Laser & Optoelectronics Progress, Volume. 58, Issue 2, 0210006(2021)

Image Feature Matching Algorithm Based on Improved ORB

Bin Zou1,2, Xiaohu Zhao1,2、*, and Zhishuai Yin1,2
Author Affiliations
  • 1Hubei Key Laboratory of Modern Auto Parts Technology, Wuhan University of Technology, Wuhan, Hubei 430070, China
  • 2Hubei Collaborative Innovation Center of Automotive Parts Technology, Wuhan University of Technology, Wuhan, Hubei 430070, China
  • show less
    Figures & Tables(7)
    Overall process of ORB algorithm fused with pyramid optical flow method
    Performance comparison before and after regional block
    Matching results after regional block
    Matching results of ORB algorithm on three data sets. (a) Matching results of ORB algorithm; (b) matching results of RANSAC algorithm
    Matching results of proposed algorithm on three data sets. (a) Matching results of proposed algorithm; (b) matching results of RANSAC algorithm
    • Table 1. Matching accuracy and time of ORB algorithm

      View table

      Table 1. Matching accuracy and time of ORB algorithm

      Data setORB algorithmRANSAC algorithmMatching accuracy /%Time /s
      Lenna1539461.40.115
      Desk18313774.00.121
      Forest18012167.00.122
    • Table 2. Matching accuracy and time of proposed algorithm

      View table

      Table 2. Matching accuracy and time of proposed algorithm

      Data setOptimized algorithmRANSAC algorithmMatching accuracy /%Time before regional block /sTime after regional block /s
      Lenna878698.90.1230.099
      Desk12812799.20.1270.104
      Forest10210098.00.1310.108
    Tools

    Get Citation

    Copy Citation Text

    Bin Zou, Xiaohu Zhao, Zhishuai Yin. Image Feature Matching Algorithm Based on Improved ORB[J]. Laser & Optoelectronics Progress, 2021, 58(2): 0210006

    Download Citation

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

    Category: Image Processing

    Received: Apr. 3, 2020

    Accepted: Jul. 7, 2020

    Published Online: Jan. 5, 2021

    The Author Email: Zhao Xiaohu (1550871442@qq.com)

    DOI:10.3788/LOP202158.0210006

    Topics