Effective venue image retrieval using robust feature extraction and model constrained matching for mobile robot localization

Yue Feng, Jinchang Ren, Jianmin Jiang, Martin Halvey, Joemon M. Jose

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

This paper describes a novel system for mobile robot localization in an indoor environment, using concepts like homography and matching borrowed from the context of stereo and content-based image retrieval techniques (CBIR). To deal with variations with respect to viewpoint and camera positions, a group of points of interest (POI) is extracted to represent the image for robust matching. To cope with illumination changes, we propose to produce a contrast image for each video frame by using the root mean square strategy, thus all the POIs are extracted from the corresponding contrast images to provide perceptually consistent measurement of image content. To achieve effective image matching, modeling of robot behavior for model constrained matching is proposed, where normalized cross correlation is employed for local matching to determine corresponding POI pairs followed by homography based global optimization using RANSAC. Meanwhile, application of specific constraints also helps to exclude irrelevant frames in the training set to further improve the efficiency and robustness. The proposed approach has been successfully applied to the Robot Vision task for the ImageCLEF workshop, and the experimental results have fully demonstrated the high-quality performance of our approaches in terms of both precision and robustness. The system and approach outlined in this paper was ranked the second best in the optional task group in ImageCLEF 2009. In addition to demonstrating the merits of our approach in isolation, we also illustrate the benefits of our proposed approach in comparison with other submissions.

LanguageEnglish
Pages1011-1027
Number of pages17
JournalMachine Vision and Applications
Volume23
Issue number5
DOIs
Publication statusPublished - 1 Sep 2012

Fingerprint

Image matching
Image retrieval
Global optimization
Mobile robots
Computer vision
Feature extraction
Lighting
Cameras
Robots

Keywords

  • content-based image retrieval
  • robot localization
  • computer vision
  • model constrained matching

Cite this

@article{99d4d028fd994f5297f44517f1ac3614,
title = "Effective venue image retrieval using robust feature extraction and model constrained matching for mobile robot localization",
abstract = "This paper describes a novel system for mobile robot localization in an indoor environment, using concepts like homography and matching borrowed from the context of stereo and content-based image retrieval techniques (CBIR). To deal with variations with respect to viewpoint and camera positions, a group of points of interest (POI) is extracted to represent the image for robust matching. To cope with illumination changes, we propose to produce a contrast image for each video frame by using the root mean square strategy, thus all the POIs are extracted from the corresponding contrast images to provide perceptually consistent measurement of image content. To achieve effective image matching, modeling of robot behavior for model constrained matching is proposed, where normalized cross correlation is employed for local matching to determine corresponding POI pairs followed by homography based global optimization using RANSAC. Meanwhile, application of specific constraints also helps to exclude irrelevant frames in the training set to further improve the efficiency and robustness. The proposed approach has been successfully applied to the Robot Vision task for the ImageCLEF workshop, and the experimental results have fully demonstrated the high-quality performance of our approaches in terms of both precision and robustness. The system and approach outlined in this paper was ranked the second best in the optional task group in ImageCLEF 2009. In addition to demonstrating the merits of our approach in isolation, we also illustrate the benefits of our proposed approach in comparison with other submissions.",
keywords = "content-based image retrieval, robot localization, computer vision, model constrained matching",
author = "Yue Feng and Jinchang Ren and Jianmin Jiang and Martin Halvey and Jose, {Joemon M.}",
year = "2012",
month = "9",
day = "1",
doi = "10.1007/s00138-011-0350-z",
language = "English",
volume = "23",
pages = "1011--1027",
journal = "Machine Vision and Applications",
issn = "0932-8092",
number = "5",

}

Effective venue image retrieval using robust feature extraction and model constrained matching for mobile robot localization. / Feng, Yue; Ren, Jinchang; Jiang, Jianmin; Halvey, Martin; Jose, Joemon M.

In: Machine Vision and Applications, Vol. 23, No. 5, 01.09.2012, p. 1011-1027.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Effective venue image retrieval using robust feature extraction and model constrained matching for mobile robot localization

AU - Feng, Yue

AU - Ren, Jinchang

AU - Jiang, Jianmin

AU - Halvey, Martin

AU - Jose, Joemon M.

PY - 2012/9/1

Y1 - 2012/9/1

N2 - This paper describes a novel system for mobile robot localization in an indoor environment, using concepts like homography and matching borrowed from the context of stereo and content-based image retrieval techniques (CBIR). To deal with variations with respect to viewpoint and camera positions, a group of points of interest (POI) is extracted to represent the image for robust matching. To cope with illumination changes, we propose to produce a contrast image for each video frame by using the root mean square strategy, thus all the POIs are extracted from the corresponding contrast images to provide perceptually consistent measurement of image content. To achieve effective image matching, modeling of robot behavior for model constrained matching is proposed, where normalized cross correlation is employed for local matching to determine corresponding POI pairs followed by homography based global optimization using RANSAC. Meanwhile, application of specific constraints also helps to exclude irrelevant frames in the training set to further improve the efficiency and robustness. The proposed approach has been successfully applied to the Robot Vision task for the ImageCLEF workshop, and the experimental results have fully demonstrated the high-quality performance of our approaches in terms of both precision and robustness. The system and approach outlined in this paper was ranked the second best in the optional task group in ImageCLEF 2009. In addition to demonstrating the merits of our approach in isolation, we also illustrate the benefits of our proposed approach in comparison with other submissions.

AB - This paper describes a novel system for mobile robot localization in an indoor environment, using concepts like homography and matching borrowed from the context of stereo and content-based image retrieval techniques (CBIR). To deal with variations with respect to viewpoint and camera positions, a group of points of interest (POI) is extracted to represent the image for robust matching. To cope with illumination changes, we propose to produce a contrast image for each video frame by using the root mean square strategy, thus all the POIs are extracted from the corresponding contrast images to provide perceptually consistent measurement of image content. To achieve effective image matching, modeling of robot behavior for model constrained matching is proposed, where normalized cross correlation is employed for local matching to determine corresponding POI pairs followed by homography based global optimization using RANSAC. Meanwhile, application of specific constraints also helps to exclude irrelevant frames in the training set to further improve the efficiency and robustness. The proposed approach has been successfully applied to the Robot Vision task for the ImageCLEF workshop, and the experimental results have fully demonstrated the high-quality performance of our approaches in terms of both precision and robustness. The system and approach outlined in this paper was ranked the second best in the optional task group in ImageCLEF 2009. In addition to demonstrating the merits of our approach in isolation, we also illustrate the benefits of our proposed approach in comparison with other submissions.

KW - content-based image retrieval

KW - robot localization

KW - computer vision

KW - model constrained matching

UR - http://www.scopus.com/inward/record.url?scp=84865375514&partnerID=8YFLogxK

U2 - 10.1007/s00138-011-0350-z

DO - 10.1007/s00138-011-0350-z

M3 - Article

VL - 23

SP - 1011

EP - 1027

JO - Machine Vision and Applications

T2 - Machine Vision and Applications

JF - Machine Vision and Applications

SN - 0932-8092

IS - 5

ER -