Details
| Original language | English |
|---|---|
| Pages (from-to) | 209-214 |
| Number of pages | 6 |
| Journal | International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives |
| Volume | 38 |
| Issue number | 4W19 |
| Publication status | Published - 5 Sept 2011 |
| Event | 2011 ISPRS Hannover Workshop on High-Resolution Earth Imaging for Geospatial Information - Hannover, Germany Duration: 14 Jun 2011 → 17 Jun 2011 |
Abstract
In this paper we propose a probabilistic supervised classification algorithm for LiDAR (Light Detection And Ranging) point clouds. Several object classes (i.e. ground, building and vegetation) can be separated reliably by considering each point's neighbourhood. Based on Conditional Random Fields (CRF) this contextual information can be incorporated into classification process in order to improve results. Since we want to perform a point-wise classification, no primarily segmentation is needed. Therefore, each 3D point is regarded as a graph's node, whereas edges represent links to the nearest neighbours. Both nodes and edges are associated with features and have effect on the classification. We use some features available from full waveform technology such as amplitude, echo width and number of echoes as well as some extracted geometrical features. The aim of the paper is to describe the CRF model set-up for irregular point clouds, present the features used for classification, and to discuss some results. The resulting overall accuracy is about 94 %.
Keywords
- 3D point cloud, Classification, Conditional Random Fields, LiDAR
ASJC Scopus subject areas
- Computer Science(all)
- Information Systems
- Social Sciences(all)
- Geography, Planning and Development
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, Vol. 38, No. 4W19, 05.09.2011, p. 209-214.
Research output: Contribution to journal › Conference article › Research › peer review
}
TY - JOUR
T1 - Conditional Random Fields for the classification of LiDAR point clouds
AU - Niemeyer, J.
AU - Mallet, C.
AU - Rottensteiner, F.
AU - Sörgel, U.
PY - 2011/9/5
Y1 - 2011/9/5
N2 - In this paper we propose a probabilistic supervised classification algorithm for LiDAR (Light Detection And Ranging) point clouds. Several object classes (i.e. ground, building and vegetation) can be separated reliably by considering each point's neighbourhood. Based on Conditional Random Fields (CRF) this contextual information can be incorporated into classification process in order to improve results. Since we want to perform a point-wise classification, no primarily segmentation is needed. Therefore, each 3D point is regarded as a graph's node, whereas edges represent links to the nearest neighbours. Both nodes and edges are associated with features and have effect on the classification. We use some features available from full waveform technology such as amplitude, echo width and number of echoes as well as some extracted geometrical features. The aim of the paper is to describe the CRF model set-up for irregular point clouds, present the features used for classification, and to discuss some results. The resulting overall accuracy is about 94 %.
AB - In this paper we propose a probabilistic supervised classification algorithm for LiDAR (Light Detection And Ranging) point clouds. Several object classes (i.e. ground, building and vegetation) can be separated reliably by considering each point's neighbourhood. Based on Conditional Random Fields (CRF) this contextual information can be incorporated into classification process in order to improve results. Since we want to perform a point-wise classification, no primarily segmentation is needed. Therefore, each 3D point is regarded as a graph's node, whereas edges represent links to the nearest neighbours. Both nodes and edges are associated with features and have effect on the classification. We use some features available from full waveform technology such as amplitude, echo width and number of echoes as well as some extracted geometrical features. The aim of the paper is to describe the CRF model set-up for irregular point clouds, present the features used for classification, and to discuss some results. The resulting overall accuracy is about 94 %.
KW - 3D point cloud
KW - Classification
KW - Conditional Random Fields
KW - LiDAR
UR - http://www.scopus.com/inward/record.url?scp=84923876399&partnerID=8YFLogxK
U2 - 10.5194/isprsarchives-XXXVIII-4-W19-209-2011
DO - 10.5194/isprsarchives-XXXVIII-4-W19-209-2011
M3 - Conference article
AN - SCOPUS:84923876399
VL - 38
SP - 209
EP - 214
JO - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
JF - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
SN - 1682-1750
IS - 4W19
T2 - 2011 ISPRS Hannover Workshop on High-Resolution Earth Imaging for Geospatial Information
Y2 - 14 June 2011 through 17 June 2011
ER -