|
|
 |
|
 |
Last name :
First name :
Position :
|
Lacoste
Caroline
PhD Student
|
| Telephone : |
(33)4-92-38-75-67 |
| Fax : |
(33)4-92-38-76-43 |
| Email : |
caroline.lacoste@sophia.inria.fr |
| Address : |
Projet Ariana
Inria Sophia-Antipolis
2004 route des lucioles, BP 93
06902 Sophia Antipolis, France |
|
My PhD work involves with unsupervised extraction of linear objects from
satellite and aerial images. The final application would be either the
production or the update of geographic data. The sought-after
cartographic item consists of the different line networks such as
hydrographic networks or road networks. The approach is to build a
Markov object process in order to model the line network present in the
image. Indeed, these recently developed models provide the same type of
stochastic properties as those of Markov fields, while incorporating
strong geometric constraints. Optimization is realized by simulated
annealing using a Reversible Jump Markov Chain Monte Carlo (RJMCMC)
algorithm. The first results have been validated by the French
Geological Survey (BRGM) which partially supports my PhD.
|
Back to people
|
|
