Loading...
Please wait, while we are loading the content...
Similar Documents
Utilisation de méthodes de classification hiérarchique pour une classification supervisée d'images satellitaires
| Content Provider | Semantic Scholar |
|---|---|
| Author | Ouarab, Nadia Smara, Youcef Rasson, Jean-Paul |
| Copyright Year | 1999 |
| Abstract | The aim of this paper consists in detecting homogeneous regions in a satellite image. These regions could be used as a set of training data in the supervised classification. The supervised classification methods need the knowledge of earth landscapes and their nature. The ground is the only reference reliable for this task. The problem appears when this reality is not available. The solution we have adopted is the automatic selection of training data. We use nonparametric tests : univariate of Wilcoxon and an approach of supports comparison. For that purpose, we develop several hierarchical clustering methods, such as Single-linkage, Complete-linkage and Averagelinkage. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://documents.irevues.inist.fr/bitstream/handle/2042/13016/ARTI1369.pdf?sequence=1 |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
