Fully Convolutional Networks for Multi-temporal SAR Image Classification

A.G. Mullissa, C. Persello, V.A. Tolpekin

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

2 Citations (Scopus)
4 Downloads (Pure)

Abstract

Classification of crop types from multi-temporal SAR data is a complex task because of the need to extract spatial and temporal features from images affected by speckle. Previous methods applied speckle filtering and then classification in
two separate processing steps. This paper introduces fully convolutional networks (FCN) for pixel-wise classification of crops from multi-temporal SAR data. It applies speckle filtering and classification in a single framework. Furthermore, it also uses dilated kernels to increase the capability to learn long distance spatial dependencies. The proposed FCN was compared with a patch-based convolutional neural network (CNN) and support vector machine (SVM) classifiers. The proposed method performed better when compared with the patch-based CNN and SVM.
Original languageEnglish
Title of host publicationIGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium
PublisherIEEE
Pages6635-3338
Number of pages4
ISBN (Electronic)978-1-5386-7150-4
DOIs
Publication statusPublished - 5 Nov 2018
Event38th IEEE International Geoscience and Remote Sensing Symposium 2018: Observing, Understanding and Forcasting the Dynamics of Our Planet - Feria Valencia Convention & Exhibition Center, Valencia, Spain
Duration: 22 Jul 201827 Jul 2018
Conference number: 38
https://www.igarss2018.org/

Conference

Conference38th IEEE International Geoscience and Remote Sensing Symposium 2018
Abbreviated titleIGARSS 2018
CountrySpain
CityValencia
Period22/07/1827/07/18
Internet address

Fingerprint

Image classification
image classification
speckle
synthetic aperture radar
Speckle
Crops
Support vector machines
Neural networks
crop
pixel
Classifiers
Pixels
Processing
method
support vector machine

Keywords

  • Deep learning
  • Fully convolutional networks
  • Remote Sensing
  • SAR
  • Sentinel-1

Cite this

Mullissa, A. G., Persello, C., & Tolpekin, V. A. (2018). Fully Convolutional Networks for Multi-temporal SAR Image Classification. In IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium (pp. 6635-3338). [8518780] IEEE. https://doi.org/10.1109/IGARSS.2018.8518780
Mullissa, A.G. ; Persello, C. ; Tolpekin, V.A. / Fully Convolutional Networks for Multi-temporal SAR Image Classification. IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2018. pp. 6635-3338
@inbook{a3bb1e48c01e4f3ea555339cb5b0662a,
title = "Fully Convolutional Networks for Multi-temporal SAR Image Classification",
abstract = "Classification of crop types from multi-temporal SAR data is a complex task because of the need to extract spatial and temporal features from images affected by speckle. Previous methods applied speckle filtering and then classification intwo separate processing steps. This paper introduces fully convolutional networks (FCN) for pixel-wise classification of crops from multi-temporal SAR data. It applies speckle filtering and classification in a single framework. Furthermore, it also uses dilated kernels to increase the capability to learn long distance spatial dependencies. The proposed FCN was compared with a patch-based convolutional neural network (CNN) and support vector machine (SVM) classifiers. The proposed method performed better when compared with the patch-based CNN and SVM.",
keywords = "Deep learning, Fully convolutional networks, Remote Sensing, SAR, Sentinel-1",
author = "A.G. Mullissa and C. Persello and V.A. Tolpekin",
year = "2018",
month = "11",
day = "5",
doi = "10.1109/IGARSS.2018.8518780",
language = "English",
pages = "6635--3338",
booktitle = "IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium",
publisher = "IEEE",
address = "United States",

}

Mullissa, AG, Persello, C & Tolpekin, VA 2018, Fully Convolutional Networks for Multi-temporal SAR Image Classification. in IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium., 8518780, IEEE, pp. 6635-3338, 38th IEEE International Geoscience and Remote Sensing Symposium 2018, Valencia, Spain, 22/07/18. https://doi.org/10.1109/IGARSS.2018.8518780

Fully Convolutional Networks for Multi-temporal SAR Image Classification. / Mullissa, A.G.; Persello, C.; Tolpekin, V.A.

IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2018. p. 6635-3338 8518780.

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

TY - CHAP

T1 - Fully Convolutional Networks for Multi-temporal SAR Image Classification

AU - Mullissa, A.G.

AU - Persello, C.

AU - Tolpekin, V.A.

PY - 2018/11/5

Y1 - 2018/11/5

N2 - Classification of crop types from multi-temporal SAR data is a complex task because of the need to extract spatial and temporal features from images affected by speckle. Previous methods applied speckle filtering and then classification intwo separate processing steps. This paper introduces fully convolutional networks (FCN) for pixel-wise classification of crops from multi-temporal SAR data. It applies speckle filtering and classification in a single framework. Furthermore, it also uses dilated kernels to increase the capability to learn long distance spatial dependencies. The proposed FCN was compared with a patch-based convolutional neural network (CNN) and support vector machine (SVM) classifiers. The proposed method performed better when compared with the patch-based CNN and SVM.

AB - Classification of crop types from multi-temporal SAR data is a complex task because of the need to extract spatial and temporal features from images affected by speckle. Previous methods applied speckle filtering and then classification intwo separate processing steps. This paper introduces fully convolutional networks (FCN) for pixel-wise classification of crops from multi-temporal SAR data. It applies speckle filtering and classification in a single framework. Furthermore, it also uses dilated kernels to increase the capability to learn long distance spatial dependencies. The proposed FCN was compared with a patch-based convolutional neural network (CNN) and support vector machine (SVM) classifiers. The proposed method performed better when compared with the patch-based CNN and SVM.

KW - Deep learning

KW - Fully convolutional networks

KW - Remote Sensing

KW - SAR

KW - Sentinel-1

UR - https://ezproxy2.utwente.nl/login?url=https://webapps.itc.utwente.nl/library/2018/chap/persello_full.pdf

UR - http://www.scopus.com/inward/record.url?scp=85064172068&partnerID=8YFLogxK

U2 - 10.1109/IGARSS.2018.8518780

DO - 10.1109/IGARSS.2018.8518780

M3 - Chapter

SP - 6635

EP - 3338

BT - IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium

PB - IEEE

ER -

Mullissa AG, Persello C, Tolpekin VA. Fully Convolutional Networks for Multi-temporal SAR Image Classification. In IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium. IEEE. 2018. p. 6635-3338. 8518780 https://doi.org/10.1109/IGARSS.2018.8518780