Title :
A data-fusion approach to partially supervised classification
Author :
Prieto, Diego Fernàndez ; Arino, Olivier
Author_Institution :
Earth Obs. Applications Dept., Eur. Space Agency, Frascati, Italy
Abstract :
Considers the problem of partially supervised classification under a data-fusion perspective. The objective is to map one class (or only few classes) of interest in multisensor remote-sensing data by using exclusively training samples belonging to such class (or classes). The proposed methodology is based on a combined use of a radial basis function (RBF)-like network and a Markov random field (MRF) approach
Keywords :
sensor fusion; terrain mapping; Markov random field; data fusion; land cover; multisensor remote sensing data; partially supervised classification; radial basis function; Earth; Forestry; Impedance; Information analysis; Layout; Markov random fields; Maximum likelihood estimation; Remote sensing; Statistical analysis; Statistical distributions;
Conference_Titel :
Geoscience and Remote Sensing Symposium, 2001. IGARSS '01. IEEE 2001 International
Conference_Location :
Sydney, NSW
Print_ISBN :
0-7803-7031-7
DOI :
10.1109/IGARSS.2001.976660
