Title :
Image fusion of radar and optical remote sensing data for land cover classification
Author :
Nsaibi, Maher ; Chaabane, Ferdaous
Author_Institution :
Unite de Rech. en Imagerie Satellitaire et ses Applic.-URISA, Ecole Super. des Telecommun. de Tunis, Tunis
Abstract :
The aim of this paper is to propose a new unsupervised land cover classification method based on probabilistic fusion theory. This method combines two different Besag Markovian auto models: a Markovian Gamma auto model that characterizes the radar texture data and a Gaussian Markov Random Field auto model to characterize the optical spectral data. An optimal Markovian neighborhood order is also applied in order to improve the speckle texture modeling.
Keywords :
Gaussian processes; Markov processes; geophysical signal processing; image classification; image fusion; image texture; optical images; probability; radar imaging; remote sensing by radar; speckle; spectral analysis; Besag Markovian auto models; Gaussian Markov random field auto model; Markovian Gamma auto model; image fusion; optical remote sensing data; optical spectral data; optimal Markovian neighborhood order; probabilistic fusion theory; radar data; speckle texture modeling; unsupervised land cover classification method; Adaptive optics; Geometrical optics; Image fusion; Laser radar; Optical filters; Optical sensors; Radar imaging; Radar remote sensing; Remote sensing; Speckle; Besag auto models; Markov neighborhood order; component: Land cover classification;
Conference_Titel :
Information and Communication Technologies: From Theory to Applications, 2008. ICTTA 2008. 3rd International Conference on
Conference_Location :
Damascus
Print_ISBN :
978-1-4244-1751-3
Electronic_ISBN :
978-1-4244-1752-0
DOI :
10.1109/ICTTA.2008.4530043
