中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

23 records (took 0.018 seconds)

    DISPARITY IMAGE FUSION METHOD FOR MULTIBAND STEREO CAMERAS
    3.
    发明申请
    DISPARITY IMAGE FUSION METHOD FOR MULTIBAND STEREO CAMERAS 有权

    公开(公告)号：US20220207776A1

    公开(公告)日：2022-06-30

    申请号：US17604288

    申请日：2020-03-05

    Applicant: DALIAN UNIVERSITY OF TECHNOLOGY ,  PENG CHENG LABORATORY

    Inventor： Wei ZHONG ,  Hong ZHANG ,  Haojie LI ,  Zhihui WANG ,  Risheng LIU ,  Xin FAN ,  Zhongxuan LUO ,  Shengquan LI

    IPC: G06T7/80 ,  G06T5/50 ,  G06F17/12

    Abstract: A disparity image fusion method for multiband stereo cameras belongs to the field of image processing and computer vision. The method obtains pixel disparity confidence information by using the intermediate output of binocular disparity estimation. The confidence information can be used to judge the disparity credibility of the position and assist disparity fusion. The confidence acquisition process makes full use of the intermediate output of calculation, and can be conveniently embedded into the traditional disparity estimation process, with high calculation efficiency and simple and easy operation. In the disparity image fusion method for multiband stereo cameras proposed by the method, the disparity diagrams participating in the fusion are obtained according to the binocular images of the corresponding bands, which makes full use of the information of each band and simultaneously avoiding introducing uncertainty and errors.

    TOPOLOGICAL PRESERVING DEFORMATION METHOD FOR 3D MODEL BASED ON MULTIPLE VOLUMETRIC HARMONIC FIELD
    6.
    发明公开
    TOPOLOGICAL PRESERVING DEFORMATION METHOD FOR 3D MODEL BASED ON MULTIPLE VOLUMETRIC HARMONIC FIELD 审中-公开

    公开(公告)号：US20240233294A1

    公开(公告)日：2024-07-11

    申请号：US17778263

    申请日：2021-05-08

    Applicant: DALIAN UNIVERSITY OF TECHNOLOGY

    Inventor： Shengfa WANG ,  Yiming ZHU ,  Xiaopeng ZHENG ,  Na LEI ,  Zhongxuan LUO ,  Fuwei CHEN ,  Yongjie WANG ,  Fan ZHANG

    IPC: G06T19/20 ,  G06T17/20

    CPC classification number: G06T19/20 ,  G06T17/20 ,  G06T2219/2021

    Abstract: The present invention discloses a topological preserving deformation method for a 3D model based on a multiple volumetric harmonic field, and belongs to the fields of computer graphics, computational mathematics, topology and differential geometry. Firstly, a tetrahedral mesh is constructed between a source 3D model and a target 3D model; then, a domain of topological transformation in a deformation process is calculated based on a traditional single volumetric harmonic field; special multiple boundary conditions are set; next, a multiple volumetric harmonic field is calculated; and finally, topological preserving surface deformation is induced. The present invention can find the domain of topological transformation in the deformation process based on a saddle point in the traditional volumetric harmonic field, and can adaptively construct the multiple volumetric harmonic field that can induce topological preserving deformation, so as to generate topological preserving deformation surfaces under the guidance of the multiple volumetric harmonic field. The method has universality and high efficiency for 3D model deformation of the same topology, requires less computational cost compared with the traditional large deformation diffeomorphism metric mapping, and can be widely used.

