81 results about "Geometric reconstruction" patented technology

Disclosed are systems and methods for compositing an augmented reality scene, the methods including the steps of extracting, by an extraction component into a memory of a data-processing machine, at least one object from a real-world image detected by a sensing device; geometrically reconstructing at least one virtual model from at least one object; and compositing AR content from at least one virtual model in order to augment the AR content on the real-world image, thereby creating AR scene. Preferably, the method further includes; extracting at least one annotation from the real-world image into the memory of the data-processing machine for modifying at least one virtual model according to at least one annotation. Preferably, the method further includes: interacting with AR scene by modifying AR content based on modification of at least one object and / or at least one annotation in the real-world image.

A method for determining a volume of ablated tissue includes the steps of supplying energy to tissue, indicating an axis within the tissue, and simulating slicing of the tissue substantially perpendicular to the axis to obtain a plurality of simulated slices. Each of the plurality of simulated slices has a thickness, a cross-sectional perimeter, and a trajectory point defined by the axis within the tissue. The method further includes the steps of determining a volume of each of the plurality of simulated slices based on the trajectory point, the cross-sectional perimeter, and the thickness of each simulated slice, and summing the volumes from each of the plurality of simulated slices to obtain the volume of the ablated tissue.

In various embodiments, methods, systems, and computer program products for processing digital images captured by a mobile device are disclosed. Myriad features enable and / or facilitate processing of such digital images using a mobile device that would otherwise be technically impossible or impractical, and furthermore address unique challenges presented by images captured using a camera rather than a traditional flat-bed scanner, paper-feed scanner, or multifunction peripheral. Notably, the presently disclosed systems and techniques enable three-dimensional reconstruction of objects depicted in image captured using a camera of a mobile device. The reconstruction corrects or compensates for perspective distortion caused by camera-based capture.

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.

Systems, computer program products, and techniques for reconstructing objects depicted in digital image data within a three-dimensional space are disclosed, according to various exemplary embodiments. The inventive concepts uniquely utilize internal features to accomplish reconstruction, thereby avoiding reliance on reconstructing objects based on information derived from location of edges. The inventive concepts thus provide an improvement over conventional object reconstruction since objects may be reconstructed even when edges are obscured or not depicted in the digital image data. In one aspect, a computer-implemented method of reconstructing an object depicted in a digital image includes: detecting a plurality of identifying features of the object, wherein the plurality of identifying features are located internally with respect to the object; and reconstructing the digital image of the object within a three dimensional coordinate space based at least in part on some or all of the plurality of identifying features.

Systems, computer program products, and techniques for detecting objects depicted in digital image data are disclosed, according to various exemplary embodiments. The inventive concepts uniquely utilize internal features to accomplish object detection, thereby avoiding reliance on detecting object edges and / or transitions between the object and other portions of the digital image data, e.g. background textures or other objects. The inventive concepts thus provide an improvement over conventional object detection since objects may be detected even when edges are obscured or not depicted in the digital image data. In one aspect, a computer-implemented method of detecting an object depicted in a digital image includes: detecting a plurality of identifying features of the object, wherein the plurality of identifying features are located internally with respect to the object; and projecting a location of one or more edges of the object based at least in part on the plurality of identifying features.

The invention provides a method for accurately reconstructing a microcosmic finite element grid model of a dissimilar material on the basis of CT (computed tomography) images. According to the method, sequence CT images are acquired through industrial CT, and micro-structural information in the CT images is mapped onto the reconstructed finite element grid model on basis of digitization and threshold segmentation, so that any detailed structural information in the dissimilar material can be represented in the reconstructed model. The method improves the reconstruction accuracy by means of contrast-limited adaptive histogram equalization, median filtering and pixel interpolation, and improves the reconstruction efficiency through image cut and pixel combination. With the method, rectangular (two-dimensional) and cuboid (three-dimensional) unit grid models with higher finite element analysis accuracy are directly reconstructed, error accumulation during reconstruction, grid partition and other links of the existing geometric reconstruction method is avoided, and reconstruction accuracy and efficiency are improved. The method can be widely applied to fields such as performance prediction and optimization design of dissimilar materials.

In various embodiments, methods, systems, and computer program products for processing digital images captured by a mobile device are disclosed. Myriad features enable and / or facilitate processing of such digital images using a mobile device that would otherwise be technically impossible or impractical, and furthermore address unique challenges presented by images captured using a camera rather than a traditional flat-bed scanner, paper-feed scanner, or multifunction peripheral. Notably, the presently disclosed systems and techniques enable three-dimensional reconstruction of objects depicted in image captured using a camera of a mobile device. The reconstruction corrects or compensates for perspective distortion caused by camera-based capture.

