A Cross Ratio Invariability-based RANSAC Registration Method

DOI

10.2991/ncce-18.2018.52How to use a DOI?

Keywords

Cross Ratio Invariability, 3D Point Cloud, Consensus Algorithm, Affine Matrix

Abstract

To improve registration performance of machine vision under complex environment, an improved RANSAC rapid point cloud registration algorithm is presented upon the principle of cross ratio invariability for internal feature points of a registered target. By analyzing the model of a registered target, the method can be used to work out the affine matrix between reference point cloud and target point cloud; cross ratio invariability of internal point data is taken as the constraint condition for registration to correct and remove feature points with wrong registration; besides, the time and accuracy of correct registration between a target image and a reference image are taken as the standards for registration effect evaluation. Lastly, it is indicated by comparing the algorithm and a traditional registration algorithm SAC-IA+ICP that the algorithm presented in the paper is substantially superior to the traditional registration algorithm, since it shows good robustness in large-scale 3D point cloud registration and significantly improved registration efficiency upon guaranteed precision.

Copyright

© 2018, 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  - Jian Liu
AU  - Wenjun Wu
PY  - 2018/05
DA  - 2018/05
TI  - A Cross Ratio Invariability-based RANSAC Registration Method
BT  - Proceedings of the 2018 International Conference on Network, Communication, Computer Engineering (NCCE 2018)
PB  - Atlantis Press
SP  - 338
EP  - 342
SN  - 1951-6851
UR  - https://doi.org/10.2991/ncce-18.2018.52
DO  - 10.2991/ncce-18.2018.52
ID  - Liu2018/05
ER  -