|
|
Publications about Markov random field model
Result of the query in the list of publications :
2 Articles |
1 - Comparative study on the performance of multi paramater SAR data for operational urban areas extraction.
C. Corbane and N. Baghdadi and X. Descombes and M. Petit. IEEE-Geoscience and Remote Sensing Letters, 6(4): pages 728-732, October 2009.
Keywords : Markov random field model, synthetic aperture radar, urban remote sensing.
@ARTICLE{COR-09,
|
| author |
= |
{Corbane, C. and Baghdadi, N. and Descombes, X. and Petit, M.}, |
| title |
= |
{Comparative study on the performance of multi paramater SAR data for operational urban areas extraction}, |
| year |
= |
{2009}, |
| month |
= |
{October}, |
| journal |
= |
{IEEE-Geoscience and Remote Sensing Letters}, |
| volume |
= |
{6}, |
| number |
= |
{4}, |
| pages |
= |
{728-732}, |
| url |
= |
{http://dx.doi.org/10.1109/LGRS.2009.2024225}, |
| keyword |
= |
{Markov random field model, synthetic aperture radar, urban remote sensing} |
| } |
Abstract :
The advent of a new generation of synthetic aperture radar (SAR) satellites, such as Advanced SAR/Environmental Satellite (C-band), Phased Array Type L-band Synthetic Aperture Radar/Advanced Land Observing Satellite (L-band), and TerraSAR-X (X-band), offers advanced potentials for the detection of urban tissue. In this letter, we analyze and compare the performance of multiple types of SAR images in terms of band frequency, polarization, incidence angle, and spatial resolution for the purpose of operational urban areas delineation. As a reference for comparison, we use a proven method for extracting textural features based on a Gaussian Markov Random Field (GMRF) model. The results of urban areas delineation are quantitatively analyzed allowing performing intrasensor and intersensors comparisons. Sensitivity of the GMRF model with respect to texture window size and to spatial resolutions of SAR images is also investigated. Intrasensor comparison shows that polarization and incidence angle play a significant role in the potential of the GMRF model for the extraction of urban areas from SAR images. Intersensors comparison evidences the better performances of X-band images, acquired at 1-m spatial resolution, when resampled to resolutions of 5 and 10 m. |
|
2 - Unsupervised parallel image classification using Markovian models.
Z. Kato and J. Zerubia and M. Berthod. Pattern Recognition, 32(4): pages 591-604, 1999.
Keywords : Markov random field model, Hierarchical model, Parameter estimation, Parallel unsupervised image classification.
@ARTICLE{jz99a,
|
| author |
= |
{Kato, Z. and Zerubia, J. and Berthod, M.}, |
| title |
= |
{Unsupervised parallel image classification using Markovian models}, |
| year |
= |
{1999}, |
| journal |
= |
{Pattern Recognition}, |
| volume |
= |
{32}, |
| number |
= |
{4}, |
| pages |
= |
{591-604}, |
| pdf |
= |
{http://dx.doi.org/10.1016/S0031-3203(98)00104-6}, |
| keyword |
= |
{Markov random field model, Hierarchical model, Parameter estimation, Parallel unsupervised image classification} |
| } |
Abstract :
This paper deals with the problem of unsupervised classification of images modeled by Markov random fields (MRF). If the model parameters are known then we have various methods to solve the segmentation problem (simulated annealing (SA), iterated conditional modes (ICM), etc). However, when the parameters are unknown, the problem becomes more difficult. One has to estimate the hidden label field parameters only from the observed image. Herein, we are interested in parameter estimation methods related to monogrid and hierarchical MRF models. The basic idea is similar to the expectation–maximization (EM) algorithm: we recursively look at the maximum a posteriori (MAP) estimate of the label field given the estimated parameters, then we look at the maximum likelihood (ML) estimate of the parameters given a tentative labeling obtained at the previous step. The only parameter supposed to be known is the number of classes, all the other parameters are estimated. The proposed algorithms have been implemented on a Connection Machine CM200. Comparative experiments have been performed on both noisy synthetic data and real images. |
|
 top of the page
These pages were generated by 
|
New demo webpage
get the bibtex
