REGISTRATION OF INDOOR 3D RANGE IMAGES USING VIRTUAL 2D SCANS

Marco Langerwisch, Bernardo Wagner

2010

Abstract

For mapping purposes, autonomous robots have to be capable to register 3D range images taken by 3D laser sensors into a common coordinate system. One approach is the Iterative Closest Point (ICP) algorithm. Due to its high computational costs, the ICP algorithm is not suitable for registering 3D range images online. This paper presents a novel approach for registering indoor 3D range images using orthogonal virtual 2D scans, utilizing the typical structure of indoor environments. The 3D registration process is split into three 2D registration steps, and hence the computational effort is reduced. First experiments show that the approach is capable of registering 3D range images much more efficient than ICP algorithm and in real time.

Download


Paper Citation


in Harvard Style

Langerwisch M. and Wagner B. (2010). REGISTRATION OF INDOOR 3D RANGE IMAGES USING VIRTUAL 2D SCANS . In Proceedings of the 7th International Conference on Informatics in Control, Automation and Robotics - Volume 2: ICINCO, ISBN 978-989-8425-01-0, pages 327-332. DOI: 10.5220/0003003303270332

in Bibtex Style

@conference{icinco10,
author={Marco Langerwisch and Bernardo Wagner},
title={REGISTRATION OF INDOOR 3D RANGE IMAGES USING VIRTUAL 2D SCANS},
booktitle={Proceedings of the 7th International Conference on Informatics in Control, Automation and Robotics - Volume 2: ICINCO,},
year={2010},
pages={327-332},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003003303270332},
isbn={978-989-8425-01-0},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 7th International Conference on Informatics in Control, Automation and Robotics - Volume 2: ICINCO,
TI - REGISTRATION OF INDOOR 3D RANGE IMAGES USING VIRTUAL 2D SCANS
SN - 978-989-8425-01-0
AU - Langerwisch M.
AU - Wagner B.
PY - 2010
SP - 327
EP - 332
DO - 10.5220/0003003303270332