The invention relates to an InISAR imaging method of a ship target in the sparse aperture under the condition of echo loss, and aims at solving the problem that 3D geometric reconstruction of scatterings in the ship target is low in precision when echoes received by a radar are incomplete. The method comprises that 1) it is supposed that radars A, B and C receive sparse aperture echoes of the target; 2) an optimal value of a translation component is estimated, and the echoes are compensated; 3) a one-dimensional distance image after translation compensation is obtained; 4) a phase error and awave path different are estimated, and the one-dimensional distance image is compensated via estimated values; 5) ISAR images of the radars A, B and C are obtained via azimuth compression; and 6) interference processing is carried out on each ISAR image of the radars A, B and C along two baselines, and a 3D InISAR reconstruction image of the ship target in the sparse aperture is realized by combining range finding information. The method can be used for the technical field of radars.

The embodiment of the invention provides a method for reconstructing a three-dimensional model of high dynamic range texture, comprising the following steps: adopting shooting devices arranged in a grouped encircling array mode to shoot multi-viewpoint and multiple exposed images and / or videos; restoring response curves of the shooting devices; performing exposure alignment on the images shot at different exposure time so as to restore the three-dimensional model based on the response curves of the devices; and restoring high dynamic texture information by calculating color information of each vertex of a grid. The embodiment of the invention also discloses a device for reconstructing the three-dimensional model of high dynamic range texture, which is composed of an acquisition module, a response curve restoring module, an exposure alignment module and a reconstructing module. The method and the device for reconstructing the three-dimensional model of high dynamic range texture can realize multi-viewpoint stereo reconstruction for the three-dimensional model of high dynamic range under the premise of not lowering geometric reconstruction precision of the model without additionally updating the hardware equipment.

Systems, computer program products, and techniques for detecting objects depicted in digital image data are disclosed, according to various exemplary embodiments. The inventive concepts uniquely utilize internal features to accomplish object detection, thereby avoiding reliance on detecting object edges and / or transitions between the object and other portions of the digital image data, e.g. background textures or other objects. The inventive concepts thus provide an improvement over conventional object detection since objects may be detected even when edges are obscured or not depicted in the digital image data. In one aspect, a computer-implemented method of detecting an object depicted in a digital image includes: detecting a plurality of identifying features of the object, wherein the plurality of identifying features are located internally with respect to the object; and projecting a location of one or more edges of the object based at least in part on the plurality of identifying features.

A method for determining a volume of ablated tissue includes the steps of supplying energy to tissue, indicating an axis within the tissue, and simulating slicing of the tissue substantially perpendicular to the axis to obtain a plurality of simulated slices. Each of the plurality of simulated slices has a thickness, a cross-sectional perimeter, and a trajectory point defined by the axis within the tissue. The method further includes the steps of determining a volume of each of the plurality of simulated slices based on the trajectory point, the cross-sectional perimeter, and the thickness of each simulated slice, and summing the volumes from each of the plurality of simulated slices to obtain the volume of the ablated tissue.

The invention provides a human body real-time modeling system and a modeling method thereof adopting a dual kinect and a rotating platform, comprising a geometric reconstruction module; a camera calibration module; a texture mapping module, etc. Through human body scanning, filtering and denoising, camera tracking and calibration, point cloud fusion and texture mapping, the final three-dimensionalhuman body model is obtained. The invention has the following advantages: the traditional ICP algorithm is improved, the speed of three-dimensional model reconstruction is accelerated, and the requirement of computer configuration is lowered; the real-timeness of human body reconstruction system is kept, the human body model is reconstructed in a short time, the requirement for stationary remaining for a long time on people is loweed, and no special clothes or props are needed; a cheap, fully automatic human body reconstruction system is built and developed, breaking through the traditionalhuman body reconstruction program funding, accuracy constraints; On the basis of geometric reconstruction, the high-quality texture is reconstructed by using the high-fidelity color image provided byKinect.

The invention discloses a method and a system for realizing a combined semantic hierarchical connection model based on panoramic area scene perception. The system comprises an ROI extraction module, apanoramic region segmentation module, a spatial information acquisition module and a multi-level modeling module; the ROI extraction module is in segmentation connection with a target instance and transmits target salient region information; the panoramic region segmentation module is connected with the interest point 3D reconstruction module and transmits region boundary information; the spatialinformation acquisition module is connected with the semantic subspace model and transmits region position correlation information, and the multi-level modeling module outputs spatial semantics and correlation degree information of each region. On the basis that the ROI is obtained through region significance for panoramic segmentation, on the premise that interest points are extracted for geometric reconstruction and element space semantic information association, multi-level modeling of scene perception is achieved according to analysis of probability symbiosis of scene composition elements.

