Title: Improved image matching algorithm based on LK optical flow and grid motion statistics

Authors: Qunpo Liu; Xiulei Xi; Weicun Zhang; Lingxiao Yang; Naohiko Hanajima

Addresses: School of Electrical Engineering and Automation, Henan Polytechnic University, Jiaozuo, Henan, China; Henan International Joint Laboratory of Direct Drive and Control of Intelligent Equipment, Jiaozuo, Henan, China ' School of Electrical Engineering and Automation, Henan Polytechnic University, Jiaozuo, Henan, China; Henan International Joint Laboratory of Direct Drive and Control of Intelligent Equipment, Jiaozuo, Henan, China ' School of Automation, University of Science and Technology Beijing, Haidian, Beijing, China ' School of Electrical Engineering and Automation, Henan Polytechnic University, Jiaozuo, Henan, China ' College of Information and Systems, Muroran Institute of Technology, Muroran, Hakkaido, Japan; Henan International Joint Laboratory of Direct Drive and Control of Intelligent Equipment, Jiaozuo, Henan, China

Abstract: In order to solve the problems of low accuracy and long time-consuming of AKAZE algorithm in the image matching of glass encapsulated electrical connectors, an improved image matching algorithm based on LK optical flow and grid motion statistics is proposed. Matching points are obtained by calculating the matching area which are made up of the feature points for conditional constraints. In the local feature matching algorithm, large amount of calculation is an urgent problem as a result of sparse neighbourhood consistency feature cannot define adjacent areas well. The false matching points are removed by improved grid motion statistics algorithm based on integrates FLANN algorithm, and then match computation time is reduced. The performance is verified by experiments based on the Mikolajczyk and actual scene data. Experiment results show that the proposed algorithm can handle the actual scene data, the CMR reaches over 93%, and the time is within 0.4s.

Keywords: positioning; glass-encapsulated electrical connector; AKAZE algorithm; LK optical flow; feature matching; grid motion statistics.

DOI: 10.1504/IJCAT.2022.123238

International Journal of Computer Applications in Technology, 2022 Vol.68 No.1, pp.49 - 57

Received: 22 Feb 2021
Accepted: 15 May 2021
Published online: 06 Jun 2022 *