Correcting Misaligned Rural Building Annotations in Open Street Map Using Convolutional Neural Networks Evidence

John E. Vargas-Munoz, Diego Marcos, Sylvain Lobry, Jefersson A. dos Santos, Alexandre X. Falcao, Devis Tuia

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

3 Citations (Scopus)

Abstract

Mapping rural buildings in developing countries is crucial to monitor and plan in those vulnerable areas. Despite the existence of some rural building annotations in OpenStreetMap (OSM), those are of insufficient quantity and quality to train models able to map large areas accurately. In particular, these annotations are very often misaligned with respect to the buildings that are present in updated aerial imagery. We propose a Markov Random Field (MRF) method to correct misaligned rural building annotations. To do so, our method uses i) the correlation between candidate aligned OSM annotations and buildings roughly detected on aerial images and ii) the local consistency of the alignment vectors.
Original languageEnglish
Title of host publication2018 IEEE International Geoscience & Remote Sensing Symposium Proceedings
Subtitle of host publicationObserving, Understanding And Forecasting The Dynamics Of Our Planet
PublisherIEEE Xplore
Pages1284-1287
ISBN (Electronic)9781538671504, 9781538671498
ISBN (Print)9781538671511
DOIs
Publication statusPublished - 5 Nov 2018
EventIGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain
Duration: 22 Jul 201827 Jul 2018

Conference

ConferenceIGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium
CountrySpain
CityValencia
Period22/07/1827/07/18

Fingerprint

field method
train
imagery
developing world
alignment
plan
method

Cite this

Vargas-Munoz, J. E., Marcos, D., Lobry, S., dos Santos, J. A., Falcao, A. X., & Tuia, D. (2018). Correcting Misaligned Rural Building Annotations in Open Street Map Using Convolutional Neural Networks Evidence. In 2018 IEEE International Geoscience & Remote Sensing Symposium Proceedings: Observing, Understanding And Forecasting The Dynamics Of Our Planet (pp. 1284-1287). IEEE Xplore. https://doi.org/10.1109/IGARSS.2018.8518711
Vargas-Munoz, John E. ; Marcos, Diego ; Lobry, Sylvain ; dos Santos, Jefersson A. ; Falcao, Alexandre X. ; Tuia, Devis. / Correcting Misaligned Rural Building Annotations in Open Street Map Using Convolutional Neural Networks Evidence. 2018 IEEE International Geoscience & Remote Sensing Symposium Proceedings: Observing, Understanding And Forecasting The Dynamics Of Our Planet. IEEE Xplore, 2018. pp. 1284-1287
@inproceedings{2ee953af175546ddbe788a7c2f0b92d7,
title = "Correcting Misaligned Rural Building Annotations in Open Street Map Using Convolutional Neural Networks Evidence",
abstract = "Mapping rural buildings in developing countries is crucial to monitor and plan in those vulnerable areas. Despite the existence of some rural building annotations in OpenStreetMap (OSM), those are of insufficient quantity and quality to train models able to map large areas accurately. In particular, these annotations are very often misaligned with respect to the buildings that are present in updated aerial imagery. We propose a Markov Random Field (MRF) method to correct misaligned rural building annotations. To do so, our method uses i) the correlation between candidate aligned OSM annotations and buildings roughly detected on aerial images and ii) the local consistency of the alignment vectors.",
author = "Vargas-Munoz, {John E.} and Diego Marcos and Sylvain Lobry and {dos Santos}, {Jefersson A.} and Falcao, {Alexandre X.} and Devis Tuia",
year = "2018",
month = "11",
day = "5",
doi = "10.1109/IGARSS.2018.8518711",
language = "English",
isbn = "9781538671511",
pages = "1284--1287",
booktitle = "2018 IEEE International Geoscience & Remote Sensing Symposium Proceedings",
publisher = "IEEE Xplore",

}

Vargas-Munoz, JE, Marcos, D, Lobry, S, dos Santos, JA, Falcao, AX & Tuia, D 2018, Correcting Misaligned Rural Building Annotations in Open Street Map Using Convolutional Neural Networks Evidence. in 2018 IEEE International Geoscience & Remote Sensing Symposium Proceedings: Observing, Understanding And Forecasting The Dynamics Of Our Planet. IEEE Xplore, pp. 1284-1287, IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium, Valencia, Spain, 22/07/18. https://doi.org/10.1109/IGARSS.2018.8518711

Correcting Misaligned Rural Building Annotations in Open Street Map Using Convolutional Neural Networks Evidence. / Vargas-Munoz, John E.; Marcos, Diego; Lobry, Sylvain; dos Santos, Jefersson A.; Falcao, Alexandre X.; Tuia, Devis.

2018 IEEE International Geoscience & Remote Sensing Symposium Proceedings: Observing, Understanding And Forecasting The Dynamics Of Our Planet. IEEE Xplore, 2018. p. 1284-1287.

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

TY - GEN

T1 - Correcting Misaligned Rural Building Annotations in Open Street Map Using Convolutional Neural Networks Evidence

AU - Vargas-Munoz, John E.

AU - Marcos, Diego

AU - Lobry, Sylvain

AU - dos Santos, Jefersson A.

AU - Falcao, Alexandre X.

AU - Tuia, Devis

PY - 2018/11/5

Y1 - 2018/11/5

N2 - Mapping rural buildings in developing countries is crucial to monitor and plan in those vulnerable areas. Despite the existence of some rural building annotations in OpenStreetMap (OSM), those are of insufficient quantity and quality to train models able to map large areas accurately. In particular, these annotations are very often misaligned with respect to the buildings that are present in updated aerial imagery. We propose a Markov Random Field (MRF) method to correct misaligned rural building annotations. To do so, our method uses i) the correlation between candidate aligned OSM annotations and buildings roughly detected on aerial images and ii) the local consistency of the alignment vectors.

AB - Mapping rural buildings in developing countries is crucial to monitor and plan in those vulnerable areas. Despite the existence of some rural building annotations in OpenStreetMap (OSM), those are of insufficient quantity and quality to train models able to map large areas accurately. In particular, these annotations are very often misaligned with respect to the buildings that are present in updated aerial imagery. We propose a Markov Random Field (MRF) method to correct misaligned rural building annotations. To do so, our method uses i) the correlation between candidate aligned OSM annotations and buildings roughly detected on aerial images and ii) the local consistency of the alignment vectors.

U2 - 10.1109/IGARSS.2018.8518711

DO - 10.1109/IGARSS.2018.8518711

M3 - Conference paper

SN - 9781538671511

SP - 1284

EP - 1287

BT - 2018 IEEE International Geoscience & Remote Sensing Symposium Proceedings

PB - IEEE Xplore

ER -

Vargas-Munoz JE, Marcos D, Lobry S, dos Santos JA, Falcao AX, Tuia D. Correcting Misaligned Rural Building Annotations in Open Street Map Using Convolutional Neural Networks Evidence. In 2018 IEEE International Geoscience & Remote Sensing Symposium Proceedings: Observing, Understanding And Forecasting The Dynamics Of Our Planet. IEEE Xplore. 2018. p. 1284-1287 https://doi.org/10.1109/IGARSS.2018.8518711