Key points are not available for this paper at this time.
This study compares several filtering techniqueprove to improve the classification of a synthetic aperture raar (SAR) image. With this aim in view, we separate thematic information and noise by speckle analysis, and we propose filters adapted to speckle models. The evaluation includes a brief theoretical analysis, a visual interpretation, a signal-to-noise ratio analysis, a comparison of required computer times, as well as classification results. The study also shows the effect of active microwaves on crop identification in the case where there is a one-date one-look SAR image, the utilization of X-and C-bands, and HH polarization.
Durand et al. (Tue,) studied this question.