Gradient-based local affine invariant feature extraction for mobile robot localization in indoor environments

Jihyo Lee, Hanseok Ko

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

In this paper, we propose a gradient-based local affine invariant feature extraction algorithm (G-LAIFE), using affine moment invariants for robot localization in real indoor environments. The proposed algorithm is an effective feature extraction algorithm that is invariant to image translation and to 3D rotation, and it is within a partial range of the image scale. Representative performance analysis confirms that the proposed G-LAIFE algorithm significantly enhances the recognition rate and is more efficient than the scale invariant feature transform (SIFT), especially in terms of 3D rotation change and computational time.

Original languageEnglish
Pages (from-to)1934-1940
Number of pages7
JournalPattern Recognition Letters
Volume29
Issue number14
DOIs
Publication statusPublished - 2008 Oct 15

Fingerprint

Mobile robots
Feature extraction
Robots

Keywords

  • 3D rotation
  • Affine invariant
  • Local feature extraction
  • Translation

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Gradient-based local affine invariant feature extraction for mobile robot localization in indoor environments. / Lee, Jihyo; Ko, Hanseok.

In: Pattern Recognition Letters, Vol. 29, No. 14, 15.10.2008, p. 1934-1940.

Research output: Contribution to journalArticle

@article{756c6eb811f14437948d031fab3567fa,
title = "Gradient-based local affine invariant feature extraction for mobile robot localization in indoor environments",
abstract = "In this paper, we propose a gradient-based local affine invariant feature extraction algorithm (G-LAIFE), using affine moment invariants for robot localization in real indoor environments. The proposed algorithm is an effective feature extraction algorithm that is invariant to image translation and to 3D rotation, and it is within a partial range of the image scale. Representative performance analysis confirms that the proposed G-LAIFE algorithm significantly enhances the recognition rate and is more efficient than the scale invariant feature transform (SIFT), especially in terms of 3D rotation change and computational time.",
keywords = "3D rotation, Affine invariant, Local feature extraction, Translation",
author = "Jihyo Lee and Hanseok Ko",
year = "2008",
month = "10",
day = "15",
doi = "10.1016/j.patrec.2008.06.006",
language = "English",
volume = "29",
pages = "1934--1940",
journal = "Pattern Recognition Letters",
issn = "0167-8655",
publisher = "Elsevier",
number = "14",

}

TY - JOUR

T1 - Gradient-based local affine invariant feature extraction for mobile robot localization in indoor environments

AU - Lee, Jihyo

AU - Ko, Hanseok

PY - 2008/10/15

Y1 - 2008/10/15

N2 - In this paper, we propose a gradient-based local affine invariant feature extraction algorithm (G-LAIFE), using affine moment invariants for robot localization in real indoor environments. The proposed algorithm is an effective feature extraction algorithm that is invariant to image translation and to 3D rotation, and it is within a partial range of the image scale. Representative performance analysis confirms that the proposed G-LAIFE algorithm significantly enhances the recognition rate and is more efficient than the scale invariant feature transform (SIFT), especially in terms of 3D rotation change and computational time.

AB - In this paper, we propose a gradient-based local affine invariant feature extraction algorithm (G-LAIFE), using affine moment invariants for robot localization in real indoor environments. The proposed algorithm is an effective feature extraction algorithm that is invariant to image translation and to 3D rotation, and it is within a partial range of the image scale. Representative performance analysis confirms that the proposed G-LAIFE algorithm significantly enhances the recognition rate and is more efficient than the scale invariant feature transform (SIFT), especially in terms of 3D rotation change and computational time.

KW - 3D rotation

KW - Affine invariant

KW - Local feature extraction

KW - Translation

UR - http://www.scopus.com/inward/record.url?scp=50149087743&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=50149087743&partnerID=8YFLogxK

U2 - 10.1016/j.patrec.2008.06.006

DO - 10.1016/j.patrec.2008.06.006

M3 - Article

VL - 29

SP - 1934

EP - 1940

JO - Pattern Recognition Letters

JF - Pattern Recognition Letters

SN - 0167-8655

IS - 14

ER -