Aerial image sequence geolocalization with road traffic as invariant feature

Gellért Máttyus, Friedrich Fraundorfer

Research output: Contribution to journalArticleResearchpeer-review

Abstract

The geolocalization of aerial images is important for extracting geospatial information (e.g. the position of buildings, streets, and cars) and for creating maps. The standard is to use an expensive aerial imaging system equipped with an accurate GPS and IMU and/or do laborious ground control point measurements. In this paper we present a novel method to recognize the geolocation of aerial images automatically without any GPS or IMU. We extract road segments in the image sequence by detecting and tracking cars. We search in a database created from a road network map for the best matches between the road database and the extracted road segments. Geometric hashing is used to retrieve a shortlist of matches. The matches in the shortlist are ranked by a verification process. The highest scoring match gives the location and orientation of the images. We show in the experiments that our method can correctly geolocalize the aerial images in various scenes: e.g. urban, suburban, and rural with motorway. Besides the current images only the road map is needed over the search area. We can search an area of 22,500 km2 containing 32,000 km of streets within minutes on a single cpu.

LanguageEnglish
Pages218-229
Number of pages12
JournalImage and Vision Computing
Volume52
DOIs
StatusPublished - 1 Aug 2016

Fingerprint

Telecommunication traffic
Antennas
Global positioning system
Railroad cars
Imaging systems
Experiments

Keywords

  • Geolocalization
  • Geometric hashing
  • Georeferencing
  • Geotagging

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Aerial image sequence geolocalization with road traffic as invariant feature. / Máttyus, Gellért; Fraundorfer, Friedrich.

In: Image and Vision Computing, Vol. 52, 01.08.2016, p. 218-229.

Research output: Contribution to journalArticleResearchpeer-review

@article{9ab1c113c38144c78b67753ea6ee3c95,
title = "Aerial image sequence geolocalization with road traffic as invariant feature",
abstract = "The geolocalization of aerial images is important for extracting geospatial information (e.g. the position of buildings, streets, and cars) and for creating maps. The standard is to use an expensive aerial imaging system equipped with an accurate GPS and IMU and/or do laborious ground control point measurements. In this paper we present a novel method to recognize the geolocation of aerial images automatically without any GPS or IMU. We extract road segments in the image sequence by detecting and tracking cars. We search in a database created from a road network map for the best matches between the road database and the extracted road segments. Geometric hashing is used to retrieve a shortlist of matches. The matches in the shortlist are ranked by a verification process. The highest scoring match gives the location and orientation of the images. We show in the experiments that our method can correctly geolocalize the aerial images in various scenes: e.g. urban, suburban, and rural with motorway. Besides the current images only the road map is needed over the search area. We can search an area of 22,500 km2 containing 32,000 km of streets within minutes on a single cpu.",
keywords = "Geolocalization, Geometric hashing, Georeferencing, Geotagging",
author = "Gell{\'e}rt M{\'a}ttyus and Friedrich Fraundorfer",
year = "2016",
month = "8",
day = "1",
doi = "10.1016/j.imavis.2016.05.014",
language = "English",
volume = "52",
pages = "218--229",
journal = "Image and Vision Computing",
issn = "0262-8856",
publisher = "Elsevier GmbH",

}

TY - JOUR

T1 - Aerial image sequence geolocalization with road traffic as invariant feature

AU - Máttyus, Gellért

AU - Fraundorfer, Friedrich

PY - 2016/8/1

Y1 - 2016/8/1

N2 - The geolocalization of aerial images is important for extracting geospatial information (e.g. the position of buildings, streets, and cars) and for creating maps. The standard is to use an expensive aerial imaging system equipped with an accurate GPS and IMU and/or do laborious ground control point measurements. In this paper we present a novel method to recognize the geolocation of aerial images automatically without any GPS or IMU. We extract road segments in the image sequence by detecting and tracking cars. We search in a database created from a road network map for the best matches between the road database and the extracted road segments. Geometric hashing is used to retrieve a shortlist of matches. The matches in the shortlist are ranked by a verification process. The highest scoring match gives the location and orientation of the images. We show in the experiments that our method can correctly geolocalize the aerial images in various scenes: e.g. urban, suburban, and rural with motorway. Besides the current images only the road map is needed over the search area. We can search an area of 22,500 km2 containing 32,000 km of streets within minutes on a single cpu.

AB - The geolocalization of aerial images is important for extracting geospatial information (e.g. the position of buildings, streets, and cars) and for creating maps. The standard is to use an expensive aerial imaging system equipped with an accurate GPS and IMU and/or do laborious ground control point measurements. In this paper we present a novel method to recognize the geolocation of aerial images automatically without any GPS or IMU. We extract road segments in the image sequence by detecting and tracking cars. We search in a database created from a road network map for the best matches between the road database and the extracted road segments. Geometric hashing is used to retrieve a shortlist of matches. The matches in the shortlist are ranked by a verification process. The highest scoring match gives the location and orientation of the images. We show in the experiments that our method can correctly geolocalize the aerial images in various scenes: e.g. urban, suburban, and rural with motorway. Besides the current images only the road map is needed over the search area. We can search an area of 22,500 km2 containing 32,000 km of streets within minutes on a single cpu.

KW - Geolocalization

KW - Geometric hashing

KW - Georeferencing

KW - Geotagging

UR - http://www.scopus.com/inward/record.url?scp=84978539267&partnerID=8YFLogxK

U2 - 10.1016/j.imavis.2016.05.014

DO - 10.1016/j.imavis.2016.05.014

M3 - Article

VL - 52

SP - 218

EP - 229

JO - Image and Vision Computing

T2 - Image and Vision Computing

JF - Image and Vision Computing

SN - 0262-8856

ER -