Details
| Original language | English |
|---|---|
| Title of host publication | 2011 Joint Urban Remote Sensing Event, JURSE 2011 - Proceedings |
| Pages | 37-40 |
| Number of pages | 4 |
| Publication status | Published - 2011 |
| Event | IEEE GRSS and ISPRS Joint Urban Remote Sensing Event, JURSE 2011 - Munich, Germany Duration: 11 Apr 2011 → 13 Apr 2011 |
Publication series
| Name | 2011 Joint Urban Remote Sensing Event, JURSE 2011 - Proceedings |
|---|
Abstract
In this paper we present a SVM-based method for automatic quality control of a road database in urban areas. The road verification is carried out by comparing the database objects to high-resolution aerial imagery. The method is trimmed to produce reliable results even if the training data selection is partly non-epresentative. A reliability metric is assigned to the SVM decision that is based on the distance of a test object to the training data. This metric can be applied to any SVM-based classification task. Our experiments show that the classifier is very reliable in only accepting road objects that are actually correct.
ASJC Scopus subject areas
- Computer Science(all)
- Computer Networks and Communications
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
2011 Joint Urban Remote Sensing Event, JURSE 2011 - Proceedings. 2011. p. 37-40 5764713 (2011 Joint Urban Remote Sensing Event, JURSE 2011 - Proceedings).
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - SVM-based road verification with partly non-representative training data
AU - Ziems, Marcel
AU - Heipke, Christian
AU - Rottensteiner, Franz
PY - 2011
Y1 - 2011
N2 - In this paper we present a SVM-based method for automatic quality control of a road database in urban areas. The road verification is carried out by comparing the database objects to high-resolution aerial imagery. The method is trimmed to produce reliable results even if the training data selection is partly non-epresentative. A reliability metric is assigned to the SVM decision that is based on the distance of a test object to the training data. This metric can be applied to any SVM-based classification task. Our experiments show that the classifier is very reliable in only accepting road objects that are actually correct.
AB - In this paper we present a SVM-based method for automatic quality control of a road database in urban areas. The road verification is carried out by comparing the database objects to high-resolution aerial imagery. The method is trimmed to produce reliable results even if the training data selection is partly non-epresentative. A reliability metric is assigned to the SVM decision that is based on the distance of a test object to the training data. This metric can be applied to any SVM-based classification task. Our experiments show that the classifier is very reliable in only accepting road objects that are actually correct.
UR - http://www.scopus.com/inward/record.url?scp=79957630302&partnerID=8YFLogxK
U2 - 10.1109/JURSE.2011.5764713
DO - 10.1109/JURSE.2011.5764713
M3 - Conference contribution
AN - SCOPUS:79957630302
SN - 9781424486571
T3 - 2011 Joint Urban Remote Sensing Event, JURSE 2011 - Proceedings
SP - 37
EP - 40
BT - 2011 Joint Urban Remote Sensing Event, JURSE 2011 - Proceedings
T2 - IEEE GRSS and ISPRS Joint Urban Remote Sensing Event, JURSE 2011
Y2 - 11 April 2011 through 13 April 2011
ER -