Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Laser & Optoelectronics Progress, Volume. 61, Issue 14, 1437009(2024)

Feature Matching Method Combining Adaptive Keyframe Strategy with Motion Information

Linbin Wu, Yunfeng Cao*, and Ning Ma
Author Affiliations
  • College of Astronautics, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, Jiangsu, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (11)
    Equations (27)
    References (21)
    Tools
    Paper Information
    Topics
    Figures & Tables(11)
    Overview of the proposed method

    Fig. 1. Overview of the proposed method

    Download full size
    Schematic of the public viewing region of the image after corresponding motion. (a) Public viewing region after pitching motion; (b) public viewing region after yaw motion; (c) public viewing region after rolling motion; (d) public viewing region after three-axis translation

    Fig. 2. Schematic of the public viewing region of the image after corresponding motion. (a) Public viewing region after pitching motion; (b) public viewing region after yaw motion; (c) public viewing region after rolling motion; (d) public viewing region after three-axis translation

    Download full size
    Schematic of the improved ORB feature extraction method

    Fig. 3. Schematic of the improved ORB feature extraction method

    Download full size
    Three-axis inertial translation from IMU measurements

    Fig. 4. Three-axis inertial translation from IMU measurements

    Download full size
    Three-axis inertial rotation from IMU measurements

    Fig. 5. Three-axis inertial rotation from IMU measurements

    Download full size
    Normalized imaging clarity index

    Fig. 6. Normalized imaging clarity index

    Download full size
    Results of the keyframes extracted by the proposed method and traditional feature point parallax method

    Fig. 7. Results of the keyframes extracted by the proposed method and traditional feature point parallax method

    Download full size
    Matching results on MH_04_difficult sequence of EuRoC dataset. (a) Matching results of the proposed method; (b) matching results of the traditional ORB feature method

    Fig. 8. Matching results on MH_04_difficult sequence of EuRoC dataset. (a) Matching results of the proposed method; (b) matching results of the traditional ORB feature method

    Download full size
    Matching results on V2_03_difficult sequence of EuRoC dataset. (a) Matching results of the proposed method; (b) matching results of the traditional ORB feature method

    Fig. 9. Matching results on V2_03_difficult sequence of EuRoC dataset. (a) Matching results of the proposed method; (b) matching results of the traditional ORB feature method

    Download full size

    • Table 1. Part of the image frame information in the MH_04_difficult sequence of the EuRoC dataset

      View table

      Table 1. Part of the image frame information in the MH_04_difficult sequence of the EuRoC dataset

      Frame groupFrameTimestampFrame time /sTime difference /s
      1Reference140363815349509708826.250.50
      Current140363815399509708826.75
      2Reference140363816349509708836.251.00
      Current140363816449509708837.25
      3Reference140363819899509708871.750.75
      Current140363819974509696072.50
      4Reference140363820074509696073.500.50
      Current140363820124509696074.00
      5Reference140363820974509696082.500.75
      Current140363821049509708883.25

    • Table 2. Part of the image frame information in the V2_03_difficult sequence of the EuRoC dataset

      View table

      Table 2. Part of the image frame information in the V2_03_difficult sequence of the EuRoC dataset

      Frame groupFrameTimestampFrame time /sTime difference /s
      1Reference141339489675576038415.250.75
      Current141339489750576051216.00
      2Reference141339490000576051218.500.75
      Current141339490075576038419.25
      3Reference141339491625576038434.500.25
      Current141339491655576038434.75
      4Reference141339494545576038457.000.50
      Current141339494595576038457.50
      5Reference141339499325576038493.100.50
      Current141339499375576038493.60
    Tools

    Get Citation

    Copy Citation Text

    Linbin Wu, Yunfeng Cao, Ning Ma. Feature Matching Method Combining Adaptive Keyframe Strategy with Motion Information[J]. Laser & Optoelectronics Progress, 2024, 61(14): 1437009

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Set citation alerts for article
    Save article for my favorites
    Paper Information

    Category: Digital Image Processing

    Received: Nov. 29, 2023

    Accepted: Dec. 28, 2023

    Published Online: Jul. 11, 2024

    The Author Email: Yunfeng Cao (cyfac@nuaa.edu.cn)

    DOI:10.3788/LOP232578

    CSTR:32186.14.LOP232578

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology