Ensemble of vector and binary descriptor for loop closure detection

Mohammed Omar Salameh, Azizi Abdullah, Shahnorbanun Sahran

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

1 Citation (Scopus)

Abstract

Loop closure detection plays an important role in vSLAM for building and updating maps of the surrounding environment. An efficient vSLAM system needs an informative descriptor for landmark description and stablemodel formaking decisions. Most of the solutions dependent on using a single descriptor for landmark description, whereas other solutions proposed to use a combination of descriptors. However, these solutions still have the limitation in correctly detecting a previously visited landmark. In this paper, an ensemble of loop closure detection is proposed using Bayesian filter models for making decisions. In this approach, a set of different keypoint descriptors is used as input to bag-of-word descriptors. After that, these descriptors, i.e., SIFT, SURF, and ORB, are used to construct Bayesian filter models and ensemble learning algorithm for loop closure detection. The proposed approach is validated on a public dataset, namely City-Center dataset (CiC). The results shown that the proposed ensemble algorithm outperforms single model and existing loop closure detection system approaches. It gives 87.96% for ensemble learning and 86.36% for the best single model and 37, 80, 81% for FAB-MAP, PIRF-Nav2.0, and RTAB-MAP, respectively.

Original languageEnglish
Title of host publicationRobot Intelligence Technology and Applications 4 - Results from the 4th International Conference on Robot Intelligence Technology and Applications
PublisherSpringer Verlag
Pages329-340
Number of pages12
Volume447
ISBN (Print)9783319312910
DOIs
Publication statusPublished - 2017
Event4th International Conference on Robot Intelligence Technology and Applications, RiTA 2015 - Bucheon, Korea, Republic of
Duration: 14 Dec 201516 Dec 2015

Publication series

NameAdvances in Intelligent Systems and Computing
Volume447
ISSN (Print)21945357

Other

Other4th International Conference on Robot Intelligence Technology and Applications, RiTA 2015
CountryKorea, Republic of
CityBucheon
Period14/12/1516/12/15

Fingerprint

Learning algorithms
Decision making

Keywords

  • Appearance-based localization
  • Ensemble learning
  • Loop closure detection

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Computer Science(all)

Cite this

Salameh, M. O., Abdullah, A., & Sahran, S. (2017). Ensemble of vector and binary descriptor for loop closure detection. In Robot Intelligence Technology and Applications 4 - Results from the 4th International Conference on Robot Intelligence Technology and Applications (Vol. 447, pp. 329-340). (Advances in Intelligent Systems and Computing; Vol. 447). Springer Verlag. https://doi.org/10.1007/978-3-319-31293-4_27

Ensemble of vector and binary descriptor for loop closure detection. / Salameh, Mohammed Omar; Abdullah, Azizi; Sahran, Shahnorbanun.

Robot Intelligence Technology and Applications 4 - Results from the 4th International Conference on Robot Intelligence Technology and Applications. Vol. 447 Springer Verlag, 2017. p. 329-340 (Advances in Intelligent Systems and Computing; Vol. 447).

