Band selection in hyperspectral imagery using sparse support vector machines

hi tins paper we propose an l(1)-norm penalized sparse support vector machine (SSVM) as an embedded approach to the hyperspectral imagery band selection prohlem. SSVMs exhihit, a model structure that includes a dearly ifiable gap between zero mid non-zero weights that permits iniportant, bands to be definitively selected in conjunction with the classification problem. The SSVM Algorithm is trained using bootstrap aggregating to obtain a sample of SSVM models to reduce variability in the band selection process. This preliminary sample approach for hand selection is followed by a secondary hand selection which involves retraining the SSVM to further reduce the set of bands retained. We propose and compare three adaptations of the SSVM band selection algorithm for the multiclass problem. Two extensions of the SSVAI Algorithm are based on pairwise band selection between classes. Their performance is validated by using one-against-one (OAO) SSVMs. The third proposed method is a combination of the filter band selection method WaLuMI in sequence with the (0A0) SSVM embedded band selection algorithm. We illustrate the perfomance of these methods on the AVIRIS Indian Pines data set and compare the results to other techniques in the literature. Additionally we illustrate the SSVM Algorithm on the Long- Wavelength Infrared (LWIR) data set consisting of ltyperspectral videos of chentical plumes.