Landcover classification with self-taught learning on archetypal dictionaries

Ribana Roscher; Christoph Romer; Bjorn Waske; Lutz Plumer

doi:10.1109/IGARSS.2015.7326282

Landcover classification with self-taught learning on archetypal dictionaries

Ribana Roscher, Christoph Romer, Bjorn Waske, Lutz Plumer

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

4 Scopus citations

Abstract

This paper introduces archetypal dictionaries for a self-taught learning framework for the application of landcover classification. Self-taught learning, an unsupervised representation learning method, is exploited to learn low-dimensional and discriminative higher-level features, which are used as input into a classification algorithm. Experiments are conducted using a multi-spectral Landsat 5 TM image of a study area in the north of Novo Progresso located in South America. Our results confirm that self-taught learning with archetypal dictionaries provide features, which can be used as input into a linear logistic regression classifier. The obtained classification accuracies are comparable to kernel-based classifier using the original features.

Original language	English
Title of host publication	2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	2358-2361
Number of pages	4
ISBN (Electronic)	9781479979295
DOIs	https://doi.org/10.1109/IGARSS.2015.7326282
State	Published - 10 Nov 2015
Externally published	Yes
Event	IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Milan, Italy Duration: 26 Jul 2015 → 31 Jul 2015

Publication series

Name	International Geoscience and Remote Sensing Symposium (IGARSS)
Volume	2015-November

Conference

Conference	IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015
Country/Territory	Italy
City	Milan
Period	26/07/15 → 31/07/15

Keywords

archetypal analysis
landcover classification
self-taught learning

Access to Document

10.1109/IGARSS.2015.7326282

Cite this

Roscher, R., Romer, C., Waske, B., & Plumer, L. (2015). Landcover classification with self-taught learning on archetypal dictionaries. In 2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings (pp. 2358-2361). Article 7326282 (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 2015-November). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IGARSS.2015.7326282

Roscher, Ribana ; Romer, Christoph ; Waske, Bjorn et al. / Landcover classification with self-taught learning on archetypal dictionaries. 2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 2358-2361 (International Geoscience and Remote Sensing Symposium (IGARSS)).

@inproceedings{70131850ba2644959866b14bb767c5df,

title = "Landcover classification with self-taught learning on archetypal dictionaries",

abstract = "This paper introduces archetypal dictionaries for a self-taught learning framework for the application of landcover classification. Self-taught learning, an unsupervised representation learning method, is exploited to learn low-dimensional and discriminative higher-level features, which are used as input into a classification algorithm. Experiments are conducted using a multi-spectral Landsat 5 TM image of a study area in the north of Novo Progresso located in South America. Our results confirm that self-taught learning with archetypal dictionaries provide features, which can be used as input into a linear logistic regression classifier. The obtained classification accuracies are comparable to kernel-based classifier using the original features.",

keywords = "archetypal analysis, landcover classification, self-taught learning",

author = "Ribana Roscher and Christoph Romer and Bjorn Waske and Lutz Plumer",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 ; Conference date: 26-07-2015 Through 31-07-2015",

year = "2015",

month = nov,

day = "10",

doi = "10.1109/IGARSS.2015.7326282",

language = "English",

series = "International Geoscience and Remote Sensing Symposium (IGARSS)",

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

pages = "2358--2361",

booktitle = "2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings",

}

Roscher, R, Romer, C, Waske, B & Plumer, L 2015, Landcover classification with self-taught learning on archetypal dictionaries. in 2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings., 7326282, International Geoscience and Remote Sensing Symposium (IGARSS), vol. 2015-November, Institute of Electrical and Electronics Engineers Inc., pp. 2358-2361, IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015, Milan, Italy, 26/07/15. https://doi.org/10.1109/IGARSS.2015.7326282

Landcover classification with self-taught learning on archetypal dictionaries. / Roscher, Ribana; Romer, Christoph; Waske, Bjorn et al.
2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2015. p. 2358-2361 7326282 (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 2015-November).

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

TY - GEN

T1 - Landcover classification with self-taught learning on archetypal dictionaries

AU - Roscher, Ribana

AU - Romer, Christoph

AU - Waske, Bjorn

AU - Plumer, Lutz

PY - 2015/11/10

Y1 - 2015/11/10

N2 - This paper introduces archetypal dictionaries for a self-taught learning framework for the application of landcover classification. Self-taught learning, an unsupervised representation learning method, is exploited to learn low-dimensional and discriminative higher-level features, which are used as input into a classification algorithm. Experiments are conducted using a multi-spectral Landsat 5 TM image of a study area in the north of Novo Progresso located in South America. Our results confirm that self-taught learning with archetypal dictionaries provide features, which can be used as input into a linear logistic regression classifier. The obtained classification accuracies are comparable to kernel-based classifier using the original features.

AB - This paper introduces archetypal dictionaries for a self-taught learning framework for the application of landcover classification. Self-taught learning, an unsupervised representation learning method, is exploited to learn low-dimensional and discriminative higher-level features, which are used as input into a classification algorithm. Experiments are conducted using a multi-spectral Landsat 5 TM image of a study area in the north of Novo Progresso located in South America. Our results confirm that self-taught learning with archetypal dictionaries provide features, which can be used as input into a linear logistic regression classifier. The obtained classification accuracies are comparable to kernel-based classifier using the original features.

KW - archetypal analysis

KW - landcover classification

KW - self-taught learning

UR - http://www.scopus.com/inward/record.url?scp=84962581216&partnerID=8YFLogxK

U2 - 10.1109/IGARSS.2015.7326282

DO - 10.1109/IGARSS.2015.7326282

M3 - Conference contribution

AN - SCOPUS:84962581216

T3 - International Geoscience and Remote Sensing Symposium (IGARSS)

SP - 2358

EP - 2361

BT - 2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015

Y2 - 26 July 2015 through 31 July 2015

ER -

Roscher R, Romer C, Waske B, Plumer L. Landcover classification with self-taught learning on archetypal dictionaries. In 2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2015. p. 2358-2361. 7326282. (International Geoscience and Remote Sensing Symposium (IGARSS)). doi: 10.1109/IGARSS.2015.7326282

Landcover classification with self-taught learning on archetypal dictionaries

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this