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

Salameh, MO, Abdullah, A & Sahran, S 2017, Ensemble of vector and binary descriptor for loop closure detection. in Robot Intelligence Technology and Applications 4 - Results from the 4th International Conference on Robot Intelligence Technology and Applications. vol. 447, Advances in Intelligent Systems and Computing, vol. 447, Springer Verlag, pp. 329-340, 4th International Conference on Robot Intelligence Technology and Applications, RiTA 2015, Bucheon, Korea, Republic of, 14/12/15. https://doi.org/10.1007/978-3-319-31293-4_27
Salameh MO, Abdullah A, Sahran S. Ensemble of vector and binary descriptor for loop closure detection. In Robot Intelligence Technology and Applications 4 - Results from the 4th International Conference on Robot Intelligence Technology and Applications. Vol. 447. Springer Verlag. 2017. p. 329-340. (Advances in Intelligent Systems and Computing). https://doi.org/10.1007/978-3-319-31293-4_27
Salameh, Mohammed Omar ; Abdullah, Azizi ; Sahran, Shahnorbanun. / Ensemble of vector and binary descriptor for loop closure detection. Robot Intelligence Technology and Applications 4 - Results from the 4th International Conference on Robot Intelligence Technology and Applications. Vol. 447 Springer Verlag, 2017. pp. 329-340 (Advances in Intelligent Systems and Computing).
@inproceedings{32c4de965eaa4a1eae687efc424a4d03,
title = "Ensemble of vector and binary descriptor for loop closure detection",
abstract = "Loop closure detection plays an important role in vSLAM for building and updating maps of the surrounding environment. An efficient vSLAM system needs an informative descriptor for landmark description and stablemodel formaking decisions. Most of the solutions dependent on using a single descriptor for landmark description, whereas other solutions proposed to use a combination of descriptors. However, these solutions still have the limitation in correctly detecting a previously visited landmark. In this paper, an ensemble of loop closure detection is proposed using Bayesian filter models for making decisions. In this approach, a set of different keypoint descriptors is used as input to bag-of-word descriptors. After that, these descriptors, i.e., SIFT, SURF, and ORB, are used to construct Bayesian filter models and ensemble learning algorithm for loop closure detection. The proposed approach is validated on a public dataset, namely City-Center dataset (CiC). The results shown that the proposed ensemble algorithm outperforms single model and existing loop closure detection system approaches. It gives 87.96{\%} for ensemble learning and 86.36{\%} for the best single model and 37, 80, 81{\%} for FAB-MAP, PIRF-Nav2.0, and RTAB-MAP, respectively.",
keywords = "Appearance-based localization, Ensemble learning, Loop closure detection",
author = "Salameh, {Mohammed Omar} and Azizi Abdullah and Shahnorbanun Sahran",
year = "2017",
doi = "10.1007/978-3-319-31293-4_27",
language = "English",
isbn = "9783319312910",
volume = "447",
series = "Advances in Intelligent Systems and Computing",
publisher = "Springer Verlag",
pages = "329--340",
booktitle = "Robot Intelligence Technology and Applications 4 - Results from the 4th International Conference on Robot Intelligence Technology and Applications",
address = "Germany",

}

TY - GEN

T1 - Ensemble of vector and binary descriptor for loop closure detection

AU - Salameh, Mohammed Omar

AU - Abdullah, Azizi

AU - Sahran, Shahnorbanun

PY - 2017

Y1 - 2017

N2 - Loop closure detection plays an important role in vSLAM for building and updating maps of the surrounding environment. An efficient vSLAM system needs an informative descriptor for landmark description and stablemodel formaking decisions. Most of the solutions dependent on using a single descriptor for landmark description, whereas other solutions proposed to use a combination of descriptors. However, these solutions still have the limitation in correctly detecting a previously visited landmark. In this paper, an ensemble of loop closure detection is proposed using Bayesian filter models for making decisions. In this approach, a set of different keypoint descriptors is used as input to bag-of-word descriptors. After that, these descriptors, i.e., SIFT, SURF, and ORB, are used to construct Bayesian filter models and ensemble learning algorithm for loop closure detection. The proposed approach is validated on a public dataset, namely City-Center dataset (CiC). The results shown that the proposed ensemble algorithm outperforms single model and existing loop closure detection system approaches. It gives 87.96% for ensemble learning and 86.36% for the best single model and 37, 80, 81% for FAB-MAP, PIRF-Nav2.0, and RTAB-MAP, respectively.

AB - Loop closure detection plays an important role in vSLAM for building and updating maps of the surrounding environment. An efficient vSLAM system needs an informative descriptor for landmark description and stablemodel formaking decisions. Most of the solutions dependent on using a single descriptor for landmark description, whereas other solutions proposed to use a combination of descriptors. However, these solutions still have the limitation in correctly detecting a previously visited landmark. In this paper, an ensemble of loop closure detection is proposed using Bayesian filter models for making decisions. In this approach, a set of different keypoint descriptors is used as input to bag-of-word descriptors. After that, these descriptors, i.e., SIFT, SURF, and ORB, are used to construct Bayesian filter models and ensemble learning algorithm for loop closure detection. The proposed approach is validated on a public dataset, namely City-Center dataset (CiC). The results shown that the proposed ensemble algorithm outperforms single model and existing loop closure detection system approaches. It gives 87.96% for ensemble learning and 86.36% for the best single model and 37, 80, 81% for FAB-MAP, PIRF-Nav2.0, and RTAB-MAP, respectively.

KW - Appearance-based localization

KW - Ensemble learning

KW - Loop closure detection

UR - http://www.scopus.com/inward/record.url?scp=84978500140&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84978500140&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-31293-4_27

DO - 10.1007/978-3-319-31293-4_27

M3 - Conference contribution

SN - 9783319312910

VL - 447

T3 - Advances in Intelligent Systems and Computing

SP - 329

EP - 340

BT - Robot Intelligence Technology and Applications 4 - Results from the 4th International Conference on Robot Intelligence Technology and Applications

PB - Springer Verlag

ER -