Title :
Parameter estimation and classification of multiscale remote sensing data
Author :
Storvik, G. ; Fjotoft, R. ; Solberg, A.H.S.
Author_Institution :
Oslo Univ., Norway
Abstract :
In this article we describe a Bayesian model for integration of multiscale image data. The approach is based on the concept of a reference resolution. Data at this and lower resolutions are connected to the reference resolution through a fully specified statistical model. Algorithms for parameter estimation and classification based on the multiscale model are proposed, and results and comparisons with singlescale classification are presented.
Keywords :
Bayes methods; geophysical signal processing; geophysical techniques; image classification; parameter estimation; radar resolution; remote sensing by radar; Bayesian model; multiscale image data; multiscale remote sensing data; parameter classification; parameter estimation; reference resolution; single-scale classification; statistical model; Artificial satellites; Bayesian methods; Data analysis; Electromagnetic spectrum; Image fusion; Image resolution; Markov random fields; Parameter estimation; Pixel; Remote sensing;
Conference_Titel :
Geoscience and Remote Sensing Symposium, 2003. IGARSS '03. Proceedings. 2003 IEEE International
Print_ISBN :
0-7803-7929-2
DOI :
10.1109/IGARSS.2003.1294007
