FEATURE SELECTION FOR SATELLITE IMAGE INDEXING

Lots of feature extraction procedures can be found in the literature. We propose to automatically select the best ones using automatic supervised feature selection algo- rithms. We demonstrate the efficiency of such a method- ology in comparing texture features sets (using Haral- ick coefficients, Gaussian Markov Random Fields, sev- eral wavelets,...) computed on satellite images. We il- lustrate the fact that geometrical features and texture fea- tures have to be combined to enhance the discriminative power of the selected features set.