Monocular SLAM Feature Point Optimization Based on ORB

DOI

10.2991/acsr.k.191223.011How to use a DOI?

Keywords

ORB-SLAM, monocular SLAM, hamming distance, homography Matrix, RANSAC

Abstract

Simultaneous localization and mapping (SLAM) of mobile robots is a hotspot in the field of computer vision and intelligent robots. Through experiments on the monocular SLAM system indoors, the existence of problems was found that the map is sparse and feature point matches are mistaken in some environments. Therefore, it is proposed to optimize the feature point acquisition and match of the ORB algorithm, filter the feature points according to the actual hardware parameters, and perform image pre-processing on the mismatch problem generated after using the Hamming distance, and then obtain the improved RANSAC algorithm, more precise match points will be got. It makes the image matching in different environments possessing better robustness, and also improves the sparse problem of the construction of point cloud.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Run Tan
AU  - Minling Zhu
AU  - Yuefan Xu
PY  - 2019
DA  - 2019/12/24
TI  - Monocular SLAM Feature Point Optimization Based on ORB
BT  - Proceedings of the 2019 International Conference on Big Data, Electronics and Communication Engineering (BDECE 2019)
PB  - Atlantis Press
SP  - 44
EP  - 50
SN  - 2352-538X
UR  - https://doi.org/10.2991/acsr.k.191223.011
DO  - 10.2991/acsr.k.191223.011
ID  - Tan2019
ER  -