Abstract
Feature fusion of remote sensing images and LiDAR points cloud data, which have strong complementarity, can effectively play the advantages of multi-class features to provide more reliable information support for the remote sensing applications, such as object classification and recognition. In this paper, we introduce a novel multi-source hierarchical conditional random field (MSHCRF) model to fuse features extracted from remote sensing images and LiDAR data for image classification. Firstly, typical features are selected to obtain the interest regions from multi-source data, then MSHCRF model is constructed to exploit up the features, category compatibility of images and the category consistency of multi-source data based on the regions, and the outputs of the model represents the optimal results of the image classification. Competitive results demonstrate the precision and robustness of the proposed method.
| Original language | English |
|---|---|
| Title of host publication | ISPRS Hannover Workshop 2013 (Volume XL-1/W1) |
| Subtitle of host publication | WG I/4, III/4, IC IV/VIII, VII/2 |
| Editors | C. Heipke, K. Jacobsen, F. Rottensteiner, U. Sörgel |
| Publisher | International Society for Photogrammetry and Remote Sensing (ISPRS) |
| Pages | 389-392 |
| Number of pages | 4 |
| Volume | XL |
| Edition | 1W1 |
| DOIs | |
| Publication status | Published - 1 Jan 2013 |
| Event | ISPRS Hannover Workshop 2013 - Hannover, Germany Duration: 21 May 2013 → 24 May 2013 |
Publication series
| Name | International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives |
|---|---|
| Publisher | Copernicus |
| ISSN (Print) | 1682-1750 |
Conference
| Conference | ISPRS Hannover Workshop 2013 |
|---|---|
| Country | Germany |
| City | Hannover |
| Period | 21/05/13 → 24/05/13 |
Fingerprint
Keywords
- Conditional random field
- Feature fusion
- Hierarchical model
- Image classification
- Multi-source data
- ITC-GOLD
Cite this
}
Multi-source hierarchical conditional random field model for feature fusion of remote sensing images and LiDAR data. / Zhang, Z.; Yang, M. Y.; Zhoua, M.
ISPRS Hannover Workshop 2013 (Volume XL-1/W1): WG I/4, III/4, IC IV/VIII, VII/2. ed. / C. Heipke; K. Jacobsen; F. Rottensteiner; U. Sörgel. Vol. XL 1W1. ed. International Society for Photogrammetry and Remote Sensing (ISPRS), 2013. p. 389-392 (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review
TY - GEN
T1 - Multi-source hierarchical conditional random field model for feature fusion of remote sensing images and LiDAR data
AU - Zhang, Z.
AU - Yang, M. Y.
AU - Zhoua, M.
PY - 2013/1/1
Y1 - 2013/1/1
N2 - Feature fusion of remote sensing images and LiDAR points cloud data, which have strong complementarity, can effectively play the advantages of multi-class features to provide more reliable information support for the remote sensing applications, such as object classification and recognition. In this paper, we introduce a novel multi-source hierarchical conditional random field (MSHCRF) model to fuse features extracted from remote sensing images and LiDAR data for image classification. Firstly, typical features are selected to obtain the interest regions from multi-source data, then MSHCRF model is constructed to exploit up the features, category compatibility of images and the category consistency of multi-source data based on the regions, and the outputs of the model represents the optimal results of the image classification. Competitive results demonstrate the precision and robustness of the proposed method.
AB - Feature fusion of remote sensing images and LiDAR points cloud data, which have strong complementarity, can effectively play the advantages of multi-class features to provide more reliable information support for the remote sensing applications, such as object classification and recognition. In this paper, we introduce a novel multi-source hierarchical conditional random field (MSHCRF) model to fuse features extracted from remote sensing images and LiDAR data for image classification. Firstly, typical features are selected to obtain the interest regions from multi-source data, then MSHCRF model is constructed to exploit up the features, category compatibility of images and the category consistency of multi-source data based on the regions, and the outputs of the model represents the optimal results of the image classification. Competitive results demonstrate the precision and robustness of the proposed method.
KW - Conditional random field
KW - Feature fusion
KW - Hierarchical model
KW - Image classification
KW - Multi-source data
KW - ITC-GOLD
UR - http://www.scopus.com/inward/record.url?scp=84924713467&partnerID=8YFLogxK
UR - https://ezproxy2.utwente.nl/login?url=https://library.itc.utwente.nl/login/2013/conf/myang_mul.pdf
U2 - 10.5194/isprsarchives-XL-1-W1-389-2013
DO - 10.5194/isprsarchives-XL-1-W1-389-2013
M3 - Conference contribution
AN - SCOPUS:84924713467
VL - XL
T3 - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
SP - 389
EP - 392
BT - ISPRS Hannover Workshop 2013 (Volume XL-1/W1)
A2 - Heipke, C.
A2 - Jacobsen, K.
A2 - Rottensteiner, F.
A2 - Sörgel, U.
PB - International Society for Photogrammetry and Remote Sensing (ISPRS)
ER -