Computer Science – Information Theory
Scientific paper
May 2001
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2001aipc..568..531s&link_type=abstract
BAYESIAN INFERENCE AND MAXIMUM ENTROPY METHODS IN SCIENCE AND ENGINEERING: 20th International Workshop. AIP Conference Proceedi
Computer Science
Information Theory
Telemetry: Remote Control, Remote Sensing, Radar, Image Processing, Information Theory And Communication Theory
Scientific paper
In this work we use stochastic geometry theory and Monte Carlo Markov Chains dynamics to extract roads in remote sensed images. The main hypothesis for this new approach is that the road network is a thin network which is composed of connected segments. To simulate and to extract thin networks in images we construct a Gibbs point process. The energy of this Gibbs process has two components: the external and the internal energies. The external energy gives the location of the network. The internal energy (Candy model) penalizes the segments which are not connected and not well aligned. The optimum configuration of the segments is estimated by the global minimum of the total energy of the Gibbs process. For the minimization procedure we design a simulated annealing algorithm based on a Metropolis-Hastings dynamics including Reversible Jumps. Finally, some results are shown on satellite images. .
Descombes X.
Stoica Radu
Zerubia Josiane
No associations
LandOfFree
Road extraction in remote sensed images using stochastic geometry framework does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Road extraction in remote sensed images using stochastic geometry framework, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Road extraction in remote sensed images using stochastic geometry framework will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-924097