Robust extraction of arbitrary-shaped features in ceiling for upward-looking camera-based SLAM

Seo Yeon Hwang, Jae-Bok Song, Mun Sang Kim

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

This paper proposes a novel scheme of extracting the features in the ceiling using an upwardlooking camera for SLAM in indoor environments. The conventional approaches based on corner or line features have difficulties in associating adjacent indistinguishable features since their descriptors do not have enough information to distinguish them. Therefore, we used various properties of the objects in the ceiling, such as the distribution of nodes, and size and orientation strength of the region of interest. Also, the similarities among adjacent features are calculated to distinguish between unique and non-unique features. The non-unique features are discarded before they are registered to the feature map. The robustness of the proposed scheme is verified by several experiments using a mobile robot. The extended Kalman filter is used to estimate both robot pose and feature position and orientation, and the experimental results show that the proposed scheme successfully works in various indoor environments.

Original languageEnglish
Title of host publicationIFAC Proceedings Volumes (IFAC-PapersOnline)
Pages8165-8170
Number of pages6
Volume18
EditionPART 1
DOIs
Publication statusPublished - 2011 Dec 1
Event18th IFAC World Congress - Milano, Italy
Duration: 2011 Aug 282011 Sep 2

Other

Other18th IFAC World Congress
CountryItaly
CityMilano
Period11/8/2811/9/2

Fingerprint

Ceilings
Cameras
Extended Kalman filters
Mobile robots
Robots
Experiments

Keywords

  • Feature extraction
  • Mobile robots
  • Monocular vision-based SLAM
  • Upward-looking camera

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Hwang, S. Y., Song, J-B., & Kim, M. S. (2011). Robust extraction of arbitrary-shaped features in ceiling for upward-looking camera-based SLAM. In IFAC Proceedings Volumes (IFAC-PapersOnline) (PART 1 ed., Vol. 18, pp. 8165-8170) https://doi.org/10.3182/20110828-6-IT-1002.00459

Robust extraction of arbitrary-shaped features in ceiling for upward-looking camera-based SLAM. / Hwang, Seo Yeon; Song, Jae-Bok; Kim, Mun Sang.

IFAC Proceedings Volumes (IFAC-PapersOnline). Vol. 18 PART 1. ed. 2011. p. 8165-8170.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hwang, SY, Song, J-B & Kim, MS 2011, Robust extraction of arbitrary-shaped features in ceiling for upward-looking camera-based SLAM. in IFAC Proceedings Volumes (IFAC-PapersOnline). PART 1 edn, vol. 18, pp. 8165-8170, 18th IFAC World Congress, Milano, Italy, 11/8/28. https://doi.org/10.3182/20110828-6-IT-1002.00459
Hwang SY, Song J-B, Kim MS. Robust extraction of arbitrary-shaped features in ceiling for upward-looking camera-based SLAM. In IFAC Proceedings Volumes (IFAC-PapersOnline). PART 1 ed. Vol. 18. 2011. p. 8165-8170 https://doi.org/10.3182/20110828-6-IT-1002.00459
Hwang, Seo Yeon ; Song, Jae-Bok ; Kim, Mun Sang. / Robust extraction of arbitrary-shaped features in ceiling for upward-looking camera-based SLAM. IFAC Proceedings Volumes (IFAC-PapersOnline). Vol. 18 PART 1. ed. 2011. pp. 8165-8170
@inproceedings{c4202fad4b514dfd84e280b4674d49f7,
title = "Robust extraction of arbitrary-shaped features in ceiling for upward-looking camera-based SLAM",
abstract = "This paper proposes a novel scheme of extracting the features in the ceiling using an upwardlooking camera for SLAM in indoor environments. The conventional approaches based on corner or line features have difficulties in associating adjacent indistinguishable features since their descriptors do not have enough information to distinguish them. Therefore, we used various properties of the objects in the ceiling, such as the distribution of nodes, and size and orientation strength of the region of interest. Also, the similarities among adjacent features are calculated to distinguish between unique and non-unique features. The non-unique features are discarded before they are registered to the feature map. The robustness of the proposed scheme is verified by several experiments using a mobile robot. The extended Kalman filter is used to estimate both robot pose and feature position and orientation, and the experimental results show that the proposed scheme successfully works in various indoor environments.",
keywords = "Feature extraction, Mobile robots, Monocular vision-based SLAM, Upward-looking camera",
author = "Hwang, {Seo Yeon} and Jae-Bok Song and Kim, {Mun Sang}",
year = "2011",
month = "12",
day = "1",
doi = "10.3182/20110828-6-IT-1002.00459",
language = "English",
isbn = "9783902661937",
volume = "18",
pages = "8165--8170",
booktitle = "IFAC Proceedings Volumes (IFAC-PapersOnline)",
edition = "PART 1",

}

TY - GEN

T1 - Robust extraction of arbitrary-shaped features in ceiling for upward-looking camera-based SLAM

AU - Hwang, Seo Yeon

AU - Song, Jae-Bok

AU - Kim, Mun Sang

PY - 2011/12/1

Y1 - 2011/12/1

N2 - This paper proposes a novel scheme of extracting the features in the ceiling using an upwardlooking camera for SLAM in indoor environments. The conventional approaches based on corner or line features have difficulties in associating adjacent indistinguishable features since their descriptors do not have enough information to distinguish them. Therefore, we used various properties of the objects in the ceiling, such as the distribution of nodes, and size and orientation strength of the region of interest. Also, the similarities among adjacent features are calculated to distinguish between unique and non-unique features. The non-unique features are discarded before they are registered to the feature map. The robustness of the proposed scheme is verified by several experiments using a mobile robot. The extended Kalman filter is used to estimate both robot pose and feature position and orientation, and the experimental results show that the proposed scheme successfully works in various indoor environments.

AB - This paper proposes a novel scheme of extracting the features in the ceiling using an upwardlooking camera for SLAM in indoor environments. The conventional approaches based on corner or line features have difficulties in associating adjacent indistinguishable features since their descriptors do not have enough information to distinguish them. Therefore, we used various properties of the objects in the ceiling, such as the distribution of nodes, and size and orientation strength of the region of interest. Also, the similarities among adjacent features are calculated to distinguish between unique and non-unique features. The non-unique features are discarded before they are registered to the feature map. The robustness of the proposed scheme is verified by several experiments using a mobile robot. The extended Kalman filter is used to estimate both robot pose and feature position and orientation, and the experimental results show that the proposed scheme successfully works in various indoor environments.

KW - Feature extraction

KW - Mobile robots

KW - Monocular vision-based SLAM

KW - Upward-looking camera

UR - http://www.scopus.com/inward/record.url?scp=84866745394&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866745394&partnerID=8YFLogxK

U2 - 10.3182/20110828-6-IT-1002.00459

DO - 10.3182/20110828-6-IT-1002.00459

M3 - Conference contribution

SN - 9783902661937

VL - 18

SP - 8165

EP - 8170

BT - IFAC Proceedings Volumes (IFAC-PapersOnline)

ER -