Details
| Original language | English |
|---|---|
| Pages (from-to) | 251-258 |
| Number of pages | 8 |
| Journal | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Volume | 4 |
| Issue number | 3 |
| Publication status | Published - 23 Apr 2018 |
| Event | 2018 ISPRS TC III Mid-Term Symposium on Developments, Technologies and Applications in Remote Sensing - Beijing, China Duration: 7 May 2018 → 10 May 2018 |
Abstract
Land cover describes the physical material of the earth's surface, whereas land use describes the socio-economic function of a piece of land. Land use information is typically collected in geospatial databases. As such databases become outdated quickly, an automatic update process is required. This paper presents a new approach to determine land cover and to classify land use objects based on convolutional neural networks (CNN). The input data are aerial images and derived data such as digital surface models. Firstly, we apply a CNN to determine the land cover for each pixel of the input image. We compare different CNN structures, all of them based on an encoder-decoder structure for obtaining dense class predictions. Secondly, we propose a new CNN-based methodology for the prediction of the land use label of objects from a geospatial database. In this context, we present a strategy for generating image patches of identical size from the input data, which are classified by a CNN. Again, we compare different CNN architectures. Our experiments show that an overall accuracy of up to 85.7 % and 77.4 % can be achieved for land cover and land use, respectively. The classification of land cover has a positive contribution to the classification of the land use classification.
Keywords
- aerial imagery, CNN, geospatial land use database, Land use classification, semantic segmentation
ASJC Scopus subject areas
- Earth and Planetary Sciences(all)
- Earth and Planetary Sciences (miscellaneous)
- Environmental Science(all)
- Environmental Science (miscellaneous)
- Physics and Astronomy(all)
- Instrumentation
Sustainable Development Goals
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. 4, No. 3, 23.04.2018, p. 251-258.
Research output: Contribution to journal › Conference article › Research › peer review
}
TY - JOUR
T1 - Classification of land cover and land use based on convolutional neural networks
AU - Yang, Chun
AU - Rottensteiner, Franz
AU - Heipke, Christian
N1 - Funding Information: We thank the Landesamt für Geoinformation und Landes-vermessung Niedersachsen(LGLN), the Landesamt für Vermessung und Geoinformation Schleswig Holstein (LVermGeo) and Landesamt für innere Verwaltung Mecklenburg-Vorpommern (LaiV-MV) for providing the test data and for their support of this project. The first author is an associate member of the Research Training Group i.c.sens (GRK 2159), funded by the German Research Foundation (DFG).
PY - 2018/4/23
Y1 - 2018/4/23
N2 - Land cover describes the physical material of the earth's surface, whereas land use describes the socio-economic function of a piece of land. Land use information is typically collected in geospatial databases. As such databases become outdated quickly, an automatic update process is required. This paper presents a new approach to determine land cover and to classify land use objects based on convolutional neural networks (CNN). The input data are aerial images and derived data such as digital surface models. Firstly, we apply a CNN to determine the land cover for each pixel of the input image. We compare different CNN structures, all of them based on an encoder-decoder structure for obtaining dense class predictions. Secondly, we propose a new CNN-based methodology for the prediction of the land use label of objects from a geospatial database. In this context, we present a strategy for generating image patches of identical size from the input data, which are classified by a CNN. Again, we compare different CNN architectures. Our experiments show that an overall accuracy of up to 85.7 % and 77.4 % can be achieved for land cover and land use, respectively. The classification of land cover has a positive contribution to the classification of the land use classification.
AB - Land cover describes the physical material of the earth's surface, whereas land use describes the socio-economic function of a piece of land. Land use information is typically collected in geospatial databases. As such databases become outdated quickly, an automatic update process is required. This paper presents a new approach to determine land cover and to classify land use objects based on convolutional neural networks (CNN). The input data are aerial images and derived data such as digital surface models. Firstly, we apply a CNN to determine the land cover for each pixel of the input image. We compare different CNN structures, all of them based on an encoder-decoder structure for obtaining dense class predictions. Secondly, we propose a new CNN-based methodology for the prediction of the land use label of objects from a geospatial database. In this context, we present a strategy for generating image patches of identical size from the input data, which are classified by a CNN. Again, we compare different CNN architectures. Our experiments show that an overall accuracy of up to 85.7 % and 77.4 % can be achieved for land cover and land use, respectively. The classification of land cover has a positive contribution to the classification of the land use classification.
KW - aerial imagery
KW - CNN
KW - geospatial land use database
KW - Land use classification
KW - semantic segmentation
UR - http://www.scopus.com/inward/record.url?scp=85046733837&partnerID=8YFLogxK
U2 - 10.5194/isprs-annals-IV-3-251-2018
DO - 10.5194/isprs-annals-IV-3-251-2018
M3 - Conference article
AN - SCOPUS:85046733837
VL - 4
SP - 251
EP - 258
JO - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
JF - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
SN - 2194-9042
IS - 3
T2 - 2018 ISPRS TC III Mid-Term Symposium on Developments, Technologies and Applications in Remote Sensing
Y2 - 7 May 2018 through 10 May 2018
ER -