Automatic Illumination-Invariant Image-to-Geometry Registration in Outdoor Environments

Christian Kehl, Simon J. Buckley, Sophie Viseur, Robert L. Gawthorpe, John A. Howell

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Image-to-geometry registration is the basis of many applications for texturing and interpreting 3D surface models. Feature-based matching is an established, automatic approach which creates 2D–3D correspondences based on salient points and their radiometric neighbourhood. This paper presents an experimental approach for assessing the accuracy of several matching algorithms in challenging imaging environments that are subject to significant outdoor illumination variations. Furthermore, a collection of accuracy assessment metrics and quality heuristics emerge from the presented approach to guide a user during the examination of registration results. As a result of the experiments, two novel salient point descriptor matching combinations outperform the standard scale-invariant feature transform (SIFT) operator on the task of image-to-image and image-to-geometry registration under varying illumination conditions.

Original languageEnglish
Pages (from-to)93-118
Number of pages26
JournalPhotogrammetric Record
Volume32
Issue number158
Early online date14 Jun 2017
DOIs
Publication statusPublished - Jun 2017

Fingerprint

Lighting
geometry
Geometry
Texturing
Imaging techniques
accuracy assessment
heuristics
transform
Experiments
registration
experiment

Keywords

  • feature-based registration
  • illumination variances
  • image-to-geometry
  • interest operators
  • mobile device imagery
  • outdoor environments

ASJC Scopus subject areas

  • Engineering (miscellaneous)
  • Computer Science Applications
  • Computers in Earth Sciences
  • Earth and Planetary Sciences (miscellaneous)

Cite this

Automatic Illumination-Invariant Image-to-Geometry Registration in Outdoor Environments. / Kehl, Christian; Buckley, Simon J.; Viseur, Sophie; Gawthorpe, Robert L.; Howell, John A.

In: Photogrammetric Record, Vol. 32, No. 158, 06.2017, p. 93-118.

Research output: Contribution to journalArticle

Kehl, Christian ; Buckley, Simon J. ; Viseur, Sophie ; Gawthorpe, Robert L. ; Howell, John A. / Automatic Illumination-Invariant Image-to-Geometry Registration in Outdoor Environments. In: Photogrammetric Record. 2017 ; Vol. 32, No. 158. pp. 93-118.
@article{ce5fe33b80cc4972a634627e9adda70d,
title = "Automatic Illumination-Invariant Image-to-Geometry Registration in Outdoor Environments",
abstract = "Image-to-geometry registration is the basis of many applications for texturing and interpreting 3D surface models. Feature-based matching is an established, automatic approach which creates 2D–3D correspondences based on salient points and their radiometric neighbourhood. This paper presents an experimental approach for assessing the accuracy of several matching algorithms in challenging imaging environments that are subject to significant outdoor illumination variations. Furthermore, a collection of accuracy assessment metrics and quality heuristics emerge from the presented approach to guide a user during the examination of registration results. As a result of the experiments, two novel salient point descriptor matching combinations outperform the standard scale-invariant feature transform (SIFT) operator on the task of image-to-image and image-to-geometry registration under varying illumination conditions.",
keywords = "feature-based registration, illumination variances, image-to-geometry, interest operators, mobile device imagery, outdoor environments",
author = "Christian Kehl and Buckley, {Simon J.} and Sophie Viseur and Gawthorpe, {Robert L.} and Howell, {John A.}",
note = "Acknowledgements This research is part of the VOM2MPS project (no. 234111/E30), funded by the Research Council of Norway (RCN) and the FORCE consortium through Petromaks 2 and SAFARI. Data are collected and provided in the framework of SAFARI (www.safaridb.com).",
year = "2017",
month = "6",
doi = "10.1111/phor.12188",
language = "English",
volume = "32",
pages = "93--118",
journal = "Photogrammetric Record",
issn = "0031-868X",
publisher = "Wiley-Blackwell",
number = "158",

}

TY - JOUR

T1 - Automatic Illumination-Invariant Image-to-Geometry Registration in Outdoor Environments

AU - Kehl, Christian

AU - Buckley, Simon J.

AU - Viseur, Sophie

AU - Gawthorpe, Robert L.

AU - Howell, John A.

N1 - Acknowledgements This research is part of the VOM2MPS project (no. 234111/E30), funded by the Research Council of Norway (RCN) and the FORCE consortium through Petromaks 2 and SAFARI. Data are collected and provided in the framework of SAFARI (www.safaridb.com).

PY - 2017/6

Y1 - 2017/6

N2 - Image-to-geometry registration is the basis of many applications for texturing and interpreting 3D surface models. Feature-based matching is an established, automatic approach which creates 2D–3D correspondences based on salient points and their radiometric neighbourhood. This paper presents an experimental approach for assessing the accuracy of several matching algorithms in challenging imaging environments that are subject to significant outdoor illumination variations. Furthermore, a collection of accuracy assessment metrics and quality heuristics emerge from the presented approach to guide a user during the examination of registration results. As a result of the experiments, two novel salient point descriptor matching combinations outperform the standard scale-invariant feature transform (SIFT) operator on the task of image-to-image and image-to-geometry registration under varying illumination conditions.

AB - Image-to-geometry registration is the basis of many applications for texturing and interpreting 3D surface models. Feature-based matching is an established, automatic approach which creates 2D–3D correspondences based on salient points and their radiometric neighbourhood. This paper presents an experimental approach for assessing the accuracy of several matching algorithms in challenging imaging environments that are subject to significant outdoor illumination variations. Furthermore, a collection of accuracy assessment metrics and quality heuristics emerge from the presented approach to guide a user during the examination of registration results. As a result of the experiments, two novel salient point descriptor matching combinations outperform the standard scale-invariant feature transform (SIFT) operator on the task of image-to-image and image-to-geometry registration under varying illumination conditions.

KW - feature-based registration

KW - illumination variances

KW - image-to-geometry

KW - interest operators

KW - mobile device imagery

KW - outdoor environments

UR - http://www.scopus.com/inward/record.url?scp=85020618637&partnerID=8YFLogxK

U2 - 10.1111/phor.12188

DO - 10.1111/phor.12188

M3 - Article

VL - 32

SP - 93

EP - 118

JO - Photogrammetric Record

JF - Photogrammetric Record

SN - 0031-868X

IS - 158

ER -