain are disclosed.

The invention discloses a large-size plate forming online measurement system and method based on machine vision, and belongs to the technical field of non-contact measurement. The large-size plate forming on-line measuring system comprises a sliding rail, a side support, a cross beam, a sensor support, a three-dimensional displacement sensor, a motion control system and an industrial personal computer. The method comprises a three-dimensional displacement sensor calibration algorithm based on a laser interferometer, a three-dimensional point cloud data splicing algorithm, a three-dimensional point cloud data preprocessing algorithm, a three-dimensional point cloud data segmentation algorithm, model reconstruction and curved surface fitting. According to the invention, a three-dimensional point cloud data splicing mode is adopted to realize measurement of a large-size plate, and the measurement range of the measurement system is increased on the premise of ensuring high precision.