The invention discloses a method for pre-estimating an equivalent thermal conductivity of an unidirectional fiber toughening composite material based on gap defect identification, wherein the method mainly comprises the following steps: processing a material micro electron microscope image by using a picture identification technology, and identifying material internal structure characteristics and gap defects; according to a fiber volume fraction ratio stability criterion, applying a geometric reconstruction technology, establishing a representative element geometric topology model, and simultaneously determining the gap defects; through a method of adding a thermal contact resistance at a representative element internal interface, introducing influence of the gap defects; and simulating to obtain the equivalent thermal conductivity by using an finite element method. According to the method, the representative element considering the gap defects is established based on image identification of an actual micro structure, and calculation of the equivalent thermal conductivity is carried out, an influence rule of gap defect position, number, thickness and other characteristic parameters on the equivalent thermal conductivity is analyzed, and a technical reference is provided for high precision pre-estimation of the thermal conductivity of same-type composite materials.

The invention belongs to the field of composite plastic forming, and particularly provides a virtual test method of the plastic forming property of a particle reinforced aluminum matrix composite. The method includes the steps that (1) geometric reconstruction is conducted on a microstructure of the particle reinforced aluminum matrix composite; (2) meshes of a geometric model, established in the step (1), of the microstructure are divided through an object-oriented finite element technology, and the model is restricted; (3) in finite element software, material attributes are given to the meshes in the step (2), and a matrix damage model, a particle breakage damage model and an interface debond damage model are coupled; (4) the plastic forming property test experimental process is simulated in finite element simulation software, and then plastic forming property indexes of the composite are acquired. According to the method, the plastic forming property of the particle reinforced aluminum matrix composite can be accurately and virtually predicted, experimental test cost of the plastic forming property of the composite can be lowered, operation is easy to conduct, and convenience is brought for technical staff to use.

The invention relates to the field of structural design of key parts of a wind generating set, in particular to a parameterized finite element model-based fan spindle structure optimization method, which comprises the following steps of: establishing a spindle system three-dimensional solid model and establishing a finite element model according to the model; establishing a first objective function to carry out topological optimization on the finite element model of the spindle system, and carrying out geometric reconstruction on the optimized model to obtain the reconstructed three-dimensional solid model of the spindle system; parameterizing the reconstructed three-dimensional solid model of the spindle system, and establishing the finite element model of the reconstructed spindle system; constructing a second objective function which enables the stress of the dangerous section of the spindle to reach an equal stress level and reduces the mass of the spindle, and solving the second objective function to obtain structural parameters of the reconstructed three-dimensional solid model of the spindle system. The design period can be greatly shortened, the rigidity of the fan spindlecan be effectively improved, the quality of the fan spindle is reduced, and the cost is reduced.

The present invention discloses a measuring system for the dynamic contact state of wheels and rails of a high-speed railway. The measuring system comprises a data acquisition system mounted on a vibration-damping frame, as well as a data processing system and a data display system mounted in an on-board control platform. The data acquisition system comprises encoding structural optical image acquisition units and an inertial measurement device; the vibration-damping frame is connected with the compartment bottom through a fixing device at the top of the vibration-damping frame and suspended between the center suspension and the axle box suspension of the frame of a bogie; the encoding structural optical image acquisition units directly face the left and right wheel-rail contact areas and form a surrounding structure; the data processing system is used for obtaining geometrical parameters of the wheel-rail contact through three-dimensional geometric reconstruction, characterization parameter identification and data interaction according to high-definition images captured by the encoding structural optical image acquisition units and data of the acceleration and the like measured by the inertial measurement device, to obtain a wheel-rail relationship spectrum; and the data display system is used for displaying the images, the data and the wheel-rail relationship spectrum through a human-computer interaction interface in real time. The measuring system for the dynamic contact state of the wheels and the rail of the high-speed railway can achieve real-time capture of the wheel-rail dynamic contact state.

