Loading...
Please wait, while we are loading the content...
Similar Documents
Segmentation d'images multispectrales par arbre de Markov caché flou
| Content Provider | Semantic Scholar |
|---|---|
| Author | Lanchantin, Pierre Fabien, Salzenstein |
| Copyright Year | 2004 |
| Abstract | A new hidden fuzzy Markov tree model is developed in an unsupervised way. Our fuzzy scheme combines the uncertainty of probabilities which models the observed data with discrete and continuous thematic classes which models the imprecision of the hidden data. Segmentation task is processed with Bayesian tools, such as the MPM (Mode of Posterior Marginals) criterion. Indeed, such fuzzy-based procedures seem to be a good answer for astronomical observations when patterns own diffuse structures and multiscale observations under dependence assumption. To validate our model, we perform the segmentation on synthetic images and raw multispectral data. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://documents.irevues.inist.fr/bitstream/handle/2042/13995/A383_37633.pdf;jsessionid=83F6BBB790759FCE5D0552753F77D497?sequence=1 |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