    METHOD BASED ON DEEP NEURAL NETWORK TO EXTRACT APPEARANCE AND GEOMETRY FEATURES FOR PULMONARY TEXTURES CLASSIFICATION
    8.
    发明申请
    METHOD BASED ON DEEP NEURAL NETWORK TO EXTRACT APPEARANCE AND GEOMETRY FEATURES FOR PULMONARY TEXTURES CLASSIFICATION 审中-公开

    公开(公告)号：US20200258218A1

    公开(公告)日：2020-08-13

    申请号：US16649650

    申请日：2019-01-07

    Applicant: Dalian University of Technology

    Inventor： Rui XU ,  Xinchen YE ,  Lin LIN ,  Haojie LI ,  Xin FAN ,  Zhongxuan LUO

    IPC: G06T7/00 ,  G06N3/04 ,  G06N3/08 ,  G06T7/41

    Abstract: Provided is a method based on deep neural network to extract appearance and geometry features for pulmonary textures classification, which belongs to the technical fields of medical image processing and computer vision. Taking 217 pulmonary computed tomography images as original data, several groups of datasets are generated through a preprocessing procedure. Each group includes a CT image patch, a corresponding image patch containing geometry information and a ground-truth label. A dual-branch residual network is constructed, including two branches separately takes CT image patches and corresponding image patches containing geometry information as input. Appearance and geometry information of pulmonary textures are learnt by the dual-branch residual network, and then they are fused to achieve high accuracy for pulmonary texture classification. Besides, the proposed network architecture is clear, easy to be constructed and implemented.

    NOVEL HIGHLY ROBUST UNDERWATER OPTICAL COMMUNICATION SYSTEM
    9.
    发明申请
    NOVEL HIGHLY ROBUST UNDERWATER OPTICAL COMMUNICATION SYSTEM 有权

    公开(公告)号：US20210336705A1

    公开(公告)日：2021-10-28

    申请号：US17257793

    申请日：2020-05-11

    Applicant: DALIAN UNIVERSITY OF TECHNOLOGY

    Inventor： Chi LIN ,  Yongda YU ,  Yichuan ZHANG ,  Lei WANG ,  Guowei WU ,  Zhehuan ZHAO ,  Zhongxuan LUO

    IPC: H04B10/61 ,  H04B10/50 ,  H04B10/516

    Abstract: The present invention belongs to the technical field of underwater communication, and provides a novel highly robust underwater optical communication system which comprises a sending module and a receiving module. The novel highly robust underwater optical communication system realizes highly robust underwater optical communication under strong interference of sunlight and artificial light sources. The system uses a new physical method irrelevant to frequency, and can be used with existing MIMO and CDMA to obtain better communication effects. The circularly polarized light is used for signal transmission, thereby avoiding the problem of channel misalignment caused by the rotation of a platform underwater. At the same time, good polarization maintaining of a marine environment makes the signal characteristics difficult to lose.

    METHOD FOR 3D SCENE DENSE RECONSTRUCTION BASED ON MONOCULAR VISUAL SLAM
    10.
    发明申请
    METHOD FOR 3D SCENE DENSE RECONSTRUCTION BASED ON MONOCULAR VISUAL SLAM 审中-公开

    公开(公告)号：US20200273190A1

    公开(公告)日：2020-08-27

    申请号：US16650331

    申请日：2019-01-07

    Applicant: DALIAN UNIVERSITY OF TECHNOLOGY

    Inventor： Xinchen YE ,  Wei ZHONG ,  Zhihui WANG ,  Haojie LI ,  Lin LIN ,  Xin FAN ,  Zhongxuan LUO

    IPC: G06T7/579 ,  G06T17/00 ,  G06T7/70 ,  G06T7/246

    Abstract: The present invention provides a method of dense 3D scene reconstruction based on monocular camera and belongs to the technical field of image processing and computer vision, which builds the reconstruction strategy with fusion of traditional geometry-based depth computation and convolutional neural network (CNN) based depth prediction, and formulates depth reconstruction model solved by efficient algorithm to obtain high-quality dense depth map. The system is easy to construct because of its low requirement for hardware resources and achieves dense reconstruction only depending on ubiquitous monocular cameras. Camera tracking of feature-based SLAM provides accurate pose estimation, while depth reconstruction model with fusion of sparse depth points and CNN-inferred depth achieves dense depth estimation and 3D scene reconstruction; The use of fast solver in depth reconstruction avoids solving inversion of large-scale sparse matrix, which improves running speed of the algorithm and ensures the real-time dense 3D scene reconstruction based on monocular camera.

Patent Agency Ranking