The invention provides a reconstruction algorithm for a finite element discretized geometric model. The reconstruction algorithm comprises the following steps: firstly, inputting a finite element discrete grid to determine a grid topological relation, and adding a steady-state thermal field value to a grid node by solving a Laplace partial differential equation; secondly, extracting surface gridsand boundaries of a model, and setting a reasonable threshold value according to a geometric discrete boundary vector value to determine an isothermal surface; performing overlapping removal, extracting an isotherm and a domain boundary, setting an objective function to perform boundary intersection operation, judging the parity of sub-domains according to an intersection result, and setting an objective function for different sub-domains to perform reasonable quadrilateral sub-domain division; and finally, carrying out discrete boundary sorting and boundary fairing on the sub-domains, obtaining control points in the sub-domains through linear interpolation to realize B-spline parameterization, and verifying the accuracy of reconstructed geometry by utilizing isogeometric analysis. Exampleresults show that the algorithm provided by the invention well solves the geometric reconstruction problem of the finite element model.

The invention provides an information processing method, device, cloud processing device and computer program product, is applied to the field of data processing technology, is capable of performing three-dimensional reconstruction and semantic segmentation at the same time, namely, acquiring semantic information at the same time of performing three-dimensional reconstruction based on RGBD data, so as to shorten calculation time and to improve scene segmentation precision. According to the invention, the method includes acquiring RGBD data collected by an image collection device; extracting key frame data in the RGBD data and performing processing and obtaining geometric reconstruction data; performing mapping treatment on the RGB data in the key frame data and the geometric reconstructiondata and obtaining three-dimensional reconstruction data; and performing semantic segmentation treatment on the RGB data in the key frame data and obtaining semantic segmentation data; performing mapping treatment on the semantic segmentation data and the three-dimensional reconstruction data and obtaining three-dimensional semantic map.

The invention relates to a tracheal tree hierarchical extraction method combining a multi-information fusion network and regional growth. The method comprises the following steps: S1, acquiring a CT image of a lung, and preprocessing the CT image; S2, grading the preprocessed CT image set of the lung, and dividing the CT image set into two training sets, namely an overall tracheal tree and a tiny tracheal branch; S3, respectively sampling the overall tracheal tree training set and the tiny tracheal branch training set to obtain an overall tracheal tree training subset and a tiny tracheal training subset; S4, constructing a multi-information fusion segmentation model, and training according to the overall tracheal tree training subset; S5, constructing a voxel classification network model, and training according to the tiny trachea training subset; S6, sequentially inputting to-be-segmented image data into the trained multi-information fusion segmentation model and the trained voxel classification network model to obtain a preliminary tracheal tree; and S7, processing the preliminary tracheal tree by a geometric reconstruction method based on a center line to obtain a final tracheal tree. According to the invention, the classification accuracy is effectively improved.

The invention discloses a BIM (Building Information Modeling) technology-based scene construction method, which relates to the technical field of garden landscape model manufacturing, and comprises the following steps of obtaining a group of site photos at multiple angles, importing the photos into Agisoft Metaphapp Pro software for processing, extracting image information by the software, and converting the image information into point cloud; editing and arranging the generated point cloud data, classifying the point information and geometrically reconstructing the point information, and storing the point cloud data in a file format of. LAS; generating a DSM / DTM model according to projection calculation; correcting the texture color of the DSM / DTM model; exporting the model into a. Obj format; importing the model in the. Obj format into 3D software for processing. According to the method for exporting the model in the SU format from the 3D software, the two kinds of software, namely the Agisoft Metaphapp Pro and the 3D software, are reasonably applied, organic combination of the two kinds of software is achieved, the advantages are complementary, and then manufacturing of a realistic site is achieved.

The invention provides a tunnel three-dimensional geometric reconstruction method considering data driving segment segmentation and model driving segment assembly. The method comprises the following steps: (1) a mapping method from tunnel three-dimensional point cloud expansion to a two-dimensional image; (2) a tunnel image segmentation method; and (3) tunnel segment reconstruction based on modelmatching. The invention provides a method for reconstructing a three-dimensional tunnel geometric model by coupling data and a model driving thought, which is used for reconstructing a multi-scale shield tunnel geometric model. Based on a data driving thought, firstly, point cloud is converted into a binary image, and then tunnel point cloud is segmented on a loop-by-loop scale and a segment scalerespectively by utilizing morphology and a template matching algorithm. The sealing grooves and the bolt holes between the segments are more obvious in the binary image, and the image itself also contains the mapping topological relation between the original point cloud and the pixels. Based on the observation that each ring of a tunnel only comprises one capping block; a segment segmentation problem is converted into a template matching problem based on least square constraint of a binary image; and then, based on the thought of model driving, the reconstruction problem of the shield tunnelsegment on the scale is converted into the matching problem of the segment point cloud and the model base, the types, sizes and positions of various segments are obtained, and assembling of the tunnelsegments is completed.

