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. 60, Issue 6, 0610007(2023)

Fast Image Registration Method Based on Improved AKAZE Algorithm

Weidong Zhao, Junde Liu*, Manman Wang, and Dan Li
Author Affiliations
  • School of Electrical and Information Engineering, Anhui University of Technology, Maanshan243032, Anhui , China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (10)
    Equations (13)
    References (20)
    Cited By (4)
    Tools
    Paper Information
    Topics
    Figures & Tables(10)
    Overall process of the algorithm

    Fig. 1. Overall process of the algorithm

    Download full size
    Illustration of sampling mode in FREAK descriptor

    Fig. 2. Illustration of sampling mode in FREAK descriptor

    Download full size
    Schematic of Si distribution

    Fig. 3. Schematic of Si distribution

    Download full size
    Diagram of moving grid

    Fig. 4. Diagram of moving grid

    Download full size
    Examples on Oxford standard dataset. (a) Different perspectives; (b) different degrees of rotation; (c) different ambiguity; (d) JPEG compression; (e) different illumination

    Fig. 5. Examples on Oxford standard dataset. (a) Different perspectives; (b) different degrees of rotation; (c) different ambiguity; (d) JPEG compression; (e) different illumination

    Download full size
    Parameter results on Oxford dataset. (a) Matching time; (b) matching accuracy

    Fig. 6. Parameter results on Oxford dataset. (a) Matching time; (b) matching accuracy

    Download full size
    Examples on RSSCN7 remote sensing image dataset. (a) Different perspectives; (b) different degrees of rotation; (c) different ambiguity; (d) JPEG compression; (e) different illumination

    Fig. 7. Examples on RSSCN7 remote sensing image dataset. (a) Different perspectives; (b) different degrees of rotation; (c) different ambiguity; (d) JPEG compression; (e) different illumination

    Download full size
    Comparison of matching results of two algorithms. (a) Different perspectives; (b) different degrees of rotation; (c) different ambiguity; (d) JPEG compression; (e) different illumination

    Fig. 8. Comparison of matching results of two algorithms. (a) Different perspectives; (b) different degrees of rotation; (c) different ambiguity; (d) JPEG compression; (e) different illumination

    Download full size
    Operation results on RSSCN7 dataset. (a) Matching time; (b) matching accuracy

    Fig. 9. Operation results on RSSCN7 dataset. (a) Matching time; (b) matching accuracy

    Download full size

    • Table 1. Comparison of feature matching time of different algorithms

      View table

      Table 1. Comparison of feature matching time of different algorithms

      Image No.AKAZEORBSIFT+FREAKKAZEAKAZE+ FREAK
      Fig. 7（a）0.2880.2610.3350.6000.106

      Fig.7（b）

      Fig.7（c）

      0.351

      0.308

      		0.3631.1081.3020.190
      0.2520.6921.0800.163
      Fig. 7（d）0.3850.2770.8400.8840.158
      Fig. 7（e）0.6630.1811.2404.8200.345
    Tools

    Get Citation

    Copy Citation Text

    Weidong Zhao, Junde Liu, Manman Wang, Dan Li. Fast Image Registration Method Based on Improved AKAZE Algorithm[J]. Laser & Optoelectronics Progress, 2023, 60(6): 0610007

    Download Citation

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

    Category: Image Processing

    Received: Dec. 13, 2021

    Accepted: Jan. 17, 2022

    Published Online: Mar. 16, 2023

    The Author Email: Junde Liu (909390702@qq.com)

    DOI:10.3788/LOP213215

    Topics

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