Abstract
This paper presents a novel application of machine learning techniques to the automatic detection of building rooftops in satellite images. The image is first segmented into homogeneous regions using the k-means algorithm. These segments are then treated as candidate rooftop regions which are presented to a novel two-stage classification process; features are extracted from each segment and submitted to an ANN which serves as the first stage of the classification procedure. New features are then extracted from the outputs of the ANN and these are presented to an SVM which then performs the second classification pass. In this way, the first classification stage acts as a preprocessing step which, when processed by the SVM significantly reduces the number of false-positives. To establish the efficacy of the proposed method, its results are compared with those obtained using an alternative approach.
| Original language | English |
|---|---|
| Title of host publication | Proceedings - UKSim-AMSS 16th International Conference on Computer Modelling and Simulation, UKSim 2014 |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 286-291 |
| Number of pages | 6 |
| ISBN (Electronic) | 9781479949236 |
| DOIs | |
| Publication status | Published - 2014 |
| Event | 16th UKSim-AMSS International Conference on Computer Modelling and Simulation, UKSim 2014 - Cambridge, United Kingdom Duration: Mar 26 2014 → Mar 28 2014 |
Other
| Other | 16th UKSim-AMSS International Conference on Computer Modelling and Simulation, UKSim 2014 |
|---|---|
| Country | United Kingdom |
| City | Cambridge |
| Period | 3/26/14 → 3/28/14 |
Fingerprint
Keywords
- Artificial neural network
- Computer vision
- Image segmentation
- Machine learning
- Rooftop detection
- Support vector machine
ASJC Scopus subject areas
- Computational Theory and Mathematics
- Applied Mathematics
- Modelling and Simulation
Cite this
Automatic rooftop detection using a two-stage classification. / Joshi, Bikash; Baluyan, Hayk; Al Hinai, Amer; Woon, Wei Lee.
Proceedings - UKSim-AMSS 16th International Conference on Computer Modelling and Simulation, UKSim 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 286-291 7046079.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Automatic rooftop detection using a two-stage classification
AU - Joshi, Bikash
AU - Baluyan, Hayk
AU - Al Hinai, Amer
AU - Woon, Wei Lee
PY - 2014
Y1 - 2014
N2 - This paper presents a novel application of machine learning techniques to the automatic detection of building rooftops in satellite images. The image is first segmented into homogeneous regions using the k-means algorithm. These segments are then treated as candidate rooftop regions which are presented to a novel two-stage classification process; features are extracted from each segment and submitted to an ANN which serves as the first stage of the classification procedure. New features are then extracted from the outputs of the ANN and these are presented to an SVM which then performs the second classification pass. In this way, the first classification stage acts as a preprocessing step which, when processed by the SVM significantly reduces the number of false-positives. To establish the efficacy of the proposed method, its results are compared with those obtained using an alternative approach.
AB - This paper presents a novel application of machine learning techniques to the automatic detection of building rooftops in satellite images. The image is first segmented into homogeneous regions using the k-means algorithm. These segments are then treated as candidate rooftop regions which are presented to a novel two-stage classification process; features are extracted from each segment and submitted to an ANN which serves as the first stage of the classification procedure. New features are then extracted from the outputs of the ANN and these are presented to an SVM which then performs the second classification pass. In this way, the first classification stage acts as a preprocessing step which, when processed by the SVM significantly reduces the number of false-positives. To establish the efficacy of the proposed method, its results are compared with those obtained using an alternative approach.
KW - Artificial neural network
KW - Computer vision
KW - Image segmentation
KW - Machine learning
KW - Rooftop detection
KW - Support vector machine
UR - http://www.scopus.com/inward/record.url?scp=84926681766&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84926681766&partnerID=8YFLogxK
U2 - 10.1109/UKSim.2014.89
DO - 10.1109/UKSim.2014.89
M3 - Conference contribution
AN - SCOPUS:84926681766
SP - 286
EP - 291
BT - Proceedings - UKSim-AMSS 16th International Conference on Computer Modelling and Simulation, UKSim 2014
PB - Institute of Electrical and Electronics Engineers Inc.
ER -