USING SUPPORT VECTOR MACHINE LEARNING TO AUTOMATICA LLY INTERPRET MODIS, ALI, AND L-BAND SAR REMOTELY SENSED IMAGERY FOR HYDROLOGY, LAND COVER, AND CRYOSPHERE APPLICATIONS

We report on efforts to automatically interpret rem otely sensed radar and multispectral imagery using machin elearned classifiers. Specifically we utilize Suppo rt Vector Machine (SVM) learning techniques for L-band SAR, EO-1/ALI, and MODIS Imagery. We share our qualitative and quantitative results thus far and d iscuss challenges experienced.