The embodiment of the invention provides a three-dimensional human body reconstruction method and device, equipment and a storage medium. The method comprises the steps: carrying out the geometric reconstruction of a human body based on a single human body image of a target human body, and obtaining a three-dimensional grid model of the target human body; based on the single human body image, performing local geometric reconstruction on a local part of the target human body to obtain a three-dimensional grid model of the local part; fusing the three-dimensional grid model of the local part with the three-dimensional grid model of the target human body to obtain an initial three-dimensional model; and performing human body texture reconstruction according to the initial three-dimensional model and the single human body image to obtain a three-dimensional human body model of the target human body. According to the embodiment of the invention, the local part in the three-dimensional grid model of the target human body is clearer and more accurate, and the reconstruction effect of the local part is improved.

The invention discloses a geometric shadow map method for triangle reconstruction, and the method comprises the steps: (1) drawing a scene in a light source space, and generating and storing a geometric shadow map; (2) carrying out the compression of the obtained geometric shadow map based on a GPU, and obtaining a compressed geometric shadow map; (3) drawing a scene in a vision field, packaging pixels, carrying out triangle consistency check, and using a stored triangle to achieve the geometric reconstruction of the depth value of a pixel shading object; (4) comparing a pixel depth value in the light source space with the depth value of the shading object, obtaining a shadow calculation result, and completing the reconstruction of the geometric shadow map. Aiming at problems of deformation and depth deviation in a shadow map method, the method employs discrete scene raw triangular patches to represent and record the information of the shading object, and the invention provides the geometric shadow map method. While generating a high-quality shadow, the method has certain advantages over other algorithms in geometric data storage and expenditure and performance, and can meet the demands of real-time application better.

The invention provides a multi-view stereoscopic vision body surface positioning tracking method. The multi-view stereoscopic vision body surface positioning tracking method comprises the following steps that S1, calibrating cameras in each group of vision assemblies; S2, performing simulation geometric reconstruction on the CT data: during reconstruction, retaining point cloud data and target point information of a body surface in the CT data, and simulating a spatial position of the CT point cloud data in a left camera coordinate system; S3, collecting body surface point cloud data through visual components on the two sides of the treatment bed; S4, projecting grating stripes to the surface of the human body by each group of structured light projectors, and modulating the phases of the grating stripes; Carrying out phase splitting by a computer; Obtaining three-dimensional information of the body surface; S5, fusing the point cloud data collected on the two sides of the body surface;And S6, performing three-dimensional accurate registration on the collected and fused body surface point cloud data and CT reconstructed body surface point cloud data. The multi-view stereoscopic vision body surface positioning tracking method has no radiation damage to human bodies and is low in equipment cost.

A method for automatically extracting CityGML LOD models from an IFC model comprises the following four steps of: analyzing the IFC model, and filtering the IFC model, the IFC model geometric reconstruction and the IFC model geometric semantic reorganization on the basis of the semantic mapping relationship; filtering the entity elements based on semantics on the basis of analysis; filtering out entity elements having a mapping relation with the CityCrMLLOD4, wherein the geometry of the entity is the self-defined geometry information of the IFC model; carrying out coordinate conversion and expression mode conversion on the geometric information; generating first layer middle geometric information; secondly, carrying out corresponding geometric reconstruction on the first layer of middle geometric information to generate second layer of middle geometric information, carrying out reorganization on the second layer of middle geometric information, outputting the reorganized geometric information and semantic information to generate a CityGML LOD4 model, and finally generating a CiTyGML LOD3-LOD1 model according to a coordinating relation of the LOD4 and LOD3-LOD1.. Automatic extraction and generation from the IFC model to the CityGML LOD4 model are achieved, and the defect of a coordination conversion method between CityGML LOD4 and CityGML LOD3- LOD1 is overcome.

The invention discloses a face feature point real-time detection method and system based on three-dimensional reconstruction, and the method comprises the steps: (1) collecting a face image frame, andreconstructing a face geometric model in real time through employing a geometric reconstruction algorithm; (2) preprocessing the human face geometric model, and extracting human face point cloud data; (3) inputting the human face point cloud data into a FacePointNet consisting of a coarse PlainFPN network and a block cascade network Cascading FPN to carry out human face three-dimensional featurepoint detection, so as to obtain human face three-dimensional feature points; Wherein the coarse PlainFPN network is used for detecting cheek edge feature points and coarse-precision internal featurepoints of a human face, and the Cascade FPN divides human face point cloud data into a plurality of blocks according to the coarse-precision internal feature points and detects the fine-precision internal feature points of each block respectively. By using the method of the invention, the accuracy and real-time performance of the face three-dimensional feature point detection method can be improved.