Title :
Comparison of some feature subset selection methods for use in remote sensing image analysis
Author_Institution :
Space Applications Inst., Joint Res. Centre, Ispra, Italy
Abstract :
As feature subset selection constitutes an important aspect of data fusion in general, this paper compares different measures of goodness and their influence on the classification results. These measures are 1) Fukunaga´s (1990) criterion, and 2) the ML criterion with a user-specified upper limit for the total error (Smits). Results are presented using publicly available multi-spectral/multi-sensor and hyperspectral images, and it is concluded that the ML criterion with a user-specified upper limit for the total error is a valid alternative to classical methods
Keywords :
geophysical signal processing; image classification; maximum likelihood estimation; remote sensing; sensor fusion; Fukunaga´s criterion; ML criterion; classification results; data fusion; goodness; hyperspectral images; multi-spectral/multi-sensor images; pattern recognition; remote sensing image analysis; subset selection methods; total error; user-specified upper limit; Costs; Data analysis; Extraterrestrial measurements; Hyperspectral imaging; Hyperspectral sensors; Image analysis; Pattern classification; Pattern recognition; Remote sensing; Training data;
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.2001.976212
