Spatial information retrieval from remote-sensing images. II. Gibbs-Markov random fields

Abstract

We present Gibbs-Markov random field (GMRF) models as a powerful and robust descriptor of spatial information in typical remote-sensing image data. This class of stochastic image models provides an intuitive description of the image data using parameters of an energy function. For the selection among several nested models and the fit of the model, we proceed in two steps of Bayesian inference. This procedure yields the most plausible model and its most likely parameters, which together describe the image content in an optimal way. Its additional application at multiple scales of the image enables us to capture all structures being present in complex remote-sensing images. The calculation of the evidences of various models applied to the resulting quasicontinuous image pyramid automatically detects such structures. We present examples for both synthetic aperture radar (SAR) and optical data.

Intensity-driven adaptive-neighborhood technique for polarimetric and interferometric SAR parameters estimation

“…2(b)]. With a single gamma distribution, the corresponding interval is (see thresholds in IDAN (15) and IDAN (16) are set in order to retain only 50%, and 95%…”

“…The first approach is based on the assumption of local stationarity and yields to different filters which search for local neighborhoods respecting this assumption and use adaptive estimators such as the local linear minimum mean square error (LLMMSE) [12], [13]. The second approach corresponds to nonstationary speckle models which require to introduce prior knowledge on the scene such as the distribution of the intensity mean in the gamma filter [14] or multilevel pdfs in multiply stochastic models [15], [16].…”

Spatial statistical methods in health

“…In general, remote sensing and image processing techniques increasingly play a key role (e.g. see Datcu et al, 1998;Schroder et al, 1998;Stein et al, 1998b) as do advances in modelling of Markov Random Fields (e.g. see Aykroyd, 1998;Cressie & Davidson, 1998;Tjelmeland & Besag, 1998).…”

A New Automatic Road Extraction Technique using Gradient Operation and Skeletal Ray Formation

“…The identification of linear features by means of digital image analysis is a generic task in remote sensing [12]. Remote sensing image data typically contain an enormous amount of information [8]. Remote sensing is used to obtain information about a target or an area or a phenomenon through the analysis of certain information obtained by the remote sensor [14].…”