Mapping human settlements with multi-seasonal sentinel-2 imagery and attention-based resnext

Chunping Qiu; Michael Schmitt; Hannes Taubenbock; Xiao Xiang Zhu

doi:10.1109/JURSE.2019.8809009

Mapping human settlements with multi-seasonal sentinel-2 imagery and attention-based resnext

Chunping Qiu, Michael Schmitt, Hannes Taubenbock, Xiao Xiang Zhu

Chair of Data Science in Earth Observation

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

7 Scopus citations

Abstract

This paper explores the potential of multi-spectral Sentinel-2 imagery for human settlement mapping, using deep learning based methods. We show first results of a study area in central Europe, with an attention-based ResNeXt to better exploit the spectral information. Reasonable mapping accuracy has been achieved, compared to the state-of-the-art products. Based on the results and comparison with the existing products, we discuss two interesting questions: How can human settlement mapping be made consistent with or complementary to the existing human settlement maps and how can further improvement in human settlement mapping be achieved by exploring deep learning-based approaches.

Original language	English
Title of host publication	2019 Joint Urban Remote Sensing Event, JURSE 2019
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728100098
DOIs	https://doi.org/10.1109/JURSE.2019.8809009
State	Published - May 2019
Event	2019 Joint Urban Remote Sensing Event, JURSE 2019 - Vannes, France Duration: 22 May 2019 → 24 May 2019

Publication series

Name	2019 Joint Urban Remote Sensing Event, JURSE 2019

Conference

Conference	2019 Joint Urban Remote Sensing Event, JURSE 2019
Country/Territory	France
City	Vannes
Period	22/05/19 → 24/05/19

Keywords

Sentinel-2
attention
classification
convolutional neural network (CNN)
human settlement (HS) mapping

Access to Document

10.1109/JURSE.2019.8809009

Cite this

Qiu, C., Schmitt, M., Taubenbock, H., & Zhu, X. X. (2019). Mapping human settlements with multi-seasonal sentinel-2 imagery and attention-based resnext. In 2019 Joint Urban Remote Sensing Event, JURSE 2019 Article 8809009 (2019 Joint Urban Remote Sensing Event, JURSE 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/JURSE.2019.8809009

@inproceedings{b2b8e2da9d364d428748bb996989b7ac,

title = "Mapping human settlements with multi-seasonal sentinel-2 imagery and attention-based resnext",

abstract = "This paper explores the potential of multi-spectral Sentinel-2 imagery for human settlement mapping, using deep learning based methods. We show first results of a study area in central Europe, with an attention-based ResNeXt to better exploit the spectral information. Reasonable mapping accuracy has been achieved, compared to the state-of-the-art products. Based on the results and comparison with the existing products, we discuss two interesting questions: How can human settlement mapping be made consistent with or complementary to the existing human settlement maps and how can further improvement in human settlement mapping be achieved by exploring deep learning-based approaches.",

keywords = "Sentinel-2, attention, classification, convolutional neural network (CNN), human settlement (HS) mapping",

author = "Chunping Qiu and Michael Schmitt and Hannes Taubenbock and Zhu, {Xiao Xiang}",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 2019 Joint Urban Remote Sensing Event, JURSE 2019 ; Conference date: 22-05-2019 Through 24-05-2019",

year = "2019",

month = may,

doi = "10.1109/JURSE.2019.8809009",

language = "English",

series = "2019 Joint Urban Remote Sensing Event, JURSE 2019",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2019 Joint Urban Remote Sensing Event, JURSE 2019",

}

Qiu, C, Schmitt, M, Taubenbock, H & Zhu, XX 2019, Mapping human settlements with multi-seasonal sentinel-2 imagery and attention-based resnext. in 2019 Joint Urban Remote Sensing Event, JURSE 2019., 8809009, 2019 Joint Urban Remote Sensing Event, JURSE 2019, Institute of Electrical and Electronics Engineers Inc., 2019 Joint Urban Remote Sensing Event, JURSE 2019, Vannes, France, 22/05/19. https://doi.org/10.1109/JURSE.2019.8809009

Mapping human settlements with multi-seasonal sentinel-2 imagery and attention-based resnext. / Qiu, Chunping; Schmitt, Michael; Taubenbock, Hannes et al.
2019 Joint Urban Remote Sensing Event, JURSE 2019. Institute of Electrical and Electronics Engineers Inc., 2019. 8809009 (2019 Joint Urban Remote Sensing Event, JURSE 2019).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Mapping human settlements with multi-seasonal sentinel-2 imagery and attention-based resnext

AU - Qiu, Chunping

AU - Schmitt, Michael

AU - Taubenbock, Hannes

AU - Zhu, Xiao Xiang

PY - 2019/5

Y1 - 2019/5

N2 - This paper explores the potential of multi-spectral Sentinel-2 imagery for human settlement mapping, using deep learning based methods. We show first results of a study area in central Europe, with an attention-based ResNeXt to better exploit the spectral information. Reasonable mapping accuracy has been achieved, compared to the state-of-the-art products. Based on the results and comparison with the existing products, we discuss two interesting questions: How can human settlement mapping be made consistent with or complementary to the existing human settlement maps and how can further improvement in human settlement mapping be achieved by exploring deep learning-based approaches.

AB - This paper explores the potential of multi-spectral Sentinel-2 imagery for human settlement mapping, using deep learning based methods. We show first results of a study area in central Europe, with an attention-based ResNeXt to better exploit the spectral information. Reasonable mapping accuracy has been achieved, compared to the state-of-the-art products. Based on the results and comparison with the existing products, we discuss two interesting questions: How can human settlement mapping be made consistent with or complementary to the existing human settlement maps and how can further improvement in human settlement mapping be achieved by exploring deep learning-based approaches.

KW - Sentinel-2

KW - attention

KW - classification

KW - convolutional neural network (CNN)

KW - human settlement (HS) mapping

UR - http://www.scopus.com/inward/record.url?scp=85072024145&partnerID=8YFLogxK

U2 - 10.1109/JURSE.2019.8809009

DO - 10.1109/JURSE.2019.8809009

M3 - Conference contribution

AN - SCOPUS:85072024145

T3 - 2019 Joint Urban Remote Sensing Event, JURSE 2019

BT - 2019 Joint Urban Remote Sensing Event, JURSE 2019

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2019 Joint Urban Remote Sensing Event, JURSE 2019

Y2 - 22 May 2019 through 24 May 2019

ER -

Mapping human settlements with multi-seasonal sentinel-2 imagery and attention-based resnext

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this